aboutsummaryrefslogtreecommitdiffstats
path: root/tools/perf/scripts/python/stackcollapse.py (unfollow)
AgeCommit message (Collapse)AuthorFilesLines
2019-10-05scripts/setlocalversion: clear local variable to make it work for shMasahiro Yamada1-1/+1
Geert Uytterhoeven reports a strange side-effect of commit 858805b336be ("kbuild: add $(BASH) to run scripts with bash-extension"), which inserts the contents of a localversion file in the build directory twice. [Steps to Reproduce] $ echo bar > localversion $ mkdir build $ cd build/ $ echo foo > localversion $ make -s -f ../Makefile defconfig include/config/kernel.release $ cat include/config/kernel.release 5.4.0-rc1foofoobar This comes down to the behavior change of local variables. The 'man sh' on my Ubuntu machine, where sh is an alias to dash, explains as follows: When a variable is made local, it inherits the initial value and exported and readonly flags from the variable with the same name in the surrounding scope, if there is one. Otherwise, the variable is initially unset. [Test Code] foo () { local res echo "res: $res" } res=1 foo [Result] $ sh test.sh res: 1 $ bash test.sh res: So, scripts/setlocalversion correctly works only for bash in spite of its hashbang being #!/bin/sh. Nobody had noticed it before because CONFIG_SHELL was previously set to bash almost all the time. Now that CONFIG_SHELL is set to sh, we must write portable and correct code. I gave the Fixes tag to the commit that uncovered the issue. Clear the variable 'res' in collect_files() to make it work for sh (and it also works on distributions where sh is an alias to bash). Fixes: 858805b336be ("kbuild: add $(BASH) to run scripts with bash-extension") Reported-by: Geert Uytterhoeven <geert+renesas@glider.be> Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com> Tested-by: Geert Uytterhoeven <geert+renesas@glider.be>
2019-10-05namespace: fix namespace.pl script to support relative pathsJacob Keller1-6/+7
The namespace.pl script does not work properly if objtree is not set to an absolute path. The do_nm function is run from within the find function, which changes directories. Because of this, appending objtree, $File::Find::dir, and $source, will return a path which is not valid from the current directory. This used to work when objtree was set to an absolute path when using "make namespacecheck". It appears to have not worked when calling ./scripts/namespace.pl directly. This behavior was changed in 7e1c04779efd ("kbuild: Use relative path for $(objtree)", 2014-05-14) Rather than fixing the Makefile to set objtree to an absolute path, just fix namespace.pl to work when srctree and objtree are relative. Also fix the script to use an absolute path for these by default. Use the File::Spec module for this purpose. It's been part of perl 5 since 5.005. The curdir() function is used to get the current directory when the objtree and srctree aren't set in the environment. rel2abs() is used to convert possibly relative objtree and srctree environment variables to absolute paths. Finally, the catfile() function is used instead of string appending paths together, since this is more robust when joining paths together. Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Acked-by: Randy Dunlap <rdunlap@infradead.org> Tested-by: Randy Dunlap <rdunlap@infradead.org> Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
2019-10-05video/logo: do not generate unneeded logo C filesMasahiro Yamada1-19/+2
Currently, all the logo C files are generated irrespective of the CONFIG options. Adding them to extra-y is wrong. What we need to do here is to add them to 'targets' so that if_changed works properly. Files listed in 'targets' are cleaned, so clean-files is unneeded. Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
2019-10-05video/logo: remove unneeded *.o pattern from clean-filesMasahiro Yamada1-1/+1
The pattern *.o is cleaned up globally by the top Makefile. Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
2019-10-05integrity: remove pointless subdir-$(CONFIG_...)Masahiro Yamada1-2/+0
The ima/ and evm/ sub-directories contain built-in objects, so obj-$(CONFIG_...) is the correct way to descend into them. subdir-$(CONFIG_...) is redundant. Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
2019-10-05integrity: remove unneeded, broken attempt to add -fshort-wcharMasahiro Yamada1-1/+0
I guess commit 15ea0e1e3e18 ("efi: Import certificates from UEFI Secure Boot") attempted to add -fshort-wchar for building load_uefi.o, but it has never worked as intended. load_uefi.o is created in the platform_certs/ sub-directory. If you really want to add -fshort-wchar, the correct code is: $(obj)/platform_certs/load_uefi.o: KBUILD_CFLAGS += -fshort-wchar But, you do not need to fix it. Commit 8c97023cf051 ("Kbuild: use -fshort-wchar globally") had already added -fshort-wchar globally. This code was unneeded in the first place. Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
2019-10-01modpost: fix static EXPORT_SYMBOL warnings for UML buildMasahiro Yamada1-4/+9
