- 24 Mar, 2022 39 commits
-
-
Aleksandr Nogikh authored
Allocate the kcov buffer during KCOV_MODE_INIT in order to untie mmapping of a kcov instance and the actual coverage collection process. Modify kcov_mmap, so that it can be reliably used any number of times once KCOV_MODE_INIT has succeeded. These changes to the user-facing interface of the tool only weaken the preconditions, so all existing user space code should remain compatible with the new version. Link: https://lkml.kernel.org/r/20220117153634.150357-3-nogikh@google.comSigned-off-by:
Aleksandr Nogikh <nogikh@google.com> Reviewed-by:
Dmitry Vyukov <dvyukov@google.com> Reviewed-by:
Andrey Konovalov <andreyknvl@gmail.com> Cc: Alexander Potapenko <glider@google.com> Cc: Marco Elver <elver@google.com> Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Cc: Taras Madan <tarasmadan@google.com> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
Aleksandr Nogikh authored
Patch series "kcov: improve mmap processing", v3. Subsequent mmaps of the same kcov descriptor currently do not update the virtual memory of the task and yet return 0 (success). This is counter-intuitive and may lead to unexpected memory access errors. Also, this unnecessarily limits the functionality of kcov to only the simplest usage scenarios. Kcov instances are effectively forever attached to their first address spaces and it becomes impossible to e.g. reuse the same kcov handle in forked child processes without mmapping the memory first. This is exactly what we tried to do in syzkaller and inadvertently came upon this behavior. This patch series addresses the problem described above. This patch (of 3): Currently all ioctls are de facto processed under a spinlock in order to serialise them. This, however, prohibits the use of vmalloc and other memory management functions in the implementations of those ioctls, unnecessary complicating any further changes to the code. Let all ioctls first be processed inside the kcov_ioctl() function which should execute the ones that are not compatible with spinlock and then pass control to kcov_ioctl_locked() for all other ones. KCOV_REMOTE_ENABLE is processed both in kcov_ioctl() and kcov_ioctl_locked() as the steps are easily separable. Although it is still compatible with a spinlock, move KCOV_INIT_TRACE handling to kcov_ioctl(), so that the changes from the next commit are easier to follow. Link: https://lkml.kernel.org/r/20220117153634.150357-1-nogikh@google.com Link: https://lkml.kernel.org/r/20220117153634.150357-2-nogikh@google.comSigned-off-by:
-
Guilherme G. Piccoli authored
The panic_print setting allows users to collect more information in a panic event, like memory stats, tasks, CPUs backtraces, etc. This is an interesting debug mechanism, but currently the print event happens *after* kmsg_dump(), meaning that pstore, for example, cannot collect a dmesg with the panic_print extra information. This patch changes that in 2 steps: (a) The panic_print setting allows to replay the existing kernel log buffer to the console (bit 5), besides the extra information dump. This functionality makes sense only at the end of the panic() function. So, we hereby allow to distinguish the two situations by a new boolean parameter in the function panic_print_sys_info(). (b) With the above change, we can safely call panic_print_sys_info() before kmsg_dump(), allowing to dump the extra information when using pstore or other kmsg dumpers. The additional messages from panic_print could overwrite the oldest messages when the buffer is full. The only reasonable solution is to use a large enough log buffer, hence we added an advice into the kernel parameters documentation about that. Link: https://lkml.kernel.org/r/20220214141308.841525-1-gpiccoli@igalia.comSigned-off-by:Guilherme G. Piccoli <gpiccoli@igalia.com> Acked-by:
Baoquan He <bhe@redhat.com> Reviewed-by:
Petr Mladek <pmladek@suse.com> Reviewed-by:
Sergey Senozhatsky <senozhatsky@chromium.org> Cc: Feng Tang <feng.tang@intel.com> Signed-off-by:
-
Guilherme G. Piccoli authored
