- 29 Dec, 2023 6 commits
-
-
Uwe Kleine-König authored
DEBUG_STACK_USAGE doesn't only have an influence on the output of sysrq-T and sysrq-P, it also enables a message at process exit. See check_stack_usage() in kernel/exit.c where this is implemented. Link: https://lkml.kernel.org/r/20231219182808.210284-2-u.kleine-koenig@pengutronix.de Signed-off-by:
Uwe Kleine-König <u.kleine-koenig@pengutronix.de> Cc: Douglas Anderson <dianders@chromium.org> Cc: Geert Uytterhoeven <geert+renesas@glider.be> Cc: Kees Cook <keescook@chromium.org> Cc: Marco Elver <elver@google.com> Cc: "Paul E. McKenney" <paulmck@kernel.org> Cc: Pengutronix Kernel Team <kernel@pengutronix.de> Cc: Petr Mladek <pmladek@suse.com> Cc: Randy Dunlap <rdunlap@infradead.org> Cc: Stephen Boyd <sboyd@kernel.org> Cc: Zhaoyang Huang <zhaoyang.huang@unisoc.com> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org>
-
Youling Tang authored
scripts/checkstack.pl lacks support for the loongarch architecture. Add support to detect "addi.{w,d} $sp, $sp, -FRAME_SIZE" stack frame generation instruction. Link: https://lkml.kernel.org/r/MW4PR84MB314514273F0B7DBCC5E35A978192A@MW4PR84MB3145.NAMPRD84.PROD.OUTLOOK.COM Signed-off-by:Youling Tang <tangyouling@kylinos.cn> Acked-by:
Huacai Chen <chenhuacai@loongson.cn> Cc: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de> Signed-off-by:
-
Youling Tang authored
An example how to invoke decodecode for loongarch64: $ echo 'Code: 380839f6 380831f9 28412bae <24000ca6> 004081ad 0014cb50 004083e8 02bff34c 58008e91' | \ ARCH=loongarch CROSS_COMPILE=loongarch64-linux-gnu- \ ./scripts/decodecode Code: 380839f6 380831f9 28412bae <24000ca6> 004081ad 0014cb50 004083e8 02bff34c 58008e91 All code ======== 0: 380839f6 ldx.w $fp, $t3, $t2 4: 380831f9 ldx.w $s2, $t3, $t0 8: 28412bae ld.h $t2, $s6, 74(0x4a) c:* 24000ca6 ldptr.w $a2, $a1, 12(0xc) <-- trapping instruction 10: 004081ad slli.w $t1, $t1, 0x0 14: 0014cb50 and $t4, $s3, $t6 18: 004083e8 slli.w $a4, $s8, 0x0 1c: 02bff34c addi.w $t0, $s3, -4(0xffc) 20: 58008e91 beq $t8, $t5, 140(0x8c) # 0xac Code starting with the faulting instruction =========================================== 0: 24000ca6 ldptr.w $a2, $a1, 12(0xc) 4: 004081ad slli.w $t1, $t1, 0x0 8: 0014cb50 and $t4, $s3, $t6 c: 004083e8 slli.w $a4, $s8, 0x0 10: 02bff34c addi.w $t0, $s3, -4(0xffc) 14: 58008e91 beq $t8, $t5, 140(0x8c) # 0xa0 Link: https://lkml.kernel.org/r/MW4PR84MB3145B99B9677BB7887BB26CD8192A@MW4PR84MB3145.NAMPRD84.PROD.OUTLOOK.COM Signed-off-by:
-
Yuntao Wang authored
temp_end represents the address of the last available byte. Therefore, the starting address of the memory segment with temp_end as its last available byte and a size of `kbuf->memsz`, that is, the value of temp_start, should be `temp_end - kbuf->memsz + 1` instead of `temp_end - kbuf->memsz`. Additionally, use the ALIGN_DOWN macro instead of open-coding it directly in locate_mem_hole_top_down() to improve code readability. Link: https://lkml.kernel.org/r/20231217033528.303333-3-ytcoode@gmail.com Signed-off-by:
Yuntao Wang <ytcoode@gmail.com> Acked-by:
Baoquan He <bhe@redhat.com> Cc: Borislav Petkov (AMD) <bp@alien8.de> Cc: Dave Hansen <dave.hansen@linux.intel.com> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Thomas Gleixner <tglx@linutronix.de> Signed-off-by:
-
Yuntao Wang authored
The end parameter received by kimage_is_destination_range() should be the last valid byte address of the target memory segment plus 1. However, in the locate_mem_hole_bottom_up() and locate_mem_hole_top_down() functions, the corresponding value passed to kimage_is_destination_range() is the last valid byte address of the target memory segment, which is 1 less. There are two ways to fix this bug. We can either correct the logic of the locate_mem_hole_bottom_up() and locate_mem_hole_top_down() functions, or we can fix kimage_is_destination_range() by making the end parameter represent the last valid byte address of the target memory segment. Here, we choose the second approach. Due to the modification to kimage_is_destination_range(), we also need to adjust its callers, such as kimage_alloc_normal_control_pages() and kimage_alloc_page(). Link: https://lkml.kernel.org/r/20231217033528.303333-2-ytcoode@gmail.com Signed-off-by:
-
Bagas Sanjaya authored
Commit 62c46d55 ("MAINTAINERS: Removing Ohad from remoteproc/rpmsg maintenance") removes his MAINTAINERS entry in regards to remoteproc subsystem due to his inactivity (the last commit with his Signed-off-by is 99c429cb ("remoteproc/wkup_m3: Use MODULE_DEVICE_TABLE to export alias") which is authored in 2015 and his last LKML message prior to 62c46d55 was [1]). Remove also his MAINTAINERS entry for hwspinlock subsystem as there is no point of Cc'ing maintainers who never respond in a long time. [1]: https://lore.kernel.org/r/CAK=Wgbbcyi36ef1-PV8VS=M6nFoQnFGUDWy6V7OCnkt0dDrtfg@mail.gmail.com/ Link: https://lkml.kernel.org/r/20231218132830.5104-2-bagasdotme@gmail.com Signed-off-by:
Bagas Sanjaya <bagasdotme@gmail.com> Acked-by:
Ohad Ben Cohen <ohad@wizery.com> Cc: Baolin Wang <baolin.wang@linux.alibaba.com> Cc: Bjorn Andersson <andersson@kernel.org> Cc: Mathieu Poirier <mathieu.poirier@linaro.org> Signed-off-by:
-
- 20 Dec, 2023 22 commits
-
-
Wang Jinchao authored
Remove second include of linux/kexec.h Link: https://lkml.kernel.org/r/202312151654+0800-wangjinchao@xfusion.com Signed-off-by:
Wang Jinchao <wangjinchao@xfusion.com> Acked-by:
-
Yuntao Wang authored
The expression `mstart + resource_size(res) - 1` is actually equivalent to `res->end`, simplify the logic of this function to improve readability. Link: https://lkml.kernel.org/r/20231212150506.31711-1-ytcoode@gmail.com Signed-off-by:
-
Yuntao Wang authored
Use ALIGN macro instead of open-coding it to improve code readability. Link: https://lkml.kernel.org/r/20231212142706.25149-1-ytcoode@gmail.com Signed-off-by:
-
Randy Dunlap authored
Use "Its" or "its" for possessive instead of "it's" (contraction for "it is"). Link: https://lkml.kernel.org/r/20231210053429.23146-1-rdunlap@infradead.org Fixes: db2aa7fd ("initramfs: allow again choice of the embedded initram compression algorithm") Signed-off-by:
Randy Dunlap <rdunlap@infradead.org> Reviewed-by:
Nicolas Schier <nicolas@fjasle.eu> Acked-by:
"Francisco Blas Izquierdo Riera (klondike)" <klondike@klondike.es> Cc: Masahiro Yamada <masahiroy@kernel.org> Cc: Nathan Chancellor <nathan@kernel.org> Signed-off-by:
-
Randy Dunlap authored
KCMP, RSEQ, CACHESTAT_SYSCALL, and PC104 depend on EXPERT but not shown in the EXPERT menu. Move some lines around so that they are displayed in the EXPERT menu. Drop one useless comment. Change "enabled" to "enable" for DEBUG_RSEQ. Link: https://lkml.kernel.org/r/20231208045819.2922-1-rdunlap@infradead.org Signed-off-by:
Mathieu Desnoyers <mathieu.desnoyers@efficios.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: "Paul E. McKenney" <paulmck@kernel.org> Cc: Boqun Feng <boqun.feng@gmail.com> Signed-off-by:
-
Kevin Hao authored
TASK_KILLABLE already includes TASK_UNINTERRUPTIBLE, so there is no need to add a separate TASK_UNINTERRUPTIBLE. Link: https://lkml.kernel.org/r/20231208084115.1973285-1-haokexin@gmail.com Signed-off-by:
Kevin Hao <haokexin@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Signed-off-by:
-
Ryusuke Konishi authored
nilfs_sufile_do_free(), which is called when log write fails or during GC, uses WARN_ONs to check for abnormal status of metadata. In the former case, these WARN_ONs will not be fired, but in the latter case they don't "never-happen". It is possible to trigger these by intentionally modifying the userland GC library to release segments that are not in the expected state. So, replace them with warning output using the dedicated macro nilfs_warn(). This replaces two potentially triggered WARN_ONs with ones that use a warning output macro. Link: https://lkml.kernel.org/r/20231207045730.5205-1-konishi.ryusuke@gmail.com Signed-off-by:
Ryusuke Konishi <konishi.ryusuke@gmail.com> Signed-off-by:
-
Randy Dunlap authored
Correct the function parameter kernel-doc notation to prevent warnings: vxfs_lookup.c:192: warning: Function parameter or member 'ctx' not described in 'vxfs_readdir' vxfs_lookup.c:192: warning: Excess function parameter 'retp' description in 'vxfs_readdir' vxfs_lookup.c:192: warning: Excess function parameter 'filler' description in 'vxfs_readdir' Link: https://lkml.kernel.org/r/20231207212035.25345-3-rdunlap@infradead.org Signed-off-by:
Christoph Hellwig <hch@lst.de> Signed-off-by:
-
Randy Dunlap authored
Correct the function parameter name to prevent kernel-doc warnings: vxfs_immed.c:32: warning: Function parameter or member 'fp' not described in 'vxfs_immed_read_folio' vxfs_immed.c:32: warning: Excess function parameter 'file' description in 'vxfs_immed_read_folio' Link: https://lkml.kernel.org/r/20231207212035.25345-2-rdunlap@infradead.org Signed-off-by:
-
Randy Dunlap authored
Fix -Wall kernel-doc warnings in vxfs_bmap.c: vxfs_bmap.c:44: warning: Function parameter or member 'bn' not described in 'vxfs_bmap_ext4' vxfs_bmap.c:44: warning: Excess function parameter 'iblock' description in 'vxfs_bmap_ext4' vxfs_bmap.c:108: warning: No description found for return value of 'vxfs_bmap_indir' vxfs_bmap.c:187: warning: No description found for return value of 'vxfs_bmap_typed' vxfs_bmap.c:251: warning: No description found for return value of 'vxfs_bmap1' Link: https://lkml.kernel.org/r/20231207212035.25345-1-rdunlap@infradead.org Signed-off-by:
-
Randy Dunlap authored
Correct kernel-doc comments in tsi721.c and tsi721_dma.c to prevent warnings from scripts/kernel-doc. tsi721_dma.c:293: warning: expecting prototype for tsi721_omsg_msix(). Prototype was for tsi721_bdma_msix() instead tsi721.c:215: warning: Function parameter or member 'data' not described in 'tsi721_cread_dma' tsi721.c:215: warning: Excess function parameter 'val' description in 'tsi721_cread_dma' tsi721.c:238: warning: Function parameter or member 'data' not described in 'tsi721_cwrite_dma' tsi721.c:238: warning: Excess function parameter 'val' description in 'tsi721_cwrite_dma' tsi721.c:2548: warning: Function parameter or member 'attr' not described in 'tsi721_query_mport' tsi721.c:2548: warning: Excess function parameter 'mbox' description in 'tsi721_query_mport' and 27 warnings like this one: tsi721.c:59: warning: No description found for return value of 'tsi721_lcread' Link: https://lkml.kernel.org/r/20231206175528.16386-1-rdunlap@infradead.org Signed-off-by:
-
Mark Rutland authored
The Kconfig help text for CONFIG_KCOV describes that recorded PC values will not be stable across machines or reboots when RANDOMIZE_BASE is selected. This was the case when KCOV was introduced in commit: 5c9a8750 ("kernel: add kcov code coverage") However, this changed in commit: 4983f0ab ("kcov: make kcov work properly with KASLR enabled") Since that commit KCOV always subtracts the KASLR offset from PC values, which ensures that these are stable across machines and across reboots even when RANDOMIZE_BASE is selected. Unfortunately, that commit failed to update the Kconfig help text, which still suggests disabling RANDOMIZE_BASE even though this is no longer necessary. Remove the stale Kconfig text. Link: https://lkml.kernel.org/r/20231204171807.3313022-1-mark.rutland@arm.com Reported-by:
Borislav Petkov <bp@alien8.de> Signed-off-by:
Mark Rutland <mark.rutland@arm.com> Reviewed-by:
Dmitry Vyukov <dvyukov@google.com> Cc: Alexander Popov <alex.popov@linux.com> Cc: Andrey Konovalov <andreyknvl@gmail.com> Signed-off-by:
-
Baoquan He authored
This was introduced in commit fba8a867 ("RISC-V: Add kexec support"). It should work on CONFIG_KEXEC_CORE, but not CONFIG_KEXEC only, since we could set CONFIG_KEXEC_FILE=y and CONFIG_KEXEC=N, or only set CONFIG_CRASH_DUMP=y and disable both CONFIG_KEXEC and CONFIG_KEXEC_FILE. In these cases, the AFLAGS won't take effect with the current ifdeffery for AFLAGS_kexec_relocate.o. So fix it now. Link: https://lkml.kernel.org/r/20231201062538.27240-1-bhe@redhat.com Signed-off-by:
-
Baoquan He authored
Then when specifying '-d' for kexec_file_load interface, loaded locations of kernel/initrd/cmdline etc can be printed out to help debug. Here replace pr_debug() with the newly added kexec_dprintk() in kexec_file loading related codes. Link: https://lkml.kernel.org/r/20231213055747.61826-8-bhe@redhat.com Signed-off-by:
-
Baoquan He authored
Then when specifying '-d' for kexec_file_load interface, loaded locations of kernel/initrd/cmdline etc can be printed out to help debug. Here replace pr_debug() with the newly added kexec_dprintk() in kexec_file loading related codes. Link: https://lkml.kernel.org/r/20231213055747.61826-7-bhe@redhat.com Signed-off-by:
-