Johannes Berg reports lots of modpost warnings on ARCH=um builds: WARNING: "rename" [vmlinux] is a static EXPORT_SYMBOL WARNING: "lseek" [vmlinux] is a static EXPORT_SYMBOL WARNING: "ftruncate64" [vmlinux] is a static EXPORT_SYMBOL WARNING: "getuid" [vmlinux] is a static EXPORT_SYMBOL WARNING: "lseek64" [vmlinux] is a static EXPORT_SYMBOL WARNING: "unlink" [vmlinux] is a static EXPORT_SYMBOL WARNING: "pwrite64" [vmlinux] is a static EXPORT_SYMBOL WARNING: "close" [vmlinux] is a static EXPORT_SYMBOL WARNING: "opendir" [vmlinux] is a static EXPORT_SYMBOL WARNING: "pread64" [vmlinux] is a static EXPORT_SYMBOL WARNING: "syscall" [vmlinux] is a static EXPORT_SYMBOL WARNING: "readdir" [vmlinux] is a static EXPORT_SYMBOL WARNING: "readdir64" [vmlinux] is a static EXPORT_SYMBOL WARNING: "futimes" [vmlinux] is a static EXPORT_SYMBOL WARNING: "__lxstat" [vmlinux] is a static EXPORT_SYMBOL WARNING: "write" [vmlinux] is a static EXPORT_SYMBOL WARNING: "closedir" [vmlinux] is a static EXPORT_SYMBOL WARNING: "__xstat" [vmlinux] is a static EXPORT_SYMBOL WARNING: "fsync" [vmlinux] is a static EXPORT_SYMBOL WARNING: "__lxstat64" [vmlinux] is a static EXPORT_SYMBOL WARNING: "__fxstat64" [vmlinux] is a static EXPORT_SYMBOL WARNING: "telldir" [vmlinux] is a static EXPORT_SYMBOL WARNING: "printf" [vmlinux] is a static EXPORT_SYMBOL WARNING: "readlink" [vmlinux] is a static EXPORT_SYMBOL WARNING: "__sprintf_chk" [vmlinux] is a static EXPORT_SYMBOL WARNING: "link" [vmlinux] is a static EXPORT_SYMBOL WARNING: "rmdir" [vmlinux] is a static EXPORT_SYMBOL WARNING: "fdatasync" [vmlinux] is a static EXPORT_SYMBOL WARNING: "truncate" [vmlinux] is a static EXPORT_SYMBOL WARNING: "statfs" [vmlinux] is a static EXPORT_SYMBOL WARNING: "__errno_location" [vmlinux] is a static EXPORT_SYMBOL WARNING: "__xmknod" [vmlinux] is a static EXPORT_SYMBOL WARNING: "open64" [vmlinux] is a static EXPORT_SYMBOL WARNING: "truncate64" [vmlinux] is a static EXPORT_SYMBOL WARNING: "open" [vmlinux] is a static EXPORT_SYMBOL WARNING: "read" [vmlinux] is a static EXPORT_SYMBOL WARNING: "chown" [vmlinux] is a static EXPORT_SYMBOL WARNING: "chmod" [vmlinux] is a static EXPORT_SYMBOL WARNING: "utime" [vmlinux] is a static EXPORT_SYMBOL WARNING: "fchmod" [vmlinux] is a static EXPORT_SYMBOL WARNING: "seekdir" [vmlinux] is a static EXPORT_SYMBOL WARNING: "ioctl" [vmlinux] is a static EXPORT_SYMBOL WARNING: "dup2" [vmlinux] is a static EXPORT_SYMBOL WARNING: "statfs64" [vmlinux] is a static EXPORT_SYMBOL WARNING: "utimes" [vmlinux] is a static EXPORT_SYMBOL WARNING: "mkdir" [vmlinux] is a static EXPORT_SYMBOL WARNING: "fchown" [vmlinux] is a static EXPORT_SYMBOL WARNING: "__guard" [vmlinux] is a static EXPORT_SYMBOL WARNING: "symlink" [vmlinux] is a static EXPORT_SYMBOL WARNING: "access" [vmlinux] is a static EXPORT_SYMBOL WARNING: "__stack_smash_handler" [vmlinux] is a static EXPORT_SYMBOL When you run "make", the modpost is run twice; before linking vmlinux, and before building modules. All the warnings above are from the second modpost. The offending symbols are defined not in vmlinux, but in the C library. The first modpost is run against the relocatable vmlinux.o, and those warnings are nicely suppressed because the SH_UNDEF entries from the symbol table clear the ->is_static flag. The second modpost is run against the executable vmlinux (+ modules), where those symbols have been resolved, but the definitions do not exist. This commit fixes it in a straightforward way; suppress the static EXPORT_SYMBOL warnings from "vmlinux". Without this commit, we see valid warnings twice anyway. For example, ARCH=arm64 defconfig shows the following warning twice: WARNING: "HYPERVISOR_platform_op" [vmlinux] is a static EXPORT_SYMBOL_GPL So, it is reasonable to suppress the second one. Fixes: 15bfc2348d54 ("modpost: check for static EXPORT_SYMBOL* functions") Reported-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com> Tested-by: Johannes Berg <johannes@sipsolutions.net> Tested-by: Denis Efremov <efremov@linux.com>
2019-10-01kbuild: correct formatting of header in kbuild module docsAlex Gaynor1-1/+2
Minor formatting fixup. Fixes: cd238effefa2 ("docs: kbuild: convert docs to ReST and rename to *.rst") Signed-off-by: Alex Gaynor <alex.gaynor@gmail.com> Signed-off-by: Matthew Garrett <mjg59@google.com> Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
2019-10-01kbuild: remove SUBDIRS supportMasahiro Yamada1-18/+2
Linux 5.3 is out. Remove the SUBDIRS support. Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
2019-10-01kbuild: remove ar-option and KBUILD_ARFLAGSMasahiro Yamada6-17/+3
Commit 40df759e2b9e ("kbuild: Fix build with binutils <= 2.19") introduced ar-option and KBUILD_ARFLAGS to deal with old binutils. According to Documentation/process/changes.rst, the current minimal supported version of binutils is 2.21 so you can assume the 'D' option is always supported. Not only GNU ar but also llvm-ar supports it. With the 'D' option hard-coded, there is no more user of ar-option or KBUILD_ARFLAGS. Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com> Reviewed-by: Nick Desaulniers <ndesaulniers@google.com> Tested-by: Nick Desaulniers <ndesaulniers@google.com>
2019-09-30Linux 5.4-rc1v5.4-rc1Linus Torvalds1-2/+2
2019-09-30csky: Move static keyword to the front of declarationKrzysztof Wilczynski1-1/+1
Move the static keyword to the front of declaration of csky_pmu_of_device_ids, and resolve the following compiler warning that can be seen when building with warnings enabled (W=1): arch/csky/kernel/perf_event.c:1340:1: warning: ‘static’ is not at beginning of declaration [-Wold-style-declaration] Signed-off-by: Krzysztof Wilczynski <kw@linux.com> Signed-off-by: Guo Ren <guoren@kernel.org>
2019-09-30csky: entry: Remove unneeded need_resched() loopValentin Schneider1-4/+0
Since the enabling and disabling of IRQs within preempt_schedule_irq() is contained in a need_resched() loop, we don't need the outer arch code loop. Signed-off-by: Valentin Schneider <valentin.schneider@arm.com> Signed-off-by: Guo Ren <guoren@kernel.org>
