linux/mm/oom_kill.c

1145 lines
30 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0-only
/*
* linux/mm/oom_kill.c
*
* Copyright (C) 1998,2000 Rik van Riel
* Thanks go out to Claus Fischer for some serious inspiration and
* for goading me into coding this file...
oom: badness heuristic rewrite This a complete rewrite of the oom killer's badness() heuristic which is used to determine which task to kill in oom conditions. The goal is to make it as simple and predictable as possible so the results are better understood and we end up killing the task which will lead to the most memory freeing while still respecting the fine-tuning from userspace. Instead of basing the heuristic on mm->total_vm for each task, the task's rss and swap space is used instead. This is a better indication of the amount of memory that will be freeable if the oom killed task is chosen and subsequently exits. This helps specifically in cases where KDE or GNOME is chosen for oom kill on desktop systems instead of a memory hogging task. The baseline for the heuristic is a proportion of memory that each task is currently using in memory plus swap compared to the amount of "allowable" memory. "Allowable," in this sense, means the system-wide resources for unconstrained oom conditions, the set of mempolicy nodes, the mems attached to current's cpuset, or a memory controller's limit. The proportion is given on a scale of 0 (never kill) to 1000 (always kill), roughly meaning that if a task has a badness() score of 500 that the task consumes approximately 50% of allowable memory resident in RAM or in swap space. The proportion is always relative to the amount of "allowable" memory and not the total amount of RAM systemwide so that mempolicies and cpusets may operate in isolation; they shall not need to know the true size of the machine on which they are running if they are bound to a specific set of nodes or mems, respectively. Root tasks are given 3% extra memory just like __vm_enough_memory() provides in LSMs. In the event of two tasks consuming similar amounts of memory, it is generally better to save root's task. Because of the change in the badness() heuristic's baseline, it is also necessary to introduce a new user interface to tune it. It's not possible to redefine the meaning of /proc/pid/oom_adj with a new scale since the ABI cannot be changed for backward compatability. Instead, a new tunable, /proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may be used to polarize the heuristic such that certain tasks are never considered for oom kill while others may always be considered. The value is added directly into the badness() score so a value of -500, for example, means to discount 50% of its memory consumption in comparison to other tasks either on the system, bound to the mempolicy, in the cpuset, or sharing the same memory controller. /proc/pid/oom_adj is changed so that its meaning is rescaled into the units used by /proc/pid/oom_score_adj, and vice versa. Changing one of these per-task tunables will rescale the value of the other to an equivalent meaning. Although /proc/pid/oom_adj was originally defined as a bitshift on the badness score, it now shares the same linear growth as /proc/pid/oom_score_adj but with different granularity. This is required so the ABI is not broken with userspace applications and allows oom_adj to be deprecated for future removal. Signed-off-by: David Rientjes <rientjes@google.com> Cc: Nick Piggin <npiggin@suse.de> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Balbir Singh <balbir@in.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 08:19:46 +08:00
* Copyright (C) 2010 Google, Inc.
* Rewritten by David Rientjes
*
* The routines in this file are used to kill a process when
[PATCH] cpusets: oom_kill tweaks This patch series extends the use of the cpuset attribute 'mem_exclusive' to support cpuset configurations that: 1) allow GFP_KERNEL allocations to come from a potentially larger set of memory nodes than GFP_USER allocations, and 2) can constrain the oom killer to tasks running in cpusets in a specified subtree of the cpuset hierarchy. Here's an example usage scenario. For a few hours or more, a large NUMA system at a University is to be divided in two halves, with a bunch of student jobs running in half the system under some form of batch manager, and with a big research project running in the other half. Each of the student jobs is placed in a small cpuset, but should share the classic Unix time share facilities, such as buffered pages of files in /bin and /usr/lib. The big research project wants no interference whatsoever from the student jobs, and has highly tuned, unusual memory and i/o patterns that intend to make full use of all the main memory on the nodes available to it. In this example, we have two big sibling cpusets, one of which is further divided into a more dynamic set of child cpusets. We want kernel memory allocations constrained by the two big cpusets, and user allocations constrained by the smaller child cpusets where present. And we require that the oom killer not operate across the two halves of this system, or else the first time a student job runs amuck, the big research project will likely be first inline to get shot. Tweaking /proc/<pid>/oom_adj is not ideal -- if the big research project really does run amuck allocating memory, it should be shot, not some other task outside the research projects mem_exclusive cpuset. I propose to extend the use of the 'mem_exclusive' flag of cpusets to manage such scenarios. Let memory allocations for user space (GFP_USER) be constrained by a tasks current cpuset, but memory allocations for kernel space (GFP_KERNEL) by constrained by the nearest mem_exclusive ancestor of the current cpuset, even though kernel space allocations will still _prefer_ to remain within the current tasks cpuset, if memory is easily available. Let the oom killer be constrained to consider only tasks that are in overlapping mem_exclusive cpusets (it won't help much to kill a task that normally cannot allocate memory on any of the same nodes as the ones on which the current task can allocate.) The current constraints imposed on setting mem_exclusive are unchanged. A cpuset may only be mem_exclusive if its parent is also mem_exclusive, and a mem_exclusive cpuset may not overlap any of its siblings memory nodes. This patch was presented on linux-mm in early July 2005, though did not generate much feedback at that time. It has been built for a variety of arch's using cross tools, and built, booted and tested for function on SN2 (ia64). There are 4 patches in this set: 1) Some minor cleanup, and some improvements to the code layout of one routine to make subsequent patches cleaner. 2) Add another GFP flag - __GFP_HARDWALL. It marks memory requests for USER space, which are tightly confined by the current tasks cpuset. 3) Now memory requests (such as KERNEL) that not marked HARDWALL can if short on memory, look in the potentially larger pool of memory defined by the nearest mem_exclusive ancestor cpuset of the current tasks cpuset. 4) Finally, modify the oom killer to skip any task whose mem_exclusive cpuset doesn't overlap ours. Patch (1), the one time I looked on an SN2 (ia64) build, actually saved 32 bytes of kernel text space. Patch (2) has no affect on the size of kernel text space (it just adds a preprocessor flag). Patches (3) and (4) added about 600 bytes each of kernel text space, mostly in kernel/cpuset.c, which matters only if CONFIG_CPUSET is enabled. This patch: This patch applies a few comment and code cleanups to mm/oom_kill.c prior to applying a few small patches to improve cpuset management of memory placement. The comment changed in oom_kill.c was seriously misleading. The code layout change in select_bad_process() makes room for adding another condition on which a process can be spared the oom killer (see the subsequent cpuset_nodes_overlap patch for this addition). Also a couple typos and spellos that bugged me, while I was here. This patch should have no material affect. Signed-off-by: Paul Jackson <pj@sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-09-07 06:18:09 +08:00
* we're seriously out of memory. This gets called from __alloc_pages()
* in mm/page_alloc.c when we really run out of memory.
*
* Since we won't call these routines often (on a well-configured
* machine) this file will double as a 'coding guide' and a signpost
* for newbie kernel hackers. It features several pointers to major
* kernel subsystems and hints as to where to find out what things do.
*/
#include <linux/oom.h>
#include <linux/mm.h>
Remove fs.h from mm.h Remove fs.h from mm.h. For this, 1) Uninline vma_wants_writenotify(). It's pretty huge anyway. 2) Add back fs.h or less bloated headers (err.h) to files that need it. As result, on x86_64 allyesconfig, fs.h dependencies cut down from 3929 files rebuilt down to 3444 (-12.3%). Cross-compile tested without regressions on my two usual configs and (sigh): alpha arm-mx1ads mips-bigsur powerpc-ebony alpha-allnoconfig arm-neponset mips-capcella powerpc-g5 alpha-defconfig arm-netwinder mips-cobalt powerpc-holly alpha-up arm-netx mips-db1000 powerpc-iseries arm arm-ns9xxx mips-db1100 powerpc-linkstation arm-assabet arm-omap_h2_1610 mips-db1200 powerpc-lite5200 arm-at91rm9200dk arm-onearm mips-db1500 powerpc-maple arm-at91rm9200ek arm-picotux200 mips-db1550 powerpc-mpc7448_hpc2 arm-at91sam9260ek arm-pleb mips-ddb5477 powerpc-mpc8272_ads arm-at91sam9261ek arm-pnx4008 mips-decstation powerpc-mpc8313_rdb arm-at91sam9263ek arm-pxa255-idp mips-e55 powerpc-mpc832x_mds arm-at91sam9rlek arm-realview mips-emma2rh powerpc-mpc832x_rdb arm-ateb9200 arm-realview-smp mips-excite powerpc-mpc834x_itx arm-badge4 arm-rpc mips-fulong powerpc-mpc834x_itxgp arm-carmeva arm-s3c2410 mips-ip22 powerpc-mpc834x_mds arm-cerfcube arm-shannon mips-ip27 powerpc-mpc836x_mds arm-clps7500 arm-shark mips-ip32 powerpc-mpc8540_ads arm-collie arm-simpad mips-jazz powerpc-mpc8544_ds arm-corgi arm-spitz mips-jmr3927 powerpc-mpc8560_ads arm-csb337 arm-trizeps4 mips-malta powerpc-mpc8568mds arm-csb637 arm-versatile mips-mipssim powerpc-mpc85xx_cds arm-ebsa110 i386 mips-mpc30x powerpc-mpc8641_hpcn arm-edb7211 i386-allnoconfig mips-msp71xx powerpc-mpc866_ads arm-em_x270 i386-defconfig mips-ocelot powerpc-mpc885_ads arm-ep93xx i386-up mips-pb1100 powerpc-pasemi arm-footbridge ia64 mips-pb1500 powerpc-pmac32 arm-fortunet ia64-allnoconfig mips-pb1550 powerpc-ppc64 arm-h3600 ia64-bigsur mips-pnx8550-jbs powerpc-prpmc2800 arm-h7201 ia64-defconfig mips-pnx8550-stb810 powerpc-ps3 arm-h7202 ia64-gensparse mips-qemu powerpc-pseries arm-hackkit ia64-sim mips-rbhma4200 powerpc-up arm-integrator ia64-sn2 mips-rbhma4500 s390 arm-iop13xx ia64-tiger mips-rm200 s390-allnoconfig arm-iop32x ia64-up mips-sb1250-swarm s390-defconfig arm-iop33x ia64-zx1 mips-sead s390-up arm-ixp2000 m68k mips-tb0219 sparc arm-ixp23xx m68k-amiga mips-tb0226 sparc-allnoconfig arm-ixp4xx m68k-apollo mips-tb0287 sparc-defconfig arm-jornada720 m68k-atari mips-workpad sparc-up arm-kafa m68k-bvme6000 mips-wrppmc sparc64 arm-kb9202 m68k-hp300 mips-yosemite sparc64-allnoconfig arm-ks8695 m68k-mac parisc sparc64-defconfig arm-lart m68k-mvme147 parisc-allnoconfig sparc64-up arm-lpd270 m68k-mvme16x parisc-defconfig um-x86_64 arm-lpd7a400 m68k-q40 parisc-up x86_64 arm-lpd7a404 m68k-sun3 powerpc x86_64-allnoconfig arm-lubbock m68k-sun3x powerpc-cell x86_64-defconfig arm-lusl7200 mips powerpc-celleb x86_64-up arm-mainstone mips-atlas powerpc-chrp32 Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-30 06:36:13 +08:00
#include <linux/err.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 16:04:11 +08:00
#include <linux/gfp.h>
#include <linux/sched.h>
#include <linux/sched/mm.h>
#include <linux/sched/coredump.h>
#include <linux/sched/task.h>
#include <linux/sched/debug.h>
#include <linux/swap.h>
#include <linux/timex.h>
#include <linux/jiffies.h>
#include <linux/cpuset.h>
#include <linux/export.h>
#include <linux/notifier.h>
#include <linux/memcontrol.h>
#include <linux/mempolicy.h>
security: Fix setting of PF_SUPERPRIV by __capable() Fix the setting of PF_SUPERPRIV by __capable() as it could corrupt the flags the target process if that is not the current process and it is trying to change its own flags in a different way at the same time. __capable() is using neither atomic ops nor locking to protect t->flags. This patch removes __capable() and introduces has_capability() that doesn't set PF_SUPERPRIV on the process being queried. This patch further splits security_ptrace() in two: (1) security_ptrace_may_access(). This passes judgement on whether one process may access another only (PTRACE_MODE_ATTACH for ptrace() and PTRACE_MODE_READ for /proc), and takes a pointer to the child process. current is the parent. (2) security_ptrace_traceme(). This passes judgement on PTRACE_TRACEME only, and takes only a pointer to the parent process. current is the child. In Smack and commoncap, this uses has_capability() to determine whether the parent will be permitted to use PTRACE_ATTACH if normal checks fail. This does not set PF_SUPERPRIV. Two of the instances of __capable() actually only act on current, and so have been changed to calls to capable(). Of the places that were using __capable(): (1) The OOM killer calls __capable() thrice when weighing the killability of a process. All of these now use has_capability(). (2) cap_ptrace() and smack_ptrace() were using __capable() to check to see whether the parent was allowed to trace any process. As mentioned above, these have been split. For PTRACE_ATTACH and /proc, capable() is now used, and for PTRACE_TRACEME, has_capability() is used. (3) cap_safe_nice() only ever saw current, so now uses capable(). (4) smack_setprocattr() rejected accesses to tasks other than current just after calling __capable(), so the order of these two tests have been switched and capable() is used instead. (5) In smack_file_send_sigiotask(), we need to allow privileged processes to receive SIGIO on files they're manipulating. (6) In smack_task_wait(), we let a process wait for a privileged process, whether or not the process doing the waiting is privileged. I've tested this with the LTP SELinux and syscalls testscripts. Signed-off-by: David Howells <dhowells@redhat.com> Acked-by: Serge Hallyn <serue@us.ibm.com> Acked-by: Casey Schaufler <casey@schaufler-ca.com> Acked-by: Andrew G. Morgan <morgan@kernel.org> Acked-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: James Morris <jmorris@namei.org>
2008-08-14 18:37:28 +08:00
#include <linux/security.h>
oom: avoid deferring oom killer if exiting task is being traced The oom killer naturally defers killing anything if it finds an eligible task that is already exiting and has yet to detach its ->mm. This avoids unnecessarily killing tasks when one is already in the exit path and may free enough memory that the oom killer is no longer needed. This is detected by PF_EXITING since threads that have already detached its ->mm are no longer considered at all. The problem with always deferring when a thread is PF_EXITING, however, is that it may never actually exit when being traced, specifically if another task is tracing it with PTRACE_O_TRACEEXIT. The oom killer does not want to defer in this case since there is no guarantee that thread will ever exit without intervention. This patch will now only defer the oom killer when a thread is PF_EXITING and no ptracer has stopped its progress in the exit path. It also ensures that a child is sacrificed for the chosen parent only if it has a different ->mm as the comment implies: this ensures that the thread group leader is always targeted appropriately. Signed-off-by: David Rientjes <rientjes@google.com> Reported-by: Oleg Nesterov <oleg@redhat.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Hugh Dickins <hughd@google.com> Cc: Andrey Vagin <avagin@openvz.org> Cc: <stable@kernel.org> [2.6.38.x] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-03-23 07:30:12 +08:00
#include <linux/ptrace.h>
#include <linux/freezer.h>
tracepoint: add tracepoints for debugging oom_score_adj oom_score_adj is used for guarding processes from OOM-Killer. One of problem is that it's inherited at fork(). When a daemon set oom_score_adj and make children, it's hard to know where the value is set. This patch adds some tracepoints useful for debugging. This patch adds 3 trace points. - creating new task - renaming a task (exec) - set oom_score_adj To debug, users need to enable some trace pointer. Maybe filtering is useful as # EVENT=/sys/kernel/debug/tracing/events/task/ # echo "oom_score_adj != 0" > $EVENT/task_newtask/filter # echo "oom_score_adj != 0" > $EVENT/task_rename/filter # echo 1 > $EVENT/enable # EVENT=/sys/kernel/debug/tracing/events/oom/ # echo 1 > $EVENT/enable output will be like this. # grep oom /sys/kernel/debug/tracing/trace bash-7699 [007] d..3 5140.744510: oom_score_adj_update: pid=7699 comm=bash oom_score_adj=-1000 bash-7699 [007] ...1 5151.818022: task_newtask: pid=7729 comm=bash clone_flags=1200011 oom_score_adj=-1000 ls-7729 [003] ...2 5151.818504: task_rename: pid=7729 oldcomm=bash newcomm=ls oom_score_adj=-1000 bash-7699 [002] ...1 5175.701468: task_newtask: pid=7730 comm=bash clone_flags=1200011 oom_score_adj=-1000 grep-7730 [007] ...2 5175.701993: task_rename: pid=7730 oldcomm=bash newcomm=grep oom_score_adj=-1000 Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Acked-by: David Rientjes <rientjes@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-11 07:08:09 +08:00
#include <linux/ftrace.h>
#include <linux/ratelimit.h>
mm, oom: introduce oom reaper This patch (of 5): This is based on the idea from Mel Gorman discussed during LSFMM 2015 and independently brought up by Oleg Nesterov. The OOM killer currently allows to kill only a single task in a good hope that the task will terminate in a reasonable time and frees up its memory. Such a task (oom victim) will get an access to memory reserves via mark_oom_victim to allow a forward progress should there be a need for additional memory during exit path. It has been shown (e.g. by Tetsuo Handa) that it is not that hard to construct workloads which break the core assumption mentioned above and the OOM victim might take unbounded amount of time to exit because it might be blocked in the uninterruptible state waiting for an event (e.g. lock) which is blocked by another task looping in the page allocator. This patch reduces the probability of such a lockup by introducing a specialized kernel thread (oom_reaper) which tries to reclaim additional memory by preemptively reaping the anonymous or swapped out memory owned by the oom victim under an assumption that such a memory won't be needed when its owner is killed and kicked from the userspace anyway. There is one notable exception to this, though, if the OOM victim was in the process of coredumping the result would be incomplete. This is considered a reasonable constrain because the overall system health is more important than debugability of a particular application. A kernel thread has been chosen because we need a reliable way of invocation so workqueue context is not appropriate because all the workers might be busy (e.g. allocating memory). Kswapd which sounds like another good fit is not appropriate as well because it might get blocked on locks during reclaim as well. oom_reaper has to take mmap_sem on the target task for reading so the solution is not 100% because the semaphore might be held or blocked for write but the probability is reduced considerably wrt. basically any lock blocking forward progress as described above. In order to prevent from blocking on the lock without any forward progress we are using only a trylock and retry 10 times with a short sleep in between. Users of mmap_sem which need it for write should be carefully reviewed to use _killable waiting as much as possible and reduce allocations requests done with the lock held to absolute minimum to reduce the risk even further. The API between oom killer and oom reaper is quite trivial. wake_oom_reaper updates mm_to_reap with cmpxchg to guarantee only NULL->mm transition and oom_reaper clear this atomically once it is done with the work. This means that only a single mm_struct can be reaped at the time. As the operation is potentially disruptive we are trying to limit it to the ncessary minimum and the reaper blocks any updates while it operates on an mm. mm_struct is pinned by mm_count to allow parallel exit_mmap and a race is detected by atomic_inc_not_zero(mm_users). Signed-off-by: Michal Hocko <mhocko@suse.com> Suggested-by: Oleg Nesterov <oleg@redhat.com> Suggested-by: Mel Gorman <mgorman@suse.de> Acked-by: Mel Gorman <mgorman@suse.de> Acked-by: David Rientjes <rientjes@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Hugh Dickins <hughd@google.com> Cc: Andrea Argangeli <andrea@kernel.org> Cc: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:24 +08:00
#include <linux/kthread.h>
#include <linux/init.h>
mm, oom_reaper: skip mm structs with mmu notifiers Andrea has noticed that the oom_reaper doesn't invalidate the range via mmu notifiers (mmu_notifier_invalidate_range_start/end) and that can corrupt the memory of the kvm guest for example. tlb_flush_mmu_tlbonly already invokes mmu notifiers but that is not sufficient as per Andrea: "mmu_notifier_invalidate_range cannot be used in replacement of mmu_notifier_invalidate_range_start/end. For KVM mmu_notifier_invalidate_range is a noop and rightfully so. A MMU notifier implementation has to implement either ->invalidate_range method or the invalidate_range_start/end methods, not both. And if you implement invalidate_range_start/end like KVM is forced to do, calling mmu_notifier_invalidate_range in common code is a noop for KVM. For those MMU notifiers that can get away only implementing ->invalidate_range, the ->invalidate_range is implicitly called by mmu_notifier_invalidate_range_end(). And only those secondary MMUs that share the same pagetable with the primary MMU (like AMD iommuv2) can get away only implementing ->invalidate_range" As the callback is allowed to sleep and the implementation is out of hand of the MM it is safer to simply bail out if there is an mmu notifier registered. In order to not fail too early make the mm_has_notifiers check under the oom_lock and have a little nap before failing to give the current oom victim some more time to exit. [akpm@linux-foundation.org: coding-style fixes] Link: http://lkml.kernel.org/r/20170913113427.2291-1-mhocko@kernel.org Fixes: aac453635549 ("mm, oom: introduce oom reaper") Signed-off-by: Michal Hocko <mhocko@suse.com> Reported-by: Andrea Arcangeli <aarcange@redhat.com> Reviewed-by: Andrea Arcangeli <aarcange@redhat.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-10-04 07:14:50 +08:00
#include <linux/mmu_notifier.h>
mm, oom: introduce oom reaper This patch (of 5): This is based on the idea from Mel Gorman discussed during LSFMM 2015 and independently brought up by Oleg Nesterov. The OOM killer currently allows to kill only a single task in a good hope that the task will terminate in a reasonable time and frees up its memory. Such a task (oom victim) will get an access to memory reserves via mark_oom_victim to allow a forward progress should there be a need for additional memory during exit path. It has been shown (e.g. by Tetsuo Handa) that it is not that hard to construct workloads which break the core assumption mentioned above and the OOM victim might take unbounded amount of time to exit because it might be blocked in the uninterruptible state waiting for an event (e.g. lock) which is blocked by another task looping in the page allocator. This patch reduces the probability of such a lockup by introducing a specialized kernel thread (oom_reaper) which tries to reclaim additional memory by preemptively reaping the anonymous or swapped out memory owned by the oom victim under an assumption that such a memory won't be needed when its owner is killed and kicked from the userspace anyway. There is one notable exception to this, though, if the OOM victim was in the process of coredumping the result would be incomplete. This is considered a reasonable constrain because the overall system health is more important than debugability of a particular application. A kernel thread has been chosen because we need a reliable way of invocation so workqueue context is not appropriate because all the workers might be busy (e.g. allocating memory). Kswapd which sounds like another good fit is not appropriate as well because it might get blocked on locks during reclaim as well. oom_reaper has to take mmap_sem on the target task for reading so the solution is not 100% because the semaphore might be held or blocked for write but the probability is reduced considerably wrt. basically any lock blocking forward progress as described above. In order to prevent from blocking on the lock without any forward progress we are using only a trylock and retry 10 times with a short sleep in between. Users of mmap_sem which need it for write should be carefully reviewed to use _killable waiting as much as possible and reduce allocations requests done with the lock held to absolute minimum to reduce the risk even further. The API between oom killer and oom reaper is quite trivial. wake_oom_reaper updates mm_to_reap with cmpxchg to guarantee only NULL->mm transition and oom_reaper clear this atomically once it is done with the work. This means that only a single mm_struct can be reaped at the time. As the operation is potentially disruptive we are trying to limit it to the ncessary minimum and the reaper blocks any updates while it operates on an mm. mm_struct is pinned by mm_count to allow parallel exit_mmap and a race is detected by atomic_inc_not_zero(mm_users). Signed-off-by: Michal Hocko <mhocko@suse.com> Suggested-by: Oleg Nesterov <oleg@redhat.com> Suggested-by: Mel Gorman <mgorman@suse.de> Acked-by: Mel Gorman <mgorman@suse.de> Acked-by: David Rientjes <rientjes@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Hugh Dickins <hughd@google.com> Cc: Andrea Argangeli <andrea@kernel.org> Cc: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:24 +08:00
#include <asm/tlb.h>
#include "internal.h"
mm: oom: show unreclaimable slab info when unreclaimable slabs > user memory The kernel may panic when an oom happens without killable process sometimes it is caused by huge unreclaimable slabs used by kernel. Although kdump could help debug such problem, however, kdump is not available on all architectures and it might be malfunction sometime. And, since kernel already panic it is worthy capturing such information in dmesg to aid touble shooting. Print out unreclaimable slab info (used size and total size) which actual memory usage is not zero (num_objs * size != 0) when unreclaimable slabs amount is greater than total user memory (LRU pages). The output looks like: Unreclaimable slab info: Name Used Total rpc_buffers 31KB 31KB rpc_tasks 7KB 7KB ebitmap_node 1964KB 1964KB avtab_node 5024KB 5024KB xfs_buf 1402KB 1402KB xfs_ili 134KB 134KB xfs_efi_item 115KB 115KB xfs_efd_item 115KB 115KB xfs_buf_item 134KB 134KB xfs_log_item_desc 342KB 342KB xfs_trans 1412KB 1412KB xfs_ifork 212KB 212KB [yang.s@alibaba-inc.com: v11] Link: http://lkml.kernel.org/r/1507656303-103845-4-git-send-email-yang.s@alibaba-inc.com Link: http://lkml.kernel.org/r/1507152550-46205-4-git-send-email-yang.s@alibaba-inc.com Signed-off-by: Yang Shi <yang.s@alibaba-inc.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Christoph Lameter <cl@linux.com> Cc: David Rientjes <rientjes@google.com> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Pekka Enberg <penberg@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-16 09:32:07 +08:00
#include "slab.h"
tracepoint: add tracepoints for debugging oom_score_adj oom_score_adj is used for guarding processes from OOM-Killer. One of problem is that it's inherited at fork(). When a daemon set oom_score_adj and make children, it's hard to know where the value is set. This patch adds some tracepoints useful for debugging. This patch adds 3 trace points. - creating new task - renaming a task (exec) - set oom_score_adj To debug, users need to enable some trace pointer. Maybe filtering is useful as # EVENT=/sys/kernel/debug/tracing/events/task/ # echo "oom_score_adj != 0" > $EVENT/task_newtask/filter # echo "oom_score_adj != 0" > $EVENT/task_rename/filter # echo 1 > $EVENT/enable # EVENT=/sys/kernel/debug/tracing/events/oom/ # echo 1 > $EVENT/enable output will be like this. # grep oom /sys/kernel/debug/tracing/trace bash-7699 [007] d..3 5140.744510: oom_score_adj_update: pid=7699 comm=bash oom_score_adj=-1000 bash-7699 [007] ...1 5151.818022: task_newtask: pid=7729 comm=bash clone_flags=1200011 oom_score_adj=-1000 ls-7729 [003] ...2 5151.818504: task_rename: pid=7729 oldcomm=bash newcomm=ls oom_score_adj=-1000 bash-7699 [002] ...1 5175.701468: task_newtask: pid=7730 comm=bash clone_flags=1200011 oom_score_adj=-1000 grep-7730 [007] ...2 5175.701993: task_rename: pid=7730 oldcomm=bash newcomm=grep oom_score_adj=-1000 Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Acked-by: David Rientjes <rientjes@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-11 07:08:09 +08:00
#define CREATE_TRACE_POINTS
#include <trace/events/oom.h>
int sysctl_panic_on_oom;
int sysctl_oom_kill_allocating_task;
int sysctl_oom_dump_tasks = 1;
/*
* Serializes oom killer invocations (out_of_memory()) from all contexts to
* prevent from over eager oom killing (e.g. when the oom killer is invoked
* from different domains).
*
* oom_killer_disable() relies on this lock to stabilize oom_killer_disabled
* and mark_oom_victim
*/
DEFINE_MUTEX(oom_lock);
mm, oom_adj: don't loop through tasks in __set_oom_adj when not necessary Currently __set_oom_adj loops through all processes in the system to keep oom_score_adj and oom_score_adj_min in sync between processes sharing their mm. This is done for any task with more that one mm_users, which includes processes with multiple threads (sharing mm and signals). However for such processes the loop is unnecessary because their signal structure is shared as well. Android updates oom_score_adj whenever a tasks changes its role (background/foreground/...) or binds to/unbinds from a service, making it more/less important. Such operation can happen frequently. We noticed that updates to oom_score_adj became more expensive and after further investigation found out that the patch mentioned in "Fixes" introduced a regression. Using Pixel 4 with a typical Android workload, write time to oom_score_adj increased from ~3.57us to ~362us. Moreover this regression linearly depends on the number of multi-threaded processes running on the system. Mark the mm with a new MMF_MULTIPROCESS flag bit when task is created with (CLONE_VM && !CLONE_THREAD && !CLONE_VFORK). Change __set_oom_adj to use MMF_MULTIPROCESS instead of mm_users to decide whether oom_score_adj update should be synchronized between multiple processes. To prevent races between clone() and __set_oom_adj(), when oom_score_adj of the process being cloned might be modified from userspace, we use oom_adj_mutex. Its scope is changed to global. The combination of (CLONE_VM && !CLONE_THREAD) is rarely used except for the case of vfork(). To prevent performance regressions of vfork(), we skip taking oom_adj_mutex and setting MMF_MULTIPROCESS when CLONE_VFORK is specified. Clearing the MMF_MULTIPROCESS flag (when the last process sharing the mm exits) is left out of this patch to keep it simple and because it is believed that this threading model is rare. Should there ever be a need for optimizing that case as well, it can be done by hooking into the exit path, likely following the mm_update_next_owner pattern. With the combination of (CLONE_VM && !CLONE_THREAD && !CLONE_VFORK) being quite rare, the regression is gone after the change is applied. [surenb@google.com: v3] Link: https://lkml.kernel.org/r/20200902012558.2335613-1-surenb@google.com Fixes: 44a70adec910 ("mm, oom_adj: make sure processes sharing mm have same view of oom_score_adj") Reported-by: Tim Murray <timmurray@google.com> Suggested-by: Michal Hocko <mhocko@kernel.org> Signed-off-by: Suren Baghdasaryan <surenb@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Acked-by: Christian Brauner <christian.brauner@ubuntu.com> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: Oleg Nesterov <oleg@redhat.com> Cc: Ingo Molnar <mingo@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Eugene Syromiatnikov <esyr@redhat.com> Cc: Christian Kellner <christian@kellner.me> Cc: Adrian Reber <areber@redhat.com> Cc: Shakeel Butt <shakeelb@google.com> Cc: Aleksa Sarai <cyphar@cyphar.com> Cc: Alexey Dobriyan <adobriyan@gmail.com> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Cc: Alexey Gladkov <gladkov.alexey@gmail.com> Cc: Michel Lespinasse <walken@google.com> Cc: Daniel Jordan <daniel.m.jordan@oracle.com> Cc: Andrei Vagin <avagin@gmail.com> Cc: Bernd Edlinger <bernd.edlinger@hotmail.de> Cc: John Johansen <john.johansen@canonical.com> Cc: Yafang Shao <laoar.shao@gmail.com> Link: https://lkml.kernel.org/r/20200824153036.3201505-1-surenb@google.com Debugged-by: Minchan Kim <minchan@kernel.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-10-14 07:58:35 +08:00
/* Serializes oom_score_adj and oom_score_adj_min updates */
DEFINE_MUTEX(oom_adj_mutex);
oom: decouple mems_allowed from oom_unkillable_task Commit ef08e3b4981a ("[PATCH] cpusets: confine oom_killer to mem_exclusive cpuset") introduces a heuristic where a potential oom-killer victim is skipped if the intersection of the potential victim and the current (the process triggered the oom) is empty based on the reason that killing such victim most probably will not help the current allocating process. However the commit 7887a3da753e ("[PATCH] oom: cpuset hint") changed the heuristic to just decrease the oom_badness scores of such potential victim based on the reason that the cpuset of such processes might have changed and previously they may have allocated memory on mems where the current allocating process can allocate from. Unintentionally 7887a3da753e ("[PATCH] oom: cpuset hint") introduced a side effect as the oom_badness is also exposed to the user space through /proc/[pid]/oom_score, so, readers with different cpusets can read different oom_score of the same process. Later, commit 6cf86ac6f36b ("oom: filter tasks not sharing the same cpuset") fixed the side effect introduced by 7887a3da753e by moving the cpuset intersection back to only oom-killer context and out of oom_badness. However the combination of ab290adbaf8f ("oom: make oom_unkillable_task() helper function") and 26ebc984913b ("oom: /proc/<pid>/oom_score treat kernel thread honestly") unintentionally brought back the cpuset intersection check into the oom_badness calculation function. Other than doing cpuset/mempolicy intersection from oom_badness, the memcg oom context is also doing cpuset/mempolicy intersection which is quite wrong and is caught by syzcaller with the following report: kasan: CONFIG_KASAN_INLINE enabled kasan: GPF could be caused by NULL-ptr deref or user memory access general protection fault: 0000 [#1] PREEMPT SMP KASAN CPU: 0 PID: 28426 Comm: syz-executor.5 Not tainted 5.2.0-rc3-next-20190607 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:__read_once_size include/linux/compiler.h:194 [inline] RIP: 0010:has_intersects_mems_allowed mm/oom_kill.c:84 [inline] RIP: 0010:oom_unkillable_task mm/oom_kill.c:168 [inline] RIP: 0010:oom_unkillable_task+0x180/0x400 mm/oom_kill.c:155 Code: c1 ea 03 80 3c 02 00 0f 85 80 02 00 00 4c 8b a3 10 07 00 00 48 b8 00 00 00 00 00 fc ff df 4d 8d 74 24 10 4c 89 f2 48 c1 ea 03 <80> 3c 02 00 0f 85 67 02 00 00 49 8b 44 24 10 4c 8d a0 68 fa ff ff RSP: 0018:ffff888000127490 EFLAGS: 00010a03 RAX: dffffc0000000000 RBX: ffff8880a4cd5438 RCX: ffffffff818dae9c RDX: 100000000c3cc602 RSI: ffffffff818dac8d RDI: 0000000000000001 RBP: ffff8880001274d0 R08: ffff888000086180 R09: ffffed1015d26be0 R10: ffffed1015d26bdf R11: ffff8880ae935efb R12: 8000000061e63007 R13: 0000000000000000 R14: 8000000061e63017 R15: 1ffff11000024ea6 FS: 00005555561f5940(0000) GS:ffff8880ae800000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000607304 CR3: 000000009237e000 CR4: 00000000001426f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000600 Call Trace: oom_evaluate_task+0x49/0x520 mm/oom_kill.c:321 mem_cgroup_scan_tasks+0xcc/0x180 mm/memcontrol.c:1169 select_bad_process mm/oom_kill.c:374 [inline] out_of_memory mm/oom_kill.c:1088 [inline] out_of_memory+0x6b2/0x1280 mm/oom_kill.c:1035 mem_cgroup_out_of_memory+0x1ca/0x230 mm/memcontrol.c:1573 mem_cgroup_oom mm/memcontrol.c:1905 [inline] try_charge+0xfbe/0x1480 mm/memcontrol.c:2468 mem_cgroup_try_charge+0x24d/0x5e0 mm/memcontrol.c:6073 mem_cgroup_try_charge_delay+0x1f/0xa0 mm/memcontrol.c:6088 do_huge_pmd_wp_page_fallback+0x24f/0x1680 mm/huge_memory.c:1201 do_huge_pmd_wp_page+0x7fc/0x2160 mm/huge_memory.c:1359 wp_huge_pmd mm/memory.c:3793 [inline] __handle_mm_fault+0x164c/0x3eb0 mm/memory.c:4006 handle_mm_fault+0x3b7/0xa90 mm/memory.c:4053 do_user_addr_fault arch/x86/mm/fault.c:1455 [inline] __do_page_fault+0x5ef/0xda0 arch/x86/mm/fault.c:1521 do_page_fault+0x71/0x57d arch/x86/mm/fault.c:1552 page_fault+0x1e/0x30 arch/x86/entry/entry_64.S:1156 RIP: 0033:0x400590 Code: 06 e9 49 01 00 00 48 8b 44 24 10 48 0b 44 24 28 75 1f 48 8b 14 24 48 8b 7c 24 20 be 04 00 00 00 e8 f5 56 00 00 48 8b 74 24 08 <89> 06 e9 1e 01 00 00 48 8b 44 24 08 48 8b 14 24 be 04 00 00 00 8b RSP: 002b:00007fff7bc49780 EFLAGS: 00010206 RAX: 0000000000000001 RBX: 0000000000760000 RCX: 0000000000000000 RDX: 0000000000000000 RSI: 000000002000cffc RDI: 0000000000000001 RBP: fffffffffffffffe R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000075 R11: 0000000000000246 R12: 0000000000760008 R13: 00000000004c55f2 R14: 0000000000000000 R15: 00007fff7bc499b0 Modules linked in: ---[ end trace a65689219582ffff ]--- RIP: 0010:__read_once_size include/linux/compiler.h:194 [inline] RIP: 0010:has_intersects_mems_allowed mm/oom_kill.c:84 [inline] RIP: 0010:oom_unkillable_task mm/oom_kill.c:168 [inline] RIP: 0010:oom_unkillable_task+0x180/0x400 mm/oom_kill.c:155 Code: c1 ea 03 80 3c 02 00 0f 85 80 02 00 00 4c 8b a3 10 07 00 00 48 b8 00 00 00 00 00 fc ff df 4d 8d 74 24 10 4c 89 f2 48 c1 ea 03 <80> 3c 02 00 0f 85 67 02 00 00 49 8b 44 24 10 4c 8d a0 68 fa ff ff RSP: 0018:ffff888000127490 EFLAGS: 00010a03 RAX: dffffc0000000000 RBX: ffff8880a4cd5438 RCX: ffffffff818dae9c RDX: 100000000c3cc602 RSI: ffffffff818dac8d RDI: 0000000000000001 RBP: ffff8880001274d0 R08: ffff888000086180 R09: ffffed1015d26be0 R10: ffffed1015d26bdf R11: ffff8880ae935efb R12: 8000000061e63007 R13: 0000000000000000 R14: 8000000061e63017 R15: 1ffff11000024ea6 FS: 00005555561f5940(0000) GS:ffff8880ae800000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000001b2f823000 CR3: 000000009237e000 CR4: 00000000001426f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000600 The fix is to decouple the cpuset/mempolicy intersection check from oom_unkillable_task() and make sure cpuset/mempolicy intersection check is only done in the global oom context. [shakeelb@google.com: change function name and update comment] Link: http://lkml.kernel.org/r/20190628152421.198994-3-shakeelb@google.com Link: http://lkml.kernel.org/r/20190624212631.87212-3-shakeelb@google.com Signed-off-by: Shakeel Butt <shakeelb@google.com> Reported-by: syzbot+d0fc9d3c166bc5e4a94b@syzkaller.appspotmail.com Acked-by: Roman Gushchin <guro@fb.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Nick Piggin <npiggin@suse.de> Cc: Paul Jackson <pj@sgi.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-07-12 12:00:31 +08:00
static inline bool is_memcg_oom(struct oom_control *oc)
{
return oc->memcg != NULL;
}
#ifdef CONFIG_NUMA
/**
oom: decouple mems_allowed from oom_unkillable_task Commit ef08e3b4981a ("[PATCH] cpusets: confine oom_killer to mem_exclusive cpuset") introduces a heuristic where a potential oom-killer victim is skipped if the intersection of the potential victim and the current (the process triggered the oom) is empty based on the reason that killing such victim most probably will not help the current allocating process. However the commit 7887a3da753e ("[PATCH] oom: cpuset hint") changed the heuristic to just decrease the oom_badness scores of such potential victim based on the reason that the cpuset of such processes might have changed and previously they may have allocated memory on mems where the current allocating process can allocate from. Unintentionally 7887a3da753e ("[PATCH] oom: cpuset hint") introduced a side effect as the oom_badness is also exposed to the user space through /proc/[pid]/oom_score, so, readers with different cpusets can read different oom_score of the same process. Later, commit 6cf86ac6f36b ("oom: filter tasks not sharing the same cpuset") fixed the side effect introduced by 7887a3da753e by moving the cpuset intersection back to only oom-killer context and out of oom_badness. However the combination of ab290adbaf8f ("oom: make oom_unkillable_task() helper function") and 26ebc984913b ("oom: /proc/<pid>/oom_score treat kernel thread honestly") unintentionally brought back the cpuset intersection check into the oom_badness calculation function. Other than doing cpuset/mempolicy intersection from oom_badness, the memcg oom context is also doing cpuset/mempolicy intersection which is quite wrong and is caught by syzcaller with the following report: kasan: CONFIG_KASAN_INLINE enabled kasan: GPF could be caused by NULL-ptr deref or user memory access general protection fault: 0000 [#1] PREEMPT SMP KASAN CPU: 0 PID: 28426 Comm: syz-executor.5 Not tainted 5.2.0-rc3-next-20190607 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:__read_once_size include/linux/compiler.h:194 [inline] RIP: 0010:has_intersects_mems_allowed mm/oom_kill.c:84 [inline] RIP: 0010:oom_unkillable_task mm/oom_kill.c:168 [inline] RIP: 0010:oom_unkillable_task+0x180/0x400 mm/oom_kill.c:155 Code: c1 ea 03 80 3c 02 00 0f 85 80 02 00 00 4c 8b a3 10 07 00 00 48 b8 00 00 00 00 00 fc ff df 4d 8d 74 24 10 4c 89 f2 48 c1 ea 03 <80> 3c 02 00 0f 85 67 02 00 00 49 8b 44 24 10 4c 8d a0 68 fa ff ff RSP: 0018:ffff888000127490 EFLAGS: 00010a03 RAX: dffffc0000000000 RBX: ffff8880a4cd5438 RCX: ffffffff818dae9c RDX: 100000000c3cc602 RSI: ffffffff818dac8d RDI: 0000000000000001 RBP: ffff8880001274d0 R08: ffff888000086180 R09: ffffed1015d26be0 R10: ffffed1015d26bdf R11: ffff8880ae935efb R12: 8000000061e63007 R13: 0000000000000000 R14: 8000000061e63017 R15: 1ffff11000024ea6 FS: 00005555561f5940(0000) GS:ffff8880ae800000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000607304 CR3: 000000009237e000 CR4: 00000000001426f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000600 Call Trace: oom_evaluate_task+0x49/0x520 mm/oom_kill.c:321 mem_cgroup_scan_tasks+0xcc/0x180 mm/memcontrol.c:1169 select_bad_process mm/oom_kill.c:374 [inline] out_of_memory mm/oom_kill.c:1088 [inline] out_of_memory+0x6b2/0x1280 mm/oom_kill.c:1035 mem_cgroup_out_of_memory+0x1ca/0x230 mm/memcontrol.c:1573 mem_cgroup_oom mm/memcontrol.c:1905 [inline] try_charge+0xfbe/0x1480 mm/memcontrol.c:2468 mem_cgroup_try_charge+0x24d/0x5e0 mm/memcontrol.c:6073 mem_cgroup_try_charge_delay+0x1f/0xa0 mm/memcontrol.c:6088 do_huge_pmd_wp_page_fallback+0x24f/0x1680 mm/huge_memory.c:1201 do_huge_pmd_wp_page+0x7fc/0x2160 mm/huge_memory.c:1359 wp_huge_pmd mm/memory.c:3793 [inline] __handle_mm_fault+0x164c/0x3eb0 mm/memory.c:4006 handle_mm_fault+0x3b7/0xa90 mm/memory.c:4053 do_user_addr_fault arch/x86/mm/fault.c:1455 [inline] __do_page_fault+0x5ef/0xda0 arch/x86/mm/fault.c:1521 do_page_fault+0x71/0x57d arch/x86/mm/fault.c:1552 page_fault+0x1e/0x30 arch/x86/entry/entry_64.S:1156 RIP: 0033:0x400590 Code: 06 e9 49 01 00 00 48 8b 44 24 10 48 0b 44 24 28 75 1f 48 8b 14 24 48 8b 7c 24 20 be 04 00 00 00 e8 f5 56 00 00 48 8b 74 24 08 <89> 06 e9 1e 01 00 00 48 8b 44 24 08 48 8b 14 24 be 04 00 00 00 8b RSP: 002b:00007fff7bc49780 EFLAGS: 00010206 RAX: 0000000000000001 RBX: 0000000000760000 RCX: 0000000000000000 RDX: 0000000000000000 RSI: 000000002000cffc RDI: 0000000000000001 RBP: fffffffffffffffe R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000075 R11: 0000000000000246 R12: 0000000000760008 R13: 00000000004c55f2 R14: 0000000000000000 R15: 00007fff7bc499b0 Modules linked in: ---[ end trace a65689219582ffff ]--- RIP: 0010:__read_once_size include/linux/compiler.h:194 [inline] RIP: 0010:has_intersects_mems_allowed mm/oom_kill.c:84 [inline] RIP: 0010:oom_unkillable_task mm/oom_kill.c:168 [inline] RIP: 0010:oom_unkillable_task+0x180/0x400 mm/oom_kill.c:155 Code: c1 ea 03 80 3c 02 00 0f 85 80 02 00 00 4c 8b a3 10 07 00 00 48 b8 00 00 00 00 00 fc ff df 4d 8d 74 24 10 4c 89 f2 48 c1 ea 03 <80> 3c 02 00 0f 85 67 02 00 00 49 8b 44 24 10 4c 8d a0 68 fa ff ff RSP: 0018:ffff888000127490 EFLAGS: 00010a03 RAX: dffffc0000000000 RBX: ffff8880a4cd5438 RCX: ffffffff818dae9c RDX: 100000000c3cc602 RSI: ffffffff818dac8d RDI: 0000000000000001 RBP: ffff8880001274d0 R08: ffff888000086180 R09: ffffed1015d26be0 R10: ffffed1015d26bdf R11: ffff8880ae935efb R12: 8000000061e63007 R13: 0000000000000000 R14: 8000000061e63017 R15: 1ffff11000024ea6 FS: 00005555561f5940(0000) GS:ffff8880ae800000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000001b2f823000 CR3: 000000009237e000 CR4: 00000000001426f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000600 The fix is to decouple the cpuset/mempolicy intersection check from oom_unkillable_task() and make sure cpuset/mempolicy intersection check is only done in the global oom context. [shakeelb@google.com: change function name and update comment] Link: http://lkml.kernel.org/r/20190628152421.198994-3-shakeelb@google.com Link: http://lkml.kernel.org/r/20190624212631.87212-3-shakeelb@google.com Signed-off-by: Shakeel Butt <shakeelb@google.com> Reported-by: syzbot+d0fc9d3c166bc5e4a94b@syzkaller.appspotmail.com Acked-by: Roman Gushchin <guro@fb.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Nick Piggin <npiggin@suse.de> Cc: Paul Jackson <pj@sgi.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-07-12 12:00:31 +08:00
* oom_cpuset_eligible() - check task eligiblity for kill
* @start: task struct of which task to consider
* @oc: pointer to struct oom_control
*
* Task eligibility is determined by whether or not a candidate task, @tsk,
* shares the same mempolicy nodes as current if it is bound by such a policy
* and whether or not it has the same set of allowed cpuset nodes.
oom: decouple mems_allowed from oom_unkillable_task Commit ef08e3b4981a ("[PATCH] cpusets: confine oom_killer to mem_exclusive cpuset") introduces a heuristic where a potential oom-killer victim is skipped if the intersection of the potential victim and the current (the process triggered the oom) is empty based on the reason that killing such victim most probably will not help the current allocating process. However the commit 7887a3da753e ("[PATCH] oom: cpuset hint") changed the heuristic to just decrease the oom_badness scores of such potential victim based on the reason that the cpuset of such processes might have changed and previously they may have allocated memory on mems where the current allocating process can allocate from. Unintentionally 7887a3da753e ("[PATCH] oom: cpuset hint") introduced a side effect as the oom_badness is also exposed to the user space through /proc/[pid]/oom_score, so, readers with different cpusets can read different oom_score of the same process. Later, commit 6cf86ac6f36b ("oom: filter tasks not sharing the same cpuset") fixed the side effect introduced by 7887a3da753e by moving the cpuset intersection back to only oom-killer context and out of oom_badness. However the combination of ab290adbaf8f ("oom: make oom_unkillable_task() helper function") and 26ebc984913b ("oom: /proc/<pid>/oom_score treat kernel thread honestly") unintentionally brought back the cpuset intersection check into the oom_badness calculation function. Other than doing cpuset/mempolicy intersection from oom_badness, the memcg oom context is also doing cpuset/mempolicy intersection which is quite wrong and is caught by syzcaller with the following report: kasan: CONFIG_KASAN_INLINE enabled kasan: GPF could be caused by NULL-ptr deref or user memory access general protection fault: 0000 [#1] PREEMPT SMP KASAN CPU: 0 PID: 28426 Comm: syz-executor.5 Not tainted 5.2.0-rc3-next-20190607 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:__read_once_size include/linux/compiler.h:194 [inline] RIP: 0010:has_intersects_mems_allowed mm/oom_kill.c:84 [inline] RIP: 0010:oom_unkillable_task mm/oom_kill.c:168 [inline] RIP: 0010:oom_unkillable_task+0x180/0x400 mm/oom_kill.c:155 Code: c1 ea 03 80 3c 02 00 0f 85 80 02 00 00 4c 8b a3 10 07 00 00 48 b8 00 00 00 00 00 fc ff df 4d 8d 74 24 10 4c 89 f2 48 c1 ea 03 <80> 3c 02 00 0f 85 67 02 00 00 49 8b 44 24 10 4c 8d a0 68 fa ff ff RSP: 0018:ffff888000127490 EFLAGS: 00010a03 RAX: dffffc0000000000 RBX: ffff8880a4cd5438 RCX: ffffffff818dae9c RDX: 100000000c3cc602 RSI: ffffffff818dac8d RDI: 0000000000000001 RBP: ffff8880001274d0 R08: ffff888000086180 R09: ffffed1015d26be0 R10: ffffed1015d26bdf R11: ffff8880ae935efb R12: 8000000061e63007 R13: 0000000000000000 R14: 8000000061e63017 R15: 1ffff11000024ea6 FS: 00005555561f5940(0000) GS:ffff8880ae800000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000607304 CR3: 000000009237e000 CR4: 00000000001426f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000600 Call Trace: oom_evaluate_task+0x49/0x520 mm/oom_kill.c:321 mem_cgroup_scan_tasks+0xcc/0x180 mm/memcontrol.c:1169 select_bad_process mm/oom_kill.c:374 [inline] out_of_memory mm/oom_kill.c:1088 [inline] out_of_memory+0x6b2/0x1280 mm/oom_kill.c:1035 mem_cgroup_out_of_memory+0x1ca/0x230 mm/memcontrol.c:1573 mem_cgroup_oom mm/memcontrol.c:1905 [inline] try_charge+0xfbe/0x1480 mm/memcontrol.c:2468 mem_cgroup_try_charge+0x24d/0x5e0 mm/memcontrol.c:6073 mem_cgroup_try_charge_delay+0x1f/0xa0 mm/memcontrol.c:6088 do_huge_pmd_wp_page_fallback+0x24f/0x1680 mm/huge_memory.c:1201 do_huge_pmd_wp_page+0x7fc/0x2160 mm/huge_memory.c:1359 wp_huge_pmd mm/memory.c:3793 [inline] __handle_mm_fault+0x164c/0x3eb0 mm/memory.c:4006 handle_mm_fault+0x3b7/0xa90 mm/memory.c:4053 do_user_addr_fault arch/x86/mm/fault.c:1455 [inline] __do_page_fault+0x5ef/0xda0 arch/x86/mm/fault.c:1521 do_page_fault+0x71/0x57d arch/x86/mm/fault.c:1552 page_fault+0x1e/0x30 arch/x86/entry/entry_64.S:1156 RIP: 0033:0x400590 Code: 06 e9 49 01 00 00 48 8b 44 24 10 48 0b 44 24 28 75 1f 48 8b 14 24 48 8b 7c 24 20 be 04 00 00 00 e8 f5 56 00 00 48 8b 74 24 08 <89> 06 e9 1e 01 00 00 48 8b 44 24 08 48 8b 14 24 be 04 00 00 00 8b RSP: 002b:00007fff7bc49780 EFLAGS: 00010206 RAX: 0000000000000001 RBX: 0000000000760000 RCX: 0000000000000000 RDX: 0000000000000000 RSI: 000000002000cffc RDI: 0000000000000001 RBP: fffffffffffffffe R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000075 R11: 0000000000000246 R12: 0000000000760008 R13: 00000000004c55f2 R14: 0000000000000000 R15: 00007fff7bc499b0 Modules linked in: ---[ end trace a65689219582ffff ]--- RIP: 0010:__read_once_size include/linux/compiler.h:194 [inline] RIP: 0010:has_intersects_mems_allowed mm/oom_kill.c:84 [inline] RIP: 0010:oom_unkillable_task mm/oom_kill.c:168 [inline] RIP: 0010:oom_unkillable_task+0x180/0x400 mm/oom_kill.c:155 Code: c1 ea 03 80 3c 02 00 0f 85 80 02 00 00 4c 8b a3 10 07 00 00 48 b8 00 00 00 00 00 fc ff df 4d 8d 74 24 10 4c 89 f2 48 c1 ea 03 <80> 3c 02 00 0f 85 67 02 00 00 49 8b 44 24 10 4c 8d a0 68 fa ff ff RSP: 0018:ffff888000127490 EFLAGS: 00010a03 RAX: dffffc0000000000 RBX: ffff8880a4cd5438 RCX: ffffffff818dae9c RDX: 100000000c3cc602 RSI: ffffffff818dac8d RDI: 0000000000000001 RBP: ffff8880001274d0 R08: ffff888000086180 R09: ffffed1015d26be0 R10: ffffed1015d26bdf R11: ffff8880ae935efb R12: 8000000061e63007 R13: 0000000000000000 R14: 8000000061e63017 R15: 1ffff11000024ea6 FS: 00005555561f5940(0000) GS:ffff8880ae800000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000001b2f823000 CR3: 000000009237e000 CR4: 00000000001426f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000600 The fix is to decouple the cpuset/mempolicy intersection check from oom_unkillable_task() and make sure cpuset/mempolicy intersection check is only done in the global oom context. [shakeelb@google.com: change function name and update comment] Link: http://lkml.kernel.org/r/20190628152421.198994-3-shakeelb@google.com Link: http://lkml.kernel.org/r/20190624212631.87212-3-shakeelb@google.com Signed-off-by: Shakeel Butt <shakeelb@google.com> Reported-by: syzbot+d0fc9d3c166bc5e4a94b@syzkaller.appspotmail.com Acked-by: Roman Gushchin <guro@fb.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Nick Piggin <npiggin@suse.de> Cc: Paul Jackson <pj@sgi.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-07-12 12:00:31 +08:00
*
* This function is assuming oom-killer context and 'current' has triggered
* the oom-killer.
*/
oom: decouple mems_allowed from oom_unkillable_task Commit ef08e3b4981a ("[PATCH] cpusets: confine oom_killer to mem_exclusive cpuset") introduces a heuristic where a potential oom-killer victim is skipped if the intersection of the potential victim and the current (the process triggered the oom) is empty based on the reason that killing such victim most probably will not help the current allocating process. However the commit 7887a3da753e ("[PATCH] oom: cpuset hint") changed the heuristic to just decrease the oom_badness scores of such potential victim based on the reason that the cpuset of such processes might have changed and previously they may have allocated memory on mems where the current allocating process can allocate from. Unintentionally 7887a3da753e ("[PATCH] oom: cpuset hint") introduced a side effect as the oom_badness is also exposed to the user space through /proc/[pid]/oom_score, so, readers with different cpusets can read different oom_score of the same process. Later, commit 6cf86ac6f36b ("oom: filter tasks not sharing the same cpuset") fixed the side effect introduced by 7887a3da753e by moving the cpuset intersection back to only oom-killer context and out of oom_badness. However the combination of ab290adbaf8f ("oom: make oom_unkillable_task() helper function") and 26ebc984913b ("oom: /proc/<pid>/oom_score treat kernel thread honestly") unintentionally brought back the cpuset intersection check into the oom_badness calculation function. Other than doing cpuset/mempolicy intersection from oom_badness, the memcg oom context is also doing cpuset/mempolicy intersection which is quite wrong and is caught by syzcaller with the following report: kasan: CONFIG_KASAN_INLINE enabled kasan: GPF could be caused by NULL-ptr deref or user memory access general protection fault: 0000 [#1] PREEMPT SMP KASAN CPU: 0 PID: 28426 Comm: syz-executor.5 Not tainted 5.2.0-rc3-next-20190607 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:__read_once_size include/linux/compiler.h:194 [inline] RIP: 0010:has_intersects_mems_allowed mm/oom_kill.c:84 [inline] RIP: 0010:oom_unkillable_task mm/oom_kill.c:168 [inline] RIP: 0010:oom_unkillable_task+0x180/0x400 mm/oom_kill.c:155 Code: c1 ea 03 80 3c 02 00 0f 85 80 02 00 00 4c 8b a3 10 07 00 00 48 b8 00 00 00 00 00 fc ff df 4d 8d 74 24 10 4c 89 f2 48 c1 ea 03 <80> 3c 02 00 0f 85 67 02 00 00 49 8b 44 24 10 4c 8d a0 68 fa ff ff RSP: 0018:ffff888000127490 EFLAGS: 00010a03 RAX: dffffc0000000000 RBX: ffff8880a4cd5438 RCX: ffffffff818dae9c RDX: 100000000c3cc602 RSI: ffffffff818dac8d RDI: 0000000000000001 RBP: ffff8880001274d0 R08: ffff888000086180 R09: ffffed1015d26be0 R10: ffffed1015d26bdf R11: ffff8880ae935efb R12: 8000000061e63007 R13: 0000000000000000 R14: 8000000061e63017 R15: 1ffff11000024ea6 FS: 00005555561f5940(0000) GS:ffff8880ae800000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000607304 CR3: 000000009237e000 CR4: 00000000001426f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000600 Call Trace: oom_evaluate_task+0x49/0x520 mm/oom_kill.c:321 mem_cgroup_scan_tasks+0xcc/0x180 mm/memcontrol.c:1169 select_bad_process mm/oom_kill.c:374 [inline] out_of_memory mm/oom_kill.c:1088 [inline] out_of_memory+0x6b2/0x1280 mm/oom_kill.c:1035 mem_cgroup_out_of_memory+0x1ca/0x230 mm/memcontrol.c:1573 mem_cgroup_oom mm/memcontrol.c:1905 [inline] try_charge+0xfbe/0x1480 mm/memcontrol.c:2468 mem_cgroup_try_charge+0x24d/0x5e0 mm/memcontrol.c:6073 mem_cgroup_try_charge_delay+0x1f/0xa0 mm/memcontrol.c:6088 do_huge_pmd_wp_page_fallback+0x24f/0x1680 mm/huge_memory.c:1201 do_huge_pmd_wp_page+0x7fc/0x2160 mm/huge_memory.c:1359 wp_huge_pmd mm/memory.c:3793 [inline] __handle_mm_fault+0x164c/0x3eb0 mm/memory.c:4006 handle_mm_fault+0x3b7/0xa90 mm/memory.c:4053 do_user_addr_fault arch/x86/mm/fault.c:1455 [inline] __do_page_fault+0x5ef/0xda0 arch/x86/mm/fault.c:1521 do_page_fault+0x71/0x57d arch/x86/mm/fault.c:1552 page_fault+0x1e/0x30 arch/x86/entry/entry_64.S:1156 RIP: 0033:0x400590 Code: 06 e9 49 01 00 00 48 8b 44 24 10 48 0b 44 24 28 75 1f 48 8b 14 24 48 8b 7c 24 20 be 04 00 00 00 e8 f5 56 00 00 48 8b 74 24 08 <89> 06 e9 1e 01 00 00 48 8b 44 24 08 48 8b 14 24 be 04 00 00 00 8b RSP: 002b:00007fff7bc49780 EFLAGS: 00010206 RAX: 0000000000000001 RBX: 0000000000760000 RCX: 0000000000000000 RDX: 0000000000000000 RSI: 000000002000cffc RDI: 0000000000000001 RBP: fffffffffffffffe R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000075 R11: 0000000000000246 R12: 0000000000760008 R13: 00000000004c55f2 R14: 0000000000000000 R15: 00007fff7bc499b0 Modules linked in: ---[ end trace a65689219582ffff ]--- RIP: 0010:__read_once_size include/linux/compiler.h:194 [inline] RIP: 0010:has_intersects_mems_allowed mm/oom_kill.c:84 [inline] RIP: 0010:oom_unkillable_task mm/oom_kill.c:168 [inline] RIP: 0010:oom_unkillable_task+0x180/0x400 mm/oom_kill.c:155 Code: c1 ea 03 80 3c 02 00 0f 85 80 02 00 00 4c 8b a3 10 07 00 00 48 b8 00 00 00 00 00 fc ff df 4d 8d 74 24 10 4c 89 f2 48 c1 ea 03 <80> 3c 02 00 0f 85 67 02 00 00 49 8b 44 24 10 4c 8d a0 68 fa ff ff RSP: 0018:ffff888000127490 EFLAGS: 00010a03 RAX: dffffc0000000000 RBX: ffff8880a4cd5438 RCX: ffffffff818dae9c RDX: 100000000c3cc602 RSI: ffffffff818dac8d RDI: 0000000000000001 RBP: ffff8880001274d0 R08: ffff888000086180 R09: ffffed1015d26be0 R10: ffffed1015d26bdf R11: ffff8880ae935efb R12: 8000000061e63007 R13: 0000000000000000 R14: 8000000061e63017 R15: 1ffff11000024ea6 FS: 00005555561f5940(0000) GS:ffff8880ae800000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000001b2f823000 CR3: 000000009237e000 CR4: 00000000001426f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000600 The fix is to decouple the cpuset/mempolicy intersection check from oom_unkillable_task() and make sure cpuset/mempolicy intersection check is only done in the global oom context. [shakeelb@google.com: change function name and update comment] Link: http://lkml.kernel.org/r/20190628152421.198994-3-shakeelb@google.com Link: http://lkml.kernel.org/r/20190624212631.87212-3-shakeelb@google.com Signed-off-by: Shakeel Butt <shakeelb@google.com> Reported-by: syzbot+d0fc9d3c166bc5e4a94b@syzkaller.appspotmail.com Acked-by: Roman Gushchin <guro@fb.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Nick Piggin <npiggin@suse.de> Cc: Paul Jackson <pj@sgi.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-07-12 12:00:31 +08:00
static bool oom_cpuset_eligible(struct task_struct *start,
struct oom_control *oc)
{
struct task_struct *tsk;
bool ret = false;
oom: decouple mems_allowed from oom_unkillable_task Commit ef08e3b4981a ("[PATCH] cpusets: confine oom_killer to mem_exclusive cpuset") introduces a heuristic where a potential oom-killer victim is skipped if the intersection of the potential victim and the current (the process triggered the oom) is empty based on the reason that killing such victim most probably will not help the current allocating process. However the commit 7887a3da753e ("[PATCH] oom: cpuset hint") changed the heuristic to just decrease the oom_badness scores of such potential victim based on the reason that the cpuset of such processes might have changed and previously they may have allocated memory on mems where the current allocating process can allocate from. Unintentionally 7887a3da753e ("[PATCH] oom: cpuset hint") introduced a side effect as the oom_badness is also exposed to the user space through /proc/[pid]/oom_score, so, readers with different cpusets can read different oom_score of the same process. Later, commit 6cf86ac6f36b ("oom: filter tasks not sharing the same cpuset") fixed the side effect introduced by 7887a3da753e by moving the cpuset intersection back to only oom-killer context and out of oom_badness. However the combination of ab290adbaf8f ("oom: make oom_unkillable_task() helper function") and 26ebc984913b ("oom: /proc/<pid>/oom_score treat kernel thread honestly") unintentionally brought back the cpuset intersection check into the oom_badness calculation function. Other than doing cpuset/mempolicy intersection from oom_badness, the memcg oom context is also doing cpuset/mempolicy intersection which is quite wrong and is caught by syzcaller with the following report: kasan: CONFIG_KASAN_INLINE enabled kasan: GPF could be caused by NULL-ptr deref or user memory access general protection fault: 0000 [#1] PREEMPT SMP KASAN CPU: 0 PID: 28426 Comm: syz-executor.5 Not tainted 5.2.0-rc3-next-20190607 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:__read_once_size include/linux/compiler.h:194 [inline] RIP: 0010:has_intersects_mems_allowed mm/oom_kill.c:84 [inline] RIP: 0010:oom_unkillable_task mm/oom_kill.c:168 [inline] RIP: 0010:oom_unkillable_task+0x180/0x400 mm/oom_kill.c:155 Code: c1 ea 03 80 3c 02 00 0f 85 80 02 00 00 4c 8b a3 10 07 00 00 48 b8 00 00 00 00 00 fc ff df 4d 8d 74 24 10 4c 89 f2 48 c1 ea 03 <80> 3c 02 00 0f 85 67 02 00 00 49 8b 44 24 10 4c 8d a0 68 fa ff ff RSP: 0018:ffff888000127490 EFLAGS: 00010a03 RAX: dffffc0000000000 RBX: ffff8880a4cd5438 RCX: ffffffff818dae9c RDX: 100000000c3cc602 RSI: ffffffff818dac8d RDI: 0000000000000001 RBP: ffff8880001274d0 R08: ffff888000086180 R09: ffffed1015d26be0 R10: ffffed1015d26bdf R11: ffff8880ae935efb R12: 8000000061e63007 R13: 0000000000000000 R14: 8000000061e63017 R15: 1ffff11000024ea6 FS: 00005555561f5940(0000) GS:ffff8880ae800000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000607304 CR3: 000000009237e000 CR4: 00000000001426f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000600 Call Trace: oom_evaluate_task+0x49/0x520 mm/oom_kill.c:321 mem_cgroup_scan_tasks+0xcc/0x180 mm/memcontrol.c:1169 select_bad_process mm/oom_kill.c:374 [inline] out_of_memory mm/oom_kill.c:1088 [inline] out_of_memory+0x6b2/0x1280 mm/oom_kill.c:1035 mem_cgroup_out_of_memory+0x1ca/0x230 mm/memcontrol.c:1573 mem_cgroup_oom mm/memcontrol.c:1905 [inline] try_charge+0xfbe/0x1480 mm/memcontrol.c:2468 mem_cgroup_try_charge+0x24d/0x5e0 mm/memcontrol.c:6073 mem_cgroup_try_charge_delay+0x1f/0xa0 mm/memcontrol.c:6088 do_huge_pmd_wp_page_fallback+0x24f/0x1680 mm/huge_memory.c:1201 do_huge_pmd_wp_page+0x7fc/0x2160 mm/huge_memory.c:1359 wp_huge_pmd mm/memory.c:3793 [inline] __handle_mm_fault+0x164c/0x3eb0 mm/memory.c:4006 handle_mm_fault+0x3b7/0xa90 mm/memory.c:4053 do_user_addr_fault arch/x86/mm/fault.c:1455 [inline] __do_page_fault+0x5ef/0xda0 arch/x86/mm/fault.c:1521 do_page_fault+0x71/0x57d arch/x86/mm/fault.c:1552 page_fault+0x1e/0x30 arch/x86/entry/entry_64.S:1156 RIP: 0033:0x400590 Code: 06 e9 49 01 00 00 48 8b 44 24 10 48 0b 44 24 28 75 1f 48 8b 14 24 48 8b 7c 24 20 be 04 00 00 00 e8 f5 56 00 00 48 8b 74 24 08 <89> 06 e9 1e 01 00 00 48 8b 44 24 08 48 8b 14 24 be 04 00 00 00 8b RSP: 002b:00007fff7bc49780 EFLAGS: 00010206 RAX: 0000000000000001 RBX: 0000000000760000 RCX: 0000000000000000 RDX: 0000000000000000 RSI: 000000002000cffc RDI: 0000000000000001 RBP: fffffffffffffffe R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000075 R11: 0000000000000246 R12: 0000000000760008 R13: 00000000004c55f2 R14: 0000000000000000 R15: 00007fff7bc499b0 Modules linked in: ---[ end trace a65689219582ffff ]--- RIP: 0010:__read_once_size include/linux/compiler.h:194 [inline] RIP: 0010:has_intersects_mems_allowed mm/oom_kill.c:84 [inline] RIP: 0010:oom_unkillable_task mm/oom_kill.c:168 [inline] RIP: 0010:oom_unkillable_task+0x180/0x400 mm/oom_kill.c:155 Code: c1 ea 03 80 3c 02 00 0f 85 80 02 00 00 4c 8b a3 10 07 00 00 48 b8 00 00 00 00 00 fc ff df 4d 8d 74 24 10 4c 89 f2 48 c1 ea 03 <80> 3c 02 00 0f 85 67 02 00 00 49 8b 44 24 10 4c 8d a0 68 fa ff ff RSP: 0018:ffff888000127490 EFLAGS: 00010a03 RAX: dffffc0000000000 RBX: ffff8880a4cd5438 RCX: ffffffff818dae9c RDX: 100000000c3cc602 RSI: ffffffff818dac8d RDI: 0000000000000001 RBP: ffff8880001274d0 R08: ffff888000086180 R09: ffffed1015d26be0 R10: ffffed1015d26bdf R11: ffff8880ae935efb R12: 8000000061e63007 R13: 0000000000000000 R14: 8000000061e63017 R15: 1ffff11000024ea6 FS: 00005555561f5940(0000) GS:ffff8880ae800000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000001b2f823000 CR3: 000000009237e000 CR4: 00000000001426f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000600 The fix is to decouple the cpuset/mempolicy intersection check from oom_unkillable_task() and make sure cpuset/mempolicy intersection check is only done in the global oom context. [shakeelb@google.com: change function name and update comment] Link: http://lkml.kernel.org/r/20190628152421.198994-3-shakeelb@google.com Link: http://lkml.kernel.org/r/20190624212631.87212-3-shakeelb@google.com Signed-off-by: Shakeel Butt <shakeelb@google.com> Reported-by: syzbot+d0fc9d3c166bc5e4a94b@syzkaller.appspotmail.com Acked-by: Roman Gushchin <guro@fb.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Nick Piggin <npiggin@suse.de> Cc: Paul Jackson <pj@sgi.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-07-12 12:00:31 +08:00
const nodemask_t *mask = oc->nodemask;
if (is_memcg_oom(oc))
return true;
rcu_read_lock();
for_each_thread(start, tsk) {
if (mask) {
/*
* If this is a mempolicy constrained oom, tsk's
* cpuset is irrelevant. Only return true if its
* mempolicy intersects current, otherwise it may be
* needlessly killed.
