Merge tag 's390-6.17-1' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux

Pull s390 updates from Alexander Gordeev:

 - Standardize on the __ASSEMBLER__ macro that is provided by GCC and
   Clang compilers and replace __ASSEMBLY__ with __ASSEMBLER__ in both
   uapi and non-uapi headers

 - Explicitly include <linux/export.h> in architecture and driver files
   which contain an EXPORT_SYMBOL() and remove the include from the
   files which do not contain the EXPORT_SYMBOL()

 - Use the full title of "z/Architecture Principles of Operation" manual
   and the name of a section where facility bits are listed

 - Use -D__DISABLE_EXPORTS for files in arch/s390/boot to avoid
   unnecessary slowing down of the build and confusing external kABI
   tools that process symtypes data

 - Print additional unrecoverable machine check information to make the
   root cause analysis easier

 - Move cmpxchg_user_key() handling to uaccess library code, since the
   generated code is large anyway and there is no benefit if it is
   inlined

 - Fix a problem when cmpxchg_user_key() is executing a code with a
   non-default key: if a system is IPL-ed with "LOAD NORMAL", and the
   previous system used storage keys where the fetch-protection bit was
   set for some pages, and the cmpxchg_user_key() is located within such
   page, a protection exception happens

 - Either the external call or emergency signal order is used to send an
   IPI to a remote CPU. Use the external order only, since it is at
   least as good and sometimes even better, than the emergency signal

 - In case of an early crash the early program check handler prints more
   or less random value of the last breaking event address, since it is
   not initialized properly. Copy the last breaking event address from
   the lowcore to pt_regs to address this

 - During STP synchronization check udelay() can not be used, since the
   first CPU modifies tod_clock_base and get_tod_clock_monotonic() might
   return a non-monotonic time. Instead, busy-loop on other CPUs, while
   the the first CPU actually handles the synchronization operation

 - When debugging the early kernel boot using QEMU with the -S flag and
   GDB attached, skip the decompressor and start directly in kernel

 - Rename PAI Crypto event 4210 according to z16 and z17 "z/Architecture
   Principles of Operation" manual

 - Remove the in-kernel time steering support in favour of the new s390
   PTP driver, which allows the kernel clock steered more precisely

 - Remove a possible false-positive warning in pte_free_defer(), which
   could be triggered in a valid case KVM guest process is initializing

* tag 's390-6.17-1' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux: (29 commits)
  s390/mm: Remove possible false-positive warning in pte_free_defer()
  s390/stp: Default to enabled
  s390/stp: Remove leap second support
  s390/time: Remove in-kernel time steering
  s390/sclp: Use monotonic clock in sclp_sync_wait()
  s390/smp: Use monotonic clock in smp_emergency_stop()
  s390/time: Use monotonic clock in get_cycles()
  s390/pai_crypto: Rename PAI Crypto event 4210
  scripts/gdb/symbols: make lx-symbols skip the s390 decompressor
  s390/boot: Introduce jump_to_kernel() function
  s390/stp: Remove udelay from stp_sync_clock()
  s390/early: Copy last breaking event address to pt_regs
  s390/smp: Remove conditional emergency signal order code usage
  s390/uaccess: Merge cmpxchg_user_key() inline assemblies
  s390/uaccess: Prevent kprobes on cmpxchg_user_key() functions
  s390/uaccess: Initialize code pages executed with non-default access key
  s390/skey: Provide infrastructure for executing with non-default access key
  s390/uaccess: Make cmpxchg_user_key() library code
  s390/page: Add memory clobber to page_set_storage_key()
  s390/page: Cleanup page_set_storage_key() inline assemblies
  ...

2 files changed, 189 insertions, 0 deletions

diff --git a/arch/s390/lib/delay.c b/arch/s390/lib/delay.c
index be14c58cb989..c1ea14e3c927 100644
--- a/arch/s390/lib/delay.c
+++ b/arch/s390/lib/delay.c
@@ -7,6 +7,7 @@
  */
 
 #include <linux/processor.h>
+#include <linux/export.h>
 #include <linux/delay.h>
 #include <asm/div64.h>
 #include <asm/timex.h>
diff --git a/arch/s390/lib/uaccess.c b/arch/s390/lib/uaccess.c
index fa7d98fa1320..1a6ba105e071 100644
--- a/arch/s390/lib/uaccess.c
+++ b/arch/s390/lib/uaccess.c
@@ -8,11 +8,13 @@
  *		 Gerald Schaefer (gerald.schaefer@de.ibm.com)
  */
 