Baoquan He authored
Then when specifying '-d' for kexec_file_load interface, loaded locations of kernel/initrd/cmdline etc can be printed out to help debug. Here replace pr_debug() with the newly added kexec_dprintk() in kexec_file loading related codes. And also replace pr_notice() with kexec_dprintk() in elf_kexec_load() because loaded location of purgatory and device tree are only printed out for debugging, it doesn't make sense to always print them out. And also remove kexec_image_info() because the content has been printed out in generic code. Link: https://lkml.kernel.org/r/20231213055747.61826-6-bhe@redhat.com Signed-off-by:
-
Baoquan He authored
Then when specifying '-d' for kexec_file_load interface, loaded locations of kernel/initrd/cmdline etc can be printed out to help debug. Here replace pr_debug() with the newly added kexec_dprintk() in kexec_file loading related codes. And also remove the kimage->segment[] printing because the generic code has done the printing. Link: https://lkml.kernel.org/r/20231213055747.61826-5-bhe@redhat.com Signed-off-by:
-
Baoquan He authored
Then when specifying '-d' for kexec_file_load interface, loaded locations of kernel/initrd/cmdline etc can be printed out to help debug. Here replace pr_debug() with the newly added kexec_dprintk() in kexec_file loading related codes. And also print out e820 memmap passed to 2nd kernel just as kexec_load interface has been doing. Link: https://lkml.kernel.org/r/20231213055747.61826-4-bhe@redhat.com Signed-off-by:
-
Baoquan He authored
Then when specifying '-d' for kexec_file_load interface, loaded locations of kernel/initrd/cmdline etc can be printed out to help debug. Here replace pr_debug() with the newly added kexec_dprintk() in kexec_file loading related codes. And also print out type/start/head of kimage and flags to help debug. Link: https://lkml.kernel.org/r/20231213055747.61826-3-bhe@redhat.com Signed-off-by:
-
Baoquan He authored
Patch series "kexec_file: print out debugging message if required", v4. Currently, specifying '-d' on kexec command will print a lot of debugging informationabout kexec/kdump loading with kexec_load interface. However, kexec_file_load prints nothing even though '-d' is specified. It's very inconvenient to debug or analyze the kexec/kdump loading when something wrong happened with kexec/kdump itself or develper want to check the kexec/kdump loading. In this patchset, a kexec_file flag is KEXEC_FILE_DEBUG added and checked in code. If it's passed in, debugging message of kexec_file code will be printed out and can be seen from console and dmesg. Otherwise, the debugging message is printed like beofre when pr_debug() is taken. Note: **** ===== 1) The code in kexec-tools utility also need be changed to support passing KEXEC_FILE_DEBUG to kernel when 'kexec -s -d' is specified. The patch link is here: ========= [PATCH] kexec_file: add kexec_file flag to support debug printing http://lists.infradead.org/pipermail/kexec/2023-November/028505.html 2) s390 also has kexec_file code, while I am not sure what debugging information is necessary. So leave it to s390 developer. Test: **** ==== Testing was done in v1 on x86_64 and arm64. For v4, tested on x86_64 again. And on x86_64, the printed messages look like below: -------------------------------------------------------------- kexec measurement buffer for the loaded kernel at 0x207fffe000. Loaded purgatory at 0x207fff9000 Loaded boot_param, command line and misc at 0x207fff3000 bufsz=0x1180 memsz=0x1180 Loaded 64bit kernel at 0x207c000000 bufsz=0xc88200 memsz=0x3c4a000 Loaded initrd at 0x2079e79000 bufsz=0x2186280 memsz=0x2186280 Final command line is: root=/dev/mapper/fedora_intel--knightslanding--lb--02-root ro rd.lvm.lv=fedora_intel-knightslanding-lb-02/root console=ttyS0,115200N81 crashkernel=256M E820 memmap: 0000000000000000-000000000009a3ff (1) 000000000009a400-000000000009ffff (2) 00000000000e0000-00000000000fffff (2) 0000000000100000-000000006ff83fff (1) 000000006ff84000-000000007ac50fff (2) ...... 