2019-09-30csky: Fixup csky_pmu.max_period assignmentMao Han1-1/+1
The csky_pmu.max_period has type u64, and BIT() can only return 32 bits unsigned long on C-SKY. The initialization for max_period will be incorrect when count_width is bigger than 32. Use BIT_ULL() Signed-off-by: Mao Han <han_mao@c-sky.com> Signed-off-by: Guo Ren <ren_guo@c-sky.com>
2019-09-30csky: Fixup add zero_fp fixup perf backtrace panicGuo Ren2-21/+31
We need set fp zero to let backtrace know the end. The patch fixup perf callchain panic problem, because backtrace didn't know what is the end of fp. Signed-off-by: Guo Ren <ren_guo@c-sky.com> Reported-by: Mao Han <han_mao@c-sky.com>
2019-09-30csky: Use generic free_initrd_mem()Mike Rapoport1-16/+0
The csky implementation of free_initrd_mem() is an open-coded version of free_reserved_area() without poisoning. Remove it and make csky use the generic version of free_initrd_mem(). Signed-off-by: Mike Rapoport <rppt@linux.ibm.com> Signed-off-by: Guo Ren <guoren@kernel.org>
2019-09-29Revert "Revert "ext4: make __ext4_get_inode_loc plug""Linus Torvalds1-0/+3
This reverts commit 72dbcf72156641fde4d8ea401e977341bfd35a05. Instead of waiting forever for entropy that may just not happen, we now try to actively generate entropy when required, and are thus hopefully avoiding the problem that caused the nice ext4 IO pattern fix to be reverted. So revert the revert. Cc: Ahmed S. Darwish <darwish.07@gmail.com> Cc: Ted Ts'o <tytso@mit.edu> Cc: Willy Tarreau <w@1wt.eu> Cc: Alexander E. Patrakov <patrakov@gmail.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-09-29random: try to actively add entropy rather than passively wait for itLinus Torvalds1-1/+61
For 5.3 we had to revert a nice ext4 IO pattern improvement, because it caused a bootup regression due to lack of entropy at bootup together with arguably broken user space that was asking for secure random numbers when it really didn't need to. See commit 72dbcf721566 (Revert "ext4: make __ext4_get_inode_loc plug"). This aims to solve the issue by actively generating entropy noise using the CPU cycle counter when waiting for the random number generator to initialize. This only works when you have a high-frequency time stamp counter available, but that's the case on all modern x86 CPU's, and on most other modern CPU's too. What we do is to generate jitter entropy from the CPU cycle counter under a somewhat complex load: calling the scheduler while also guaranteeing a certain amount of timing noise by also triggering a timer. I'm sure we can tweak this, and that people will want to look at other alternatives, but there's been a number of papers written on jitter entropy, and this should really be fairly conservative by crediting one bit of entropy for every timer-induced jump in the cycle counter. Not because the timer itself would be all that unpredictable, but because the interaction between the timer and the loop is going to be. Even if (and perhaps particularly if) the timer actually happens on another CPU, the cacheline interaction between the loop that reads the cycle counter and the timer itself firing is going to add perturbations to the cycle counter values that get mixed into the entropy pool. As Thomas pointed out, with a modern out-of-order CPU, even quite simple loops show a fair amount of hard-to-predict timing variability even in the absense of external interrupts. But this tries to take that further by actually having a fairly complex interaction. This is not going to solve the entropy issue for architectures that have no CPU cycle counter, but it's not clear how (and if) that is solvable, and the hardware in question is largely starting to be irrelevant. And by doing this we can at least avoid some of the even more contentious approaches (like making the entropy waiting time out in order to avoid the possibly unbounded waiting). Cc: Ahmed Darwish <darwish.07@gmail.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Theodore Ts'o <tytso@mit.edu> Cc: Nicholas Mc Guire <hofrat@opentech.at> Cc: Andy Lutomirski <luto@kernel.org> Cc: Kees Cook <keescook@chromium.org> Cc: Willy Tarreau <w@1wt.eu> Cc: Alexander E. Patrakov <patrakov@gmail.com> Cc: Lennart Poettering <mzxreary@0pointer.de> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-09-29Documentation/process: Clarify disclosure rulesThomas Gleixner1-7/+33
The role of the contact list provided by the disclosing party and how it affects the disclosure process and the ability to include experts into the development process is not really well explained. Neither is it entirely clear when the disclosing party will be informed about the fact that a developer who is not covered by an employer NDA needs to be brought in and disclosed. Explain the role of the contact list and the information policy along with an eventual conflict resolution better. Reported-by: Dave Hansen <dave.hansen@linux.intel.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Acked-by: Dave Hansen <dave.hansen@linux.intel.com> Link: https://lore.kernel.org/r/alpine.DEB.2.21.1909251028390.10825@nanos.tec.linutronix.de Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2019-09-28selftests/ftrace: Fix same probe error testSteven Rostedt (VMware)1-1/+1
The "same probe" selftest that tests that adding the same probe fails doesn't add the same probe and passes, which fails the test. Fixes: b78b94b82122 ("selftests/ftrace: Update kprobe event error testcase") Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2019-09-28mm, tracing: Print symbol name for call_site in trace eventsChangbin Du1-3/+4