Currently the "panic_print" parameter/sysctl allows some interesting debug information to be printed during a panic event. This is useful for example in cases the user cannot kdump due to resource limits, or if the user collects panic logs in a serial output (or pstore) and prefers a fast reboot instead of a kdump. Happens that currently there's no way to see all CPUs backtraces in a panic using "panic_print" on architectures that support that. We do have "oops_all_cpu_backtrace" sysctl, but although partially overlapping in the functionality, they are orthogonal in nature: "panic_print" is a panic tuning (and we have panics without oopses, like direct calls to panic() or maybe other paths that don't go through oops_enter() function), and the original purpose of "oops_all_cpu_backtrace" is to provide more information on oopses for cases in which the users desire to continue running the kernel even after an oops, i.e., used in non-panic scenarios. So, we hereby introduce an additional bit for "panic_print" to allow dumping the CPUs backtraces during a panic event. Link: https://lkml.kernel.org/r/20211109202848.610874-3-gpiccoli@igalia.comSigned-off-by:
Feng Tang <feng.tang@intel.com> Cc: Iurii Zaikin <yzaikin@google.com> Cc: Kees Cook <keescook@chromium.org> Cc: Luis Chamberlain <mcgrof@kernel.org> Cc: Samuel Iglesias Gonsalvez <siglesias@igalia.com> Signed-off-by:
-
Guilherme G. Piccoli authored
Patch series "Some improvements on panic_print". This is a mix of a documentation fix with some additions to the "panic_print" syscall / parameter. The goal here is being able to collect all CPUs backtraces during a panic event and also to enable "panic_print" in a kdump event - details of the reasoning and design choices in the patches. This patch (of 3): Commit de6da1e8 ("panic: add an option to replay all the printk message in buffer") added a new bit to the sysctl/kernel parameter "panic_print", but the documentation was added only in kernel-parameters.txt, not in the sysctl guide. Fix it here by adding bit 5 to sysctl admin-guide documentation. [rdunlap@infradead.org: fix table format warning] Link: https://lkml.kernel.org/r/20220109055635.6999-1-rdunlap@infradead.org Link: https://lkml.kernel.org/r/20211109202848.610874-1-gpiccoli@igalia.com Link: https://lkml.kernel.org/r/20211109202848.610874-2-gpiccoli@igalia.com Fixes: de6da1e8 ("panic: add an option to replay all the printk message in buffer") Signed-off-by:
-
Lukas Bulwahn authored
make clang-analyzer on x86_64 defconfig caught my attention with: kernel/taskstats.c:120:2: warning: Value stored to 'rc' is never read \ [clang-analyzer-deadcode.DeadStores] rc = 0; ^ Commit d94a0415 ("taskstats: free skb, avoid returns in send_cpu_listeners") made send_cpu_listeners() not return a value and hence, the rc variable remained only to be used within the loop where it is always assigned before read and it does not need any other initialisation. So, simply remove this unneeded dead initializing assignment. As compilers will detect this unneeded assignment and optimize this anyway, the resulting object code is identical before and after this change. No functional change. No change to object code. [akpm@linux-foundation.org: reduce scope of `rc'] Link: https://lkml.kernel.org/r/20220307093942.21310-1-lukas.bulwahn@gmail.comSigned-off-by:Lukas Bulwahn <lukas.bulwahn@gmail.com> Reviewed-by:
Nick Desaulniers <ndesaulniers@google.com> Cc: Balbir Singh <bsingharora@gmail.com> Cc: Tom Rix <trix@redhat.com> Cc: Nathan Chancellor <natechancellor@gmail.com> Signed-off-by:
-
Tiezhu Yang authored
panic_on_warn is unset inside panic(), so no need to unset it before calling panic() in end_report(). Link: https://lkml.kernel.org/r/1644324666-15947-6-git-send-email-yangtiezhu@loongson.cnSigned-off-by:
Tiezhu Yang <yangtiezhu@loongson.cn> Reviewed-by:
Marco Elver <elver@google.com> Cc: Andrey Ryabinin <ryabinin.a.a@gmail.com> Cc: Baoquan He <bhe@redhat.com> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Xuefeng Li <lixuefeng@loongson.cn> Signed-off-by:
-
Tiezhu Yang authored
panic_on_warn is unset inside panic(), so no need to unset it before calling panic() in ubsan_epilogue(). Link: https://lkml.kernel.org/r/1644324666-15947-5-git-send-email-yangtiezhu@loongson.cnSigned-off-by:
-
Tiezhu Yang authored