*/
ret = mempolicy_nodemask_intersects(tsk, mask);
} else {
/*
* This is not a mempolicy constrained oom, so only
* check the mems of tsk's cpuset.
*/
ret = cpuset_mems_allowed_intersects(current, tsk);
}
if (ret)
break;
}
rcu_read_unlock();
return ret;
}
#else
oom: decouple mems_allowed from oom_unkillable_task Commit ef08e3b4981a ("[PATCH] cpusets: confine oom_killer to mem_exclusive cpuset") introduces a heuristic where a potential oom-killer victim is skipped if the intersection of the potential victim and the current (the process triggered the oom) is empty based on the reason that killing such victim most probably will not help the current allocating process. However the commit 7887a3da753e ("[PATCH] oom: cpuset hint") changed the heuristic to just decrease the oom_badness scores of such potential victim based on the reason that the cpuset of such processes might have changed and previously they may have allocated memory on mems where the current allocating process can allocate from. Unintentionally 7887a3da753e ("[PATCH] oom: cpuset hint") introduced a side effect as the oom_badness is also exposed to the user space through /proc/[pid]/oom_score, so, readers with different cpusets can read different oom_score of the same process. Later, commit 6cf86ac6f36b ("oom: filter tasks not sharing the same cpuset") fixed the side effect introduced by 7887a3da753e by moving the cpuset intersection back to only oom-killer context and out of oom_badness. However the combination of ab290adbaf8f ("oom: make oom_unkillable_task() helper function") and 26ebc984913b ("oom: /proc/<pid>/oom_score treat kernel thread honestly") unintentionally brought back the cpuset intersection check into the oom_badness calculation function. Other than doing cpuset/mempolicy intersection from oom_badness, the memcg oom context is also doing cpuset/mempolicy intersection which is quite wrong and is caught by syzcaller with the following report: kasan: CONFIG_KASAN_INLINE enabled kasan: GPF could be caused by NULL-ptr deref or user memory access general protection fault: 0000 [#1] PREEMPT SMP KASAN CPU: 0 PID: 28426 Comm: syz-executor.5 Not tainted 5.2.0-rc3-next-20190607 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:__read_once_size include/linux/compiler.h:194 [inline] RIP: 0010:has_intersects_mems_allowed mm/oom_kill.c:84 [inline] RIP: 0010:oom_unkillable_task mm/oom_kill.c:168 [inline] RIP: 0010:oom_unkillable_task+0x180/0x400 mm/oom_kill.c:155 Code: c1 ea 03 80 3c 02 00 0f 85 80 02 00 00 4c 8b a3 10 07 00 00 48 b8 00 00 00 00 00 fc ff df 4d 8d 74 24 10 4c 89 f2 48 c1 ea 03 <80> 3c 02 00 0f 85 67 02 00 00 49 8b 44 24 10 4c 8d a0 68 fa ff ff RSP: 0018:ffff888000127490 EFLAGS: 00010a03 RAX: dffffc0000000000 RBX: ffff8880a4cd5438 RCX: ffffffff818dae9c RDX: 100000000c3cc602 RSI: ffffffff818dac8d RDI: 0000000000000001 RBP: ffff8880001274d0 R08: ffff888000086180 R09: ffffed1015d26be0 R10: ffffed1015d26bdf R11: ffff8880ae935efb R12: 8000000061e63007 R13: 0000000000000000 R14: 8000000061e63017 R15: 1ffff11000024ea6 FS: 00005555561f5940(0000) GS:ffff8880ae800000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000607304 CR3: 000000009237e000 CR4: 00000000001426f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000600 Call Trace: oom_evaluate_task+0x49/0x520 mm/oom_kill.c:321 mem_cgroup_scan_tasks+0xcc/0x180 mm/memcontrol.c:1169 select_bad_process mm/oom_kill.c:374 [inline] out_of_memory mm/oom_kill.c:1088 [inline] out_of_memory+0x6b2/0x1280 mm/oom_kill.c:1035 mem_cgroup_out_of_memory+0x1ca/0x230 mm/memcontrol.c:1573 mem_cgroup_oom mm/memcontrol.c:1905 [inline] try_charge+0xfbe/0x1480 mm/memcontrol.c:2468 mem_cgroup_try_charge+0x24d/0x5e0 mm/memcontrol.c:6073 mem_cgroup_try_charge_delay+0x1f/0xa0 mm/memcontrol.c:6088 do_huge_pmd_wp_page_fallback+0x24f/0x1680 mm/huge_memory.c:1201 do_huge_pmd_wp_page+0x7fc/0x2160 mm/huge_memory.c:1359 wp_huge_pmd mm/memory.c:3793 [inline] __handle_mm_fault+0x164c/0x3eb0 mm/memory.c:4006 handle_mm_fault+0x3b7/0xa90 mm/memory.c:4053 do_user_addr_fault arch/x86/mm/fault.c:1455 [inline] __do_page_fault+0x5ef/0xda0 arch/x86/mm/fault.c:1521 do_page_fault+0x71/0x57d arch/x86/mm/fault.c:1552 page_fault+0x1e/0x30 arch/x86/entry/entry_64.S:1156 RIP: 0033:0x400590 Code: 06 e9 49 01 00 00 48 8b 44 24 10 48 0b 44 24 28 75 1f 48 8b 14 24 48 8b 7c 24 20 be 04 00 00 00 e8 f5 56 00 00 48 8b 74 24 08 <89> 06 e9 1e 01 00 00 48 8b 44 24 08 48 8b 14 24 be 04 00 00 00 8b RSP: 002b:00007fff7bc49780 EFLAGS: 00010206 RAX: 0000000000000001 RBX: 0000000000760000 RCX: 0000000000000000 RDX: 0000000000000000 RSI: 000000002000cffc RDI: 0000000000000001 RBP: fffffffffffffffe R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000075 R11: 0000000000000246 R12: 0000000000760008 R13: 00000000004c55f2 R14: 0000000000000000 R15: 00007fff7bc499b0 Modules linked in: ---[ end trace a65689219582ffff ]--- RIP: 0010:__read_once_size include/linux/compiler.h:194 [inline] RIP: 0010:has_intersects_mems_allowed mm/oom_kill.c:84 [inline] RIP: 0010:oom_unkillable_task mm/oom_kill.c:168 [inline] RIP: 0010:oom_unkillable_task+0x180/0x400 mm/oom_kill.c:155 Code: c1 ea 03 80 3c 02 00 0f 85 80 02 00 00 4c 8b a3 10 07 00 00 48 b8 00 00 00 00 00 fc ff df 4d 8d 74 24 10 4c 89 f2 48 c1 ea 03 <80> 3c 02 00 0f 85 67 02 00 00 49 8b 44 24 10 4c 8d a0 68 fa ff ff RSP: 0018:ffff888000127490 EFLAGS: 00010a03 RAX: dffffc0000000000 RBX: ffff8880a4cd5438 RCX: ffffffff818dae9c RDX: 100000000c3cc602 RSI: ffffffff818dac8d RDI: 0000000000000001 RBP: ffff8880001274d0 R08: ffff888000086180 R09: ffffed1015d26be0 R10: ffffed1015d26bdf R11: ffff8880ae935efb R12: 8000000061e63007 R13: 0000000000000000 R14: 8000000061e63017 R15: 1ffff11000024ea6 FS: 00005555561f5940(0000) GS:ffff8880ae800000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000001b2f823000 CR3: 000000009237e000 CR4: 00000000001426f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000600 The fix is to decouple the cpuset/mempolicy intersection check from oom_unkillable_task() and make sure cpuset/mempolicy intersection check is only done in the global oom context. [shakeelb@google.com: change function name and update comment] Link: http://lkml.kernel.org/r/20190628152421.198994-3-shakeelb@google.com Link: http://lkml.kernel.org/r/20190624212631.87212-3-shakeelb@google.com Signed-off-by: Shakeel Butt <shakeelb@google.com> Reported-by: syzbot+d0fc9d3c166bc5e4a94b@syzkaller.appspotmail.com Acked-by: Roman Gushchin <guro@fb.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Nick Piggin <npiggin@suse.de> Cc: Paul Jackson <pj@sgi.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-07-12 12:00:31 +08:00
static bool oom_cpuset_eligible(struct task_struct *tsk, struct oom_control *oc)
{
return true;
}
#endif /* CONFIG_NUMA */
/*
* The process p may have detached its own ->mm while exiting or through
* kthread_use_mm(), but one or more of its subthreads may still have a valid
* pointer. Return p, or any of its subthreads with a valid ->mm, with
* task_lock() held.
*/
struct task_struct *find_lock_task_mm(struct task_struct *p)
oom: introduce find_lock_task_mm() to fix !mm false positives Almost all ->mm == NULL checks in oom_kill.c are wrong. The current code assumes that the task without ->mm has already released its memory and ignores the process. However this is not necessarily true when this process is multithreaded, other live sub-threads can use this ->mm. - Remove the "if (!p->mm)" check in select_bad_process(), it is just wrong. - Add the new helper, find_lock_task_mm(), which finds the live thread which uses the memory and takes task_lock() to pin ->mm - change oom_badness() to use this helper instead of just checking ->mm != NULL. - As David pointed out, select_bad_process() must never choose the task without ->mm, but no matter what oom_badness() returns the task can be chosen if nothing else has been found yet. Change oom_badness() to return int, change it to return -1 if find_lock_task_mm() fails, and change select_bad_process() to check points >= 0. Note! This patch is not enough, we need more changes. - oom_badness() was fixed, but oom_kill_task() still ignores the task without ->mm - oom_forkbomb_penalty() should use find_lock_task_mm() too, and it also needs other changes to actually find the first first-descendant children This will be addressed later. [kosaki.motohiro@jp.fujitsu.com: use in badness(), __oom_kill_task()] Signed-off-by: Oleg Nesterov <oleg@redhat.com> Signed-off-by: David Rientjes <rientjes@google.com> Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 08:18:45 +08:00
{
struct task_struct *t;
oom: introduce find_lock_task_mm() to fix !mm false positives Almost all ->mm == NULL checks in oom_kill.c are wrong. The current code assumes that the task without ->mm has already released its memory and ignores the process. However this is not necessarily true when this process is multithreaded, other live sub-threads can use this ->mm. - Remove the "if (!p->mm)" check in select_bad_process(), it is just wrong. - Add the new helper, find_lock_task_mm(), which finds the live thread which uses the memory and takes task_lock() to pin ->mm - change oom_badness() to use this helper instead of just checking ->mm != NULL. - As David pointed out, select_bad_process() must never choose the task without ->mm, but no matter what oom_badness() returns the task can be chosen if nothing else has been found yet. Change oom_badness() to return int, change it to return -1 if find_lock_task_mm() fails, and change select_bad_process() to check points >= 0. Note! This patch is not enough, we need more changes. - oom_badness() was fixed, but oom_kill_task() still ignores the task without ->mm - oom_forkbomb_penalty() should use find_lock_task_mm() too, and it also needs other changes to actually find the first first-descendant children This will be addressed later. [kosaki.motohiro@jp.fujitsu.com: use in badness(), __oom_kill_task()] Signed-off-by: Oleg Nesterov <oleg@redhat.com> Signed-off-by: David Rientjes <rientjes@google.com> Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 08:18:45 +08:00
rcu_read_lock();
for_each_thread(p, t) {
oom: introduce find_lock_task_mm() to fix !mm false positives Almost all ->mm == NULL checks in oom_kill.c are wrong. The current code assumes that the task without ->mm has already released its memory and ignores the process. However this is not necessarily true when this process is multithreaded, other live sub-threads can use this ->mm. - Remove the "if (!p->mm)" check in select_bad_process(), it is just wrong. - Add the new helper, find_lock_task_mm(), which finds the live thread which uses the memory and takes task_lock() to pin ->mm - change oom_badness() to use this helper instead of just checking ->mm != NULL. - As David pointed out, select_bad_process() must never choose the task without ->mm, but no matter what oom_badness() returns the task can be chosen if nothing else has been found yet. Change oom_badness() to return int, change it to return -1 if find_lock_task_mm() fails, and change select_bad_process() to check points >= 0. Note! This patch is not enough, we need more changes. - oom_badness() was fixed, but oom_kill_task() still ignores the task without ->mm - oom_forkbomb_penalty() should use find_lock_task_mm() too, and it also needs other changes to actually find the first first-descendant children This will be addressed later. [kosaki.motohiro@jp.fujitsu.com: use in badness(), __oom_kill_task()] Signed-off-by: Oleg Nesterov <oleg@redhat.com> Signed-off-by: David Rientjes <rientjes@google.com> Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 08:18:45 +08:00
task_lock(t);
if (likely(t->mm))
goto found;
oom: introduce find_lock_task_mm() to fix !mm false positives Almost all ->mm == NULL checks in oom_kill.c are wrong. The current code assumes that the task without ->mm has already released its memory and ignores the process. However this is not necessarily true when this process is multithreaded, other live sub-threads can use this ->mm. - Remove the "if (!p->mm)" check in select_bad_process(), it is just wrong. - Add the new helper, find_lock_task_mm(), which finds the live thread which uses the memory and takes task_lock() to pin ->mm - change oom_badness() to use this helper instead of just checking ->mm != NULL. - As David pointed out, select_bad_process() must never choose the task without ->mm, but no matter what oom_badness() returns the task can be chosen if nothing else has been found yet. Change oom_badness() to return int, change it to return -1 if find_lock_task_mm() fails, and change select_bad_process() to check points >= 0. Note! This patch is not enough, we need more changes. - oom_badness() was fixed, but oom_kill_task() still ignores the task without ->mm - oom_forkbomb_penalty() should use find_lock_task_mm() too, and it also needs other changes to actually find the first first-descendant children This will be addressed later. [kosaki.motohiro@jp.fujitsu.com: use in badness(), __oom_kill_task()] Signed-off-by: Oleg Nesterov <oleg@redhat.com> Signed-off-by: David Rientjes <rientjes@google.com> Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 08:18:45 +08:00
task_unlock(t);
}
t = NULL;
found:
rcu_read_unlock();
oom: introduce find_lock_task_mm() to fix !mm false positives Almost all ->mm == NULL checks in oom_kill.c are wrong. The current code assumes that the task without ->mm has already released its memory and ignores the process. However this is not necessarily true when this process is multithreaded, other live sub-threads can use this ->mm. - Remove the "if (!p->mm)" check in select_bad_process(), it is just wrong. - Add the new helper, find_lock_task_mm(), which finds the live thread which uses the memory and takes task_lock() to pin ->mm - change oom_badness() to use this helper instead of just checking ->mm != NULL. - As David pointed out, select_bad_process() must never choose the task without ->mm, but no matter what oom_badness() returns the task can be chosen if nothing else has been found yet. Change oom_badness() to return int, change it to return -1 if find_lock_task_mm() fails, and change select_bad_process() to check points >= 0. Note! This patch is not enough, we need more changes. - oom_badness() was fixed, but oom_kill_task() still ignores the task without ->mm - oom_forkbomb_penalty() should use find_lock_task_mm() too, and it also needs other changes to actually find the first first-descendant children This will be addressed later. [kosaki.motohiro@jp.fujitsu.com: use in badness(), __oom_kill_task()] Signed-off-by: Oleg Nesterov <oleg@redhat.com> Signed-off-by: David Rientjes <rientjes@google.com> Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 08:18:45 +08:00
return t;
oom: introduce find_lock_task_mm() to fix !mm false positives Almost all ->mm == NULL checks in oom_kill.c are wrong. The current code assumes that the task without ->mm has already released its memory and ignores the process. However this is not necessarily true when this process is multithreaded, other live sub-threads can use this ->mm. - Remove the "if (!p->mm)" check in select_bad_process(), it is just wrong. - Add the new helper, find_lock_task_mm(), which finds the live thread which uses the memory and takes task_lock() to pin ->mm - change oom_badness() to use this helper instead of just checking ->mm != NULL. - As David pointed out, select_bad_process() must never choose the task without ->mm, but no matter what oom_badness() returns the task can be chosen if nothing else has been found yet. Change oom_badness() to return int, change it to return -1 if find_lock_task_mm() fails, and change select_bad_process() to check points >= 0. Note! This patch is not enough, we need more changes. - oom_badness() was fixed, but oom_kill_task() still ignores the task without ->mm - oom_forkbomb_penalty() should use find_lock_task_mm() too, and it also needs other changes to actually find the first first-descendant children This will be addressed later. [kosaki.motohiro@jp.fujitsu.com: use in badness(), __oom_kill_task()] Signed-off-by: Oleg Nesterov <oleg@redhat.com> Signed-off-by: David Rientjes <rientjes@google.com> Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 08:18:45 +08:00
}
/*
* order == -1 means the oom kill is required by sysrq, otherwise only
* for display purposes.
*/
static inline bool is_sysrq_oom(struct oom_control *oc)
{
return oc->order == -1;
}
/* return true if the task is not adequate as candidate victim task. */
oom: decouple mems_allowed from oom_unkillable_task Commit ef08e3b4981a ("[PATCH] cpusets: confine oom_killer to mem_exclusive cpuset") introduces a heuristic where a potential oom-killer victim is skipped if the intersection of the potential victim and the current (the process triggered the oom) is empty based on the reason that killing such victim most probably will not help the current allocating process. However the commit 7887a3da753e ("[PATCH] oom: cpuset hint") changed the heuristic to just decrease the oom_badness scores of such potential victim based on the reason that the cpuset of such processes might have changed and previously they may have allocated memory on mems where the current allocating process can allocate from. Unintentionally 7887a3da753e ("[PATCH] oom: cpuset hint") introduced a side effect as the oom_badness is also exposed to the user space through /proc/[pid]/oom_score, so, readers with different cpusets can read different oom_score of the same process. Later, commit 6cf86ac6f36b ("oom: filter tasks not sharing the same cpuset") fixed the side effect introduced by 7887a3da753e by moving the cpuset intersection back to only oom-killer context and out of oom_badness. However the combination of ab290adbaf8f ("oom: make oom_unkillable_task() helper function") and 26ebc984913b ("oom: /proc/<pid>/oom_score treat kernel thread honestly") unintentionally brought back the cpuset intersection check into the oom_badness calculation function. Other than doing cpuset/mempolicy intersection from oom_badness, the memcg oom context is also doing cpuset/mempolicy intersection which is quite wrong and is caught by syzcaller with the following report: kasan: CONFIG_KASAN_INLINE enabled kasan: GPF could be caused by NULL-ptr deref or user memory access general protection fault: 0000 [#1] PREEMPT SMP KASAN CPU: 0 PID: 28426 Comm: syz-executor.5 Not tainted 5.2.0-rc3-next-20190607 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:__read_once_size include/linux/compiler.h:194 [inline] RIP: 0010:has_intersects_mems_allowed mm/oom_kill.c:84 [inline] RIP: 0010:oom_unkillable_task mm/oom_kill.c:168 [inline] RIP: 0010:oom_unkillable_task+0x180/0x400 mm/oom_kill.c:155 Code: c1 ea 03 80 3c 02 00 0f 85 80 02 00 00 4c 8b a3 10 07 00 00 48 b8 00 00 00 00 00 fc ff df 4d 8d 74 24 10 4c 89 f2 48 c1 ea 03 <80> 3c 02 00 0f 85 67 02 00 00 49 8b 44 24 10 4c 8d a0 68 fa ff ff RSP: 0018:ffff888000127490 EFLAGS: 00010a03 RAX: dffffc0000000000 RBX: ffff8880a4cd5438 RCX: ffffffff818dae9c RDX: 100000000c3cc602 RSI: ffffffff818dac8d RDI: 0000000000000001 RBP: ffff8880001274d0 R08: ffff888000086180 R09: ffffed1015d26be0 R10: ffffed1015d26bdf R11: ffff8880ae935efb R12: 8000000061e63007 R13: 0000000000000000 R14: 8000000061e63017 R15: 1ffff11000024ea6 FS: 00005555561f5940(0000) GS:ffff8880ae800000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000607304 CR3: 000000009237e000 CR4: 00000000001426f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000600 Call Trace: oom_evaluate_task+0x49/0x520 mm/oom_kill.c:321 mem_cgroup_scan_tasks+0xcc/0x180 mm/memcontrol.c:1169 select_bad_process mm/oom_kill.c:374 [inline] out_of_memory mm/oom_kill.c:1088 [inline] out_of_memory+0x6b2/0x1280 mm/oom_kill.c:1035 mem_cgroup_out_of_memory+0x1ca/0x230 mm/memcontrol.c:1573 mem_cgroup_oom mm/memcontrol.c:1905 [inline] try_charge+0xfbe/0x1480 mm/memcontrol.c:2468 mem_cgroup_try_charge+0x24d/0x5e0 mm/memcontrol.c:6073 mem_cgroup_try_charge_delay+0x1f/0xa0 mm/memcontrol.c:6088 do_huge_pmd_wp_page_fallback+0x24f/0x1680 mm/huge_memory.c:1201 do_huge_pmd_wp_page+0x7fc/0x2160 mm/huge_memory.c:1359 wp_huge_pmd mm/memory.c:3793 [inline] __handle_mm_fault+0x164c/0x3eb0 mm/memory.c:4006 handle_mm_fault+0x3b7/0xa90 mm/memory.c:4053 do_user_addr_fault arch/x86/mm/fault.c:1455 [inline] __do_page_fault+0x5ef/0xda0 arch/x86/mm/fault.c:1521 do_page_fault+0x71/0x57d arch/x86/mm/fault.c:1552 page_fault+0x1e/0x30 arch/x86/entry/entry_64.S:1156 RIP: 0033:0x400590 Code: 06 e9 49 01 00 00 48 8b 44 24 10 48 0b 44 24 28 75 1f 48 8b 14 24 48 8b 7c 24 20 be 04 00 00 00 e8 f5 56 00 00 48 8b 74 24 08 <89> 06 e9 1e 01 00 00 48 8b 44 24 08 48 8b 14 24 be 04 00 00 00 8b RSP: 002b:00007fff7bc49780 EFLAGS: 00010206 RAX: 0000000000000001 RBX: 0000000000760000 RCX: 0000000000000000 RDX: 0000000000000000 RSI: 000000002000cffc RDI: 0000000000000001 RBP: fffffffffffffffe R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000075 R11: 0000000000000246 R12: 0000000000760008 R13: 00000000004c55f2 R14: 0000000000000000 R15: 00007fff7bc499b0 Modules linked in: ---[ end trace a65689219582ffff ]--- RIP: 0010:__read_once_size include/linux/compiler.h:194 [inline] RIP: 0010:has_intersects_mems_allowed mm/oom_kill.c:84 [inline] RIP: 0010:oom_unkillable_task mm/oom_kill.c:168 [inline] RIP: 0010:oom_unkillable_task+0x180/0x400 mm/oom_kill.c:155 Code: c1 ea 03 80 3c 02 00 0f 85 80 02 00 00 4c 8b a3 10 07 00 00 48 b8 00 00 00 00 00 fc ff df 4d 8d 74 24 10 4c 89 f2 48 c1 ea 03 <80> 3c 02 00 0f 85 67 02 00 00 49 8b 44 24 10 4c 8d a0 68 fa ff ff RSP: 0018:ffff888000127490 EFLAGS: 00010a03 RAX: dffffc0000000000 RBX: ffff8880a4cd5438 RCX: ffffffff818dae9c RDX: 100000000c3cc602 RSI: ffffffff818dac8d RDI: 0000000000000001 RBP: ffff8880001274d0 R08: ffff888000086180 R09: ffffed1015d26be0 R10: ffffed1015d26bdf R11: ffff8880ae935efb R12: 8000000061e63007 R13: 0000000000000000 R14: 8000000061e63017 R15: 1ffff11000024ea6 FS: 00005555561f5940(0000) GS:ffff8880ae800000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000001b2f823000 CR3: 000000009237e000 CR4: 00000000001426f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000600 The fix is to decouple the cpuset/mempolicy intersection check from oom_unkillable_task() and make sure cpuset/mempolicy intersection check is only done in the global oom context. [shakeelb@google.com: change function name and update comment] Link: http://lkml.kernel.org/r/20190628152421.198994-3-shakeelb@google.com Link: http://lkml.kernel.org/r/20190624212631.87212-3-shakeelb@google.com Signed-off-by: Shakeel Butt <shakeelb@google.com> Reported-by: syzbot+d0fc9d3c166bc5e4a94b@syzkaller.appspotmail.com Acked-by: Roman Gushchin <guro@fb.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Nick Piggin <npiggin@suse.de> Cc: Paul Jackson <pj@sgi.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-07-12 12:00:31 +08:00
static bool oom_unkillable_task(struct task_struct *p)
{
if (is_global_init(p))
return true;
if (p->flags & PF_KTHREAD)
return true;
return false;
}
/**
* Check whether unreclaimable slab amount is greater than
* all user memory(LRU pages).
* dump_unreclaimable_slab() could help in the case that
* oom due to too much unreclaimable slab used by kernel.
*/
static bool should_dump_unreclaim_slab(void)
mm: oom: show unreclaimable slab info when unreclaimable slabs > user memory The kernel may panic when an oom happens without killable process sometimes it is caused by huge unreclaimable slabs used by kernel. Although kdump could help debug such problem, however, kdump is not available on all architectures and it might be malfunction sometime. And, since kernel already panic it is worthy capturing such information in dmesg to aid touble shooting. Print out unreclaimable slab info (used size and total size) which actual memory usage is not zero (num_objs * size != 0) when unreclaimable slabs amount is greater than total user memory (LRU pages). The output looks like: Unreclaimable slab info: Name Used Total rpc_buffers 31KB 31KB rpc_tasks 7KB 7KB ebitmap_node 1964KB 1964KB avtab_node 5024KB 5024KB xfs_buf 1402KB 1402KB xfs_ili 134KB 134KB xfs_efi_item 115KB 115KB xfs_efd_item 115KB 115KB xfs_buf_item 134KB 134KB xfs_log_item_desc 342KB 342KB xfs_trans 1412KB 1412KB xfs_ifork 212KB 212KB [yang.s@alibaba-inc.com: v11] Link: http://lkml.kernel.org/r/1507656303-103845-4-git-send-email-yang.s@alibaba-inc.com Link: http://lkml.kernel.org/r/1507152550-46205-4-git-send-email-yang.s@alibaba-inc.com Signed-off-by: Yang Shi <yang.s@alibaba-inc.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Christoph Lameter <cl@linux.com> Cc: David Rientjes <rientjes@google.com> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Pekka Enberg <penberg@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-16 09:32:07 +08:00
{
unsigned long nr_lru;
nr_lru = global_node_page_state(NR_ACTIVE_ANON) +
global_node_page_state(NR_INACTIVE_ANON) +
global_node_page_state(NR_ACTIVE_FILE) +
global_node_page_state(NR_INACTIVE_FILE) +
global_node_page_state(NR_ISOLATED_ANON) +
global_node_page_state(NR_ISOLATED_FILE) +
global_node_page_state(NR_UNEVICTABLE);
return (global_node_page_state_pages(NR_SLAB_UNRECLAIMABLE_B) > nr_lru);
mm: oom: show unreclaimable slab info when unreclaimable slabs > user memory The kernel may panic when an oom happens without killable process sometimes it is caused by huge unreclaimable slabs used by kernel. Although kdump could help debug such problem, however, kdump is not available on all architectures and it might be malfunction sometime. And, since kernel already panic it is worthy capturing such information in dmesg to aid touble shooting. Print out unreclaimable slab info (used size and total size) which actual memory usage is not zero (num_objs * size != 0) when unreclaimable slabs amount is greater than total user memory (LRU pages). The output looks like: Unreclaimable slab info: Name Used Total rpc_buffers 31KB 31KB rpc_tasks 7KB 7KB ebitmap_node 1964KB 1964KB avtab_node 5024KB 5024KB xfs_buf 1402KB 1402KB xfs_ili 134KB 134KB xfs_efi_item 115KB 115KB xfs_efd_item 115KB 115KB xfs_buf_item 134KB 134KB xfs_log_item_desc 342KB 342KB xfs_trans 1412KB 1412KB xfs_ifork 212KB 212KB [yang.s@alibaba-inc.com: v11] Link: http://lkml.kernel.org/r/1507656303-103845-4-git-send-email-yang.s@alibaba-inc.com Link: http://lkml.kernel.org/r/1507152550-46205-4-git-send-email-yang.s@alibaba-inc.com Signed-off-by: Yang Shi <yang.s@alibaba-inc.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Christoph Lameter <cl@linux.com> Cc: David Rientjes <rientjes@google.com> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Pekka Enberg <penberg@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-16 09:32:07 +08:00
}
/**
oom: badness heuristic rewrite This a complete rewrite of the oom killer's badness() heuristic which is used to determine which task to kill in oom conditions. The goal is to make it as simple and predictable as possible so the results are better understood and we end up killing the task which will lead to the most memory freeing while still respecting the fine-tuning from userspace. Instead of basing the heuristic on mm->total_vm for each task, the task's rss and swap space is used instead. This is a better indication of the amount of memory that will be freeable if the oom killed task is chosen and subsequently exits. This helps specifically in cases where KDE or GNOME is chosen for oom kill on desktop systems instead of a memory hogging task. The baseline for the heuristic is a proportion of memory that each task is currently using in memory plus swap compared to the amount of "allowable" memory. "Allowable," in this sense, means the system-wide resources for unconstrained oom conditions, the set of mempolicy nodes, the mems attached to current's cpuset, or a memory controller's limit. The proportion is given on a scale of 0 (never kill) to 1000 (always kill), roughly meaning that if a task has a badness() score of 500 that the task consumes approximately 50% of allowable memory resident in RAM or in swap space. The proportion is always relative to the amount of "allowable" memory and not the total amount of RAM systemwide so that mempolicies and cpusets may operate in isolation; they shall not need to know the true size of the machine on which they are running if they are bound to a specific set of nodes or mems, respectively. Root tasks are given 3% extra memory just like __vm_enough_memory() provides in LSMs. In the event of two tasks consuming similar amounts of memory, it is generally better to save root's task. Because of the change in the badness() heuristic's baseline, it is also necessary to introduce a new user interface to tune it. It's not possible to redefine the meaning of /proc/pid/oom_adj with a new scale since the ABI cannot be changed for backward compatability. Instead, a new tunable, /proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may be used to polarize the heuristic such that certain tasks are never considered for oom kill while others may always be considered. The value is added directly into the badness() score so a value of -500, for example, means to discount 50% of its memory consumption in comparison to other tasks either on the system, bound to the mempolicy, in the cpuset, or sharing the same memory controller. /proc/pid/oom_adj is changed so that its meaning is rescaled into the units used by /proc/pid/oom_score_adj, and vice versa. Changing one of these per-task tunables will rescale the value of the other to an equivalent meaning. Although /proc/pid/oom_adj was originally defined as a bitshift on the badness score, it now shares the same linear growth as /proc/pid/oom_score_adj but with different granularity. This is required so the ABI is not broken with userspace applications and allows oom_adj to be deprecated for future removal. Signed-off-by: David Rientjes <rientjes@google.com> Cc: Nick Piggin <npiggin@suse.de> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Balbir Singh <balbir@in.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 08:19:46 +08:00
* oom_badness - heuristic function to determine which candidate task to kill
* @p: task struct of which task we should calculate
oom: badness heuristic rewrite This a complete rewrite of the oom killer's badness() heuristic which is used to determine which task to kill in oom conditions. The goal is to make it as simple and predictable as possible so the results are better understood and we end up killing the task which will lead to the most memory freeing while still respecting the fine-tuning from userspace. Instead of basing the heuristic on mm->total_vm for each task, the task's rss and swap space is used instead. This is a better indication of the amount of memory that will be freeable if the oom killed task is chosen and subsequently exits. This helps specifically in cases where KDE or GNOME is chosen for oom kill on desktop systems instead of a memory hogging task. The baseline for the heuristic is a proportion of memory that each task is currently using in memory plus swap compared to the amount of "allowable" memory. "Allowable," in this sense, means the system-wide resources for unconstrained oom conditions, the set of mempolicy nodes, the mems attached to current's cpuset, or a memory controller's limit. The proportion is given on a scale of 0 (never kill) to 1000 (always kill), roughly meaning that if a task has a badness() score of 500 that the task consumes approximately 50% of allowable memory resident in RAM or in swap space. The proportion is always relative to the amount of "allowable" memory and not the total amount of RAM systemwide so that mempolicies and cpusets may operate in isolation; they shall not need to know the true size of the machine on which they are running if they are bound to a specific set of nodes or mems, respectively. Root tasks are given 3% extra memory just like __vm_enough_memory() provides in LSMs. In the event of two tasks consuming similar amounts of memory, it is generally better to save root's task. Because of the change in the badness() heuristic's baseline, it is also necessary to introduce a new user interface to tune it. It's not possible to redefine the meaning of /proc/pid/oom_adj with a new scale since the ABI cannot be changed for backward compatability. Instead, a new tunable, /proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may be used to polarize the heuristic such that certain tasks are never considered for oom kill while others may always be considered. The value is added directly into the badness() score so a value of -500, for example, means to discount 50% of its memory consumption in comparison to other tasks either on the system, bound to the mempolicy, in the cpuset, or sharing the same memory controller. /proc/pid/oom_adj is changed so that its meaning is rescaled into the units used by /proc/pid/oom_score_adj, and vice versa. Changing one of these per-task tunables will rescale the value of the other to an equivalent meaning. Although /proc/pid/oom_adj was originally defined as a bitshift on the badness score, it now shares the same linear growth as /proc/pid/oom_score_adj but with different granularity. This is required so the ABI is not broken with userspace applications and allows oom_adj to be deprecated for future removal. Signed-off-by: David Rientjes <rientjes@google.com> Cc: Nick Piggin <npiggin@suse.de> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Balbir Singh <balbir@in.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 08:19:46 +08:00
* @totalpages: total present RAM allowed for page allocation
*
oom: badness heuristic rewrite This a complete rewrite of the oom killer's badness() heuristic which is used to determine which task to kill in oom conditions. The goal is to make it as simple and predictable as possible so the results are better understood and we end up killing the task which will lead to the most memory freeing while still respecting the fine-tuning from userspace. Instead of basing the heuristic on mm->total_vm for each task, the task's rss and swap space is used instead. This is a better indication of the amount of memory that will be freeable if the oom killed task is chosen and subsequently exits. This helps specifically in cases where KDE or GNOME is chosen for oom kill on desktop systems instead of a memory hogging task. The baseline for the heuristic is a proportion of memory that each task is currently using in memory plus swap compared to the amount of "allowable" memory. "Allowable," in this sense, means the system-wide resources for unconstrained oom conditions, the set of mempolicy nodes, the mems attached to current's cpuset, or a memory controller's limit. The proportion is given on a scale of 0 (never kill) to 1000 (always kill), roughly meaning that if a task has a badness() score of 500 that the task consumes approximately 50% of allowable memory resident in RAM or in swap space. The proportion is always relative to the amount of "allowable" memory and not the total amount of RAM systemwide so that mempolicies and cpusets may operate in isolation; they shall not need to know the true size of the machine on which they are running if they are bound to a specific set of nodes or mems, respectively. Root tasks are given 3% extra memory just like __vm_enough_memory() provides in LSMs. In the event of two tasks consuming similar amounts of memory, it is generally better to save root's task. Because of the change in the badness() heuristic's baseline, it is also necessary to introduce a new user interface to tune it. It's not possible to redefine the meaning of /proc/pid/oom_adj with a new scale since the ABI cannot be changed for backward compatability. Instead, a new tunable, /proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may be used to polarize the heuristic such that certain tasks are never considered for oom kill while others may always be considered. The value is added directly into the badness() score so a value of -500, for example, means to discount 50% of its memory consumption in comparison to other tasks either on the system, bound to the mempolicy, in the cpuset, or sharing the same memory controller. /proc/pid/oom_adj is changed so that its meaning is rescaled into the units used by /proc/pid/oom_score_adj, and vice versa. Changing one of these per-task tunables will rescale the value of the other to an equivalent meaning. Although /proc/pid/oom_adj was originally defined as a bitshift on the badness score, it now shares the same linear growth as /proc/pid/oom_score_adj but with different granularity. This is required so the ABI is not broken with userspace applications and allows oom_adj to be deprecated for future removal. Signed-off-by: David Rientjes <rientjes@google.com> Cc: Nick Piggin <npiggin@suse.de> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Balbir Singh <balbir@in.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 08:19:46 +08:00
* The heuristic for determining which task to kill is made to be as simple and
* predictable as possible. The goal is to return the highest value for the
* task consuming the most memory to avoid subsequent oom failures.
*/
mm, oom: make the calculation of oom badness more accurate Recently we found an issue on our production environment that when memcg oom is triggered the oom killer doesn't chose the process with largest resident memory but chose the first scanned process. Note that all processes in this memcg have the same oom_score_adj, so the oom killer should chose the process with largest resident memory. Bellow is part of the oom info, which is enough to analyze this issue. [7516987.983223] memory: usage 16777216kB, limit 16777216kB, failcnt 52843037 [7516987.983224] memory+swap: usage 16777216kB, limit 9007199254740988kB, failcnt 0 [7516987.983225] kmem: usage 301464kB, limit 9007199254740988kB, failcnt 0 [...] [7516987.983293] [ pid ] uid tgid total_vm rss pgtables_bytes swapents oom_score_adj name [7516987.983510] [ 5740] 0 5740 257 1 32768 0 -998 pause [7516987.983574] [58804] 0 58804 4594 771 81920 0 -998 entry_point.bas [7516987.983577] [58908] 0 58908 7089 689 98304 0 -998 cron [7516987.983580] [58910] 0 58910 16235 5576 163840 0 -998 supervisord [7516987.983590] [59620] 0 59620 18074 1395 188416 0 -998 sshd [7516987.983594] [59622] 0 59622 18680 6679 188416 0 -998 python [7516987.983598] [59624] 0 59624 1859266 5161 548864 0 -998 odin-agent [7516987.983600] [59625] 0 59625 707223 9248 983040 0 -998 filebeat [7516987.983604] [59627] 0 59627 416433 64239 774144 0 -998 odin-log-agent [7516987.983607] [59631] 0 59631 180671 15012 385024 0 -998 python3 [7516987.983612] [61396] 0 61396 791287 3189 352256 0 -998 client [7516987.983615] [61641] 0 61641 1844642 29089 946176 0 -998 client [7516987.983765] [ 9236] 0 9236 2642 467 53248 0 -998 php_scanner [7516987.983911] [42898] 0 42898 15543 838 167936 0 -998 su [7516987.983915] [42900] 1000 42900 3673 867 77824 0 -998 exec_script_vr2 [7516987.983918] [42925] 1000 42925 36475 19033 335872 0 -998 python [7516987.983921] [57146] 1000 57146 3673 848 73728 0 -998 exec_script_J2p [7516987.983925] [57195] 1000 57195 186359 22958 491520 0 -998 python2 [7516987.983928] [58376] 1000 58376 275764 14402 290816 0 -998 rosmaster [7516987.983931] [58395] 1000 58395 155166 4449 245760 0 -998 rosout [7516987.983935] [58406] 1000 58406 18285584 3967322 37101568 0 -998 data_sim [7516987.984221] oom-kill:constraint=CONSTRAINT_MEMCG,nodemask=(null),cpuset=3aa16c9482ae3a6f6b78bda68a55d32c87c99b985e0f11331cddf05af6c4d753,mems_allowed=0-1,oom_memcg=/kubepods/podf1c273d3-9b36-11ea-b3df-246e9693c184,task_memcg=/kubepods/podf1c273d3-9b36-11ea-b3df-246e9693c184/1f246a3eeea8f70bf91141eeaf1805346a666e225f823906485ea0b6c37dfc3d,task=pause,pid=5740,uid=0 [7516987.984254] Memory cgroup out of memory: Killed process 5740 (pause) total-vm:1028kB, anon-rss:4kB, file-rss:0kB, shmem-rss:0kB [7516988.092344] oom_reaper: reaped process 5740 (pause), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kB We can find that the first scanned process 5740 (pause) was killed, but its rss is only one page. That is because, when we calculate the oom badness in oom_badness(), we always ignore the negtive point and convert all of these negtive points to 1. Now as oom_score_adj of all the processes in this targeted memcg have the same value -998, the points of these processes are all negtive value. As a result, the first scanned process will be killed. The oom_socre_adj (-998) in this memcg is set by kubelet, because it is a a Guaranteed pod, which has higher priority to prevent from being killed by system oom. To fix this issue, we should make the calculation of oom point more accurate. We can achieve it by convert the chosen_point from 'unsigned long' to 'long'. [cai@lca.pw: reported a issue in the previous version] [mhocko@suse.com: fixed the issue reported by Cai] [mhocko@suse.com: add the comment in proc_oom_score()] [laoar.shao@gmail.com: v3] Link: http://lkml.kernel.org/r/1594396651-9931-1-git-send-email-laoar.shao@gmail.com Signed-off-by: Yafang Shao <laoar.shao@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Tested-by: Naresh Kamboju <naresh.kamboju@linaro.org> Acked-by: Michal Hocko <mhocko@suse.com> Cc: David Rientjes <rientjes@google.com> Cc: Qian Cai <cai@lca.pw> Link: http://lkml.kernel.org/r/1594309987-9919-1-git-send-email-laoar.shao@gmail.com Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-08-12 09:31:22 +08:00
long oom_badness(struct task_struct *p, unsigned long totalpages)
{
long points;
long adj;
oom: move oom_adj value from task_struct to signal_struct Currently, OOM logic callflow is here. __out_of_memory() select_bad_process() for each task badness() calculate badness of one task oom_kill_process() search child oom_kill_task() kill target task and mm shared tasks with it example, process-A have two thread, thread-A and thread-B and it have very fat memory and each thread have following oom_adj and oom_score. thread-A: oom_adj = OOM_DISABLE, oom_score = 0 thread-B: oom_adj = 0, oom_score = very-high Then, select_bad_process() select thread-B, but oom_kill_task() refuse kill the task because thread-A have OOM_DISABLE. Thus __out_of_memory() call select_bad_process() again. but select_bad_process() select the same task. It mean kernel fall in livelock. The fact is, select_bad_process() must select killable task. otherwise OOM logic go into livelock. And root cause is, oom_adj shouldn't be per-thread value. it should be per-process value because OOM-killer kill a process, not thread. Thus This patch moves oomkilladj (now more appropriately named oom_adj) from struct task_struct to struct signal_struct. it naturally prevent select_bad_process() choose wrong task. Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Paul Menage <menage@google.com> Cc: David Rientjes <rientjes@google.com> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Rik van Riel <riel@redhat.com> Cc: Oleg Nesterov <oleg@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-09-22 08:03:13 +08:00
oom: decouple mems_allowed from oom_unkillable_task Commit ef08e3b4981a ("[PATCH] cpusets: confine oom_killer to mem_exclusive cpuset") introduces a heuristic where a potential oom-killer victim is skipped if the intersection of the potential victim and the current (the process triggered the oom) is empty based on the reason that killing such victim most probably will not help the current allocating process. However the commit 7887a3da753e ("[PATCH] oom: cpuset hint") changed the heuristic to just decrease the oom_badness scores of such potential victim based on the reason that the cpuset of such processes might have changed and previously they may have allocated memory on mems where the current allocating process can allocate from. Unintentionally 7887a3da753e ("[PATCH] oom: cpuset hint") introduced a side effect as the oom_badness is also exposed to the user space through /proc/[pid]/oom_score, so, readers with different cpusets can read different oom_score of the same process. Later, commit 6cf86ac6f36b ("oom: filter tasks not sharing the same cpuset") fixed the side effect introduced by 7887a3da753e by moving the cpuset intersection back to only oom-killer context and out of oom_badness. However the combination of ab290adbaf8f ("oom: make oom_unkillable_task() helper function") and 26ebc984913b ("oom: /proc/<pid>/oom_score treat kernel thread honestly") unintentionally brought back the cpuset intersection check into the oom_badness calculation function. Other than doing cpuset/mempolicy intersection from oom_badness, the memcg oom context is also doing cpuset/mempolicy intersection which is quite wrong and is caught by syzcaller with the following report: kasan: CONFIG_KASAN_INLINE enabled kasan: GPF could be caused by NULL-ptr deref or user memory access general protection fault: 0000 [#1] PREEMPT SMP KASAN CPU: 0 PID: 28426 Comm: syz-executor.5 Not tainted 5.2.0-rc3-next-20190607 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:__read_once_size include/linux/compiler.h:194 [inline] RIP: 0010:has_intersects_mems_allowed mm/oom_kill.c:84 [inline] RIP: 0010:oom_unkillable_task mm/oom_kill.c:168 [inline] RIP: 0010:oom_unkillable_task+0x180/0x400 mm/oom_kill.c:155 Code: c1 ea 03 80 3c 02 00 0f 85 80 02 00 00 4c 8b a3 10 07 00 00 48 b8 00 00 00 00 00 fc ff df 4d 8d 74 24 10 4c 89 f2 48 c1 ea 03 <80> 3c 02 00 0f 85 67 02 00 00 49 8b 44 24 10 4c 8d a0 68 fa ff ff RSP: 0018:ffff888000127490 EFLAGS: 00010a03 RAX: dffffc0000000000 RBX: ffff8880a4cd5438 RCX: ffffffff818dae9c RDX: 100000000c3cc602 RSI: ffffffff818dac8d RDI: 0000000000000001 RBP: ffff8880001274d0 R08: ffff888000086180 R09: ffffed1015d26be0 R10: ffffed1015d26bdf R11: ffff8880ae935efb R12: 8000000061e63007 R13: 0000000000000000 R14: 8000000061e63017 R15: 1ffff11000024ea6 FS: 00005555561f5940(0000) GS:ffff8880ae800000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000607304 CR3: 000000009237e000 CR4: 00000000001426f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000600 Call Trace: oom_evaluate_task+0x49/0x520 mm/oom_kill.c:321 mem_cgroup_scan_tasks+0xcc/0x180 mm/memcontrol.c:1169 select_bad_process mm/oom_kill.c:374 [inline] out_of_memory mm/oom_kill.c:1088 [inline] out_of_memory+0x6b2/0x1280 mm/oom_kill.c:1035 mem_cgroup_out_of_memory+0x1ca/0x230 mm/memcontrol.c:1573 mem_cgroup_oom mm/memcontrol.c:1905 [inline] try_charge+0xfbe/0x1480 mm/memcontrol.c:2468 mem_cgroup_try_charge+0x24d/0x5e0 mm/memcontrol.c:6073 mem_cgroup_try_charge_delay+0x1f/0xa0 mm/memcontrol.c:6088 do_huge_pmd_wp_page_fallback+0x24f/0x1680 mm/huge_memory.c:1201 do_huge_pmd_wp_page+0x7fc/0x2160 mm/huge_memory.c:1359 wp_huge_pmd mm/memory.c:3793 [inline] __handle_mm_fault+0x164c/0x3eb0 mm/memory.c:4006 handle_mm_fault+0x3b7/0xa90 mm/memory.c:4053 do_user_addr_fault arch/x86/mm/fault.c:1455 [inline] __do_page_fault+0x5ef/0xda0 arch/x86/mm/fault.c:1521 do_page_fault+0x71/0x57d arch/x86/mm/fault.c:1552 page_fault+0x1e/0x30 arch/x86/entry/entry_64.S:1156 RIP: 0033:0x400590 Code: 06 e9 49 01 00 00 48 8b 44 24 10 48 0b 44 24 28 75 1f 48 8b 14 24 48 8b 7c 24 20 be 04 00 00 00 e8 f5 56 00 00 48 8b 74 24 08 <89> 06 e9 1e 01 00 00 48 8b 44 24 08 48 8b 14 24 be 04 00 00 00 8b RSP: 002b:00007fff7bc49780 EFLAGS: 00010206 RAX: 0000000000000001 RBX: 0000000000760000 RCX: 0000000000000000 RDX: 0000000000000000 RSI: 000000002000cffc RDI: 0000000000000001 RBP: fffffffffffffffe R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000075 R11: 0000000000000246 R12: 0000000000760008 R13: 00000000004c55f2 R14: 0000000000000000 R15: 00007fff7bc499b0 Modules linked in: ---[ end trace a65689219582ffff ]--- RIP: 0010:__read_once_size include/linux/compiler.h:194 [inline] RIP: 0010:has_intersects_mems_allowed mm/oom_kill.c:84 [inline] RIP: 0010:oom_unkillable_task mm/oom_kill.c:168 [inline] RIP: 0010:oom_unkillable_task+0x180/0x400 mm/oom_kill.c:155 Code: c1 ea 03 80 3c 02 00 0f 85 80 02 00 00 4c 8b a3 10 07 00 00 48 b8 00 00 00 00 00 fc ff df 4d 8d 74 24 10 4c 89 f2 48 c1 ea 03 <80> 3c 02 00 0f 85 67 02 00 00 49 8b 44 24 10 4c 8d a0 68 fa ff ff RSP: 0018:ffff888000127490 EFLAGS: 00010a03 RAX: dffffc0000000000 RBX: ffff8880a4cd5438 RCX: ffffffff818dae9c RDX: 100000000c3cc602 RSI: ffffffff818dac8d RDI: 0000000000000001 RBP: ffff8880001274d0 R08: ffff888000086180 R09: ffffed1015d26be0 R10: ffffed1015d26bdf R11: ffff8880ae935efb R12: 8000000061e63007 R13: 0000000000000000 R14: 8000000061e63017 R15: 1ffff11000024ea6 FS: 00005555561f5940(0000) GS:ffff8880ae800000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000001b2f823000 CR3: 000000009237e000 CR4: 00000000001426f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000600 The fix is to decouple the cpuset/mempolicy intersection check from oom_unkillable_task() and make sure cpuset/mempolicy intersection check is only done in the global oom context. [shakeelb@google.com: change function name and update comment] Link: http://lkml.kernel.org/r/20190628152421.198994-3-shakeelb@google.com Link: http://lkml.kernel.org/r/20190624212631.87212-3-shakeelb@google.com Signed-off-by: Shakeel Butt <shakeelb@google.com> Reported-by: syzbot+d0fc9d3c166bc5e4a94b@syzkaller.appspotmail.com Acked-by: Roman Gushchin <guro@fb.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Nick Piggin <npiggin@suse.de> Cc: Paul Jackson <pj@sgi.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-07-12 12:00:31 +08:00
if (oom_unkillable_task(p))
mm, oom: make the calculation of oom badness more accurate Recently we found an issue on our production environment that when memcg oom is triggered the oom killer doesn't chose the process with largest resident memory but chose the first scanned process. Note that all processes in this memcg have the same oom_score_adj, so the oom killer should chose the process with largest resident memory. Bellow is part of the oom info, which is enough to analyze this issue. [7516987.983223] memory: usage 16777216kB, limit 16777216kB, failcnt 52843037 [7516987.983224] memory+swap: usage 16777216kB, limit 9007199254740988kB, failcnt 0 [7516987.983225] kmem: usage 301464kB, limit 9007199254740988kB, failcnt 0 [...] [7516987.983293] [ pid ] uid tgid total_vm rss pgtables_bytes swapents oom_score_adj name [7516987.983510] [ 5740] 0 5740 257 1 32768 0 -998 pause [7516987.983574] [58804] 0 58804 4594 771 81920 0 -998 entry_point.bas [7516987.983577] [58908] 0 58908 7089 689 98304 0 -998 cron [7516987.983580] [58910] 0 58910 16235 5576 163840 0 -998 supervisord [7516987.983590] [59620] 0 59620 18074 1395 188416 0 -998 sshd [7516987.983594] [59622] 0 59622 18680 6679 188416 0 -998 python [7516987.983598] [59624] 0 59624 1859266 5161 548864 0 -998 odin-agent [7516987.983600] [59625] 0 59625 707223 9248 983040 0 -998 filebeat [7516987.983604] [59627] 0 59627 416433 64239 774144 0 -998 odin-log-agent [7516987.983607] [59631] 0 59631 180671 15012 385024 0 -998 python3 [7516987.983612] [61396] 0 61396 791287 3189 352256 0 -998 client [7516987.983615] [61641] 0 61641 1844642 29089 946176 0 -998 client [7516987.983765] [ 9236] 0 9236 2642 467 53248 0 -998 php_scanner [7516987.983911] [42898] 0 42898 15543 838 167936 0 -998 su [7516987.983915] [42900] 1000 42900 3673 867 77824 0 -998 exec_script_vr2 [7516987.983918] [42925] 1000 42925 36475 19033 335872 0 -998 python [7516987.983921] [57146] 1000 57146 3673 848 73728 0 -998 exec_script_J2p [7516987.983925] [57195] 1000 57195 186359 22958 491520 0 -998 python2 [7516987.983928] [58376] 1000 58376 275764 14402 290816 0 -998 rosmaster [7516987.983931] [58395] 1000 58395 155166 4449 245760 0 -998 rosout [7516987.983935] [58406] 1000 58406 18285584 3967322 37101568 0 -998 data_sim [7516987.984221] oom-kill:constraint=CONSTRAINT_MEMCG,nodemask=(null),cpuset=3aa16c9482ae3a6f6b78bda68a55d32c87c99b985e0f11331cddf05af6c4d753,mems_allowed=0-1,oom_memcg=/kubepods/podf1c273d3-9b36-11ea-b3df-246e9693c184,task_memcg=/kubepods/podf1c273d3-9b36-11ea-b3df-246e9693c184/1f246a3eeea8f70bf91141eeaf1805346a666e225f823906485ea0b6c37dfc3d,task=pause,pid=5740,uid=0 [7516987.984254] Memory cgroup out of memory: Killed process 5740 (pause) total-vm:1028kB, anon-rss:4kB, file-rss:0kB, shmem-rss:0kB [7516988.092344] oom_reaper: reaped process 5740 (pause), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kB We can find that the first scanned process 5740 (pause) was killed, but its rss is only one page. That is because, when we calculate the oom badness in oom_badness(), we always ignore the negtive point and convert all of these negtive points to 1. Now as oom_score_adj of all the processes in this targeted memcg have the same value -998, the points of these processes are all negtive value. As a result, the first scanned process will be killed. The oom_socre_adj (-998) in this memcg is set by kubelet, because it is a a Guaranteed pod, which has higher priority to prevent from being killed by system oom. To fix this issue, we should make the calculation of oom point more accurate. We can achieve it by convert the chosen_point from 'unsigned long' to 'long'. [cai@lca.pw: reported a issue in the previous version] [mhocko@suse.com: fixed the issue reported by Cai] [mhocko@suse.com: add the comment in proc_oom_score()] [laoar.shao@gmail.com: v3] Link: http://lkml.kernel.org/r/1594396651-9931-1-git-send-email-laoar.shao@gmail.com Signed-off-by: Yafang Shao <laoar.shao@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Tested-by: Naresh Kamboju <naresh.kamboju@linaro.org> Acked-by: Michal Hocko <mhocko@suse.com> Cc: David Rientjes <rientjes@google.com> Cc: Qian Cai <cai@lca.pw> Link: http://lkml.kernel.org/r/1594309987-9919-1-git-send-email-laoar.shao@gmail.com Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-08-12 09:31:22 +08:00
return LONG_MIN;
oom: introduce find_lock_task_mm() to fix !mm false positives Almost all ->mm == NULL checks in oom_kill.c are wrong. The current code assumes that the task without ->mm has already released its memory and ignores the process. However this is not necessarily true when this process is multithreaded, other live sub-threads can use this ->mm. - Remove the "if (!p->mm)" check in select_bad_process(), it is just wrong. - Add the new helper, find_lock_task_mm(), which finds the live thread which uses the memory and takes task_lock() to pin ->mm - change oom_badness() to use this helper instead of just checking ->mm != NULL. - As David pointed out, select_bad_process() must never choose the task without ->mm, but no matter what oom_badness() returns the task can be chosen if nothing else has been found yet. Change oom_badness() to return int, change it to return -1 if find_lock_task_mm() fails, and change select_bad_process() to check points >= 0. Note! This patch is not enough, we need more changes. - oom_badness() was fixed, but oom_kill_task() still ignores the task without ->mm - oom_forkbomb_penalty() should use find_lock_task_mm() too, and it also needs other changes to actually find the first first-descendant children This will be addressed later. [kosaki.motohiro@jp.fujitsu.com: use in badness(), __oom_kill_task()] Signed-off-by: Oleg Nesterov <oleg@redhat.com> Signed-off-by: David Rientjes <rientjes@google.com> Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 08:18:45 +08:00
p = find_lock_task_mm(p);
if (!p)
mm, oom: make the calculation of oom badness more accurate Recently we found an issue on our production environment that when memcg oom is triggered the oom killer doesn't chose the process with largest resident memory but chose the first scanned process. Note that all processes in this memcg have the same oom_score_adj, so the oom killer should chose the process with largest resident memory. Bellow is part of the oom info, which is enough to analyze this issue. [7516987.983223] memory: usage 16777216kB, limit 16777216kB, failcnt 52843037 [7516987.983224] memory+swap: usage 16777216kB, limit 9007199254740988kB, failcnt 0 [7516987.983225] kmem: usage 301464kB, limit 9007199254740988kB, failcnt 0 [...] [7516987.983293] [ pid ] uid tgid total_vm rss pgtables_bytes swapents oom_score_adj name [7516987.983510] [ 5740] 0 5740 257 1 32768 0 -998 pause [7516987.983574] [58804] 0 58804 4594 771 81920 0 -998 entry_point.bas [7516987.983577] [58908] 0 58908 7089 689 98304 0 -998 cron [7516987.983580] [58910] 0 58910 16235 5576 163840 0 -998 supervisord [7516987.983590] [59620] 0 59620 18074 1395 188416 0 -998 sshd [7516987.983594] [59622] 0 59622 18680 6679 188416 0 -998 python [7516987.983598] [59624] 0 59624 1859266 5161 548864 0 -998 odin-agent [7516987.983600] [59625] 0 59625 707223 9248 983040 0 -998 filebeat [7516987.983604] [59627] 0 59627 416433 64239 774144 0 -998 odin-log-agent [7516987.983607] [59631] 0 59631 180671 15012 385024 0 -998 python3 [7516987.983612] [61396] 0 61396 791287 3189 352256 0 -998 client [7516987.983615] [61641] 0 61641 1844642 29089 946176 0 -998 client [7516987.983765] [ 9236] 0 9236 2642 467 53248 0 -998 php_scanner [7516987.983911] [42898] 0 42898 15543 838 167936 0 -998 su [7516987.983915] [42900] 1000 42900 3673 867 77824 0 -998 exec_script_vr2 [7516987.983918] [42925] 1000 42925 36475 19033 335872 0 -998 python [7516987.983921] [57146] 1000 57146 3673 848 73728 0 -998 exec_script_J2p [7516987.983925] [57195] 1000 57195 186359 22958 491520 0 -998 python2 [7516987.983928] [58376] 1000 58376 275764 14402 290816 0 -998 rosmaster [7516987.983931] [58395] 1000 58395 155166 4449 245760 0 -998 rosout [7516987.983935] [58406] 1000 58406 18285584 3967322 37101568 0 -998 data_sim [7516987.984221] oom-kill:constraint=CONSTRAINT_MEMCG,nodemask=(null),cpuset=3aa16c9482ae3a6f6b78bda68a55d32c87c99b985e0f11331cddf05af6c4d753,mems_allowed=0-1,oom_memcg=/kubepods/podf1c273d3-9b36-11ea-b3df-246e9693c184,task_memcg=/kubepods/podf1c273d3-9b36-11ea-b3df-246e9693c184/1f246a3eeea8f70bf91141eeaf1805346a666e225f823906485ea0b6c37dfc3d,task=pause,pid=5740,uid=0 [7516987.984254] Memory cgroup out of memory: Killed process 5740 (pause) total-vm:1028kB, anon-rss:4kB, file-rss:0kB, shmem-rss:0kB [7516988.092344] oom_reaper: reaped process 5740 (pause), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kB We can find that the first scanned process 5740 (pause) was killed, but its rss is only one page. That is because, when we calculate the oom badness in oom_badness(), we always ignore the negtive point and convert all of these negtive points to 1. Now as oom_score_adj of all the processes in this targeted memcg have the same value -998, the points of these processes are all negtive value. As a result, the first scanned process will be killed. The oom_socre_adj (-998) in this memcg is set by kubelet, because it is a a Guaranteed pod, which has higher priority to prevent from being killed by system oom. To fix this issue, we should make the calculation of oom point more accurate. We can achieve it by convert the chosen_point from 'unsigned long' to 'long'. [cai@lca.pw: reported a issue in the previous version] [mhocko@suse.com: fixed the issue reported by Cai] [mhocko@suse.com: add the comment in proc_oom_score()] [laoar.shao@gmail.com: v3] Link: http://lkml.kernel.org/r/1594396651-9931-1-git-send-email-laoar.shao@gmail.com Signed-off-by: Yafang Shao <laoar.shao@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Tested-by: Naresh Kamboju <naresh.kamboju@linaro.org> Acked-by: Michal Hocko <mhocko@suse.com> Cc: David Rientjes <rientjes@google.com> Cc: Qian Cai <cai@lca.pw> Link: http://lkml.kernel.org/r/1594309987-9919-1-git-send-email-laoar.shao@gmail.com Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-08-12 09:31:22 +08:00
return LONG_MIN;
/*
* Do not even consider tasks which are explicitly marked oom
* unkillable or have been already oom reaped or the are in
* the middle of vfork
*/
adj = (long)p->signal->oom_score_adj;
if (adj == OOM_SCORE_ADJ_MIN ||
test_bit(MMF_OOM_SKIP, &p->mm->flags) ||
in_vfork(p)) {
task_unlock(p);
mm, oom: make the calculation of oom badness more accurate Recently we found an issue on our production environment that when memcg oom is triggered the oom killer doesn't chose the process with largest resident memory but chose the first scanned process. Note that all processes in this memcg have the same oom_score_adj, so the oom killer should chose the process with largest resident memory. Bellow is part of the oom info, which is enough to analyze this issue. [7516987.983223] memory: usage 16777216kB, limit 16777216kB, failcnt 52843037 [7516987.983224] memory+swap: usage 16777216kB, limit 9007199254740988kB, failcnt 0 [7516987.983225] kmem: usage 301464kB, limit 9007199254740988kB, failcnt 0 [...] [7516987.983293] [ pid ] uid tgid total_vm rss pgtables_bytes swapents oom_score_adj name [7516987.983510] [ 5740] 0 5740 257 1 32768 0 -998 pause [7516987.983574] [58804] 0 58804 4594 771 81920 0 -998 entry_point.bas [7516987.983577] [58908] 0 58908 7089 689 98304 0 -998 cron [7516987.983580] [58910] 0 58910 16235 5576 163840 0 -998 supervisord [7516987.983590] [59620] 0 59620 18074 1395 188416 0 -998 sshd [7516987.983594] [59622] 0 59622 18680 6679 188416 0 -998 python [7516987.983598] [59624] 0 59624 1859266 5161 548864 0 -998 odin-agent [7516987.983600] [59625] 0 59625 707223 9248 983040 0 -998 filebeat [7516987.983604] [59627] 0 59627 416433 64239 774144 0 -998 odin-log-agent [7516987.983607] [59631] 0 59631 180671 15012 385024 0 -998 python3 [7516987.983612] [61396] 0 61396 791287 3189 352256 0 -998 client [7516987.983615] [61641] 0 61641 1844642 29089 946176 0 -998 client [7516987.983765] [ 9236] 0 9236 2642 467 53248 0 -998 php_scanner [7516987.983911] [42898] 0 42898 15543 838 167936 0 -998 su [7516987.983915] [42900] 1000 42900 3673 867 77824 0 -998 exec_script_vr2 [7516987.983918] [42925] 1000 42925 36475 19033 335872 0 -998 python [7516987.983921] [57146] 1000 57146 3673 848 73728 0 -998 exec_script_J2p [7516987.983925] [57195] 1000 57195 186359 22958 491520 0 -998 python2 [7516987.983928] [58376] 1000 58376 275764 14402 290816 0 -998 rosmaster [7516987.983931] [58395] 1000 58395 155166 4449 245760 0 -998 rosout [7516987.983935] [58406] 1000 58406 18285584 3967322 37101568 0 -998 data_sim [7516987.984221] oom-kill:constraint=CONSTRAINT_MEMCG,nodemask=(null),cpuset=3aa16c9482ae3a6f6b78bda68a55d32c87c99b985e0f11331cddf05af6c4d753,mems_allowed=0-1,oom_memcg=/kubepods/podf1c273d3-9b36-11ea-b3df-246e9693c184,task_memcg=/kubepods/podf1c273d3-9b36-11ea-b3df-246e9693c184/1f246a3eeea8f70bf91141eeaf1805346a666e225f823906485ea0b6c37dfc3d,task=pause,pid=5740,uid=0 [7516987.984254] Memory cgroup out of memory: Killed process 5740 (pause) total-vm:1028kB, anon-rss:4kB, file-rss:0kB, shmem-rss:0kB [7516988.092344] oom_reaper: reaped process 5740 (pause), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kB We can find that the first scanned process 5740 (pause) was killed, but its rss is only one page. That is because, when we calculate the oom badness in oom_badness(), we always ignore the negtive point and convert all of these negtive points to 1. Now as oom_score_adj of all the processes in this targeted memcg have the same value -998, the points of these processes are all negtive value. As a result, the first scanned process will be killed. The oom_socre_adj (-998) in this memcg is set by kubelet, because it is a a Guaranteed pod, which has higher priority to prevent from being killed by system oom. To fix this issue, we should make the calculation of oom point more accurate. We can achieve it by convert the chosen_point from 'unsigned long' to 'long'. [cai@lca.pw: reported a issue in the previous version] [mhocko@suse.com: fixed the issue reported by Cai] [mhocko@suse.com: add the comment in proc_oom_score()] [laoar.shao@gmail.com: v3] Link: http://lkml.kernel.org/r/1594396651-9931-1-git-send-email-laoar.shao@gmail.com Signed-off-by: Yafang Shao <laoar.shao@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Tested-by: Naresh Kamboju <naresh.kamboju@linaro.org> Acked-by: Michal Hocko <mhocko@suse.com> Cc: David Rientjes <rientjes@google.com> Cc: Qian Cai <cai@lca.pw> Link: http://lkml.kernel.org/r/1594309987-9919-1-git-send-email-laoar.shao@gmail.com Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-08-12 09:31:22 +08:00
return LONG_MIN;
}
/*
oom: badness heuristic rewrite This a complete rewrite of the oom killer's badness() heuristic which is used to determine which task to kill in oom conditions. The goal is to make it as simple and predictable as possible so the results are better understood and we end up killing the task which will lead to the most memory freeing while still respecting the fine-tuning from userspace. Instead of basing the heuristic on mm->total_vm for each task, the task's rss and swap space is used instead. This is a better indication of the amount of memory that will be freeable if the oom killed task is chosen and subsequently exits. This helps specifically in cases where KDE or GNOME is chosen for oom kill on desktop systems instead of a memory hogging task. The baseline for the heuristic is a proportion of memory that each task is currently using in memory plus swap compared to the amount of "allowable" memory. "Allowable," in this sense, means the system-wide resources for unconstrained oom conditions, the set of mempolicy nodes, the mems attached to current's cpuset, or a memory controller's limit. The proportion is given on a scale of 0 (never kill) to 1000 (always kill), roughly meaning that if a task has a badness() score of 500 that the task consumes approximately 50% of allowable memory resident in RAM or in swap space. The proportion is always relative to the amount of "allowable" memory and not the total amount of RAM systemwide so that mempolicies and cpusets may operate in isolation; they shall not need to know the true size of the machine on which they are running if they are bound to a specific set of nodes or mems, respectively. Root tasks are given 3% extra memory just like __vm_enough_memory() provides in LSMs. In the event of two tasks consuming similar amounts of memory, it is generally better to save root's task. Because of the change in the badness() heuristic's baseline, it is also necessary to introduce a new user interface to tune it. It's not possible to redefine the meaning of /proc/pid/oom_adj with a new scale since the ABI cannot be changed for backward compatability. Instead, a new tunable, /proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may be used to polarize the heuristic such that certain tasks are never considered for oom kill while others may always be considered. The value is added directly into the badness() score so a value of -500, for example, means to discount 50% of its memory consumption in comparison to other tasks either on the system, bound to the mempolicy, in the cpuset, or sharing the same memory controller. /proc/pid/oom_adj is changed so that its meaning is rescaled into the units used by /proc/pid/oom_score_adj, and vice versa. Changing one of these per-task tunables will rescale the value of the other to an equivalent meaning. Although /proc/pid/oom_adj was originally defined as a bitshift on the badness score, it now shares the same linear growth as /proc/pid/oom_score_adj but with different granularity. This is required so the ABI is not broken with userspace applications and allows oom_adj to be deprecated for future removal. Signed-off-by: David Rientjes <rientjes@google.com> Cc: Nick Piggin <npiggin@suse.de> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Balbir Singh <balbir@in.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 08:19:46 +08:00
* The baseline for the badness score is the proportion of RAM that each
* task's rss, pagetable and swap space use.