To improve the readability of raw slab trace points, print the call_site ip using '%pS'. Then we can grep events with function names. [002] .... 808.188897: kmem_cache_free: call_site=putname+0x47/0x50 ptr=00000000cef40c80 [002] .... 808.188898: kfree: call_site=security_cred_free+0x42/0x50 ptr=0000000062400820 [002] .... 808.188904: kmem_cache_free: call_site=put_cred_rcu+0x88/0xa0 ptr=0000000058d74ef8 [002] .... 808.188913: kmem_cache_alloc: call_site=prepare_creds+0x26/0x100 ptr=0000000058d74ef8 bytes_req=168 bytes_alloc=576 gfp_flags=GFP_KERNEL [002] .... 808.188917: kmalloc: call_site=security_prepare_creds+0x77/0xa0 ptr=0000000062400820 bytes_req=8 bytes_alloc=336 gfp_flags=GFP_KERNEL|__GFP_ZERO [002] .... 808.188920: kmem_cache_alloc: call_site=getname_flags+0x4f/0x1e0 ptr=00000000cef40c80 bytes_req=4096 bytes_alloc=4480 gfp_flags=GFP_KERNEL [002] .... 808.188925: kmem_cache_free: call_site=putname+0x47/0x50 ptr=00000000cef40c80 [002] .... 808.188926: kfree: call_site=security_cred_free+0x42/0x50 ptr=0000000062400820 [002] .... 808.188931: kmem_cache_free: call_site=put_cred_rcu+0x88/0xa0 ptr=0000000058d74ef8 Link: http://lkml.kernel.org/r/20190914103215.23301-1-changbin.du@gmail.com Signed-off-by: Changbin Du <changbin.du@gmail.com> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2019-09-28tracing: Have error path in predicate_parse() free its allocated memoryNavid Emamdoost1-2/+4
In predicate_parse, there is an error path that is not going to out_free instead it returns directly which leads to a memory leak. Link: http://lkml.kernel.org/r/20190920225800.3870-1-navid.emamdoost@gmail.com Signed-off-by: Navid Emamdoost <navid.emamdoost@gmail.com> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2019-09-28tracing: Fix clang -Wint-in-bool-context warnings in IF_ASSIGN macroNathan Chancellor1-5/+5
After r372664 in clang, the IF_ASSIGN macro causes a couple hundred warnings along the lines of: kernel/trace/trace_output.c:1331:2: warning: converting the enum constant to a boolean [-Wint-in-bool-context] kernel/trace/trace.h:409:3: note: expanded from macro 'trace_assign_type' IF_ASSIGN(var, ent, struct ftrace_graph_ret_entry, ^ kernel/trace/trace.h:371:14: note: expanded from macro 'IF_ASSIGN' WARN_ON(id && (entry)->type != id); \ ^ 264 warnings generated. This warning can catch issues with constructs like: if (state == A || B) where the developer really meant: if (state == A || state == B) This is currently the only occurrence of the warning in the kernel tree across defconfig, allyesconfig, allmodconfig for arm32, arm64, and x86_64. Add the implicit '!= 0' to the WARN_ON statement to fix the warnings and find potential issues in the future. Link: https://github.com/llvm/llvm-project/commit/28b38c277a2941e9e891b2db30652cfd962f070b Link: https://github.com/ClangBuiltLinux/linux/issues/686 Link: http://lkml.kernel.org/r/20190926162258.466321-1-natechancellor@gmail.com Reviewed-by: Nick Desaulniers <ndesaulniers@google.com> Signed-off-by: Nathan Chancellor <natechancellor@gmail.com> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2019-09-28tracing/probe: Fix to check the difference of nr_args before adding probeMasami Hiramatsu1-0/+16
Steven reported that a test triggered: ================================================================== BUG: KASAN: slab-out-of-bounds in trace_kprobe_create+0xa9e/0xe40 Read of size 8 at addr ffff8880c4f25a48 by task ftracetest/4798 CPU: 2 PID: 4798 Comm: ftracetest Not tainted 5.3.0-rc6-test+ #30 Hardware name: Hewlett-Packard HP Compaq Pro 6300 SFF/339A, BIOS K01 v03.03 07/14/2016 Call Trace: dump_stack+0x7c/0xc0 ? trace_kprobe_create+0xa9e/0xe40 print_address_description+0x6c/0x332 ? trace_kprobe_create+0xa9e/0xe40 ? trace_kprobe_create+0xa9e/0xe40 __kasan_report.cold.6+0x1a/0x3b ? trace_kprobe_create+0xa9e/0xe40 kasan_report+0xe/0x12 trace_kprobe_create+0xa9e/0xe40 ? print_kprobe_event+0x280/0x280 ? match_held_lock+0x1b/0x240 ? find_held_lock+0xac/0xd0 ? fs_reclaim_release.part.112+0x5/0x20 ? lock_downgrade+0x350/0x350 ? kasan_unpoison_shadow+0x30/0x40 ? __kasan_kmalloc.constprop.6+0xc1/0xd0 ? trace_kprobe_create+0xe40/0xe40 ? trace_kprobe_create+0xe40/0xe40 create_or_delete_trace_kprobe+0x2e/0x60 trace_run_command+0xc3/0xe0 ? trace_panic_handler+0x20/0x20 ? kasan_unpoison_shadow+0x30/0x40 trace_parse_run_command+0xdc/0x163 vfs_write+0xe1/0x240 ksys_write+0xba/0x150 ? __ia32_sys_read+0x50/0x50 ? tracer_hardirqs_on+0x61/0x180 ? trace_hardirqs_off_caller+0x43/0x110 ? mark_held_locks+0x29/0xa0 ? do_syscall_64+0x14/0x260 do_syscall_64+0x68/0x260 Fix to check the difference of nr_args before adding probe on existing probes. This also may set the error log index bigger than the number of command parameters. In that case it sets the error position is next to the last parameter. Link: http://lkml.kernel.org/r/156966474783.3478.13217501608215769150.stgit@devnote2 Fixes: ca89bc071d5e ("tracing/kprobe: Add multi-probe per event support") Reported-by: Steven Rostedt (VMware) <rostedt@goodmis.org> Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2019-09-28mm, page_alloc: allow hugepage fallback to remote nodes when madvisedDavid Rientjes1-0/+11