+#include <linux/kprobes.h>
 #include <linux/uaccess.h>
 #include <linux/export.h>
 #include <linux/mm.h>
 #include <asm/asm-extable.h>
 #include <asm/ctlreg.h>
+#include <asm/skey.h>
 
 #ifdef CONFIG_DEBUG_ENTRY
 void debug_user_asce(int exit)
@@ -145,3 +147,189 @@ unsigned long _copy_to_user_key(void __user *to, const void *from,
 	return raw_copy_to_user_key(to, from, n, key);
 }
 EXPORT_SYMBOL(_copy_to_user_key);
+
+#define CMPXCHG_USER_KEY_MAX_LOOPS 128
+
+static nokprobe_inline int __cmpxchg_user_key_small(unsigned long address, unsigned int *uval,
+						    unsigned int old, unsigned int new,
+						    unsigned int mask, unsigned long key)
+{
+	unsigned long count;
+	unsigned int prev;
+	bool sacf_flag;
+	int rc = 0;
+
+	skey_regions_initialize();
+	sacf_flag = enable_sacf_uaccess();
+	asm_inline volatile(
+		"20:	spka	0(%[key])\n"
+		"	sacf	256\n"
+		"	llill	%[count],%[max_loops]\n"
+		"0:	l	%[prev],%[address]\n"
+		"1:	nr	%[prev],%[mask]\n"
+		"	xilf	%[mask],0xffffffff\n"
+		"	or	%[new],%[prev]\n"
+		"	or	%[prev],%[tmp]\n"
+		"2:	lr	%[tmp],%[prev]\n"
+		"3:	cs	%[prev],%[new],%[address]\n"
+		"4:	jnl	5f\n"
+		"	xr	%[tmp],%[prev]\n"
+		"	xr	%[new],%[tmp]\n"
+		"	nr	%[tmp],%[mask]\n"
+		"	jnz	5f\n"
+		"	brct	%[count],2b\n"
+		"5:	sacf	768\n"
+		"	spka	%[default_key]\n"
+		"21:\n"
+		EX_TABLE_UA_LOAD_REG(0b, 5b, %[rc], %[prev])
+		EX_TABLE_UA_LOAD_REG(1b, 5b, %[rc], %[prev])
+		EX_TABLE_UA_LOAD_REG(3b, 5b, %[rc], %[prev])
+		EX_TABLE_UA_LOAD_REG(4b, 5b, %[rc], %[prev])
+		SKEY_REGION(20b, 21b)
+		: [rc] "+&d" (rc),
+		[prev] "=&d" (prev),
+		[address] "+Q" (*(int *)address),
+		[tmp] "+&d" (old),
+		[new] "+&d" (new),
+		[mask] "+&d" (mask),
+		[count] "=a" (count)
+		: [key] "%[count]" (key << 4),
+		[default_key] "J" (PAGE_DEFAULT_KEY),
+		[max_loops] "J" (CMPXCHG_USER_KEY_MAX_LOOPS)
+		: "memory", "cc");
+	disable_sacf_uaccess(sacf_flag);
+	*uval = prev;
+	if (!count)
+		rc = -EAGAIN;
+	return rc;
+}
+
+int __kprobes __cmpxchg_user_key1(unsigned long address, unsigned char *uval,
+				  unsigned char old, unsigned char new, unsigned long key)
+{
+	unsigned int prev, shift, mask, _old, _new;
+	int rc;
+
+	shift = (3 ^ (address & 3)) << 3;
+	address ^= address & 3;
+	_old = (unsigned int)old << shift;
+	_new = (unsigned int)new << shift;
+	mask = ~(0xff << shift);
+	rc = __cmpxchg_user_key_small(address, &prev, _old, _new, mask, key);
+	*uval = prev >> shift;
+	return rc;
+}
+EXPORT_SYMBOL(__cmpxchg_user_key1);
+
+int __kprobes __cmpxchg_user_key2(unsigned long address, unsigned short *uval,
+				  unsigned short old, unsigned short new, unsigned long key)
+{
+	unsigned int prev, shift, mask, _old, _new;
+	int rc;
+
+	shift = (2 ^ (address & 2)) << 3;
+	address ^= address & 2;
+	_old = (unsigned int)old << shift;