*/
mm: account pmd page tables to the process Dave noticed that unprivileged process can allocate significant amount of memory -- >500 MiB on x86_64 -- and stay unnoticed by oom-killer and memory cgroup. The trick is to allocate a lot of PMD page tables. Linux kernel doesn't account PMD tables to the process, only PTE. The use-cases below use few tricks to allocate a lot of PMD page tables while keeping VmRSS and VmPTE low. oom_score for the process will be 0. #include <errno.h> #include <stdio.h> #include <stdlib.h> #include <unistd.h> #include <sys/mman.h> #include <sys/prctl.h> #define PUD_SIZE (1UL << 30) #define PMD_SIZE (1UL << 21) #define NR_PUD 130000 int main(void) { char *addr = NULL; unsigned long i; prctl(PR_SET_THP_DISABLE); for (i = 0; i < NR_PUD ; i++) { addr = mmap(addr + PUD_SIZE, PUD_SIZE, PROT_WRITE|PROT_READ, MAP_ANONYMOUS|MAP_PRIVATE, -1, 0); if (addr == MAP_FAILED) { perror("mmap"); break; } *addr = 'x'; munmap(addr, PMD_SIZE); mmap(addr, PMD_SIZE, PROT_WRITE|PROT_READ, MAP_ANONYMOUS|MAP_PRIVATE|MAP_FIXED, -1, 0); if (addr == MAP_FAILED) perror("re-mmap"), exit(1); } printf("PID %d consumed %lu KiB in PMD page tables\n", getpid(), i * 4096 >> 10); return pause(); } The patch addresses the issue by account PMD tables to the process the same way we account PTE. The main place where PMD tables is accounted is __pmd_alloc() and free_pmd_range(). But there're few corner cases: - HugeTLB can share PMD page tables. The patch handles by accounting the table to all processes who share it. - x86 PAE pre-allocates few PMD tables on fork. - Architectures with FIRST_USER_ADDRESS > 0. We need to adjust sanity check on exit(2). Accounting only happens on configuration where PMD page table's level is present (PMD is not folded). As with nr_ptes we use per-mm counter. The counter value is used to calculate baseline for badness score by oom-killer. Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Reported-by: Dave Hansen <dave.hansen@linux.intel.com> Cc: Hugh Dickins <hughd@google.com> Reviewed-by: Cyrill Gorcunov <gorcunov@openvz.org> Cc: Pavel Emelyanov <xemul@openvz.org> Cc: David Rientjes <rientjes@google.com> Tested-by: Sedat Dilek <sedat.dilek@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-02-12 07:26:50 +08:00
points = get_mm_rss(p->mm) + get_mm_counter(p->mm, MM_SWAPENTS) +
mm: consolidate page table accounting Currently, we account page tables separately for each page table level, but that's redundant -- we only make use of total memory allocated to page tables for oom_badness calculation. We also provide the information to userspace, but it has dubious value there too. This patch switches page table accounting to single counter. mm->pgtables_bytes is now used to account all page table levels. We use bytes, because page table size for different levels of page table tree may be different. The change has user-visible effect: we don't have VmPMD and VmPUD reported in /proc/[pid]/status. Not sure if anybody uses them. (As alternative, we can always report 0 kB for them.) OOM-killer report is also slightly changed: we now report pgtables_bytes instead of nr_ptes, nr_pmd, nr_puds. Apart from reducing number of counters per-mm, the benefit is that we now calculate oom_badness() more correctly for machines which have different size of page tables depending on level or where page tables are less than a page in size. The only downside can be debuggability because we do not know which page table level could leak. But I do not remember many bugs that would be caught by separate counters so I wouldn't lose sleep over this. [akpm@linux-foundation.org: fix mm/huge_memory.c] Link: http://lkml.kernel.org/r/20171006100651.44742-2-kirill.shutemov@linux.intel.com Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> [kirill.shutemov@linux.intel.com: fix build] Link: http://lkml.kernel.org/r/20171016150113.ikfxy3e7zzfvsr4w@black.fi.intel.com Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-16 09:35:40 +08:00
mm_pgtables_bytes(p->mm) / PAGE_SIZE;
oom: badness heuristic rewrite This a complete rewrite of the oom killer's badness() heuristic which is used to determine which task to kill in oom conditions. The goal is to make it as simple and predictable as possible so the results are better understood and we end up killing the task which will lead to the most memory freeing while still respecting the fine-tuning from userspace. Instead of basing the heuristic on mm->total_vm for each task, the task's rss and swap space is used instead. This is a better indication of the amount of memory that will be freeable if the oom killed task is chosen and subsequently exits. This helps specifically in cases where KDE or GNOME is chosen for oom kill on desktop systems instead of a memory hogging task. The baseline for the heuristic is a proportion of memory that each task is currently using in memory plus swap compared to the amount of "allowable" memory. "Allowable," in this sense, means the system-wide resources for unconstrained oom conditions, the set of mempolicy nodes, the mems attached to current's cpuset, or a memory controller's limit. The proportion is given on a scale of 0 (never kill) to 1000 (always kill), roughly meaning that if a task has a badness() score of 500 that the task consumes approximately 50% of allowable memory resident in RAM or in swap space. The proportion is always relative to the amount of "allowable" memory and not the total amount of RAM systemwide so that mempolicies and cpusets may operate in isolation; they shall not need to know the true size of the machine on which they are running if they are bound to a specific set of nodes or mems, respectively. Root tasks are given 3% extra memory just like __vm_enough_memory() provides in LSMs. In the event of two tasks consuming similar amounts of memory, it is generally better to save root's task. Because of the change in the badness() heuristic's baseline, it is also necessary to introduce a new user interface to tune it. It's not possible to redefine the meaning of /proc/pid/oom_adj with a new scale since the ABI cannot be changed for backward compatability. Instead, a new tunable, /proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may be used to polarize the heuristic such that certain tasks are never considered for oom kill while others may always be considered. The value is added directly into the badness() score so a value of -500, for example, means to discount 50% of its memory consumption in comparison to other tasks either on the system, bound to the mempolicy, in the cpuset, or sharing the same memory controller. /proc/pid/oom_adj is changed so that its meaning is rescaled into the units used by /proc/pid/oom_score_adj, and vice versa. Changing one of these per-task tunables will rescale the value of the other to an equivalent meaning. Although /proc/pid/oom_adj was originally defined as a bitshift on the badness score, it now shares the same linear growth as /proc/pid/oom_score_adj but with different granularity. This is required so the ABI is not broken with userspace applications and allows oom_adj to be deprecated for future removal. Signed-off-by: David Rientjes <rientjes@google.com> Cc: Nick Piggin <npiggin@suse.de> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Balbir Singh <balbir@in.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 08:19:46 +08:00
task_unlock(p);
/* Normalize to oom_score_adj units */
adj *= totalpages / 1000;
points += adj;
mm, oom: make the calculation of oom badness more accurate Recently we found an issue on our production environment that when memcg oom is triggered the oom killer doesn't chose the process with largest resident memory but chose the first scanned process. Note that all processes in this memcg have the same oom_score_adj, so the oom killer should chose the process with largest resident memory. Bellow is part of the oom info, which is enough to analyze this issue. [7516987.983223] memory: usage 16777216kB, limit 16777216kB, failcnt 52843037 [7516987.983224] memory+swap: usage 16777216kB, limit 9007199254740988kB, failcnt 0 [7516987.983225] kmem: usage 301464kB, limit 9007199254740988kB, failcnt 0 [...] [7516987.983293] [ pid ] uid tgid total_vm rss pgtables_bytes swapents oom_score_adj name [7516987.983510] [ 5740] 0 5740 257 1 32768 0 -998 pause [7516987.983574] [58804] 0 58804 4594 771 81920 0 -998 entry_point.bas [7516987.983577] [58908] 0 58908 7089 689 98304 0 -998 cron [7516987.983580] [58910] 0 58910 16235 5576 163840 0 -998 supervisord [7516987.983590] [59620] 0 59620 18074 1395 188416 0 -998 sshd [7516987.983594] [59622] 0 59622 18680 6679 188416 0 -998 python [7516987.983598] [59624] 0 59624 1859266 5161 548864 0 -998 odin-agent [7516987.983600] [59625] 0 59625 707223 9248 983040 0 -998 filebeat [7516987.983604] [59627] 0 59627 416433 64239 774144 0 -998 odin-log-agent [7516987.983607] [59631] 0 59631 180671 15012 385024 0 -998 python3 [7516987.983612] [61396] 0 61396 791287 3189 352256 0 -998 client [7516987.983615] [61641] 0 61641 1844642 29089 946176 0 -998 client [7516987.983765] [ 9236] 0 9236 2642 467 53248 0 -998 php_scanner [7516987.983911] [42898] 0 42898 15543 838 167936 0 -998 su [7516987.983915] [42900] 1000 42900 3673 867 77824 0 -998 exec_script_vr2 [7516987.983918] [42925] 1000 42925 36475 19033 335872 0 -998 python [7516987.983921] [57146] 1000 57146 3673 848 73728 0 -998 exec_script_J2p [7516987.983925] [57195] 1000 57195 186359 22958 491520 0 -998 python2 [7516987.983928] [58376] 1000 58376 275764 14402 290816 0 -998 rosmaster [7516987.983931] [58395] 1000 58395 155166 4449 245760 0 -998 rosout [7516987.983935] [58406] 1000 58406 18285584 3967322 37101568 0 -998 data_sim [7516987.984221] oom-kill:constraint=CONSTRAINT_MEMCG,nodemask=(null),cpuset=3aa16c9482ae3a6f6b78bda68a55d32c87c99b985e0f11331cddf05af6c4d753,mems_allowed=0-1,oom_memcg=/kubepods/podf1c273d3-9b36-11ea-b3df-246e9693c184,task_memcg=/kubepods/podf1c273d3-9b36-11ea-b3df-246e9693c184/1f246a3eeea8f70bf91141eeaf1805346a666e225f823906485ea0b6c37dfc3d,task=pause,pid=5740,uid=0 [7516987.984254] Memory cgroup out of memory: Killed process 5740 (pause) total-vm:1028kB, anon-rss:4kB, file-rss:0kB, shmem-rss:0kB [7516988.092344] oom_reaper: reaped process 5740 (pause), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kB We can find that the first scanned process 5740 (pause) was killed, but its rss is only one page. That is because, when we calculate the oom badness in oom_badness(), we always ignore the negtive point and convert all of these negtive points to 1. Now as oom_score_adj of all the processes in this targeted memcg have the same value -998, the points of these processes are all negtive value. As a result, the first scanned process will be killed. The oom_socre_adj (-998) in this memcg is set by kubelet, because it is a a Guaranteed pod, which has higher priority to prevent from being killed by system oom. To fix this issue, we should make the calculation of oom point more accurate. We can achieve it by convert the chosen_point from 'unsigned long' to 'long'. [cai@lca.pw: reported a issue in the previous version] [mhocko@suse.com: fixed the issue reported by Cai] [mhocko@suse.com: add the comment in proc_oom_score()] [laoar.shao@gmail.com: v3] Link: http://lkml.kernel.org/r/1594396651-9931-1-git-send-email-laoar.shao@gmail.com Signed-off-by: Yafang Shao <laoar.shao@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Tested-by: Naresh Kamboju <naresh.kamboju@linaro.org> Acked-by: Michal Hocko <mhocko@suse.com> Cc: David Rientjes <rientjes@google.com> Cc: Qian Cai <cai@lca.pw> Link: http://lkml.kernel.org/r/1594309987-9919-1-git-send-email-laoar.shao@gmail.com Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-08-12 09:31:22 +08:00
return points;
}
mm, oom: reorganize the oom report in dump_header OOM report contains several sections. The first one is the allocation context that has triggered the OOM. Then we have cpuset context followed by the stack trace of the OOM path. The tird one is the OOM memory information. Followed by the current memory state of all system tasks. At last, we will show oom eligible tasks and the information about the chosen oom victim. One thing that makes parsing more awkward than necessary is that we do not have a single and easily parsable line about the oom context. This patch is reorganizing the oom report to 1) who invoked oom and what was the allocation request [ 515.902945] tuned invoked oom-killer: gfp_mask=0x6200ca(GFP_HIGHUSER_MOVABLE), order=0, oom_score_adj=0 2) OOM stack trace [ 515.904273] CPU: 24 PID: 1809 Comm: tuned Not tainted 4.20.0-rc3+ #3 [ 515.905518] Hardware name: Inspur SA5212M4/YZMB-00370-107, BIOS 4.1.10 11/14/2016 [ 515.906821] Call Trace: [ 515.908062] dump_stack+0x5a/0x73 [ 515.909311] dump_header+0x55/0x28c [ 515.914260] oom_kill_process+0x2d8/0x300 [ 515.916708] out_of_memory+0x145/0x4a0 [ 515.917932] __alloc_pages_slowpath+0x7d2/0xa16 [ 515.919157] __alloc_pages_nodemask+0x277/0x290 [ 515.920367] filemap_fault+0x3d0/0x6c0 [ 515.921529] ? filemap_map_pages+0x2b8/0x420 [ 515.922709] ext4_filemap_fault+0x2c/0x40 [ext4] [ 515.923884] __do_fault+0x20/0x80 [ 515.925032] __handle_mm_fault+0xbc0/0xe80 [ 515.926195] handle_mm_fault+0xfa/0x210 [ 515.927357] __do_page_fault+0x233/0x4c0 [ 515.928506] do_page_fault+0x32/0x140 [ 515.929646] ? page_fault+0x8/0x30 [ 515.930770] page_fault+0x1e/0x30 3) OOM memory information [ 515.958093] Mem-Info: [ 515.959647] active_anon:26501758 inactive_anon:1179809 isolated_anon:0 active_file:4402672 inactive_file:483963 isolated_file:1344 unevictable:0 dirty:4886753 writeback:0 unstable:0 slab_reclaimable:148442 slab_unreclaimable:18741 mapped:1347 shmem:1347 pagetables:58669 bounce:0 free:88663 free_pcp:0 free_cma:0 ... 4) current memory state of all system tasks [ 516.079544] [ 744] 0 744 9211 1345 114688 82 0 systemd-journal [ 516.082034] [ 787] 0 787 31764 0 143360 92 0 lvmetad [ 516.084465] [ 792] 0 792 10930 1 110592 208 -1000 systemd-udevd [ 516.086865] [ 1199] 0 1199 13866 0 131072 112 -1000 auditd [ 516.089190] [ 1222] 0 1222 31990 1 110592 157 0 smartd [ 516.091477] [ 1225] 0 1225 4864 85 81920 43 0 irqbalance [ 516.093712] [ 1226] 0 1226 52612 0 258048 426 0 abrtd [ 516.112128] [ 1280] 0 1280 109774 55 299008 400 0 NetworkManager [ 516.113998] [ 1295] 0 1295 28817 37 69632 24 0 ksmtuned [ 516.144596] [ 10718] 0 10718 2622484 1721372 15998976 267219 0 panic [ 516.145792] [ 10719] 0 10719 2622484 1164767 9818112 53576 0 panic [ 516.146977] [ 10720] 0 10720 2622484 1174361 9904128 53709 0 panic [ 516.148163] [ 10721] 0 10721 2622484 1209070 10194944 54824 0 panic [ 516.149329] [ 10722] 0 10722 2622484 1745799 14774272 91138 0 panic 5) oom context (contrains and the chosen victim). oom-kill:constraint=CONSTRAINT_NONE,nodemask=(null),cpuset=/,mems_allowed=0-1,task=panic,pid=10737,uid=0 An admin can easily get the full oom context at a single line which makes parsing much easier. Link: http://lkml.kernel.org/r/1542799799-36184-1-git-send-email-ufo19890607@gmail.com Signed-off-by: yuzhoujian <yuzhoujian@didichuxing.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: David Rientjes <rientjes@google.com> Cc: "Kirill A . Shutemov" <kirill.shutemov@linux.intel.com> Cc: Roman Gushchin <guro@fb.com> Cc: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp> Cc: Yang Shi <yang.s@alibaba-inc.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-12-28 16:36:07 +08:00
static const char * const oom_constraint_text[] = {
[CONSTRAINT_NONE] = "CONSTRAINT_NONE",
[CONSTRAINT_CPUSET] = "CONSTRAINT_CPUSET",
[CONSTRAINT_MEMORY_POLICY] = "CONSTRAINT_MEMORY_POLICY",
[CONSTRAINT_MEMCG] = "CONSTRAINT_MEMCG",
};
/*
* Determine the type of allocation constraint.
*/
static enum oom_constraint constrained_alloc(struct oom_control *oc)
{
struct zone *zone;
mm: have zonelist contains structs with both a zone pointer and zone_idx Filtering zonelists requires very frequent use of zone_idx(). This is costly as it involves a lookup of another structure and a substraction operation. As the zone_idx is often required, it should be quickly accessible. The node idx could also be stored here if it was found that accessing zone->node is significant which may be the case on workloads where nodemasks are heavily used. This patch introduces a struct zoneref to store a zone pointer and a zone index. The zonelist then consists of an array of these struct zonerefs which are looked up as necessary. Helpers are given for accessing the zone index as well as the node index. [kamezawa.hiroyu@jp.fujitsu.com: Suggested struct zoneref instead of embedding information in pointers] [hugh@veritas.com: mm-have-zonelist: fix memcg ooms] [hugh@veritas.com: just return do_try_to_free_pages] [hugh@veritas.com: do_try_to_free_pages gfp_mask redundant] Signed-off-by: Mel Gorman <mel@csn.ul.ie> Acked-by: Christoph Lameter <clameter@sgi.com> Acked-by: David Rientjes <rientjes@google.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Mel Gorman <mel@csn.ul.ie> Cc: Christoph Lameter <clameter@sgi.com> Cc: Nick Piggin <nickpiggin@yahoo.com.au> Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-28 17:12:17 +08:00
struct zoneref *z;
enum zone_type highest_zoneidx = gfp_zone(oc->gfp_mask);
oom: badness heuristic rewrite This a complete rewrite of the oom killer's badness() heuristic which is used to determine which task to kill in oom conditions. The goal is to make it as simple and predictable as possible so the results are better understood and we end up killing the task which will lead to the most memory freeing while still respecting the fine-tuning from userspace. Instead of basing the heuristic on mm->total_vm for each task, the task's rss and swap space is used instead. This is a better indication of the amount of memory that will be freeable if the oom killed task is chosen and subsequently exits. This helps specifically in cases where KDE or GNOME is chosen for oom kill on desktop systems instead of a memory hogging task. The baseline for the heuristic is a proportion of memory that each task is currently using in memory plus swap compared to the amount of "allowable" memory. "Allowable," in this sense, means the system-wide resources for unconstrained oom conditions, the set of mempolicy nodes, the mems attached to current's cpuset, or a memory controller's limit. The proportion is given on a scale of 0 (never kill) to 1000 (always kill), roughly meaning that if a task has a badness() score of 500 that the task consumes approximately 50% of allowable memory resident in RAM or in swap space. The proportion is always relative to the amount of "allowable" memory and not the total amount of RAM systemwide so that mempolicies and cpusets may operate in isolation; they shall not need to know the true size of the machine on which they are running if they are bound to a specific set of nodes or mems, respectively. Root tasks are given 3% extra memory just like __vm_enough_memory() provides in LSMs. In the event of two tasks consuming similar amounts of memory, it is generally better to save root's task. Because of the change in the badness() heuristic's baseline, it is also necessary to introduce a new user interface to tune it. It's not possible to redefine the meaning of /proc/pid/oom_adj with a new scale since the ABI cannot be changed for backward compatability. Instead, a new tunable, /proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may be used to polarize the heuristic such that certain tasks are never considered for oom kill while others may always be considered. The value is added directly into the badness() score so a value of -500, for example, means to discount 50% of its memory consumption in comparison to other tasks either on the system, bound to the mempolicy, in the cpuset, or sharing the same memory controller. /proc/pid/oom_adj is changed so that its meaning is rescaled into the units used by /proc/pid/oom_score_adj, and vice versa. Changing one of these per-task tunables will rescale the value of the other to an equivalent meaning. Although /proc/pid/oom_adj was originally defined as a bitshift on the badness score, it now shares the same linear growth as /proc/pid/oom_score_adj but with different granularity. This is required so the ABI is not broken with userspace applications and allows oom_adj to be deprecated for future removal. Signed-off-by: David Rientjes <rientjes@google.com> Cc: Nick Piggin <npiggin@suse.de> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Balbir Singh <balbir@in.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 08:19:46 +08:00
bool cpuset_limited = false;
int nid;
if (is_memcg_oom(oc)) {
oc->totalpages = mem_cgroup_get_max(oc->memcg) ?: 1;
return CONSTRAINT_MEMCG;
}
oom: badness heuristic rewrite This a complete rewrite of the oom killer's badness() heuristic which is used to determine which task to kill in oom conditions. The goal is to make it as simple and predictable as possible so the results are better understood and we end up killing the task which will lead to the most memory freeing while still respecting the fine-tuning from userspace. Instead of basing the heuristic on mm->total_vm for each task, the task's rss and swap space is used instead. This is a better indication of the amount of memory that will be freeable if the oom killed task is chosen and subsequently exits. This helps specifically in cases where KDE or GNOME is chosen for oom kill on desktop systems instead of a memory hogging task. The baseline for the heuristic is a proportion of memory that each task is currently using in memory plus swap compared to the amount of "allowable" memory. "Allowable," in this sense, means the system-wide resources for unconstrained oom conditions, the set of mempolicy nodes, the mems attached to current's cpuset, or a memory controller's limit. The proportion is given on a scale of 0 (never kill) to 1000 (always kill), roughly meaning that if a task has a badness() score of 500 that the task consumes approximately 50% of allowable memory resident in RAM or in swap space. The proportion is always relative to the amount of "allowable" memory and not the total amount of RAM systemwide so that mempolicies and cpusets may operate in isolation; they shall not need to know the true size of the machine on which they are running if they are bound to a specific set of nodes or mems, respectively. Root tasks are given 3% extra memory just like __vm_enough_memory() provides in LSMs. In the event of two tasks consuming similar amounts of memory, it is generally better to save root's task. Because of the change in the badness() heuristic's baseline, it is also necessary to introduce a new user interface to tune it. It's not possible to redefine the meaning of /proc/pid/oom_adj with a new scale since the ABI cannot be changed for backward compatability. Instead, a new tunable, /proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may be used to polarize the heuristic such that certain tasks are never considered for oom kill while others may always be considered. The value is added directly into the badness() score so a value of -500, for example, means to discount 50% of its memory consumption in comparison to other tasks either on the system, bound to the mempolicy, in the cpuset, or sharing the same memory controller. /proc/pid/oom_adj is changed so that its meaning is rescaled into the units used by /proc/pid/oom_score_adj, and vice versa. Changing one of these per-task tunables will rescale the value of the other to an equivalent meaning. Although /proc/pid/oom_adj was originally defined as a bitshift on the badness score, it now shares the same linear growth as /proc/pid/oom_score_adj but with different granularity. This is required so the ABI is not broken with userspace applications and allows oom_adj to be deprecated for future removal. Signed-off-by: David Rientjes <rientjes@google.com> Cc: Nick Piggin <npiggin@suse.de> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Balbir Singh <balbir@in.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 08:19:46 +08:00
/* Default to all available memory */
oc->totalpages = totalram_pages() + total_swap_pages;
if (!IS_ENABLED(CONFIG_NUMA))
return CONSTRAINT_NONE;
oom: badness heuristic rewrite This a complete rewrite of the oom killer's badness() heuristic which is used to determine which task to kill in oom conditions. The goal is to make it as simple and predictable as possible so the results are better understood and we end up killing the task which will lead to the most memory freeing while still respecting the fine-tuning from userspace. Instead of basing the heuristic on mm->total_vm for each task, the task's rss and swap space is used instead. This is a better indication of the amount of memory that will be freeable if the oom killed task is chosen and subsequently exits. This helps specifically in cases where KDE or GNOME is chosen for oom kill on desktop systems instead of a memory hogging task. The baseline for the heuristic is a proportion of memory that each task is currently using in memory plus swap compared to the amount of "allowable" memory. "Allowable," in this sense, means the system-wide resources for unconstrained oom conditions, the set of mempolicy nodes, the mems attached to current's cpuset, or a memory controller's limit. The proportion is given on a scale of 0 (never kill) to 1000 (always kill), roughly meaning that if a task has a badness() score of 500 that the task consumes approximately 50% of allowable memory resident in RAM or in swap space. The proportion is always relative to the amount of "allowable" memory and not the total amount of RAM systemwide so that mempolicies and cpusets may operate in isolation; they shall not need to know the true size of the machine on which they are running if they are bound to a specific set of nodes or mems, respectively. Root tasks are given 3% extra memory just like __vm_enough_memory() provides in LSMs. In the event of two tasks consuming similar amounts of memory, it is generally better to save root's task. Because of the change in the badness() heuristic's baseline, it is also necessary to introduce a new user interface to tune it. It's not possible to redefine the meaning of /proc/pid/oom_adj with a new scale since the ABI cannot be changed for backward compatability. Instead, a new tunable, /proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may be used to polarize the heuristic such that certain tasks are never considered for oom kill while others may always be considered. The value is added directly into the badness() score so a value of -500, for example, means to discount 50% of its memory consumption in comparison to other tasks either on the system, bound to the mempolicy, in the cpuset, or sharing the same memory controller. /proc/pid/oom_adj is changed so that its meaning is rescaled into the units used by /proc/pid/oom_score_adj, and vice versa. Changing one of these per-task tunables will rescale the value of the other to an equivalent meaning. Although /proc/pid/oom_adj was originally defined as a bitshift on the badness score, it now shares the same linear growth as /proc/pid/oom_score_adj but with different granularity. This is required so the ABI is not broken with userspace applications and allows oom_adj to be deprecated for future removal. Signed-off-by: David Rientjes <rientjes@google.com> Cc: Nick Piggin <npiggin@suse.de> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Balbir Singh <balbir@in.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 08:19:46 +08:00
if (!oc->zonelist)
oom: badness heuristic rewrite This a complete rewrite of the oom killer's badness() heuristic which is used to determine which task to kill in oom conditions. The goal is to make it as simple and predictable as possible so the results are better understood and we end up killing the task which will lead to the most memory freeing while still respecting the fine-tuning from userspace. Instead of basing the heuristic on mm->total_vm for each task, the task's rss and swap space is used instead. This is a better indication of the amount of memory that will be freeable if the oom killed task is chosen and subsequently exits. This helps specifically in cases where KDE or GNOME is chosen for oom kill on desktop systems instead of a memory hogging task. The baseline for the heuristic is a proportion of memory that each task is currently using in memory plus swap compared to the amount of "allowable" memory. "Allowable," in this sense, means the system-wide resources for unconstrained oom conditions, the set of mempolicy nodes, the mems attached to current's cpuset, or a memory controller's limit. The proportion is given on a scale of 0 (never kill) to 1000 (always kill), roughly meaning that if a task has a badness() score of 500 that the task consumes approximately 50% of allowable memory resident in RAM or in swap space. The proportion is always relative to the amount of "allowable" memory and not the total amount of RAM systemwide so that mempolicies and cpusets may operate in isolation; they shall not need to know the true size of the machine on which they are running if they are bound to a specific set of nodes or mems, respectively. Root tasks are given 3% extra memory just like __vm_enough_memory() provides in LSMs. In the event of two tasks consuming similar amounts of memory, it is generally better to save root's task. Because of the change in the badness() heuristic's baseline, it is also necessary to introduce a new user interface to tune it. It's not possible to redefine the meaning of /proc/pid/oom_adj with a new scale since the ABI cannot be changed for backward compatability. Instead, a new tunable, /proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may be used to polarize the heuristic such that certain tasks are never considered for oom kill while others may always be considered. The value is added directly into the badness() score so a value of -500, for example, means to discount 50% of its memory consumption in comparison to other tasks either on the system, bound to the mempolicy, in the cpuset, or sharing the same memory controller. /proc/pid/oom_adj is changed so that its meaning is rescaled into the units used by /proc/pid/oom_score_adj, and vice versa. Changing one of these per-task tunables will rescale the value of the other to an equivalent meaning. Although /proc/pid/oom_adj was originally defined as a bitshift on the badness score, it now shares the same linear growth as /proc/pid/oom_score_adj but with different granularity. This is required so the ABI is not broken with userspace applications and allows oom_adj to be deprecated for future removal. Signed-off-by: David Rientjes <rientjes@google.com> Cc: Nick Piggin <npiggin@suse.de> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Balbir Singh <balbir@in.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 08:19:46 +08:00
return CONSTRAINT_NONE;
/*
* Reach here only when __GFP_NOFAIL is used. So, we should avoid
* to kill current.We have to random task kill in this case.
* Hopefully, CONSTRAINT_THISNODE...but no way to handle it, now.
*/
if (oc->gfp_mask & __GFP_THISNODE)
return CONSTRAINT_NONE;
/*
oom: badness heuristic rewrite This a complete rewrite of the oom killer's badness() heuristic which is used to determine which task to kill in oom conditions. The goal is to make it as simple and predictable as possible so the results are better understood and we end up killing the task which will lead to the most memory freeing while still respecting the fine-tuning from userspace. Instead of basing the heuristic on mm->total_vm for each task, the task's rss and swap space is used instead. This is a better indication of the amount of memory that will be freeable if the oom killed task is chosen and subsequently exits. This helps specifically in cases where KDE or GNOME is chosen for oom kill on desktop systems instead of a memory hogging task. The baseline for the heuristic is a proportion of memory that each task is currently using in memory plus swap compared to the amount of "allowable" memory. "Allowable," in this sense, means the system-wide resources for unconstrained oom conditions, the set of mempolicy nodes, the mems attached to current's cpuset, or a memory controller's limit. The proportion is given on a scale of 0 (never kill) to 1000 (always kill), roughly meaning that if a task has a badness() score of 500 that the task consumes approximately 50% of allowable memory resident in RAM or in swap space. The proportion is always relative to the amount of "allowable" memory and not the total amount of RAM systemwide so that mempolicies and cpusets may operate in isolation; they shall not need to know the true size of the machine on which they are running if they are bound to a specific set of nodes or mems, respectively. Root tasks are given 3% extra memory just like __vm_enough_memory() provides in LSMs. In the event of two tasks consuming similar amounts of memory, it is generally better to save root's task. Because of the change in the badness() heuristic's baseline, it is also necessary to introduce a new user interface to tune it. It's not possible to redefine the meaning of /proc/pid/oom_adj with a new scale since the ABI cannot be changed for backward compatability. Instead, a new tunable, /proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may be used to polarize the heuristic such that certain tasks are never considered for oom kill while others may always be considered. The value is added directly into the badness() score so a value of -500, for example, means to discount 50% of its memory consumption in comparison to other tasks either on the system, bound to the mempolicy, in the cpuset, or sharing the same memory controller. /proc/pid/oom_adj is changed so that its meaning is rescaled into the units used by /proc/pid/oom_score_adj, and vice versa. Changing one of these per-task tunables will rescale the value of the other to an equivalent meaning. Although /proc/pid/oom_adj was originally defined as a bitshift on the badness score, it now shares the same linear growth as /proc/pid/oom_score_adj but with different granularity. This is required so the ABI is not broken with userspace applications and allows oom_adj to be deprecated for future removal. Signed-off-by: David Rientjes <rientjes@google.com> Cc: Nick Piggin <npiggin@suse.de> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Balbir Singh <balbir@in.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 08:19:46 +08:00
* This is not a __GFP_THISNODE allocation, so a truncated nodemask in
* the page allocator means a mempolicy is in effect. Cpuset policy
* is enforced in get_page_from_freelist().
*/
if (oc->nodemask &&
!nodes_subset(node_states[N_MEMORY], *oc->nodemask)) {
oc->totalpages = total_swap_pages;
for_each_node_mask(nid, *oc->nodemask)
oc->totalpages += node_present_pages(nid);
return CONSTRAINT_MEMORY_POLICY;
oom: badness heuristic rewrite This a complete rewrite of the oom killer's badness() heuristic which is used to determine which task to kill in oom conditions. The goal is to make it as simple and predictable as possible so the results are better understood and we end up killing the task which will lead to the most memory freeing while still respecting the fine-tuning from userspace. Instead of basing the heuristic on mm->total_vm for each task, the task's rss and swap space is used instead. This is a better indication of the amount of memory that will be freeable if the oom killed task is chosen and subsequently exits. This helps specifically in cases where KDE or GNOME is chosen for oom kill on desktop systems instead of a memory hogging task. The baseline for the heuristic is a proportion of memory that each task is currently using in memory plus swap compared to the amount of "allowable" memory. "Allowable," in this sense, means the system-wide resources for unconstrained oom conditions, the set of mempolicy nodes, the mems attached to current's cpuset, or a memory controller's limit. The proportion is given on a scale of 0 (never kill) to 1000 (always kill), roughly meaning that if a task has a badness() score of 500 that the task consumes approximately 50% of allowable memory resident in RAM or in swap space. The proportion is always relative to the amount of "allowable" memory and not the total amount of RAM systemwide so that mempolicies and cpusets may operate in isolation; they shall not need to know the true size of the machine on which they are running if they are bound to a specific set of nodes or mems, respectively. Root tasks are given 3% extra memory just like __vm_enough_memory() provides in LSMs. In the event of two tasks consuming similar amounts of memory, it is generally better to save root's task. Because of the change in the badness() heuristic's baseline, it is also necessary to introduce a new user interface to tune it. It's not possible to redefine the meaning of /proc/pid/oom_adj with a new scale since the ABI cannot be changed for backward compatability. Instead, a new tunable, /proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may be used to polarize the heuristic such that certain tasks are never considered for oom kill while others may always be considered. The value is added directly into the badness() score so a value of -500, for example, means to discount 50% of its memory consumption in comparison to other tasks either on the system, bound to the mempolicy, in the cpuset, or sharing the same memory controller. /proc/pid/oom_adj is changed so that its meaning is rescaled into the units used by /proc/pid/oom_score_adj, and vice versa. Changing one of these per-task tunables will rescale the value of the other to an equivalent meaning. Although /proc/pid/oom_adj was originally defined as a bitshift on the badness score, it now shares the same linear growth as /proc/pid/oom_score_adj but with different granularity. This is required so the ABI is not broken with userspace applications and allows oom_adj to be deprecated for future removal. Signed-off-by: David Rientjes <rientjes@google.com> Cc: Nick Piggin <npiggin@suse.de> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Balbir Singh <balbir@in.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 08:19:46 +08:00
}
/* Check this allocation failure is caused by cpuset's wall function */
for_each_zone_zonelist_nodemask(zone, z, oc->zonelist,
highest_zoneidx, oc->nodemask)
if (!cpuset_zone_allowed(zone, oc->gfp_mask))
oom: badness heuristic rewrite This a complete rewrite of the oom killer's badness() heuristic which is used to determine which task to kill in oom conditions. The goal is to make it as simple and predictable as possible so the results are better understood and we end up killing the task which will lead to the most memory freeing while still respecting the fine-tuning from userspace. Instead of basing the heuristic on mm->total_vm for each task, the task's rss and swap space is used instead. This is a better indication of the amount of memory that will be freeable if the oom killed task is chosen and subsequently exits. This helps specifically in cases where KDE or GNOME is chosen for oom kill on desktop systems instead of a memory hogging task. The baseline for the heuristic is a proportion of memory that each task is currently using in memory plus swap compared to the amount of "allowable" memory. "Allowable," in this sense, means the system-wide resources for unconstrained oom conditions, the set of mempolicy nodes, the mems attached to current's cpuset, or a memory controller's limit. The proportion is given on a scale of 0 (never kill) to 1000 (always kill), roughly meaning that if a task has a badness() score of 500 that the task consumes approximately 50% of allowable memory resident in RAM or in swap space. The proportion is always relative to the amount of "allowable" memory and not the total amount of RAM systemwide so that mempolicies and cpusets may operate in isolation; they shall not need to know the true size of the machine on which they are running if they are bound to a specific set of nodes or mems, respectively. Root tasks are given 3% extra memory just like __vm_enough_memory() provides in LSMs. In the event of two tasks consuming similar amounts of memory, it is generally better to save root's task. Because of the change in the badness() heuristic's baseline, it is also necessary to introduce a new user interface to tune it. It's not possible to redefine the meaning of /proc/pid/oom_adj with a new scale since the ABI cannot be changed for backward compatability. Instead, a new tunable, /proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may be used to polarize the heuristic such that certain tasks are never considered for oom kill while others may always be considered. The value is added directly into the badness() score so a value of -500, for example, means to discount 50% of its memory consumption in comparison to other tasks either on the system, bound to the mempolicy, in the cpuset, or sharing the same memory controller. /proc/pid/oom_adj is changed so that its meaning is rescaled into the units used by /proc/pid/oom_score_adj, and vice versa. Changing one of these per-task tunables will rescale the value of the other to an equivalent meaning. Although /proc/pid/oom_adj was originally defined as a bitshift on the badness score, it now shares the same linear growth as /proc/pid/oom_score_adj but with different granularity. This is required so the ABI is not broken with userspace applications and allows oom_adj to be deprecated for future removal. Signed-off-by: David Rientjes <rientjes@google.com> Cc: Nick Piggin <npiggin@suse.de> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Balbir Singh <balbir@in.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 08:19:46 +08:00
cpuset_limited = true;
oom: badness heuristic rewrite This a complete rewrite of the oom killer's badness() heuristic which is used to determine which task to kill in oom conditions. The goal is to make it as simple and predictable as possible so the results are better understood and we end up killing the task which will lead to the most memory freeing while still respecting the fine-tuning from userspace. Instead of basing the heuristic on mm->total_vm for each task, the task's rss and swap space is used instead. This is a better indication of the amount of memory that will be freeable if the oom killed task is chosen and subsequently exits. This helps specifically in cases where KDE or GNOME is chosen for oom kill on desktop systems instead of a memory hogging task. The baseline for the heuristic is a proportion of memory that each task is currently using in memory plus swap compared to the amount of "allowable" memory. "Allowable," in this sense, means the system-wide resources for unconstrained oom conditions, the set of mempolicy nodes, the mems attached to current's cpuset, or a memory controller's limit. The proportion is given on a scale of 0 (never kill) to 1000 (always kill), roughly meaning that if a task has a badness() score of 500 that the task consumes approximately 50% of allowable memory resident in RAM or in swap space. The proportion is always relative to the amount of "allowable" memory and not the total amount of RAM systemwide so that mempolicies and cpusets may operate in isolation; they shall not need to know the true size of the machine on which they are running if they are bound to a specific set of nodes or mems, respectively. Root tasks are given 3% extra memory just like __vm_enough_memory() provides in LSMs. In the event of two tasks consuming similar amounts of memory, it is generally better to save root's task. Because of the change in the badness() heuristic's baseline, it is also necessary to introduce a new user interface to tune it. It's not possible to redefine the meaning of /proc/pid/oom_adj with a new scale since the ABI cannot be changed for backward compatability. Instead, a new tunable, /proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may be used to polarize the heuristic such that certain tasks are never considered for oom kill while others may always be considered. The value is added directly into the badness() score so a value of -500, for example, means to discount 50% of its memory consumption in comparison to other tasks either on the system, bound to the mempolicy, in the cpuset, or sharing the same memory controller. /proc/pid/oom_adj is changed so that its meaning is rescaled into the units used by /proc/pid/oom_score_adj, and vice versa. Changing one of these per-task tunables will rescale the value of the other to an equivalent meaning. Although /proc/pid/oom_adj was originally defined as a bitshift on the badness score, it now shares the same linear growth as /proc/pid/oom_score_adj but with different granularity. This is required so the ABI is not broken with userspace applications and allows oom_adj to be deprecated for future removal. Signed-off-by: David Rientjes <rientjes@google.com> Cc: Nick Piggin <npiggin@suse.de> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Balbir Singh <balbir@in.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 08:19:46 +08:00
if (cpuset_limited) {
oc->totalpages = total_swap_pages;
oom: badness heuristic rewrite This a complete rewrite of the oom killer's badness() heuristic which is used to determine which task to kill in oom conditions. The goal is to make it as simple and predictable as possible so the results are better understood and we end up killing the task which will lead to the most memory freeing while still respecting the fine-tuning from userspace. Instead of basing the heuristic on mm->total_vm for each task, the task's rss and swap space is used instead. This is a better indication of the amount of memory that will be freeable if the oom killed task is chosen and subsequently exits. This helps specifically in cases where KDE or GNOME is chosen for oom kill on desktop systems instead of a memory hogging task. The baseline for the heuristic is a proportion of memory that each task is currently using in memory plus swap compared to the amount of "allowable" memory. "Allowable," in this sense, means the system-wide resources for unconstrained oom conditions, the set of mempolicy nodes, the mems attached to current's cpuset, or a memory controller's limit. The proportion is given on a scale of 0 (never kill) to 1000 (always kill), roughly meaning that if a task has a badness() score of 500 that the task consumes approximately 50% of allowable memory resident in RAM or in swap space. The proportion is always relative to the amount of "allowable" memory and not the total amount of RAM systemwide so that mempolicies and cpusets may operate in isolation; they shall not need to know the true size of the machine on which they are running if they are bound to a specific set of nodes or mems, respectively. Root tasks are given 3% extra memory just like __vm_enough_memory() provides in LSMs. In the event of two tasks consuming similar amounts of memory, it is generally better to save root's task. Because of the change in the badness() heuristic's baseline, it is also necessary to introduce a new user interface to tune it. It's not possible to redefine the meaning of /proc/pid/oom_adj with a new scale since the ABI cannot be changed for backward compatability. Instead, a new tunable, /proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may be used to polarize the heuristic such that certain tasks are never considered for oom kill while others may always be considered. The value is added directly into the badness() score so a value of -500, for example, means to discount 50% of its memory consumption in comparison to other tasks either on the system, bound to the mempolicy, in the cpuset, or sharing the same memory controller. /proc/pid/oom_adj is changed so that its meaning is rescaled into the units used by /proc/pid/oom_score_adj, and vice versa. Changing one of these per-task tunables will rescale the value of the other to an equivalent meaning. Although /proc/pid/oom_adj was originally defined as a bitshift on the badness score, it now shares the same linear growth as /proc/pid/oom_score_adj but with different granularity. This is required so the ABI is not broken with userspace applications and allows oom_adj to be deprecated for future removal. Signed-off-by: David Rientjes <rientjes@google.com> Cc: Nick Piggin <npiggin@suse.de> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Balbir Singh <balbir@in.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 08:19:46 +08:00
for_each_node_mask(nid, cpuset_current_mems_allowed)
oc->totalpages += node_present_pages(nid);
oom: badness heuristic rewrite This a complete rewrite of the oom killer's badness() heuristic which is used to determine which task to kill in oom conditions. The goal is to make it as simple and predictable as possible so the results are better understood and we end up killing the task which will lead to the most memory freeing while still respecting the fine-tuning from userspace. Instead of basing the heuristic on mm->total_vm for each task, the task's rss and swap space is used instead. This is a better indication of the amount of memory that will be freeable if the oom killed task is chosen and subsequently exits. This helps specifically in cases where KDE or GNOME is chosen for oom kill on desktop systems instead of a memory hogging task. The baseline for the heuristic is a proportion of memory that each task is currently using in memory plus swap compared to the amount of "allowable" memory. "Allowable," in this sense, means the system-wide resources for unconstrained oom conditions, the set of mempolicy nodes, the mems attached to current's cpuset, or a memory controller's limit. The proportion is given on a scale of 0 (never kill) to 1000 (always kill), roughly meaning that if a task has a badness() score of 500 that the task consumes approximately 50% of allowable memory resident in RAM or in swap space. The proportion is always relative to the amount of "allowable" memory and not the total amount of RAM systemwide so that mempolicies and cpusets may operate in isolation; they shall not need to know the true size of the machine on which they are running if they are bound to a specific set of nodes or mems, respectively. Root tasks are given 3% extra memory just like __vm_enough_memory() provides in LSMs. In the event of two tasks consuming similar amounts of memory, it is generally better to save root's task. Because of the change in the badness() heuristic's baseline, it is also necessary to introduce a new user interface to tune it. It's not possible to redefine the meaning of /proc/pid/oom_adj with a new scale since the ABI cannot be changed for backward compatability. Instead, a new tunable, /proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may be used to polarize the heuristic such that certain tasks are never considered for oom kill while others may always be considered. The value is added directly into the badness() score so a value of -500, for example, means to discount 50% of its memory consumption in comparison to other tasks either on the system, bound to the mempolicy, in the cpuset, or sharing the same memory controller. /proc/pid/oom_adj is changed so that its meaning is rescaled into the units used by /proc/pid/oom_score_adj, and vice versa. Changing one of these per-task tunables will rescale the value of the other to an equivalent meaning. Although /proc/pid/oom_adj was originally defined as a bitshift on the badness score, it now shares the same linear growth as /proc/pid/oom_score_adj but with different granularity. This is required so the ABI is not broken with userspace applications and allows oom_adj to be deprecated for future removal. Signed-off-by: David Rientjes <rientjes@google.com> Cc: Nick Piggin <npiggin@suse.de> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Balbir Singh <balbir@in.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 08:19:46 +08:00
return CONSTRAINT_CPUSET;
}
return CONSTRAINT_NONE;
}
static int oom_evaluate_task(struct task_struct *task, void *arg)
{
struct oom_control *oc = arg;
mm, oom: make the calculation of oom badness more accurate Recently we found an issue on our production environment that when memcg oom is triggered the oom killer doesn't chose the process with largest resident memory but chose the first scanned process. Note that all processes in this memcg have the same oom_score_adj, so the oom killer should chose the process with largest resident memory. Bellow is part of the oom info, which is enough to analyze this issue. [7516987.983223] memory: usage 16777216kB, limit 16777216kB, failcnt 52843037 [7516987.983224] memory+swap: usage 16777216kB, limit 9007199254740988kB, failcnt 0 [7516987.983225] kmem: usage 301464kB, limit 9007199254740988kB, failcnt 0 [...] [7516987.983293] [ pid ] uid tgid total_vm rss pgtables_bytes swapents oom_score_adj name [7516987.983510] [ 5740] 0 5740 257 1 32768 0 -998 pause [7516987.983574] [58804] 0 58804 4594 771 81920 0 -998 entry_point.bas [7516987.983577] [58908] 0 58908 7089 689 98304 0 -998 cron [7516987.983580] [58910] 0 58910 16235 5576 163840 0 -998 supervisord [7516987.983590] [59620] 0 59620 18074 1395 188416 0 -998 sshd [7516987.983594] [59622] 0 59622 18680 6679 188416 0 -998 python [7516987.983598] [59624] 0 59624 1859266 5161 548864 0 -998 odin-agent [7516987.983600] [59625] 0 59625 707223 9248 983040 0 -998 filebeat [7516987.983604] [59627] 0 59627 416433 64239 774144 0 -998 odin-log-agent [7516987.983607] [59631] 0 59631 180671 15012 385024 0 -998 python3 [7516987.983612] [61396] 0 61396 791287 3189 352256 0 -998 client [7516987.983615] [61641] 0 61641 1844642 29089 946176 0 -998 client [7516987.983765] [ 9236] 0 9236 2642 467 53248 0 -998 php_scanner [7516987.983911] [42898] 0 42898 15543 838 167936 0 -998 su [7516987.983915] [42900] 1000 42900 3673 867 77824 0 -998 exec_script_vr2 [7516987.983918] [42925] 1000 42925 36475 19033 335872 0 -998 python [7516987.983921] [57146] 1000 57146 3673 848 73728 0 -998 exec_script_J2p [7516987.983925] [57195] 1000 57195 186359 22958 491520 0 -998 python2 [7516987.983928] [58376] 1000 58376 275764 14402 290816 0 -998 rosmaster [7516987.983931] [58395] 1000 58395 155166 4449 245760 0 -998 rosout [7516987.983935] [58406] 1000 58406 18285584 3967322 37101568 0 -998 data_sim [7516987.984221] oom-kill:constraint=CONSTRAINT_MEMCG,nodemask=(null),cpuset=3aa16c9482ae3a6f6b78bda68a55d32c87c99b985e0f11331cddf05af6c4d753,mems_allowed=0-1,oom_memcg=/kubepods/podf1c273d3-9b36-11ea-b3df-246e9693c184,task_memcg=/kubepods/podf1c273d3-9b36-11ea-b3df-246e9693c184/1f246a3eeea8f70bf91141eeaf1805346a666e225f823906485ea0b6c37dfc3d,task=pause,pid=5740,uid=0 [7516987.984254] Memory cgroup out of memory: Killed process 5740 (pause) total-vm:1028kB, anon-rss:4kB, file-rss:0kB, shmem-rss:0kB [7516988.092344] oom_reaper: reaped process 5740 (pause), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kB We can find that the first scanned process 5740 (pause) was killed, but its rss is only one page. That is because, when we calculate the oom badness in oom_badness(), we always ignore the negtive point and convert all of these negtive points to 1. Now as oom_score_adj of all the processes in this targeted memcg have the same value -998, the points of these processes are all negtive value. As a result, the first scanned process will be killed. The oom_socre_adj (-998) in this memcg is set by kubelet, because it is a a Guaranteed pod, which has higher priority to prevent from being killed by system oom. To fix this issue, we should make the calculation of oom point more accurate. We can achieve it by convert the chosen_point from 'unsigned long' to 'long'. [cai@lca.pw: reported a issue in the previous version] [mhocko@suse.com: fixed the issue reported by Cai] [mhocko@suse.com: add the comment in proc_oom_score()] [laoar.shao@gmail.com: v3] Link: http://lkml.kernel.org/r/1594396651-9931-1-git-send-email-laoar.shao@gmail.com Signed-off-by: Yafang Shao <laoar.shao@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Tested-by: Naresh Kamboju <naresh.kamboju@linaro.org> Acked-by: Michal Hocko <mhocko@suse.com> Cc: David Rientjes <rientjes@google.com> Cc: Qian Cai <cai@lca.pw> Link: http://lkml.kernel.org/r/1594309987-9919-1-git-send-email-laoar.shao@gmail.com Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-08-12 09:31:22 +08:00
long points;
oom: decouple mems_allowed from oom_unkillable_task Commit ef08e3b4981a ("[PATCH] cpusets: confine oom_killer to mem_exclusive cpuset") introduces a heuristic where a potential oom-killer victim is skipped if the intersection of the potential victim and the current (the process triggered the oom) is empty based on the reason that killing such victim most probably will not help the current allocating process. However the commit 7887a3da753e ("[PATCH] oom: cpuset hint") changed the heuristic to just decrease the oom_badness scores of such potential victim based on the reason that the cpuset of such processes might have changed and previously they may have allocated memory on mems where the current allocating process can allocate from. Unintentionally 7887a3da753e ("[PATCH] oom: cpuset hint") introduced a side effect as the oom_badness is also exposed to the user space through /proc/[pid]/oom_score, so, readers with different cpusets can read different oom_score of the same process. Later, commit 6cf86ac6f36b ("oom: filter tasks not sharing the same cpuset") fixed the side effect introduced by 7887a3da753e by moving the cpuset intersection back to only oom-killer context and out of oom_badness. However the combination of ab290adbaf8f ("oom: make oom_unkillable_task() helper function") and 26ebc984913b ("oom: /proc/<pid>/oom_score treat kernel thread honestly") unintentionally brought back the cpuset intersection check into the oom_badness calculation function. Other than doing cpuset/mempolicy intersection from oom_badness, the memcg oom context is also doing cpuset/mempolicy intersection which is quite wrong and is caught by syzcaller with the following report: kasan: CONFIG_KASAN_INLINE enabled kasan: GPF could be caused by NULL-ptr deref or user memory access general protection fault: 0000 [#1] PREEMPT SMP KASAN CPU: 0 PID: 28426 Comm: syz-executor.5 Not tainted 5.2.0-rc3-next-20190607 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:__read_once_size include/linux/compiler.h:194 [inline] RIP: 0010:has_intersects_mems_allowed mm/oom_kill.c:84 [inline] RIP: 0010:oom_unkillable_task mm/oom_kill.c:168 [inline] RIP: 0010:oom_unkillable_task+0x180/0x400 mm/oom_kill.c:155 Code: c1 ea 03 80 3c 02 00 0f 85 80 02 00 00 4c 8b a3 10 07 00 00 48 b8 00 00 00 00 00 fc ff df 4d 8d 74 24 10 4c 89 f2 48 c1 ea 03 <80> 3c 02 00 0f 85 67 02 00 00 49 8b 44 24 10 4c 8d a0 68 fa ff ff RSP: 0018:ffff888000127490 EFLAGS: 00010a03 RAX: dffffc0000000000 RBX: ffff8880a4cd5438 RCX: ffffffff818dae9c RDX: 100000000c3cc602 RSI: ffffffff818dac8d RDI: 0000000000000001 RBP: ffff8880001274d0 R08: ffff888000086180 R09: ffffed1015d26be0 R10: ffffed1015d26bdf R11: ffff8880ae935efb R12: 8000000061e63007 R13: 0000000000000000 R14: 8000000061e63017 R15: 1ffff11000024ea6 FS: 00005555561f5940(0000) GS:ffff8880ae800000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000607304 CR3: 000000009237e000 CR4: 00000000001426f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000600 Call Trace: oom_evaluate_task+0x49/0x520 mm/oom_kill.c:321 mem_cgroup_scan_tasks+0xcc/0x180 mm/memcontrol.c:1169 select_bad_process mm/oom_kill.c:374 [inline] out_of_memory mm/oom_kill.c:1088 [inline] out_of_memory+0x6b2/0x1280 mm/oom_kill.c:1035 mem_cgroup_out_of_memory+0x1ca/0x230 mm/memcontrol.c:1573 mem_cgroup_oom mm/memcontrol.c:1905 [inline] try_charge+0xfbe/0x1480 mm/memcontrol.c:2468 mem_cgroup_try_charge+0x24d/0x5e0 mm/memcontrol.c:6073 mem_cgroup_try_charge_delay+0x1f/0xa0 mm/memcontrol.c:6088 do_huge_pmd_wp_page_fallback+0x24f/0x1680 mm/huge_memory.c:1201 do_huge_pmd_wp_page+0x7fc/0x2160 mm/huge_memory.c:1359 wp_huge_pmd mm/memory.c:3793 [inline] __handle_mm_fault+0x164c/0x3eb0 mm/memory.c:4006 handle_mm_fault+0x3b7/0xa90 mm/memory.c:4053 do_user_addr_fault arch/x86/mm/fault.c:1455 [inline] __do_page_fault+0x5ef/0xda0 arch/x86/mm/fault.c:1521 do_page_fault+0x71/0x57d arch/x86/mm/fault.c:1552 page_fault+0x1e/0x30 arch/x86/entry/entry_64.S:1156 RIP: 0033:0x400590 Code: 06 e9 49 01 00 00 48 8b 44 24 10 48 0b 44 24 28 75 1f 48 8b 14 24 48 8b 7c 24 20 be 04 00 00 00 e8 f5 56 00 00 48 8b 74 24 08 <89> 06 e9 1e 01 00 00 48 8b 44 24 08 48 8b 14 24 be 04 00 00 00 8b RSP: 002b:00007fff7bc49780 EFLAGS: 00010206 RAX: 0000000000000001 RBX: 0000000000760000 RCX: 0000000000000000 RDX: 0000000000000000 RSI: 000000002000cffc RDI: 0000000000000001 RBP: fffffffffffffffe R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000075 R11: 0000000000000246 R12: 0000000000760008 R13: 00000000004c55f2 R14: 0000000000000000 R15: 00007fff7bc499b0 Modules linked in: ---[ end trace a65689219582ffff ]--- RIP: 0010:__read_once_size include/linux/compiler.h:194 [inline] RIP: 0010:has_intersects_mems_allowed mm/oom_kill.c:84 [inline] RIP: 0010:oom_unkillable_task mm/oom_kill.c:168 [inline] RIP: 0010:oom_unkillable_task+0x180/0x400 mm/oom_kill.c:155 Code: c1 ea 03 80 3c 02 00 0f 85 80 02 00 00 4c 8b a3 10 07 00 00 48 b8 00 00 00 00 00 fc ff df 4d 8d 74 24 10 4c 89 f2 48 c1 ea 03 <80> 3c 02 00 0f 85 67 02 00 00 49 8b 44 24 10 4c 8d a0 68 fa ff ff RSP: 0018:ffff888000127490 EFLAGS: 00010a03 RAX: dffffc0000000000 RBX: ffff8880a4cd5438 RCX: ffffffff818dae9c RDX: 100000000c3cc602 RSI: ffffffff818dac8d RDI: 0000000000000001 RBP: ffff8880001274d0 R08: ffff888000086180 R09: ffffed1015d26be0 R10: ffffed1015d26bdf R11: ffff8880ae935efb R12: 8000000061e63007 R13: 0000000000000000 R14: 8000000061e63017 R15: 1ffff11000024ea6 FS: 00005555561f5940(0000) GS:ffff8880ae800000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000001b2f823000 CR3: 000000009237e000 CR4: 00000000001426f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000600 The fix is to decouple the cpuset/mempolicy intersection check from oom_unkillable_task() and make sure cpuset/mempolicy intersection check is only done in the global oom context. [shakeelb@google.com: change function name and update comment] Link: http://lkml.kernel.org/r/20190628152421.198994-3-shakeelb@google.com Link: http://lkml.kernel.org/r/20190624212631.87212-3-shakeelb@google.com Signed-off-by: Shakeel Butt <shakeelb@google.com> Reported-by: syzbot+d0fc9d3c166bc5e4a94b@syzkaller.appspotmail.com Acked-by: Roman Gushchin <guro@fb.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Nick Piggin <npiggin@suse.de> Cc: Paul Jackson <pj@sgi.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-07-12 12:00:31 +08:00
if (oom_unkillable_task(task))
goto next;
/* p may not have freeable memory in nodemask */
if (!is_memcg_oom(oc) && !oom_cpuset_eligible(task, oc))
goto next;
/*
* This task already has access to memory reserves and is being killed.
* Don't allow any other task to have access to the reserves unless
* the task has MMF_OOM_SKIP because chances that it would release
* any memory is quite low.
*/
if (!is_sysrq_oom(oc) && tsk_is_oom_victim(task)) {
if (test_bit(MMF_OOM_SKIP, &task->signal->oom_mm->flags))
goto next;
goto abort;
}
mm, oom: fix race when specifying a thread as the oom origin test_set_oom_score_adj() and compare_swap_oom_score_adj() are used to specify that current should be killed first if an oom condition occurs in between the two calls. The usage is short oom_score_adj = test_set_oom_score_adj(OOM_SCORE_ADJ_MAX); ... compare_swap_oom_score_adj(OOM_SCORE_ADJ_MAX, oom_score_adj); to store the thread's oom_score_adj, temporarily change it to the maximum score possible, and then restore the old value if it is still the same. This happens to still be racy, however, if the user writes OOM_SCORE_ADJ_MAX to /proc/pid/oom_score_adj in between the two calls. The compare_swap_oom_score_adj() will then incorrectly reset the old value prior to the write of OOM_SCORE_ADJ_MAX. To fix this, introduce a new oom_flags_t member in struct signal_struct that will be used for per-thread oom killer flags. KSM and swapoff can now use a bit in this member to specify that threads should be killed first in oom conditions without playing around with oom_score_adj. This also allows the correct oom_score_adj to always be shown when reading /proc/pid/oom_score. Signed-off-by: David Rientjes <rientjes@google.com> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Reviewed-by: Michal Hocko <mhocko@suse.cz> Cc: Anton Vorontsov <anton.vorontsov@linaro.org> Cc: Oleg Nesterov <oleg@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-12-12 08:02:56 +08:00
/*
* If task is allocating a lot of memory and has been marked to be
* killed first if it triggers an oom, then select it.