000000207fff6150-000000207fff615f (128) 000000207fff6160-000000207fff714f (1) 000000207fff7150-000000207fff715f (128) 000000207fff7160-000000207fff814f (1) 000000207fff8150-000000207fff815f (128) 000000207fff8160-000000207fffffff (1) nr_segments = 5 segment[0]: buf=0x000000004e5ece74 bufsz=0x211 mem=0x207fffe000 memsz=0x1000 segment[1]: buf=0x000000009e871498 bufsz=0x4000 mem=0x207fff9000 memsz=0x5000 segment[2]: buf=0x00000000d879f1fe bufsz=0x1180 mem=0x207fff3000 memsz=0x2000 segment[3]: buf=0x000000001101cd86 bufsz=0xc88200 mem=0x207c000000 memsz=0x3c4a000 segment[4]: buf=0x00000000c6e38ac7 bufsz=0x2186280 mem=0x2079e79000 memsz=0x2187000 kexec_file_load: type:0, start:0x207fff91a0 head:0x109e004002 flags:0x8 --------------------------------------------------------------------------- This patch (of 7): When specifying 'kexec -c -d', kexec_load interface will print loading information, e.g the regions where kernel/initrd/purgatory/cmdline are put, the memmap passed to 2nd kernel taken as system RAM ranges, and printing all contents of struct kexec_segment, etc. These are very helpful for analyzing or positioning what's happening when kexec/kdump itself failed. The debugging printing for kexec_load interface is made in user space utility kexec-tools. Whereas, with kexec_file_load interface, 'kexec -s -d' print nothing. Because kexec_file code is mostly implemented in kernel space, and the debugging printing functionality is missed. It's not convenient when debugging kexec/kdump loading and jumping with kexec_file_load interface. Now add KEXEC_FILE_DEBUG to kexec_file flag to control the debugging message printing. And add global variable kexec_file_dbg_print and macro kexec_dprintk() to facilitate the printing. This is a preparation, later kexec_dprintk() will be used to replace the existing pr_debug(). Once 'kexec -s -d' is specified, it will print out kexec/kdump loading information. If '-d' is not specified, it regresses to pr_debug(). Link: https://lkml.kernel.org/r/20231213055747.61826-1-bhe@redhat.com Link: https://lkml.kernel.org/r/20231213055747.61826-2-bhe@redhat.com Signed-off-by:
-
Arnd Bergmann authored
Over the years we went from > 1000 of warnings to under 100 earlier this year, and I sent patches to address all the ones that I saw with compile testing randcom configs on arm64, arm and x86 kernels. This is a really useful warning, as it catches real bugs when there are mismatched prototypes. In particular with kernel control flow integrity enabled, those are no longer allowed. I have done extensive testing to ensure that there are no new build errors or warnings on any configuration of x86, arm and arm64 builds. I also made sure that at least both the normal defconfig and an allmodconfig build is clean for arc, csky, loongarch, m68k, microblaze, openrisc, parisc, powerpc, riscv, s390, and xtensa, with the respective maintainers doing most of the patches. At this point, there are five architectures with a number of known regressions: alpha, nios2, mips, sh and sparc. In the previous version of this patch, I had turned off the missing prototype warnings for the 15 architectures that still had issues, but since there are only five left, I think we can leave the rest to the maintainers (Cc'd here) as well. Link: https://lkml.kernel.org/r/20231123110506.707903-7-arnd@kernel.org Link: https://lore.kernel.org/lkml/20230810141947.1236730-1-arnd@kernel.org/ Signed-off-by:
Arnd Bergmann <arnd@arndb.de> Reviewed-by:
Kees Cook <keescook@chromium.org> Acked-by: Palmer Dabbelt <palmer@rivosinc.com> # RISC-V Cc: Richard Henderson <richard.henderson@linaro.org> Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Cc: Matt Turner <mattst88@gmail.com> Cc: Dinh Nguyen <dinguyen@kernel.org> Cc: Thomas Bogendoerfer <tsbogend@alpha.franken.de> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Cc: Rich Felker <dalias@libc.org> Cc: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de> Cc: "David S. Miller" <davem@davemloft.net> Cc: David Woodhouse <dwmw2@infradead.org> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: Masahiro Yamada <masahiroy@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: Nathan Chancellor <nathan@kernel.org> Cc: Nicolas Schier <nicolas@fjasle.eu> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Richard Weinberger <richard@nod.at> Cc: Stephen Rothwell <sfr@canb.auug.org.au> Cc: Tudor Ambarus <tudor.ambarus@linaro.org> Cc: Zhihao Cheng <chengzhihao1@huawei.com> Signed-off-by:
-
Arnd Bergmann authored
My earlier patch removed __weak function declarations that used to be turned into wild branches by the linker, instead causing a link failure when the called functions are unavailable: mips-linux-ld: arch/mips/mm/cache.o: in function `cpu_cache_init': cache.c:(.text+0x670): undefined reference to `r3k_cache_init' The __weak method seems suboptimal, so rather than putting that back, make the function calls conditional on the Kconfig symbol that controls the compilation. [akpm@linux-foundation.org: fix whitespace while we're in there] Link: https://lkml.kernel.org/r/20231214205506.310402-1-arnd@kernel.org Fixes: 66445677 ("mips: move cache declarations into header") Signed-off-by:
kernelci.org bot <bot@kernelci.org> Cc: Jiaxun Yang <jiaxun.yang@flygoat.com> Cc: Thomas Bogendoerfer <tsbogend@alpha.franken.de> Cc: Zi Yan <ziy@nvidia.com> Signed-off-by:
-
- 13 Dec, 2023 12 commits
-
-
Andrew Morton authored
-
Yu Zhao authored
In the effort to reduce zombie memcgs [1], it was discovered that the memcg LRU doesn't apply enough pressure on offlined memcgs. Specifically, instead of rotating them to the tail of the current generation (MEMCG_LRU_TAIL) for a second attempt, it moves them to the next generation (MEMCG_LRU_YOUNG) after the first attempt. Not applying enough pressure on offlined memcgs can cause them to build up, and this can be particularly harmful to memory-constrained systems. On Pixel 8 Pro, launching apps for 50 cycles: Before After Change Zombie memcgs 45 35 -22% [1] https://lore.kernel.org/CABdmKX2M6koq4Q0Cmp_-=wbP0Qa190HdEGGaHfxNS05gAkUtPA@mail.gmail.com/ Link: https://lkml.kernel.org/r/20231208061407.2125867-4-yuzhao@google.com Fixes: e4dde56c ("mm: multi-gen LRU: per-node lru_gen_folio lists") Signed-off-by:Yu Zhao <yuzhao@google.com> Reported-by:
T.J. Mercier <tjmercier@google.com> Tested-by:
-
Yu Zhao authored
While investigating kswapd "consuming 100% CPU" [1] (also see "mm/mglru: try to stop at high watermarks"), it was discovered that the memcg LRU can breach the thrashing protection imposed by min_ttl_ms. Before the memcg LRU: kswapd() shrink_node_memcgs() mem_cgroup_iter() inc_max_seq() // always hit a different memcg lru_gen_age_node() mem_cgroup_iter() check the timestamp of the oldest generation After the memcg LRU: kswapd() shrink_many() restart: iterate the memcg LRU: inc_max_seq() // occasionally hit the same memcg if raced with lru_gen_rotate_memcg(): goto restart lru_gen_age_node() mem_cgroup_iter() check the timestamp of the oldest generation Specifically, when the restart happens in shrink_many(), it needs to stick with the (memcg LRU) generation it began with. In other words, it should neither re-read memcg_lru->seq nor age an lruvec of a different generation. Otherwise it can hit the same memcg multiple times without giving lru_gen_age_node() a chance to check the timestamp of that memcg's oldest generation (against min_ttl_ms). [1] https://lore.kernel.org/CAK8fFZ4DY+GtBA40Pm7Nn5xCHy+51w3sfxPqkqpqakSXYyX+Wg@mail.gmail.com/ Link: https://lkml.kernel.org/r/20231208061407.2125867-3-yuzhao@google.com Fixes: e4dde56c ("mm: multi-gen LRU: per-node lru_gen_folio lists") Signed-off-by: -
Yu Zhao authored
The initial MGLRU patchset didn't include the memcg LRU support, and it relied on should_abort_scan(), added by commit f76c8337 ("mm: multi-gen LRU: optimize multiple memcgs"), to "backoff to avoid overshooting their aggregate reclaim target by too much". Later on when the memcg LRU was added, should_abort_scan() was deemed unnecessary, and the test results [1] showed no side effects after it was removed by commit a579086c ("mm: multi-gen LRU: remove eviction fairness safeguard"). However, that test used memory.reclaim, which sets nr_to_reclaim to SWAP_CLUSTER_MAX. So it can overshoot only by SWAP_CLUSTER_MAX-1 pages, i.e., from nr_reclaimed=nr_to_reclaim-1 to nr_reclaimed=nr_to_reclaim+SWAP_CLUSTER_MAX-1. Compared with the batch size kswapd sets to nr_to_reclaim, SWAP_CLUSTER_MAX is tiny. Therefore that test isn't able to reproduce the worst case scenario, i.e., kswapd overshooting GBs on large systems and "consuming 100% CPU" (see the Closes tag). Bring back a simplified version of should_abort_scan() on top of the memcg