For systems configured to always try hard to allocate transparent hugepages (thp defrag setting of "always") or for memory that has been explicitly madvised to MADV_HUGEPAGE, it is often better to fallback to remote memory to allocate the hugepage if the local allocation fails first. The point is to allow the initial call to __alloc_pages_node() to attempt to defragment local memory to make a hugepage available, if possible, rather than immediately fallback to remote memory. Local hugepages will always have a better access latency than remote (huge)pages, so an attempt to make a hugepage available locally is always preferred. If memory compaction cannot be successful locally, however, it is likely better to fallback to remote memory. This could take on two forms: either allow immediate fallback to remote memory or do per-zone watermark checks. It would be possible to fallback only when per-zone watermarks fail for order-0 memory, since that would require local reclaim for all subsequent faults so remote huge allocation is likely better than thrashing the local zone for large workloads. In this case, it is assumed that because the system is configured to try hard to allocate hugepages or the vma is advised to explicitly want to try hard for hugepages that remote allocation is better when local allocation and memory compaction have both failed. Signed-off-by: David Rientjes <rientjes@google.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Michal Hocko <mhocko@suse.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: Stefan Priebe - Profihost AG <s.priebe@profihost.ag> Cc: "Kirill A. Shutemov" <kirill@shutemov.name> Cc: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-09-28mm, page_alloc: avoid expensive reclaim when compaction may not succeedDavid Rientjes1-0/+22
Memory compaction has a couple significant drawbacks as the allocation order increases, specifically: - isolate_freepages() is responsible for finding free pages to use as migration targets and is implemented as a linear scan of memory starting at the end of a zone, - failing order-0 watermark checks in memory compaction does not account for how far below the watermarks the zone actually is: to enable migration, there must be *some* free memory available. Per the above, watermarks are not always suffficient if isolate_freepages() cannot find the free memory but it could require hundreds of MBs of reclaim to even reach this threshold (read: potentially very expensive reclaim with no indication compaction can be successful), and - if compaction at this order has failed recently so that it does not even run as a result of deferred compaction, looping through reclaim can often be pointless. For hugepage allocations, these are quite substantial drawbacks because these are very high order allocations (order-9 on x86) and falling back to doing reclaim can potentially be *very* expensive without any indication that compaction would even be successful. Reclaim itself is unlikely to free entire pageblocks and certainly no reliance should be put on it to do so in isolation (recall lumpy reclaim). This means we should avoid reclaim and simply fail hugepage allocation if compaction is deferred. It is also not helpful to thrash a zone by doing excessive reclaim if compaction may not be able to access that memory. If order-0 watermarks fail and the allocation order is sufficiently large, it is likely better to fail the allocation rather than thrashing the zone. Signed-off-by: David Rientjes <rientjes@google.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Michal Hocko <mhocko@suse.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: Stefan Priebe - Profihost AG <s.priebe@profihost.ag> Cc: "Kirill A. Shutemov" <kirill@shutemov.name> Cc: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-09-28Revert "Revert "Revert "mm, thp: consolidate THP gfp handling into ↵David Rientjes4-22/+51
alloc_hugepage_direct_gfpmask"" This reverts commit 92717d429b38e4f9f934eed7e605cc42858f1839. Since commit a8282608c88e ("Revert "mm, thp: restore node-local hugepage allocations"") is reverted in this series, it is better to restore the previous 5.2 behavior between the thp allocation and the page allocator rather than to attempt any consolidation or cleanup for a policy that is now reverted. It's less risky during an rc cycle and subsequent patches in this series further modify the same policy that the pre-5.3 behavior implements. Consolidation and cleanup can be done subsequent to a sane default page allocation strategy, so this patch reverts a cleanup done on a strategy that is now reverted and thus is the least risky option. Signed-off-by: David Rientjes <rientjes@google.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Michal Hocko <mhocko@suse.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: Stefan Priebe - Profihost AG <s.priebe@profihost.ag> Cc: "Kirill A. Shutemov" <kirill@shutemov.name> Cc: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-09-28Revert "Revert "mm, thp: restore node-local hugepage allocations""David Rientjes3-29/+17