In the current code, the following three places need to unset panic_on_warn before calling panic() to avoid recursive panics: kernel/kcsan/report.c: print_report() kernel/sched/core.c: __schedule_bug() mm/kfence/report.c: kfence_report_error() In order to avoid copy-pasting "panic_on_warn = 0" all over the places, it is better to move it inside panic() and then remove it from the other places. Link: https://lkml.kernel.org/r/1644324666-15947-4-git-send-email-yangtiezhu@loongson.cnSigned-off-by:
-
Tiezhu Yang authored
Except cp and makedumpfile, add scp example to write out the dump file. Link: https://lkml.kernel.org/r/1644324666-15947-3-git-send-email-yangtiezhu@loongson.cnSigned-off-by:
-
Tiezhu Yang authored
Patch series "Update doc and fix some issues about kdump", v2. This patch (of 5): After commit 6a108a14 ("kconfig: rename CONFIG_EMBEDDED to CONFIG_EXPERT"), "Configure standard kernel features (for small systems)" is not exist, we should use "Configure standard kernel features (expert users)" now. Link: https://lkml.kernel.org/r/1644324666-15947-1-git-send-email-yangtiezhu@loongson.cn Link: https://lkml.kernel.org/r/1644324666-15947-2-git-send-email-yangtiezhu@loongson.cnSigned-off-by:
-
Jisheng Zhang authored
Replace the conditional compilation using "#ifdef CONFIG_KEXEC_CORE" by a check for "IS_ENABLED(CONFIG_KEXEC_CORE)", to simplify the code and increase compile coverage. Link: https://lkml.kernel.org/r/20211206160514.2000-5-jszhang@kernel.orgSigned-off-by:
Jisheng Zhang <jszhang@kernel.org> Acked-by:
Catalin Marinas <catalin.marinas@arm.com> Acked-by:
-
Jisheng Zhang authored
Replace the conditional compilation using "#ifdef CONFIG_KEXEC_CORE" by a check for "IS_ENABLED(CONFIG_KEXEC_CORE)", to simplify the code and increase compile coverage. Link: https://lkml.kernel.org/r/20211206160514.2000-4-jszhang@kernel.orgSigned-off-by:
-
Jisheng Zhang authored
Replace the conditional compilation using "#ifdef CONFIG_KEXEC_CORE" by a check for "IS_ENABLED(CONFIG_KEXEC_CORE)", to simplify the code and increase compile coverage. Link: https://lkml.kernel.org/r/20211206160514.2000-3-jszhang@kernel.orgSigned-off-by:
Palmer Dabbelt <palmer@rivosinc.com> Acked-by:
-
Jisheng Zhang authored
Patch series "kexec: use IS_ENABLED(CONFIG_KEXEC_CORE) instead of #ifdef", v2. Replace the conditional compilation using "#ifdef CONFIG_KEXEC_CORE" by a check for "IS_ENABLED(CONFIG_KEXEC_CORE)", to simplify the code and increase compile coverage. I only modified x86, arm, arm64 and riscv, other architectures such as sh, powerpc and s390 are better to be kept kexec code as-is so they are not touched. This patch (of 5): Make the forward declarations of crashk_res, crashk_low_res and crash_notes always visible. Code referring to these symbols can then just check for IS_ENABLED(CONFIG_KEXEC_CORE), instead of requiring conditional compilation using an #ifdef, thus preparing to increase compile coverage and simplify the code. Link: https://lkml.kernel.org/r/20211206160514.2000-1-jszhang@kernel.org Link: https://lkml.kernel.org/r/20211206160514.2000-2-jszhang@kernel.orgSigned-off-by:
-
Sebastian Andrzej Siewior authored
All callers of cgroup_rstat_flush_locked() acquire cgroup_rstat_lock either with spin_lock_irq() or spin_lock_irqsave(). cgroup_rstat_flush_locked() itself acquires cgroup_rstat_cpu_lock which is a raw_spin_lock. This lock is also acquired in cgroup_rstat_updated() in IRQ context and therefore requires _irqsave() locking suffix in cgroup_rstat_flush_locked(). Since there is no difference between spin_lock_t and raw_spin_lock_t on !RT lockdep does not complain here. On RT lockdep complains because the interrupts were not disabled here and a deadlock is possible. Acquire the raw_spin_lock_t with disabled interrupts. Link: https://lkml.kernel.org/r/20220301122143.1521823-2-bigeasy@linutronix.deSigned-off-by:
Sebastian Andrzej Siewior <bigeasy@linutronix.de> Acked-by:
Tejun Heo <tj@kernel.org> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Zefan Li <lizefan.x@bytedance.com> From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Subject: cgroup: add a comment to cgroup_rstat_flush_locked(). Add a comment why spin_lock_irq() -> raw_spin_lock_irqsave() is needed. Link: https://lkml.kernel.org/r/Yh+DOK73hfVV5ThX@linutronix.deSigned-off-by:
-
Helge Deller authored