*/
if (oom_task_origin(task)) {
mm, oom: make the calculation of oom badness more accurate Recently we found an issue on our production environment that when memcg oom is triggered the oom killer doesn't chose the process with largest resident memory but chose the first scanned process. Note that all processes in this memcg have the same oom_score_adj, so the oom killer should chose the process with largest resident memory. Bellow is part of the oom info, which is enough to analyze this issue. [7516987.983223] memory: usage 16777216kB, limit 16777216kB, failcnt 52843037 [7516987.983224] memory+swap: usage 16777216kB, limit 9007199254740988kB, failcnt 0 [7516987.983225] kmem: usage 301464kB, limit 9007199254740988kB, failcnt 0 [...] [7516987.983293] [ pid ] uid tgid total_vm rss pgtables_bytes swapents oom_score_adj name [7516987.983510] [ 5740] 0 5740 257 1 32768 0 -998 pause [7516987.983574] [58804] 0 58804 4594 771 81920 0 -998 entry_point.bas [7516987.983577] [58908] 0 58908 7089 689 98304 0 -998 cron [7516987.983580] [58910] 0 58910 16235 5576 163840 0 -998 supervisord [7516987.983590] [59620] 0 59620 18074 1395 188416 0 -998 sshd [7516987.983594] [59622] 0 59622 18680 6679 188416 0 -998 python [7516987.983598] [59624] 0 59624 1859266 5161 548864 0 -998 odin-agent [7516987.983600] [59625] 0 59625 707223 9248 983040 0 -998 filebeat [7516987.983604] [59627] 0 59627 416433 64239 774144 0 -998 odin-log-agent [7516987.983607] [59631] 0 59631 180671 15012 385024 0 -998 python3 [7516987.983612] [61396] 0 61396 791287 3189 352256 0 -998 client [7516987.983615] [61641] 0 61641 1844642 29089 946176 0 -998 client [7516987.983765] [ 9236] 0 9236 2642 467 53248 0 -998 php_scanner [7516987.983911] [42898] 0 42898 15543 838 167936 0 -998 su [7516987.983915] [42900] 1000 42900 3673 867 77824 0 -998 exec_script_vr2 [7516987.983918] [42925] 1000 42925 36475 19033 335872 0 -998 python [7516987.983921] [57146] 1000 57146 3673 848 73728 0 -998 exec_script_J2p [7516987.983925] [57195] 1000 57195 186359 22958 491520 0 -998 python2 [7516987.983928] [58376] 1000 58376 275764 14402 290816 0 -998 rosmaster [7516987.983931] [58395] 1000 58395 155166 4449 245760 0 -998 rosout [7516987.983935] [58406] 1000 58406 18285584 3967322 37101568 0 -998 data_sim [7516987.984221] oom-kill:constraint=CONSTRAINT_MEMCG,nodemask=(null),cpuset=3aa16c9482ae3a6f6b78bda68a55d32c87c99b985e0f11331cddf05af6c4d753,mems_allowed=0-1,oom_memcg=/kubepods/podf1c273d3-9b36-11ea-b3df-246e9693c184,task_memcg=/kubepods/podf1c273d3-9b36-11ea-b3df-246e9693c184/1f246a3eeea8f70bf91141eeaf1805346a666e225f823906485ea0b6c37dfc3d,task=pause,pid=5740,uid=0 [7516987.984254] Memory cgroup out of memory: Killed process 5740 (pause) total-vm:1028kB, anon-rss:4kB, file-rss:0kB, shmem-rss:0kB [7516988.092344] oom_reaper: reaped process 5740 (pause), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kB We can find that the first scanned process 5740 (pause) was killed, but its rss is only one page. That is because, when we calculate the oom badness in oom_badness(), we always ignore the negtive point and convert all of these negtive points to 1. Now as oom_score_adj of all the processes in this targeted memcg have the same value -998, the points of these processes are all negtive value. As a result, the first scanned process will be killed. The oom_socre_adj (-998) in this memcg is set by kubelet, because it is a a Guaranteed pod, which has higher priority to prevent from being killed by system oom. To fix this issue, we should make the calculation of oom point more accurate. We can achieve it by convert the chosen_point from 'unsigned long' to 'long'. [cai@lca.pw: reported a issue in the previous version] [mhocko@suse.com: fixed the issue reported by Cai] [mhocko@suse.com: add the comment in proc_oom_score()] [laoar.shao@gmail.com: v3] Link: http://lkml.kernel.org/r/1594396651-9931-1-git-send-email-laoar.shao@gmail.com Signed-off-by: Yafang Shao <laoar.shao@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Tested-by: Naresh Kamboju <naresh.kamboju@linaro.org> Acked-by: Michal Hocko <mhocko@suse.com> Cc: David Rientjes <rientjes@google.com> Cc: Qian Cai <cai@lca.pw> Link: http://lkml.kernel.org/r/1594309987-9919-1-git-send-email-laoar.shao@gmail.com Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-08-12 09:31:22 +08:00
points = LONG_MAX;
goto select;
}
mm, oom: fix race when specifying a thread as the oom origin test_set_oom_score_adj() and compare_swap_oom_score_adj() are used to specify that current should be killed first if an oom condition occurs in between the two calls. The usage is short oom_score_adj = test_set_oom_score_adj(OOM_SCORE_ADJ_MAX); ... compare_swap_oom_score_adj(OOM_SCORE_ADJ_MAX, oom_score_adj); to store the thread's oom_score_adj, temporarily change it to the maximum score possible, and then restore the old value if it is still the same. This happens to still be racy, however, if the user writes OOM_SCORE_ADJ_MAX to /proc/pid/oom_score_adj in between the two calls. The compare_swap_oom_score_adj() will then incorrectly reset the old value prior to the write of OOM_SCORE_ADJ_MAX. To fix this, introduce a new oom_flags_t member in struct signal_struct that will be used for per-thread oom killer flags. KSM and swapoff can now use a bit in this member to specify that threads should be killed first in oom conditions without playing around with oom_score_adj. This also allows the correct oom_score_adj to always be shown when reading /proc/pid/oom_score. Signed-off-by: David Rientjes <rientjes@google.com> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Reviewed-by: Michal Hocko <mhocko@suse.cz> Cc: Anton Vorontsov <anton.vorontsov@linaro.org> Cc: Oleg Nesterov <oleg@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-12-12 08:02:56 +08:00
oom: decouple mems_allowed from oom_unkillable_task Commit ef08e3b4981a ("[PATCH] cpusets: confine oom_killer to mem_exclusive cpuset") introduces a heuristic where a potential oom-killer victim is skipped if the intersection of the potential victim and the current (the process triggered the oom) is empty based on the reason that killing such victim most probably will not help the current allocating process. However the commit 7887a3da753e ("[PATCH] oom: cpuset hint") changed the heuristic to just decrease the oom_badness scores of such potential victim based on the reason that the cpuset of such processes might have changed and previously they may have allocated memory on mems where the current allocating process can allocate from. Unintentionally 7887a3da753e ("[PATCH] oom: cpuset hint") introduced a side effect as the oom_badness is also exposed to the user space through /proc/[pid]/oom_score, so, readers with different cpusets can read different oom_score of the same process. Later, commit 6cf86ac6f36b ("oom: filter tasks not sharing the same cpuset") fixed the side effect introduced by 7887a3da753e by moving the cpuset intersection back to only oom-killer context and out of oom_badness. However the combination of ab290adbaf8f ("oom: make oom_unkillable_task() helper function") and 26ebc984913b ("oom: /proc/<pid>/oom_score treat kernel thread honestly") unintentionally brought back the cpuset intersection check into the oom_badness calculation function. Other than doing cpuset/mempolicy intersection from oom_badness, the memcg oom context is also doing cpuset/mempolicy intersection which is quite wrong and is caught by syzcaller with the following report: kasan: CONFIG_KASAN_INLINE enabled kasan: GPF could be caused by NULL-ptr deref or user memory access general protection fault: 0000 [#1] PREEMPT SMP KASAN CPU: 0 PID: 28426 Comm: syz-executor.5 Not tainted 5.2.0-rc3-next-20190607 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:__read_once_size include/linux/compiler.h:194 [inline] RIP: 0010:has_intersects_mems_allowed mm/oom_kill.c:84 [inline] RIP: 0010:oom_unkillable_task mm/oom_kill.c:168 [inline] RIP: 0010:oom_unkillable_task+0x180/0x400 mm/oom_kill.c:155 Code: c1 ea 03 80 3c 02 00 0f 85 80 02 00 00 4c 8b a3 10 07 00 00 48 b8 00 00 00 00 00 fc ff df 4d 8d 74 24 10 4c 89 f2 48 c1 ea 03 <80> 3c 02 00 0f 85 67 02 00 00 49 8b 44 24 10 4c 8d a0 68 fa ff ff RSP: 0018:ffff888000127490 EFLAGS: 00010a03 RAX: dffffc0000000000 RBX: ffff8880a4cd5438 RCX: ffffffff818dae9c RDX: 100000000c3cc602 RSI: ffffffff818dac8d RDI: 0000000000000001 RBP: ffff8880001274d0 R08: ffff888000086180 R09: ffffed1015d26be0 R10: ffffed1015d26bdf R11: ffff8880ae935efb R12: 8000000061e63007 R13: 0000000000000000 R14: 8000000061e63017 R15: 1ffff11000024ea6 FS: 00005555561f5940(0000) GS:ffff8880ae800000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000607304 CR3: 000000009237e000 CR4: 00000000001426f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000600 Call Trace: oom_evaluate_task+0x49/0x520 mm/oom_kill.c:321 mem_cgroup_scan_tasks+0xcc/0x180 mm/memcontrol.c:1169 select_bad_process mm/oom_kill.c:374 [inline] out_of_memory mm/oom_kill.c:1088 [inline] out_of_memory+0x6b2/0x1280 mm/oom_kill.c:1035 mem_cgroup_out_of_memory+0x1ca/0x230 mm/memcontrol.c:1573 mem_cgroup_oom mm/memcontrol.c:1905 [inline] try_charge+0xfbe/0x1480 mm/memcontrol.c:2468 mem_cgroup_try_charge+0x24d/0x5e0 mm/memcontrol.c:6073 mem_cgroup_try_charge_delay+0x1f/0xa0 mm/memcontrol.c:6088 do_huge_pmd_wp_page_fallback+0x24f/0x1680 mm/huge_memory.c:1201 do_huge_pmd_wp_page+0x7fc/0x2160 mm/huge_memory.c:1359 wp_huge_pmd mm/memory.c:3793 [inline] __handle_mm_fault+0x164c/0x3eb0 mm/memory.c:4006 handle_mm_fault+0x3b7/0xa90 mm/memory.c:4053 do_user_addr_fault arch/x86/mm/fault.c:1455 [inline] __do_page_fault+0x5ef/0xda0 arch/x86/mm/fault.c:1521 do_page_fault+0x71/0x57d arch/x86/mm/fault.c:1552 page_fault+0x1e/0x30 arch/x86/entry/entry_64.S:1156 RIP: 0033:0x400590 Code: 06 e9 49 01 00 00 48 8b 44 24 10 48 0b 44 24 28 75 1f 48 8b 14 24 48 8b 7c 24 20 be 04 00 00 00 e8 f5 56 00 00 48 8b 74 24 08 <89> 06 e9 1e 01 00 00 48 8b 44 24 08 48 8b 14 24 be 04 00 00 00 8b RSP: 002b:00007fff7bc49780 EFLAGS: 00010206 RAX: 0000000000000001 RBX: 0000000000760000 RCX: 0000000000000000 RDX: 0000000000000000 RSI: 000000002000cffc RDI: 0000000000000001 RBP: fffffffffffffffe R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000075 R11: 0000000000000246 R12: 0000000000760008 R13: 00000000004c55f2 R14: 0000000000000000 R15: 00007fff7bc499b0 Modules linked in: ---[ end trace a65689219582ffff ]--- RIP: 0010:__read_once_size include/linux/compiler.h:194 [inline] RIP: 0010:has_intersects_mems_allowed mm/oom_kill.c:84 [inline] RIP: 0010:oom_unkillable_task mm/oom_kill.c:168 [inline] RIP: 0010:oom_unkillable_task+0x180/0x400 mm/oom_kill.c:155 Code: c1 ea 03 80 3c 02 00 0f 85 80 02 00 00 4c 8b a3 10 07 00 00 48 b8 00 00 00 00 00 fc ff df 4d 8d 74 24 10 4c 89 f2 48 c1 ea 03 <80> 3c 02 00 0f 85 67 02 00 00 49 8b 44 24 10 4c 8d a0 68 fa ff ff RSP: 0018:ffff888000127490 EFLAGS: 00010a03 RAX: dffffc0000000000 RBX: ffff8880a4cd5438 RCX: ffffffff818dae9c RDX: 100000000c3cc602 RSI: ffffffff818dac8d RDI: 0000000000000001 RBP: ffff8880001274d0 R08: ffff888000086180 R09: ffffed1015d26be0 R10: ffffed1015d26bdf R11: ffff8880ae935efb R12: 8000000061e63007 R13: 0000000000000000 R14: 8000000061e63017 R15: 1ffff11000024ea6 FS: 00005555561f5940(0000) GS:ffff8880ae800000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000001b2f823000 CR3: 000000009237e000 CR4: 00000000001426f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000600 The fix is to decouple the cpuset/mempolicy intersection check from oom_unkillable_task() and make sure cpuset/mempolicy intersection check is only done in the global oom context. [shakeelb@google.com: change function name and update comment] Link: http://lkml.kernel.org/r/20190628152421.198994-3-shakeelb@google.com Link: http://lkml.kernel.org/r/20190624212631.87212-3-shakeelb@google.com Signed-off-by: Shakeel Butt <shakeelb@google.com> Reported-by: syzbot+d0fc9d3c166bc5e4a94b@syzkaller.appspotmail.com Acked-by: Roman Gushchin <guro@fb.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Nick Piggin <npiggin@suse.de> Cc: Paul Jackson <pj@sgi.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-07-12 12:00:31 +08:00
points = oom_badness(task, oc->totalpages);
mm, oom: make the calculation of oom badness more accurate Recently we found an issue on our production environment that when memcg oom is triggered the oom killer doesn't chose the process with largest resident memory but chose the first scanned process. Note that all processes in this memcg have the same oom_score_adj, so the oom killer should chose the process with largest resident memory. Bellow is part of the oom info, which is enough to analyze this issue. [7516987.983223] memory: usage 16777216kB, limit 16777216kB, failcnt 52843037 [7516987.983224] memory+swap: usage 16777216kB, limit 9007199254740988kB, failcnt 0 [7516987.983225] kmem: usage 301464kB, limit 9007199254740988kB, failcnt 0 [...] [7516987.983293] [ pid ] uid tgid total_vm rss pgtables_bytes swapents oom_score_adj name [7516987.983510] [ 5740] 0 5740 257 1 32768 0 -998 pause [7516987.983574] [58804] 0 58804 4594 771 81920 0 -998 entry_point.bas [7516987.983577] [58908] 0 58908 7089 689 98304 0 -998 cron [7516987.983580] [58910] 0 58910 16235 5576 163840 0 -998 supervisord [7516987.983590] [59620] 0 59620 18074 1395 188416 0 -998 sshd [7516987.983594] [59622] 0 59622 18680 6679 188416 0 -998 python [7516987.983598] [59624] 0 59624 1859266 5161 548864 0 -998 odin-agent [7516987.983600] [59625] 0 59625 707223 9248 983040 0 -998 filebeat [7516987.983604] [59627] 0 59627 416433 64239 774144 0 -998 odin-log-agent [7516987.983607] [59631] 0 59631 180671 15012 385024 0 -998 python3 [7516987.983612] [61396] 0 61396 791287 3189 352256 0 -998 client [7516987.983615] [61641] 0 61641 1844642 29089 946176 0 -998 client [7516987.983765] [ 9236] 0 9236 2642 467 53248 0 -998 php_scanner [7516987.983911] [42898] 0 42898 15543 838 167936 0 -998 su [7516987.983915] [42900] 1000 42900 3673 867 77824 0 -998 exec_script_vr2 [7516987.983918] [42925] 1000 42925 36475 19033 335872 0 -998 python [7516987.983921] [57146] 1000 57146 3673 848 73728 0 -998 exec_script_J2p [7516987.983925] [57195] 1000 57195 186359 22958 491520 0 -998 python2 [7516987.983928] [58376] 1000 58376 275764 14402 290816 0 -998 rosmaster [7516987.983931] [58395] 1000 58395 155166 4449 245760 0 -998 rosout [7516987.983935] [58406] 1000 58406 18285584 3967322 37101568 0 -998 data_sim [7516987.984221] oom-kill:constraint=CONSTRAINT_MEMCG,nodemask=(null),cpuset=3aa16c9482ae3a6f6b78bda68a55d32c87c99b985e0f11331cddf05af6c4d753,mems_allowed=0-1,oom_memcg=/kubepods/podf1c273d3-9b36-11ea-b3df-246e9693c184,task_memcg=/kubepods/podf1c273d3-9b36-11ea-b3df-246e9693c184/1f246a3eeea8f70bf91141eeaf1805346a666e225f823906485ea0b6c37dfc3d,task=pause,pid=5740,uid=0 [7516987.984254] Memory cgroup out of memory: Killed process 5740 (pause) total-vm:1028kB, anon-rss:4kB, file-rss:0kB, shmem-rss:0kB [7516988.092344] oom_reaper: reaped process 5740 (pause), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kB We can find that the first scanned process 5740 (pause) was killed, but its rss is only one page. That is because, when we calculate the oom badness in oom_badness(), we always ignore the negtive point and convert all of these negtive points to 1. Now as oom_score_adj of all the processes in this targeted memcg have the same value -998, the points of these processes are all negtive value. As a result, the first scanned process will be killed. The oom_socre_adj (-998) in this memcg is set by kubelet, because it is a a Guaranteed pod, which has higher priority to prevent from being killed by system oom. To fix this issue, we should make the calculation of oom point more accurate. We can achieve it by convert the chosen_point from 'unsigned long' to 'long'. [cai@lca.pw: reported a issue in the previous version] [mhocko@suse.com: fixed the issue reported by Cai] [mhocko@suse.com: add the comment in proc_oom_score()] [laoar.shao@gmail.com: v3] Link: http://lkml.kernel.org/r/1594396651-9931-1-git-send-email-laoar.shao@gmail.com Signed-off-by: Yafang Shao <laoar.shao@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Tested-by: Naresh Kamboju <naresh.kamboju@linaro.org> Acked-by: Michal Hocko <mhocko@suse.com> Cc: David Rientjes <rientjes@google.com> Cc: Qian Cai <cai@lca.pw> Link: http://lkml.kernel.org/r/1594309987-9919-1-git-send-email-laoar.shao@gmail.com Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-08-12 09:31:22 +08:00
if (points == LONG_MIN || points < oc->chosen_points)
goto next;
select:
if (oc->chosen)
put_task_struct(oc->chosen);
get_task_struct(task);
oc->chosen = task;
oc->chosen_points = points;
next:
return 0;
abort:
if (oc->chosen)
put_task_struct(oc->chosen);
oc->chosen = (void *)-1UL;
return 1;
}
/*
* Simple selection loop. We choose the process with the highest number of
* 'points'. In case scan was aborted, oc->chosen is set to -1.
*/
static void select_bad_process(struct oom_control *oc)
{
mm, oom: make the calculation of oom badness more accurate Recently we found an issue on our production environment that when memcg oom is triggered the oom killer doesn't chose the process with largest resident memory but chose the first scanned process. Note that all processes in this memcg have the same oom_score_adj, so the oom killer should chose the process with largest resident memory. Bellow is part of the oom info, which is enough to analyze this issue. [7516987.983223] memory: usage 16777216kB, limit 16777216kB, failcnt 52843037 [7516987.983224] memory+swap: usage 16777216kB, limit 9007199254740988kB, failcnt 0 [7516987.983225] kmem: usage 301464kB, limit 9007199254740988kB, failcnt 0 [...] [7516987.983293] [ pid ] uid tgid total_vm rss pgtables_bytes swapents oom_score_adj name [7516987.983510] [ 5740] 0 5740 257 1 32768 0 -998 pause [7516987.983574] [58804] 0 58804 4594 771 81920 0 -998 entry_point.bas [7516987.983577] [58908] 0 58908 7089 689 98304 0 -998 cron [7516987.983580] [58910] 0 58910 16235 5576 163840 0 -998 supervisord [7516987.983590] [59620] 0 59620 18074 1395 188416 0 -998 sshd [7516987.983594] [59622] 0 59622 18680 6679 188416 0 -998 python [7516987.983598] [59624] 0 59624 1859266 5161 548864 0 -998 odin-agent [7516987.983600] [59625] 0 59625 707223 9248 983040 0 -998 filebeat [7516987.983604] [59627] 0 59627 416433 64239 774144 0 -998 odin-log-agent [7516987.983607] [59631] 0 59631 180671 15012 385024 0 -998 python3 [7516987.983612] [61396] 0 61396 791287 3189 352256 0 -998 client [7516987.983615] [61641] 0 61641 1844642 29089 946176 0 -998 client [7516987.983765] [ 9236] 0 9236 2642 467 53248 0 -998 php_scanner [7516987.983911] [42898] 0 42898 15543 838 167936 0 -998 su [7516987.983915] [42900] 1000 42900 3673 867 77824 0 -998 exec_script_vr2 [7516987.983918] [42925] 1000 42925 36475 19033 335872 0 -998 python [7516987.983921] [57146] 1000 57146 3673 848 73728 0 -998 exec_script_J2p [7516987.983925] [57195] 1000 57195 186359 22958 491520 0 -998 python2 [7516987.983928] [58376] 1000 58376 275764 14402 290816 0 -998 rosmaster [7516987.983931] [58395] 1000 58395 155166 4449 245760 0 -998 rosout [7516987.983935] [58406] 1000 58406 18285584 3967322 37101568 0 -998 data_sim [7516987.984221] oom-kill:constraint=CONSTRAINT_MEMCG,nodemask=(null),cpuset=3aa16c9482ae3a6f6b78bda68a55d32c87c99b985e0f11331cddf05af6c4d753,mems_allowed=0-1,oom_memcg=/kubepods/podf1c273d3-9b36-11ea-b3df-246e9693c184,task_memcg=/kubepods/podf1c273d3-9b36-11ea-b3df-246e9693c184/1f246a3eeea8f70bf91141eeaf1805346a666e225f823906485ea0b6c37dfc3d,task=pause,pid=5740,uid=0 [7516987.984254] Memory cgroup out of memory: Killed process 5740 (pause) total-vm:1028kB, anon-rss:4kB, file-rss:0kB, shmem-rss:0kB [7516988.092344] oom_reaper: reaped process 5740 (pause), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kB We can find that the first scanned process 5740 (pause) was killed, but its rss is only one page. That is because, when we calculate the oom badness in oom_badness(), we always ignore the negtive point and convert all of these negtive points to 1. Now as oom_score_adj of all the processes in this targeted memcg have the same value -998, the points of these processes are all negtive value. As a result, the first scanned process will be killed. The oom_socre_adj (-998) in this memcg is set by kubelet, because it is a a Guaranteed pod, which has higher priority to prevent from being killed by system oom. To fix this issue, we should make the calculation of oom point more accurate. We can achieve it by convert the chosen_point from 'unsigned long' to 'long'. [cai@lca.pw: reported a issue in the previous version] [mhocko@suse.com: fixed the issue reported by Cai] [mhocko@suse.com: add the comment in proc_oom_score()] [laoar.shao@gmail.com: v3] Link: http://lkml.kernel.org/r/1594396651-9931-1-git-send-email-laoar.shao@gmail.com Signed-off-by: Yafang Shao <laoar.shao@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Tested-by: Naresh Kamboju <naresh.kamboju@linaro.org> Acked-by: Michal Hocko <mhocko@suse.com> Cc: David Rientjes <rientjes@google.com> Cc: Qian Cai <cai@lca.pw> Link: http://lkml.kernel.org/r/1594309987-9919-1-git-send-email-laoar.shao@gmail.com Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-08-12 09:31:22 +08:00
oc->chosen_points = LONG_MIN;
if (is_memcg_oom(oc))
mem_cgroup_scan_tasks(oc->memcg, oom_evaluate_task, oc);
else {
struct task_struct *p;
rcu_read_lock();
for_each_process(p)
if (oom_evaluate_task(p, oc))
break;
rcu_read_unlock();
}
}
static int dump_task(struct task_struct *p, void *arg)
{
struct oom_control *oc = arg;
struct task_struct *task;
oom: decouple mems_allowed from oom_unkillable_task Commit ef08e3b4981a ("[PATCH] cpusets: confine oom_killer to mem_exclusive cpuset") introduces a heuristic where a potential oom-killer victim is skipped if the intersection of the potential victim and the current (the process triggered the oom) is empty based on the reason that killing such victim most probably will not help the current allocating process. However the commit 7887a3da753e ("[PATCH] oom: cpuset hint") changed the heuristic to just decrease the oom_badness scores of such potential victim based on the reason that the cpuset of such processes might have changed and previously they may have allocated memory on mems where the current allocating process can allocate from. Unintentionally 7887a3da753e ("[PATCH] oom: cpuset hint") introduced a side effect as the oom_badness is also exposed to the user space through /proc/[pid]/oom_score, so, readers with different cpusets can read different oom_score of the same process. Later, commit 6cf86ac6f36b ("oom: filter tasks not sharing the same cpuset") fixed the side effect introduced by 7887a3da753e by moving the cpuset intersection back to only oom-killer context and out of oom_badness. However the combination of ab290adbaf8f ("oom: make oom_unkillable_task() helper function") and 26ebc984913b ("oom: /proc/<pid>/oom_score treat kernel thread honestly") unintentionally brought back the cpuset intersection check into the oom_badness calculation function. Other than doing cpuset/mempolicy intersection from oom_badness, the memcg oom context is also doing cpuset/mempolicy intersection which is quite wrong and is caught by syzcaller with the following report: kasan: CONFIG_KASAN_INLINE enabled kasan: GPF could be caused by NULL-ptr deref or user memory access general protection fault: 0000 [#1] PREEMPT SMP KASAN CPU: 0 PID: 28426 Comm: syz-executor.5 Not tainted 5.2.0-rc3-next-20190607 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:__read_once_size include/linux/compiler.h:194 [inline] RIP: 0010:has_intersects_mems_allowed mm/oom_kill.c:84 [inline] RIP: 0010:oom_unkillable_task mm/oom_kill.c:168 [inline] RIP: 0010:oom_unkillable_task+0x180/0x400 mm/oom_kill.c:155 Code: c1 ea 03 80 3c 02 00 0f 85 80 02 00 00 4c 8b a3 10 07 00 00 48 b8 00 00 00 00 00 fc ff df 4d 8d 74 24 10 4c 89 f2 48 c1 ea 03 <80> 3c 02 00 0f 85 67 02 00 00 49 8b 44 24 10 4c 8d a0 68 fa ff ff RSP: 0018:ffff888000127490 EFLAGS: 00010a03 RAX: dffffc0000000000 RBX: ffff8880a4cd5438 RCX: ffffffff818dae9c RDX: 100000000c3cc602 RSI: ffffffff818dac8d RDI: 0000000000000001 RBP: ffff8880001274d0 R08: ffff888000086180 R09: ffffed1015d26be0 R10: ffffed1015d26bdf R11: ffff8880ae935efb R12: 8000000061e63007 R13: 0000000000000000 R14: 8000000061e63017 R15: 1ffff11000024ea6 FS: 00005555561f5940(0000) GS:ffff8880ae800000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000607304 CR3: 000000009237e000 CR4: 00000000001426f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000600 Call Trace: oom_evaluate_task+0x49/0x520 mm/oom_kill.c:321 mem_cgroup_scan_tasks+0xcc/0x180 mm/memcontrol.c:1169 select_bad_process mm/oom_kill.c:374 [inline] out_of_memory mm/oom_kill.c:1088 [inline] out_of_memory+0x6b2/0x1280 mm/oom_kill.c:1035 mem_cgroup_out_of_memory+0x1ca/0x230 mm/memcontrol.c:1573 mem_cgroup_oom mm/memcontrol.c:1905 [inline] try_charge+0xfbe/0x1480 mm/memcontrol.c:2468 mem_cgroup_try_charge+0x24d/0x5e0 mm/memcontrol.c:6073 mem_cgroup_try_charge_delay+0x1f/0xa0 mm/memcontrol.c:6088 do_huge_pmd_wp_page_fallback+0x24f/0x1680 mm/huge_memory.c:1201 do_huge_pmd_wp_page+0x7fc/0x2160 mm/huge_memory.c:1359 wp_huge_pmd mm/memory.c:3793 [inline] __handle_mm_fault+0x164c/0x3eb0 mm/memory.c:4006 handle_mm_fault+0x3b7/0xa90 mm/memory.c:4053 do_user_addr_fault arch/x86/mm/fault.c:1455 [inline] __do_page_fault+0x5ef/0xda0 arch/x86/mm/fault.c:1521 do_page_fault+0x71/0x57d arch/x86/mm/fault.c:1552 page_fault+0x1e/0x30 arch/x86/entry/entry_64.S:1156 RIP: 0033:0x400590 Code: 06 e9 49 01 00 00 48 8b 44 24 10 48 0b 44 24 28 75 1f 48 8b 14 24 48 8b 7c 24 20 be 04 00 00 00 e8 f5 56 00 00 48 8b 74 24 08 <89> 06 e9 1e 01 00 00 48 8b 44 24 08 48 8b 14 24 be 04 00 00 00 8b RSP: 002b:00007fff7bc49780 EFLAGS: 00010206 RAX: 0000000000000001 RBX: 0000000000760000 RCX: 0000000000000000 RDX: 0000000000000000 RSI: 000000002000cffc RDI: 0000000000000001 RBP: fffffffffffffffe R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000075 R11: 0000000000000246 R12: 0000000000760008 R13: 00000000004c55f2 R14: 0000000000000000 R15: 00007fff7bc499b0 Modules linked in: ---[ end trace a65689219582ffff ]--- RIP: 0010:__read_once_size include/linux/compiler.h:194 [inline] RIP: 0010:has_intersects_mems_allowed mm/oom_kill.c:84 [inline] RIP: 0010:oom_unkillable_task mm/oom_kill.c:168 [inline] RIP: 0010:oom_unkillable_task+0x180/0x400 mm/oom_kill.c:155 Code: c1 ea 03 80 3c 02 00 0f 85 80 02 00 00 4c 8b a3 10 07 00 00 48 b8 00 00 00 00 00 fc ff df 4d 8d 74 24 10 4c 89 f2 48 c1 ea 03 <80> 3c 02 00 0f 85 67 02 00 00 49 8b 44 24 10 4c 8d a0 68 fa ff ff RSP: 0018:ffff888000127490 EFLAGS: 00010a03 RAX: dffffc0000000000 RBX: ffff8880a4cd5438 RCX: ffffffff818dae9c RDX: 100000000c3cc602 RSI: ffffffff818dac8d RDI: 0000000000000001 RBP: ffff8880001274d0 R08: ffff888000086180 R09: ffffed1015d26be0 R10: ffffed1015d26bdf R11: ffff8880ae935efb R12: 8000000061e63007 R13: 0000000000000000 R14: 8000000061e63017 R15: 1ffff11000024ea6 FS: 00005555561f5940(0000) GS:ffff8880ae800000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000001b2f823000 CR3: 000000009237e000 CR4: 00000000001426f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000600 The fix is to decouple the cpuset/mempolicy intersection check from oom_unkillable_task() and make sure cpuset/mempolicy intersection check is only done in the global oom context. [shakeelb@google.com: change function name and update comment] Link: http://lkml.kernel.org/r/20190628152421.198994-3-shakeelb@google.com Link: http://lkml.kernel.org/r/20190624212631.87212-3-shakeelb@google.com Signed-off-by: Shakeel Butt <shakeelb@google.com> Reported-by: syzbot+d0fc9d3c166bc5e4a94b@syzkaller.appspotmail.com Acked-by: Roman Gushchin <guro@fb.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Nick Piggin <npiggin@suse.de> Cc: Paul Jackson <pj@sgi.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-07-12 12:00:31 +08:00
if (oom_unkillable_task(p))
return 0;
/* p may not have freeable memory in nodemask */
if (!is_memcg_oom(oc) && !oom_cpuset_eligible(p, oc))
return 0;
task = find_lock_task_mm(p);
if (!task) {
/*
* This is a kthread or all of p's threads have already
* detached their mm's. There's no need to report
* them; they can't be oom killed anyway.
*/
return 0;
}
pr_info("[%7d] %5d %5d %8lu %8lu %8ld %8lu %5hd %s\n",
task->pid, from_kuid(&init_user_ns, task_uid(task)),
task->tgid, task->mm->total_vm, get_mm_rss(task->mm),
mm_pgtables_bytes(task->mm),
get_mm_counter(task->mm, MM_SWAPENTS),
task->signal->oom_score_adj, task->comm);
task_unlock(task);
return 0;
}
/**
* dump_tasks - dump current memory state of all system tasks
* @oc: pointer to struct oom_control
*
* Dumps the current memory state of all eligible tasks. Tasks not in the same
* memcg, not in the same cpuset, or bound to a disjoint set of mempolicy nodes
* are not shown.
mm: consolidate page table accounting Currently, we account page tables separately for each page table level, but that's redundant -- we only make use of total memory allocated to page tables for oom_badness calculation. We also provide the information to userspace, but it has dubious value there too. This patch switches page table accounting to single counter. mm->pgtables_bytes is now used to account all page table levels. We use bytes, because page table size for different levels of page table tree may be different. The change has user-visible effect: we don't have VmPMD and VmPUD reported in /proc/[pid]/status. Not sure if anybody uses them. (As alternative, we can always report 0 kB for them.) OOM-killer report is also slightly changed: we now report pgtables_bytes instead of nr_ptes, nr_pmd, nr_puds. Apart from reducing number of counters per-mm, the benefit is that we now calculate oom_badness() more correctly for machines which have different size of page tables depending on level or where page tables are less than a page in size. The only downside can be debuggability because we do not know which page table level could leak. But I do not remember many bugs that would be caught by separate counters so I wouldn't lose sleep over this. [akpm@linux-foundation.org: fix mm/huge_memory.c] Link: http://lkml.kernel.org/r/20171006100651.44742-2-kirill.shutemov@linux.intel.com Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> [kirill.shutemov@linux.intel.com: fix build] Link: http://lkml.kernel.org/r/20171016150113.ikfxy3e7zzfvsr4w@black.fi.intel.com Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-16 09:35:40 +08:00
* State information includes task's pid, uid, tgid, vm size, rss,
* pgtables_bytes, swapents, oom_score_adj value, and name.
*/
static void dump_tasks(struct oom_control *oc)
{
pr_info("Tasks state (memory values in pages):\n");
pr_info("[ pid ] uid tgid total_vm rss pgtables_bytes swapents oom_score_adj name\n");
if (is_memcg_oom(oc))
mem_cgroup_scan_tasks(oc->memcg, dump_task, oc);
else {
struct task_struct *p;
rcu_read_lock();
for_each_process(p)
dump_task(p, oc);
rcu_read_unlock();
}
}
mm, oom: reorganize the oom report in dump_header OOM report contains several sections. The first one is the allocation context that has triggered the OOM. Then we have cpuset context followed by the stack trace of the OOM path. The tird one is the OOM memory information. Followed by the current memory state of all system tasks. At last, we will show oom eligible tasks and the information about the chosen oom victim. One thing that makes parsing more awkward than necessary is that we do not have a single and easily parsable line about the oom context. This patch is reorganizing the oom report to 1) who invoked oom and what was the allocation request [ 515.902945] tuned invoked oom-killer: gfp_mask=0x6200ca(GFP_HIGHUSER_MOVABLE), order=0, oom_score_adj=0 2) OOM stack trace [ 515.904273] CPU: 24 PID: 1809 Comm: tuned Not tainted 4.20.0-rc3+ #3 [ 515.905518] Hardware name: Inspur SA5212M4/YZMB-00370-107, BIOS 4.1.10 11/14/2016 [ 515.906821] Call Trace: [ 515.908062] dump_stack+0x5a/0x73 [ 515.909311] dump_header+0x55/0x28c [ 515.914260] oom_kill_process+0x2d8/0x300 [ 515.916708] out_of_memory+0x145/0x4a0 [ 515.917932] __alloc_pages_slowpath+0x7d2/0xa16 [ 515.919157] __alloc_pages_nodemask+0x277/0x290 [ 515.920367] filemap_fault+0x3d0/0x6c0 [ 515.921529] ? filemap_map_pages+0x2b8/0x420 [ 515.922709] ext4_filemap_fault+0x2c/0x40 [ext4] [ 515.923884] __do_fault+0x20/0x80 [ 515.925032] __handle_mm_fault+0xbc0/0xe80 [ 515.926195] handle_mm_fault+0xfa/0x210 [ 515.927357] __do_page_fault+0x233/0x4c0 [ 515.928506] do_page_fault+0x32/0x140 [ 515.929646] ? page_fault+0x8/0x30 [ 515.930770] page_fault+0x1e/0x30 3) OOM memory information [ 515.958093] Mem-Info: [ 515.959647] active_anon:26501758 inactive_anon:1179809 isolated_anon:0 active_file:4402672 inactive_file:483963 isolated_file:1344 unevictable:0 dirty:4886753 writeback:0 unstable:0 slab_reclaimable:148442 slab_unreclaimable:18741 mapped:1347 shmem:1347 pagetables:58669 bounce:0 free:88663 free_pcp:0 free_cma:0 ... 4) current memory state of all system tasks [ 516.079544] [ 744] 0 744 9211 1345 114688 82 0 systemd-journal [ 516.082034] [ 787] 0 787 31764 0 143360 92 0 lvmetad [ 516.084465] [ 792] 0 792 10930 1 110592 208 -1000 systemd-udevd [ 516.086865] [ 1199] 0 1199 13866 0 131072 112 -1000 auditd [ 516.089190] [ 1222] 0 1222 31990 1 110592 157 0 smartd [ 516.091477] [ 1225] 0 1225 4864 85 81920 43 0 irqbalance [ 516.093712] [ 1226] 0 1226 52612 0 258048 426 0 abrtd [ 516.112128] [ 1280] 0 1280 109774 55 299008 400 0 NetworkManager [ 516.113998] [ 1295] 0 1295 28817 37 69632 24 0 ksmtuned [ 516.144596] [ 10718] 0 10718 2622484 1721372 15998976 267219 0 panic [ 516.145792] [ 10719] 0 10719 2622484 1164767 9818112 53576 0 panic [ 516.146977] [ 10720] 0 10720 2622484 1174361 9904128 53709 0 panic [ 516.148163] [ 10721] 0 10721 2622484 1209070 10194944 54824 0 panic [ 516.149329] [ 10722] 0 10722 2622484 1745799 14774272 91138 0 panic 5) oom context (contrains and the chosen victim). oom-kill:constraint=CONSTRAINT_NONE,nodemask=(null),cpuset=/,mems_allowed=0-1,task=panic,pid=10737,uid=0 An admin can easily get the full oom context at a single line which makes parsing much easier. Link: http://lkml.kernel.org/r/1542799799-36184-1-git-send-email-ufo19890607@gmail.com Signed-off-by: yuzhoujian <yuzhoujian@didichuxing.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: David Rientjes <rientjes@google.com> Cc: "Kirill A . Shutemov" <kirill.shutemov@linux.intel.com> Cc: Roman Gushchin <guro@fb.com> Cc: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp> Cc: Yang Shi <yang.s@alibaba-inc.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-12-28 16:36:07 +08:00
static void dump_oom_summary(struct oom_control *oc, struct task_struct *victim)
{
/* one line summary of the oom killer context. */
pr_info("oom-kill:constraint=%s,nodemask=%*pbl",
oom_constraint_text[oc->constraint],
nodemask_pr_args(oc->nodemask));
cpuset_print_current_mems_allowed();
mm, oom: add oom victim's memcg to the oom context information The current oom report doesn't display victim's memcg context during the global OOM situation. While this information is not strictly needed, it can be really helpful for containerized environments to locate which container has lost a process. Now that we have a single line for the oom context, we can trivially add both the oom memcg (this can be either global_oom or a specific memcg which hits its hard limits) and task_memcg which is the victim's memcg. Below is the single line output in the oom report after this patch. - global oom context information: oom-kill:constraint=<constraint>,nodemask=<nodemask>,cpuset=<cpuset>,mems_allowed=<mems_allowed>,global_oom,task_memcg=<memcg>,task=<comm>,pid=<pid>,uid=<uid> - memcg oom context information: oom-kill:constraint=<constraint>,nodemask=<nodemask>,cpuset=<cpuset>,mems_allowed=<mems_allowed>,oom_memcg=<memcg>,task_memcg=<memcg>,task=<comm>,pid=<pid>,uid=<uid> [penguin-kernel@I-love.SAKURA.ne.jp: use pr_cont() in mem_cgroup_print_oom_context()] Link: http://lkml.kernel.org/r/201812190723.wBJ7NdkN032628@www262.sakura.ne.jp Link: http://lkml.kernel.org/r/1542799799-36184-2-git-send-email-ufo19890607@gmail.com Signed-off-by: yuzhoujian <yuzhoujian@didichuxing.com> Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Acked-by: Michal Hocko <mhocko@suse.com> Cc: David Rientjes <rientjes@google.com> Cc: "Kirill A . Shutemov" <kirill.shutemov@linux.intel.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp> Cc: Roman Gushchin <guro@fb.com> Cc: Yang Shi <yang.s@alibaba-inc.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-12-28 16:36:10 +08:00
mem_cgroup_print_oom_context(oc->memcg, victim);
mm, oom: reorganize the oom report in dump_header OOM report contains several sections. The first one is the allocation context that has triggered the OOM. Then we have cpuset context followed by the stack trace of the OOM path. The tird one is the OOM memory information. Followed by the current memory state of all system tasks. At last, we will show oom eligible tasks and the information about the chosen oom victim. One thing that makes parsing more awkward than necessary is that we do not have a single and easily parsable line about the oom context. This patch is reorganizing the oom report to 1) who invoked oom and what was the allocation request [ 515.902945] tuned invoked oom-killer: gfp_mask=0x6200ca(GFP_HIGHUSER_MOVABLE), order=0, oom_score_adj=0 2) OOM stack trace [ 515.904273] CPU: 24 PID: 1809 Comm: tuned Not tainted 4.20.0-rc3+ #3 [ 515.905518] Hardware name: Inspur SA5212M4/YZMB-00370-107, BIOS 4.1.10 11/14/2016 [ 515.906821] Call Trace: [ 515.908062] dump_stack+0x5a/0x73 [ 515.909311] dump_header+0x55/0x28c [ 515.914260] oom_kill_process+0x2d8/0x300 [ 515.916708] out_of_memory+0x145/0x4a0 [ 515.917932] __alloc_pages_slowpath+0x7d2/0xa16 [ 515.919157] __alloc_pages_nodemask+0x277/0x290 [ 515.920367] filemap_fault+0x3d0/0x6c0 [ 515.921529] ? filemap_map_pages+0x2b8/0x420 [ 515.922709] ext4_filemap_fault+0x2c/0x40 [ext4] [ 515.923884] __do_fault+0x20/0x80 [ 515.925032] __handle_mm_fault+0xbc0/0xe80 [ 515.926195] handle_mm_fault+0xfa/0x210 [ 515.927357] __do_page_fault+0x233/0x4c0 [ 515.928506] do_page_fault+0x32/0x140 [ 515.929646] ? page_fault+0x8/0x30 [ 515.930770] page_fault+0x1e/0x30 3) OOM memory information [ 515.958093] Mem-Info: [ 515.959647] active_anon:26501758 inactive_anon:1179809 isolated_anon:0 active_file:4402672 inactive_file:483963 isolated_file:1344 unevictable:0 dirty:4886753 writeback:0 unstable:0 slab_reclaimable:148442 slab_unreclaimable:18741 mapped:1347 shmem:1347 pagetables:58669 bounce:0 free:88663 free_pcp:0 free_cma:0 ... 4) current memory state of all system tasks [ 516.079544] [ 744] 0 744 9211 1345 114688 82 0 systemd-journal [ 516.082034] [ 787] 0 787 31764 0 143360 92 0 lvmetad [ 516.084465] [ 792] 0 792 10930 1 110592 208 -1000 systemd-udevd [ 516.086865] [ 1199] 0 1199 13866 0 131072 112 -1000 auditd [ 516.089190] [ 1222] 0 1222 31990 1 110592 157 0 smartd [ 516.091477] [ 1225] 0 1225 4864 85 81920 43 0 irqbalance [ 516.093712] [ 1226] 0 1226 52612 0 258048 426 0 abrtd [ 516.112128] [ 1280] 0 1280 109774 55 299008 400 0 NetworkManager [ 516.113998] [ 1295] 0 1295 28817 37 69632 24 0 ksmtuned [ 516.144596] [ 10718] 0 10718 2622484 1721372 15998976 267219 0 panic [ 516.145792] [ 10719] 0 10719 2622484 1164767 9818112 53576 0 panic [ 516.146977] [ 10720] 0 10720 2622484 1174361 9904128 53709 0 panic [ 516.148163] [ 10721] 0 10721 2622484 1209070 10194944 54824 0 panic [ 516.149329] [ 10722] 0 10722 2622484 1745799 14774272 91138 0 panic 5) oom context (contrains and the chosen victim). oom-kill:constraint=CONSTRAINT_NONE,nodemask=(null),cpuset=/,mems_allowed=0-1,task=panic,pid=10737,uid=0 An admin can easily get the full oom context at a single line which makes parsing much easier. Link: http://lkml.kernel.org/r/1542799799-36184-1-git-send-email-ufo19890607@gmail.com Signed-off-by: yuzhoujian <yuzhoujian@didichuxing.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: David Rientjes <rientjes@google.com> Cc: "Kirill A . Shutemov" <kirill.shutemov@linux.intel.com> Cc: Roman Gushchin <guro@fb.com> Cc: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp> Cc: Yang Shi <yang.s@alibaba-inc.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-12-28 16:36:07 +08:00
pr_cont(",task=%s,pid=%d,uid=%d\n", victim->comm, victim->pid,
from_kuid(&init_user_ns, task_uid(victim)));
}
static void dump_header(struct oom_control *oc, struct task_struct *p)
{
mm, oom: reorganize the oom report in dump_header OOM report contains several sections. The first one is the allocation context that has triggered the OOM. Then we have cpuset context followed by the stack trace of the OOM path. The tird one is the OOM memory information. Followed by the current memory state of all system tasks. At last, we will show oom eligible tasks and the information about the chosen oom victim. One thing that makes parsing more awkward than necessary is that we do not have a single and easily parsable line about the oom context. This patch is reorganizing the oom report to 1) who invoked oom and what was the allocation request [ 515.902945] tuned invoked oom-killer: gfp_mask=0x6200ca(GFP_HIGHUSER_MOVABLE), order=0, oom_score_adj=0 2) OOM stack trace [ 515.904273] CPU: 24 PID: 1809 Comm: tuned Not tainted 4.20.0-rc3+ #3 [ 515.905518] Hardware name: Inspur SA5212M4/YZMB-00370-107, BIOS 4.1.10 11/14/2016 [ 515.906821] Call Trace: [ 515.908062] dump_stack+0x5a/0x73 [ 515.909311] dump_header+0x55/0x28c [ 515.914260] oom_kill_process+0x2d8/0x300 [ 515.916708] out_of_memory+0x145/0x4a0 [ 515.917932] __alloc_pages_slowpath+0x7d2/0xa16 [ 515.919157] __alloc_pages_nodemask+0x277/0x290 [ 515.920367] filemap_fault+0x3d0/0x6c0 [ 515.921529] ? filemap_map_pages+0x2b8/0x420 [ 515.922709] ext4_filemap_fault+0x2c/0x40 [ext4] [ 515.923884] __do_fault+0x20/0x80 [ 515.925032] __handle_mm_fault+0xbc0/0xe80 [ 515.926195] handle_mm_fault+0xfa/0x210 [ 515.927357] __do_page_fault+0x233/0x4c0 [ 515.928506] do_page_fault+0x32/0x140 [ 515.929646] ? page_fault+0x8/0x30 [ 515.930770] page_fault+0x1e/0x30 3) OOM memory information [ 515.958093] Mem-Info: [ 515.959647] active_anon:26501758 inactive_anon:1179809 isolated_anon:0 active_file:4402672 inactive_file:483963 isolated_file:1344 unevictable:0 dirty:4886753 writeback:0 unstable:0 slab_reclaimable:148442 slab_unreclaimable:18741 mapped:1347 shmem:1347 pagetables:58669 bounce:0 free:88663 free_pcp:0 free_cma:0 ... 4) current memory state of all system tasks [ 516.079544] [ 744] 0 744 9211 1345 114688 82 0 systemd-journal [ 516.082034] [ 787] 0 787 31764 0 143360 92 0 lvmetad [ 516.084465] [ 792] 0 792 10930 1 110592 208 -1000 systemd-udevd [ 516.086865] [ 1199] 0 1199 13866 0 131072 112 -1000 auditd [ 516.089190] [ 1222] 0 1222 31990 1 110592 157 0 smartd [ 516.091477] [ 1225] 0 1225 4864 85 81920 43 0 irqbalance [ 516.093712] [ 1226] 0 1226 52612 0 258048 426 0 abrtd [ 516.112128] [ 1280] 0 1280 109774 55 299008 400 0 NetworkManager [ 516.113998] [ 1295] 0 1295 28817 37 69632 24 0 ksmtuned [ 516.144596] [ 10718] 0 10718 2622484 1721372 15998976 267219 0 panic [ 516.145792] [ 10719] 0 10719 2622484 1164767 9818112 53576 0 panic [ 516.146977] [ 10720] 0 10720 2622484 1174361 9904128 53709 0 panic [ 516.148163] [ 10721] 0 10721 2622484 1209070 10194944 54824 0 panic [ 516.149329] [ 10722] 0 10722 2622484 1745799 14774272 91138 0 panic 5) oom context (contrains and the chosen victim). oom-kill:constraint=CONSTRAINT_NONE,nodemask=(null),cpuset=/,mems_allowed=0-1,task=panic,pid=10737,uid=0 An admin can easily get the full oom context at a single line which makes parsing much easier. Link: http://lkml.kernel.org/r/1542799799-36184-1-git-send-email-ufo19890607@gmail.com Signed-off-by: yuzhoujian <yuzhoujian@didichuxing.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: David Rientjes <rientjes@google.com> Cc: "Kirill A . Shutemov" <kirill.shutemov@linux.intel.com> Cc: Roman Gushchin <guro@fb.com> Cc: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp> Cc: Yang Shi <yang.s@alibaba-inc.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-12-28 16:36:07 +08:00
pr_warn("%s invoked oom-killer: gfp_mask=%#x(%pGg), order=%d, oom_score_adj=%hd\n",
current->comm, oc->gfp_mask, &oc->gfp_mask, oc->order,
current->signal->oom_score_adj);
if (!IS_ENABLED(CONFIG_COMPACTION) && oc->order)
pr_warn("COMPACTION is disabled!!!\n");
dump_stack();
mm: oom: show unreclaimable slab info when unreclaimable slabs > user memory The kernel may panic when an oom happens without killable process sometimes it is caused by huge unreclaimable slabs used by kernel. Although kdump could help debug such problem, however, kdump is not available on all architectures and it might be malfunction sometime. And, since kernel already panic it is worthy capturing such information in dmesg to aid touble shooting. Print out unreclaimable slab info (used size and total size) which actual memory usage is not zero (num_objs * size != 0) when unreclaimable slabs amount is greater than total user memory (LRU pages). The output looks like: Unreclaimable slab info: Name Used Total rpc_buffers 31KB 31KB rpc_tasks 7KB 7KB ebitmap_node 1964KB 1964KB avtab_node 5024KB 5024KB xfs_buf 1402KB 1402KB xfs_ili 134KB 134KB xfs_efi_item 115KB 115KB xfs_efd_item 115KB 115KB xfs_buf_item 134KB 134KB xfs_log_item_desc 342KB 342KB xfs_trans 1412KB 1412KB xfs_ifork 212KB 212KB [yang.s@alibaba-inc.com: v11] Link: http://lkml.kernel.org/r/1507656303-103845-4-git-send-email-yang.s@alibaba-inc.com Link: http://lkml.kernel.org/r/1507152550-46205-4-git-send-email-yang.s@alibaba-inc.com Signed-off-by: Yang Shi <yang.s@alibaba-inc.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Christoph Lameter <cl@linux.com> Cc: David Rientjes <rientjes@google.com> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Pekka Enberg <penberg@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-16 09:32:07 +08:00
if (is_memcg_oom(oc))
mm, oom: add oom victim's memcg to the oom context information The current oom report doesn't display victim's memcg context during the global OOM situation. While this information is not strictly needed, it can be really helpful for containerized environments to locate which container has lost a process. Now that we have a single line for the oom context, we can trivially add both the oom memcg (this can be either global_oom or a specific memcg which hits its hard limits) and task_memcg which is the victim's memcg. Below is the single line output in the oom report after this patch. - global oom context information: oom-kill:constraint=<constraint>,nodemask=<nodemask>,cpuset=<cpuset>,mems_allowed=<mems_allowed>,global_oom,task_memcg=<memcg>,task=<comm>,pid=<pid>,uid=<uid> - memcg oom context information: oom-kill:constraint=<constraint>,nodemask=<nodemask>,cpuset=<cpuset>,mems_allowed=<mems_allowed>,oom_memcg=<memcg>,task_memcg=<memcg>,task=<comm>,pid=<pid>,uid=<uid> [penguin-kernel@I-love.SAKURA.ne.jp: use pr_cont() in mem_cgroup_print_oom_context()] Link: http://lkml.kernel.org/r/201812190723.wBJ7NdkN032628@www262.sakura.ne.jp Link: http://lkml.kernel.org/r/1542799799-36184-2-git-send-email-ufo19890607@gmail.com Signed-off-by: yuzhoujian <yuzhoujian@didichuxing.com> Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Acked-by: Michal Hocko <mhocko@suse.com> Cc: David Rientjes <rientjes@google.com> Cc: "Kirill A . Shutemov" <kirill.shutemov@linux.intel.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp> Cc: Roman Gushchin <guro@fb.com> Cc: Yang Shi <yang.s@alibaba-inc.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-12-28 16:36:10 +08:00
mem_cgroup_print_oom_meminfo(oc->memcg);
mm: oom: show unreclaimable slab info when unreclaimable slabs > user memory The kernel may panic when an oom happens without killable process sometimes it is caused by huge unreclaimable slabs used by kernel. Although kdump could help debug such problem, however, kdump is not available on all architectures and it might be malfunction sometime. And, since kernel already panic it is worthy capturing such information in dmesg to aid touble shooting. Print out unreclaimable slab info (used size and total size) which actual memory usage is not zero (num_objs * size != 0) when unreclaimable slabs amount is greater than total user memory (LRU pages). The output looks like: Unreclaimable slab info: Name Used Total rpc_buffers 31KB 31KB rpc_tasks 7KB 7KB ebitmap_node 1964KB 1964KB avtab_node 5024KB 5024KB xfs_buf 1402KB 1402KB xfs_ili 134KB 134KB xfs_efi_item 115KB 115KB xfs_efd_item 115KB 115KB xfs_buf_item 134KB 134KB xfs_log_item_desc 342KB 342KB xfs_trans 1412KB 1412KB xfs_ifork 212KB 212KB [yang.s@alibaba-inc.com: v11] Link: http://lkml.kernel.org/r/1507656303-103845-4-git-send-email-yang.s@alibaba-inc.com Link: http://lkml.kernel.org/r/1507152550-46205-4-git-send-email-yang.s@alibaba-inc.com Signed-off-by: Yang Shi <yang.s@alibaba-inc.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Christoph Lameter <cl@linux.com> Cc: David Rientjes <rientjes@google.com> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Pekka Enberg <penberg@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-16 09:32:07 +08:00
else {
show_mem(SHOW_MEM_FILTER_NODES, oc->nodemask);
if (should_dump_unreclaim_slab())
mm: oom: show unreclaimable slab info when unreclaimable slabs > user memory The kernel may panic when an oom happens without killable process sometimes it is caused by huge unreclaimable slabs used by kernel. Although kdump could help debug such problem, however, kdump is not available on all architectures and it might be malfunction sometime. And, since kernel already panic it is worthy capturing such information in dmesg to aid touble shooting. Print out unreclaimable slab info (used size and total size) which actual memory usage is not zero (num_objs * size != 0) when unreclaimable slabs amount is greater than total user memory (LRU pages). The output looks like: Unreclaimable slab info: Name Used Total rpc_buffers 31KB 31KB rpc_tasks 7KB 7KB ebitmap_node 1964KB 1964KB avtab_node 5024KB 5024KB xfs_buf 1402KB 1402KB xfs_ili 134KB 134KB xfs_efi_item 115KB 115KB xfs_efd_item 115KB 115KB xfs_buf_item 134KB 134KB xfs_log_item_desc 342KB 342KB xfs_trans 1412KB 1412KB xfs_ifork 212KB 212KB [yang.s@alibaba-inc.com: v11] Link: http://lkml.kernel.org/r/1507656303-103845-4-git-send-email-yang.s@alibaba-inc.com Link: http://lkml.kernel.org/r/1507152550-46205-4-git-send-email-yang.s@alibaba-inc.com Signed-off-by: Yang Shi <yang.s@alibaba-inc.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Christoph Lameter <cl@linux.com> Cc: David Rientjes <rientjes@google.com> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Pekka Enberg <penberg@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-16 09:32:07 +08:00
dump_unreclaimable_slab();
}
if (sysctl_oom_dump_tasks)
dump_tasks(oc);
mm, oom: reorganize the oom report in dump_header OOM report contains several sections. The first one is the allocation context that has triggered the OOM. Then we have cpuset context followed by the stack trace of the OOM path. The tird one is the OOM memory information. Followed by the current memory state of all system tasks. At last, we will show oom eligible tasks and the information about the chosen oom victim. One thing that makes parsing more awkward than necessary is that we do not have a single and easily parsable line about the oom context. This patch is reorganizing the oom report to 1) who invoked oom and what was the allocation request [ 515.902945] tuned invoked oom-killer: gfp_mask=0x6200ca(GFP_HIGHUSER_MOVABLE), order=0, oom_score_adj=0 2) OOM stack trace [ 515.904273] CPU: 24 PID: 1809 Comm: tuned Not tainted 4.20.0-rc3+ #3 [ 515.905518] Hardware name: Inspur SA5212M4/YZMB-00370-107, BIOS 4.1.10 11/14/2016 [ 515.906821] Call Trace: [ 515.908062] dump_stack+0x5a/0x73 [ 515.909311] dump_header+0x55/0x28c [ 515.914260] oom_kill_process+0x2d8/0x300 [ 515.916708] out_of_memory+0x145/0x4a0 [ 515.917932] __alloc_pages_slowpath+0x7d2/0xa16 [ 515.919157] __alloc_pages_nodemask+0x277/0x290 [ 515.920367] filemap_fault+0x3d0/0x6c0 [ 515.921529] ? filemap_map_pages+0x2b8/0x420 [ 515.922709] ext4_filemap_fault+0x2c/0x40 [ext4] [ 515.923884] __do_fault+0x20/0x80 [ 515.925032] __handle_mm_fault+0xbc0/0xe80 [ 515.926195] handle_mm_fault+0xfa/0x210 [ 515.927357] __do_page_fault+0x233/0x4c0 [ 515.928506] do_page_fault+0x32/0x140 [ 515.929646] ? page_fault+0x8/0x30 [ 515.930770] page_fault+0x1e/0x30 3) OOM memory information [ 515.958093] Mem-Info: [ 515.959647] active_anon:26501758 inactive_anon:1179809 isolated_anon:0 active_file:4402672 inactive_file:483963 isolated_file:1344 unevictable:0 dirty:4886753 writeback:0 unstable:0 slab_reclaimable:148442 slab_unreclaimable:18741 mapped:1347 shmem:1347 pagetables:58669 bounce:0 free:88663 free_pcp:0 free_cma:0 ... 4) current memory state of all system tasks [ 516.079544] [ 744] 0 744 9211 1345 114688 82 0 systemd-journal [ 516.082034] [ 787] 0 787 31764 0 143360 92 0 lvmetad [ 516.084465] [ 792] 0 792 10930 1 110592 208 -1000 systemd-udevd [ 516.086865] [ 1199] 0 1199 13866 0 131072 112 -1000 auditd [ 516.089190] [ 1222] 0 1222 31990 1 110592 157 0 smartd [ 516.091477] [ 1225] 0 1225 4864 85 81920 43 0 irqbalance [ 516.093712] [ 1226] 0 1226 52612 0 258048 426 0 abrtd [ 516.112128] [ 1280] 0 1280 109774 55 299008 400 0 NetworkManager [ 516.113998] [ 1295] 0 1295 28817 37 69632 24 0 ksmtuned [ 516.144596] [ 10718] 0 10718 2622484 1721372 15998976 267219 0 panic [ 516.145792] [ 10719] 0 10719 2622484 1164767 9818112 53576 0 panic [ 516.146977] [ 10720] 0 10720 2622484 1174361 9904128 53709 0 panic [ 516.148163] [ 10721] 0 10721 2622484 1209070 10194944 54824 0 panic [ 516.149329] [ 10722] 0 10722 2622484 1745799 14774272 91138 0 panic 5) oom context (contrains and the chosen victim). oom-kill:constraint=CONSTRAINT_NONE,nodemask=(null),cpuset=/,mems_allowed=0-1,task=panic,pid=10737,uid=0 An admin can easily get the full oom context at a single line which makes parsing much easier. Link: http://lkml.kernel.org/r/1542799799-36184-1-git-send-email-ufo19890607@gmail.com Signed-off-by: yuzhoujian <yuzhoujian@didichuxing.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: David Rientjes <rientjes@google.com> Cc: "Kirill A . Shutemov" <kirill.shutemov@linux.intel.com> Cc: Roman Gushchin <guro@fb.com> Cc: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp> Cc: Yang Shi <yang.s@alibaba-inc.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-12-28 16:36:07 +08:00
if (p)
dump_oom_summary(oc, p);
}
OOM, PM: OOM killed task shouldn't escape PM suspend PM freezer relies on having all tasks frozen by the time devices are getting frozen so that no task will touch them while they are getting frozen. But OOM killer is allowed to kill an already frozen task in order to handle OOM situtation. In order to protect from late wake ups OOM killer is disabled after all tasks are frozen. This, however, still keeps a window open when a killed task didn't manage to die by the time freeze_processes finishes. Reduce the race window by checking all tasks after OOM killer has been disabled. This is still not race free completely unfortunately because oom_killer_disable cannot stop an already ongoing OOM killer so a task might still wake up from the fridge and get killed without freeze_processes noticing. Full synchronization of OOM and freezer is, however, too heavy weight for this highly unlikely case. Introduce and check oom_kills counter which gets incremented early when the allocator enters __alloc_pages_may_oom path and only check all the tasks if the counter changes during the freezing attempt. The counter is updated so early to reduce the race window since allocator checked oom_killer_disabled which is set by PM-freezing code. A false positive will push the PM-freezer into a slow path but that is not a big deal. Changes since v1 - push the re-check loop out of freeze_processes into check_frozen_processes and invert the condition to make the code more readable as per Rafael Fixes: f660daac474c6f (oom: thaw threads if oom killed thread is frozen before deferring) Cc: 3.2+ <stable@vger.kernel.org> # 3.2+ Signed-off-by: Michal Hocko <mhocko@suse.cz> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-10-21 00:12:32 +08:00
/*
oom, PM: make OOM detection in the freezer path raceless Commit 5695be142e20 ("OOM, PM: OOM killed task shouldn't escape PM suspend") has left a race window when OOM killer manages to note_oom_kill after freeze_processes checks the counter. The race window is quite small and really unlikely and partial solution deemed sufficient at the time of submission. Tejun wasn't happy about this partial solution though and insisted on a full solution. That requires the full OOM and freezer's task freezing exclusion, though. This is done by this patch which introduces oom_sem RW lock and turns oom_killer_disable() into a full OOM barrier. oom_killer_disabled check is moved from the allocation path to the OOM level and we take oom_sem for reading for both the check and the whole OOM invocation. oom_killer_disable() takes oom_sem for writing so it waits for all currently running OOM killer invocations. Then it disable all the further OOMs by setting oom_killer_disabled and checks for any oom victims. Victims are counted via mark_tsk_oom_victim resp. unmark_oom_victim. The last victim wakes up all waiters enqueued by oom_killer_disable(). Therefore this function acts as the full OOM barrier. The page fault path is covered now as well although it was assumed to be safe before. As per Tejun, "We used to have freezing points deep in file system code which may be reacheable from page fault." so it would be better and more robust to not rely on freezing points here. Same applies to the memcg OOM killer. out_of_memory tells the caller whether the OOM was allowed to trigger and the callers are supposed to handle the situation. The page allocation path simply fails the allocation same as before. The page fault path will retry the fault (more on that later) and Sysrq OOM trigger will simply complain to the log. Normally there wouldn't be any unfrozen user tasks after try_to_freeze_tasks so the function will not block. But if there was an OOM killer racing with try_to_freeze_tasks and the OOM victim didn't finish yet then we have to wait for it. This should complete in a finite time, though, because - the victim cannot loop in the page fault handler (it would die on the way out from the exception) - it cannot loop in the page allocator because all the further allocation would fail and __GFP_NOFAIL allocations are not acceptable at this stage - it shouldn't be blocked on any locks held by frozen tasks (try_to_freeze expects lockless context) and kernel threads and work queues are not frozen yet Signed-off-by: Michal Hocko <mhocko@suse.cz> Suggested-by: Tejun Heo <tj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-02-12 07:26:24 +08:00
* Number of OOM victims in flight
OOM, PM: OOM killed task shouldn't escape PM suspend PM freezer relies on having all tasks frozen by the time devices are getting frozen so that no task will touch them while they are getting frozen. But OOM killer is allowed to kill an already frozen task in order to handle OOM situtation. In order to protect from late wake ups OOM killer is disabled after all tasks are frozen. This, however, still keeps a window open when a killed task didn't manage to die by the time freeze_processes finishes. Reduce the race window by checking all tasks after OOM killer has been disabled. This is still not race free completely unfortunately because oom_killer_disable cannot stop an already ongoing OOM killer so a task might still wake up from the fridge and get killed without freeze_processes noticing. Full synchronization of OOM and freezer is, however, too heavy weight for this highly unlikely case. Introduce and check oom_kills counter which gets incremented early when the allocator enters __alloc_pages_may_oom path and only check all the tasks if the counter changes during the freezing attempt. The counter is updated so early to reduce the race window since allocator checked oom_killer_disabled which is set by PM-freezing code. A false positive will push the PM-freezer into a slow path but that is not a big deal. Changes since v1 - push the re-check loop out of freeze_processes into check_frozen_processes and invert the condition to make the code more readable as per Rafael Fixes: f660daac474c6f (oom: thaw threads if oom killed thread is frozen before deferring) Cc: 3.2+ <stable@vger.kernel.org> # 3.2+ Signed-off-by: Michal Hocko <mhocko@suse.cz> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-10-21 00:12:32 +08:00
*/
oom, PM: make OOM detection in the freezer path raceless Commit 5695be142e20 ("OOM, PM: OOM killed task shouldn't escape PM suspend") has left a race window when OOM killer manages to note_oom_kill after freeze_processes checks the counter. The race window is quite small and really unlikely and partial solution deemed sufficient at the time of submission. Tejun wasn't happy about this partial solution though and insisted on a full solution. That requires the full OOM and freezer's task freezing exclusion, though. This is done by this patch which introduces oom_sem RW lock and turns oom_killer_disable() into a full OOM barrier. oom_killer_disabled check is moved from the allocation path to the OOM level and we take oom_sem for reading for both the check and the whole OOM invocation. oom_killer_disable() takes oom_sem for writing so it waits for all currently running OOM killer invocations. Then it disable all the further OOMs by setting oom_killer_disabled and checks for any oom victims. Victims are counted via mark_tsk_oom_victim resp. unmark_oom_victim. The last victim wakes up all waiters enqueued by oom_killer_disable(). Therefore this function acts as the full OOM barrier. The page fault path is covered now as well although it was assumed to be safe before. As per Tejun, "We used to have freezing points deep in file system code which may be reacheable from page fault." so it would be better and more robust to not rely on freezing points here. Same applies to the memcg OOM killer. out_of_memory tells the caller whether the OOM was allowed to trigger and the callers are supposed to handle the situation. The page allocation path simply fails the allocation same as before. The page fault path will retry the fault (more on that later) and Sysrq OOM trigger will simply complain to the log. Normally there wouldn't be any unfrozen user tasks after try_to_freeze_tasks so the function will not block. But if there was an OOM killer racing with try_to_freeze_tasks and the OOM victim didn't finish yet then we have to wait for it. This should complete in a finite time, though, because - the victim cannot loop in the page fault handler (it would die on the way out from the exception) - it cannot loop in the page allocator because all the further allocation would fail and __GFP_NOFAIL allocations are not acceptable at this stage - it shouldn't be blocked on any locks held by frozen tasks (try_to_freeze expects lockless context) and kernel threads and work queues are not frozen yet Signed-off-by: Michal Hocko <mhocko@suse.cz> Suggested-by: Tejun Heo <tj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-02-12 07:26:24 +08:00
static atomic_t oom_victims = ATOMIC_INIT(0);
static DECLARE_WAIT_QUEUE_HEAD(oom_victims_wait);
OOM, PM: OOM killed task shouldn't escape PM suspend PM freezer relies on having all tasks frozen by the time devices are getting frozen so that no task will touch them while they are getting frozen. But OOM killer is allowed to kill an already frozen task in order to handle OOM situtation. In order to protect from late wake ups OOM killer is disabled after all tasks are frozen. This, however, still keeps a window open when a killed task didn't manage to die by the time freeze_processes finishes. Reduce the race window by checking all tasks after OOM killer has been disabled. This is still not race free completely unfortunately because oom_killer_disable cannot stop an already ongoing OOM killer so a task might still wake up from the fridge and get killed without freeze_processes noticing. Full synchronization of OOM and freezer is, however, too heavy weight for this highly unlikely case. Introduce and check oom_kills counter which gets incremented early when the allocator enters __alloc_pages_may_oom path and only check all the tasks if the counter changes during the freezing attempt. The counter is updated so early to reduce the race window since allocator checked oom_killer_disabled which is set by PM-freezing code. A false positive will push the PM-freezer into a slow path but that is not a big deal. Changes since v1 - push the re-check loop out of freeze_processes into check_frozen_processes and invert the condition to make the code more readable as per Rafael Fixes: f660daac474c6f (oom: thaw threads if oom killed thread is frozen before deferring) Cc: 3.2+ <stable@vger.kernel.org> # 3.2+ Signed-off-by: Michal Hocko <mhocko@suse.cz> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-10-21 00:12:32 +08:00
static bool oom_killer_disabled __read_mostly;
OOM, PM: OOM killed task shouldn't escape PM suspend PM freezer relies on having all tasks frozen by the time devices are getting frozen so that no task will touch them while they are getting frozen. But OOM killer is allowed to kill an already frozen task in order to handle OOM situtation. In order to protect from late wake ups OOM killer is disabled after all tasks are frozen. This, however, still keeps a window open when a killed task didn't manage to die by the time freeze_processes finishes. Reduce the race window by checking all tasks after OOM killer has been disabled. This is still not race free completely unfortunately because oom_killer_disable cannot stop an already ongoing OOM killer so a task might still wake up from the fridge and get killed without freeze_processes noticing. Full synchronization of OOM and freezer is, however, too heavy weight for this highly unlikely case. Introduce and check oom_kills counter which gets incremented early when the allocator enters __alloc_pages_may_oom path and only check all the tasks if the counter changes during the freezing attempt. The counter is updated so early to reduce the race window since allocator checked oom_killer_disabled which is set by PM-freezing code. A false positive will push the PM-freezer into a slow path but that is not a big deal. Changes since v1 - push the re-check loop out of freeze_processes into check_frozen_processes and invert the condition to make the code more readable as per Rafael Fixes: f660daac474c6f (oom: thaw threads if oom killed thread is frozen before deferring) Cc: 3.2+ <stable@vger.kernel.org> # 3.2+ Signed-off-by: Michal Hocko <mhocko@suse.cz> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-10-21 00:12:32 +08:00
#define K(x) ((x) << (PAGE_SHIFT-10))
/*
* task->mm can be NULL if the task is the exited group leader. So to
* determine whether the task is using a particular mm, we examine all the
* task's threads: if one of those is using this mm then this task was also
* using it.
*/
mm, oom_adj: make sure processes sharing mm have same view of oom_score_adj oom_score_adj is shared for the thread groups (via struct signal) but this is not sufficient to cover processes sharing mm (CLONE_VM without CLONE_SIGHAND) and so we can easily end up in a situation when some processes update their oom_score_adj and confuse the oom killer. In the worst case some of those processes might hide from the oom killer altogether via OOM_SCORE_ADJ_MIN while others are eligible. OOM killer would then pick up those eligible but won't be allowed to kill others sharing the same mm so the mm wouldn't release the mm and so the memory. It would be ideal to have the oom_score_adj per mm_struct because that is the natural entity OOM killer considers. But this will not work because some programs are doing vfork() set_oom_adj() exec() We can achieve the same though. oom_score_adj write handler can set the oom_score_adj for all processes sharing the same mm if the task is not in the middle of vfork. As a result all the processes will share the same oom_score_adj. The current implementation is rather pessimistic and checks all the existing processes by default if there is more than 1 holder of the mm but we do not have any reliable way to check for external users yet. Link: http://lkml.kernel.org/r/1466426628-15074-5-git-send-email-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Acked-by: Oleg Nesterov <oleg@redhat.com> Cc: Vladimir Davydov <vdavydov@virtuozzo.com> Cc: David Rientjes <rientjes@google.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-07-29 06:44:43 +08:00
bool process_shares_mm(struct task_struct *p, struct mm_struct *mm)
{
struct task_struct *t;
for_each_thread(p, t) {
struct mm_struct *t_mm = READ_ONCE(t->mm);
if (t_mm)
return t_mm == mm;
}
return false;
}
mm, oom: introduce oom reaper This patch (of 5): This is based on the idea from Mel Gorman discussed during LSFMM 2015 and independently brought up by Oleg Nesterov. The OOM killer currently allows to kill only a single task in a good hope that the task will terminate in a reasonable time and frees up its memory. Such a task (oom victim) will get an access to memory reserves via mark_oom_victim to allow a forward progress should there be a need for additional memory during exit path. It has been shown (e.g. by Tetsuo Handa) that it is not that hard to construct workloads which break the core assumption mentioned above and the OOM victim might take unbounded amount of time to exit because it might be blocked in the uninterruptible state waiting for an event (e.g. lock) which is blocked by another task looping in the page allocator. This patch reduces the probability of such a lockup by introducing a specialized kernel thread (oom_reaper) which tries to reclaim additional memory by preemptively reaping the anonymous or swapped out memory owned by the oom victim under an assumption that such a memory won't be needed when its owner is killed and kicked from the userspace anyway. There is one notable exception to this, though, if the OOM victim was in the process of coredumping the result would be incomplete. This is considered a reasonable constrain because the overall system health is more important than debugability of a particular application. A kernel thread has been chosen because we need a reliable way of invocation so workqueue context is not appropriate because all the workers might be busy (e.g. allocating memory). Kswapd which sounds like another good fit is not appropriate as well because it might get blocked on locks during reclaim as well. oom_reaper has to take mmap_sem on the target task for reading so the solution is not 100% because the semaphore might be held or blocked for write but the probability is reduced considerably wrt. basically any lock blocking forward progress as described above. In order to prevent from blocking on the lock without any forward progress we are using only a trylock and retry 10 times with a short sleep in between. Users of mmap_sem which need it for write should be carefully reviewed to use _killable waiting as much as possible and reduce allocations requests done with the lock held to absolute minimum to reduce the risk even further. The API between oom killer and oom reaper is quite trivial. wake_oom_reaper updates mm_to_reap with cmpxchg to guarantee only NULL->mm transition and oom_reaper clear this atomically once it is done with the work. This means that only a single mm_struct can be reaped at the time. As the operation is potentially disruptive we are trying to limit it to the ncessary minimum and the reaper blocks any updates while it operates on an mm. mm_struct is pinned by mm_count to allow parallel exit_mmap and a race is detected by atomic_inc_not_zero(mm_users). Signed-off-by: Michal Hocko <mhocko@suse.com> Suggested-by: Oleg Nesterov <oleg@redhat.com> Suggested-by: Mel Gorman <mgorman@suse.de> Acked-by: Mel Gorman <mgorman@suse.de> Acked-by: David Rientjes <rientjes@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Hugh Dickins <hughd@google.com> Cc: Andrea Argangeli <andrea@kernel.org> Cc: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:24 +08:00
#ifdef CONFIG_MMU
/*
* OOM Reaper kernel thread which tries to reap the memory used by the OOM
* victim (if that is possible) to help the OOM killer to move on.