LRU, so that kswapd stops when all eligible zones are above their respective high watermarks plus a small delta to lower the chance of KSWAPD_HIGH_WMARK_HIT_QUICKLY. Note that this only applies to order-0 reclaim, meaning compaction-induced reclaim can still run wild (which is a different problem). On Android, launching 55 apps sequentially: Before After Change pgpgin 838377172 802955040 -4% pgpgout 38037080 34336300 -10% [1] https://lore.kernel.org/20221222041905.2431096-1-yuzhao@google.com/ Link: https://lkml.kernel.org/r/20231208061407.2125867-2-yuzhao@google.com Fixes: a579086c ("mm: multi-gen LRU: remove eviction fairness safeguard") Signed-off-by:
Charan Teja Kalla <quic_charante@quicinc.com> Reported-by:
Jaroslav Pulchart <jaroslav.pulchart@gooddata.com> Closes: https://lore.kernel.org/CAK8fFZ4DY+GtBA40Pm7Nn5xCHy+51w3sfxPqkqpqakSXYyX+Wg@mail.gmail.com/ Tested-by:
Kalesh Singh <kaleshsingh@google.com> Cc: Hillf Danton <hdanton@sina.com> Cc: Kairui Song <ryncsn@gmail.com> Cc: T.J. Mercier <tjmercier@google.com> Cc: <stable@vger.kernel.org> Signed-off-by:
-
Yu Zhao authored
Unmapped folios accessed through file descriptors can be underprotected. Those folios are added to the oldest generation based on: 1. The fact that they are less costly to reclaim (no need to walk the rmap and flush the TLB) and have less impact on performance (don't cause major PFs and can be non-blocking if needed again). 2. The observation that they are likely to be single-use. E.g., for client use cases like Android, its apps parse configuration files and store the data in heap (anon); for server use cases like MySQL, it reads from InnoDB files and holds the cached data for tables in buffer pools (anon). However, the oldest generation can be very short lived, and if so, it doesn't provide the PID controller with enough time to respond to a surge of refaults. (Note that the PID controller uses weighted refaults and those from evicted generations only take a half of the whole weight.) In other words, for a short lived generation, the moving average smooths out the spike quickly. To fix the problem: 1. For folios that are already on LRU, if they can be beyond the tracking range of tiers, i.e., five accesses through file descriptors, move them to the second oldest generation to give them more time to age. (Note that tiers are used by the PID controller to statistically determine whether folios accessed multiple times through file descriptors are worth protecting.) 2. When adding unmapped folios to LRU, adjust the placement of them so that they are not too close to the tail. The effect of this is similar to the above. On Android, launching 55 apps sequentially: Before After Change workingset_refault_anon 25641024 25598972 0% workingset_refault_file 115016834 106178438 -8% Link: https://lkml.kernel.org/r/20231208061407.2125867-1-yuzhao@google.com Fixes: ac35a490 ("mm: multi-gen LRU: minimal implementation") Signed-off-by: -
David Stevens authored
Split folios during the second loop of shmem_undo_range. It's not sufficient to only split folios when dealing with partial pages, since it's possible for a THP to be faulted in after that point. Calling truncate_inode_folio in that situation can result in throwing away data outside of the range being targeted. [akpm@linux-foundation.org: tidy up comment layout] Link: https://lkml.kernel.org/r/20230418084031.3439795-1-stevensd@google.com Fixes: b9a8a419 ("truncate,shmem: Handle truncates that split large folios") Signed-off-by:
David Stevens <stevensd@chromium.org> Cc: Matthew Wilcox (Oracle) <willy@infradead.org> Cc: Suleiman Souhlal <suleiman@google.com> Cc: <stable@vger.kernel.org> Signed-off-by:
-
John Hubbard authored
This reverts commit 9fc96c7c ("selftests: error out if kernel header files are not yet built"). It turns out that requiring the kernel headers to be built as a prerequisite to building selftests, does not work in many cases. For example, Peter Zijlstra writes: "My biggest beef with the whole thing is that I simply do not want to use 'make headers', it doesn't work for me. I have a ton of output directories and I don't care to build tools into the output dirs, in fact some of them flat out refuse to work that way (bpf comes to mind)." [1] Therefore, stop erroring out on the selftests build. Additional patches will be required in order to change over to not requiring the kernel headers. [1] https://lore.kernel.org/20231208221007.GO28727@noisy.programming.kicks-ass.net Link: https://lkml.kernel.org/r/20231209020144.244759-1-jhubbard@nvidia.com Fixes: 9fc96c7c ("selftests: error out if kernel header files are not yet built") Signed-off-by:
John Hubbard <jhubbard@nvidia.com> Cc: Anders Roxell <anders.roxell@linaro.org> Cc: Muhammad Usama Anjum <usama.anjum@collabora.com> Cc: David Hildenbrand <david@redhat.com> Cc: Peter Xu <peterx@redhat.com> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Nathan Chancellor <nathan@kernel.org> Cc: Shuah Khan <shuah@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Marcos Paulo de Souza <mpdesouza@suse.com> Cc: <stable@vger.kernel.org> Signed-off-by:
-
Yuntao Wang authored
If crash_base is equal to CRASH_ADDR_LOW_MAX, it also indicates that the crashkernel memory is allocated from high memory. However, the current check only considers the case where crash_base is greater than CRASH_ADDR_LOW_MAX. Fix it. The runtime effects is that crashkernel high memory is successfully reserved, whereas the crashkernel low memory is bypassed in this case, then kdump kernel bootup will fail because of no low memory under 4G. This patch also includes some minor cleanups. Link: https://lkml.kernel.org/r/20231209141438.77233-1-ytcoode@gmail.com Fixes: 0ab97169 ("crash_core: add generic function to do reservation") Signed-off-by:
-
Baoquan He authored
With the current ifdeffery CONFIG_KEXEC, get_cmdline_acpi_rsdp() is only available when kexec_load interface is taken, while kexec_file_load interface can't make use of it. Now change it to CONFIG_KEXEC_CORE. Link: https://lkml.kernel.org/r/20231208073036.7884-6-bhe@redhat.com Signed-off-by:
-
Baoquan He authored
The select of KEXEC for CRASH_DUMP in kernel/Kconfig.kexec will be dropped, then compiling errors will be triggered if below config items are set: === CONFIG_CRASH_CORE=y CONFIG_KEXEC_CORE=y CONFIG_CRASH_DUMP=y === Here, change the dependency of building kexec_core related object files, and the ifdeffery on SuperH from CONFIG_KEXEC to CONFIG_KEXEC_CORE. Link: https://lkml.kernel.org/r/20231208073036.7884-5-bhe@redhat.com Signed-off-by:
-
Baoquan He authored
The select of KEXEC for CRASH_DUMP in kernel/Kconfig.kexec will be dropped, then compiling errors will be triggered if below config items are set: === CONFIG_CRASH_CORE=y CONFIG_KEXEC_CORE=y CONFIG_CRASH_DUMP=y === -------------------------------------------------------------------- mipsel-linux-ld: kernel/kexec_core.o: in function `kimage_free': kernel/kexec_core.c:(.text+0x2200): undefined reference to `machine_kexec_cleanup' mipsel-linux-ld: kernel/kexec_core.o: in function `__crash_kexec': kernel/kexec_core.c:(.text+0x2480): undefined reference to `machine_crash_shutdown' mipsel-linux-ld: kernel/kexec_core.c:(.text+0x2488): undefined reference to `machine_kexec' mipsel-linux-ld: kernel/kexec_core.o: in function `kernel_kexec': kernel/kexec_core.c:(.text+0x29b8): undefined reference to `machine_shutdown' mipsel-linux-ld: kernel/kexec_core.c:(.text+0x29c0): undefined reference to `machine_kexec' -------------------------------------------------------------------- Here, change the dependency of building kexec_core related object files, and the ifdeffery in mips from CONFIG_KEXEC to CONFIG_KEXEC_CORE. Link: https://lkml.kernel.org/r/20231208073036.7884-4-bhe@redhat.com Signed-off-by:
kernel test robot <lkp@intel.com> Closes: https://lore.kernel.org/oe-kbuild-all/202311302042.sn8cDPIX-lkp@intel.com/ Cc: Eric DeVolder <eric_devolder@yahoo.com> Cc: Ignat Korchagin <ignat@cloudflare.com> Cc: Stephen Rothwell <sfr@canb.auug.org.au> Signed-off-by:
-
Baoquan He authored
The select of KEXEC for CRASH_DUMP in kernel/Kconfig.kexec will be dropped, then compiling errors will be triggered if below config items are set: === CONFIG_CRASH_CORE=y CONFIG_KEXEC_CORE=y CONFIG_CRASH_DUMP=y === Here, change the dependency of buinding machine_kexec.o relocate_kernel.o and the ifdeffery in asm/kexe.h to CONFIG_KEXEC_CORE. Link: https://lkml.kernel.org/r/20231208073036.7884-3-bhe@redhat.com Signed-off-by:
-