The put_user(val,ptr) macro wants a pointer to a simple type, but in fat_ioctl_filldir() the d_name field references an "array of chars". Be more accurate and explicitly give the pointer to the first character of the d_name[] array. I noticed that issue while trying to optimize the parisc put_user() macro and used an intermediate variable to store the pointer. In that case I got this error: In file included from include/linux/uaccess.h:11, from include/linux/compat.h:17, from fs/fat/dir.c:18: fs/fat/dir.c: In function `fat_ioctl_filldir': fs/fat/dir.c:725:33: error: invalid initializer 725 | if (put_user(0, d2->d_name) || \ | ^~ include/asm/uaccess.h:152:33: note: in definition of macro `__put_user' 152 | __typeof__(ptr) __ptr = ptr; \ | ^~~ fs/fat/dir.c:759:1: note: in expansion of macro `FAT_IOCTL_FILLDIR_FUNC' 759 | FAT_IOCTL_FILLDIR_FUNC(fat_ioctl_filldir, __fat_dirent) Andreas Schwab <schwab@linux-m68k.org> suggested to use __typeof__(&*(ptr)) __ptr = ptr; instead. This works, but nevertheless it's probably reasonable to fix the original caller too. Link: https://lkml.kernel.org/r/Ygo+A9MREmC1H3kr@p100Signed-off-by:Helge Deller <deller@gmx.de> Acked-by:
OGAWA Hirofumi <hirofumi@mail.parknet.co.jp> Cc: David Laight <David.Laight@aculab.com> Cc: Andreas Schwab <schwab@linux-m68k.org> Signed-off-by:
-
Qinghua Jin authored
Testcase: 1. create a minix file system and mount it 2. open a file on the file system with O_RDWR|O_CREAT|O_TRUNC|O_DIRECT 3. open fails with -EINVAL but leaves an empty file behind. All other open() failures don't leave the failed open files behind. It is hard to check the direct_IO op before creating the inode. Just as ext4 and btrfs do, this patch will resolve the issue by allowing to create the file with O_DIRECT but returning error when writing the file. Link: https://lkml.kernel.org/r/20220107133626.413379-1-qhjin.dev@gmail.comSigned-off-by:
Qinghua Jin <qhjin.dev@gmail.com> Reported-by:
Colin Ian King <colin.king@intel.com> Reviewed-by:
Jan Kara <jack@suse.cz> Acked-by:
Christian Brauner <christian.brauner@ubuntu.com> Signed-off-by:
-
Andrei Vagin authored
head, tail, ring_size are declared as unsigned int, so all local variables that operate with these fields have to be unsigned to avoid signed integer overflow. Right now, it isn't an issue because the maximum pipe size is limited by 1U<<31. Link: https://lkml.kernel.org/r/20220106171946.36128-1-avagin@gmail.comSigned-off-by:
Andrei Vagin <avagin@gmail.com> Suggested-by:
Dmitry Safonov <0x7f454c46@gmail.com> Acked-by:
-
Andrei Vagin authored
Right now, kcalloc is used to allocate a pipe_buffer array. The size of the pipe_buffer struct is 40 bytes. kcalloc allows allocating reliably chunks with sizes less or equal to PAGE_ALLOC_COSTLY_ORDER (3). It means that the maximum pipe size is 3.2MB in this case. In CRIU, we use pipes to dump processes memory. CRIU freezes a target process, injects a parasite code into it and then this code splices memory into pipes. If a maximum pipe size is small, we need to do many iterations or create many pipes. kvcalloc attempt to allocate physically contiguous memory, but upon failure, fall back to non-contiguous (vmalloc) allocation and so it isn't limited by PAGE_ALLOC_COSTLY_ORDER. The maximum pipe size for non-root users is limited by the /proc/sys/fs/pipe-max-size sysctl that is 1MB by default, so only the root user will be able to trigger vmalloc allocations. Link: https://lkml.kernel.org/r/20220104171058.22580-1-avagin@gmail.comSigned-off-by:
-
Randy Dunlap authored
initcall_blacklist() should return 1 to indicate that it handled its cmdline arguments. set_debug_rodata() should return 1 to indicate that it handled its cmdline arguments. Print a warning if the option string is invalid. This prevents these strings from being added to the 'init' program's environment as they are not init arguments/parameters. Link: https://lkml.kernel.org/r/20220221050901.23985-1-rdunlap@infradead.orgSigned-off-by:
Randy Dunlap <rdunlap@infradead.org> Reported-by:
Igor Zhbanov <i.zhbanov@omprussia.ru> Cc: Ingo Molnar <mingo@kernel.org> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by:
-
Randy Dunlap authored