*/
static struct task_struct *oom_reaper_th;
static DECLARE_WAIT_QUEUE_HEAD(oom_reaper_wait);
static struct task_struct *oom_reaper_list;
mm, oom_reaper: implement OOM victims queuing wake_oom_reaper has allowed only 1 oom victim to be queued. The main reason for that was the simplicity as other solutions would require some way of queuing. The current approach is racy and that was deemed sufficient as the oom_reaper is considered a best effort approach to help with oom handling when the OOM victim cannot terminate in a reasonable time. The race could lead to missing an oom victim which can get stuck out_of_memory wake_oom_reaper cmpxchg // OK oom_reaper oom_reap_task __oom_reap_task oom_victim terminates atomic_inc_not_zero // fail out_of_memory wake_oom_reaper cmpxchg // fails task_to_reap = NULL This race requires 2 OOM invocations in a short time period which is not very likely but certainly not impossible. E.g. the original victim might have not released a lot of memory for some reason. The situation would improve considerably if wake_oom_reaper used a more robust queuing. This is what this patch implements. This means adding oom_reaper_list list_head into task_struct (eat a hole before embeded thread_struct for that purpose) and a oom_reaper_lock spinlock for queuing synchronization. wake_oom_reaper will then add the task on the queue and oom_reaper will dequeue it. Signed-off-by: Michal Hocko <mhocko@suse.com> Cc: Vladimir Davydov <vdavydov@virtuozzo.com> Cc: Andrea Argangeli <andrea@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Hugh Dickins <hughd@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Mel Gorman <mgorman@suse.de> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Rik van Riel <riel@redhat.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:33 +08:00
static DEFINE_SPINLOCK(oom_reaper_lock);
mm, oom: distinguish blockable mode for mmu notifiers There are several blockable mmu notifiers which might sleep in mmu_notifier_invalidate_range_start and that is a problem for the oom_reaper because it needs to guarantee a forward progress so it cannot depend on any sleepable locks. Currently we simply back off and mark an oom victim with blockable mmu notifiers as done after a short sleep. That can result in selecting a new oom victim prematurely because the previous one still hasn't torn its memory down yet. We can do much better though. Even if mmu notifiers use sleepable locks there is no reason to automatically assume those locks are held. Moreover majority of notifiers only care about a portion of the address space and there is absolutely zero reason to fail when we are unmapping an unrelated range. Many notifiers do really block and wait for HW which is harder to handle and we have to bail out though. This patch handles the low hanging fruit. __mmu_notifier_invalidate_range_start gets a blockable flag and callbacks are not allowed to sleep if the flag is set to false. This is achieved by using trylock instead of the sleepable lock for most callbacks and continue as long as we do not block down the call chain. I think we can improve that even further because there is a common pattern to do a range lookup first and then do something about that. The first part can be done without a sleeping lock in most cases AFAICS. The oom_reaper end then simply retries if there is at least one notifier which couldn't make any progress in !blockable mode. A retry loop is already implemented to wait for the mmap_sem and this is basically the same thing. The simplest way for driver developers to test this code path is to wrap userspace code which uses these notifiers into a memcg and set the hard limit to hit the oom. This can be done e.g. after the test faults in all the mmu notifier managed memory and set the hard limit to something really small. Then we are looking for a proper process tear down. [akpm@linux-foundation.org: coding style fixes] [akpm@linux-foundation.org: minor code simplification] Link: http://lkml.kernel.org/r/20180716115058.5559-1-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Acked-by: Christian König <christian.koenig@amd.com> # AMD notifiers Acked-by: Leon Romanovsky <leonro@mellanox.com> # mlx and umem_odp Reported-by: David Rientjes <rientjes@google.com> Cc: "David (ChunMing) Zhou" <David1.Zhou@amd.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Alex Deucher <alexander.deucher@amd.com> Cc: David Airlie <airlied@linux.ie> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Doug Ledford <dledford@redhat.com> Cc: Jason Gunthorpe <jgg@ziepe.ca> Cc: Mike Marciniszyn <mike.marciniszyn@intel.com> Cc: Dennis Dalessandro <dennis.dalessandro@intel.com> Cc: Sudeep Dutt <sudeep.dutt@intel.com> Cc: Ashutosh Dixit <ashutosh.dixit@intel.com> Cc: Dimitri Sivanich <sivanich@sgi.com> Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com> Cc: Juergen Gross <jgross@suse.com> Cc: "Jérôme Glisse" <jglisse@redhat.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-08-22 12:52:33 +08:00
bool __oom_reap_task_mm(struct mm_struct *mm)
mm, oom: introduce oom reaper This patch (of 5): This is based on the idea from Mel Gorman discussed during LSFMM 2015 and independently brought up by Oleg Nesterov. The OOM killer currently allows to kill only a single task in a good hope that the task will terminate in a reasonable time and frees up its memory. Such a task (oom victim) will get an access to memory reserves via mark_oom_victim to allow a forward progress should there be a need for additional memory during exit path. It has been shown (e.g. by Tetsuo Handa) that it is not that hard to construct workloads which break the core assumption mentioned above and the OOM victim might take unbounded amount of time to exit because it might be blocked in the uninterruptible state waiting for an event (e.g. lock) which is blocked by another task looping in the page allocator. This patch reduces the probability of such a lockup by introducing a specialized kernel thread (oom_reaper) which tries to reclaim additional memory by preemptively reaping the anonymous or swapped out memory owned by the oom victim under an assumption that such a memory won't be needed when its owner is killed and kicked from the userspace anyway. There is one notable exception to this, though, if the OOM victim was in the process of coredumping the result would be incomplete. This is considered a reasonable constrain because the overall system health is more important than debugability of a particular application. A kernel thread has been chosen because we need a reliable way of invocation so workqueue context is not appropriate because all the workers might be busy (e.g. allocating memory). Kswapd which sounds like another good fit is not appropriate as well because it might get blocked on locks during reclaim as well. oom_reaper has to take mmap_sem on the target task for reading so the solution is not 100% because the semaphore might be held or blocked for write but the probability is reduced considerably wrt. basically any lock blocking forward progress as described above. In order to prevent from blocking on the lock without any forward progress we are using only a trylock and retry 10 times with a short sleep in between. Users of mmap_sem which need it for write should be carefully reviewed to use _killable waiting as much as possible and reduce allocations requests done with the lock held to absolute minimum to reduce the risk even further. The API between oom killer and oom reaper is quite trivial. wake_oom_reaper updates mm_to_reap with cmpxchg to guarantee only NULL->mm transition and oom_reaper clear this atomically once it is done with the work. This means that only a single mm_struct can be reaped at the time. As the operation is potentially disruptive we are trying to limit it to the ncessary minimum and the reaper blocks any updates while it operates on an mm. mm_struct is pinned by mm_count to allow parallel exit_mmap and a race is detected by atomic_inc_not_zero(mm_users). Signed-off-by: Michal Hocko <mhocko@suse.com> Suggested-by: Oleg Nesterov <oleg@redhat.com> Suggested-by: Mel Gorman <mgorman@suse.de> Acked-by: Mel Gorman <mgorman@suse.de> Acked-by: David Rientjes <rientjes@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Hugh Dickins <hughd@google.com> Cc: Andrea Argangeli <andrea@kernel.org> Cc: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:24 +08:00
{
struct vm_area_struct *vma;
mm, oom: distinguish blockable mode for mmu notifiers There are several blockable mmu notifiers which might sleep in mmu_notifier_invalidate_range_start and that is a problem for the oom_reaper because it needs to guarantee a forward progress so it cannot depend on any sleepable locks. Currently we simply back off and mark an oom victim with blockable mmu notifiers as done after a short sleep. That can result in selecting a new oom victim prematurely because the previous one still hasn't torn its memory down yet. We can do much better though. Even if mmu notifiers use sleepable locks there is no reason to automatically assume those locks are held. Moreover majority of notifiers only care about a portion of the address space and there is absolutely zero reason to fail when we are unmapping an unrelated range. Many notifiers do really block and wait for HW which is harder to handle and we have to bail out though. This patch handles the low hanging fruit. __mmu_notifier_invalidate_range_start gets a blockable flag and callbacks are not allowed to sleep if the flag is set to false. This is achieved by using trylock instead of the sleepable lock for most callbacks and continue as long as we do not block down the call chain. I think we can improve that even further because there is a common pattern to do a range lookup first and then do something about that. The first part can be done without a sleeping lock in most cases AFAICS. The oom_reaper end then simply retries if there is at least one notifier which couldn't make any progress in !blockable mode. A retry loop is already implemented to wait for the mmap_sem and this is basically the same thing. The simplest way for driver developers to test this code path is to wrap userspace code which uses these notifiers into a memcg and set the hard limit to hit the oom. This can be done e.g. after the test faults in all the mmu notifier managed memory and set the hard limit to something really small. Then we are looking for a proper process tear down. [akpm@linux-foundation.org: coding style fixes] [akpm@linux-foundation.org: minor code simplification] Link: http://lkml.kernel.org/r/20180716115058.5559-1-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Acked-by: Christian König <christian.koenig@amd.com> # AMD notifiers Acked-by: Leon Romanovsky <leonro@mellanox.com> # mlx and umem_odp Reported-by: David Rientjes <rientjes@google.com> Cc: "David (ChunMing) Zhou" <David1.Zhou@amd.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Alex Deucher <alexander.deucher@amd.com> Cc: David Airlie <airlied@linux.ie> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Doug Ledford <dledford@redhat.com> Cc: Jason Gunthorpe <jgg@ziepe.ca> Cc: Mike Marciniszyn <mike.marciniszyn@intel.com> Cc: Dennis Dalessandro <dennis.dalessandro@intel.com> Cc: Sudeep Dutt <sudeep.dutt@intel.com> Cc: Ashutosh Dixit <ashutosh.dixit@intel.com> Cc: Dimitri Sivanich <sivanich@sgi.com> Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com> Cc: Juergen Gross <jgross@suse.com> Cc: "Jérôme Glisse" <jglisse@redhat.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-08-22 12:52:33 +08:00
bool ret = true;
mm, oom: fix concurrent munlock and oom reaper unmap, v3 Since exit_mmap() is done without the protection of mm->mmap_sem, it is possible for the oom reaper to concurrently operate on an mm until MMF_OOM_SKIP is set. This allows munlock_vma_pages_all() to concurrently run while the oom reaper is operating on a vma. Since munlock_vma_pages_range() depends on clearing VM_LOCKED from vm_flags before actually doing the munlock to determine if any other vmas are locking the same memory, the check for VM_LOCKED in the oom reaper is racy. This is especially noticeable on architectures such as powerpc where clearing a huge pmd requires serialize_against_pte_lookup(). If the pmd is zapped by the oom reaper during follow_page_mask() after the check for pmd_none() is bypassed, this ends up deferencing a NULL ptl or a kernel oops. Fix this by manually freeing all possible memory from the mm before doing the munlock and then setting MMF_OOM_SKIP. The oom reaper can not run on the mm anymore so the munlock is safe to do in exit_mmap(). It also matches the logic that the oom reaper currently uses for determining when to set MMF_OOM_SKIP itself, so there's no new risk of excessive oom killing. This issue fixes CVE-2018-1000200. Link: http://lkml.kernel.org/r/alpine.DEB.2.21.1804241526320.238665@chino.kir.corp.google.com Fixes: 212925802454 ("mm: oom: let oom_reap_task and exit_mmap run concurrently") Signed-off-by: David Rientjes <rientjes@google.com> Suggested-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: <stable@vger.kernel.org> [4.14+] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-05-12 07:02:04 +08:00
/*
* Tell all users of get_user/copy_from_user etc... that the content
* is no longer stable. No barriers really needed because unmapping
* should imply barriers already and the reader would hit a page fault
* if it stumbled over a reaped memory.
*/
set_bit(MMF_UNSTABLE, &mm->flags);
for (vma = mm->mmap ; vma; vma = vma->vm_next) {
mm: introduce MADV_COLD Patch series "Introduce MADV_COLD and MADV_PAGEOUT", v7. - Background The Android terminology used for forking a new process and starting an app from scratch is a cold start, while resuming an existing app is a hot start. While we continually try to improve the performance of cold starts, hot starts will always be significantly less power hungry as well as faster so we are trying to make hot start more likely than cold start. To increase hot start, Android userspace manages the order that apps should be killed in a process called ActivityManagerService. ActivityManagerService tracks every Android app or service that the user could be interacting with at any time and translates that into a ranked list for lmkd(low memory killer daemon). They are likely to be killed by lmkd if the system has to reclaim memory. In that sense they are similar to entries in any other cache. Those apps are kept alive for opportunistic performance improvements but those performance improvements will vary based on the memory requirements of individual workloads. - Problem Naturally, cached apps were dominant consumers of memory on the system. However, they were not significant consumers of swap even though they are good candidate for swap. Under investigation, swapping out only begins once the low zone watermark is hit and kswapd wakes up, but the overall allocation rate in the system might trip lmkd thresholds and cause a cached process to be killed(we measured performance swapping out vs. zapping the memory by killing a process. Unsurprisingly, zapping is 10x times faster even though we use zram which is much faster than real storage) so kill from lmkd will often satisfy the high zone watermark, resulting in very few pages actually being moved to swap. - Approach The approach we chose was to use a new interface to allow userspace to proactively reclaim entire processes by leveraging platform information. This allowed us to bypass the inaccuracy of the kernel’s LRUs for pages that are known to be cold from userspace and to avoid races with lmkd by reclaiming apps as soon as they entered the cached state. Additionally, it could provide many chances for platform to use much information to optimize memory efficiency. To achieve the goal, the patchset introduce two new options for madvise. One is MADV_COLD which will deactivate activated pages and the other is MADV_PAGEOUT which will reclaim private pages instantly. These new options complement MADV_DONTNEED and MADV_FREE by adding non-destructive ways to gain some free memory space. MADV_PAGEOUT is similar to MADV_DONTNEED in a way that it hints the kernel that memory region is not currently needed and should be reclaimed immediately; MADV_COLD is similar to MADV_FREE in a way that it hints the kernel that memory region is not currently needed and should be reclaimed when memory pressure rises. This patch (of 5): When a process expects no accesses to a certain memory range, it could give a hint to kernel that the pages can be reclaimed when memory pressure happens but data should be preserved for future use. This could reduce workingset eviction so it ends up increasing performance. This patch introduces the new MADV_COLD hint to madvise(2) syscall. MADV_COLD can be used by a process to mark a memory range as not expected to be used in the near future. The hint can help kernel in deciding which pages to evict early during memory pressure. It works for every LRU pages like MADV_[DONTNEED|FREE]. IOW, It moves active file page -> inactive file LRU active anon page -> inacdtive anon LRU Unlike MADV_FREE, it doesn't move active anonymous pages to inactive file LRU's head because MADV_COLD is a little bit different symantic. MADV_FREE means it's okay to discard when the memory pressure because the content of the page is *garbage* so freeing such pages is almost zero overhead since we don't need to swap out and access afterward causes just minor fault. Thus, it would make sense to put those freeable pages in inactive file LRU to compete other used-once pages. It makes sense for implmentaion point of view, too because it's not swapbacked memory any longer until it would be re-dirtied. Even, it could give a bonus to make them be reclaimed on swapless system. However, MADV_COLD doesn't mean garbage so reclaiming them requires swap-out/in in the end so it's bigger cost. Since we have designed VM LRU aging based on cost-model, anonymous cold pages would be better to position inactive anon's LRU list, not file LRU. Furthermore, it would help to avoid unnecessary scanning if system doesn't have a swap device. Let's start simpler way without adding complexity at this moment. However, keep in mind, too that it's a caveat that workloads with a lot of pages cache are likely to ignore MADV_COLD on anonymous memory because we rarely age anonymous LRU lists. * man-page material MADV_COLD (since Linux x.x) Pages in the specified regions will be treated as less-recently-accessed compared to pages in the system with similar access frequencies. In contrast to MADV_FREE, the contents of the region are preserved regardless of subsequent writes to pages. MADV_COLD cannot be applied to locked pages, Huge TLB pages, or VM_PFNMAP pages. [akpm@linux-foundation.org: resolve conflicts with hmm.git] Link: http://lkml.kernel.org/r/20190726023435.214162-2-minchan@kernel.org Signed-off-by: Minchan Kim <minchan@kernel.org> Reported-by: kbuild test robot <lkp@intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: James E.J. Bottomley <James.Bottomley@HansenPartnership.com> Cc: Richard Henderson <rth@twiddle.net> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Chris Zankel <chris@zankel.net> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Daniel Colascione <dancol@google.com> Cc: Dave Hansen <dave.hansen@intel.com> Cc: Hillf Danton <hdanton@sina.com> Cc: Joel Fernandes (Google) <joel@joelfernandes.org> Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc: Oleksandr Natalenko <oleksandr@redhat.com> Cc: Shakeel Butt <shakeelb@google.com> Cc: Sonny Rao <sonnyrao@google.com> Cc: Suren Baghdasaryan <surenb@google.com> Cc: Tim Murray <timmurray@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-09-26 07:49:08 +08:00
if (!can_madv_lru_vma(vma))
mm, oom: fix concurrent munlock and oom reaper unmap, v3 Since exit_mmap() is done without the protection of mm->mmap_sem, it is possible for the oom reaper to concurrently operate on an mm until MMF_OOM_SKIP is set. This allows munlock_vma_pages_all() to concurrently run while the oom reaper is operating on a vma. Since munlock_vma_pages_range() depends on clearing VM_LOCKED from vm_flags before actually doing the munlock to determine if any other vmas are locking the same memory, the check for VM_LOCKED in the oom reaper is racy. This is especially noticeable on architectures such as powerpc where clearing a huge pmd requires serialize_against_pte_lookup(). If the pmd is zapped by the oom reaper during follow_page_mask() after the check for pmd_none() is bypassed, this ends up deferencing a NULL ptl or a kernel oops. Fix this by manually freeing all possible memory from the mm before doing the munlock and then setting MMF_OOM_SKIP. The oom reaper can not run on the mm anymore so the munlock is safe to do in exit_mmap(). It also matches the logic that the oom reaper currently uses for determining when to set MMF_OOM_SKIP itself, so there's no new risk of excessive oom killing. This issue fixes CVE-2018-1000200. Link: http://lkml.kernel.org/r/alpine.DEB.2.21.1804241526320.238665@chino.kir.corp.google.com Fixes: 212925802454 ("mm: oom: let oom_reap_task and exit_mmap run concurrently") Signed-off-by: David Rientjes <rientjes@google.com> Suggested-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: <stable@vger.kernel.org> [4.14+] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-05-12 07:02:04 +08:00
continue;
/*
* Only anonymous pages have a good chance to be dropped
* without additional steps which we cannot afford as we
* are OOM already.
*
* We do not even care about fs backed pages because all
* which are reclaimable have already been reclaimed and
* we do not want to block exit_mmap by keeping mm ref
* count elevated without a good reason.
*/
if (vma_is_anonymous(vma) || !(vma->vm_flags & VM_SHARED)) {
struct mmu_notifier_range range;
mm, oom: fix concurrent munlock and oom reaper unmap, v3 Since exit_mmap() is done without the protection of mm->mmap_sem, it is possible for the oom reaper to concurrently operate on an mm until MMF_OOM_SKIP is set. This allows munlock_vma_pages_all() to concurrently run while the oom reaper is operating on a vma. Since munlock_vma_pages_range() depends on clearing VM_LOCKED from vm_flags before actually doing the munlock to determine if any other vmas are locking the same memory, the check for VM_LOCKED in the oom reaper is racy. This is especially noticeable on architectures such as powerpc where clearing a huge pmd requires serialize_against_pte_lookup(). If the pmd is zapped by the oom reaper during follow_page_mask() after the check for pmd_none() is bypassed, this ends up deferencing a NULL ptl or a kernel oops. Fix this by manually freeing all possible memory from the mm before doing the munlock and then setting MMF_OOM_SKIP. The oom reaper can not run on the mm anymore so the munlock is safe to do in exit_mmap(). It also matches the logic that the oom reaper currently uses for determining when to set MMF_OOM_SKIP itself, so there's no new risk of excessive oom killing. This issue fixes CVE-2018-1000200. Link: http://lkml.kernel.org/r/alpine.DEB.2.21.1804241526320.238665@chino.kir.corp.google.com Fixes: 212925802454 ("mm: oom: let oom_reap_task and exit_mmap run concurrently") Signed-off-by: David Rientjes <rientjes@google.com> Suggested-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: <stable@vger.kernel.org> [4.14+] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-05-12 07:02:04 +08:00
struct mmu_gather tlb;
mm/mmu_notifier: contextual information for event triggering invalidation CPU page table update can happens for many reasons, not only as a result of a syscall (munmap(), mprotect(), mremap(), madvise(), ...) but also as a result of kernel activities (memory compression, reclaim, migration, ...). Users of mmu notifier API track changes to the CPU page table and take specific action for them. While current API only provide range of virtual address affected by the change, not why the changes is happening. This patchset do the initial mechanical convertion of all the places that calls mmu_notifier_range_init to also provide the default MMU_NOTIFY_UNMAP event as well as the vma if it is know (most invalidation happens against a given vma). Passing down the vma allows the users of mmu notifier to inspect the new vma page protection. The MMU_NOTIFY_UNMAP is always the safe default as users of mmu notifier should assume that every for the range is going away when that event happens. A latter patch do convert mm call path to use a more appropriate events for each call. This is done as 2 patches so that no call site is forgotten especialy as it uses this following coccinelle patch: %<---------------------------------------------------------------------- @@ identifier I1, I2, I3, I4; @@ static inline void mmu_notifier_range_init(struct mmu_notifier_range *I1, +enum mmu_notifier_event event, +unsigned flags, +struct vm_area_struct *vma, struct mm_struct *I2, unsigned long I3, unsigned long I4) { ... } @@ @@ -#define mmu_notifier_range_init(range, mm, start, end) +#define mmu_notifier_range_init(range, event, flags, vma, mm, start, end) @@ expression E1, E3, E4; identifier I1; @@ <... mmu_notifier_range_init(E1, +MMU_NOTIFY_UNMAP, 0, I1, I1->vm_mm, E3, E4) ...> @@ expression E1, E2, E3, E4; identifier FN, VMA; @@ FN(..., struct vm_area_struct *VMA, ...) { <... mmu_notifier_range_init(E1, +MMU_NOTIFY_UNMAP, 0, VMA, E2, E3, E4) ...> } @@ expression E1, E2, E3, E4; identifier FN, VMA; @@ FN(...) { struct vm_area_struct *VMA; <... mmu_notifier_range_init(E1, +MMU_NOTIFY_UNMAP, 0, VMA, E2, E3, E4) ...> } @@ expression E1, E2, E3, E4; identifier FN; @@ FN(...) { <... mmu_notifier_range_init(E1, +MMU_NOTIFY_UNMAP, 0, NULL, E2, E3, E4) ...> } ---------------------------------------------------------------------->% Applied with: spatch --all-includes --sp-file mmu-notifier.spatch fs/proc/task_mmu.c --in-place spatch --sp-file mmu-notifier.spatch --dir kernel/events/ --in-place spatch --sp-file mmu-notifier.spatch --dir mm --in-place Link: http://lkml.kernel.org/r/20190326164747.24405-6-jglisse@redhat.com Signed-off-by: Jérôme Glisse <jglisse@redhat.com> Reviewed-by: Ralph Campbell <rcampbell@nvidia.com> Reviewed-by: Ira Weiny <ira.weiny@intel.com> Cc: Christian König <christian.koenig@amd.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Jan Kara <jack@suse.cz> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Peter Xu <peterx@redhat.com> Cc: Felix Kuehling <Felix.Kuehling@amd.com> Cc: Jason Gunthorpe <jgg@mellanox.com> Cc: Ross Zwisler <zwisler@kernel.org> Cc: Dan Williams <dan.j.williams@intel.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Radim Krcmar <rkrcmar@redhat.com> Cc: Michal Hocko <mhocko@kernel.org> Cc: Christian Koenig <christian.koenig@amd.com> Cc: John Hubbard <jhubbard@nvidia.com> Cc: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-05-14 08:20:49 +08:00
mmu_notifier_range_init(&range, MMU_NOTIFY_UNMAP, 0,
vma, mm, vma->vm_start,
vma->vm_end);
tlb_gather_mmu(&tlb, mm, range.start, range.end);
if (mmu_notifier_invalidate_range_start_nonblock(&range)) {
tlb_finish_mmu(&tlb);
mm, oom: distinguish blockable mode for mmu notifiers There are several blockable mmu notifiers which might sleep in mmu_notifier_invalidate_range_start and that is a problem for the oom_reaper because it needs to guarantee a forward progress so it cannot depend on any sleepable locks. Currently we simply back off and mark an oom victim with blockable mmu notifiers as done after a short sleep. That can result in selecting a new oom victim prematurely because the previous one still hasn't torn its memory down yet. We can do much better though. Even if mmu notifiers use sleepable locks there is no reason to automatically assume those locks are held. Moreover majority of notifiers only care about a portion of the address space and there is absolutely zero reason to fail when we are unmapping an unrelated range. Many notifiers do really block and wait for HW which is harder to handle and we have to bail out though. This patch handles the low hanging fruit. __mmu_notifier_invalidate_range_start gets a blockable flag and callbacks are not allowed to sleep if the flag is set to false. This is achieved by using trylock instead of the sleepable lock for most callbacks and continue as long as we do not block down the call chain. I think we can improve that even further because there is a common pattern to do a range lookup first and then do something about that. The first part can be done without a sleeping lock in most cases AFAICS. The oom_reaper end then simply retries if there is at least one notifier which couldn't make any progress in !blockable mode. A retry loop is already implemented to wait for the mmap_sem and this is basically the same thing. The simplest way for driver developers to test this code path is to wrap userspace code which uses these notifiers into a memcg and set the hard limit to hit the oom. This can be done e.g. after the test faults in all the mmu notifier managed memory and set the hard limit to something really small. Then we are looking for a proper process tear down. [akpm@linux-foundation.org: coding style fixes] [akpm@linux-foundation.org: minor code simplification] Link: http://lkml.kernel.org/r/20180716115058.5559-1-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Acked-by: Christian König <christian.koenig@amd.com> # AMD notifiers Acked-by: Leon Romanovsky <leonro@mellanox.com> # mlx and umem_odp Reported-by: David Rientjes <rientjes@google.com> Cc: "David (ChunMing) Zhou" <David1.Zhou@amd.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Alex Deucher <alexander.deucher@amd.com> Cc: David Airlie <airlied@linux.ie> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Doug Ledford <dledford@redhat.com> Cc: Jason Gunthorpe <jgg@ziepe.ca> Cc: Mike Marciniszyn <mike.marciniszyn@intel.com> Cc: Dennis Dalessandro <dennis.dalessandro@intel.com> Cc: Sudeep Dutt <sudeep.dutt@intel.com> Cc: Ashutosh Dixit <ashutosh.dixit@intel.com> Cc: Dimitri Sivanich <sivanich@sgi.com> Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com> Cc: Juergen Gross <jgross@suse.com> Cc: "Jérôme Glisse" <jglisse@redhat.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-08-22 12:52:33 +08:00
ret = false;
continue;
}
unmap_page_range(&tlb, vma, range.start, range.end, NULL);
mmu_notifier_invalidate_range_end(&range);
tlb_finish_mmu(&tlb);
mm, oom: fix concurrent munlock and oom reaper unmap, v3 Since exit_mmap() is done without the protection of mm->mmap_sem, it is possible for the oom reaper to concurrently operate on an mm until MMF_OOM_SKIP is set. This allows munlock_vma_pages_all() to concurrently run while the oom reaper is operating on a vma. Since munlock_vma_pages_range() depends on clearing VM_LOCKED from vm_flags before actually doing the munlock to determine if any other vmas are locking the same memory, the check for VM_LOCKED in the oom reaper is racy. This is especially noticeable on architectures such as powerpc where clearing a huge pmd requires serialize_against_pte_lookup(). If the pmd is zapped by the oom reaper during follow_page_mask() after the check for pmd_none() is bypassed, this ends up deferencing a NULL ptl or a kernel oops. Fix this by manually freeing all possible memory from the mm before doing the munlock and then setting MMF_OOM_SKIP. The oom reaper can not run on the mm anymore so the munlock is safe to do in exit_mmap(). It also matches the logic that the oom reaper currently uses for determining when to set MMF_OOM_SKIP itself, so there's no new risk of excessive oom killing. This issue fixes CVE-2018-1000200. Link: http://lkml.kernel.org/r/alpine.DEB.2.21.1804241526320.238665@chino.kir.corp.google.com Fixes: 212925802454 ("mm: oom: let oom_reap_task and exit_mmap run concurrently") Signed-off-by: David Rientjes <rientjes@google.com> Suggested-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: <stable@vger.kernel.org> [4.14+] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-05-12 07:02:04 +08:00
}
}
mm, oom: distinguish blockable mode for mmu notifiers There are several blockable mmu notifiers which might sleep in mmu_notifier_invalidate_range_start and that is a problem for the oom_reaper because it needs to guarantee a forward progress so it cannot depend on any sleepable locks. Currently we simply back off and mark an oom victim with blockable mmu notifiers as done after a short sleep. That can result in selecting a new oom victim prematurely because the previous one still hasn't torn its memory down yet. We can do much better though. Even if mmu notifiers use sleepable locks there is no reason to automatically assume those locks are held. Moreover majority of notifiers only care about a portion of the address space and there is absolutely zero reason to fail when we are unmapping an unrelated range. Many notifiers do really block and wait for HW which is harder to handle and we have to bail out though. This patch handles the low hanging fruit. __mmu_notifier_invalidate_range_start gets a blockable flag and callbacks are not allowed to sleep if the flag is set to false. This is achieved by using trylock instead of the sleepable lock for most callbacks and continue as long as we do not block down the call chain. I think we can improve that even further because there is a common pattern to do a range lookup first and then do something about that. The first part can be done without a sleeping lock in most cases AFAICS. The oom_reaper end then simply retries if there is at least one notifier which couldn't make any progress in !blockable mode. A retry loop is already implemented to wait for the mmap_sem and this is basically the same thing. The simplest way for driver developers to test this code path is to wrap userspace code which uses these notifiers into a memcg and set the hard limit to hit the oom. This can be done e.g. after the test faults in all the mmu notifier managed memory and set the hard limit to something really small. Then we are looking for a proper process tear down. [akpm@linux-foundation.org: coding style fixes] [akpm@linux-foundation.org: minor code simplification] Link: http://lkml.kernel.org/r/20180716115058.5559-1-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Acked-by: Christian König <christian.koenig@amd.com> # AMD notifiers Acked-by: Leon Romanovsky <leonro@mellanox.com> # mlx and umem_odp Reported-by: David Rientjes <rientjes@google.com> Cc: "David (ChunMing) Zhou" <David1.Zhou@amd.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Alex Deucher <alexander.deucher@amd.com> Cc: David Airlie <airlied@linux.ie> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Doug Ledford <dledford@redhat.com> Cc: Jason Gunthorpe <jgg@ziepe.ca> Cc: Mike Marciniszyn <mike.marciniszyn@intel.com> Cc: Dennis Dalessandro <dennis.dalessandro@intel.com> Cc: Sudeep Dutt <sudeep.dutt@intel.com> Cc: Ashutosh Dixit <ashutosh.dixit@intel.com> Cc: Dimitri Sivanich <sivanich@sgi.com> Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com> Cc: Juergen Gross <jgross@suse.com> Cc: "Jérôme Glisse" <jglisse@redhat.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-08-22 12:52:33 +08:00
return ret;
mm, oom: fix concurrent munlock and oom reaper unmap, v3 Since exit_mmap() is done without the protection of mm->mmap_sem, it is possible for the oom reaper to concurrently operate on an mm until MMF_OOM_SKIP is set. This allows munlock_vma_pages_all() to concurrently run while the oom reaper is operating on a vma. Since munlock_vma_pages_range() depends on clearing VM_LOCKED from vm_flags before actually doing the munlock to determine if any other vmas are locking the same memory, the check for VM_LOCKED in the oom reaper is racy. This is especially noticeable on architectures such as powerpc where clearing a huge pmd requires serialize_against_pte_lookup(). If the pmd is zapped by the oom reaper during follow_page_mask() after the check for pmd_none() is bypassed, this ends up deferencing a NULL ptl or a kernel oops. Fix this by manually freeing all possible memory from the mm before doing the munlock and then setting MMF_OOM_SKIP. The oom reaper can not run on the mm anymore so the munlock is safe to do in exit_mmap(). It also matches the logic that the oom reaper currently uses for determining when to set MMF_OOM_SKIP itself, so there's no new risk of excessive oom killing. This issue fixes CVE-2018-1000200. Link: http://lkml.kernel.org/r/alpine.DEB.2.21.1804241526320.238665@chino.kir.corp.google.com Fixes: 212925802454 ("mm: oom: let oom_reap_task and exit_mmap run concurrently") Signed-off-by: David Rientjes <rientjes@google.com> Suggested-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: <stable@vger.kernel.org> [4.14+] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-05-12 07:02:04 +08:00
}
/*
* Reaps the address space of the give task.
*
* Returns true on success and false if none or part of the address space
* has been reclaimed and the caller should retry later.
*/
mm, oom: fix concurrent munlock and oom reaper unmap, v3 Since exit_mmap() is done without the protection of mm->mmap_sem, it is possible for the oom reaper to concurrently operate on an mm until MMF_OOM_SKIP is set. This allows munlock_vma_pages_all() to concurrently run while the oom reaper is operating on a vma. Since munlock_vma_pages_range() depends on clearing VM_LOCKED from vm_flags before actually doing the munlock to determine if any other vmas are locking the same memory, the check for VM_LOCKED in the oom reaper is racy. This is especially noticeable on architectures such as powerpc where clearing a huge pmd requires serialize_against_pte_lookup(). If the pmd is zapped by the oom reaper during follow_page_mask() after the check for pmd_none() is bypassed, this ends up deferencing a NULL ptl or a kernel oops. Fix this by manually freeing all possible memory from the mm before doing the munlock and then setting MMF_OOM_SKIP. The oom reaper can not run on the mm anymore so the munlock is safe to do in exit_mmap(). It also matches the logic that the oom reaper currently uses for determining when to set MMF_OOM_SKIP itself, so there's no new risk of excessive oom killing. This issue fixes CVE-2018-1000200. Link: http://lkml.kernel.org/r/alpine.DEB.2.21.1804241526320.238665@chino.kir.corp.google.com Fixes: 212925802454 ("mm: oom: let oom_reap_task and exit_mmap run concurrently") Signed-off-by: David Rientjes <rientjes@google.com> Suggested-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: <stable@vger.kernel.org> [4.14+] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-05-12 07:02:04 +08:00
static bool oom_reap_task_mm(struct task_struct *tsk, struct mm_struct *mm)
{
mm, oom: introduce oom reaper This patch (of 5): This is based on the idea from Mel Gorman discussed during LSFMM 2015 and independently brought up by Oleg Nesterov. The OOM killer currently allows to kill only a single task in a good hope that the task will terminate in a reasonable time and frees up its memory. Such a task (oom victim) will get an access to memory reserves via mark_oom_victim to allow a forward progress should there be a need for additional memory during exit path. It has been shown (e.g. by Tetsuo Handa) that it is not that hard to construct workloads which break the core assumption mentioned above and the OOM victim might take unbounded amount of time to exit because it might be blocked in the uninterruptible state waiting for an event (e.g. lock) which is blocked by another task looping in the page allocator. This patch reduces the probability of such a lockup by introducing a specialized kernel thread (oom_reaper) which tries to reclaim additional memory by preemptively reaping the anonymous or swapped out memory owned by the oom victim under an assumption that such a memory won't be needed when its owner is killed and kicked from the userspace anyway. There is one notable exception to this, though, if the OOM victim was in the process of coredumping the result would be incomplete. This is considered a reasonable constrain because the overall system health is more important than debugability of a particular application. A kernel thread has been chosen because we need a reliable way of invocation so workqueue context is not appropriate because all the workers might be busy (e.g. allocating memory). Kswapd which sounds like another good fit is not appropriate as well because it might get blocked on locks during reclaim as well. oom_reaper has to take mmap_sem on the target task for reading so the solution is not 100% because the semaphore might be held or blocked for write but the probability is reduced considerably wrt. basically any lock blocking forward progress as described above. In order to prevent from blocking on the lock without any forward progress we are using only a trylock and retry 10 times with a short sleep in between. Users of mmap_sem which need it for write should be carefully reviewed to use _killable waiting as much as possible and reduce allocations requests done with the lock held to absolute minimum to reduce the risk even further. The API between oom killer and oom reaper is quite trivial. wake_oom_reaper updates mm_to_reap with cmpxchg to guarantee only NULL->mm transition and oom_reaper clear this atomically once it is done with the work. This means that only a single mm_struct can be reaped at the time. As the operation is potentially disruptive we are trying to limit it to the ncessary minimum and the reaper blocks any updates while it operates on an mm. mm_struct is pinned by mm_count to allow parallel exit_mmap and a race is detected by atomic_inc_not_zero(mm_users). Signed-off-by: Michal Hocko <mhocko@suse.com> Suggested-by: Oleg Nesterov <oleg@redhat.com> Suggested-by: Mel Gorman <mgorman@suse.de> Acked-by: Mel Gorman <mgorman@suse.de> Acked-by: David Rientjes <rientjes@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Hugh Dickins <hughd@google.com> Cc: Andrea Argangeli <andrea@kernel.org> Cc: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:24 +08:00
bool ret = true;
mmap locking API: use coccinelle to convert mmap_sem rwsem call sites This change converts the existing mmap_sem rwsem calls to use the new mmap locking API instead. The change is generated using coccinelle with the following rule: // spatch --sp-file mmap_lock_api.cocci --in-place --include-headers --dir . @@ expression mm; @@ ( -init_rwsem +mmap_init_lock | -down_write +mmap_write_lock | -down_write_killable +mmap_write_lock_killable | -down_write_trylock +mmap_write_trylock | -up_write +mmap_write_unlock | -downgrade_write +mmap_write_downgrade | -down_read +mmap_read_lock | -down_read_killable +mmap_read_lock_killable | -down_read_trylock +mmap_read_trylock | -up_read +mmap_read_unlock ) -(&mm->mmap_sem) +(mm) Signed-off-by: Michel Lespinasse <walken@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Reviewed-by: Daniel Jordan <daniel.m.jordan@oracle.com> Reviewed-by: Laurent Dufour <ldufour@linux.ibm.com> Reviewed-by: Vlastimil Babka <vbabka@suse.cz> Cc: Davidlohr Bueso <dbueso@suse.de> Cc: David Rientjes <rientjes@google.com> Cc: Hugh Dickins <hughd@google.com> Cc: Jason Gunthorpe <jgg@ziepe.ca> Cc: Jerome Glisse <jglisse@redhat.com> Cc: John Hubbard <jhubbard@nvidia.com> Cc: Liam Howlett <Liam.Howlett@oracle.com> Cc: Matthew Wilcox <willy@infradead.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Ying Han <yinghan@google.com> Link: http://lkml.kernel.org/r/20200520052908.204642-5-walken@google.com Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-06-09 12:33:25 +08:00
if (!mmap_read_trylock(mm)) {
mm/oom_kill.c: add tracepoints for oom reaper-related events During the debugging of the problem described in https://lkml.org/lkml/2017/5/17/542 and fixed by Tetsuo Handa in https://lkml.org/lkml/2017/5/19/383 , I've found that the existing debug output is not really useful to understand issues related to the oom reaper. So, I assume, that adding some tracepoints might help with debugging of similar issues. Trace the following events: 1) a process is marked as an oom victim, 2) a process is added to the oom reaper list, 3) the oom reaper starts reaping process's mm, 4) the oom reaper finished reaping, 5) the oom reaper skips reaping. How it works in practice? Below is an example which show how the problem mentioned above can be found: one process is added twice to the oom_reaper list: $ cd /sys/kernel/debug/tracing $ echo "oom:mark_victim" > set_event $ echo "oom:wake_reaper" >> set_event $ echo "oom:skip_task_reaping" >> set_event $ echo "oom:start_task_reaping" >> set_event $ echo "oom:finish_task_reaping" >> set_event $ cat trace_pipe allocate-502 [001] .... 91.836405: mark_victim: pid=502 allocate-502 [001] .N.. 91.837356: wake_reaper: pid=502 allocate-502 [000] .N.. 91.871149: wake_reaper: pid=502 oom_reaper-23 [000] .... 91.871177: start_task_reaping: pid=502 oom_reaper-23 [000] .N.. 91.879511: finish_task_reaping: pid=502 oom_reaper-23 [000] .... 91.879580: skip_task_reaping: pid=502 Link: http://lkml.kernel.org/r/20170530185231.GA13412@castle Signed-off-by: Roman Gushchin <guro@fb.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-07-11 06:49:05 +08:00
trace_skip_task_reaping(tsk->pid);
mm, oom: remove oom_lock from oom_reaper oom_reaper used to rely on the oom_lock since e2fe14564d33 ("oom_reaper: close race with exiting task"). We do not really need the lock anymore though. 212925802454 ("mm: oom: let oom_reap_task and exit_mmap run concurrently") has removed serialization with the exit path based on the mm reference count and so we do not really rely on the oom_lock anymore. Tetsuo was arguing that at least MMF_OOM_SKIP should be set under the lock to prevent from races when the page allocator didn't manage to get the freed (reaped) memory in __alloc_pages_may_oom but it sees the flag later on and move on to another victim. Although this is possible in principle let's wait for it to actually happen in real life before we make the locking more complex again. Therefore remove the oom_lock for oom_reaper paths (both exit_mmap and oom_reap_task_mm). The reaper serializes with exit_mmap by mmap_sem + MMF_OOM_SKIP flag. There is no synchronization with out_of_memory path now. [mhocko@kernel.org: oom_reap_task_mm should return false when __oom_reap_task_mm did] Link: http://lkml.kernel.org/r/20180724141747.GP28386@dhcp22.suse.cz Link: http://lkml.kernel.org/r/20180719075922.13784-1-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Suggested-by: David Rientjes <rientjes@google.com> Acked-by: David Rientjes <rientjes@google.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-08-22 12:52:37 +08:00
return false;
mm, oom_reaper: skip mm structs with mmu notifiers Andrea has noticed that the oom_reaper doesn't invalidate the range via mmu notifiers (mmu_notifier_invalidate_range_start/end) and that can corrupt the memory of the kvm guest for example. tlb_flush_mmu_tlbonly already invokes mmu notifiers but that is not sufficient as per Andrea: "mmu_notifier_invalidate_range cannot be used in replacement of mmu_notifier_invalidate_range_start/end. For KVM mmu_notifier_invalidate_range is a noop and rightfully so. A MMU notifier implementation has to implement either ->invalidate_range method or the invalidate_range_start/end methods, not both. And if you implement invalidate_range_start/end like KVM is forced to do, calling mmu_notifier_invalidate_range in common code is a noop for KVM. For those MMU notifiers that can get away only implementing ->invalidate_range, the ->invalidate_range is implicitly called by mmu_notifier_invalidate_range_end(). And only those secondary MMUs that share the same pagetable with the primary MMU (like AMD iommuv2) can get away only implementing ->invalidate_range" As the callback is allowed to sleep and the implementation is out of hand of the MM it is safer to simply bail out if there is an mmu notifier registered. In order to not fail too early make the mm_has_notifiers check under the oom_lock and have a little nap before failing to give the current oom victim some more time to exit. [akpm@linux-foundation.org: coding-style fixes] Link: http://lkml.kernel.org/r/20170913113427.2291-1-mhocko@kernel.org Fixes: aac453635549 ("mm, oom: introduce oom reaper") Signed-off-by: Michal Hocko <mhocko@suse.com> Reported-by: Andrea Arcangeli <aarcange@redhat.com> Reviewed-by: Andrea Arcangeli <aarcange@redhat.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-10-04 07:14:50 +08:00
}
mm, oom_reaper: make sure that mmput_async is called only when memory was reaped Tetsuo is worried that mmput_async might still lead to a premature new oom victim selection due to the following race: __oom_reap_task exit_mm find_lock_task_mm atomic_inc(mm->mm_users) # = 2 task_unlock task_lock task->mm = NULL up_read(&mm->mmap_sem) < somebody write locks mmap_sem > task_unlock mmput atomic_dec_and_test # = 1 exit_oom_victim down_read_trylock # failed - no reclaim mmput_async # Takes unpredictable amount of time < new OOM situation > the final __mmput will be executed in the delayed context which might happen far in the future. Such a race is highly unlikely because the write holder of mmap_sem would have to be an external task (all direct holders are already killed or exiting) and it usually have to pin mm_users in order to do anything reasonable. We can, however, make sure that the mmput_async is only called when we do not back off and reap some memory. That would reduce the impact of the delayed __mmput because the real content would be already freed. Pin mm_count to keep it alive after we drop task_lock and before we try to get mmap_sem. If the mmap_sem succeeds we can try to grab mm_users reference and then go on with unmapping the address space. It is not clear whether this race is possible at all but it is better to be more robust and do not pin mm_users unless we are sure we are actually doing some real work during __oom_reap_task. Link: http://lkml.kernel.org/r/1465306987-30297-1-git-send-email-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Reported-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-07-27 06:24:50 +08:00
/*
mm: oom: let oom_reap_task and exit_mmap run concurrently This is purely required because exit_aio() may block and exit_mmap() may never start, if the oom_reap_task cannot start running on a mm with mm_users == 0. At the same time if the OOM reaper doesn't wait at all for the memory of the current OOM candidate to be freed by exit_mmap->unmap_vmas, it would generate a spurious OOM kill. If it wasn't because of the exit_aio or similar blocking functions in the last mmput, it would be enough to change the oom_reap_task() in the case it finds mm_users == 0, to wait for a timeout or to wait for __mmput to set MMF_OOM_SKIP itself, but it's not just exit_mmap the problem here so the concurrency of exit_mmap and oom_reap_task is apparently warranted. It's a non standard runtime, exit_mmap() runs without mmap_sem, and oom_reap_task runs with the mmap_sem for reading as usual (kind of MADV_DONTNEED). The race between the two is solved with a combination of tsk_is_oom_victim() (serialized by task_lock) and MMF_OOM_SKIP (serialized by a dummy down_write/up_write cycle on the same lines of the ksm_exit method). If the oom_reap_task() may be running concurrently during exit_mmap, exit_mmap will wait it to finish in down_write (before taking down mm structures that would make the oom_reap_task fail with use after free). If exit_mmap comes first, oom_reap_task() will skip the mm if MMF_OOM_SKIP is already set and in turn all memory is already freed and furthermore the mm data structures may already have been taken down by free_pgtables. [aarcange@redhat.com: incremental one liner] Link: http://lkml.kernel.org/r/20170726164319.GC29716@redhat.com [rientjes@google.com: remove unused mmput_async] Link: http://lkml.kernel.org/r/alpine.DEB.2.10.1708141733130.50317@chino.kir.corp.google.com [aarcange@redhat.com: microoptimization] Link: http://lkml.kernel.org/r/20170817171240.GB5066@redhat.com Link: http://lkml.kernel.org/r/20170726162912.GA29716@redhat.com Fixes: 26db62f179d1 ("oom: keep mm of the killed task available") Signed-off-by: Andrea Arcangeli <aarcange@redhat.com> Signed-off-by: David Rientjes <rientjes@google.com> Reported-by: David Rientjes <rientjes@google.com> Tested-by: David Rientjes <rientjes@google.com> Reviewed-by: Michal Hocko <mhocko@suse.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Hugh Dickins <hughd@google.com> Cc: "Kirill A. Shutemov" <kirill@shutemov.name> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-09-07 07:25:00 +08:00
* MMF_OOM_SKIP is set by exit_mmap when the OOM reaper can't
* work on the mm anymore. The check for MMF_OOM_SKIP must run
* under mmap_lock for reading because it serializes against the
* mmap_write_lock();mmap_write_unlock() cycle in exit_mmap().
mm, oom_reaper: make sure that mmput_async is called only when memory was reaped Tetsuo is worried that mmput_async might still lead to a premature new oom victim selection due to the following race: __oom_reap_task exit_mm find_lock_task_mm atomic_inc(mm->mm_users) # = 2 task_unlock task_lock task->mm = NULL up_read(&mm->mmap_sem) < somebody write locks mmap_sem > task_unlock mmput atomic_dec_and_test # = 1 exit_oom_victim down_read_trylock # failed - no reclaim mmput_async # Takes unpredictable amount of time < new OOM situation > the final __mmput will be executed in the delayed context which might happen far in the future. Such a race is highly unlikely because the write holder of mmap_sem would have to be an external task (all direct holders are already killed or exiting) and it usually have to pin mm_users in order to do anything reasonable. We can, however, make sure that the mmput_async is only called when we do not back off and reap some memory. That would reduce the impact of the delayed __mmput because the real content would be already freed. Pin mm_count to keep it alive after we drop task_lock and before we try to get mmap_sem. If the mmap_sem succeeds we can try to grab mm_users reference and then go on with unmapping the address space. It is not clear whether this race is possible at all but it is better to be more robust and do not pin mm_users unless we are sure we are actually doing some real work during __oom_reap_task. Link: http://lkml.kernel.org/r/1465306987-30297-1-git-send-email-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Reported-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-07-27 06:24:50 +08:00
*/
mm: oom: let oom_reap_task and exit_mmap run concurrently This is purely required because exit_aio() may block and exit_mmap() may never start, if the oom_reap_task cannot start running on a mm with mm_users == 0. At the same time if the OOM reaper doesn't wait at all for the memory of the current OOM candidate to be freed by exit_mmap->unmap_vmas, it would generate a spurious OOM kill. If it wasn't because of the exit_aio or similar blocking functions in the last mmput, it would be enough to change the oom_reap_task() in the case it finds mm_users == 0, to wait for a timeout or to wait for __mmput to set MMF_OOM_SKIP itself, but it's not just exit_mmap the problem here so the concurrency of exit_mmap and oom_reap_task is apparently warranted. It's a non standard runtime, exit_mmap() runs without mmap_sem, and oom_reap_task runs with the mmap_sem for reading as usual (kind of MADV_DONTNEED). The race between the two is solved with a combination of tsk_is_oom_victim() (serialized by task_lock) and MMF_OOM_SKIP (serialized by a dummy down_write/up_write cycle on the same lines of the ksm_exit method). If the oom_reap_task() may be running concurrently during exit_mmap, exit_mmap will wait it to finish in down_write (before taking down mm structures that would make the oom_reap_task fail with use after free). If exit_mmap comes first, oom_reap_task() will skip the mm if MMF_OOM_SKIP is already set and in turn all memory is already freed and furthermore the mm data structures may already have been taken down by free_pgtables. [aarcange@redhat.com: incremental one liner] Link: http://lkml.kernel.org/r/20170726164319.GC29716@redhat.com [rientjes@google.com: remove unused mmput_async] Link: http://lkml.kernel.org/r/alpine.DEB.2.10.1708141733130.50317@chino.kir.corp.google.com [aarcange@redhat.com: microoptimization] Link: http://lkml.kernel.org/r/20170817171240.GB5066@redhat.com Link: http://lkml.kernel.org/r/20170726162912.GA29716@redhat.com Fixes: 26db62f179d1 ("oom: keep mm of the killed task available") Signed-off-by: Andrea Arcangeli <aarcange@redhat.com> Signed-off-by: David Rientjes <rientjes@google.com> Reported-by: David Rientjes <rientjes@google.com> Tested-by: David Rientjes <rientjes@google.com> Reviewed-by: Michal Hocko <mhocko@suse.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Hugh Dickins <hughd@google.com> Cc: "Kirill A. Shutemov" <kirill@shutemov.name> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-09-07 07:25:00 +08:00
if (test_bit(MMF_OOM_SKIP, &mm->flags)) {
mm/oom_kill.c: add tracepoints for oom reaper-related events During the debugging of the problem described in https://lkml.org/lkml/2017/5/17/542 and fixed by Tetsuo Handa in https://lkml.org/lkml/2017/5/19/383 , I've found that the existing debug output is not really useful to understand issues related to the oom reaper. So, I assume, that adding some tracepoints might help with debugging of similar issues. Trace the following events: 1) a process is marked as an oom victim, 2) a process is added to the oom reaper list, 3) the oom reaper starts reaping process's mm, 4) the oom reaper finished reaping, 5) the oom reaper skips reaping. How it works in practice? Below is an example which show how the problem mentioned above can be found: one process is added twice to the oom_reaper list: $ cd /sys/kernel/debug/tracing $ echo "oom:mark_victim" > set_event $ echo "oom:wake_reaper" >> set_event $ echo "oom:skip_task_reaping" >> set_event $ echo "oom:start_task_reaping" >> set_event $ echo "oom:finish_task_reaping" >> set_event $ cat trace_pipe allocate-502 [001] .... 91.836405: mark_victim: pid=502 allocate-502 [001] .N.. 91.837356: wake_reaper: pid=502 allocate-502 [000] .N.. 91.871149: wake_reaper: pid=502 oom_reaper-23 [000] .... 91.871177: start_task_reaping: pid=502 oom_reaper-23 [000] .N.. 91.879511: finish_task_reaping: pid=502 oom_reaper-23 [000] .... 91.879580: skip_task_reaping: pid=502 Link: http://lkml.kernel.org/r/20170530185231.GA13412@castle Signed-off-by: Roman Gushchin <guro@fb.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-07-11 06:49:05 +08:00
trace_skip_task_reaping(tsk->pid);
goto out_unlock;
mm, oom: introduce oom reaper This patch (of 5): This is based on the idea from Mel Gorman discussed during LSFMM 2015 and independently brought up by Oleg Nesterov. The OOM killer currently allows to kill only a single task in a good hope that the task will terminate in a reasonable time and frees up its memory. Such a task (oom victim) will get an access to memory reserves via mark_oom_victim to allow a forward progress should there be a need for additional memory during exit path. It has been shown (e.g. by Tetsuo Handa) that it is not that hard to construct workloads which break the core assumption mentioned above and the OOM victim might take unbounded amount of time to exit because it might be blocked in the uninterruptible state waiting for an event (e.g. lock) which is blocked by another task looping in the page allocator. This patch reduces the probability of such a lockup by introducing a specialized kernel thread (oom_reaper) which tries to reclaim additional memory by preemptively reaping the anonymous or swapped out memory owned by the oom victim under an assumption that such a memory won't be needed when its owner is killed and kicked from the userspace anyway. There is one notable exception to this, though, if the OOM victim was in the process of coredumping the result would be incomplete. This is considered a reasonable constrain because the overall system health is more important than debugability of a particular application. A kernel thread has been chosen because we need a reliable way of invocation so workqueue context is not appropriate because all the workers might be busy (e.g. allocating memory). Kswapd which sounds like another good fit is not appropriate as well because it might get blocked on locks during reclaim as well. oom_reaper has to take mmap_sem on the target task for reading so the solution is not 100% because the semaphore might be held or blocked for write but the probability is reduced considerably wrt. basically any lock blocking forward progress as described above. In order to prevent from blocking on the lock without any forward progress we are using only a trylock and retry 10 times with a short sleep in between. Users of mmap_sem which need it for write should be carefully reviewed to use _killable waiting as much as possible and reduce allocations requests done with the lock held to absolute minimum to reduce the risk even further. The API between oom killer and oom reaper is quite trivial. wake_oom_reaper updates mm_to_reap with cmpxchg to guarantee only NULL->mm transition and oom_reaper clear this atomically once it is done with the work. This means that only a single mm_struct can be reaped at the time. As the operation is potentially disruptive we are trying to limit it to the ncessary minimum and the reaper blocks any updates while it operates on an mm. mm_struct is pinned by mm_count to allow parallel exit_mmap and a race is detected by atomic_inc_not_zero(mm_users). Signed-off-by: Michal Hocko <mhocko@suse.com> Suggested-by: Oleg Nesterov <oleg@redhat.com> Suggested-by: Mel Gorman <mgorman@suse.de> Acked-by: Mel Gorman <mgorman@suse.de> Acked-by: David Rientjes <rientjes@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Hugh Dickins <hughd@google.com> Cc: Andrea Argangeli <andrea@kernel.org> Cc: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:24 +08:00
}
mm/oom_kill.c: add tracepoints for oom reaper-related events During the debugging of the problem described in https://lkml.org/lkml/2017/5/17/542 and fixed by Tetsuo Handa in https://lkml.org/lkml/2017/5/19/383 , I've found that the existing debug output is not really useful to understand issues related to the oom reaper. So, I assume, that adding some tracepoints might help with debugging of similar issues. Trace the following events: 1) a process is marked as an oom victim, 2) a process is added to the oom reaper list, 3) the oom reaper starts reaping process's mm, 4) the oom reaper finished reaping, 5) the oom reaper skips reaping. How it works in practice? Below is an example which show how the problem mentioned above can be found: one process is added twice to the oom_reaper list: $ cd /sys/kernel/debug/tracing $ echo "oom:mark_victim" > set_event $ echo "oom:wake_reaper" >> set_event $ echo "oom:skip_task_reaping" >> set_event $ echo "oom:start_task_reaping" >> set_event $ echo "oom:finish_task_reaping" >> set_event $ cat trace_pipe allocate-502 [001] .... 91.836405: mark_victim: pid=502 allocate-502 [001] .N.. 91.837356: wake_reaper: pid=502 allocate-502 [000] .N.. 91.871149: wake_reaper: pid=502 oom_reaper-23 [000] .... 91.871177: start_task_reaping: pid=502 oom_reaper-23 [000] .N.. 91.879511: finish_task_reaping: pid=502 oom_reaper-23 [000] .... 91.879580: skip_task_reaping: pid=502 Link: http://lkml.kernel.org/r/20170530185231.GA13412@castle Signed-off-by: Roman Gushchin <guro@fb.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-07-11 06:49:05 +08:00
trace_start_task_reaping(tsk->pid);
mm, oom: distinguish blockable mode for mmu notifiers There are several blockable mmu notifiers which might sleep in mmu_notifier_invalidate_range_start and that is a problem for the oom_reaper because it needs to guarantee a forward progress so it cannot depend on any sleepable locks. Currently we simply back off and mark an oom victim with blockable mmu notifiers as done after a short sleep. That can result in selecting a new oom victim prematurely because the previous one still hasn't torn its memory down yet. We can do much better though. Even if mmu notifiers use sleepable locks there is no reason to automatically assume those locks are held. Moreover majority of notifiers only care about a portion of the address space and there is absolutely zero reason to fail when we are unmapping an unrelated range. Many notifiers do really block and wait for HW which is harder to handle and we have to bail out though. This patch handles the low hanging fruit. __mmu_notifier_invalidate_range_start gets a blockable flag and callbacks are not allowed to sleep if the flag is set to false. This is achieved by using trylock instead of the sleepable lock for most callbacks and continue as long as we do not block down the call chain. I think we can improve that even further because there is a common pattern to do a range lookup first and then do something about that. The first part can be done without a sleeping lock in most cases AFAICS. The oom_reaper end then simply retries if there is at least one notifier which couldn't make any progress in !blockable mode. A retry loop is already implemented to wait for the mmap_sem and this is basically the same thing. The simplest way for driver developers to test this code path is to wrap userspace code which uses these notifiers into a memcg and set the hard limit to hit the oom. This can be done e.g. after the test faults in all the mmu notifier managed memory and set the hard limit to something really small. Then we are looking for a proper process tear down. [akpm@linux-foundation.org: coding style fixes] [akpm@linux-foundation.org: minor code simplification] Link: http://lkml.kernel.org/r/20180716115058.5559-1-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Acked-by: Christian König <christian.koenig@amd.com> # AMD notifiers Acked-by: Leon Romanovsky <leonro@mellanox.com> # mlx and umem_odp Reported-by: David Rientjes <rientjes@google.com> Cc: "David (ChunMing) Zhou" <David1.Zhou@amd.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Alex Deucher <alexander.deucher@amd.com> Cc: David Airlie <airlied@linux.ie> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Doug Ledford <dledford@redhat.com> Cc: Jason Gunthorpe <jgg@ziepe.ca> Cc: Mike Marciniszyn <mike.marciniszyn@intel.com> Cc: Dennis Dalessandro <dennis.dalessandro@intel.com> Cc: Sudeep Dutt <sudeep.dutt@intel.com> Cc: Ashutosh Dixit <ashutosh.dixit@intel.com> Cc: Dimitri Sivanich <sivanich@sgi.com> Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com> Cc: Juergen Gross <jgross@suse.com> Cc: "Jérôme Glisse" <jglisse@redhat.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-08-22 12:52:33 +08:00
/* failed to reap part of the address space. Try again later */
ret = __oom_reap_task_mm(mm);
if (!ret)
goto out_finish;
mm, oom: introduce oom reaper This patch (of 5): This is based on the idea from Mel Gorman discussed during LSFMM 2015 and independently brought up by Oleg Nesterov. The OOM killer currently allows to kill only a single task in a good hope that the task will terminate in a reasonable time and frees up its memory. Such a task (oom victim) will get an access to memory reserves via mark_oom_victim to allow a forward progress should there be a need for additional memory during exit path. It has been shown (e.g. by Tetsuo Handa) that it is not that hard to construct workloads which break the core assumption mentioned above and the OOM victim might take unbounded amount of time to exit because it might be blocked in the uninterruptible state waiting for an event (e.g. lock) which is blocked by another task looping in the page allocator. This patch reduces the probability of such a lockup by introducing a specialized kernel thread (oom_reaper) which tries to reclaim additional memory by preemptively reaping the anonymous or swapped out memory owned by the oom victim under an assumption that such a memory won't be needed when its owner is killed and kicked from the userspace anyway. There is one notable exception to this, though, if the OOM victim was in the process of coredumping the result would be incomplete. This is considered a reasonable constrain because the overall system health is more important than debugability of a particular application. A kernel thread has been chosen because we need a reliable way of invocation so workqueue context is not appropriate because all the workers might be busy (e.g. allocating memory). Kswapd which sounds like another good fit is not appropriate as well because it might get blocked on locks during reclaim as well. oom_reaper has to take mmap_sem on the target task for reading so the solution is not 100% because the semaphore might be held or blocked for write but the probability is reduced considerably wrt. basically any lock blocking forward progress as described above. In order to prevent from blocking on the lock without any forward progress we are using only a trylock and retry 10 times with a short sleep in between. Users of mmap_sem which need it for write should be carefully reviewed to use _killable waiting as much as possible and reduce allocations requests done with the lock held to absolute minimum to reduce the risk even further. The API between oom killer and oom reaper is quite trivial. wake_oom_reaper updates mm_to_reap with cmpxchg to guarantee only NULL->mm transition and oom_reaper clear this atomically once it is done with the work. This means that only a single mm_struct can be reaped at the time. As the operation is potentially disruptive we are trying to limit it to the ncessary minimum and the reaper blocks any updates while it operates on an mm. mm_struct is pinned by mm_count to allow parallel exit_mmap and a race is detected by atomic_inc_not_zero(mm_users). Signed-off-by: Michal Hocko <mhocko@suse.com> Suggested-by: Oleg Nesterov <oleg@redhat.com> Suggested-by: Mel Gorman <mgorman@suse.de> Acked-by: Mel Gorman <mgorman@suse.de> Acked-by: David Rientjes <rientjes@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Hugh Dickins <hughd@google.com> Cc: Andrea Argangeli <andrea@kernel.org> Cc: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:24 +08:00
pr_info("oom_reaper: reaped process %d (%s), now anon-rss:%lukB, file-rss:%lukB, shmem-rss:%lukB\n",
task_pid_nr(tsk), tsk->comm,
K(get_mm_counter(mm, MM_ANONPAGES)),
K(get_mm_counter(mm, MM_FILEPAGES)),
K(get_mm_counter(mm, MM_SHMEMPAGES)));
out_finish:
trace_finish_task_reaping(tsk->pid);
out_unlock:
mmap locking API: use coccinelle to convert mmap_sem rwsem call sites This change converts the existing mmap_sem rwsem calls to use the new mmap locking API instead. The change is generated using coccinelle with the following rule: // spatch --sp-file mmap_lock_api.cocci --in-place --include-headers --dir . @@ expression mm; @@ ( -init_rwsem +mmap_init_lock | -down_write +mmap_write_lock | -down_write_killable +mmap_write_lock_killable | -down_write_trylock +mmap_write_trylock | -up_write +mmap_write_unlock | -downgrade_write +mmap_write_downgrade | -down_read +mmap_read_lock | -down_read_killable +mmap_read_lock_killable | -down_read_trylock +mmap_read_trylock | -up_read +mmap_read_unlock ) -(&mm->mmap_sem) +(mm) Signed-off-by: Michel Lespinasse <walken@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Reviewed-by: Daniel Jordan <daniel.m.jordan@oracle.com> Reviewed-by: Laurent Dufour <ldufour@linux.ibm.com> Reviewed-by: Vlastimil Babka <vbabka@suse.cz> Cc: Davidlohr Bueso <dbueso@suse.de> Cc: David Rientjes <rientjes@google.com> Cc: Hugh Dickins <hughd@google.com> Cc: Jason Gunthorpe <jgg@ziepe.ca> Cc: Jerome Glisse <jglisse@redhat.com> Cc: John Hubbard <jhubbard@nvidia.com> Cc: Liam Howlett <Liam.Howlett@oracle.com> Cc: Matthew Wilcox <willy@infradead.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Ying Han <yinghan@google.com> Link: http://lkml.kernel.org/r/20200520052908.204642-5-walken@google.com Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-06-09 12:33:25 +08:00
mmap_read_unlock(mm);
oom: clear TIF_MEMDIE after oom_reaper managed to unmap the address space When oom_reaper manages to unmap all the eligible vmas there shouldn't be much of the freable memory held by the oom victim left anymore so it makes sense to clear the TIF_MEMDIE flag for the victim and allow the OOM killer to select another task. The lack of TIF_MEMDIE also means that the victim cannot access memory reserves anymore but that shouldn't be a problem because it would get the access again if it needs to allocate and hits the OOM killer again due to the fatal_signal_pending resp. PF_EXITING check. We can safely hide the task from the OOM killer because it is clearly not a good candidate anymore as everyhing reclaimable has been torn down already. This patch will allow to cap the time an OOM victim can keep TIF_MEMDIE and thus hold off further global OOM killer actions granted the oom reaper is able to take mmap_sem for the associated mm struct. This is not guaranteed now but further steps should make sure that mmap_sem for write should be blocked killable which will help to reduce such a lock contention. This is not done by this patch. Note that exit_oom_victim might be called on a remote task from __oom_reap_task now so we have to check and clear the flag atomically otherwise we might race and underflow oom_victims or wake up waiters too early. Signed-off-by: Michal Hocko <mhocko@suse.com> Suggested-by: Johannes Weiner <hannes@cmpxchg.org> Suggested-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Andrea Argangeli <andrea@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Hugh Dickins <hughd@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:27 +08:00
mm, oom: introduce oom reaper This patch (of 5): This is based on the idea from Mel Gorman discussed during LSFMM 2015 and independently brought up by Oleg Nesterov. The OOM killer currently allows to kill only a single task in a good hope that the task will terminate in a reasonable time and frees up its memory. Such a task (oom victim) will get an access to memory reserves via mark_oom_victim to allow a forward progress should there be a need for additional memory during exit path. It has been shown (e.g. by Tetsuo Handa) that it is not that hard to construct workloads which break the core assumption mentioned above and the OOM victim might take unbounded amount of time to exit because it might be blocked in the uninterruptible state waiting for an event (e.g. lock) which is blocked by another task looping in the page allocator. This patch reduces the probability of such a lockup by introducing a specialized kernel thread (oom_reaper) which tries to reclaim additional memory by preemptively reaping the anonymous or swapped out memory owned by the oom victim under an assumption that such a memory won't be needed when its owner is killed and kicked from the userspace anyway. There is one notable exception to this, though, if the OOM victim was in the process of coredumping the result would be incomplete. This is considered a reasonable constrain because the overall system health is more important than debugability of a particular application. A kernel thread has been chosen because we need a reliable way of invocation so workqueue context is not appropriate because all the workers might be busy (e.g. allocating memory). Kswapd which sounds like another good fit is not appropriate as well because it might get blocked on locks during reclaim as well. oom_reaper has to take mmap_sem on the target task for reading so the solution is not 100% because the semaphore might be held or blocked for write but the probability is reduced considerably wrt. basically any lock blocking forward progress as described above. In order to prevent from blocking on the lock without any forward progress we are using only a trylock and retry 10 times with a short sleep in between. Users of mmap_sem which need it for write should be carefully reviewed to use _killable waiting as much as possible and reduce allocations requests done with the lock held to absolute minimum to reduce the risk even further. The API between oom killer and oom reaper is quite trivial. wake_oom_reaper updates mm_to_reap with cmpxchg to guarantee only NULL->mm transition and oom_reaper clear this atomically once it is done with the work. This means that only a single mm_struct can be reaped at the time. As the operation is potentially disruptive we are trying to limit it to the ncessary minimum and the reaper blocks any updates while it operates on an mm. mm_struct is pinned by mm_count to allow parallel exit_mmap and a race is detected by atomic_inc_not_zero(mm_users). Signed-off-by: Michal Hocko <mhocko@suse.com> Suggested-by: Oleg Nesterov <oleg@redhat.com> Suggested-by: Mel Gorman <mgorman@suse.de> Acked-by: Mel Gorman <mgorman@suse.de> Acked-by: David Rientjes <rientjes@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Hugh Dickins <hughd@google.com> Cc: Andrea Argangeli <andrea@kernel.org> Cc: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:24 +08:00
return ret;
}
#define MAX_OOM_REAP_RETRIES 10
oom: clear TIF_MEMDIE after oom_reaper managed to unmap the address space When oom_reaper manages to unmap all the eligible vmas there shouldn't be much of the freable memory held by the oom victim left anymore so it makes sense to clear the TIF_MEMDIE flag for the victim and allow the OOM killer to select another task. The lack of TIF_MEMDIE also means that the victim cannot access memory reserves anymore but that shouldn't be a problem because it would get the access again if it needs to allocate and hits the OOM killer again due to the fatal_signal_pending resp. PF_EXITING check. We can safely hide the task from the OOM killer because it is clearly not a good candidate anymore as everyhing reclaimable has been torn down already. This patch will allow to cap the time an OOM victim can keep TIF_MEMDIE and thus hold off further global OOM killer actions granted the oom reaper is able to take mmap_sem for the associated mm struct. This is not guaranteed now but further steps should make sure that mmap_sem for write should be blocked killable which will help to reduce such a lock contention. This is not done by this patch. Note that exit_oom_victim might be called on a remote task from __oom_reap_task now so we have to check and clear the flag atomically otherwise we might race and underflow oom_victims or wake up waiters too early. Signed-off-by: Michal Hocko <mhocko@suse.com> Suggested-by: Johannes Weiner <hannes@cmpxchg.org> Suggested-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Andrea Argangeli <andrea@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Hugh Dickins <hughd@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:27 +08:00
static void oom_reap_task(struct task_struct *tsk)
mm, oom: introduce oom reaper This patch (of 5): This is based on the idea from Mel Gorman discussed during LSFMM 2015 and independently brought up by Oleg Nesterov. The OOM killer currently allows to kill only a single task in a good hope that the task will terminate in a reasonable time and frees up its memory. Such a task (oom victim) will get an access to memory reserves via mark_oom_victim to allow a forward progress should there be a need for additional memory during exit path. It has been shown (e.g. by Tetsuo Handa) that it is not that hard to construct workloads which break the core assumption mentioned above and the OOM victim might take unbounded amount of time to exit because it might be blocked in the uninterruptible state waiting for an event (e.g. lock) which is blocked by another task looping in the page allocator. This patch reduces the probability of such a lockup by introducing a specialized kernel thread (oom_reaper) which tries to reclaim additional memory by preemptively reaping the anonymous or swapped out memory owned by the oom victim under an assumption that such a memory won't be needed when its owner is killed and kicked from the userspace anyway. There is one notable exception to this, though, if the OOM victim was in the process of coredumping the result would be incomplete. This is considered a reasonable constrain because the overall system health is more important than debugability of a particular application. A kernel thread has been chosen because we need a reliable way of invocation so workqueue context is not appropriate because all the workers might be busy (e.g. allocating memory). Kswapd which sounds like another good fit is not appropriate as well because it might get blocked on locks during reclaim as well. oom_reaper has to take mmap_sem on the target task for reading so the solution is not 100% because the semaphore might be held or blocked for write but the probability is reduced considerably wrt. basically any lock blocking forward progress as described above. In order to prevent from blocking on the lock without any forward progress we are using only a trylock and retry 10 times with a short sleep in between. Users of mmap_sem which need it for write should be carefully reviewed to use _killable waiting as much as possible and reduce allocations requests done with the lock held to absolute minimum to reduce the risk even further. The API between oom killer and oom reaper is quite trivial. wake_oom_reaper updates mm_to_reap with cmpxchg to guarantee only NULL->mm transition and oom_reaper clear this atomically once it is done with the work. This means that only a single mm_struct can be reaped at the time. As the operation is potentially disruptive we are trying to limit it to the ncessary minimum and the reaper blocks any updates while it operates on an mm. mm_struct is pinned by mm_count to allow parallel exit_mmap and a race is detected by atomic_inc_not_zero(mm_users). Signed-off-by: Michal Hocko <mhocko@suse.com> Suggested-by: Oleg Nesterov <oleg@redhat.com> Suggested-by: Mel Gorman <mgorman@suse.de> Acked-by: Mel Gorman <mgorman@suse.de> Acked-by: David Rientjes <rientjes@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Hugh Dickins <hughd@google.com> Cc: Andrea Argangeli <andrea@kernel.org> Cc: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:24 +08:00
{
int attempts = 0;
oom: keep mm of the killed task available oom_reap_task has to call exit_oom_victim in order to make sure that the oom vicim will not block the oom killer for ever. This is, however, opening new problems (e.g oom_killer_disable exclusion - see commit 74070542099c ("oom, suspend: fix oom_reaper vs. oom_killer_disable race")). exit_oom_victim should be only called from the victim's context ideally. One way to achieve this would be to rely on per mm_struct flags. We already have MMF_OOM_REAPED to hide a task from the oom killer since "mm, oom: hide mm which is shared with kthread or global init". The problem is that the exit path: do_exit exit_mm tsk->mm = NULL; mmput __mmput exit_oom_victim doesn't guarantee that exit_oom_victim will get called in a bounded amount of time. At least exit_aio depends on IO which might get blocked due to lack of memory and who knows what else is lurking there. This patch takes a different approach. We remember tsk->mm into the signal_struct and bind it to the signal struct life time for all oom victims. __oom_reap_task_mm as well as oom_scan_process_thread do not have to rely on find_lock_task_mm anymore and they will have a reliable reference to the mm struct. As a result all the oom specific communication inside the OOM killer can be done via tsk->signal->oom_mm. Increasing the signal_struct for something as unlikely as the oom killer is far from ideal but this approach will make the code much more reasonable and long term we even might want to move task->mm into the signal_struct anyway. In the next step we might want to make the oom killer exclusion and access to memory reserves completely independent which would be also nice. Link: http://lkml.kernel.org/r/1472119394-11342-4-git-send-email-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: David Rientjes <rientjes@google.com> Cc: Vladimir Davydov <vdavydov@parallels.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-10-08 07:58:51 +08:00
struct mm_struct *mm = tsk->signal->oom_mm;
mm, oom: introduce oom reaper This patch (of 5): This is based on the idea from Mel Gorman discussed during LSFMM 2015 and independently brought up by Oleg Nesterov. The OOM killer currently allows to kill only a single task in a good hope that the task will terminate in a reasonable time and frees up its memory. Such a task (oom victim) will get an access to memory reserves via mark_oom_victim to allow a forward progress should there be a need for additional memory during exit path. It has been shown (e.g. by Tetsuo Handa) that it is not that hard to construct workloads which break the core assumption mentioned above and the OOM victim might take unbounded amount of time to exit because it might be blocked in the uninterruptible state waiting for an event (e.g. lock) which is blocked by another task looping in the page allocator. This patch reduces the probability of such a lockup by introducing a specialized kernel thread (oom_reaper) which tries to reclaim additional memory by preemptively reaping the anonymous or swapped out memory owned by the oom victim under an assumption that such a memory won't be needed when its owner is killed and kicked from the userspace anyway. There is one notable exception to this, though, if the OOM victim was in the process of coredumping the result would be incomplete. This is considered a reasonable constrain because the overall system health is more important than debugability of a particular application. A kernel thread has been chosen because we need a reliable way of invocation so workqueue context is not appropriate because all the workers might be busy (e.g. allocating memory). Kswapd which sounds like another good fit is not appropriate as well because it might get blocked on locks during reclaim as well. oom_reaper has to take mmap_sem on the target task for reading so the solution is not 100% because the semaphore might be held or blocked for write but the probability is reduced considerably wrt. basically any lock blocking forward progress as described above. In order to prevent from blocking on the lock without any forward progress we are using only a trylock and retry 10 times with a short sleep in between. Users of mmap_sem which need it for write should be carefully reviewed to use _killable waiting as much as possible and reduce allocations requests done with the lock held to absolute minimum to reduce the risk even further. The API between oom killer and oom reaper is quite trivial. wake_oom_reaper updates mm_to_reap with cmpxchg to guarantee only NULL->mm transition and oom_reaper clear this atomically once it is done with the work. This means that only a single mm_struct can be reaped at the time. As the operation is potentially disruptive we are trying to limit it to the ncessary minimum and the reaper blocks any updates while it operates on an mm. mm_struct is pinned by mm_count to allow parallel exit_mmap and a race is detected by atomic_inc_not_zero(mm_users). Signed-off-by: Michal Hocko <mhocko@suse.com> Suggested-by: Oleg Nesterov <oleg@redhat.com> Suggested-by: Mel Gorman <mgorman@suse.de> Acked-by: Mel Gorman <mgorman@suse.de> Acked-by: David Rientjes <rientjes@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Hugh Dickins <hughd@google.com> Cc: Andrea Argangeli <andrea@kernel.org> Cc: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:24 +08:00
/* Retry the mmap_read_trylock(mm) a few times */
mm, oom: fix concurrent munlock and oom reaper unmap, v3 Since exit_mmap() is done without the protection of mm->mmap_sem, it is possible for the oom reaper to concurrently operate on an mm until MMF_OOM_SKIP is set. This allows munlock_vma_pages_all() to concurrently run while the oom reaper is operating on a vma. Since munlock_vma_pages_range() depends on clearing VM_LOCKED from vm_flags before actually doing the munlock to determine if any other vmas are locking the same memory, the check for VM_LOCKED in the oom reaper is racy. This is especially noticeable on architectures such as powerpc where clearing a huge pmd requires serialize_against_pte_lookup(). If the pmd is zapped by the oom reaper during follow_page_mask() after the check for pmd_none() is bypassed, this ends up deferencing a NULL ptl or a kernel oops. Fix this by manually freeing all possible memory from the mm before doing the munlock and then setting MMF_OOM_SKIP. The oom reaper can not run on the mm anymore so the munlock is safe to do in exit_mmap(). It also matches the logic that the oom reaper currently uses for determining when to set MMF_OOM_SKIP itself, so there's no new risk of excessive oom killing. This issue fixes CVE-2018-1000200. Link: http://lkml.kernel.org/r/alpine.DEB.2.21.1804241526320.238665@chino.kir.corp.google.com Fixes: 212925802454 ("mm: oom: let oom_reap_task and exit_mmap run concurrently") Signed-off-by: David Rientjes <rientjes@google.com> Suggested-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: <stable@vger.kernel.org> [4.14+] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-05-12 07:02:04 +08:00
while (attempts++ < MAX_OOM_REAP_RETRIES && !oom_reap_task_mm(tsk, mm))
mm, oom: introduce oom reaper This patch (of 5): This is based on the idea from Mel Gorman discussed during LSFMM 2015 and independently brought up by Oleg Nesterov. The OOM killer currently allows to kill only a single task in a good hope that the task will terminate in a reasonable time and frees up its memory. Such a task (oom victim) will get an access to memory reserves via mark_oom_victim to allow a forward progress should there be a need for additional memory during exit path. It has been shown (e.g. by Tetsuo Handa) that it is not that hard to construct workloads which break the core assumption mentioned above and the OOM victim might take unbounded amount of time to exit because it might be blocked in the uninterruptible state waiting for an event (e.g. lock) which is blocked by another task looping in the page allocator. This patch reduces the probability of such a lockup by introducing a specialized kernel thread (oom_reaper) which tries to reclaim additional memory by preemptively reaping the anonymous or swapped out memory owned by the oom victim under an assumption that such a memory won't be needed when its owner is killed and kicked from the userspace anyway. There is one notable exception to this, though, if the OOM victim was in the process of coredumping the result would be incomplete. This is considered a reasonable constrain because the overall system health is more important than debugability of a particular application. A kernel thread has been chosen because we need a reliable way of invocation so workqueue context is not appropriate because all the workers might be busy (e.g. allocating memory). Kswapd which sounds like another good fit is not appropriate as well because it might get blocked on locks during reclaim as well. oom_reaper has to take mmap_sem on the target task for reading so the solution is not 100% because the semaphore might be held or blocked for write but the probability is reduced considerably wrt. basically any lock blocking forward progress as described above. In order to prevent from blocking on the lock without any forward progress we are using only a trylock and retry 10 times with a short sleep in between. Users of mmap_sem which need it for write should be carefully reviewed to use _killable waiting as much as possible and reduce allocations requests done with the lock held to absolute minimum to reduce the risk even further. The API between oom killer and oom reaper is quite trivial. wake_oom_reaper updates mm_to_reap with cmpxchg to guarantee only NULL->mm transition and oom_reaper clear this atomically once it is done with the work. This means that only a single mm_struct can be reaped at the time. As the operation is potentially disruptive we are trying to limit it to the ncessary minimum and the reaper blocks any updates while it operates on an mm. mm_struct is pinned by mm_count to allow parallel exit_mmap and a race is detected by atomic_inc_not_zero(mm_users). Signed-off-by: Michal Hocko <mhocko@suse.com> Suggested-by: Oleg Nesterov <oleg@redhat.com> Suggested-by: Mel Gorman <mgorman@suse.de> Acked-by: Mel Gorman <mgorman@suse.de> Acked-by: David Rientjes <rientjes@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Hugh Dickins <hughd@google.com> Cc: Andrea Argangeli <andrea@kernel.org> Cc: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:24 +08:00
schedule_timeout_idle(HZ/10);
if (attempts <= MAX_OOM_REAP_RETRIES ||
test_bit(MMF_OOM_SKIP, &mm->flags))
goto done;
pr_info("oom_reaper: unable to reap pid:%d (%s)\n",
task_pid_nr(tsk), tsk->comm);
sched_show_task(tsk);
debug_show_all_locks();
done:
mm, oom_reaper: clear TIF_MEMDIE for all tasks queued for oom_reaper Right now the oom reaper will clear TIF_MEMDIE only for tasks which were successfully reaped. This is the safest option because we know that such an oom victim would only block forward progress of the oom killer without a good reason because it is highly unlikely it would release much more memory. Basically most of its memory has been already torn down. We can relax this assumption to catch more corner cases though. The first obvious one is when the oom victim clears its mm and gets stuck later on. oom_reaper would back of on find_lock_task_mm returning NULL. We can safely try to clear TIF_MEMDIE in this case because such a task would be ignored by the oom killer anyway. The flag would be cleared by that time already most of the time anyway. The less obvious one is when the oom reaper fails due to mmap_sem contention. Even if we clear TIF_MEMDIE for this task then it is not very likely that we would select another task too easily because we haven't reaped the last victim and so it would be still the #1 candidate. There is a rare race condition possible when the current victim terminates before the next select_bad_process but considering that oom_reap_task had retried several times before giving up then this sounds like a borderline thing. After this patch we should have a guarantee that the OOM killer will not be block for unbounded amount of time for most cases. Signed-off-by: Michal Hocko <mhocko@suse.com> Cc: Raushaniya Maksudova <rmaksudova@parallels.com> Cc: Michael S. Tsirkin <mst@redhat.com> Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: David Rientjes <rientjes@google.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Daniel Vetter <daniel.vetter@intel.com> Cc: Oleg Nesterov <oleg@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-20 08:13:15 +08:00
tsk->oom_reaper_list = NULL;
oom: keep mm of the killed task available oom_reap_task has to call exit_oom_victim in order to make sure that the oom vicim will not block the oom killer for ever. This is, however, opening new problems (e.g oom_killer_disable exclusion - see commit 74070542099c ("oom, suspend: fix oom_reaper vs. oom_killer_disable race")). exit_oom_victim should be only called from the victim's context ideally. One way to achieve this would be to rely on per mm_struct flags. We already have MMF_OOM_REAPED to hide a task from the oom killer since "mm, oom: hide mm which is shared with kthread or global init". The problem is that the exit path: do_exit exit_mm tsk->mm = NULL; mmput __mmput exit_oom_victim doesn't guarantee that exit_oom_victim will get called in a bounded amount of time. At least exit_aio depends on IO which might get blocked due to lack of memory and who knows what else is lurking there. This patch takes a different approach. We remember tsk->mm into the signal_struct and bind it to the signal struct life time for all oom victims. __oom_reap_task_mm as well as oom_scan_process_thread do not have to rely on find_lock_task_mm anymore and they will have a reliable reference to the mm struct. As a result all the oom specific communication inside the OOM killer can be done via tsk->signal->oom_mm. Increasing the signal_struct for something as unlikely as the oom killer is far from ideal but this approach will make the code much more reasonable and long term we even might want to move task->mm into the signal_struct anyway. In the next step we might want to make the oom killer exclusion and access to memory reserves completely independent which would be also nice. Link: http://lkml.kernel.org/r/1472119394-11342-4-git-send-email-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: David Rientjes <rientjes@google.com> Cc: Vladimir Davydov <vdavydov@parallels.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-10-08 07:58:51 +08:00
/*
* Hide this mm from OOM killer because it has been either reaped or
* somebody can't call mmap_write_unlock(mm).