+	_new = (unsigned int)new << shift;
+	mask = ~(0xffff << shift);
+	rc = __cmpxchg_user_key_small(address, &prev, _old, _new, mask, key);
+	*uval = prev >> shift;
+	return rc;
+}
+EXPORT_SYMBOL(__cmpxchg_user_key2);
+
+int __kprobes __cmpxchg_user_key4(unsigned long address, unsigned int *uval,
+				  unsigned int old, unsigned int new, unsigned long key)
+{
+	unsigned int prev = old;
+	bool sacf_flag;
+	int rc = 0;
+
+	skey_regions_initialize();
+	sacf_flag = enable_sacf_uaccess();
+	asm_inline volatile(
+		"20:	spka	0(%[key])\n"
+		"	sacf	256\n"
+		"0:	cs	%[prev],%[new],%[address]\n"
+		"1:	sacf	768\n"
+		"	spka	%[default_key]\n"
+		"21:\n"
+		EX_TABLE_UA_LOAD_REG(0b, 1b, %[rc], %[prev])
+		EX_TABLE_UA_LOAD_REG(1b, 1b, %[rc], %[prev])
+		SKEY_REGION(20b, 21b)
+		: [rc] "+&d" (rc),
+		[prev] "+&d" (prev),
+		[address] "+Q" (*(int *)address)
+		: [new] "d" (new),
+		[key] "a" (key << 4),
+		[default_key] "J" (PAGE_DEFAULT_KEY)
+		: "memory", "cc");
+	disable_sacf_uaccess(sacf_flag);
+	*uval = prev;
+	return rc;
+}
+EXPORT_SYMBOL(__cmpxchg_user_key4);
+
+int __kprobes __cmpxchg_user_key8(unsigned long address, unsigned long *uval,
+				  unsigned long old, unsigned long new, unsigned long key)
+{
+	unsigned long prev = old;
+	bool sacf_flag;
+	int rc = 0;
+
+	skey_regions_initialize();
+	sacf_flag = enable_sacf_uaccess();
+	asm_inline volatile(
+		"20:	spka	0(%[key])\n"
+		"	sacf	256\n"
+		"0:	csg	%[prev],%[new],%[address]\n"
+		"1:	sacf	768\n"
+		"	spka	%[default_key]\n"
+		"21:\n"
+		EX_TABLE_UA_LOAD_REG(0b, 1b, %[rc], %[prev])
+		EX_TABLE_UA_LOAD_REG(1b, 1b, %[rc], %[prev])
+		SKEY_REGION(20b, 21b)
+		: [rc] "+&d" (rc),
+		[prev] "+&d" (prev),
+		[address] "+QS" (*(long *)address)
+		: [new] "d" (new),
+		[key] "a" (key << 4),
+		[default_key] "J" (PAGE_DEFAULT_KEY)
+		: "memory", "cc");
+	disable_sacf_uaccess(sacf_flag);
+	*uval = prev;
+	return rc;
+}
+EXPORT_SYMBOL(__cmpxchg_user_key8);
+
+int __kprobes __cmpxchg_user_key16(unsigned long address, __uint128_t *uval,
+				   __uint128_t old, __uint128_t new, unsigned long key)
+{
+	__uint128_t prev = old;
+	bool sacf_flag;
+	int rc = 0;
+
+	skey_regions_initialize();
+	sacf_flag = enable_sacf_uaccess();
+	asm_inline volatile(
+		"20:	spka	0(%[key])\n"
+		"	sacf	256\n"
+		"0:	cdsg	%[prev],%[new],%[address]\n"
+		"1:	sacf	768\n"
+		"	spka	%[default_key]\n"
+		"21:\n"
+		EX_TABLE_UA_LOAD_REGPAIR(0b, 1b, %[rc], %[prev])
+		EX_TABLE_UA_LOAD_REGPAIR(1b, 1b, %[rc], %[prev])
+		SKEY_REGION(20b, 21b)
+		: [rc] "+&d" (rc),
+		[prev] "+&d" (prev),
+		[address] "+QS" (*(__int128_t *)address)
+		: [new] "d" (new),
+		[key] "a" (key << 4),
+		[default_key] "J" (PAGE_DEFAULT_KEY)
+		: "memory", "cc");
+	disable_sacf_uaccess(sacf_flag);
+	*uval = prev;
+	return rc;
+}
+EXPORT_SYMBOL(__cmpxchg_user_key16);