Igor noted in [1] that there are quite a few __setup() handling functions that return incorrect values. Doing this can be harmless, but it can also cause strings to be added to init's argument or environment list, polluting them. Since __setup() handling and return values are not documented, first add documentation for that. Also add more documentation for early_param() handling and return values. For __setup() functions, returning 0 (not handled) has questionable value if it is just a malformed option value, as in rodata=junk since returning 0 would just cause "rodata=junk" to be added to init's environment unnecessarily: Run /sbin/init as init process with arguments: /sbin/init with environment: HOME=/ TERM=linux splash=native rodata=junk Also, there are no recommendations on whether to print a warning when an unknown parameter value is seen. I am not addressing that here. [1] lore.kernel.org/r/64644a2f-4a20-bab3-1e15-3b2cdd0defe3@omprussia.ru Link: https://lkml.kernel.org/r/20220221050852.1147-1-rdunlap@infradead.orgSigned-off-by: -
Mark-PK Tsai authored
Use ktime_us_delta() to make the initcall_debug log more precise than right shifting the result of ktime_to_ns() by 10 bits. Link: https://lkml.kernel.org/r/20220209053350.15771-1-mark-pk.tsai@mediatek.comSigned-off-by:
Mark-PK Tsai <mark-pk.tsai@mediatek.com> Reviewed-by:
Andrew Halaney <ahalaney@redhat.com> Tested-by:
-
Sagar Patel authored
Commit 0ee3e7b8 ("checkpatch: get default codespell dictionary path from package location") introduced the ability to search for the codespell dictionary rather than hardcoding its path. codespell requires Python 3.6 or above, but on some systems, the python executable is a Python 2.7 interpreter. In this case, searching for the dictionary fails, subsequently making codespell fail: No codespell typos will be found - file '/usr/share/codespell/dictionary.txt': No such file or directory So, use python3 to remove ambiguity. In addition, when searching for dictionary.txt, do not check if the codespell executable exists since, - checkpatch.pl only uses dictionary.txt, not the codespell executable. - codespell can be installed via a Python package manager, in which case the codespell executable may not be present in a typical $PATH, but a dictionary does exist. Link: https://lkml.kernel.org/r/20220309180048.147672-1-sagarmp@cs.unc.eduSigned-off-by:
Sagar Patel <sagarmp@cs.unc.edu> Reviewed-by:
Peter Ujfalusi <peter.ujfalusi@linux.intel.com> Cc: Joe Perches <joe@perches.com> Signed-off-by:
-
Joe Perches authored
Add early_param as another exception to the blank line preferred after function/struct/union declaration or definition test. Link: https://lkml.kernel.org/r/3bd6ada59f411a7685d7e64eeb670540d4bfdcde.camel@perches.comSigned-off-by:
Joe Perches <joe@perches.com> Cc: Dwaipayan Ray <dwaipayanray1@gmail.com> Cc: Lukas Bulwahn <lukas.bulwahn@gmail.com> Signed-off-by:
-
Joe Perches authored
Single line code like: if (foo) bar; should generally be written: if (foo) bar; Add a --fix test to do so. This fix is not done when an ASSIGN_IN_IF in the same line exists. Link: https://lkml.kernel.org/r/20220128185924.80137-2-joe@perches.comSigned-off-by:
-
Joe Perches authored
There is no effective difference. Given the large number of uses of "GPL v2", emit this message only for patches as a trivial treeside sed could be done one day. Ref: commit bf7fbeea ("module: Cure the MODULE_LICENSE "GPL" vs. "GPL v2" bogosity") Link: https://lkml.kernel.org/r/20220128185924.80137-1-joe@perches.comSigned-off-by:
-
Randy Dunlap authored
Fix kernel-doc warings in lib/bitmap.c: lib/bitmap.c:498: warning: Function parameter or member 'buf' not described in 'bitmap_print_to_buf' lib/bitmap.c:498: warning: Function parameter or member 'maskp' not described in 'bitmap_print_to_buf' lib/bitmap.c:498: warning: Function parameter or member 'nmaskbits' not described in 'bitmap_print_to_buf' lib/bitmap.c:498: warning: Function parameter or member 'off' not described in 'bitmap_print_to_buf' lib/bitmap.c:498: warning: Function parameter or member 'count' not described in 'bitmap_print_to_buf' lib/bitmap.c:561: warning: contents before sections lib/bitmap.c:606: warning: Function parameter or member 'buf' not described in 'bitmap_print_list_to_buf' lib/bitmap.c:606: warning: Function parameter or member 'maskp' not described in 'bitmap_print_list_to_buf' lib/bitmap.c:606: warning: Function parameter or member 'nmaskbits' not described in 'bitmap_print_list_to_buf' lib/bitmap.c:606: warning: Function parameter or