oom: keep mm of the killed task available oom_reap_task has to call exit_oom_victim in order to make sure that the oom vicim will not block the oom killer for ever. This is, however, opening new problems (e.g oom_killer_disable exclusion - see commit 74070542099c ("oom, suspend: fix oom_reaper vs. oom_killer_disable race")). exit_oom_victim should be only called from the victim's context ideally. One way to achieve this would be to rely on per mm_struct flags. We already have MMF_OOM_REAPED to hide a task from the oom killer since "mm, oom: hide mm which is shared with kthread or global init". The problem is that the exit path: do_exit exit_mm tsk->mm = NULL; mmput __mmput exit_oom_victim doesn't guarantee that exit_oom_victim will get called in a bounded amount of time. At least exit_aio depends on IO which might get blocked due to lack of memory and who knows what else is lurking there. This patch takes a different approach. We remember tsk->mm into the signal_struct and bind it to the signal struct life time for all oom victims. __oom_reap_task_mm as well as oom_scan_process_thread do not have to rely on find_lock_task_mm anymore and they will have a reliable reference to the mm struct. As a result all the oom specific communication inside the OOM killer can be done via tsk->signal->oom_mm. Increasing the signal_struct for something as unlikely as the oom killer is far from ideal but this approach will make the code much more reasonable and long term we even might want to move task->mm into the signal_struct anyway. In the next step we might want to make the oom killer exclusion and access to memory reserves completely independent which would be also nice. Link: http://lkml.kernel.org/r/1472119394-11342-4-git-send-email-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: David Rientjes <rientjes@google.com> Cc: Vladimir Davydov <vdavydov@parallels.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-10-08 07:58:51 +08:00
*/
set_bit(MMF_OOM_SKIP, &mm->flags);
oom: keep mm of the killed task available oom_reap_task has to call exit_oom_victim in order to make sure that the oom vicim will not block the oom killer for ever. This is, however, opening new problems (e.g oom_killer_disable exclusion - see commit 74070542099c ("oom, suspend: fix oom_reaper vs. oom_killer_disable race")). exit_oom_victim should be only called from the victim's context ideally. One way to achieve this would be to rely on per mm_struct flags. We already have MMF_OOM_REAPED to hide a task from the oom killer since "mm, oom: hide mm which is shared with kthread or global init". The problem is that the exit path: do_exit exit_mm tsk->mm = NULL; mmput __mmput exit_oom_victim doesn't guarantee that exit_oom_victim will get called in a bounded amount of time. At least exit_aio depends on IO which might get blocked due to lack of memory and who knows what else is lurking there. This patch takes a different approach. We remember tsk->mm into the signal_struct and bind it to the signal struct life time for all oom victims. __oom_reap_task_mm as well as oom_scan_process_thread do not have to rely on find_lock_task_mm anymore and they will have a reliable reference to the mm struct. As a result all the oom specific communication inside the OOM killer can be done via tsk->signal->oom_mm. Increasing the signal_struct for something as unlikely as the oom killer is far from ideal but this approach will make the code much more reasonable and long term we even might want to move task->mm into the signal_struct anyway. In the next step we might want to make the oom killer exclusion and access to memory reserves completely independent which would be also nice. Link: http://lkml.kernel.org/r/1472119394-11342-4-git-send-email-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: David Rientjes <rientjes@google.com> Cc: Vladimir Davydov <vdavydov@parallels.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-10-08 07:58:51 +08:00
mm, oom: introduce oom reaper This patch (of 5): This is based on the idea from Mel Gorman discussed during LSFMM 2015 and independently brought up by Oleg Nesterov. The OOM killer currently allows to kill only a single task in a good hope that the task will terminate in a reasonable time and frees up its memory. Such a task (oom victim) will get an access to memory reserves via mark_oom_victim to allow a forward progress should there be a need for additional memory during exit path. It has been shown (e.g. by Tetsuo Handa) that it is not that hard to construct workloads which break the core assumption mentioned above and the OOM victim might take unbounded amount of time to exit because it might be blocked in the uninterruptible state waiting for an event (e.g. lock) which is blocked by another task looping in the page allocator. This patch reduces the probability of such a lockup by introducing a specialized kernel thread (oom_reaper) which tries to reclaim additional memory by preemptively reaping the anonymous or swapped out memory owned by the oom victim under an assumption that such a memory won't be needed when its owner is killed and kicked from the userspace anyway. There is one notable exception to this, though, if the OOM victim was in the process of coredumping the result would be incomplete. This is considered a reasonable constrain because the overall system health is more important than debugability of a particular application. A kernel thread has been chosen because we need a reliable way of invocation so workqueue context is not appropriate because all the workers might be busy (e.g. allocating memory). Kswapd which sounds like another good fit is not appropriate as well because it might get blocked on locks during reclaim as well. oom_reaper has to take mmap_sem on the target task for reading so the solution is not 100% because the semaphore might be held or blocked for write but the probability is reduced considerably wrt. basically any lock blocking forward progress as described above. In order to prevent from blocking on the lock without any forward progress we are using only a trylock and retry 10 times with a short sleep in between. Users of mmap_sem which need it for write should be carefully reviewed to use _killable waiting as much as possible and reduce allocations requests done with the lock held to absolute minimum to reduce the risk even further. The API between oom killer and oom reaper is quite trivial. wake_oom_reaper updates mm_to_reap with cmpxchg to guarantee only NULL->mm transition and oom_reaper clear this atomically once it is done with the work. This means that only a single mm_struct can be reaped at the time. As the operation is potentially disruptive we are trying to limit it to the ncessary minimum and the reaper blocks any updates while it operates on an mm. mm_struct is pinned by mm_count to allow parallel exit_mmap and a race is detected by atomic_inc_not_zero(mm_users). Signed-off-by: Michal Hocko <mhocko@suse.com> Suggested-by: Oleg Nesterov <oleg@redhat.com> Suggested-by: Mel Gorman <mgorman@suse.de> Acked-by: Mel Gorman <mgorman@suse.de> Acked-by: David Rientjes <rientjes@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Hugh Dickins <hughd@google.com> Cc: Andrea Argangeli <andrea@kernel.org> Cc: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:24 +08:00
/* Drop a reference taken by wake_oom_reaper */
oom: clear TIF_MEMDIE after oom_reaper managed to unmap the address space When oom_reaper manages to unmap all the eligible vmas there shouldn't be much of the freable memory held by the oom victim left anymore so it makes sense to clear the TIF_MEMDIE flag for the victim and allow the OOM killer to select another task. The lack of TIF_MEMDIE also means that the victim cannot access memory reserves anymore but that shouldn't be a problem because it would get the access again if it needs to allocate and hits the OOM killer again due to the fatal_signal_pending resp. PF_EXITING check. We can safely hide the task from the OOM killer because it is clearly not a good candidate anymore as everyhing reclaimable has been torn down already. This patch will allow to cap the time an OOM victim can keep TIF_MEMDIE and thus hold off further global OOM killer actions granted the oom reaper is able to take mmap_sem for the associated mm struct. This is not guaranteed now but further steps should make sure that mmap_sem for write should be blocked killable which will help to reduce such a lock contention. This is not done by this patch. Note that exit_oom_victim might be called on a remote task from __oom_reap_task now so we have to check and clear the flag atomically otherwise we might race and underflow oom_victims or wake up waiters too early. Signed-off-by: Michal Hocko <mhocko@suse.com> Suggested-by: Johannes Weiner <hannes@cmpxchg.org> Suggested-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Andrea Argangeli <andrea@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Hugh Dickins <hughd@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:27 +08:00
put_task_struct(tsk);
mm, oom: introduce oom reaper This patch (of 5): This is based on the idea from Mel Gorman discussed during LSFMM 2015 and independently brought up by Oleg Nesterov. The OOM killer currently allows to kill only a single task in a good hope that the task will terminate in a reasonable time and frees up its memory. Such a task (oom victim) will get an access to memory reserves via mark_oom_victim to allow a forward progress should there be a need for additional memory during exit path. It has been shown (e.g. by Tetsuo Handa) that it is not that hard to construct workloads which break the core assumption mentioned above and the OOM victim might take unbounded amount of time to exit because it might be blocked in the uninterruptible state waiting for an event (e.g. lock) which is blocked by another task looping in the page allocator. This patch reduces the probability of such a lockup by introducing a specialized kernel thread (oom_reaper) which tries to reclaim additional memory by preemptively reaping the anonymous or swapped out memory owned by the oom victim under an assumption that such a memory won't be needed when its owner is killed and kicked from the userspace anyway. There is one notable exception to this, though, if the OOM victim was in the process of coredumping the result would be incomplete. This is considered a reasonable constrain because the overall system health is more important than debugability of a particular application. A kernel thread has been chosen because we need a reliable way of invocation so workqueue context is not appropriate because all the workers might be busy (e.g. allocating memory). Kswapd which sounds like another good fit is not appropriate as well because it might get blocked on locks during reclaim as well. oom_reaper has to take mmap_sem on the target task for reading so the solution is not 100% because the semaphore might be held or blocked for write but the probability is reduced considerably wrt. basically any lock blocking forward progress as described above. In order to prevent from blocking on the lock without any forward progress we are using only a trylock and retry 10 times with a short sleep in between. Users of mmap_sem which need it for write should be carefully reviewed to use _killable waiting as much as possible and reduce allocations requests done with the lock held to absolute minimum to reduce the risk even further. The API between oom killer and oom reaper is quite trivial. wake_oom_reaper updates mm_to_reap with cmpxchg to guarantee only NULL->mm transition and oom_reaper clear this atomically once it is done with the work. This means that only a single mm_struct can be reaped at the time. As the operation is potentially disruptive we are trying to limit it to the ncessary minimum and the reaper blocks any updates while it operates on an mm. mm_struct is pinned by mm_count to allow parallel exit_mmap and a race is detected by atomic_inc_not_zero(mm_users). Signed-off-by: Michal Hocko <mhocko@suse.com> Suggested-by: Oleg Nesterov <oleg@redhat.com> Suggested-by: Mel Gorman <mgorman@suse.de> Acked-by: Mel Gorman <mgorman@suse.de> Acked-by: David Rientjes <rientjes@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Hugh Dickins <hughd@google.com> Cc: Andrea Argangeli <andrea@kernel.org> Cc: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:24 +08:00
}
static int oom_reaper(void *unused)
{
while (true) {
mm, oom_reaper: implement OOM victims queuing wake_oom_reaper has allowed only 1 oom victim to be queued. The main reason for that was the simplicity as other solutions would require some way of queuing. The current approach is racy and that was deemed sufficient as the oom_reaper is considered a best effort approach to help with oom handling when the OOM victim cannot terminate in a reasonable time. The race could lead to missing an oom victim which can get stuck out_of_memory wake_oom_reaper cmpxchg // OK oom_reaper oom_reap_task __oom_reap_task oom_victim terminates atomic_inc_not_zero // fail out_of_memory wake_oom_reaper cmpxchg // fails task_to_reap = NULL This race requires 2 OOM invocations in a short time period which is not very likely but certainly not impossible. E.g. the original victim might have not released a lot of memory for some reason. The situation would improve considerably if wake_oom_reaper used a more robust queuing. This is what this patch implements. This means adding oom_reaper_list list_head into task_struct (eat a hole before embeded thread_struct for that purpose) and a oom_reaper_lock spinlock for queuing synchronization. wake_oom_reaper will then add the task on the queue and oom_reaper will dequeue it. Signed-off-by: Michal Hocko <mhocko@suse.com> Cc: Vladimir Davydov <vdavydov@virtuozzo.com> Cc: Andrea Argangeli <andrea@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Hugh Dickins <hughd@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Mel Gorman <mgorman@suse.de> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Rik van Riel <riel@redhat.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:33 +08:00
struct task_struct *tsk = NULL;
mm, oom: introduce oom reaper This patch (of 5): This is based on the idea from Mel Gorman discussed during LSFMM 2015 and independently brought up by Oleg Nesterov. The OOM killer currently allows to kill only a single task in a good hope that the task will terminate in a reasonable time and frees up its memory. Such a task (oom victim) will get an access to memory reserves via mark_oom_victim to allow a forward progress should there be a need for additional memory during exit path. It has been shown (e.g. by Tetsuo Handa) that it is not that hard to construct workloads which break the core assumption mentioned above and the OOM victim might take unbounded amount of time to exit because it might be blocked in the uninterruptible state waiting for an event (e.g. lock) which is blocked by another task looping in the page allocator. This patch reduces the probability of such a lockup by introducing a specialized kernel thread (oom_reaper) which tries to reclaim additional memory by preemptively reaping the anonymous or swapped out memory owned by the oom victim under an assumption that such a memory won't be needed when its owner is killed and kicked from the userspace anyway. There is one notable exception to this, though, if the OOM victim was in the process of coredumping the result would be incomplete. This is considered a reasonable constrain because the overall system health is more important than debugability of a particular application. A kernel thread has been chosen because we need a reliable way of invocation so workqueue context is not appropriate because all the workers might be busy (e.g. allocating memory). Kswapd which sounds like another good fit is not appropriate as well because it might get blocked on locks during reclaim as well. oom_reaper has to take mmap_sem on the target task for reading so the solution is not 100% because the semaphore might be held or blocked for write but the probability is reduced considerably wrt. basically any lock blocking forward progress as described above. In order to prevent from blocking on the lock without any forward progress we are using only a trylock and retry 10 times with a short sleep in between. Users of mmap_sem which need it for write should be carefully reviewed to use _killable waiting as much as possible and reduce allocations requests done with the lock held to absolute minimum to reduce the risk even further. The API between oom killer and oom reaper is quite trivial. wake_oom_reaper updates mm_to_reap with cmpxchg to guarantee only NULL->mm transition and oom_reaper clear this atomically once it is done with the work. This means that only a single mm_struct can be reaped at the time. As the operation is potentially disruptive we are trying to limit it to the ncessary minimum and the reaper blocks any updates while it operates on an mm. mm_struct is pinned by mm_count to allow parallel exit_mmap and a race is detected by atomic_inc_not_zero(mm_users). Signed-off-by: Michal Hocko <mhocko@suse.com> Suggested-by: Oleg Nesterov <oleg@redhat.com> Suggested-by: Mel Gorman <mgorman@suse.de> Acked-by: Mel Gorman <mgorman@suse.de> Acked-by: David Rientjes <rientjes@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Hugh Dickins <hughd@google.com> Cc: Andrea Argangeli <andrea@kernel.org> Cc: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:24 +08:00
wait_event_freezable(oom_reaper_wait, oom_reaper_list != NULL);
mm, oom_reaper: implement OOM victims queuing wake_oom_reaper has allowed only 1 oom victim to be queued. The main reason for that was the simplicity as other solutions would require some way of queuing. The current approach is racy and that was deemed sufficient as the oom_reaper is considered a best effort approach to help with oom handling when the OOM victim cannot terminate in a reasonable time. The race could lead to missing an oom victim which can get stuck out_of_memory wake_oom_reaper cmpxchg // OK oom_reaper oom_reap_task __oom_reap_task oom_victim terminates atomic_inc_not_zero // fail out_of_memory wake_oom_reaper cmpxchg // fails task_to_reap = NULL This race requires 2 OOM invocations in a short time period which is not very likely but certainly not impossible. E.g. the original victim might have not released a lot of memory for some reason. The situation would improve considerably if wake_oom_reaper used a more robust queuing. This is what this patch implements. This means adding oom_reaper_list list_head into task_struct (eat a hole before embeded thread_struct for that purpose) and a oom_reaper_lock spinlock for queuing synchronization. wake_oom_reaper will then add the task on the queue and oom_reaper will dequeue it. Signed-off-by: Michal Hocko <mhocko@suse.com> Cc: Vladimir Davydov <vdavydov@virtuozzo.com> Cc: Andrea Argangeli <andrea@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Hugh Dickins <hughd@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Mel Gorman <mgorman@suse.de> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Rik van Riel <riel@redhat.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:33 +08:00
spin_lock(&oom_reaper_lock);
if (oom_reaper_list != NULL) {
tsk = oom_reaper_list;
oom_reaper_list = tsk->oom_reaper_list;
mm, oom_reaper: implement OOM victims queuing wake_oom_reaper has allowed only 1 oom victim to be queued. The main reason for that was the simplicity as other solutions would require some way of queuing. The current approach is racy and that was deemed sufficient as the oom_reaper is considered a best effort approach to help with oom handling when the OOM victim cannot terminate in a reasonable time. The race could lead to missing an oom victim which can get stuck out_of_memory wake_oom_reaper cmpxchg // OK oom_reaper oom_reap_task __oom_reap_task oom_victim terminates atomic_inc_not_zero // fail out_of_memory wake_oom_reaper cmpxchg // fails task_to_reap = NULL This race requires 2 OOM invocations in a short time period which is not very likely but certainly not impossible. E.g. the original victim might have not released a lot of memory for some reason. The situation would improve considerably if wake_oom_reaper used a more robust queuing. This is what this patch implements. This means adding oom_reaper_list list_head into task_struct (eat a hole before embeded thread_struct for that purpose) and a oom_reaper_lock spinlock for queuing synchronization. wake_oom_reaper will then add the task on the queue and oom_reaper will dequeue it. Signed-off-by: Michal Hocko <mhocko@suse.com> Cc: Vladimir Davydov <vdavydov@virtuozzo.com> Cc: Andrea Argangeli <andrea@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Hugh Dickins <hughd@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Mel Gorman <mgorman@suse.de> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Rik van Riel <riel@redhat.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:33 +08:00
}
spin_unlock(&oom_reaper_lock);
if (tsk)
oom_reap_task(tsk);
mm, oom: introduce oom reaper This patch (of 5): This is based on the idea from Mel Gorman discussed during LSFMM 2015 and independently brought up by Oleg Nesterov. The OOM killer currently allows to kill only a single task in a good hope that the task will terminate in a reasonable time and frees up its memory. Such a task (oom victim) will get an access to memory reserves via mark_oom_victim to allow a forward progress should there be a need for additional memory during exit path. It has been shown (e.g. by Tetsuo Handa) that it is not that hard to construct workloads which break the core assumption mentioned above and the OOM victim might take unbounded amount of time to exit because it might be blocked in the uninterruptible state waiting for an event (e.g. lock) which is blocked by another task looping in the page allocator. This patch reduces the probability of such a lockup by introducing a specialized kernel thread (oom_reaper) which tries to reclaim additional memory by preemptively reaping the anonymous or swapped out memory owned by the oom victim under an assumption that such a memory won't be needed when its owner is killed and kicked from the userspace anyway. There is one notable exception to this, though, if the OOM victim was in the process of coredumping the result would be incomplete. This is considered a reasonable constrain because the overall system health is more important than debugability of a particular application. A kernel thread has been chosen because we need a reliable way of invocation so workqueue context is not appropriate because all the workers might be busy (e.g. allocating memory). Kswapd which sounds like another good fit is not appropriate as well because it might get blocked on locks during reclaim as well. oom_reaper has to take mmap_sem on the target task for reading so the solution is not 100% because the semaphore might be held or blocked for write but the probability is reduced considerably wrt. basically any lock blocking forward progress as described above. In order to prevent from blocking on the lock without any forward progress we are using only a trylock and retry 10 times with a short sleep in between. Users of mmap_sem which need it for write should be carefully reviewed to use _killable waiting as much as possible and reduce allocations requests done with the lock held to absolute minimum to reduce the risk even further. The API between oom killer and oom reaper is quite trivial. wake_oom_reaper updates mm_to_reap with cmpxchg to guarantee only NULL->mm transition and oom_reaper clear this atomically once it is done with the work. This means that only a single mm_struct can be reaped at the time. As the operation is potentially disruptive we are trying to limit it to the ncessary minimum and the reaper blocks any updates while it operates on an mm. mm_struct is pinned by mm_count to allow parallel exit_mmap and a race is detected by atomic_inc_not_zero(mm_users). Signed-off-by: Michal Hocko <mhocko@suse.com> Suggested-by: Oleg Nesterov <oleg@redhat.com> Suggested-by: Mel Gorman <mgorman@suse.de> Acked-by: Mel Gorman <mgorman@suse.de> Acked-by: David Rientjes <rientjes@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Hugh Dickins <hughd@google.com> Cc: Andrea Argangeli <andrea@kernel.org> Cc: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:24 +08:00
}
return 0;
}
static void wake_oom_reaper(struct task_struct *tsk)
mm, oom: introduce oom reaper This patch (of 5): This is based on the idea from Mel Gorman discussed during LSFMM 2015 and independently brought up by Oleg Nesterov. The OOM killer currently allows to kill only a single task in a good hope that the task will terminate in a reasonable time and frees up its memory. Such a task (oom victim) will get an access to memory reserves via mark_oom_victim to allow a forward progress should there be a need for additional memory during exit path. It has been shown (e.g. by Tetsuo Handa) that it is not that hard to construct workloads which break the core assumption mentioned above and the OOM victim might take unbounded amount of time to exit because it might be blocked in the uninterruptible state waiting for an event (e.g. lock) which is blocked by another task looping in the page allocator. This patch reduces the probability of such a lockup by introducing a specialized kernel thread (oom_reaper) which tries to reclaim additional memory by preemptively reaping the anonymous or swapped out memory owned by the oom victim under an assumption that such a memory won't be needed when its owner is killed and kicked from the userspace anyway. There is one notable exception to this, though, if the OOM victim was in the process of coredumping the result would be incomplete. This is considered a reasonable constrain because the overall system health is more important than debugability of a particular application. A kernel thread has been chosen because we need a reliable way of invocation so workqueue context is not appropriate because all the workers might be busy (e.g. allocating memory). Kswapd which sounds like another good fit is not appropriate as well because it might get blocked on locks during reclaim as well. oom_reaper has to take mmap_sem on the target task for reading so the solution is not 100% because the semaphore might be held or blocked for write but the probability is reduced considerably wrt. basically any lock blocking forward progress as described above. In order to prevent from blocking on the lock without any forward progress we are using only a trylock and retry 10 times with a short sleep in between. Users of mmap_sem which need it for write should be carefully reviewed to use _killable waiting as much as possible and reduce allocations requests done with the lock held to absolute minimum to reduce the risk even further. The API between oom killer and oom reaper is quite trivial. wake_oom_reaper updates mm_to_reap with cmpxchg to guarantee only NULL->mm transition and oom_reaper clear this atomically once it is done with the work. This means that only a single mm_struct can be reaped at the time. As the operation is potentially disruptive we are trying to limit it to the ncessary minimum and the reaper blocks any updates while it operates on an mm. mm_struct is pinned by mm_count to allow parallel exit_mmap and a race is detected by atomic_inc_not_zero(mm_users). Signed-off-by: Michal Hocko <mhocko@suse.com> Suggested-by: Oleg Nesterov <oleg@redhat.com> Suggested-by: Mel Gorman <mgorman@suse.de> Acked-by: Mel Gorman <mgorman@suse.de> Acked-by: David Rientjes <rientjes@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Hugh Dickins <hughd@google.com> Cc: Andrea Argangeli <andrea@kernel.org> Cc: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:24 +08:00
{
oom, oom_reaper: do not enqueue same task twice Arkadiusz reported that enabling memcg's group oom killing causes strange memcg statistics where there is no task in a memcg despite the number of tasks in that memcg is not 0. It turned out that there is a bug in wake_oom_reaper() which allows enqueuing same task twice which makes impossible to decrease the number of tasks in that memcg due to a refcount leak. This bug existed since the OOM reaper became invokable from task_will_free_mem(current) path in out_of_memory() in Linux 4.7, T1@P1 |T2@P1 |T3@P1 |OOM reaper ----------+----------+----------+------------ # Processing an OOM victim in a different memcg domain. try_charge() mem_cgroup_out_of_memory() mutex_lock(&oom_lock) try_charge() mem_cgroup_out_of_memory() mutex_lock(&oom_lock) try_charge() mem_cgroup_out_of_memory() mutex_lock(&oom_lock) out_of_memory() oom_kill_process(P1) do_send_sig_info(SIGKILL, @P1) mark_oom_victim(T1@P1) wake_oom_reaper(T1@P1) # T1@P1 is enqueued. mutex_unlock(&oom_lock) out_of_memory() mark_oom_victim(T2@P1) wake_oom_reaper(T2@P1) # T2@P1 is enqueued. mutex_unlock(&oom_lock) out_of_memory() mark_oom_victim(T1@P1) wake_oom_reaper(T1@P1) # T1@P1 is enqueued again due to oom_reaper_list == T2@P1 && T1@P1->oom_reaper_list == NULL. mutex_unlock(&oom_lock) # Completed processing an OOM victim in a different memcg domain. spin_lock(&oom_reaper_lock) # T1P1 is dequeued. spin_unlock(&oom_reaper_lock) but memcg's group oom killing made it easier to trigger this bug by calling wake_oom_reaper() on the same task from one out_of_memory() request. Fix this bug using an approach used by commit 855b018325737f76 ("oom, oom_reaper: disable oom_reaper for oom_kill_allocating_task"). As a side effect of this patch, this patch also avoids enqueuing multiple threads sharing memory via task_will_free_mem(current) path. Link: http://lkml.kernel.org/r/e865a044-2c10-9858-f4ef-254bc71d6cc2@i-love.sakura.ne.jp Link: http://lkml.kernel.org/r/5ee34fc6-1485-34f8-8790-903ddabaa809@i-love.sakura.ne.jp Fixes: af8e15cc85a25315 ("oom, oom_reaper: do not enqueue task if it is on the oom_reaper_list head") Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Reported-by: Arkadiusz Miskiewicz <arekm@maven.pl> Tested-by: Arkadiusz Miskiewicz <arekm@maven.pl> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: Roman Gushchin <guro@fb.com> Cc: Tejun Heo <tj@kernel.org> Cc: Aleksa Sarai <asarai@suse.de> Cc: Jay Kamat <jgkamat@fb.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-02-02 06:20:31 +08:00
/* mm is already queued? */
if (test_and_set_bit(MMF_OOM_REAP_QUEUED, &tsk->signal->oom_mm->flags))
mm, oom: introduce oom reaper This patch (of 5): This is based on the idea from Mel Gorman discussed during LSFMM 2015 and independently brought up by Oleg Nesterov. The OOM killer currently allows to kill only a single task in a good hope that the task will terminate in a reasonable time and frees up its memory. Such a task (oom victim) will get an access to memory reserves via mark_oom_victim to allow a forward progress should there be a need for additional memory during exit path. It has been shown (e.g. by Tetsuo Handa) that it is not that hard to construct workloads which break the core assumption mentioned above and the OOM victim might take unbounded amount of time to exit because it might be blocked in the uninterruptible state waiting for an event (e.g. lock) which is blocked by another task looping in the page allocator. This patch reduces the probability of such a lockup by introducing a specialized kernel thread (oom_reaper) which tries to reclaim additional memory by preemptively reaping the anonymous or swapped out memory owned by the oom victim under an assumption that such a memory won't be needed when its owner is killed and kicked from the userspace anyway. There is one notable exception to this, though, if the OOM victim was in the process of coredumping the result would be incomplete. This is considered a reasonable constrain because the overall system health is more important than debugability of a particular application. A kernel thread has been chosen because we need a reliable way of invocation so workqueue context is not appropriate because all the workers might be busy (e.g. allocating memory). Kswapd which sounds like another good fit is not appropriate as well because it might get blocked on locks during reclaim as well. oom_reaper has to take mmap_sem on the target task for reading so the solution is not 100% because the semaphore might be held or blocked for write but the probability is reduced considerably wrt. basically any lock blocking forward progress as described above. In order to prevent from blocking on the lock without any forward progress we are using only a trylock and retry 10 times with a short sleep in between. Users of mmap_sem which need it for write should be carefully reviewed to use _killable waiting as much as possible and reduce allocations requests done with the lock held to absolute minimum to reduce the risk even further. The API between oom killer and oom reaper is quite trivial. wake_oom_reaper updates mm_to_reap with cmpxchg to guarantee only NULL->mm transition and oom_reaper clear this atomically once it is done with the work. This means that only a single mm_struct can be reaped at the time. As the operation is potentially disruptive we are trying to limit it to the ncessary minimum and the reaper blocks any updates while it operates on an mm. mm_struct is pinned by mm_count to allow parallel exit_mmap and a race is detected by atomic_inc_not_zero(mm_users). Signed-off-by: Michal Hocko <mhocko@suse.com> Suggested-by: Oleg Nesterov <oleg@redhat.com> Suggested-by: Mel Gorman <mgorman@suse.de> Acked-by: Mel Gorman <mgorman@suse.de> Acked-by: David Rientjes <rientjes@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Hugh Dickins <hughd@google.com> Cc: Andrea Argangeli <andrea@kernel.org> Cc: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:24 +08:00
return;
oom: clear TIF_MEMDIE after oom_reaper managed to unmap the address space When oom_reaper manages to unmap all the eligible vmas there shouldn't be much of the freable memory held by the oom victim left anymore so it makes sense to clear the TIF_MEMDIE flag for the victim and allow the OOM killer to select another task. The lack of TIF_MEMDIE also means that the victim cannot access memory reserves anymore but that shouldn't be a problem because it would get the access again if it needs to allocate and hits the OOM killer again due to the fatal_signal_pending resp. PF_EXITING check. We can safely hide the task from the OOM killer because it is clearly not a good candidate anymore as everyhing reclaimable has been torn down already. This patch will allow to cap the time an OOM victim can keep TIF_MEMDIE and thus hold off further global OOM killer actions granted the oom reaper is able to take mmap_sem for the associated mm struct. This is not guaranteed now but further steps should make sure that mmap_sem for write should be blocked killable which will help to reduce such a lock contention. This is not done by this patch. Note that exit_oom_victim might be called on a remote task from __oom_reap_task now so we have to check and clear the flag atomically otherwise we might race and underflow oom_victims or wake up waiters too early. Signed-off-by: Michal Hocko <mhocko@suse.com> Suggested-by: Johannes Weiner <hannes@cmpxchg.org> Suggested-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Andrea Argangeli <andrea@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Hugh Dickins <hughd@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:27 +08:00
get_task_struct(tsk);
mm, oom: introduce oom reaper This patch (of 5): This is based on the idea from Mel Gorman discussed during LSFMM 2015 and independently brought up by Oleg Nesterov. The OOM killer currently allows to kill only a single task in a good hope that the task will terminate in a reasonable time and frees up its memory. Such a task (oom victim) will get an access to memory reserves via mark_oom_victim to allow a forward progress should there be a need for additional memory during exit path. It has been shown (e.g. by Tetsuo Handa) that it is not that hard to construct workloads which break the core assumption mentioned above and the OOM victim might take unbounded amount of time to exit because it might be blocked in the uninterruptible state waiting for an event (e.g. lock) which is blocked by another task looping in the page allocator. This patch reduces the probability of such a lockup by introducing a specialized kernel thread (oom_reaper) which tries to reclaim additional memory by preemptively reaping the anonymous or swapped out memory owned by the oom victim under an assumption that such a memory won't be needed when its owner is killed and kicked from the userspace anyway. There is one notable exception to this, though, if the OOM victim was in the process of coredumping the result would be incomplete. This is considered a reasonable constrain because the overall system health is more important than debugability of a particular application. A kernel thread has been chosen because we need a reliable way of invocation so workqueue context is not appropriate because all the workers might be busy (e.g. allocating memory). Kswapd which sounds like another good fit is not appropriate as well because it might get blocked on locks during reclaim as well. oom_reaper has to take mmap_sem on the target task for reading so the solution is not 100% because the semaphore might be held or blocked for write but the probability is reduced considerably wrt. basically any lock blocking forward progress as described above. In order to prevent from blocking on the lock without any forward progress we are using only a trylock and retry 10 times with a short sleep in between. Users of mmap_sem which need it for write should be carefully reviewed to use _killable waiting as much as possible and reduce allocations requests done with the lock held to absolute minimum to reduce the risk even further. The API between oom killer and oom reaper is quite trivial. wake_oom_reaper updates mm_to_reap with cmpxchg to guarantee only NULL->mm transition and oom_reaper clear this atomically once it is done with the work. This means that only a single mm_struct can be reaped at the time. As the operation is potentially disruptive we are trying to limit it to the ncessary minimum and the reaper blocks any updates while it operates on an mm. mm_struct is pinned by mm_count to allow parallel exit_mmap and a race is detected by atomic_inc_not_zero(mm_users). Signed-off-by: Michal Hocko <mhocko@suse.com> Suggested-by: Oleg Nesterov <oleg@redhat.com> Suggested-by: Mel Gorman <mgorman@suse.de> Acked-by: Mel Gorman <mgorman@suse.de> Acked-by: David Rientjes <rientjes@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Hugh Dickins <hughd@google.com> Cc: Andrea Argangeli <andrea@kernel.org> Cc: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:24 +08:00
mm, oom_reaper: implement OOM victims queuing wake_oom_reaper has allowed only 1 oom victim to be queued. The main reason for that was the simplicity as other solutions would require some way of queuing. The current approach is racy and that was deemed sufficient as the oom_reaper is considered a best effort approach to help with oom handling when the OOM victim cannot terminate in a reasonable time. The race could lead to missing an oom victim which can get stuck out_of_memory wake_oom_reaper cmpxchg // OK oom_reaper oom_reap_task __oom_reap_task oom_victim terminates atomic_inc_not_zero // fail out_of_memory wake_oom_reaper cmpxchg // fails task_to_reap = NULL This race requires 2 OOM invocations in a short time period which is not very likely but certainly not impossible. E.g. the original victim might have not released a lot of memory for some reason. The situation would improve considerably if wake_oom_reaper used a more robust queuing. This is what this patch implements. This means adding oom_reaper_list list_head into task_struct (eat a hole before embeded thread_struct for that purpose) and a oom_reaper_lock spinlock for queuing synchronization. wake_oom_reaper will then add the task on the queue and oom_reaper will dequeue it. Signed-off-by: Michal Hocko <mhocko@suse.com> Cc: Vladimir Davydov <vdavydov@virtuozzo.com> Cc: Andrea Argangeli <andrea@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Hugh Dickins <hughd@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Mel Gorman <mgorman@suse.de> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Rik van Riel <riel@redhat.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:33 +08:00
spin_lock(&oom_reaper_lock);
tsk->oom_reaper_list = oom_reaper_list;
oom_reaper_list = tsk;
mm, oom_reaper: implement OOM victims queuing wake_oom_reaper has allowed only 1 oom victim to be queued. The main reason for that was the simplicity as other solutions would require some way of queuing. The current approach is racy and that was deemed sufficient as the oom_reaper is considered a best effort approach to help with oom handling when the OOM victim cannot terminate in a reasonable time. The race could lead to missing an oom victim which can get stuck out_of_memory wake_oom_reaper cmpxchg // OK oom_reaper oom_reap_task __oom_reap_task oom_victim terminates atomic_inc_not_zero // fail out_of_memory wake_oom_reaper cmpxchg // fails task_to_reap = NULL This race requires 2 OOM invocations in a short time period which is not very likely but certainly not impossible. E.g. the original victim might have not released a lot of memory for some reason. The situation would improve considerably if wake_oom_reaper used a more robust queuing. This is what this patch implements. This means adding oom_reaper_list list_head into task_struct (eat a hole before embeded thread_struct for that purpose) and a oom_reaper_lock spinlock for queuing synchronization. wake_oom_reaper will then add the task on the queue and oom_reaper will dequeue it. Signed-off-by: Michal Hocko <mhocko@suse.com> Cc: Vladimir Davydov <vdavydov@virtuozzo.com> Cc: Andrea Argangeli <andrea@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Hugh Dickins <hughd@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Mel Gorman <mgorman@suse.de> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Rik van Riel <riel@redhat.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:33 +08:00
spin_unlock(&oom_reaper_lock);
mm/oom_kill.c: add tracepoints for oom reaper-related events During the debugging of the problem described in https://lkml.org/lkml/2017/5/17/542 and fixed by Tetsuo Handa in https://lkml.org/lkml/2017/5/19/383 , I've found that the existing debug output is not really useful to understand issues related to the oom reaper. So, I assume, that adding some tracepoints might help with debugging of similar issues. Trace the following events: 1) a process is marked as an oom victim, 2) a process is added to the oom reaper list, 3) the oom reaper starts reaping process's mm, 4) the oom reaper finished reaping, 5) the oom reaper skips reaping. How it works in practice? Below is an example which show how the problem mentioned above can be found: one process is added twice to the oom_reaper list: $ cd /sys/kernel/debug/tracing $ echo "oom:mark_victim" > set_event $ echo "oom:wake_reaper" >> set_event $ echo "oom:skip_task_reaping" >> set_event $ echo "oom:start_task_reaping" >> set_event $ echo "oom:finish_task_reaping" >> set_event $ cat trace_pipe allocate-502 [001] .... 91.836405: mark_victim: pid=502 allocate-502 [001] .N.. 91.837356: wake_reaper: pid=502 allocate-502 [000] .N.. 91.871149: wake_reaper: pid=502 oom_reaper-23 [000] .... 91.871177: start_task_reaping: pid=502 oom_reaper-23 [000] .N.. 91.879511: finish_task_reaping: pid=502 oom_reaper-23 [000] .... 91.879580: skip_task_reaping: pid=502 Link: http://lkml.kernel.org/r/20170530185231.GA13412@castle Signed-off-by: Roman Gushchin <guro@fb.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-07-11 06:49:05 +08:00
trace_wake_reaper(tsk->pid);
mm, oom_reaper: implement OOM victims queuing wake_oom_reaper has allowed only 1 oom victim to be queued. The main reason for that was the simplicity as other solutions would require some way of queuing. The current approach is racy and that was deemed sufficient as the oom_reaper is considered a best effort approach to help with oom handling when the OOM victim cannot terminate in a reasonable time. The race could lead to missing an oom victim which can get stuck out_of_memory wake_oom_reaper cmpxchg // OK oom_reaper oom_reap_task __oom_reap_task oom_victim terminates atomic_inc_not_zero // fail out_of_memory wake_oom_reaper cmpxchg // fails task_to_reap = NULL This race requires 2 OOM invocations in a short time period which is not very likely but certainly not impossible. E.g. the original victim might have not released a lot of memory for some reason. The situation would improve considerably if wake_oom_reaper used a more robust queuing. This is what this patch implements. This means adding oom_reaper_list list_head into task_struct (eat a hole before embeded thread_struct for that purpose) and a oom_reaper_lock spinlock for queuing synchronization. wake_oom_reaper will then add the task on the queue and oom_reaper will dequeue it. Signed-off-by: Michal Hocko <mhocko@suse.com> Cc: Vladimir Davydov <vdavydov@virtuozzo.com> Cc: Andrea Argangeli <andrea@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Hugh Dickins <hughd@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Mel Gorman <mgorman@suse.de> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Rik van Riel <riel@redhat.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:33 +08:00
wake_up(&oom_reaper_wait);
mm, oom: introduce oom reaper This patch (of 5): This is based on the idea from Mel Gorman discussed during LSFMM 2015 and independently brought up by Oleg Nesterov. The OOM killer currently allows to kill only a single task in a good hope that the task will terminate in a reasonable time and frees up its memory. Such a task (oom victim) will get an access to memory reserves via mark_oom_victim to allow a forward progress should there be a need for additional memory during exit path. It has been shown (e.g. by Tetsuo Handa) that it is not that hard to construct workloads which break the core assumption mentioned above and the OOM victim might take unbounded amount of time to exit because it might be blocked in the uninterruptible state waiting for an event (e.g. lock) which is blocked by another task looping in the page allocator. This patch reduces the probability of such a lockup by introducing a specialized kernel thread (oom_reaper) which tries to reclaim additional memory by preemptively reaping the anonymous or swapped out memory owned by the oom victim under an assumption that such a memory won't be needed when its owner is killed and kicked from the userspace anyway. There is one notable exception to this, though, if the OOM victim was in the process of coredumping the result would be incomplete. This is considered a reasonable constrain because the overall system health is more important than debugability of a particular application. A kernel thread has been chosen because we need a reliable way of invocation so workqueue context is not appropriate because all the workers might be busy (e.g. allocating memory). Kswapd which sounds like another good fit is not appropriate as well because it might get blocked on locks during reclaim as well. oom_reaper has to take mmap_sem on the target task for reading so the solution is not 100% because the semaphore might be held or blocked for write but the probability is reduced considerably wrt. basically any lock blocking forward progress as described above. In order to prevent from blocking on the lock without any forward progress we are using only a trylock and retry 10 times with a short sleep in between. Users of mmap_sem which need it for write should be carefully reviewed to use _killable waiting as much as possible and reduce allocations requests done with the lock held to absolute minimum to reduce the risk even further. The API between oom killer and oom reaper is quite trivial. wake_oom_reaper updates mm_to_reap with cmpxchg to guarantee only NULL->mm transition and oom_reaper clear this atomically once it is done with the work. This means that only a single mm_struct can be reaped at the time. As the operation is potentially disruptive we are trying to limit it to the ncessary minimum and the reaper blocks any updates while it operates on an mm. mm_struct is pinned by mm_count to allow parallel exit_mmap and a race is detected by atomic_inc_not_zero(mm_users). Signed-off-by: Michal Hocko <mhocko@suse.com> Suggested-by: Oleg Nesterov <oleg@redhat.com> Suggested-by: Mel Gorman <mgorman@suse.de> Acked-by: Mel Gorman <mgorman@suse.de> Acked-by: David Rientjes <rientjes@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Hugh Dickins <hughd@google.com> Cc: Andrea Argangeli <andrea@kernel.org> Cc: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:24 +08:00
}
static int __init oom_init(void)
{
oom_reaper_th = kthread_run(oom_reaper, NULL, "oom_reaper");
return 0;
}
subsys_initcall(oom_init)
#else
static inline void wake_oom_reaper(struct task_struct *tsk)
{
}
#endif /* CONFIG_MMU */
mm, oom: introduce oom reaper This patch (of 5): This is based on the idea from Mel Gorman discussed during LSFMM 2015 and independently brought up by Oleg Nesterov. The OOM killer currently allows to kill only a single task in a good hope that the task will terminate in a reasonable time and frees up its memory. Such a task (oom victim) will get an access to memory reserves via mark_oom_victim to allow a forward progress should there be a need for additional memory during exit path. It has been shown (e.g. by Tetsuo Handa) that it is not that hard to construct workloads which break the core assumption mentioned above and the OOM victim might take unbounded amount of time to exit because it might be blocked in the uninterruptible state waiting for an event (e.g. lock) which is blocked by another task looping in the page allocator. This patch reduces the probability of such a lockup by introducing a specialized kernel thread (oom_reaper) which tries to reclaim additional memory by preemptively reaping the anonymous or swapped out memory owned by the oom victim under an assumption that such a memory won't be needed when its owner is killed and kicked from the userspace anyway. There is one notable exception to this, though, if the OOM victim was in the process of coredumping the result would be incomplete. This is considered a reasonable constrain because the overall system health is more important than debugability of a particular application. A kernel thread has been chosen because we need a reliable way of invocation so workqueue context is not appropriate because all the workers might be busy (e.g. allocating memory). Kswapd which sounds like another good fit is not appropriate as well because it might get blocked on locks during reclaim as well. oom_reaper has to take mmap_sem on the target task for reading so the solution is not 100% because the semaphore might be held or blocked for write but the probability is reduced considerably wrt. basically any lock blocking forward progress as described above. In order to prevent from blocking on the lock without any forward progress we are using only a trylock and retry 10 times with a short sleep in between. Users of mmap_sem which need it for write should be carefully reviewed to use _killable waiting as much as possible and reduce allocations requests done with the lock held to absolute minimum to reduce the risk even further. The API between oom killer and oom reaper is quite trivial. wake_oom_reaper updates mm_to_reap with cmpxchg to guarantee only NULL->mm transition and oom_reaper clear this atomically once it is done with the work. This means that only a single mm_struct can be reaped at the time. As the operation is potentially disruptive we are trying to limit it to the ncessary minimum and the reaper blocks any updates while it operates on an mm. mm_struct is pinned by mm_count to allow parallel exit_mmap and a race is detected by atomic_inc_not_zero(mm_users). Signed-off-by: Michal Hocko <mhocko@suse.com> Suggested-by: Oleg Nesterov <oleg@redhat.com> Suggested-by: Mel Gorman <mgorman@suse.de> Acked-by: Mel Gorman <mgorman@suse.de> Acked-by: David Rientjes <rientjes@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Hugh Dickins <hughd@google.com> Cc: Andrea Argangeli <andrea@kernel.org> Cc: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:24 +08:00
oom: add helpers for setting and clearing TIF_MEMDIE This patchset addresses a race which was described in the changelog for 5695be142e20 ("OOM, PM: OOM killed task shouldn't escape PM suspend"): : PM freezer relies on having all tasks frozen by the time devices are : getting frozen so that no task will touch them while they are getting : frozen. But OOM killer is allowed to kill an already frozen task in order : to handle OOM situtation. In order to protect from late wake ups OOM : killer is disabled after all tasks are frozen. This, however, still keeps : a window open when a killed task didn't manage to die by the time : freeze_processes finishes. The original patch hasn't closed the race window completely because that would require a more complex solution as it can be seen by this patchset. The primary motivation was to close the race condition between OOM killer and PM freezer _completely_. As Tejun pointed out, even though the race condition is unlikely the harder it would be to debug weird bugs deep in the PM freezer when the debugging options are reduced considerably. I can only speculate what might happen when a task is still runnable unexpectedly. On a plus side and as a side effect the oom enable/disable has a better (full barrier) semantic without polluting hot paths. I have tested the series in KVM with 100M RAM: - many small tasks (20M anon mmap) which are triggering OOM continually - s2ram which resumes automatically is triggered in a loop echo processors > /sys/power/pm_test while true do echo mem > /sys/power/state sleep 1s done - simple module which allocates and frees 20M in 8K chunks. If it sees freezing(current) then it tries another round of allocation before calling try_to_freeze - debugging messages of PM stages and OOM killer enable/disable/fail added and unmark_oom_victim is delayed by 1s after it clears TIF_MEMDIE and before it wakes up waiters. - rebased on top of the current mmotm which means some necessary updates in mm/oom_kill.c. mark_tsk_oom_victim is now called under task_lock but I think this should be OK because __thaw_task shouldn't interfere with any locking down wake_up_process. Oleg? As expected there are no OOM killed tasks after oom is disabled and allocations requested by the kernel thread are failing after all the tasks are frozen and OOM disabled. I wasn't able to catch a race where oom_killer_disable would really have to wait but I kinda expected the race is really unlikely. [ 242.609330] Killed process 2992 (mem_eater) total-vm:24412kB, anon-rss:2164kB, file-rss:4kB [ 243.628071] Unmarking 2992 OOM victim. oom_victims: 1 [ 243.636072] (elapsed 2.837 seconds) done. [ 243.641985] Trying to disable OOM killer [ 243.643032] Waiting for concurent OOM victims [ 243.644342] OOM killer disabled [ 243.645447] Freezing remaining freezable tasks ... (elapsed 0.005 seconds) done. [ 243.652983] Suspending console(s) (use no_console_suspend to debug) [ 243.903299] kmem_eater: page allocation failure: order:1, mode:0x204010 [...] [ 243.992600] PM: suspend of devices complete after 336.667 msecs [ 243.993264] PM: late suspend of devices complete after 0.660 msecs [ 243.994713] PM: noirq suspend of devices complete after 1.446 msecs [ 243.994717] ACPI: Preparing to enter system sleep state S3 [ 243.994795] PM: Saving platform NVS memory [ 243.994796] Disabling non-boot CPUs ... The first 2 patches are simple cleanups for OOM. They should go in regardless the rest IMO. Patches 3 and 4 are trivial printk -> pr_info conversion and they should go in ditto. The main patch is the last one and I would appreciate acks from Tejun and Rafael. I think the OOM part should be OK (except for __thaw_task vs. task_lock where a look from Oleg would appreciated) but I am not so sure I haven't screwed anything in the freezer code. I have found several surprises there. This patch (of 5): This patch is just a preparatory and it doesn't introduce any functional change. Note: I am utterly unhappy about lowmemory killer abusing TIF_MEMDIE just to wait for the oom victim and to prevent from new killing. This is just a side effect of the flag. The primary meaning is to give the oom victim access to the memory reserves and that shouldn't be necessary here. Signed-off-by: Michal Hocko <mhocko@suse.cz> Cc: Tejun Heo <tj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-02-12 07:26:12 +08:00
/**
* mark_oom_victim - mark the given task as OOM victim
oom: add helpers for setting and clearing TIF_MEMDIE This patchset addresses a race which was described in the changelog for 5695be142e20 ("OOM, PM: OOM killed task shouldn't escape PM suspend"): : PM freezer relies on having all tasks frozen by the time devices are : getting frozen so that no task will touch them while they are getting : frozen. But OOM killer is allowed to kill an already frozen task in order : to handle OOM situtation. In order to protect from late wake ups OOM : killer is disabled after all tasks are frozen. This, however, still keeps : a window open when a killed task didn't manage to die by the time : freeze_processes finishes. The original patch hasn't closed the race window completely because that would require a more complex solution as it can be seen by this patchset. The primary motivation was to close the race condition between OOM killer and PM freezer _completely_. As Tejun pointed out, even though the race condition is unlikely the harder it would be to debug weird bugs deep in the PM freezer when the debugging options are reduced considerably. I can only speculate what might happen when a task is still runnable unexpectedly. On a plus side and as a side effect the oom enable/disable has a better (full barrier) semantic without polluting hot paths. I have tested the series in KVM with 100M RAM: - many small tasks (20M anon mmap) which are triggering OOM continually - s2ram which resumes automatically is triggered in a loop echo processors > /sys/power/pm_test while true do echo mem > /sys/power/state sleep 1s done - simple module which allocates and frees 20M in 8K chunks. If it sees freezing(current) then it tries another round of allocation before calling try_to_freeze - debugging messages of PM stages and OOM killer enable/disable/fail added and unmark_oom_victim is delayed by 1s after it clears TIF_MEMDIE and before it wakes up waiters. - rebased on top of the current mmotm which means some necessary updates in mm/oom_kill.c. mark_tsk_oom_victim is now called under task_lock but I think this should be OK because __thaw_task shouldn't interfere with any locking down wake_up_process. Oleg? As expected there are no OOM killed tasks after oom is disabled and allocations requested by the kernel thread are failing after all the tasks are frozen and OOM disabled. I wasn't able to catch a race where oom_killer_disable would really have to wait but I kinda expected the race is really unlikely. [ 242.609330] Killed process 2992 (mem_eater) total-vm:24412kB, anon-rss:2164kB, file-rss:4kB [ 243.628071] Unmarking 2992 OOM victim. oom_victims: 1 [ 243.636072] (elapsed 2.837 seconds) done. [ 243.641985] Trying to disable OOM killer [ 243.643032] Waiting for concurent OOM victims [ 243.644342] OOM killer disabled [ 243.645447] Freezing remaining freezable tasks ... (elapsed 0.005 seconds) done. [ 243.652983] Suspending console(s) (use no_console_suspend to debug) [ 243.903299] kmem_eater: page allocation failure: order:1, mode:0x204010 [...] [ 243.992600] PM: suspend of devices complete after 336.667 msecs [ 243.993264] PM: late suspend of devices complete after 0.660 msecs [ 243.994713] PM: noirq suspend of devices complete after 1.446 msecs [ 243.994717] ACPI: Preparing to enter system sleep state S3 [ 243.994795] PM: Saving platform NVS memory [ 243.994796] Disabling non-boot CPUs ... The first 2 patches are simple cleanups for OOM. They should go in regardless the rest IMO. Patches 3 and 4 are trivial printk -> pr_info conversion and they should go in ditto. The main patch is the last one and I would appreciate acks from Tejun and Rafael. I think the OOM part should be OK (except for __thaw_task vs. task_lock where a look from Oleg would appreciated) but I am not so sure I haven't screwed anything in the freezer code. I have found several surprises there. This patch (of 5): This patch is just a preparatory and it doesn't introduce any functional change. Note: I am utterly unhappy about lowmemory killer abusing TIF_MEMDIE just to wait for the oom victim and to prevent from new killing. This is just a side effect of the flag. The primary meaning is to give the oom victim access to the memory reserves and that shouldn't be necessary here. Signed-off-by: Michal Hocko <mhocko@suse.cz> Cc: Tejun Heo <tj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-02-12 07:26:12 +08:00
* @tsk: task to mark
oom, PM: make OOM detection in the freezer path raceless Commit 5695be142e20 ("OOM, PM: OOM killed task shouldn't escape PM suspend") has left a race window when OOM killer manages to note_oom_kill after freeze_processes checks the counter. The race window is quite small and really unlikely and partial solution deemed sufficient at the time of submission. Tejun wasn't happy about this partial solution though and insisted on a full solution. That requires the full OOM and freezer's task freezing exclusion, though. This is done by this patch which introduces oom_sem RW lock and turns oom_killer_disable() into a full OOM barrier. oom_killer_disabled check is moved from the allocation path to the OOM level and we take oom_sem for reading for both the check and the whole OOM invocation. oom_killer_disable() takes oom_sem for writing so it waits for all currently running OOM killer invocations. Then it disable all the further OOMs by setting oom_killer_disabled and checks for any oom victims. Victims are counted via mark_tsk_oom_victim resp. unmark_oom_victim. The last victim wakes up all waiters enqueued by oom_killer_disable(). Therefore this function acts as the full OOM barrier. The page fault path is covered now as well although it was assumed to be safe before. As per Tejun, "We used to have freezing points deep in file system code which may be reacheable from page fault." so it would be better and more robust to not rely on freezing points here. Same applies to the memcg OOM killer. out_of_memory tells the caller whether the OOM was allowed to trigger and the callers are supposed to handle the situation. The page allocation path simply fails the allocation same as before. The page fault path will retry the fault (more on that later) and Sysrq OOM trigger will simply complain to the log. Normally there wouldn't be any unfrozen user tasks after try_to_freeze_tasks so the function will not block. But if there was an OOM killer racing with try_to_freeze_tasks and the OOM victim didn't finish yet then we have to wait for it. This should complete in a finite time, though, because - the victim cannot loop in the page fault handler (it would die on the way out from the exception) - it cannot loop in the page allocator because all the further allocation would fail and __GFP_NOFAIL allocations are not acceptable at this stage - it shouldn't be blocked on any locks held by frozen tasks (try_to_freeze expects lockless context) and kernel threads and work queues are not frozen yet Signed-off-by: Michal Hocko <mhocko@suse.cz> Suggested-by: Tejun Heo <tj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-02-12 07:26:24 +08:00
*
* Has to be called with oom_lock held and never after
oom, PM: make OOM detection in the freezer path raceless Commit 5695be142e20 ("OOM, PM: OOM killed task shouldn't escape PM suspend") has left a race window when OOM killer manages to note_oom_kill after freeze_processes checks the counter. The race window is quite small and really unlikely and partial solution deemed sufficient at the time of submission. Tejun wasn't happy about this partial solution though and insisted on a full solution. That requires the full OOM and freezer's task freezing exclusion, though. This is done by this patch which introduces oom_sem RW lock and turns oom_killer_disable() into a full OOM barrier. oom_killer_disabled check is moved from the allocation path to the OOM level and we take oom_sem for reading for both the check and the whole OOM invocation. oom_killer_disable() takes oom_sem for writing so it waits for all currently running OOM killer invocations. Then it disable all the further OOMs by setting oom_killer_disabled and checks for any oom victims. Victims are counted via mark_tsk_oom_victim resp. unmark_oom_victim. The last victim wakes up all waiters enqueued by oom_killer_disable(). Therefore this function acts as the full OOM barrier. The page fault path is covered now as well although it was assumed to be safe before. As per Tejun, "We used to have freezing points deep in file system code which may be reacheable from page fault." so it would be better and more robust to not rely on freezing points here. Same applies to the memcg OOM killer. out_of_memory tells the caller whether the OOM was allowed to trigger and the callers are supposed to handle the situation. The page allocation path simply fails the allocation same as before. The page fault path will retry the fault (more on that later) and Sysrq OOM trigger will simply complain to the log. Normally there wouldn't be any unfrozen user tasks after try_to_freeze_tasks so the function will not block. But if there was an OOM killer racing with try_to_freeze_tasks and the OOM victim didn't finish yet then we have to wait for it. This should complete in a finite time, though, because - the victim cannot loop in the page fault handler (it would die on the way out from the exception) - it cannot loop in the page allocator because all the further allocation would fail and __GFP_NOFAIL allocations are not acceptable at this stage - it shouldn't be blocked on any locks held by frozen tasks (try_to_freeze expects lockless context) and kernel threads and work queues are not frozen yet Signed-off-by: Michal Hocko <mhocko@suse.cz> Suggested-by: Tejun Heo <tj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-02-12 07:26:24 +08:00
* oom has been disabled already.