This reverts commit a8282608c88e08b1782141026eab61204c1e533f. The commit references the original intended semantic for MADV_HUGEPAGE which has subsequently taken on three unique purposes: - enables or disables thp for a range of memory depending on the system's config (is thp "enabled" set to "always" or "madvise"), - determines the synchronous compaction behavior for thp allocations at fault (is thp "defrag" set to "always", "defer+madvise", or "madvise"), and - reverts a previous MADV_NOHUGEPAGE (there is no madvise mode to only clear previous hugepage advice). These are the three purposes that currently exist in 5.2 and over the past several years that userspace has been written around. Adding a NUMA locality preference adds a fourth dimension to an already conflated advice mode. Based on the semantic that MADV_HUGEPAGE has provided over the past several years, there exist workloads that use the tunable based on these principles: specifically that the allocation should attempt to defragment a local node before falling back. It is agreed that remote hugepages typically (but not always) have a better access latency than remote native pages, although on Naples this is at parity for intersocket. The revert commit that this patch reverts allows hugepage allocation to immediately allocate remotely when local memory is fragmented. This is contrary to the semantic of MADV_HUGEPAGE over the past several years: that is, memory compaction should be attempted locally before falling back. The performance degradation of remote hugepages over local hugepages on Rome, for example, is 53.5% increased access latency. For this reason, the goal is to revert back to the 5.2 and previous behavior that would attempt local defragmentation before falling back. With the patch that is reverted by this patch, we see performance degradations at the tail because the allocator happily allocates the remote hugepage rather than even attempting to make a local hugepage available. zone_reclaim_mode is not a solution to this problem since it does not only impact hugepage allocations but rather changes the memory allocation strategy for *all* page allocations. Signed-off-by: David Rientjes <rientjes@google.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Michal Hocko <mhocko@suse.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: Stefan Priebe - Profihost AG <s.priebe@profihost.ag> Cc: "Kirill A. Shutemov" <kirill@shutemov.name> Cc: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-09-28i2c: slave-eeprom: Add read only modeBjörn Ardö1-3/+11
Add read-only versions of all EEPROMs. These versions are read-only on the i2c side, but can be written from the sysfs side. Signed-off-by: Björn Ardö <bjorn.ardo@axis.com> Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
2019-09-28i2c: i801: Bring back Block Process Call support for certain platformsJarkko Nikula1-0/+1
Commit b84398d6d7f9 ("i2c: i801: Use iTCO version 6 in Cannon Lake PCH and beyond") looks like to drop by accident Block Write-Block Read Process Call support for Intel Sunrisepoint, Lewisburg, Denverton and Kaby Lake. That support was added for above and newer platforms by the commit 315cd67c9453 ("i2c: i801: Add Block Write-Block Read Process Call support") so bring it back for above platforms. Fixes: b84398d6d7f9 ("i2c: i801: Use iTCO version 6 in Cannon Lake PCH and beyond") Signed-off-by: Jarkko Nikula <jarkko.nikula@linux.intel.com> Reviewed-by: Alexander Sverdlin <alexander.sverdlin@nokia.com> Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
2019-09-28i2c: riic: Clear NACK in tend isrChris Brandt1-0/+1
The NACKF flag should be cleared in INTRIICNAKI interrupt processing as description in HW manual. This issue shows up quickly when PREEMPT_RT is applied and a device is probed that is not plugged in (like a touchscreen controller). The result is endless interrupts that halt system boot. Fixes: 310c18a41450 ("i2c: riic: add driver") Cc: stable@vger.kernel.org Reported-by: Chien Nguyen <chien.nguyen.eb@rvc.renesas.com> Signed-off-by: Chris Brandt <chris.brandt@renesas.com> Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
2019-09-28i2c: qcom-geni: Disable DMA processing on the Lenovo Yoga C630Lee Jones1-4/+8
We have a production-level laptop (Lenovo Yoga C630) which is exhibiting a rather horrific bug. When I2C HID devices are being scanned for at boot-time the QCom Geni based I2C (Serial Engine) attempts to use DMA. When it does, the laptop reboots and the user never sees the OS. Attempts are being made to debug the reason for the spontaneous reboot. No luck so far, hence the requirement for this hot-fix. This workaround will be removed once we have a viable fix. Signed-off-by: Lee Jones <lee.jones@linaro.org> Tested-by: Bjorn Andersson <bjorn.andersson@linaro.org> Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
2019-09-28iommu/amd: Lock code paths traversing protection_domain->dev_listJoerg Roedel1-1/+24
The traversing of this list requires protection_domain->lock to be taken to avoid nasty races with attach/detach code. Make sure the lock is held on all code-paths traversing this list. Reported-by: Filippo Sironi <sironi@amazon.de> Fixes: 92d420ec028d ("iommu/amd: Relax locking in dma_ops path") Reviewed-by: Filippo Sironi <sironi@amazon.de> Reviewed-by: Jerry Snitselaar <jsnitsel@redhat.com> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2019-09-28iommu/amd: Lock dev_data in attach/detach code pathsJoerg Roedel2-0/+12
Make sure that attaching a detaching a device can't race against each other and protect the iommu_dev_data with a spin_lock in these code paths. Fixes: 92d420ec028d ("iommu/amd: Relax locking in dma_ops path") Reviewed-by: Filippo Sironi <sironi@amazon.de> Reviewed-by: Jerry Snitselaar <jsnitsel@redhat.com> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2019-09-28iommu/amd: Check for busy devices earlier in attach_device()Joerg Roedel1-18/+7
Check early in attach_device whether the device is already attached to a domain. This also simplifies the code path so that __attach_device() can be removed. Fixes: 92d420ec028d ("iommu/amd: Relax locking in dma_ops path") Reviewed-by: Filippo Sironi <sironi@amazon.de> Reviewed-by: Jerry Snitselaar <jsnitsel@redhat.com> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2019-09-28iommu/amd: Take domain->lock for complete attach/detach pathJoerg Roedel1-39/+26