member 'off' not described in 'bitmap_print_list_to_buf' lib/bitmap.c:606: warning: Function parameter or member 'count' not described in 'bitmap_print_list_to_buf' lib/bitmap.c:819: warning: missing initial short description on line: * bitmap_parselist_user() This still leaves 15 warnings for function return values not described, similar to this one: bitmap.c:890: warning: No description found for return value of 'bitmap_parse' Link: https://lkml.kernel.org/r/20220306065823.5153-1-rdunlap@infradead.org Fixes: 1fae5629 ("cpumask: introduce cpumap_print_list/bitmask_to_buf to support large bitmask and list") Fixes: 4b060420 ("bitmap, irq: add smp_affinity_list interface to /proc/irq") Signed-off-by:
-
Feng Tang authored
0Day robots reported there is compiling issue for 'csky' ARCH when CONFIG_DEBUG_FORCE_DATA_SECTION_ALIGNED is enabled [1]: All errors (new ones prefixed by >>): {standard input}: Assembler messages: >> {standard input}:2277: Error: pcrel offset for branch to .LS000B too far (0x3c) Which was discussed in [2]. And as there is no solution for csky yet, add some dependency for this config to limit it to several ARCHs which have no compiling issue so far. [1]. https://lore.kernel.org/lkml/202202271612.W32UJAj2-lkp@intel.com/ [2]. https://www.spinics.net/lists/linux-kbuild/msg30298.html Link: https://lkml.kernel.org/r/20220304021100.GN4548@shbuild999.sh.intel.comReported-by:kernel test robot <lkp@intel.com> Signed-off-by:
-
Andy Shevchenko authored
It's not obvious that bitfield.h doesn't guarantee the bits.h inclusion and the example in the former is confusing. Some developers think that it's okay to just include bitfield.h to get it working. Change example to explicitly include necessary headers in order to avoid confusion. Link: https://lkml.kernel.org/r/20220207123341.47533-1-andriy.shevchenko@linux.intel.com Fixes: 3e9b3112 ("add basic register-field manipulation macros") Depends-on: 8bd9cb51 ("locking/atomics, asm-generic: Move some macros from <linux/bitops.h> to a new <linux/bits.h> file") Signed-off-by:
Andy Shevchenko <andriy.shevchenko@linux.intel.com> Reported-by:
Jan Dąbroś <jsd@semihalf.com> Cc: Peter Zijlstra <peterz@infradead.org> Signed-off-by:
-
Christophe Leroy authored
Building a kernel with CONFIG_CC_OPTIMISE_FOR_SIZE leads to __ilog2_u32() being duplicated 50 times and __ilog2_u64() 3 times in vmlinux on a tiny powerpc32 config. __ilog2_u32() being 2 instructions it is not worth being kept out of line, so force inlining. Allthough the u64 version is a bit bigger, there is still a small benefit in keeping it inlined. On a 64 bits config there's a real benefit. With this change the size of vmlinux text is reduced by 1 kbytes, which is approx 50% more than the size of the removed functions. Before the patch there is for instance: c00d2a94 <__ilog2_u32>: c00d2a94: 7c 63 00 34 cntlzw r3,r3 c00d2a98: 20 63 00 1f subfic r3,r3,31 c00d2a9c: 4e 80 00 20 blr c00d36d8 <__order_base_2>: c00d36d8: 28 03 00 01 cmplwi r3,1 c00d36dc: 40 81 00 2c ble c00d3708 <__order_base_2+0x30> c00d36e0: 94 21 ff f0 stwu r1,-16(r1) c00d36e4: 7c 08 02 a6 mflr r0 c00d36e8: 38 63 ff ff addi r3,r3,-1 c00d36ec: 90 01 00 14 stw r0,20(r1) c00d36f0: 4b ff f3 a5 bl c00d2a94 <__ilog2_u32> c00d36f4: 80 01 00 14 lwz r0,20(r1) c00d36f8: 38 63 00 01 addi r3,r3,1 c00d36fc: 7c 08 03 a6 mtlr r0 c00d3700: 38 21 00 10 addi r1,r1,16 c00d3704: 4e 80 00 20 blr c00d3708: 38 60 00 00 li r3,0 c00d370c: 4e 80 00 20 blr With the patch it has become: c00d356c <__order_base_2>: c00d356c: 28 03 00 01 cmplwi r3,1 c00d3570: 40 81 00 14 ble c00d3584 <__order_base_2+0x18> c00d3574: 38 63 ff ff addi r3,r3,-1 c00d3578: 7c 63 00 34 cntlzw r3,r3 c00d357c: 20 63 00 20 subfic r3,r3,32 c00d3580: 4e 80 00 20 blr c00d3584: 38 60 00 00 li r3,0 c00d3588: 4e 80 00 20 blr No more need for __order_base_2() to setup a stack frame and save/restore caller address. And the following 'add 1' is merged in the subtract. Another typical use of it: c080ff28 <hugepagesz_setup>: ... c080fff8: 7f c3 f3 78 mr r3,r30 c080fffc: 4b 8f 81 f1 bl c01081ec <__ilog2_u32> c0810000: 38 63 ff f2 addi r3,r3,-14 ... Becomes c080ff1c <hugepagesz_setup>: ... c080ffec: 7f c3 00 34 cntlzw r3,r30 c080fff0: 20 63 00 11 subfic r3,r3,17 ... Here no need to move r30 argument to r3 then substract 14 to result. Just work on r30 and merge the 'sub 14' with the 'sub from 31'. Link: https://lkml.kernel.org/r/803a2ac3d923ebcfd0dd40f5886b05cae7bb0aba.1644243860.git.christophe.leroy@csgroup.euSigned-off-by:
Christophe Leroy <christophe.leroy@csgroup.eu> Signed-off-by:
-
Rasmus Villemoes authored
(1) compiler_types.h is unconditionally included via an -include flag (see scripts/Makefile.lib), and it defines __compiler_offsetof unconditionally. So testing for definedness of __compiler_offsetof is mostly pointless. (2) Every relevant compiler provides __builtin_offsetof (even sparse has had that for 14 years), and if for whatever reason one would end up picking up the poor man's fallback definition (C file compiler with completely custom CFLAGS?), newer clang versions won't treat the result as an Integer Constant Expression, so if used in place where such is required (static initializer or static_assert), one would get errors like t.c:11:16: error: static_assert expression is not an integral constant expression t.c:11:16: note: cast that performs the conversions of a reinterpret_cast is not allowed in a constant expression t.c:4:33: note: expanded from macro 'offsetof' #define offsetof(TYPE, MEMBER) ((size_t)&((TYPE *)0)->MEMBER) So just define offsetof unconditionally and directly in terms of __builtin_offsetof. Link: https://lkml.kernel.org/r/20220202102147.326672-1-linux@rasmusvillemoes.dkSigned-off-by:Rasmus Villemoes <linux@rasmusvillemoes.dk> Reviewed-by:
Miguel Ojeda <ojeda@kernel.org> Reviewed-by:
Nathan Chancellor <nathan@kernel.org> Reviewed-by:
Kees Cook <keescook@chromium.org> Acked-by:
-
Kees Cook authored
Currently it's not possible to enable DEBUG_INFO for an all*config build, since it is marked as "depends on !COMPILE_TEST". This generally makes sense because a debug build of an all*config target ends up taking much longer and the output is much larger. Having this be "default off" makes sense. However, there are cases where enabling DEBUG_INFO for such builds is useful for doing treewide A/B comparisons of build options, etc. Make DEBUG_INFO selectable from any of the DWARF version choice options, with DEBUG_INFO_NONE being the default for COMPILE_TEST. The mutually exclusive relationship between DWARF5 and BTF must be inverted, but the result remains the same. Additionally moves DEBUG_KERNEL and DEBUG_MISC up to the top of the menu because they were enabling features _above_ it, making it weird to navigate menuconfig. [keescook@chromium.org: make DEBUG_INFO always default=n] Link: https://lkml.kernel.org/r/20220128214131.580131-1-keescook@chromium.org Link: https://lore.kernel.org/lkml/YfRY6+CaQxX7O8vF@dev-arch.archlinux-ax161 Link: https://lkml.kernel.org/r/20220125075126.891825-1-keescook@chromium.orgSigned-off-by:
Arnd Bergmann <arnd@arndb.de> Reviewed-by:
Masahiro Yamada <masahiroy@kernel.org> Cc: Nick Desaulniers <ndesaulniers@google.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by:
-
Miaohe Lin authored
Use helper macro __ATTR_RW to define kobj_attribute to make code more clear. Minor readability improvement. Link: https://lkml.kernel.org/r/20220222112034.48298-1-linmiaohe@huawei.comSigned-off-by:
Miaohe Lin <linmiaohe@huawei.com> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by:
-
Bjorn Helgaas authored
Several attributes depend on __CHECKER__, but previously there was no clue in the tree about when __CHECKER__ might be defined. Add hints at the most common places (__kernel, __user, __iomem, __bitwise) and in the sparse documentation. Link: https://lkml.kernel.org/r/20220310220927.245704-3-helgaas@kernel.orgSigned-off-by:
Bjorn Helgaas <bhelgaas@google.com> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Nathan Chancellor <nathan@kernel.org> Cc: Nick Desaulniers <ndesaulniers@google.com> Cc: "Michael S . Tsirkin" <mst@redhat.com> Signed-off-by:
-
Bjorn Helgaas authored
There are no users of "__bitwise__" except the definition of "__bitwise". Remove __bitwise__ and define __bitwise directly. This is a follow-up to 05de9700 ("linux/types.h: enable endian checks for all sparse builds"). [akpm@linux-foundation.org: change the tools/include/linux/types.h definition also] Link: https://lkml.kernel.org/r/20220310220927.245704-2-helgaas@kernel.orgSigned-off-by:
-
Yang Li authored