oom: keep mm of the killed task available oom_reap_task has to call exit_oom_victim in order to make sure that the oom vicim will not block the oom killer for ever. This is, however, opening new problems (e.g oom_killer_disable exclusion - see commit 74070542099c ("oom, suspend: fix oom_reaper vs. oom_killer_disable race")). exit_oom_victim should be only called from the victim's context ideally. One way to achieve this would be to rely on per mm_struct flags. We already have MMF_OOM_REAPED to hide a task from the oom killer since "mm, oom: hide mm which is shared with kthread or global init". The problem is that the exit path: do_exit exit_mm tsk->mm = NULL; mmput __mmput exit_oom_victim doesn't guarantee that exit_oom_victim will get called in a bounded amount of time. At least exit_aio depends on IO which might get blocked due to lack of memory and who knows what else is lurking there. This patch takes a different approach. We remember tsk->mm into the signal_struct and bind it to the signal struct life time for all oom victims. __oom_reap_task_mm as well as oom_scan_process_thread do not have to rely on find_lock_task_mm anymore and they will have a reliable reference to the mm struct. As a result all the oom specific communication inside the OOM killer can be done via tsk->signal->oom_mm. Increasing the signal_struct for something as unlikely as the oom killer is far from ideal but this approach will make the code much more reasonable and long term we even might want to move task->mm into the signal_struct anyway. In the next step we might want to make the oom killer exclusion and access to memory reserves completely independent which would be also nice. Link: http://lkml.kernel.org/r/1472119394-11342-4-git-send-email-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: David Rientjes <rientjes@google.com> Cc: Vladimir Davydov <vdavydov@parallels.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-10-08 07:58:51 +08:00
*
* tsk->mm has to be non NULL and caller has to guarantee it is stable (either
* under task_lock or operate on the current).
oom: add helpers for setting and clearing TIF_MEMDIE This patchset addresses a race which was described in the changelog for 5695be142e20 ("OOM, PM: OOM killed task shouldn't escape PM suspend"): : PM freezer relies on having all tasks frozen by the time devices are : getting frozen so that no task will touch them while they are getting : frozen. But OOM killer is allowed to kill an already frozen task in order : to handle OOM situtation. In order to protect from late wake ups OOM : killer is disabled after all tasks are frozen. This, however, still keeps : a window open when a killed task didn't manage to die by the time : freeze_processes finishes. The original patch hasn't closed the race window completely because that would require a more complex solution as it can be seen by this patchset. The primary motivation was to close the race condition between OOM killer and PM freezer _completely_. As Tejun pointed out, even though the race condition is unlikely the harder it would be to debug weird bugs deep in the PM freezer when the debugging options are reduced considerably. I can only speculate what might happen when a task is still runnable unexpectedly. On a plus side and as a side effect the oom enable/disable has a better (full barrier) semantic without polluting hot paths. I have tested the series in KVM with 100M RAM: - many small tasks (20M anon mmap) which are triggering OOM continually - s2ram which resumes automatically is triggered in a loop echo processors > /sys/power/pm_test while true do echo mem > /sys/power/state sleep 1s done - simple module which allocates and frees 20M in 8K chunks. If it sees freezing(current) then it tries another round of allocation before calling try_to_freeze - debugging messages of PM stages and OOM killer enable/disable/fail added and unmark_oom_victim is delayed by 1s after it clears TIF_MEMDIE and before it wakes up waiters. - rebased on top of the current mmotm which means some necessary updates in mm/oom_kill.c. mark_tsk_oom_victim is now called under task_lock but I think this should be OK because __thaw_task shouldn't interfere with any locking down wake_up_process. Oleg? As expected there are no OOM killed tasks after oom is disabled and allocations requested by the kernel thread are failing after all the tasks are frozen and OOM disabled. I wasn't able to catch a race where oom_killer_disable would really have to wait but I kinda expected the race is really unlikely. [ 242.609330] Killed process 2992 (mem_eater) total-vm:24412kB, anon-rss:2164kB, file-rss:4kB [ 243.628071] Unmarking 2992 OOM victim. oom_victims: 1 [ 243.636072] (elapsed 2.837 seconds) done. [ 243.641985] Trying to disable OOM killer [ 243.643032] Waiting for concurent OOM victims [ 243.644342] OOM killer disabled [ 243.645447] Freezing remaining freezable tasks ... (elapsed 0.005 seconds) done. [ 243.652983] Suspending console(s) (use no_console_suspend to debug) [ 243.903299] kmem_eater: page allocation failure: order:1, mode:0x204010 [...] [ 243.992600] PM: suspend of devices complete after 336.667 msecs [ 243.993264] PM: late suspend of devices complete after 0.660 msecs [ 243.994713] PM: noirq suspend of devices complete after 1.446 msecs [ 243.994717] ACPI: Preparing to enter system sleep state S3 [ 243.994795] PM: Saving platform NVS memory [ 243.994796] Disabling non-boot CPUs ... The first 2 patches are simple cleanups for OOM. They should go in regardless the rest IMO. Patches 3 and 4 are trivial printk -> pr_info conversion and they should go in ditto. The main patch is the last one and I would appreciate acks from Tejun and Rafael. I think the OOM part should be OK (except for __thaw_task vs. task_lock where a look from Oleg would appreciated) but I am not so sure I haven't screwed anything in the freezer code. I have found several surprises there. This patch (of 5): This patch is just a preparatory and it doesn't introduce any functional change. Note: I am utterly unhappy about lowmemory killer abusing TIF_MEMDIE just to wait for the oom victim and to prevent from new killing. This is just a side effect of the flag. The primary meaning is to give the oom victim access to the memory reserves and that shouldn't be necessary here. Signed-off-by: Michal Hocko <mhocko@suse.cz> Cc: Tejun Heo <tj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-02-12 07:26:12 +08:00
*/
static void mark_oom_victim(struct task_struct *tsk)
oom: add helpers for setting and clearing TIF_MEMDIE This patchset addresses a race which was described in the changelog for 5695be142e20 ("OOM, PM: OOM killed task shouldn't escape PM suspend"): : PM freezer relies on having all tasks frozen by the time devices are : getting frozen so that no task will touch them while they are getting : frozen. But OOM killer is allowed to kill an already frozen task in order : to handle OOM situtation. In order to protect from late wake ups OOM : killer is disabled after all tasks are frozen. This, however, still keeps : a window open when a killed task didn't manage to die by the time : freeze_processes finishes. The original patch hasn't closed the race window completely because that would require a more complex solution as it can be seen by this patchset. The primary motivation was to close the race condition between OOM killer and PM freezer _completely_. As Tejun pointed out, even though the race condition is unlikely the harder it would be to debug weird bugs deep in the PM freezer when the debugging options are reduced considerably. I can only speculate what might happen when a task is still runnable unexpectedly. On a plus side and as a side effect the oom enable/disable has a better (full barrier) semantic without polluting hot paths. I have tested the series in KVM with 100M RAM: - many small tasks (20M anon mmap) which are triggering OOM continually - s2ram which resumes automatically is triggered in a loop echo processors > /sys/power/pm_test while true do echo mem > /sys/power/state sleep 1s done - simple module which allocates and frees 20M in 8K chunks. If it sees freezing(current) then it tries another round of allocation before calling try_to_freeze - debugging messages of PM stages and OOM killer enable/disable/fail added and unmark_oom_victim is delayed by 1s after it clears TIF_MEMDIE and before it wakes up waiters. - rebased on top of the current mmotm which means some necessary updates in mm/oom_kill.c. mark_tsk_oom_victim is now called under task_lock but I think this should be OK because __thaw_task shouldn't interfere with any locking down wake_up_process. Oleg? As expected there are no OOM killed tasks after oom is disabled and allocations requested by the kernel thread are failing after all the tasks are frozen and OOM disabled. I wasn't able to catch a race where oom_killer_disable would really have to wait but I kinda expected the race is really unlikely. [ 242.609330] Killed process 2992 (mem_eater) total-vm:24412kB, anon-rss:2164kB, file-rss:4kB [ 243.628071] Unmarking 2992 OOM victim. oom_victims: 1 [ 243.636072] (elapsed 2.837 seconds) done. [ 243.641985] Trying to disable OOM killer [ 243.643032] Waiting for concurent OOM victims [ 243.644342] OOM killer disabled [ 243.645447] Freezing remaining freezable tasks ... (elapsed 0.005 seconds) done. [ 243.652983] Suspending console(s) (use no_console_suspend to debug) [ 243.903299] kmem_eater: page allocation failure: order:1, mode:0x204010 [...] [ 243.992600] PM: suspend of devices complete after 336.667 msecs [ 243.993264] PM: late suspend of devices complete after 0.660 msecs [ 243.994713] PM: noirq suspend of devices complete after 1.446 msecs [ 243.994717] ACPI: Preparing to enter system sleep state S3 [ 243.994795] PM: Saving platform NVS memory [ 243.994796] Disabling non-boot CPUs ... The first 2 patches are simple cleanups for OOM. They should go in regardless the rest IMO. Patches 3 and 4 are trivial printk -> pr_info conversion and they should go in ditto. The main patch is the last one and I would appreciate acks from Tejun and Rafael. I think the OOM part should be OK (except for __thaw_task vs. task_lock where a look from Oleg would appreciated) but I am not so sure I haven't screwed anything in the freezer code. I have found several surprises there. This patch (of 5): This patch is just a preparatory and it doesn't introduce any functional change. Note: I am utterly unhappy about lowmemory killer abusing TIF_MEMDIE just to wait for the oom victim and to prevent from new killing. This is just a side effect of the flag. The primary meaning is to give the oom victim access to the memory reserves and that shouldn't be necessary here. Signed-off-by: Michal Hocko <mhocko@suse.cz> Cc: Tejun Heo <tj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-02-12 07:26:12 +08:00
{
oom: keep mm of the killed task available oom_reap_task has to call exit_oom_victim in order to make sure that the oom vicim will not block the oom killer for ever. This is, however, opening new problems (e.g oom_killer_disable exclusion - see commit 74070542099c ("oom, suspend: fix oom_reaper vs. oom_killer_disable race")). exit_oom_victim should be only called from the victim's context ideally. One way to achieve this would be to rely on per mm_struct flags. We already have MMF_OOM_REAPED to hide a task from the oom killer since "mm, oom: hide mm which is shared with kthread or global init". The problem is that the exit path: do_exit exit_mm tsk->mm = NULL; mmput __mmput exit_oom_victim doesn't guarantee that exit_oom_victim will get called in a bounded amount of time. At least exit_aio depends on IO which might get blocked due to lack of memory and who knows what else is lurking there. This patch takes a different approach. We remember tsk->mm into the signal_struct and bind it to the signal struct life time for all oom victims. __oom_reap_task_mm as well as oom_scan_process_thread do not have to rely on find_lock_task_mm anymore and they will have a reliable reference to the mm struct. As a result all the oom specific communication inside the OOM killer can be done via tsk->signal->oom_mm. Increasing the signal_struct for something as unlikely as the oom killer is far from ideal but this approach will make the code much more reasonable and long term we even might want to move task->mm into the signal_struct anyway. In the next step we might want to make the oom killer exclusion and access to memory reserves completely independent which would be also nice. Link: http://lkml.kernel.org/r/1472119394-11342-4-git-send-email-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: David Rientjes <rientjes@google.com> Cc: Vladimir Davydov <vdavydov@parallels.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-10-08 07:58:51 +08:00
struct mm_struct *mm = tsk->mm;
oom, PM: make OOM detection in the freezer path raceless Commit 5695be142e20 ("OOM, PM: OOM killed task shouldn't escape PM suspend") has left a race window when OOM killer manages to note_oom_kill after freeze_processes checks the counter. The race window is quite small and really unlikely and partial solution deemed sufficient at the time of submission. Tejun wasn't happy about this partial solution though and insisted on a full solution. That requires the full OOM and freezer's task freezing exclusion, though. This is done by this patch which introduces oom_sem RW lock and turns oom_killer_disable() into a full OOM barrier. oom_killer_disabled check is moved from the allocation path to the OOM level and we take oom_sem for reading for both the check and the whole OOM invocation. oom_killer_disable() takes oom_sem for writing so it waits for all currently running OOM killer invocations. Then it disable all the further OOMs by setting oom_killer_disabled and checks for any oom victims. Victims are counted via mark_tsk_oom_victim resp. unmark_oom_victim. The last victim wakes up all waiters enqueued by oom_killer_disable(). Therefore this function acts as the full OOM barrier. The page fault path is covered now as well although it was assumed to be safe before. As per Tejun, "We used to have freezing points deep in file system code which may be reacheable from page fault." so it would be better and more robust to not rely on freezing points here. Same applies to the memcg OOM killer. out_of_memory tells the caller whether the OOM was allowed to trigger and the callers are supposed to handle the situation. The page allocation path simply fails the allocation same as before. The page fault path will retry the fault (more on that later) and Sysrq OOM trigger will simply complain to the log. Normally there wouldn't be any unfrozen user tasks after try_to_freeze_tasks so the function will not block. But if there was an OOM killer racing with try_to_freeze_tasks and the OOM victim didn't finish yet then we have to wait for it. This should complete in a finite time, though, because - the victim cannot loop in the page fault handler (it would die on the way out from the exception) - it cannot loop in the page allocator because all the further allocation would fail and __GFP_NOFAIL allocations are not acceptable at this stage - it shouldn't be blocked on any locks held by frozen tasks (try_to_freeze expects lockless context) and kernel threads and work queues are not frozen yet Signed-off-by: Michal Hocko <mhocko@suse.cz> Suggested-by: Tejun Heo <tj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-02-12 07:26:24 +08:00
WARN_ON(oom_killer_disabled);
/* OOM killer might race with memcg OOM */
if (test_and_set_tsk_thread_flag(tsk, TIF_MEMDIE))
return;
oom: keep mm of the killed task available oom_reap_task has to call exit_oom_victim in order to make sure that the oom vicim will not block the oom killer for ever. This is, however, opening new problems (e.g oom_killer_disable exclusion - see commit 74070542099c ("oom, suspend: fix oom_reaper vs. oom_killer_disable race")). exit_oom_victim should be only called from the victim's context ideally. One way to achieve this would be to rely on per mm_struct flags. We already have MMF_OOM_REAPED to hide a task from the oom killer since "mm, oom: hide mm which is shared with kthread or global init". The problem is that the exit path: do_exit exit_mm tsk->mm = NULL; mmput __mmput exit_oom_victim doesn't guarantee that exit_oom_victim will get called in a bounded amount of time. At least exit_aio depends on IO which might get blocked due to lack of memory and who knows what else is lurking there. This patch takes a different approach. We remember tsk->mm into the signal_struct and bind it to the signal struct life time for all oom victims. __oom_reap_task_mm as well as oom_scan_process_thread do not have to rely on find_lock_task_mm anymore and they will have a reliable reference to the mm struct. As a result all the oom specific communication inside the OOM killer can be done via tsk->signal->oom_mm. Increasing the signal_struct for something as unlikely as the oom killer is far from ideal but this approach will make the code much more reasonable and long term we even might want to move task->mm into the signal_struct anyway. In the next step we might want to make the oom killer exclusion and access to memory reserves completely independent which would be also nice. Link: http://lkml.kernel.org/r/1472119394-11342-4-git-send-email-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: David Rientjes <rientjes@google.com> Cc: Vladimir Davydov <vdavydov@parallels.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-10-08 07:58:51 +08:00
/* oom_mm is bound to the signal struct life time. */
mm, oom_reaper: fix memory corruption David Rientjes has reported the following memory corruption while the oom reaper tries to unmap the victims address space BUG: Bad page map in process oom_reaper pte:6353826300000000 pmd:00000000 addr:00007f50cab1d000 vm_flags:08100073 anon_vma:ffff9eea335603f0 mapping: (null) index:7f50cab1d file: (null) fault: (null) mmap: (null) readpage: (null) CPU: 2 PID: 1001 Comm: oom_reaper Call Trace: unmap_page_range+0x1068/0x1130 __oom_reap_task_mm+0xd5/0x16b oom_reaper+0xff/0x14c kthread+0xc1/0xe0 Tetsuo Handa has noticed that the synchronization inside exit_mmap is insufficient. We only synchronize with the oom reaper if tsk_is_oom_victim which is not true if the final __mmput is called from a different context than the oom victim exit path. This can trivially happen from context of any task which has grabbed mm reference (e.g. to read /proc/<pid>/ file which requires mm etc.). The race would look like this oom_reaper oom_victim task mmget_not_zero do_exit mmput __oom_reap_task_mm mmput __mmput exit_mmap remove_vma unmap_page_range Fix this issue by providing a new mm_is_oom_victim() helper which operates on the mm struct rather than a task. Any context which operates on a remote mm struct should use this helper in place of tsk_is_oom_victim. The flag is set in mark_oom_victim and never cleared so it is stable in the exit_mmap path. Debugged by Tetsuo Handa. Link: http://lkml.kernel.org/r/20171210095130.17110-1-mhocko@kernel.org Fixes: 212925802454 ("mm: oom: let oom_reap_task and exit_mmap run concurrently") Signed-off-by: Michal Hocko <mhocko@suse.com> Reported-by: David Rientjes <rientjes@google.com> Acked-by: David Rientjes <rientjes@google.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Andrea Argangeli <andrea@kernel.org> Cc: <stable@vger.kernel.org> [4.14] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-12-15 07:33:15 +08:00
if (!cmpxchg(&tsk->signal->oom_mm, NULL, mm)) {
mmgrab(tsk->signal->oom_mm);
mm, oom_reaper: fix memory corruption David Rientjes has reported the following memory corruption while the oom reaper tries to unmap the victims address space BUG: Bad page map in process oom_reaper pte:6353826300000000 pmd:00000000 addr:00007f50cab1d000 vm_flags:08100073 anon_vma:ffff9eea335603f0 mapping: (null) index:7f50cab1d file: (null) fault: (null) mmap: (null) readpage: (null) CPU: 2 PID: 1001 Comm: oom_reaper Call Trace: unmap_page_range+0x1068/0x1130 __oom_reap_task_mm+0xd5/0x16b oom_reaper+0xff/0x14c kthread+0xc1/0xe0 Tetsuo Handa has noticed that the synchronization inside exit_mmap is insufficient. We only synchronize with the oom reaper if tsk_is_oom_victim which is not true if the final __mmput is called from a different context than the oom victim exit path. This can trivially happen from context of any task which has grabbed mm reference (e.g. to read /proc/<pid>/ file which requires mm etc.). The race would look like this oom_reaper oom_victim task mmget_not_zero do_exit mmput __oom_reap_task_mm mmput __mmput exit_mmap remove_vma unmap_page_range Fix this issue by providing a new mm_is_oom_victim() helper which operates on the mm struct rather than a task. Any context which operates on a remote mm struct should use this helper in place of tsk_is_oom_victim. The flag is set in mark_oom_victim and never cleared so it is stable in the exit_mmap path. Debugged by Tetsuo Handa. Link: http://lkml.kernel.org/r/20171210095130.17110-1-mhocko@kernel.org Fixes: 212925802454 ("mm: oom: let oom_reap_task and exit_mmap run concurrently") Signed-off-by: Michal Hocko <mhocko@suse.com> Reported-by: David Rientjes <rientjes@google.com> Acked-by: David Rientjes <rientjes@google.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Andrea Argangeli <andrea@kernel.org> Cc: <stable@vger.kernel.org> [4.14] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-12-15 07:33:15 +08:00
set_bit(MMF_OOM_VICTIM, &mm->flags);
}
oom: keep mm of the killed task available oom_reap_task has to call exit_oom_victim in order to make sure that the oom vicim will not block the oom killer for ever. This is, however, opening new problems (e.g oom_killer_disable exclusion - see commit 74070542099c ("oom, suspend: fix oom_reaper vs. oom_killer_disable race")). exit_oom_victim should be only called from the victim's context ideally. One way to achieve this would be to rely on per mm_struct flags. We already have MMF_OOM_REAPED to hide a task from the oom killer since "mm, oom: hide mm which is shared with kthread or global init". The problem is that the exit path: do_exit exit_mm tsk->mm = NULL; mmput __mmput exit_oom_victim doesn't guarantee that exit_oom_victim will get called in a bounded amount of time. At least exit_aio depends on IO which might get blocked due to lack of memory and who knows what else is lurking there. This patch takes a different approach. We remember tsk->mm into the signal_struct and bind it to the signal struct life time for all oom victims. __oom_reap_task_mm as well as oom_scan_process_thread do not have to rely on find_lock_task_mm anymore and they will have a reliable reference to the mm struct. As a result all the oom specific communication inside the OOM killer can be done via tsk->signal->oom_mm. Increasing the signal_struct for something as unlikely as the oom killer is far from ideal but this approach will make the code much more reasonable and long term we even might want to move task->mm into the signal_struct anyway. In the next step we might want to make the oom killer exclusion and access to memory reserves completely independent which would be also nice. Link: http://lkml.kernel.org/r/1472119394-11342-4-git-send-email-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: David Rientjes <rientjes@google.com> Cc: Vladimir Davydov <vdavydov@parallels.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-10-08 07:58:51 +08:00
/*
* Make sure that the task is woken up from uninterruptible sleep
* if it is frozen because OOM killer wouldn't be able to free
* any memory and livelock. freezing_slow_path will tell the freezer
* that TIF_MEMDIE tasks should be ignored.
*/
__thaw_task(tsk);
oom, PM: make OOM detection in the freezer path raceless Commit 5695be142e20 ("OOM, PM: OOM killed task shouldn't escape PM suspend") has left a race window when OOM killer manages to note_oom_kill after freeze_processes checks the counter. The race window is quite small and really unlikely and partial solution deemed sufficient at the time of submission. Tejun wasn't happy about this partial solution though and insisted on a full solution. That requires the full OOM and freezer's task freezing exclusion, though. This is done by this patch which introduces oom_sem RW lock and turns oom_killer_disable() into a full OOM barrier. oom_killer_disabled check is moved from the allocation path to the OOM level and we take oom_sem for reading for both the check and the whole OOM invocation. oom_killer_disable() takes oom_sem for writing so it waits for all currently running OOM killer invocations. Then it disable all the further OOMs by setting oom_killer_disabled and checks for any oom victims. Victims are counted via mark_tsk_oom_victim resp. unmark_oom_victim. The last victim wakes up all waiters enqueued by oom_killer_disable(). Therefore this function acts as the full OOM barrier. The page fault path is covered now as well although it was assumed to be safe before. As per Tejun, "We used to have freezing points deep in file system code which may be reacheable from page fault." so it would be better and more robust to not rely on freezing points here. Same applies to the memcg OOM killer. out_of_memory tells the caller whether the OOM was allowed to trigger and the callers are supposed to handle the situation. The page allocation path simply fails the allocation same as before. The page fault path will retry the fault (more on that later) and Sysrq OOM trigger will simply complain to the log. Normally there wouldn't be any unfrozen user tasks after try_to_freeze_tasks so the function will not block. But if there was an OOM killer racing with try_to_freeze_tasks and the OOM victim didn't finish yet then we have to wait for it. This should complete in a finite time, though, because - the victim cannot loop in the page fault handler (it would die on the way out from the exception) - it cannot loop in the page allocator because all the further allocation would fail and __GFP_NOFAIL allocations are not acceptable at this stage - it shouldn't be blocked on any locks held by frozen tasks (try_to_freeze expects lockless context) and kernel threads and work queues are not frozen yet Signed-off-by: Michal Hocko <mhocko@suse.cz> Suggested-by: Tejun Heo <tj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-02-12 07:26:24 +08:00
atomic_inc(&oom_victims);
mm/oom_kill.c: add tracepoints for oom reaper-related events During the debugging of the problem described in https://lkml.org/lkml/2017/5/17/542 and fixed by Tetsuo Handa in https://lkml.org/lkml/2017/5/19/383 , I've found that the existing debug output is not really useful to understand issues related to the oom reaper. So, I assume, that adding some tracepoints might help with debugging of similar issues. Trace the following events: 1) a process is marked as an oom victim, 2) a process is added to the oom reaper list, 3) the oom reaper starts reaping process's mm, 4) the oom reaper finished reaping, 5) the oom reaper skips reaping. How it works in practice? Below is an example which show how the problem mentioned above can be found: one process is added twice to the oom_reaper list: $ cd /sys/kernel/debug/tracing $ echo "oom:mark_victim" > set_event $ echo "oom:wake_reaper" >> set_event $ echo "oom:skip_task_reaping" >> set_event $ echo "oom:start_task_reaping" >> set_event $ echo "oom:finish_task_reaping" >> set_event $ cat trace_pipe allocate-502 [001] .... 91.836405: mark_victim: pid=502 allocate-502 [001] .N.. 91.837356: wake_reaper: pid=502 allocate-502 [000] .N.. 91.871149: wake_reaper: pid=502 oom_reaper-23 [000] .... 91.871177: start_task_reaping: pid=502 oom_reaper-23 [000] .N.. 91.879511: finish_task_reaping: pid=502 oom_reaper-23 [000] .... 91.879580: skip_task_reaping: pid=502 Link: http://lkml.kernel.org/r/20170530185231.GA13412@castle Signed-off-by: Roman Gushchin <guro@fb.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-07-11 06:49:05 +08:00
trace_mark_victim(tsk->pid);
oom: add helpers for setting and clearing TIF_MEMDIE This patchset addresses a race which was described in the changelog for 5695be142e20 ("OOM, PM: OOM killed task shouldn't escape PM suspend"): : PM freezer relies on having all tasks frozen by the time devices are : getting frozen so that no task will touch them while they are getting : frozen. But OOM killer is allowed to kill an already frozen task in order : to handle OOM situtation. In order to protect from late wake ups OOM : killer is disabled after all tasks are frozen. This, however, still keeps : a window open when a killed task didn't manage to die by the time : freeze_processes finishes. The original patch hasn't closed the race window completely because that would require a more complex solution as it can be seen by this patchset. The primary motivation was to close the race condition between OOM killer and PM freezer _completely_. As Tejun pointed out, even though the race condition is unlikely the harder it would be to debug weird bugs deep in the PM freezer when the debugging options are reduced considerably. I can only speculate what might happen when a task is still runnable unexpectedly. On a plus side and as a side effect the oom enable/disable has a better (full barrier) semantic without polluting hot paths. I have tested the series in KVM with 100M RAM: - many small tasks (20M anon mmap) which are triggering OOM continually - s2ram which resumes automatically is triggered in a loop echo processors > /sys/power/pm_test while true do echo mem > /sys/power/state sleep 1s done - simple module which allocates and frees 20M in 8K chunks. If it sees freezing(current) then it tries another round of allocation before calling try_to_freeze - debugging messages of PM stages and OOM killer enable/disable/fail added and unmark_oom_victim is delayed by 1s after it clears TIF_MEMDIE and before it wakes up waiters. - rebased on top of the current mmotm which means some necessary updates in mm/oom_kill.c. mark_tsk_oom_victim is now called under task_lock but I think this should be OK because __thaw_task shouldn't interfere with any locking down wake_up_process. Oleg? As expected there are no OOM killed tasks after oom is disabled and allocations requested by the kernel thread are failing after all the tasks are frozen and OOM disabled. I wasn't able to catch a race where oom_killer_disable would really have to wait but I kinda expected the race is really unlikely. [ 242.609330] Killed process 2992 (mem_eater) total-vm:24412kB, anon-rss:2164kB, file-rss:4kB [ 243.628071] Unmarking 2992 OOM victim. oom_victims: 1 [ 243.636072] (elapsed 2.837 seconds) done. [ 243.641985] Trying to disable OOM killer [ 243.643032] Waiting for concurent OOM victims [ 243.644342] OOM killer disabled [ 243.645447] Freezing remaining freezable tasks ... (elapsed 0.005 seconds) done. [ 243.652983] Suspending console(s) (use no_console_suspend to debug) [ 243.903299] kmem_eater: page allocation failure: order:1, mode:0x204010 [...] [ 243.992600] PM: suspend of devices complete after 336.667 msecs [ 243.993264] PM: late suspend of devices complete after 0.660 msecs [ 243.994713] PM: noirq suspend of devices complete after 1.446 msecs [ 243.994717] ACPI: Preparing to enter system sleep state S3 [ 243.994795] PM: Saving platform NVS memory [ 243.994796] Disabling non-boot CPUs ... The first 2 patches are simple cleanups for OOM. They should go in regardless the rest IMO. Patches 3 and 4 are trivial printk -> pr_info conversion and they should go in ditto. The main patch is the last one and I would appreciate acks from Tejun and Rafael. I think the OOM part should be OK (except for __thaw_task vs. task_lock where a look from Oleg would appreciated) but I am not so sure I haven't screwed anything in the freezer code. I have found several surprises there. This patch (of 5): This patch is just a preparatory and it doesn't introduce any functional change. Note: I am utterly unhappy about lowmemory killer abusing TIF_MEMDIE just to wait for the oom victim and to prevent from new killing. This is just a side effect of the flag. The primary meaning is to give the oom victim access to the memory reserves and that shouldn't be necessary here. Signed-off-by: Michal Hocko <mhocko@suse.cz> Cc: Tejun Heo <tj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-02-12 07:26:12 +08:00
}
/**
* exit_oom_victim - note the exit of an OOM victim
oom: add helpers for setting and clearing TIF_MEMDIE This patchset addresses a race which was described in the changelog for 5695be142e20 ("OOM, PM: OOM killed task shouldn't escape PM suspend"): : PM freezer relies on having all tasks frozen by the time devices are : getting frozen so that no task will touch them while they are getting : frozen. But OOM killer is allowed to kill an already frozen task in order : to handle OOM situtation. In order to protect from late wake ups OOM : killer is disabled after all tasks are frozen. This, however, still keeps : a window open when a killed task didn't manage to die by the time : freeze_processes finishes. The original patch hasn't closed the race window completely because that would require a more complex solution as it can be seen by this patchset. The primary motivation was to close the race condition between OOM killer and PM freezer _completely_. As Tejun pointed out, even though the race condition is unlikely the harder it would be to debug weird bugs deep in the PM freezer when the debugging options are reduced considerably. I can only speculate what might happen when a task is still runnable unexpectedly. On a plus side and as a side effect the oom enable/disable has a better (full barrier) semantic without polluting hot paths. I have tested the series in KVM with 100M RAM: - many small tasks (20M anon mmap) which are triggering OOM continually - s2ram which resumes automatically is triggered in a loop echo processors > /sys/power/pm_test while true do echo mem > /sys/power/state sleep 1s done - simple module which allocates and frees 20M in 8K chunks. If it sees freezing(current) then it tries another round of allocation before calling try_to_freeze - debugging messages of PM stages and OOM killer enable/disable/fail added and unmark_oom_victim is delayed by 1s after it clears TIF_MEMDIE and before it wakes up waiters. - rebased on top of the current mmotm which means some necessary updates in mm/oom_kill.c. mark_tsk_oom_victim is now called under task_lock but I think this should be OK because __thaw_task shouldn't interfere with any locking down wake_up_process. Oleg? As expected there are no OOM killed tasks after oom is disabled and allocations requested by the kernel thread are failing after all the tasks are frozen and OOM disabled. I wasn't able to catch a race where oom_killer_disable would really have to wait but I kinda expected the race is really unlikely. [ 242.609330] Killed process 2992 (mem_eater) total-vm:24412kB, anon-rss:2164kB, file-rss:4kB [ 243.628071] Unmarking 2992 OOM victim. oom_victims: 1 [ 243.636072] (elapsed 2.837 seconds) done. [ 243.641985] Trying to disable OOM killer [ 243.643032] Waiting for concurent OOM victims [ 243.644342] OOM killer disabled [ 243.645447] Freezing remaining freezable tasks ... (elapsed 0.005 seconds) done. [ 243.652983] Suspending console(s) (use no_console_suspend to debug) [ 243.903299] kmem_eater: page allocation failure: order:1, mode:0x204010 [...] [ 243.992600] PM: suspend of devices complete after 336.667 msecs [ 243.993264] PM: late suspend of devices complete after 0.660 msecs [ 243.994713] PM: noirq suspend of devices complete after 1.446 msecs [ 243.994717] ACPI: Preparing to enter system sleep state S3 [ 243.994795] PM: Saving platform NVS memory [ 243.994796] Disabling non-boot CPUs ... The first 2 patches are simple cleanups for OOM. They should go in regardless the rest IMO. Patches 3 and 4 are trivial printk -> pr_info conversion and they should go in ditto. The main patch is the last one and I would appreciate acks from Tejun and Rafael. I think the OOM part should be OK (except for __thaw_task vs. task_lock where a look from Oleg would appreciated) but I am not so sure I haven't screwed anything in the freezer code. I have found several surprises there. This patch (of 5): This patch is just a preparatory and it doesn't introduce any functional change. Note: I am utterly unhappy about lowmemory killer abusing TIF_MEMDIE just to wait for the oom victim and to prevent from new killing. This is just a side effect of the flag. The primary meaning is to give the oom victim access to the memory reserves and that shouldn't be necessary here. Signed-off-by: Michal Hocko <mhocko@suse.cz> Cc: Tejun Heo <tj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-02-12 07:26:12 +08:00
*/
void exit_oom_victim(void)
oom: add helpers for setting and clearing TIF_MEMDIE This patchset addresses a race which was described in the changelog for 5695be142e20 ("OOM, PM: OOM killed task shouldn't escape PM suspend"): : PM freezer relies on having all tasks frozen by the time devices are : getting frozen so that no task will touch them while they are getting : frozen. But OOM killer is allowed to kill an already frozen task in order : to handle OOM situtation. In order to protect from late wake ups OOM : killer is disabled after all tasks are frozen. This, however, still keeps : a window open when a killed task didn't manage to die by the time : freeze_processes finishes. The original patch hasn't closed the race window completely because that would require a more complex solution as it can be seen by this patchset. The primary motivation was to close the race condition between OOM killer and PM freezer _completely_. As Tejun pointed out, even though the race condition is unlikely the harder it would be to debug weird bugs deep in the PM freezer when the debugging options are reduced considerably. I can only speculate what might happen when a task is still runnable unexpectedly. On a plus side and as a side effect the oom enable/disable has a better (full barrier) semantic without polluting hot paths. I have tested the series in KVM with 100M RAM: - many small tasks (20M anon mmap) which are triggering OOM continually - s2ram which resumes automatically is triggered in a loop echo processors > /sys/power/pm_test while true do echo mem > /sys/power/state sleep 1s done - simple module which allocates and frees 20M in 8K chunks. If it sees freezing(current) then it tries another round of allocation before calling try_to_freeze - debugging messages of PM stages and OOM killer enable/disable/fail added and unmark_oom_victim is delayed by 1s after it clears TIF_MEMDIE and before it wakes up waiters. - rebased on top of the current mmotm which means some necessary updates in mm/oom_kill.c. mark_tsk_oom_victim is now called under task_lock but I think this should be OK because __thaw_task shouldn't interfere with any locking down wake_up_process. Oleg? As expected there are no OOM killed tasks after oom is disabled and allocations requested by the kernel thread are failing after all the tasks are frozen and OOM disabled. I wasn't able to catch a race where oom_killer_disable would really have to wait but I kinda expected the race is really unlikely. [ 242.609330] Killed process 2992 (mem_eater) total-vm:24412kB, anon-rss:2164kB, file-rss:4kB [ 243.628071] Unmarking 2992 OOM victim. oom_victims: 1 [ 243.636072] (elapsed 2.837 seconds) done. [ 243.641985] Trying to disable OOM killer [ 243.643032] Waiting for concurent OOM victims [ 243.644342] OOM killer disabled [ 243.645447] Freezing remaining freezable tasks ... (elapsed 0.005 seconds) done. [ 243.652983] Suspending console(s) (use no_console_suspend to debug) [ 243.903299] kmem_eater: page allocation failure: order:1, mode:0x204010 [...] [ 243.992600] PM: suspend of devices complete after 336.667 msecs [ 243.993264] PM: late suspend of devices complete after 0.660 msecs [ 243.994713] PM: noirq suspend of devices complete after 1.446 msecs [ 243.994717] ACPI: Preparing to enter system sleep state S3 [ 243.994795] PM: Saving platform NVS memory [ 243.994796] Disabling non-boot CPUs ... The first 2 patches are simple cleanups for OOM. They should go in regardless the rest IMO. Patches 3 and 4 are trivial printk -> pr_info conversion and they should go in ditto. The main patch is the last one and I would appreciate acks from Tejun and Rafael. I think the OOM part should be OK (except for __thaw_task vs. task_lock where a look from Oleg would appreciated) but I am not so sure I haven't screwed anything in the freezer code. I have found several surprises there. This patch (of 5): This patch is just a preparatory and it doesn't introduce any functional change. Note: I am utterly unhappy about lowmemory killer abusing TIF_MEMDIE just to wait for the oom victim and to prevent from new killing. This is just a side effect of the flag. The primary meaning is to give the oom victim access to the memory reserves and that shouldn't be necessary here. Signed-off-by: Michal Hocko <mhocko@suse.cz> Cc: Tejun Heo <tj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-02-12 07:26:12 +08:00
{
clear_thread_flag(TIF_MEMDIE);
oom, PM: make OOM detection in the freezer path raceless Commit 5695be142e20 ("OOM, PM: OOM killed task shouldn't escape PM suspend") has left a race window when OOM killer manages to note_oom_kill after freeze_processes checks the counter. The race window is quite small and really unlikely and partial solution deemed sufficient at the time of submission. Tejun wasn't happy about this partial solution though and insisted on a full solution. That requires the full OOM and freezer's task freezing exclusion, though. This is done by this patch which introduces oom_sem RW lock and turns oom_killer_disable() into a full OOM barrier. oom_killer_disabled check is moved from the allocation path to the OOM level and we take oom_sem for reading for both the check and the whole OOM invocation. oom_killer_disable() takes oom_sem for writing so it waits for all currently running OOM killer invocations. Then it disable all the further OOMs by setting oom_killer_disabled and checks for any oom victims. Victims are counted via mark_tsk_oom_victim resp. unmark_oom_victim. The last victim wakes up all waiters enqueued by oom_killer_disable(). Therefore this function acts as the full OOM barrier. The page fault path is covered now as well although it was assumed to be safe before. As per Tejun, "We used to have freezing points deep in file system code which may be reacheable from page fault." so it would be better and more robust to not rely on freezing points here. Same applies to the memcg OOM killer. out_of_memory tells the caller whether the OOM was allowed to trigger and the callers are supposed to handle the situation. The page allocation path simply fails the allocation same as before. The page fault path will retry the fault (more on that later) and Sysrq OOM trigger will simply complain to the log. Normally there wouldn't be any unfrozen user tasks after try_to_freeze_tasks so the function will not block. But if there was an OOM killer racing with try_to_freeze_tasks and the OOM victim didn't finish yet then we have to wait for it. This should complete in a finite time, though, because - the victim cannot loop in the page fault handler (it would die on the way out from the exception) - it cannot loop in the page allocator because all the further allocation would fail and __GFP_NOFAIL allocations are not acceptable at this stage - it shouldn't be blocked on any locks held by frozen tasks (try_to_freeze expects lockless context) and kernel threads and work queues are not frozen yet Signed-off-by: Michal Hocko <mhocko@suse.cz> Suggested-by: Tejun Heo <tj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-02-12 07:26:24 +08:00
if (!atomic_dec_return(&oom_victims))
oom, PM: make OOM detection in the freezer path raceless Commit 5695be142e20 ("OOM, PM: OOM killed task shouldn't escape PM suspend") has left a race window when OOM killer manages to note_oom_kill after freeze_processes checks the counter. The race window is quite small and really unlikely and partial solution deemed sufficient at the time of submission. Tejun wasn't happy about this partial solution though and insisted on a full solution. That requires the full OOM and freezer's task freezing exclusion, though. This is done by this patch which introduces oom_sem RW lock and turns oom_killer_disable() into a full OOM barrier. oom_killer_disabled check is moved from the allocation path to the OOM level and we take oom_sem for reading for both the check and the whole OOM invocation. oom_killer_disable() takes oom_sem for writing so it waits for all currently running OOM killer invocations. Then it disable all the further OOMs by setting oom_killer_disabled and checks for any oom victims. Victims are counted via mark_tsk_oom_victim resp. unmark_oom_victim. The last victim wakes up all waiters enqueued by oom_killer_disable(). Therefore this function acts as the full OOM barrier. The page fault path is covered now as well although it was assumed to be safe before. As per Tejun, "We used to have freezing points deep in file system code which may be reacheable from page fault." so it would be better and more robust to not rely on freezing points here. Same applies to the memcg OOM killer. out_of_memory tells the caller whether the OOM was allowed to trigger and the callers are supposed to handle the situation. The page allocation path simply fails the allocation same as before. The page fault path will retry the fault (more on that later) and Sysrq OOM trigger will simply complain to the log. Normally there wouldn't be any unfrozen user tasks after try_to_freeze_tasks so the function will not block. But if there was an OOM killer racing with try_to_freeze_tasks and the OOM victim didn't finish yet then we have to wait for it. This should complete in a finite time, though, because - the victim cannot loop in the page fault handler (it would die on the way out from the exception) - it cannot loop in the page allocator because all the further allocation would fail and __GFP_NOFAIL allocations are not acceptable at this stage - it shouldn't be blocked on any locks held by frozen tasks (try_to_freeze expects lockless context) and kernel threads and work queues are not frozen yet Signed-off-by: Michal Hocko <mhocko@suse.cz> Suggested-by: Tejun Heo <tj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-02-12 07:26:24 +08:00
wake_up_all(&oom_victims_wait);
}
oom, suspend: fix oom_killer_disable vs. pm suspend properly Commit 74070542099c ("oom, suspend: fix oom_reaper vs. oom_killer_disable race") has workaround an existing race between oom_killer_disable and oom_reaper by adding another round of try_to_freeze_tasks after the oom killer was disabled. This was the easiest thing to do for a late 4.7 fix. Let's fix it properly now. After "oom: keep mm of the killed task available" we no longer have to call exit_oom_victim from the oom reaper because we have stable mm available and hide the oom_reaped mm by MMF_OOM_SKIP flag. So let's remove exit_oom_victim and the race described in the above commit doesn't exist anymore if. Unfortunately this alone is not sufficient for the oom_killer_disable usecase because now we do not have any reliable way to reach exit_oom_victim (the victim might get stuck on a way to exit for an unbounded amount of time). OOM killer can cope with that by checking mm flags and move on to another victim but we cannot do the same for oom_killer_disable as we would lose the guarantee of no further interference of the victim with the rest of the system. What we can do instead is to cap the maximum time the oom_killer_disable waits for victims. The only current user of this function (pm suspend) already has a concept of timeout for back off so we can reuse the same value there. Let's drop set_freezable for the oom_reaper kthread because it is no longer needed as the reaper doesn't wake or thaw any processes. Link: http://lkml.kernel.org/r/1472119394-11342-7-git-send-email-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: David Rientjes <rientjes@google.com> Cc: Vladimir Davydov <vdavydov@parallels.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-10-08 07:59:00 +08:00
/**
* oom_killer_enable - enable OOM killer
*/
void oom_killer_enable(void)
{
oom_killer_disabled = false;
oom: improve oom disable handling Tetsuo has reported that sysrq triggered OOM killer will print a misleading information when no tasks are selected: sysrq: SysRq : Manual OOM execution Out of memory: Kill process 4468 ((agetty)) score 0 or sacrifice child Killed process 4468 ((agetty)) total-vm:43704kB, anon-rss:1760kB, file-rss:0kB, shmem-rss:0kB sysrq: SysRq : Manual OOM execution Out of memory: Kill process 4469 (systemd-cgroups) score 0 or sacrifice child Killed process 4469 (systemd-cgroups) total-vm:10704kB, anon-rss:120kB, file-rss:0kB, shmem-rss:0kB sysrq: SysRq : Manual OOM execution sysrq: OOM request ignored because killer is disabled sysrq: SysRq : Manual OOM execution sysrq: OOM request ignored because killer is disabled sysrq: SysRq : Manual OOM execution sysrq: OOM request ignored because killer is disabled The real reason is that there are no eligible tasks for the OOM killer to select but since commit 7c5f64f84483 ("mm: oom: deduplicate victim selection code for memcg and global oom") the semantic of out_of_memory has changed without updating moom_callback. This patch updates moom_callback to tell that no task was eligible which is the case for both oom killer disabled and no eligible tasks. In order to help distinguish first case from the second add printk to both oom_killer_{enable,disable}. This information is useful on its own because it might help debugging potential memory allocation failures. Fixes: 7c5f64f84483 ("mm: oom: deduplicate victim selection code for memcg and global oom") Link: http://lkml.kernel.org/r/20170404134705.6361-1-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Reported-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-05-04 05:54:57 +08:00
pr_info("OOM killer enabled.\n");
oom, suspend: fix oom_killer_disable vs. pm suspend properly Commit 74070542099c ("oom, suspend: fix oom_reaper vs. oom_killer_disable race") has workaround an existing race between oom_killer_disable and oom_reaper by adding another round of try_to_freeze_tasks after the oom killer was disabled. This was the easiest thing to do for a late 4.7 fix. Let's fix it properly now. After "oom: keep mm of the killed task available" we no longer have to call exit_oom_victim from the oom reaper because we have stable mm available and hide the oom_reaped mm by MMF_OOM_SKIP flag. So let's remove exit_oom_victim and the race described in the above commit doesn't exist anymore if. Unfortunately this alone is not sufficient for the oom_killer_disable usecase because now we do not have any reliable way to reach exit_oom_victim (the victim might get stuck on a way to exit for an unbounded amount of time). OOM killer can cope with that by checking mm flags and move on to another victim but we cannot do the same for oom_killer_disable as we would lose the guarantee of no further interference of the victim with the rest of the system. What we can do instead is to cap the maximum time the oom_killer_disable waits for victims. The only current user of this function (pm suspend) already has a concept of timeout for back off so we can reuse the same value there. Let's drop set_freezable for the oom_reaper kthread because it is no longer needed as the reaper doesn't wake or thaw any processes. Link: http://lkml.kernel.org/r/1472119394-11342-7-git-send-email-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: David Rientjes <rientjes@google.com> Cc: Vladimir Davydov <vdavydov@parallels.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-10-08 07:59:00 +08:00
}
oom, PM: make OOM detection in the freezer path raceless Commit 5695be142e20 ("OOM, PM: OOM killed task shouldn't escape PM suspend") has left a race window when OOM killer manages to note_oom_kill after freeze_processes checks the counter. The race window is quite small and really unlikely and partial solution deemed sufficient at the time of submission. Tejun wasn't happy about this partial solution though and insisted on a full solution. That requires the full OOM and freezer's task freezing exclusion, though. This is done by this patch which introduces oom_sem RW lock and turns oom_killer_disable() into a full OOM barrier. oom_killer_disabled check is moved from the allocation path to the OOM level and we take oom_sem for reading for both the check and the whole OOM invocation. oom_killer_disable() takes oom_sem for writing so it waits for all currently running OOM killer invocations. Then it disable all the further OOMs by setting oom_killer_disabled and checks for any oom victims. Victims are counted via mark_tsk_oom_victim resp. unmark_oom_victim. The last victim wakes up all waiters enqueued by oom_killer_disable(). Therefore this function acts as the full OOM barrier. The page fault path is covered now as well although it was assumed to be safe before. As per Tejun, "We used to have freezing points deep in file system code which may be reacheable from page fault." so it would be better and more robust to not rely on freezing points here. Same applies to the memcg OOM killer. out_of_memory tells the caller whether the OOM was allowed to trigger and the callers are supposed to handle the situation. The page allocation path simply fails the allocation same as before. The page fault path will retry the fault (more on that later) and Sysrq OOM trigger will simply complain to the log. Normally there wouldn't be any unfrozen user tasks after try_to_freeze_tasks so the function will not block. But if there was an OOM killer racing with try_to_freeze_tasks and the OOM victim didn't finish yet then we have to wait for it. This should complete in a finite time, though, because - the victim cannot loop in the page fault handler (it would die on the way out from the exception) - it cannot loop in the page allocator because all the further allocation would fail and __GFP_NOFAIL allocations are not acceptable at this stage - it shouldn't be blocked on any locks held by frozen tasks (try_to_freeze expects lockless context) and kernel threads and work queues are not frozen yet Signed-off-by: Michal Hocko <mhocko@suse.cz> Suggested-by: Tejun Heo <tj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-02-12 07:26:24 +08:00
/**
* oom_killer_disable - disable OOM killer
oom, suspend: fix oom_killer_disable vs. pm suspend properly Commit 74070542099c ("oom, suspend: fix oom_reaper vs. oom_killer_disable race") has workaround an existing race between oom_killer_disable and oom_reaper by adding another round of try_to_freeze_tasks after the oom killer was disabled. This was the easiest thing to do for a late 4.7 fix. Let's fix it properly now. After "oom: keep mm of the killed task available" we no longer have to call exit_oom_victim from the oom reaper because we have stable mm available and hide the oom_reaped mm by MMF_OOM_SKIP flag. So let's remove exit_oom_victim and the race described in the above commit doesn't exist anymore if. Unfortunately this alone is not sufficient for the oom_killer_disable usecase because now we do not have any reliable way to reach exit_oom_victim (the victim might get stuck on a way to exit for an unbounded amount of time). OOM killer can cope with that by checking mm flags and move on to another victim but we cannot do the same for oom_killer_disable as we would lose the guarantee of no further interference of the victim with the rest of the system. What we can do instead is to cap the maximum time the oom_killer_disable waits for victims. The only current user of this function (pm suspend) already has a concept of timeout for back off so we can reuse the same value there. Let's drop set_freezable for the oom_reaper kthread because it is no longer needed as the reaper doesn't wake or thaw any processes. Link: http://lkml.kernel.org/r/1472119394-11342-7-git-send-email-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: David Rientjes <rientjes@google.com> Cc: Vladimir Davydov <vdavydov@parallels.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-10-08 07:59:00 +08:00
* @timeout: maximum timeout to wait for oom victims in jiffies
oom, PM: make OOM detection in the freezer path raceless Commit 5695be142e20 ("OOM, PM: OOM killed task shouldn't escape PM suspend") has left a race window when OOM killer manages to note_oom_kill after freeze_processes checks the counter. The race window is quite small and really unlikely and partial solution deemed sufficient at the time of submission. Tejun wasn't happy about this partial solution though and insisted on a full solution. That requires the full OOM and freezer's task freezing exclusion, though. This is done by this patch which introduces oom_sem RW lock and turns oom_killer_disable() into a full OOM barrier. oom_killer_disabled check is moved from the allocation path to the OOM level and we take oom_sem for reading for both the check and the whole OOM invocation. oom_killer_disable() takes oom_sem for writing so it waits for all currently running OOM killer invocations. Then it disable all the further OOMs by setting oom_killer_disabled and checks for any oom victims. Victims are counted via mark_tsk_oom_victim resp. unmark_oom_victim. The last victim wakes up all waiters enqueued by oom_killer_disable(). Therefore this function acts as the full OOM barrier. The page fault path is covered now as well although it was assumed to be safe before. As per Tejun, "We used to have freezing points deep in file system code which may be reacheable from page fault." so it would be better and more robust to not rely on freezing points here. Same applies to the memcg OOM killer. out_of_memory tells the caller whether the OOM was allowed to trigger and the callers are supposed to handle the situation. The page allocation path simply fails the allocation same as before. The page fault path will retry the fault (more on that later) and Sysrq OOM trigger will simply complain to the log. Normally there wouldn't be any unfrozen user tasks after try_to_freeze_tasks so the function will not block. But if there was an OOM killer racing with try_to_freeze_tasks and the OOM victim didn't finish yet then we have to wait for it. This should complete in a finite time, though, because - the victim cannot loop in the page fault handler (it would die on the way out from the exception) - it cannot loop in the page allocator because all the further allocation would fail and __GFP_NOFAIL allocations are not acceptable at this stage - it shouldn't be blocked on any locks held by frozen tasks (try_to_freeze expects lockless context) and kernel threads and work queues are not frozen yet Signed-off-by: Michal Hocko <mhocko@suse.cz> Suggested-by: Tejun Heo <tj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-02-12 07:26:24 +08:00
*
* Forces all page allocations to fail rather than trigger OOM killer.
oom, suspend: fix oom_killer_disable vs. pm suspend properly Commit 74070542099c ("oom, suspend: fix oom_reaper vs. oom_killer_disable race") has workaround an existing race between oom_killer_disable and oom_reaper by adding another round of try_to_freeze_tasks after the oom killer was disabled. This was the easiest thing to do for a late 4.7 fix. Let's fix it properly now. After "oom: keep mm of the killed task available" we no longer have to call exit_oom_victim from the oom reaper because we have stable mm available and hide the oom_reaped mm by MMF_OOM_SKIP flag. So let's remove exit_oom_victim and the race described in the above commit doesn't exist anymore if. Unfortunately this alone is not sufficient for the oom_killer_disable usecase because now we do not have any reliable way to reach exit_oom_victim (the victim might get stuck on a way to exit for an unbounded amount of time). OOM killer can cope with that by checking mm flags and move on to another victim but we cannot do the same for oom_killer_disable as we would lose the guarantee of no further interference of the victim with the rest of the system. What we can do instead is to cap the maximum time the oom_killer_disable waits for victims. The only current user of this function (pm suspend) already has a concept of timeout for back off so we can reuse the same value there. Let's drop set_freezable for the oom_reaper kthread because it is no longer needed as the reaper doesn't wake or thaw any processes. Link: http://lkml.kernel.org/r/1472119394-11342-7-git-send-email-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: David Rientjes <rientjes@google.com> Cc: Vladimir Davydov <vdavydov@parallels.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-10-08 07:59:00 +08:00
* Will block and wait until all OOM victims are killed or the given
* timeout expires.
oom, PM: make OOM detection in the freezer path raceless Commit 5695be142e20 ("OOM, PM: OOM killed task shouldn't escape PM suspend") has left a race window when OOM killer manages to note_oom_kill after freeze_processes checks the counter. The race window is quite small and really unlikely and partial solution deemed sufficient at the time of submission. Tejun wasn't happy about this partial solution though and insisted on a full solution. That requires the full OOM and freezer's task freezing exclusion, though. This is done by this patch which introduces oom_sem RW lock and turns oom_killer_disable() into a full OOM barrier. oom_killer_disabled check is moved from the allocation path to the OOM level and we take oom_sem for reading for both the check and the whole OOM invocation. oom_killer_disable() takes oom_sem for writing so it waits for all currently running OOM killer invocations. Then it disable all the further OOMs by setting oom_killer_disabled and checks for any oom victims. Victims are counted via mark_tsk_oom_victim resp. unmark_oom_victim. The last victim wakes up all waiters enqueued by oom_killer_disable(). Therefore this function acts as the full OOM barrier. The page fault path is covered now as well although it was assumed to be safe before. As per Tejun, "We used to have freezing points deep in file system code which may be reacheable from page fault." so it would be better and more robust to not rely on freezing points here. Same applies to the memcg OOM killer. out_of_memory tells the caller whether the OOM was allowed to trigger and the callers are supposed to handle the situation. The page allocation path simply fails the allocation same as before. The page fault path will retry the fault (more on that later) and Sysrq OOM trigger will simply complain to the log. Normally there wouldn't be any unfrozen user tasks after try_to_freeze_tasks so the function will not block. But if there was an OOM killer racing with try_to_freeze_tasks and the OOM victim didn't finish yet then we have to wait for it. This should complete in a finite time, though, because - the victim cannot loop in the page fault handler (it would die on the way out from the exception) - it cannot loop in the page allocator because all the further allocation would fail and __GFP_NOFAIL allocations are not acceptable at this stage - it shouldn't be blocked on any locks held by frozen tasks (try_to_freeze expects lockless context) and kernel threads and work queues are not frozen yet Signed-off-by: Michal Hocko <mhocko@suse.cz> Suggested-by: Tejun Heo <tj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-02-12 07:26:24 +08:00
*
* The function cannot be called when there are runnable user tasks because
* the userspace would see unexpected allocation failures as a result. Any
* new usage of this function should be consulted with MM people.
*
* Returns true if successful and false if the OOM killer cannot be
* disabled.