The code-paths before __attach_device() and __detach_device() are called also access and modify domain state, so take the domain lock there too. This allows to get rid of the __detach_device() function. Fixes: 92d420ec028d ("iommu/amd: Relax locking in dma_ops path") Reviewed-by: Filippo Sironi <sironi@amazon.de> Reviewed-by: Jerry Snitselaar <jsnitsel@redhat.com> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2019-09-28iommu/amd: Remove amd_iommu_devtable_lockJoerg Roedel1-17/+6
The lock is not necessary because the device table does not contain shared state that needs protection. Locking is only needed on an individual entry basis, and that needs to happen on the iommu_dev_data level. Fixes: 92d420ec028d ("iommu/amd: Relax locking in dma_ops path") Reviewed-by: Filippo Sironi <sironi@amazon.de> Reviewed-by: Jerry Snitselaar <jsnitsel@redhat.com> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2019-09-28iommu/amd: Remove domain->updatedJoerg Roedel2-25/+25
This struct member was used to track whether a domain change requires updates to the device-table and IOMMU cache flushes. The problem is, that access to this field is racy since locking in the common mapping code-paths has been eliminated. Move the updated field to the stack to get rid of all potential races and remove the field from the struct. Fixes: 92d420ec028d ("iommu/amd: Relax locking in dma_ops path") Reviewed-by: Filippo Sironi <sironi@amazon.de> Reviewed-by: Jerry Snitselaar <jsnitsel@redhat.com> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2019-09-27net: tap: clean up an indentation issueColin Ian King1-1/+1
There is a statement that is indented too deeply, remove the extraneous tab. Signed-off-by: Colin Ian King <colin.king@canonical.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-09-27nfp: abm: fix memory leak in nfp_abm_u32_knode_replaceNavid Emamdoost1-4/+10
In nfp_abm_u32_knode_replace if the allocation for match fails it should go to the error handling instead of returning. Updated other gotos to have correct errno returned, too. Signed-off-by: Navid Emamdoost <navid.emamdoost@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-09-27mmc: host: sdhci-pci: Add Genesys Logic GL975x supportBen Chuang5-1/+361
Add support for the GL9750 and GL9755 chipsets. Enable v4 mode and wait 5ms after set 1.8V signal enable for GL9750/ GL9755. Fix the value of SDHCI_MAX_CURRENT register and use the vendor tuning flow for GL9750. Co-developed-by: Michael K Johnson <johnsonm@danlj.org> Signed-off-by: Michael K Johnson <johnsonm@danlj.org> Signed-off-by: Ben Chuang <ben.chuang@genesyslogic.com.tw> Acked-by: Adrian Hunter <adrian.hunter@intel.com> Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
2019-09-27tcp: better handle TCP_USER_TIMEOUT in SYN_SENT stateEric Dumazet1-2/+3
Yuchung Cheng and Marek Majkowski independently reported a weird behavior of TCP_USER_TIMEOUT option when used at connect() time. When the TCP_USER_TIMEOUT is reached, tcp_write_timeout() believes the flow should live, and the following condition in tcp_clamp_rto_to_user_timeout() programs one jiffie timers : remaining = icsk->icsk_user_timeout - elapsed; if (remaining <= 0) return 1; /* user timeout has passed; fire ASAP */ This silly situation ends when the max syn rtx count is reached. This patch makes sure we honor both TCP_SYNCNT and TCP_USER_TIMEOUT, avoiding these spurious SYN packets. Fixes: b701a99e431d ("tcp: Add tcp_clamp_rto_to_user_timeout() helper to improve accuracy") Signed-off-by: Eric Dumazet <edumazet@google.com> Reported-by: Yuchung Cheng <ycheng@google.com> Reported-by: Marek Majkowski <marek@cloudflare.com> Cc: Jon Maxwell <jmaxwell37@gmail.com> Link: https://marc.info/?l=linux-netdev&m=156940118307949&w=2 Acked-by: Jon Maxwell <jmaxwell37@gmail.com> Tested-by: Marek Majkowski <marek@cloudflare.com> Signed-off-by: Marek Majkowski <marek@cloudflare.com> Acked-by: Yuchung Cheng <ycheng@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-09-27sk_buff: drop all skb extensions on free and skb scrubbingFlorian Westphal3-3/+12
Now that we have a 3rd extension, add a new helper that drops the extension space and use it when we need to scrub an sk_buff. At this time, scrubbing clears secpath and bridge netfilter data, but retains the tc skb extension, after this patch all three get cleared. NAPI reuse/free assumes we can only have a secpath attached to skb, but it seems better to clear all extensions there as well. v2: add unlikely hint (Eric Dumazet) Fixes: 95a7233c452a ("net: openvswitch: Set OvS recirc_id from tc chain index") Signed-off-by: Florian Westphal <fw@strlen.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-09-27tcp_bbr: fix quantization code to not raise cwnd if not probing bandwidthKevin(Yudong) Yang1-4/+4
There was a bug in the previous logic that attempted to ensure gain cycling gets inflight above BDP even for small BDPs. This code correctly raised and lowered target inflight values during the gain cycle. And this code correctly ensured that cwnd was raised when probing bandwidth. However, it did not correspondingly ensure that cwnd was *not* raised in this way when *not* probing for bandwidth. The result was that small-BDP flows that were always cwnd-bound could go for many cycles with a fixed cwnd, and not probe or yield bandwidth at all. This meant that multiple small-BDP flows could fail to converge in their bandwidth allocations. Fixes: 3c346b233c68 ("tcp_bbr: fix bw probing to raise in-flight data for very small BDPs") Signed-off-by: Kevin(Yudong) Yang <yyd@google.com> Acked-by: Neal Cardwell <ncardwell@google.com> Acked-by: Yuchung Cheng <ycheng@google.com> Acked-by: Soheil Hassas Yeganeh <soheil@google.com> Acked-by: Priyaranjan Jha <priyarjha@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-09-27mlxsw: spectrum_flower: Fail in case user specifies multiple mirror actionsDanielle Ratson1-0/+6