Fix a spelling problem to remove warnings found by running scripts/kernel-doc, which is caused by using 'make W=1'. fs/proc/vmcore.c:492: warning: Function parameter or member 'size' not described in 'vmcore_alloc_buf' fs/proc/vmcore.c:492: warning: Excess function parameter 'sizez' description in 'vmcore_alloc_buf' Link: https://lkml.kernel.org/r/20220129011449.105278-1-yang.lee@linux.alibaba.comSigned-off-by:
Yang Li <yang.lee@linux.alibaba.com> Reported-by:
Abaci Robot <abaci@linux.alibaba.com> Acked-by:
-
David Hildenbrand authored
Lockdep noticed that there is chance for a deadlock if we have concurrent mmap, concurrent read, and the addition/removal of a callback. As nicely explained by Boqun: "Lockdep warned about the above sequences because rw_semaphore is a fair read-write lock, and the following can cause a deadlock: TASK 1 TASK 2 TASK 3 ====== ====== ====== down_write(mmap_lock); down_read(vmcore_cb_rwsem) down_write(vmcore_cb_rwsem); // blocked down_read(vmcore_cb_rwsem); // cannot get the lock because of the fairness down_read(mmap_lock); // blocked IOW, a reader can block another read if there is a writer queued by the second reader and the lock is fair" To fix this, convert to srcu to make this deadlock impossible. We need srcu as our callbacks can sleep. With this change, I cannot trigger any lockdep warnings. ====================================================== WARNING: possible circular locking dependency detected 5.17.0-0.rc0.20220117git0c947b89.68.test.fc36.x86_64 #1 Not tainted ------------------------------------------------------ makedumpfile/542 is trying to acquire lock: ffffffff832d2eb8 (vmcore_cb_rwsem){.+.+}-{3:3}, at: mmap_vmcore+0x340/0x580 but task is already holding lock: ffff8880af226438 (&mm->mmap_lock#2){++++}-{3:3}, at: vm_mmap_pgoff+0x84/0x150 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #1 (&mm->mmap_lock#2){++++}-{3:3}: lock_acquire+0xc3/0x1a0 __might_fault+0x4e/0x70 _copy_to_user+0x1f/0x90 __copy_oldmem_page+0x72/0xc0 read_from_oldmem+0x77/0x1e0 read_vmcore+0x2c2/0x310 proc_reg_read+0x47/0xa0 vfs_read+0x101/0x340 __x64_sys_pread64+0x5d/0xa0 do_syscall_64+0x43/0x90 entry_SYSCALL_64_after_hwframe+0x44/0xae -> #0 (vmcore_cb_rwsem){.+.+}-{3:3}: validate_chain+0x9f4/0x2670 __lock_acquire+0x8f7/0xbc0 lock_acquire+0xc3/0x1a0 down_read+0x4a/0x140 mmap_vmcore+0x340/0x580 proc_reg_mmap+0x3e/0x90 mmap_region+0x504/0x880 do_mmap+0x38a/0x520 vm_mmap_pgoff+0xc1/0x150 ksys_mmap_pgoff+0x178/0x200 do_syscall_64+0x43/0x90 entry_SYSCALL_64_after_hwframe+0x44/0xae other info that might help us debug this: Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(&mm->mmap_lock#2); lock(vmcore_cb_rwsem); lock(&mm->mmap_lock#2); lock(vmcore_cb_rwsem); *** DEADLOCK *** 1 lock held by makedumpfile/542: #0: ffff8880af226438 (&mm->mmap_lock#2){++++}-{3:3}, at: vm_mmap_pgoff+0x84/0x150 stack backtrace: CPU: 0 PID: 542 Comm: makedumpfile Not tainted 5.17.0-0.rc0.20220117git0c947b89.68.test.fc36.x86_64 #1 Hardware name: Red Hat KVM, BIOS 0.5.1 01/01/2011 Call Trace: __lock_acquire+0x8f7/0xbc0 lock_acquire+0xc3/0x1a0 down_read+0x4a/0x140 mmap_vmcore+0x340/0x580 proc_reg_mmap+0x3e/0x90 mmap_region+0x504/0x880 do_mmap+0x38a/0x520 vm_mmap_pgoff+0xc1/0x150 ksys_mmap_pgoff+0x178/0x200 do_syscall_64+0x43/0x90 Link: https://lkml.kernel.org/r/20220119193417.100385-1-david@redhat.com Fixes: cc5f2704 ("proc/vmcore: convert oldmem_pfn_is_ram callback to more generic vmcore callbacks") Signed-off-by:David Hildenbrand <david@redhat.com> Reported-by:
-
Hao Lee authored
It's not a standard approach that use __get_free_page() to alloc path buffer directly. We'd better use kmalloc and PATH_MAX. PAGE_SIZE is different on different archs. An unlinked file with very long canonical pathname will readlink differently because "(deleted)" eats into a buffer. --adobriyan [akpm@linux-foundation.org: remove now-unneeded cast] Link: https://lkml.kernel.org/r/Ye1fCxyZZ0I5lgOL@localhost.localdomainSigned-off-by:
Hao Lee <haolee.swjtu@gmail.com> Signed-off-by:
Alexey Dobriyan <adobriyan@gmail.com> Cc: Christian Brauner <christian.brauner@ubuntu.com> Cc: Kees Cook <keescook@chromium.org> Cc: James Morris <jamorris@linux.microsoft.com> Signed-off-by:
-
- 20 Mar, 2022 1 commit
-
-
Linus Torvalds authored
Signed-off-by:
-