The ASIC can only mirror a packet to one port, but when user is trying to set more than one mirror action, it doesn't fail. Add a check if more than one mirror action was specified per rule and if so, fail for not being supported. Fixes: d0d13c1858a11 ("mlxsw: spectrum_acl: Add support for mirror action") Signed-off-by: Danielle Ratson <danieller@mellanox.com> Acked-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: Ido Schimmel <idosch@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-09-27Documentation: Clarify trap's descriptionIdo Schimmel1-1/+2
Alex noted that the below description might not be obvious to all users. Clarify it by adding an example. Fixes: f3047ca01f12 ("Documentation: Add devlink-trap documentation") Reported-by: Alex Kushnarov <alexanderk@mellanox.com> Reviewed-by: Alex Kushnarov <alexanderk@mellanox.com> Acked-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: Ido Schimmel <idosch@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-09-27mlxsw: spectrum: Clear VLAN filters during port initializationIdo Schimmel2-7/+9
When a port is created, its VLAN filters are not cleared by the firmware. This causes tagged packets to be later dropped by the ingress STP filters, which default to DISCARD state. The above did not matter much until commit b5ce611fd96e ("mlxsw: spectrum: Add devlink-trap support") where we exposed the drop reason to users. Without this patch, the drop reason users will see is not consistent. If a port is enslaved to a VLAN-aware bridge and a packet with an invalid VLAN tries to ingress the bridge, it will be dropped due to ingress STP filter. If the VLAN is later enabled and then disabled, the packet will be dropped by the ingress VLAN filter despite the above being a seemingly NOP operation. Fix this by clearing all the VLAN filters during port initialization. Adjust the test accordingly. Fixes: b5ce611fd96e ("mlxsw: spectrum: Add devlink-trap support") Reported-by: Alex Kushnarov <alexanderk@mellanox.com> Tested-by: Alex Kushnarov <alexanderk@mellanox.com> Acked-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: Ido Schimmel <idosch@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-09-27net: ena: clean up indentation issueColin Ian King1-2/+2
There memset is indented incorrectly, remove the extraneous tabs. Signed-off-by: Colin Ian King <colin.king@canonical.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-09-27NFC: st95hf: clean up indentation issueColin Ian King1-1/+1
The return statement is indented incorrectly, add in a missing tab and remove an extraneous space after the return Signed-off-by: Colin Ian King <colin.king@canonical.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-09-27mmc: tegra: Implement ->set_dma_mask()Nicolin Chen1-20/+28
The SDHCI controller on Tegra186 supports 40-bit addressing, which is usually enough to address all of system memory. However, if the SDHCI controller is behind an IOMMU, the address space can go beyond. This happens on Tegra186 and later where the ARM SMMU has an input address space of 48 bits. If the DMA API is backed by this ARM SMMU, the top- down IOVA allocator will cause IOV addresses to be returned that the SDHCI controller cannot access. Unfortunately, prior to the introduction of the ->set_dma_mask() host operation, the SDHCI core would set either a 64-bit DMA mask if the controller claimed to support 64-bit addressing, or a 32-bit DMA mask otherwise. Since the full 64 bits cannot be addressed on Tegra, this had to be worked around in commit 68481a7e1c84 ("mmc: tegra: Mark 64 bit dma broken on Tegra186") by setting the SDHCI_QUIRK2_BROKEN_64_BIT_DMA quirk, which effectively restricts the DMA mask to 32 bits. One disadvantage of this is that dma_map_*() APIs will now try to use the swiotlb to bounce DMA to addresses beyond of the controller's DMA mask. This in turn caused degraded performance and can lead to situations where the swiotlb buffer is exhausted, which in turn leads to DMA transfers to fail. With the recent introduction of the ->set_dma_mask() host operation, this can now be properly fixed. For each generation of Tegra, the exact supported DMA mask can be configured. This kills two birds with one stone: it avoids the use of bounce buffers because system memory never exceeds the addressable memory range of the SDHCI controllers on these devices, and at the same time when an IOMMU is involved, it prevents IOV addresses from being allocated beyond the addressible range of the controllers. Since the DMA mask is now properly handled, the 64-bit DMA quirk can be removed. Signed-off-by: Nicolin Chen <nicoleotsuka@gmail.com> [treding@nvidia.com: provide more background in commit message] Tested-by: Nicolin Chen <nicoleotsuka@gmail.com> Acked-by: Adrian Hunter <adrian.hunter@intel.com> Signed-off-by: Thierry Reding <treding@nvidia.com> Cc: stable@vger.kernel.org # v4.15 + Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
generated by cgit v1.2.3 (git 2.39.1) at 2026-01-07 10:44:42 +0000