*/
oom, suspend: fix oom_killer_disable vs. pm suspend properly Commit 74070542099c ("oom, suspend: fix oom_reaper vs. oom_killer_disable race") has workaround an existing race between oom_killer_disable and oom_reaper by adding another round of try_to_freeze_tasks after the oom killer was disabled. This was the easiest thing to do for a late 4.7 fix. Let's fix it properly now. After "oom: keep mm of the killed task available" we no longer have to call exit_oom_victim from the oom reaper because we have stable mm available and hide the oom_reaped mm by MMF_OOM_SKIP flag. So let's remove exit_oom_victim and the race described in the above commit doesn't exist anymore if. Unfortunately this alone is not sufficient for the oom_killer_disable usecase because now we do not have any reliable way to reach exit_oom_victim (the victim might get stuck on a way to exit for an unbounded amount of time). OOM killer can cope with that by checking mm flags and move on to another victim but we cannot do the same for oom_killer_disable as we would lose the guarantee of no further interference of the victim with the rest of the system. What we can do instead is to cap the maximum time the oom_killer_disable waits for victims. The only current user of this function (pm suspend) already has a concept of timeout for back off so we can reuse the same value there. Let's drop set_freezable for the oom_reaper kthread because it is no longer needed as the reaper doesn't wake or thaw any processes. Link: http://lkml.kernel.org/r/1472119394-11342-7-git-send-email-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: David Rientjes <rientjes@google.com> Cc: Vladimir Davydov <vdavydov@parallels.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-10-08 07:59:00 +08:00
bool oom_killer_disable(signed long timeout)
oom, PM: make OOM detection in the freezer path raceless Commit 5695be142e20 ("OOM, PM: OOM killed task shouldn't escape PM suspend") has left a race window when OOM killer manages to note_oom_kill after freeze_processes checks the counter. The race window is quite small and really unlikely and partial solution deemed sufficient at the time of submission. Tejun wasn't happy about this partial solution though and insisted on a full solution. That requires the full OOM and freezer's task freezing exclusion, though. This is done by this patch which introduces oom_sem RW lock and turns oom_killer_disable() into a full OOM barrier. oom_killer_disabled check is moved from the allocation path to the OOM level and we take oom_sem for reading for both the check and the whole OOM invocation. oom_killer_disable() takes oom_sem for writing so it waits for all currently running OOM killer invocations. Then it disable all the further OOMs by setting oom_killer_disabled and checks for any oom victims. Victims are counted via mark_tsk_oom_victim resp. unmark_oom_victim. The last victim wakes up all waiters enqueued by oom_killer_disable(). Therefore this function acts as the full OOM barrier. The page fault path is covered now as well although it was assumed to be safe before. As per Tejun, "We used to have freezing points deep in file system code which may be reacheable from page fault." so it would be better and more robust to not rely on freezing points here. Same applies to the memcg OOM killer. out_of_memory tells the caller whether the OOM was allowed to trigger and the callers are supposed to handle the situation. The page allocation path simply fails the allocation same as before. The page fault path will retry the fault (more on that later) and Sysrq OOM trigger will simply complain to the log. Normally there wouldn't be any unfrozen user tasks after try_to_freeze_tasks so the function will not block. But if there was an OOM killer racing with try_to_freeze_tasks and the OOM victim didn't finish yet then we have to wait for it. This should complete in a finite time, though, because - the victim cannot loop in the page fault handler (it would die on the way out from the exception) - it cannot loop in the page allocator because all the further allocation would fail and __GFP_NOFAIL allocations are not acceptable at this stage - it shouldn't be blocked on any locks held by frozen tasks (try_to_freeze expects lockless context) and kernel threads and work queues are not frozen yet Signed-off-by: Michal Hocko <mhocko@suse.cz> Suggested-by: Tejun Heo <tj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-02-12 07:26:24 +08:00
{
oom, suspend: fix oom_killer_disable vs. pm suspend properly Commit 74070542099c ("oom, suspend: fix oom_reaper vs. oom_killer_disable race") has workaround an existing race between oom_killer_disable and oom_reaper by adding another round of try_to_freeze_tasks after the oom killer was disabled. This was the easiest thing to do for a late 4.7 fix. Let's fix it properly now. After "oom: keep mm of the killed task available" we no longer have to call exit_oom_victim from the oom reaper because we have stable mm available and hide the oom_reaped mm by MMF_OOM_SKIP flag. So let's remove exit_oom_victim and the race described in the above commit doesn't exist anymore if. Unfortunately this alone is not sufficient for the oom_killer_disable usecase because now we do not have any reliable way to reach exit_oom_victim (the victim might get stuck on a way to exit for an unbounded amount of time). OOM killer can cope with that by checking mm flags and move on to another victim but we cannot do the same for oom_killer_disable as we would lose the guarantee of no further interference of the victim with the rest of the system. What we can do instead is to cap the maximum time the oom_killer_disable waits for victims. The only current user of this function (pm suspend) already has a concept of timeout for back off so we can reuse the same value there. Let's drop set_freezable for the oom_reaper kthread because it is no longer needed as the reaper doesn't wake or thaw any processes. Link: http://lkml.kernel.org/r/1472119394-11342-7-git-send-email-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: David Rientjes <rientjes@google.com> Cc: Vladimir Davydov <vdavydov@parallels.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-10-08 07:59:00 +08:00
signed long ret;
oom, PM: make OOM detection in the freezer path raceless Commit 5695be142e20 ("OOM, PM: OOM killed task shouldn't escape PM suspend") has left a race window when OOM killer manages to note_oom_kill after freeze_processes checks the counter. The race window is quite small and really unlikely and partial solution deemed sufficient at the time of submission. Tejun wasn't happy about this partial solution though and insisted on a full solution. That requires the full OOM and freezer's task freezing exclusion, though. This is done by this patch which introduces oom_sem RW lock and turns oom_killer_disable() into a full OOM barrier. oom_killer_disabled check is moved from the allocation path to the OOM level and we take oom_sem for reading for both the check and the whole OOM invocation. oom_killer_disable() takes oom_sem for writing so it waits for all currently running OOM killer invocations. Then it disable all the further OOMs by setting oom_killer_disabled and checks for any oom victims. Victims are counted via mark_tsk_oom_victim resp. unmark_oom_victim. The last victim wakes up all waiters enqueued by oom_killer_disable(). Therefore this function acts as the full OOM barrier. The page fault path is covered now as well although it was assumed to be safe before. As per Tejun, "We used to have freezing points deep in file system code which may be reacheable from page fault." so it would be better and more robust to not rely on freezing points here. Same applies to the memcg OOM killer. out_of_memory tells the caller whether the OOM was allowed to trigger and the callers are supposed to handle the situation. The page allocation path simply fails the allocation same as before. The page fault path will retry the fault (more on that later) and Sysrq OOM trigger will simply complain to the log. Normally there wouldn't be any unfrozen user tasks after try_to_freeze_tasks so the function will not block. But if there was an OOM killer racing with try_to_freeze_tasks and the OOM victim didn't finish yet then we have to wait for it. This should complete in a finite time, though, because - the victim cannot loop in the page fault handler (it would die on the way out from the exception) - it cannot loop in the page allocator because all the further allocation would fail and __GFP_NOFAIL allocations are not acceptable at this stage - it shouldn't be blocked on any locks held by frozen tasks (try_to_freeze expects lockless context) and kernel threads and work queues are not frozen yet Signed-off-by: Michal Hocko <mhocko@suse.cz> Suggested-by: Tejun Heo <tj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-02-12 07:26:24 +08:00
/*
* Make sure to not race with an ongoing OOM killer. Check that the
* current is not killed (possibly due to sharing the victim's memory).
oom, PM: make OOM detection in the freezer path raceless Commit 5695be142e20 ("OOM, PM: OOM killed task shouldn't escape PM suspend") has left a race window when OOM killer manages to note_oom_kill after freeze_processes checks the counter. The race window is quite small and really unlikely and partial solution deemed sufficient at the time of submission. Tejun wasn't happy about this partial solution though and insisted on a full solution. That requires the full OOM and freezer's task freezing exclusion, though. This is done by this patch which introduces oom_sem RW lock and turns oom_killer_disable() into a full OOM barrier. oom_killer_disabled check is moved from the allocation path to the OOM level and we take oom_sem for reading for both the check and the whole OOM invocation. oom_killer_disable() takes oom_sem for writing so it waits for all currently running OOM killer invocations. Then it disable all the further OOMs by setting oom_killer_disabled and checks for any oom victims. Victims are counted via mark_tsk_oom_victim resp. unmark_oom_victim. The last victim wakes up all waiters enqueued by oom_killer_disable(). Therefore this function acts as the full OOM barrier. The page fault path is covered now as well although it was assumed to be safe before. As per Tejun, "We used to have freezing points deep in file system code which may be reacheable from page fault." so it would be better and more robust to not rely on freezing points here. Same applies to the memcg OOM killer. out_of_memory tells the caller whether the OOM was allowed to trigger and the callers are supposed to handle the situation. The page allocation path simply fails the allocation same as before. The page fault path will retry the fault (more on that later) and Sysrq OOM trigger will simply complain to the log. Normally there wouldn't be any unfrozen user tasks after try_to_freeze_tasks so the function will not block. But if there was an OOM killer racing with try_to_freeze_tasks and the OOM victim didn't finish yet then we have to wait for it. This should complete in a finite time, though, because - the victim cannot loop in the page fault handler (it would die on the way out from the exception) - it cannot loop in the page allocator because all the further allocation would fail and __GFP_NOFAIL allocations are not acceptable at this stage - it shouldn't be blocked on any locks held by frozen tasks (try_to_freeze expects lockless context) and kernel threads and work queues are not frozen yet Signed-off-by: Michal Hocko <mhocko@suse.cz> Suggested-by: Tejun Heo <tj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-02-12 07:26:24 +08:00
*/
if (mutex_lock_killable(&oom_lock))
oom, PM: make OOM detection in the freezer path raceless Commit 5695be142e20 ("OOM, PM: OOM killed task shouldn't escape PM suspend") has left a race window when OOM killer manages to note_oom_kill after freeze_processes checks the counter. The race window is quite small and really unlikely and partial solution deemed sufficient at the time of submission. Tejun wasn't happy about this partial solution though and insisted on a full solution. That requires the full OOM and freezer's task freezing exclusion, though. This is done by this patch which introduces oom_sem RW lock and turns oom_killer_disable() into a full OOM barrier. oom_killer_disabled check is moved from the allocation path to the OOM level and we take oom_sem for reading for both the check and the whole OOM invocation. oom_killer_disable() takes oom_sem for writing so it waits for all currently running OOM killer invocations. Then it disable all the further OOMs by setting oom_killer_disabled and checks for any oom victims. Victims are counted via mark_tsk_oom_victim resp. unmark_oom_victim. The last victim wakes up all waiters enqueued by oom_killer_disable(). Therefore this function acts as the full OOM barrier. The page fault path is covered now as well although it was assumed to be safe before. As per Tejun, "We used to have freezing points deep in file system code which may be reacheable from page fault." so it would be better and more robust to not rely on freezing points here. Same applies to the memcg OOM killer. out_of_memory tells the caller whether the OOM was allowed to trigger and the callers are supposed to handle the situation. The page allocation path simply fails the allocation same as before. The page fault path will retry the fault (more on that later) and Sysrq OOM trigger will simply complain to the log. Normally there wouldn't be any unfrozen user tasks after try_to_freeze_tasks so the function will not block. But if there was an OOM killer racing with try_to_freeze_tasks and the OOM victim didn't finish yet then we have to wait for it. This should complete in a finite time, though, because - the victim cannot loop in the page fault handler (it would die on the way out from the exception) - it cannot loop in the page allocator because all the further allocation would fail and __GFP_NOFAIL allocations are not acceptable at this stage - it shouldn't be blocked on any locks held by frozen tasks (try_to_freeze expects lockless context) and kernel threads and work queues are not frozen yet Signed-off-by: Michal Hocko <mhocko@suse.cz> Suggested-by: Tejun Heo <tj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-02-12 07:26:24 +08:00
return false;
oom_killer_disabled = true;
mutex_unlock(&oom_lock);
oom, PM: make OOM detection in the freezer path raceless Commit 5695be142e20 ("OOM, PM: OOM killed task shouldn't escape PM suspend") has left a race window when OOM killer manages to note_oom_kill after freeze_processes checks the counter. The race window is quite small and really unlikely and partial solution deemed sufficient at the time of submission. Tejun wasn't happy about this partial solution though and insisted on a full solution. That requires the full OOM and freezer's task freezing exclusion, though. This is done by this patch which introduces oom_sem RW lock and turns oom_killer_disable() into a full OOM barrier. oom_killer_disabled check is moved from the allocation path to the OOM level and we take oom_sem for reading for both the check and the whole OOM invocation. oom_killer_disable() takes oom_sem for writing so it waits for all currently running OOM killer invocations. Then it disable all the further OOMs by setting oom_killer_disabled and checks for any oom victims. Victims are counted via mark_tsk_oom_victim resp. unmark_oom_victim. The last victim wakes up all waiters enqueued by oom_killer_disable(). Therefore this function acts as the full OOM barrier. The page fault path is covered now as well although it was assumed to be safe before. As per Tejun, "We used to have freezing points deep in file system code which may be reacheable from page fault." so it would be better and more robust to not rely on freezing points here. Same applies to the memcg OOM killer. out_of_memory tells the caller whether the OOM was allowed to trigger and the callers are supposed to handle the situation. The page allocation path simply fails the allocation same as before. The page fault path will retry the fault (more on that later) and Sysrq OOM trigger will simply complain to the log. Normally there wouldn't be any unfrozen user tasks after try_to_freeze_tasks so the function will not block. But if there was an OOM killer racing with try_to_freeze_tasks and the OOM victim didn't finish yet then we have to wait for it. This should complete in a finite time, though, because - the victim cannot loop in the page fault handler (it would die on the way out from the exception) - it cannot loop in the page allocator because all the further allocation would fail and __GFP_NOFAIL allocations are not acceptable at this stage - it shouldn't be blocked on any locks held by frozen tasks (try_to_freeze expects lockless context) and kernel threads and work queues are not frozen yet Signed-off-by: Michal Hocko <mhocko@suse.cz> Suggested-by: Tejun Heo <tj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-02-12 07:26:24 +08:00
oom, suspend: fix oom_killer_disable vs. pm suspend properly Commit 74070542099c ("oom, suspend: fix oom_reaper vs. oom_killer_disable race") has workaround an existing race between oom_killer_disable and oom_reaper by adding another round of try_to_freeze_tasks after the oom killer was disabled. This was the easiest thing to do for a late 4.7 fix. Let's fix it properly now. After "oom: keep mm of the killed task available" we no longer have to call exit_oom_victim from the oom reaper because we have stable mm available and hide the oom_reaped mm by MMF_OOM_SKIP flag. So let's remove exit_oom_victim and the race described in the above commit doesn't exist anymore if. Unfortunately this alone is not sufficient for the oom_killer_disable usecase because now we do not have any reliable way to reach exit_oom_victim (the victim might get stuck on a way to exit for an unbounded amount of time). OOM killer can cope with that by checking mm flags and move on to another victim but we cannot do the same for oom_killer_disable as we would lose the guarantee of no further interference of the victim with the rest of the system. What we can do instead is to cap the maximum time the oom_killer_disable waits for victims. The only current user of this function (pm suspend) already has a concept of timeout for back off so we can reuse the same value there. Let's drop set_freezable for the oom_reaper kthread because it is no longer needed as the reaper doesn't wake or thaw any processes. Link: http://lkml.kernel.org/r/1472119394-11342-7-git-send-email-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: David Rientjes <rientjes@google.com> Cc: Vladimir Davydov <vdavydov@parallels.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-10-08 07:59:00 +08:00
ret = wait_event_interruptible_timeout(oom_victims_wait,
!atomic_read(&oom_victims), timeout);
if (ret <= 0) {
oom_killer_enable();
return false;
}
oom: improve oom disable handling Tetsuo has reported that sysrq triggered OOM killer will print a misleading information when no tasks are selected: sysrq: SysRq : Manual OOM execution Out of memory: Kill process 4468 ((agetty)) score 0 or sacrifice child Killed process 4468 ((agetty)) total-vm:43704kB, anon-rss:1760kB, file-rss:0kB, shmem-rss:0kB sysrq: SysRq : Manual OOM execution Out of memory: Kill process 4469 (systemd-cgroups) score 0 or sacrifice child Killed process 4469 (systemd-cgroups) total-vm:10704kB, anon-rss:120kB, file-rss:0kB, shmem-rss:0kB sysrq: SysRq : Manual OOM execution sysrq: OOM request ignored because killer is disabled sysrq: SysRq : Manual OOM execution sysrq: OOM request ignored because killer is disabled sysrq: SysRq : Manual OOM execution sysrq: OOM request ignored because killer is disabled The real reason is that there are no eligible tasks for the OOM killer to select but since commit 7c5f64f84483 ("mm: oom: deduplicate victim selection code for memcg and global oom") the semantic of out_of_memory has changed without updating moom_callback. This patch updates moom_callback to tell that no task was eligible which is the case for both oom killer disabled and no eligible tasks. In order to help distinguish first case from the second add printk to both oom_killer_{enable,disable}. This information is useful on its own because it might help debugging potential memory allocation failures. Fixes: 7c5f64f84483 ("mm: oom: deduplicate victim selection code for memcg and global oom") Link: http://lkml.kernel.org/r/20170404134705.6361-1-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Reported-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-05-04 05:54:57 +08:00
pr_info("OOM killer disabled.\n");
oom, PM: make OOM detection in the freezer path raceless Commit 5695be142e20 ("OOM, PM: OOM killed task shouldn't escape PM suspend") has left a race window when OOM killer manages to note_oom_kill after freeze_processes checks the counter. The race window is quite small and really unlikely and partial solution deemed sufficient at the time of submission. Tejun wasn't happy about this partial solution though and insisted on a full solution. That requires the full OOM and freezer's task freezing exclusion, though. This is done by this patch which introduces oom_sem RW lock and turns oom_killer_disable() into a full OOM barrier. oom_killer_disabled check is moved from the allocation path to the OOM level and we take oom_sem for reading for both the check and the whole OOM invocation. oom_killer_disable() takes oom_sem for writing so it waits for all currently running OOM killer invocations. Then it disable all the further OOMs by setting oom_killer_disabled and checks for any oom victims. Victims are counted via mark_tsk_oom_victim resp. unmark_oom_victim. The last victim wakes up all waiters enqueued by oom_killer_disable(). Therefore this function acts as the full OOM barrier. The page fault path is covered now as well although it was assumed to be safe before. As per Tejun, "We used to have freezing points deep in file system code which may be reacheable from page fault." so it would be better and more robust to not rely on freezing points here. Same applies to the memcg OOM killer. out_of_memory tells the caller whether the OOM was allowed to trigger and the callers are supposed to handle the situation. The page allocation path simply fails the allocation same as before. The page fault path will retry the fault (more on that later) and Sysrq OOM trigger will simply complain to the log. Normally there wouldn't be any unfrozen user tasks after try_to_freeze_tasks so the function will not block. But if there was an OOM killer racing with try_to_freeze_tasks and the OOM victim didn't finish yet then we have to wait for it. This should complete in a finite time, though, because - the victim cannot loop in the page fault handler (it would die on the way out from the exception) - it cannot loop in the page allocator because all the further allocation would fail and __GFP_NOFAIL allocations are not acceptable at this stage - it shouldn't be blocked on any locks held by frozen tasks (try_to_freeze expects lockless context) and kernel threads and work queues are not frozen yet Signed-off-by: Michal Hocko <mhocko@suse.cz> Suggested-by: Tejun Heo <tj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-02-12 07:26:24 +08:00
return true;
}
static inline bool __task_will_free_mem(struct task_struct *task)
{
struct signal_struct *sig = task->signal;
/*
* A coredumping process may sleep for an extended period in exit_mm(),
* so the oom killer cannot assume that the process will promptly exit
* and release memory.
*/
if (sig->flags & SIGNAL_GROUP_COREDUMP)
return false;
if (sig->flags & SIGNAL_GROUP_EXIT)
return true;
if (thread_group_empty(task) && (task->flags & PF_EXITING))
return true;
return false;
}
/*
* Checks whether the given task is dying or exiting and likely to
* release its address space. This means that all threads and processes
* sharing the same mm have to be killed or exiting.
* Caller has to make sure that task->mm is stable (hold task_lock or
* it operates on the current).
*/
static bool task_will_free_mem(struct task_struct *task)
{
struct mm_struct *mm = task->mm;
struct task_struct *p;
bool ret = true;
/*
* Skip tasks without mm because it might have passed its exit_mm and
* exit_oom_victim. oom_reaper could have rescued that but do not rely
* on that for now. We can consider find_lock_task_mm in future.
*/
if (!mm)
return false;
if (!__task_will_free_mem(task))
return false;
mm, oom: task_will_free_mem should skip oom_reaped tasks The 0-day robot has encountered the following: Out of memory: Kill process 3914 (trinity-c0) score 167 or sacrifice child Killed process 3914 (trinity-c0) total-vm:55864kB, anon-rss:1512kB, file-rss:1088kB, shmem-rss:25616kB oom_reaper: reaped process 3914 (trinity-c0), now anon-rss:0kB, file-rss:0kB, shmem-rss:26488kB oom_reaper: reaped process 3914 (trinity-c0), now anon-rss:0kB, file-rss:0kB, shmem-rss:26900kB oom_reaper: reaped process 3914 (trinity-c0), now anon-rss:0kB, file-rss:0kB, shmem-rss:26900kB oom_reaper: reaped process 3914 (trinity-c0), now anon-rss:0kB, file-rss:0kB, shmem-rss:27296kB oom_reaper: reaped process 3914 (trinity-c0), now anon-rss:0kB, file-rss:0kB, shmem-rss:28148kB oom_reaper is trying to reap the same task again and again. This is possible only when the oom killer is bypassed because of task_will_free_mem because we skip over tasks with MMF_OOM_REAPED already set during select_bad_process. Teach task_will_free_mem to skip over MMF_OOM_REAPED tasks as well because they will be unlikely to free anything more. Analyzed by Tetsuo Handa. Link: http://lkml.kernel.org/r/1466426628-15074-9-git-send-email-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Acked-by: Oleg Nesterov <oleg@redhat.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Vladimir Davydov <vdavydov@virtuozzo.com> Cc: David Rientjes <rientjes@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-07-29 06:44:55 +08:00
/*
* This task has already been drained by the oom reaper so there are
* only small chances it will free some more
*/
if (test_bit(MMF_OOM_SKIP, &mm->flags))
mm, oom: task_will_free_mem should skip oom_reaped tasks The 0-day robot has encountered the following: Out of memory: Kill process 3914 (trinity-c0) score 167 or sacrifice child Killed process 3914 (trinity-c0) total-vm:55864kB, anon-rss:1512kB, file-rss:1088kB, shmem-rss:25616kB oom_reaper: reaped process 3914 (trinity-c0), now anon-rss:0kB, file-rss:0kB, shmem-rss:26488kB oom_reaper: reaped process 3914 (trinity-c0), now anon-rss:0kB, file-rss:0kB, shmem-rss:26900kB oom_reaper: reaped process 3914 (trinity-c0), now anon-rss:0kB, file-rss:0kB, shmem-rss:26900kB oom_reaper: reaped process 3914 (trinity-c0), now anon-rss:0kB, file-rss:0kB, shmem-rss:27296kB oom_reaper: reaped process 3914 (trinity-c0), now anon-rss:0kB, file-rss:0kB, shmem-rss:28148kB oom_reaper is trying to reap the same task again and again. This is possible only when the oom killer is bypassed because of task_will_free_mem because we skip over tasks with MMF_OOM_REAPED already set during select_bad_process. Teach task_will_free_mem to skip over MMF_OOM_REAPED tasks as well because they will be unlikely to free anything more. Analyzed by Tetsuo Handa. Link: http://lkml.kernel.org/r/1466426628-15074-9-git-send-email-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Acked-by: Oleg Nesterov <oleg@redhat.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Vladimir Davydov <vdavydov@virtuozzo.com> Cc: David Rientjes <rientjes@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-07-29 06:44:55 +08:00
return false;
if (atomic_read(&mm->mm_users) <= 1)
return true;
/*
* Make sure that all tasks which share the mm with the given tasks
* are dying as well to make sure that a) nobody pins its mm and
* b) the task is also reapable by the oom reaper.
*/
rcu_read_lock();
for_each_process(p) {
if (!process_shares_mm(p, mm))
continue;
if (same_thread_group(task, p))
continue;
ret = __task_will_free_mem(p);
if (!ret)
break;
}
rcu_read_unlock();
return ret;
}
mm, oom: remove 'prefer children over parent' heuristic Since the start of the git history of Linux, the kernel after selecting the worst process to be oom-killed, prefer to kill its child (if the child does not share mm with the parent). Later it was changed to prefer to kill a child who is worst. If the parent is still the worst then the parent will be killed. This heuristic assumes that the children did less work than their parent and by killing one of them, the work lost will be less. However this is very workload dependent. If there is a workload which can benefit from this heuristic, can use oom_score_adj to prefer children to be killed before the parent. The select_bad_process() has already selected the worst process in the system/memcg. There is no need to recheck the badness of its children and hoping to find a worse candidate. That's a lot of unneeded racy work. Also the heuristic is dangerous because it make fork bomb like workloads to recover much later because we constantly pick and kill processes which are not memory hogs. So, let's remove this whole heuristic. [akpm@linux-foundation.org: coding-style fixes] Link: http://lkml.kernel.org/r/20190121215850.221745-2-shakeelb@google.com Signed-off-by: Shakeel Butt <shakeelb@google.com> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: Roman Gushchin <guro@fb.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: David Rientjes <rientjes@google.com> Cc: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-03-06 07:46:12 +08:00
static void __oom_kill_process(struct task_struct *victim, const char *message)
{
mm, oom: refactor oom_kill_process() Patch series "introduce memory.oom.group", v2. This is a tiny implementation of cgroup-aware OOM killer, which adds an ability to kill a cgroup as a single unit and so guarantee the integrity of the workload. Although it has only a limited functionality in comparison to what now resides in the mm tree (it doesn't change the victim task selection algorithm, doesn't look at memory stas on cgroup level, etc), it's also much simpler and more straightforward. So, hopefully, we can avoid having long debates here, as we had with the full implementation. As it doesn't prevent any futher development, and implements an useful and complete feature, it looks as a sane way forward. This patch (of 2): oom_kill_process() consists of two logical parts: the first one is responsible for considering task's children as a potential victim and printing the debug information. The second half is responsible for sending SIGKILL to all tasks sharing the mm struct with the given victim. This commit splits oom_kill_process() with an intention to re-use the the second half: __oom_kill_process(). The cgroup-aware OOM killer will kill multiple tasks belonging to the victim cgroup. We don't need to print the debug information for the each task, as well as play with task selection (considering task's children), so we can't use the existing oom_kill_process(). Link: http://lkml.kernel.org/r/20171130152824.1591-2-guro@fb.com Link: http://lkml.kernel.org/r/20180802003201.817-3-guro@fb.com Signed-off-by: Roman Gushchin <guro@fb.com> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Acked-by: David Rientjes <rientjes@google.com> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: David Rientjes <rientjes@google.com> Cc: Tejun Heo <tj@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-08-22 12:53:50 +08:00
struct task_struct *p;
struct mm_struct *mm;
bool can_oom_reap = true;
p = find_lock_task_mm(victim);
if (!p) {
pr_info("%s: OOM victim %d (%s) is already exiting. Skip killing the task\n",
message, task_pid_nr(victim), victim->comm);
put_task_struct(victim);
return;
} else if (victim != p) {
get_task_struct(p);
put_task_struct(victim);
victim = p;
}
/* Get a reference to safely compare mm after task_unlock(victim) */
mm = victim->mm;
mmgrab(mm);
/* Raise event before sending signal: task reaper must see this */
count_vm_event(OOM_KILL);
memcg_memory_event_mm(mm, MEMCG_OOM_KILL);
mm/oom_kill.c: reverse the order of setting TIF_MEMDIE and sending SIGKILL It was confirmed that a local unprivileged user can consume all memory reserves and hang up that system using time lag between the OOM killer sets TIF_MEMDIE on an OOM victim and sends SIGKILL to that victim, for printk() inside for_each_process() loop at oom_kill_process() can consume many seconds when there are many thread groups sharing the same memory. Before starting oom-depleter process: Node 0 DMA: 3*4kB (UM) 6*8kB (U) 4*16kB (UEM) 0*32kB 0*64kB 1*128kB (M) 2*256kB (EM) 2*512kB (UE) 2*1024kB (EM) 1*2048kB (E) 1*4096kB (M) = 9980kB Node 0 DMA32: 31*4kB (UEM) 27*8kB (UE) 32*16kB (UE) 13*32kB (UE) 14*64kB (UM) 7*128kB (UM) 8*256kB (UM) 8*512kB (UM) 3*1024kB (U) 4*2048kB (UM) 362*4096kB (UM) = 1503220kB As of invoking the OOM killer: Node 0 DMA: 11*4kB (UE) 8*8kB (UEM) 6*16kB (UE) 2*32kB (EM) 0*64kB 1*128kB (U) 3*256kB (UEM) 2*512kB (UE) 3*1024kB (UEM) 1*2048kB (U) 0*4096kB = 7308kB Node 0 DMA32: 1049*4kB (UEM) 507*8kB (UE) 151*16kB (UE) 53*32kB (UEM) 83*64kB (UEM) 52*128kB (EM) 25*256kB (UEM) 11*512kB (M) 6*1024kB (UM) 1*2048kB (M) 0*4096kB = 44556kB Between the thread group leader got TIF_MEMDIE and receives SIGKILL: Node 0 DMA: 0*4kB 0*8kB 0*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 0kB Node 0 DMA32: 0*4kB 0*8kB 0*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 0kB The oom-depleter's thread group leader which got TIF_MEMDIE started memset() in user space after the OOM killer set TIF_MEMDIE, and it was free to abuse ALLOC_NO_WATERMARKS by TIF_MEMDIE for memset() in user space until SIGKILL is delivered. If SIGKILL is delivered before TIF_MEMDIE is set, the oom-depleter can terminate without touching memory reserves. Although the possibility of hitting this time lag is very small for 3.19 and earlier kernels because TIF_MEMDIE is set immediately before sending SIGKILL, preemption or long interrupts (an extreme example is SysRq-t) can step between and allow memory allocations which are not needed for terminating the OOM victim. Fixes: 83363b917a29 ("oom: make sure that TIF_MEMDIE is set under task_lock") Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Acked-by: Michal Hocko <mhocko@suse.com> Cc: David Rientjes <rientjes@google.com> Cc: <stable@vger.kernel.org> [4.0+] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-11-06 10:47:44 +08:00
/*
mm, oom: do not rely on TIF_MEMDIE for memory reserves access For ages we have been relying on TIF_MEMDIE thread flag to mark OOM victims and then, among other things, to give these threads full access to memory reserves. There are few shortcomings of this implementation, though. First of all and the most serious one is that the full access to memory reserves is quite dangerous because we leave no safety room for the system to operate and potentially do last emergency steps to move on. Secondly this flag is per task_struct while the OOM killer operates on mm_struct granularity so all processes sharing the given mm are killed. Giving the full access to all these task_structs could lead to a quick memory reserves depletion. We have tried to reduce this risk by giving TIF_MEMDIE only to the main thread and the currently allocating task but that doesn't really solve this problem while it surely opens up a room for corner cases - e.g. GFP_NO{FS,IO} requests might loop inside the allocator without access to memory reserves because a particular thread was not the group leader. Now that we have the oom reaper and that all oom victims are reapable after 1b51e65eab64 ("oom, oom_reaper: allow to reap mm shared by the kthreads") we can be more conservative and grant only partial access to memory reserves because there are reasonable chances of the parallel memory freeing. We still want some access to reserves because we do not want other consumers to eat up the victim's freed memory. oom victims will still contend with __GFP_HIGH users but those shouldn't be so aggressive to starve oom victims completely. Introduce ALLOC_OOM flag and give all tsk_is_oom_victim tasks access to the half of the reserves. This makes the access to reserves independent on which task has passed through mark_oom_victim. Also drop any usage of TIF_MEMDIE from the page allocator proper and replace it by tsk_is_oom_victim as well which will make page_alloc.c completely TIF_MEMDIE free finally. CONFIG_MMU=n doesn't have oom reaper so let's stick to the original ALLOC_NO_WATERMARKS approach. There is a demand to make the oom killer memcg aware which will imply many tasks killed at once. This change will allow such a usecase without worrying about complete memory reserves depletion. Link: http://lkml.kernel.org/r/20170810075019.28998-2-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Acked-by: Mel Gorman <mgorman@techsingularity.net> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Roman Gushchin <guro@fb.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-09-07 07:24:50 +08:00
* We should send SIGKILL before granting access to memory reserves
* in order to prevent the OOM victim from depleting the memory
* reserves from the user space under its control.
mm/oom_kill.c: reverse the order of setting TIF_MEMDIE and sending SIGKILL It was confirmed that a local unprivileged user can consume all memory reserves and hang up that system using time lag between the OOM killer sets TIF_MEMDIE on an OOM victim and sends SIGKILL to that victim, for printk() inside for_each_process() loop at oom_kill_process() can consume many seconds when there are many thread groups sharing the same memory. Before starting oom-depleter process: Node 0 DMA: 3*4kB (UM) 6*8kB (U) 4*16kB (UEM) 0*32kB 0*64kB 1*128kB (M) 2*256kB (EM) 2*512kB (UE) 2*1024kB (EM) 1*2048kB (E) 1*4096kB (M) = 9980kB Node 0 DMA32: 31*4kB (UEM) 27*8kB (UE) 32*16kB (UE) 13*32kB (UE) 14*64kB (UM) 7*128kB (UM) 8*256kB (UM) 8*512kB (UM) 3*1024kB (U) 4*2048kB (UM) 362*4096kB (UM) = 1503220kB As of invoking the OOM killer: Node 0 DMA: 11*4kB (UE) 8*8kB (UEM) 6*16kB (UE) 2*32kB (EM) 0*64kB 1*128kB (U) 3*256kB (UEM) 2*512kB (UE) 3*1024kB (UEM) 1*2048kB (U) 0*4096kB = 7308kB Node 0 DMA32: 1049*4kB (UEM) 507*8kB (UE) 151*16kB (UE) 53*32kB (UEM) 83*64kB (UEM) 52*128kB (EM) 25*256kB (UEM) 11*512kB (M) 6*1024kB (UM) 1*2048kB (M) 0*4096kB = 44556kB Between the thread group leader got TIF_MEMDIE and receives SIGKILL: Node 0 DMA: 0*4kB 0*8kB 0*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 0kB Node 0 DMA32: 0*4kB 0*8kB 0*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 0kB The oom-depleter's thread group leader which got TIF_MEMDIE started memset() in user space after the OOM killer set TIF_MEMDIE, and it was free to abuse ALLOC_NO_WATERMARKS by TIF_MEMDIE for memset() in user space until SIGKILL is delivered. If SIGKILL is delivered before TIF_MEMDIE is set, the oom-depleter can terminate without touching memory reserves. Although the possibility of hitting this time lag is very small for 3.19 and earlier kernels because TIF_MEMDIE is set immediately before sending SIGKILL, preemption or long interrupts (an extreme example is SysRq-t) can step between and allow memory allocations which are not needed for terminating the OOM victim. Fixes: 83363b917a29 ("oom: make sure that TIF_MEMDIE is set under task_lock") Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Acked-by: Michal Hocko <mhocko@suse.com> Cc: David Rientjes <rientjes@google.com> Cc: <stable@vger.kernel.org> [4.0+] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-11-06 10:47:44 +08:00
*/
do_send_sig_info(SIGKILL, SEND_SIG_PRIV, victim, PIDTYPE_TGID);
mark_oom_victim(victim);
mm/oom: add oom_score_adj and pgtables to Killed process message For an OOM event: print oom_score_adj value for the OOM Killed process to document what the oom score adjust value was at the time the process was OOM Killed. The adjustment value can be set by user code and it affects the resulting oom_score so it is used to influence kill process selection. When eligible tasks are not printed (sysctl oom_dump_tasks = 0) printing this value is the only documentation of the value for the process being killed. Having this value on the Killed process message is useful to document if a miscconfiguration occurred or to confirm that the oom_score_adj configuration applies as expected. An example which illustates both misconfiguration and validation that the oom_score_adj was applied as expected is: Aug 14 23:00:02 testserver kernel: Out of memory: Killed process 2692 (systemd-udevd) total-vm:1056800kB, anon-rss:1052760kB, file-rss:4kB, shmem-rss:0kB pgtables:22kB oom_score_adj:1000 The systemd-udevd is a critical system application that should have an oom_score_adj of -1000. It was miconfigured to have a adjustment of 1000 making it a highly favored OOM kill target process. The output documents both the misconfiguration and the fact that the process was correctly targeted by OOM due to the miconfiguration. This can be quite helpful for triage and problem determination. The addition of the pgtables_bytes shows page table usage by the process and is a useful measure of the memory size of the process. Link: http://lkml.kernel.org/r/20190822173157.1569-1-echron@arista.com Signed-off-by: Edward Chron <echron@arista.com> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: David Rientjes <rientjes@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-09-24 06:37:11 +08:00
pr_err("%s: Killed process %d (%s) total-vm:%lukB, anon-rss:%lukB, file-rss:%lukB, shmem-rss:%lukB, UID:%u pgtables:%lukB oom_score_adj:%hd\n",
message, task_pid_nr(victim), victim->comm, K(mm->total_vm),
K(get_mm_counter(mm, MM_ANONPAGES)),
K(get_mm_counter(mm, MM_FILEPAGES)),
K(get_mm_counter(mm, MM_SHMEMPAGES)),
from_kuid(&init_user_ns, task_uid(victim)),
mm_pgtables_bytes(mm) >> 10, victim->signal->oom_score_adj);
task_unlock(victim);
/*
* Kill all user processes sharing victim->mm in other thread groups, if
* any. They don't get access to memory reserves, though, to avoid
* depletion of all memory. This prevents mm->mmap_lock livelock when an
* oom killed thread cannot exit because it requires the semaphore and
* its contended by another thread trying to allocate memory itself.
* That thread will now get access to memory reserves since it has a
* pending fatal signal.
*/
rcu_read_lock();
for_each_process(p) {
if (!process_shares_mm(p, mm))
continue;
if (same_thread_group(p, victim))
continue;
if (is_global_init(p)) {
mm, oom: introduce oom reaper This patch (of 5): This is based on the idea from Mel Gorman discussed during LSFMM 2015 and independently brought up by Oleg Nesterov. The OOM killer currently allows to kill only a single task in a good hope that the task will terminate in a reasonable time and frees up its memory. Such a task (oom victim) will get an access to memory reserves via mark_oom_victim to allow a forward progress should there be a need for additional memory during exit path. It has been shown (e.g. by Tetsuo Handa) that it is not that hard to construct workloads which break the core assumption mentioned above and the OOM victim might take unbounded amount of time to exit because it might be blocked in the uninterruptible state waiting for an event (e.g. lock) which is blocked by another task looping in the page allocator. This patch reduces the probability of such a lockup by introducing a specialized kernel thread (oom_reaper) which tries to reclaim additional memory by preemptively reaping the anonymous or swapped out memory owned by the oom victim under an assumption that such a memory won't be needed when its owner is killed and kicked from the userspace anyway. There is one notable exception to this, though, if the OOM victim was in the process of coredumping the result would be incomplete. This is considered a reasonable constrain because the overall system health is more important than debugability of a particular application. A kernel thread has been chosen because we need a reliable way of invocation so workqueue context is not appropriate because all the workers might be busy (e.g. allocating memory). Kswapd which sounds like another good fit is not appropriate as well because it might get blocked on locks during reclaim as well. oom_reaper has to take mmap_sem on the target task for reading so the solution is not 100% because the semaphore might be held or blocked for write but the probability is reduced considerably wrt. basically any lock blocking forward progress as described above. In order to prevent from blocking on the lock without any forward progress we are using only a trylock and retry 10 times with a short sleep in between. Users of mmap_sem which need it for write should be carefully reviewed to use _killable waiting as much as possible and reduce allocations requests done with the lock held to absolute minimum to reduce the risk even further. The API between oom killer and oom reaper is quite trivial. wake_oom_reaper updates mm_to_reap with cmpxchg to guarantee only NULL->mm transition and oom_reaper clear this atomically once it is done with the work. This means that only a single mm_struct can be reaped at the time. As the operation is potentially disruptive we are trying to limit it to the ncessary minimum and the reaper blocks any updates while it operates on an mm. mm_struct is pinned by mm_count to allow parallel exit_mmap and a race is detected by atomic_inc_not_zero(mm_users). Signed-off-by: Michal Hocko <mhocko@suse.com> Suggested-by: Oleg Nesterov <oleg@redhat.com> Suggested-by: Mel Gorman <mgorman@suse.de> Acked-by: Mel Gorman <mgorman@suse.de> Acked-by: David Rientjes <rientjes@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Hugh Dickins <hughd@google.com> Cc: Andrea Argangeli <andrea@kernel.org> Cc: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:24 +08:00
can_oom_reap = false;
set_bit(MMF_OOM_SKIP, &mm->flags);
pr_info("oom killer %d (%s) has mm pinned by %d (%s)\n",
task_pid_nr(victim), victim->comm,
task_pid_nr(p), p->comm);
continue;
mm, oom: introduce oom reaper This patch (of 5): This is based on the idea from Mel Gorman discussed during LSFMM 2015 and independently brought up by Oleg Nesterov. The OOM killer currently allows to kill only a single task in a good hope that the task will terminate in a reasonable time and frees up its memory. Such a task (oom victim) will get an access to memory reserves via mark_oom_victim to allow a forward progress should there be a need for additional memory during exit path. It has been shown (e.g. by Tetsuo Handa) that it is not that hard to construct workloads which break the core assumption mentioned above and the OOM victim might take unbounded amount of time to exit because it might be blocked in the uninterruptible state waiting for an event (e.g. lock) which is blocked by another task looping in the page allocator. This patch reduces the probability of such a lockup by introducing a specialized kernel thread (oom_reaper) which tries to reclaim additional memory by preemptively reaping the anonymous or swapped out memory owned by the oom victim under an assumption that such a memory won't be needed when its owner is killed and kicked from the userspace anyway. There is one notable exception to this, though, if the OOM victim was in the process of coredumping the result would be incomplete. This is considered a reasonable constrain because the overall system health is more important than debugability of a particular application. A kernel thread has been chosen because we need a reliable way of invocation so workqueue context is not appropriate because all the workers might be busy (e.g. allocating memory). Kswapd which sounds like another good fit is not appropriate as well because it might get blocked on locks during reclaim as well. oom_reaper has to take mmap_sem on the target task for reading so the solution is not 100% because the semaphore might be held or blocked for write but the probability is reduced considerably wrt. basically any lock blocking forward progress as described above. In order to prevent from blocking on the lock without any forward progress we are using only a trylock and retry 10 times with a short sleep in between. Users of mmap_sem which need it for write should be carefully reviewed to use _killable waiting as much as possible and reduce allocations requests done with the lock held to absolute minimum to reduce the risk even further. The API between oom killer and oom reaper is quite trivial. wake_oom_reaper updates mm_to_reap with cmpxchg to guarantee only NULL->mm transition and oom_reaper clear this atomically once it is done with the work. This means that only a single mm_struct can be reaped at the time. As the operation is potentially disruptive we are trying to limit it to the ncessary minimum and the reaper blocks any updates while it operates on an mm. mm_struct is pinned by mm_count to allow parallel exit_mmap and a race is detected by atomic_inc_not_zero(mm_users). Signed-off-by: Michal Hocko <mhocko@suse.com> Suggested-by: Oleg Nesterov <oleg@redhat.com> Suggested-by: Mel Gorman <mgorman@suse.de> Acked-by: Mel Gorman <mgorman@suse.de> Acked-by: David Rientjes <rientjes@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Hugh Dickins <hughd@google.com> Cc: Andrea Argangeli <andrea@kernel.org> Cc: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:24 +08:00
}
/*
* No kthead_use_mm() user needs to read from the userspace so
* we are ok to reap it.
*/
if (unlikely(p->flags & PF_KTHREAD))
continue;
do_send_sig_info(SIGKILL, SEND_SIG_PRIV, p, PIDTYPE_TGID);
}
rcu_read_unlock();
mm, oom: introduce oom reaper This patch (of 5): This is based on the idea from Mel Gorman discussed during LSFMM 2015 and independently brought up by Oleg Nesterov. The OOM killer currently allows to kill only a single task in a good hope that the task will terminate in a reasonable time and frees up its memory. Such a task (oom victim) will get an access to memory reserves via mark_oom_victim to allow a forward progress should there be a need for additional memory during exit path. It has been shown (e.g. by Tetsuo Handa) that it is not that hard to construct workloads which break the core assumption mentioned above and the OOM victim might take unbounded amount of time to exit because it might be blocked in the uninterruptible state waiting for an event (e.g. lock) which is blocked by another task looping in the page allocator. This patch reduces the probability of such a lockup by introducing a specialized kernel thread (oom_reaper) which tries to reclaim additional memory by preemptively reaping the anonymous or swapped out memory owned by the oom victim under an assumption that such a memory won't be needed when its owner is killed and kicked from the userspace anyway. There is one notable exception to this, though, if the OOM victim was in the process of coredumping the result would be incomplete. This is considered a reasonable constrain because the overall system health is more important than debugability of a particular application. A kernel thread has been chosen because we need a reliable way of invocation so workqueue context is not appropriate because all the workers might be busy (e.g. allocating memory). Kswapd which sounds like another good fit is not appropriate as well because it might get blocked on locks during reclaim as well. oom_reaper has to take mmap_sem on the target task for reading so the solution is not 100% because the semaphore might be held or blocked for write but the probability is reduced considerably wrt. basically any lock blocking forward progress as described above. In order to prevent from blocking on the lock without any forward progress we are using only a trylock and retry 10 times with a short sleep in between. Users of mmap_sem which need it for write should be carefully reviewed to use _killable waiting as much as possible and reduce allocations requests done with the lock held to absolute minimum to reduce the risk even further. The API between oom killer and oom reaper is quite trivial. wake_oom_reaper updates mm_to_reap with cmpxchg to guarantee only NULL->mm transition and oom_reaper clear this atomically once it is done with the work. This means that only a single mm_struct can be reaped at the time. As the operation is potentially disruptive we are trying to limit it to the ncessary minimum and the reaper blocks any updates while it operates on an mm. mm_struct is pinned by mm_count to allow parallel exit_mmap and a race is detected by atomic_inc_not_zero(mm_users). Signed-off-by: Michal Hocko <mhocko@suse.com> Suggested-by: Oleg Nesterov <oleg@redhat.com> Suggested-by: Mel Gorman <mgorman@suse.de> Acked-by: Mel Gorman <mgorman@suse.de> Acked-by: David Rientjes <rientjes@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Hugh Dickins <hughd@google.com> Cc: Andrea Argangeli <andrea@kernel.org> Cc: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:24 +08:00
if (can_oom_reap)
oom: clear TIF_MEMDIE after oom_reaper managed to unmap the address space When oom_reaper manages to unmap all the eligible vmas there shouldn't be much of the freable memory held by the oom victim left anymore so it makes sense to clear the TIF_MEMDIE flag for the victim and allow the OOM killer to select another task. The lack of TIF_MEMDIE also means that the victim cannot access memory reserves anymore but that shouldn't be a problem because it would get the access again if it needs to allocate and hits the OOM killer again due to the fatal_signal_pending resp. PF_EXITING check. We can safely hide the task from the OOM killer because it is clearly not a good candidate anymore as everyhing reclaimable has been torn down already. This patch will allow to cap the time an OOM victim can keep TIF_MEMDIE and thus hold off further global OOM killer actions granted the oom reaper is able to take mmap_sem for the associated mm struct. This is not guaranteed now but further steps should make sure that mmap_sem for write should be blocked killable which will help to reduce such a lock contention. This is not done by this patch. Note that exit_oom_victim might be called on a remote task from __oom_reap_task now so we have to check and clear the flag atomically otherwise we might race and underflow oom_victims or wake up waiters too early. Signed-off-by: Michal Hocko <mhocko@suse.com> Suggested-by: Johannes Weiner <hannes@cmpxchg.org> Suggested-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Andrea Argangeli <andrea@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Hugh Dickins <hughd@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:27 +08:00
wake_oom_reaper(victim);
mm, oom: introduce oom reaper This patch (of 5): This is based on the idea from Mel Gorman discussed during LSFMM 2015 and independently brought up by Oleg Nesterov. The OOM killer currently allows to kill only a single task in a good hope that the task will terminate in a reasonable time and frees up its memory. Such a task (oom victim) will get an access to memory reserves via mark_oom_victim to allow a forward progress should there be a need for additional memory during exit path. It has been shown (e.g. by Tetsuo Handa) that it is not that hard to construct workloads which break the core assumption mentioned above and the OOM victim might take unbounded amount of time to exit because it might be blocked in the uninterruptible state waiting for an event (e.g. lock) which is blocked by another task looping in the page allocator. This patch reduces the probability of such a lockup by introducing a specialized kernel thread (oom_reaper) which tries to reclaim additional memory by preemptively reaping the anonymous or swapped out memory owned by the oom victim under an assumption that such a memory won't be needed when its owner is killed and kicked from the userspace anyway. There is one notable exception to this, though, if the OOM victim was in the process of coredumping the result would be incomplete. This is considered a reasonable constrain because the overall system health is more important than debugability of a particular application. A kernel thread has been chosen because we need a reliable way of invocation so workqueue context is not appropriate because all the workers might be busy (e.g. allocating memory). Kswapd which sounds like another good fit is not appropriate as well because it might get blocked on locks during reclaim as well. oom_reaper has to take mmap_sem on the target task for reading so the solution is not 100% because the semaphore might be held or blocked for write but the probability is reduced considerably wrt. basically any lock blocking forward progress as described above. In order to prevent from blocking on the lock without any forward progress we are using only a trylock and retry 10 times with a short sleep in between. Users of mmap_sem which need it for write should be carefully reviewed to use _killable waiting as much as possible and reduce allocations requests done with the lock held to absolute minimum to reduce the risk even further. The API between oom killer and oom reaper is quite trivial. wake_oom_reaper updates mm_to_reap with cmpxchg to guarantee only NULL->mm transition and oom_reaper clear this atomically once it is done with the work. This means that only a single mm_struct can be reaped at the time. As the operation is potentially disruptive we are trying to limit it to the ncessary minimum and the reaper blocks any updates while it operates on an mm. mm_struct is pinned by mm_count to allow parallel exit_mmap and a race is detected by atomic_inc_not_zero(mm_users). Signed-off-by: Michal Hocko <mhocko@suse.com> Suggested-by: Oleg Nesterov <oleg@redhat.com> Suggested-by: Mel Gorman <mgorman@suse.de> Acked-by: Mel Gorman <mgorman@suse.de> Acked-by: David Rientjes <rientjes@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Hugh Dickins <hughd@google.com> Cc: Andrea Argangeli <andrea@kernel.org> Cc: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-26 05:20:24 +08:00
mmdrop(mm);
put_task_struct(victim);
}
#undef K
mm, oom: introduce memory.oom.group For some workloads an intervention from the OOM killer can be painful. Killing a random task can bring the workload into an inconsistent state. Historically, there are two common solutions for this problem: 1) enabling panic_on_oom, 2) using a userspace daemon to monitor OOMs and kill all outstanding processes. Both approaches have their downsides: rebooting on each OOM is an obvious waste of capacity, and handling all in userspace is tricky and requires a userspace agent, which will monitor all cgroups for OOMs. In most cases an in-kernel after-OOM cleaning-up mechanism can eliminate the necessity of enabling panic_on_oom. Also, it can simplify the cgroup management for userspace applications. This commit introduces a new knob for cgroup v2 memory controller: memory.oom.group. The knob determines whether the cgroup should be treated as an indivisible workload by the OOM killer. If set, all tasks belonging to the cgroup or to its descendants (if the memory cgroup is not a leaf cgroup) are killed together or not at all. To determine which cgroup has to be killed, we do traverse the cgroup hierarchy from the victim task's cgroup up to the OOMing cgroup (or root) and looking for the highest-level cgroup with memory.oom.group set. Tasks with the OOM protection (oom_score_adj set to -1000) are treated as an exception and are never killed. This patch doesn't change the OOM victim selection algorithm. Link: http://lkml.kernel.org/r/20180802003201.817-4-guro@fb.com Signed-off-by: Roman Gushchin <guro@fb.com> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: David Rientjes <rientjes@google.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Tejun Heo <tj@kernel.org> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-08-22 12:53:54 +08:00
/*
* Kill provided task unless it's secured by setting
* oom_score_adj to OOM_SCORE_ADJ_MIN.
*/
mm, oom: remove 'prefer children over parent' heuristic Since the start of the git history of Linux, the kernel after selecting the worst process to be oom-killed, prefer to kill its child (if the child does not share mm with the parent). Later it was changed to prefer to kill a child who is worst. If the parent is still the worst then the parent will be killed. This heuristic assumes that the children did less work than their parent and by killing one of them, the work lost will be less. However this is very workload dependent. If there is a workload which can benefit from this heuristic, can use oom_score_adj to prefer children to be killed before the parent. The select_bad_process() has already selected the worst process in the system/memcg. There is no need to recheck the badness of its children and hoping to find a worse candidate. That's a lot of unneeded racy work. Also the heuristic is dangerous because it make fork bomb like workloads to recover much later because we constantly pick and kill processes which are not memory hogs. So, let's remove this whole heuristic. [akpm@linux-foundation.org: coding-style fixes] Link: http://lkml.kernel.org/r/20190121215850.221745-2-shakeelb@google.com Signed-off-by: Shakeel Butt <shakeelb@google.com> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: Roman Gushchin <guro@fb.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: David Rientjes <rientjes@google.com> Cc: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-03-06 07:46:12 +08:00
static int oom_kill_memcg_member(struct task_struct *task, void *message)
mm, oom: introduce memory.oom.group For some workloads an intervention from the OOM killer can be painful. Killing a random task can bring the workload into an inconsistent state. Historically, there are two common solutions for this problem: 1) enabling panic_on_oom, 2) using a userspace daemon to monitor OOMs and kill all outstanding processes. Both approaches have their downsides: rebooting on each OOM is an obvious waste of capacity, and handling all in userspace is tricky and requires a userspace agent, which will monitor all cgroups for OOMs. In most cases an in-kernel after-OOM cleaning-up mechanism can eliminate the necessity of enabling panic_on_oom. Also, it can simplify the cgroup management for userspace applications. This commit introduces a new knob for cgroup v2 memory controller: memory.oom.group. The knob determines whether the cgroup should be treated as an indivisible workload by the OOM killer. If set, all tasks belonging to the cgroup or to its descendants (if the memory cgroup is not a leaf cgroup) are killed together or not at all. To determine which cgroup has to be killed, we do traverse the cgroup hierarchy from the victim task's cgroup up to the OOMing cgroup (or root) and looking for the highest-level cgroup with memory.oom.group set. Tasks with the OOM protection (oom_score_adj set to -1000) are treated as an exception and are never killed. This patch doesn't change the OOM victim selection algorithm. Link: http://lkml.kernel.org/r/20180802003201.817-4-guro@fb.com Signed-off-by: Roman Gushchin <guro@fb.com> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: David Rientjes <rientjes@google.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Tejun Heo <tj@kernel.org> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-08-22 12:53:54 +08:00
{
mm,oom: don't kill global init via memory.oom.group Since setting global init process to some memory cgroup is technically possible, oom_kill_memcg_member() must check it. Tasks in /test1 are going to be killed due to memory.oom.group set Memory cgroup out of memory: Killed process 1 (systemd) total-vm:43400kB, anon-rss:1228kB, file-rss:3992kB, shmem-rss:0kB oom_reaper: reaped process 1 (systemd), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kB Kernel panic - not syncing: Attempted to kill init! exitcode=0x0000008b #include <stdio.h> #include <string.h> #include <unistd.h> #include <sys/types.h> #include <sys/stat.h> #include <fcntl.h> int main(int argc, char *argv[]) { static char buffer[10485760]; static int pipe_fd[2] = { EOF, EOF }; unsigned int i; int fd; char buf[64] = { }; if (pipe(pipe_fd)) return 1; if (chdir("/sys/fs/cgroup/")) return 1; fd = open("cgroup.subtree_control", O_WRONLY); write(fd, "+memory", 7); close(fd); mkdir("test1", 0755); fd = open("test1/memory.oom.group", O_WRONLY); write(fd, "1", 1); close(fd); fd = open("test1/cgroup.procs", O_WRONLY); write(fd, "1", 1); snprintf(buf, sizeof(buf) - 1, "%d", getpid()); write(fd, buf, strlen(buf)); close(fd); snprintf(buf, sizeof(buf) - 1, "%lu", sizeof(buffer) * 5); fd = open("test1/memory.max", O_WRONLY); write(fd, buf, strlen(buf)); close(fd); for (i = 0; i < 10; i++) if (fork() == 0) { char c; close(pipe_fd[1]); read(pipe_fd[0], &c, 1); memset(buffer, 0, sizeof(buffer)); sleep(3); _exit(0); } close(pipe_fd[0]); close(pipe_fd[1]); sleep(3); return 0; } [ 37.052923][ T9185] a.out invoked oom-killer: gfp_mask=0xcc0(GFP_KERNEL), order=0, oom_score_adj=0 [ 37.056169][ T9185] CPU: 4 PID: 9185 Comm: a.out Kdump: loaded Not tainted 5.0.0-rc4-next-20190131 #280 [ 37.059205][ T9185] Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS 6.00 04/13/2018 [ 37.062954][ T9185] Call Trace: [ 37.063976][ T9185] dump_stack+0x67/0x95 [ 37.065263][ T9185] dump_header+0x51/0x570 [ 37.066619][ T9185] ? trace_hardirqs_on+0x3f/0x110 [ 37.068171][ T9185] ? _raw_spin_unlock_irqrestore+0x3d/0x70 [ 37.069967][ T9185] oom_kill_process+0x18d/0x210 [ 37.071515][ T9185] out_of_memory+0x11b/0x380 [ 37.072936][ T9185] mem_cgroup_out_of_memory+0xb6/0xd0 [ 37.074601][ T9185] try_charge+0x790/0x820 [ 37.076021][ T9185] mem_cgroup_try_charge+0x42/0x1d0 [ 37.077629][ T9185] mem_cgroup_try_charge_delay+0x11/0x30 [ 37.079370][ T9185] do_anonymous_page+0x105/0x5e0 [ 37.080939][ T9185] __handle_mm_fault+0x9cb/0x1070 [ 37.082485][ T9185] handle_mm_fault+0x1b2/0x3a0 [ 37.083819][ T9185] ? handle_mm_fault+0x47/0x3a0 [ 37.085181][ T9185] __do_page_fault+0x255/0x4c0 [ 37.086529][ T9185] do_page_fault+0x28/0x260 [ 37.087788][ T9185] ? page_fault+0x8/0x30 [ 37.088978][ T9185] page_fault+0x1e/0x30 [ 37.090142][ T9185] RIP: 0033:0x7f8b183aefe0 [ 37.091433][ T9185] Code: 20 f3 44 0f 7f 44 17 d0 f3 44 0f 7f 47 30 f3 44 0f 7f 44 17 c0 48 01 fa 48 83 e2 c0 48 39 d1 74 a3 66 0f 1f 84 00 00 00 00 00 <66> 44 0f 7f 01 66 44 0f 7f 41 10 66 44 0f 7f 41 20 66 44 0f 7f 41 [ 37.096917][ T9185] RSP: 002b:00007fffc5d329e8 EFLAGS: 00010206 [ 37.098615][ T9185] RAX: 00000000006010e0 RBX: 0000000000000008 RCX: 0000000000c30000 [ 37.100905][ T9185] RDX: 00000000010010c0 RSI: 0000000000000000 RDI: 00000000006010e0 [ 37.103349][ T9185] RBP: 0000000000000000 R08: 00007f8b188f4740 R09: 0000000000000000 [ 37.105797][ T9185] R10: 00007fffc5d32420 R11: 00007f8b183aef40 R12: 0000000000000005 [ 37.108228][ T9185] R13: 0000000000000000 R14: ffffffffffffffff R15: 0000000000000000 [ 37.110840][ T9185] memory: usage 51200kB, limit 51200kB, failcnt 125 [ 37.113045][ T9185] memory+swap: usage 0kB, limit 9007199254740988kB, failcnt 0 [ 37.115808][ T9185] kmem: usage 0kB, limit 9007199254740988kB, failcnt 0 [ 37.117660][ T9185] Memory cgroup stats for /test1: cache:0KB rss:49484KB rss_huge:30720KB shmem:0KB mapped_file:0KB dirty:0KB writeback:0KB inactive_anon:0KB active_anon:49700KB inactive_file:0KB active_file:0KB unevictable:0KB [ 37.123371][ T9185] oom-kill:constraint=CONSTRAINT_NONE,nodemask=(null),cpuset=/,mems_allowed=0,oom_memcg=/test1,task_memcg=/test1,task=a.out,pid=9188,uid=0 [ 37.128158][ T9185] Memory cgroup out of memory: Killed process 9188 (a.out) total-vm:14456kB, anon-rss:10324kB, file-rss:504kB, shmem-rss:0kB [ 37.132710][ T9185] Tasks in /test1 are going to be killed due to memory.oom.group set [ 37.132833][ T54] oom_reaper: reaped process 9188 (a.out), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kB [ 37.135498][ T9185] Memory cgroup out of memory: Killed process 1 (systemd) total-vm:43400kB, anon-rss:1228kB, file-rss:3992kB, shmem-rss:0kB [ 37.143434][ T9185] Memory cgroup out of memory: Killed process 9182 (a.out) total-vm:14456kB, anon-rss:76kB, file-rss:588kB, shmem-rss:0kB [ 37.144328][ T54] oom_reaper: reaped process 1 (systemd), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kB [ 37.147585][ T9185] Memory cgroup out of memory: Killed process 9183 (a.out) total-vm:14456kB, anon-rss:6228kB, file-rss:512kB, shmem-rss:0kB [ 37.157222][ T9185] Memory cgroup out of memory: Killed process 9184 (a.out) total-vm:14456kB, anon-rss:6228kB, file-rss:508kB, shmem-rss:0kB [ 37.157259][ T9185] Memory cgroup out of memory: Killed process 9185 (a.out) total-vm:14456kB, anon-rss:6228kB, file-rss:512kB, shmem-rss:0kB [ 37.157291][ T9185] Memory cgroup out of memory: Killed process 9186 (a.out) total-vm:14456kB, anon-rss:4180kB, file-rss:508kB, shmem-rss:0kB [ 37.157306][ T54] oom_reaper: reaped process 9183 (a.out), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kB [ 37.157328][ T9185] Memory cgroup out of memory: Killed process 9187 (a.out) total-vm:14456kB, anon-rss:4180kB, file-rss:512kB, shmem-rss:0kB [ 37.157452][ T9185] Memory cgroup out of memory: Killed process 9189 (a.out) total-vm:14456kB, anon-rss:6228kB, file-rss:512kB, shmem-rss:0kB [ 37.158733][ T9185] Memory cgroup out of memory: Killed process 9190 (a.out) total-vm:14456kB, anon-rss:552kB, file-rss:512kB, shmem-rss:0kB [ 37.160083][ T54] oom_reaper: reaped process 9186 (a.out), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kB [ 37.160187][ T54] oom_reaper: reaped process 9189 (a.out), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kB [ 37.206941][ T54] oom_reaper: reaped process 9185 (a.out), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kB [ 37.212300][ T9185] Memory cgroup out of memory: Killed process 9191 (a.out) total-vm:14456kB, anon-rss:4180kB, file-rss:512kB, shmem-rss:0kB [ 37.212317][ T54] oom_reaper: reaped process 9190 (a.out), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kB [ 37.218860][ T9185] Memory cgroup out of memory: Killed process 9192 (a.out) total-vm:14456kB, anon-rss:1080kB, file-rss:512kB, shmem-rss:0kB [ 37.227667][ T54] oom_reaper: reaped process 9192 (a.out), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kB [ 37.292323][ T9193] abrt-hook-ccpp (9193) used greatest stack depth: 10480 bytes left [ 37.351843][ T1] Kernel panic - not syncing: Attempted to kill init! exitcode=0x0000008b [ 37.354833][ T1] CPU: 7 PID: 1 Comm: systemd Kdump: loaded Not tainted 5.0.0-rc4-next-20190131 #280 [ 37.357876][ T1] Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS 6.00 04/13/2018 [ 37.361685][ T1] Call Trace: [ 37.363239][ T1] dump_stack+0x67/0x95 [ 37.365010][ T1] panic+0xfc/0x2b0 [ 37.366853][ T1] do_exit+0xd55/0xd60 [ 37.368595][ T1] do_group_exit+0x47/0xc0 [ 37.370415][ T1] get_signal+0x32a/0x920 [ 37.372449][ T1] ? _raw_spin_unlock_irqrestore+0x3d/0x70 [ 37.374596][ T1] do_signal+0x32/0x6e0 [ 37.376430][ T1] ? exit_to_usermode_loop+0x26/0x9b [ 37.378418][ T1] ? prepare_exit_to_usermode+0xa8/0xd0 [ 37.380571][ T1] exit_to_usermode_loop+0x3e/0x9b [ 37.382588][ T1] prepare_exit_to_usermode+0xa8/0xd0 [ 37.384594][ T1] ? page_fault+0x8/0x30 [ 37.386453][ T1] retint_user+0x8/0x18 [ 37.388160][ T1] RIP: 0033:0x7f42c06974a8 [ 37.389922][ T1] Code: Bad RIP value. [ 37.391788][ T1] RSP: 002b:00007ffc3effd388 EFLAGS: 00010213 [ 37.394075][ T1] RAX: 000000000000000e RBX: 00007ffc3effd390 RCX: 0000000000000000 [ 37.396963][ T1] RDX: 000000000000002a RSI: 00007ffc3effd390 RDI: 0000000000000004 [ 37.399550][ T1] RBP: 00007ffc3effd680 R08: 0000000000000000 R09: 0000000000000000 [ 37.402334][ T1] R10: 00000000ffffffff R11: 0000000000000246 R12: 0000000000000001 [ 37.404890][ T1] R13: ffffffffffffffff R14: 0000000000000884 R15: 000056460b1ac3b0 Link: http://lkml.kernel.org/r/201902010336.x113a4EO027170@www262.sakura.ne.jp Fixes: 3d8b38eb81cac813 ("mm, oom: introduce memory.oom.group") Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Roman Gushchin <guro@fb.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-03-06 07:48:22 +08:00
if (task->signal->oom_score_adj != OOM_SCORE_ADJ_MIN &&
!is_global_init(task)) {
mm, oom: introduce memory.oom.group For some workloads an intervention from the OOM killer can be painful. Killing a random task can bring the workload into an inconsistent state. Historically, there are two common solutions for this problem: 1) enabling panic_on_oom, 2) using a userspace daemon to monitor OOMs and kill all outstanding processes. Both approaches have their downsides: rebooting on each OOM is an obvious waste of capacity, and handling all in userspace is tricky and requires a userspace agent, which will monitor all cgroups for OOMs. In most cases an in-kernel after-OOM cleaning-up mechanism can eliminate the necessity of enabling panic_on_oom. Also, it can simplify the cgroup management for userspace applications. This commit introduces a new knob for cgroup v2 memory controller: memory.oom.group. The knob determines whether the cgroup should be treated as an indivisible workload by the OOM killer. If set, all tasks belonging to the cgroup or to its descendants (if the memory cgroup is not a leaf cgroup) are killed together or not at all. To determine which cgroup has to be killed, we do traverse the cgroup hierarchy from the victim task's cgroup up to the OOMing cgroup (or root) and looking for the highest-level cgroup with memory.oom.group set. Tasks with the OOM protection (oom_score_adj set to -1000) are treated as an exception and are never killed. This patch doesn't change the OOM victim selection algorithm. Link: http://lkml.kernel.org/r/20180802003201.817-4-guro@fb.com Signed-off-by: Roman Gushchin <guro@fb.com> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: David Rientjes <rientjes@google.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Tejun Heo <tj@kernel.org> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-08-22 12:53:54 +08:00
get_task_struct(task);
mm, oom: remove 'prefer children over parent' heuristic Since the start of the git history of Linux, the kernel after selecting the worst process to be oom-killed, prefer to kill its child (if the child does not share mm with the parent). Later it was changed to prefer to kill a child who is worst. If the parent is still the worst then the parent will be killed. This heuristic assumes that the children did less work than their parent and by killing one of them, the work lost will be less. However this is very workload dependent. If there is a workload which can benefit from this heuristic, can use oom_score_adj to prefer children to be killed before the parent. The select_bad_process() has already selected the worst process in the system/memcg. There is no need to recheck the badness of its children and hoping to find a worse candidate. That's a lot of unneeded racy work. Also the heuristic is dangerous because it make fork bomb like workloads to recover much later because we constantly pick and kill processes which are not memory hogs. So, let's remove this whole heuristic. [akpm@linux-foundation.org: coding-style fixes] Link: http://lkml.kernel.org/r/20190121215850.221745-2-shakeelb@google.com Signed-off-by: Shakeel Butt <shakeelb@google.com> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: Roman Gushchin <guro@fb.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: David Rientjes <rientjes@google.com> Cc: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-03-06 07:46:12 +08:00
__oom_kill_process(task, message);
mm, oom: introduce memory.oom.group For some workloads an intervention from the OOM killer can be painful. Killing a random task can bring the workload into an inconsistent state. Historically, there are two common solutions for this problem: 1) enabling panic_on_oom, 2) using a userspace daemon to monitor OOMs and kill all outstanding processes. Both approaches have their downsides: rebooting on each OOM is an obvious waste of capacity, and handling all in userspace is tricky and requires a userspace agent, which will monitor all cgroups for OOMs. In most cases an in-kernel after-OOM cleaning-up mechanism can eliminate the necessity of enabling panic_on_oom. Also, it can simplify the cgroup management for userspace applications. This commit introduces a new knob for cgroup v2 memory controller: memory.oom.group. The knob determines whether the cgroup should be treated as an indivisible workload by the OOM killer. If set, all tasks belonging to the cgroup or to its descendants (if the memory cgroup is not a leaf cgroup) are killed together or not at all. To determine which cgroup has to be killed, we do traverse the cgroup hierarchy from the victim task's cgroup up to the OOMing cgroup (or root) and looking for the highest-level cgroup with memory.oom.group set. Tasks with the OOM protection (oom_score_adj set to -1000) are treated as an exception and are never killed. This patch doesn't change the OOM victim selection algorithm. Link: http://lkml.kernel.org/r/20180802003201.817-4-guro@fb.com Signed-off-by: Roman Gushchin <guro@fb.com> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: David Rientjes <rientjes@google.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Tejun Heo <tj@kernel.org> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-08-22 12:53:54 +08:00
}
return 0;
}
mm, oom: refactor oom_kill_process() Patch series "introduce memory.oom.group", v2. This is a tiny implementation of cgroup-aware OOM killer, which adds an ability to kill a cgroup as a single unit and so guarantee the integrity of the workload. Although it has only a limited functionality in comparison to what now resides in the mm tree (it doesn't change the victim task selection algorithm, doesn't look at memory stas on cgroup level, etc), it's also much simpler and more straightforward. So, hopefully, we can avoid having long debates here, as we had with the full implementation. As it doesn't prevent any futher development, and implements an useful and complete feature, it looks as a sane way forward. This patch (of 2): oom_kill_process() consists of two logical parts: the first one is responsible for considering task's children as a potential victim and printing the debug information. The second half is responsible for sending SIGKILL to all tasks sharing the mm struct with the given victim. This commit splits oom_kill_process() with an intention to re-use the the second half: __oom_kill_process(). The cgroup-aware OOM killer will kill multiple tasks belonging to the victim cgroup. We don't need to print the debug information for the each task, as well as play with task selection (considering task's children), so we can't use the existing oom_kill_process(). Link: http://lkml.kernel.org/r/20171130152824.1591-2-guro@fb.com Link: http://lkml.kernel.org/r/20180802003201.817-3-guro@fb.com Signed-off-by: Roman Gushchin <guro@fb.com> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Acked-by: David Rientjes <rientjes@google.com> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: David Rientjes <rientjes@google.com> Cc: Tejun Heo <tj@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-08-22 12:53:50 +08:00
static void oom_kill_process(struct oom_control *oc, const char *message)
{
mm, oom: remove 'prefer children over parent' heuristic Since the start of the git history of Linux, the kernel after selecting the worst process to be oom-killed, prefer to kill its child (if the child does not share mm with the parent). Later it was changed to prefer to kill a child who is worst. If the parent is still the worst then the parent will be killed. This heuristic assumes that the children did less work than their parent and by killing one of them, the work lost will be less. However this is very workload dependent. If there is a workload which can benefit from this heuristic, can use oom_score_adj to prefer children to be killed before the parent. The select_bad_process() has already selected the worst process in the system/memcg. There is no need to recheck the badness of its children and hoping to find a worse candidate. That's a lot of unneeded racy work. Also the heuristic is dangerous because it make fork bomb like workloads to recover much later because we constantly pick and kill processes which are not memory hogs. So, let's remove this whole heuristic. [akpm@linux-foundation.org: coding-style fixes] Link: http://lkml.kernel.org/r/20190121215850.221745-2-shakeelb@google.com Signed-off-by: Shakeel Butt <shakeelb@google.com> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: Roman Gushchin <guro@fb.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: David Rientjes <rientjes@google.com> Cc: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-03-06 07:46:12 +08:00
struct task_struct *victim = oc->chosen;
mm, oom: introduce memory.oom.group For some workloads an intervention from the OOM killer can be painful. Killing a random task can bring the workload into an inconsistent state. Historically, there are two common solutions for this problem: 1) enabling panic_on_oom, 2) using a userspace daemon to monitor OOMs and kill all outstanding processes. Both approaches have their downsides: rebooting on each OOM is an obvious waste of capacity, and handling all in userspace is tricky and requires a userspace agent, which will monitor all cgroups for OOMs. In most cases an in-kernel after-OOM cleaning-up mechanism can eliminate the necessity of enabling panic_on_oom. Also, it can simplify the cgroup management for userspace applications. This commit introduces a new knob for cgroup v2 memory controller: memory.oom.group. The knob determines whether the cgroup should be treated as an indivisible workload by the OOM killer. If set, all tasks belonging to the cgroup or to its descendants (if the memory cgroup is not a leaf cgroup) are killed together or not at all. To determine which cgroup has to be killed, we do traverse the cgroup hierarchy from the victim task's cgroup up to the OOMing cgroup (or root) and looking for the highest-level cgroup with memory.oom.group set. Tasks with the OOM protection (oom_score_adj set to -1000) are treated as an exception and are never killed. This patch doesn't change the OOM victim selection algorithm. Link: http://lkml.kernel.org/r/20180802003201.817-4-guro@fb.com Signed-off-by: Roman Gushchin <guro@fb.com> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: David Rientjes <rientjes@google.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Tejun Heo <tj@kernel.org> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-08-22 12:53:54 +08:00
struct mem_cgroup *oom_group;
mm, oom: refactor oom_kill_process() Patch series "introduce memory.oom.group", v2. This is a tiny implementation of cgroup-aware OOM killer, which adds an ability to kill a cgroup as a single unit and so guarantee the integrity of the workload. Although it has only a limited functionality in comparison to what now resides in the mm tree (it doesn't change the victim task selection algorithm, doesn't look at memory stas on cgroup level, etc), it's also much simpler and more straightforward. So, hopefully, we can avoid having long debates here, as we had with the full implementation. As it doesn't prevent any futher development, and implements an useful and complete feature, it looks as a sane way forward. This patch (of 2): oom_kill_process() consists of two logical parts: the first one is responsible for considering task's children as a potential victim and printing the debug information. The second half is responsible for sending SIGKILL to all tasks sharing the mm struct with the given victim. This commit splits oom_kill_process() with an intention to re-use the the second half: __oom_kill_process(). The cgroup-aware OOM killer will kill multiple tasks belonging to the victim cgroup. We don't need to print the debug information for the each task, as well as play with task selection (considering task's children), so we can't use the existing oom_kill_process(). Link: http://lkml.kernel.org/r/20171130152824.1591-2-guro@fb.com Link: http://lkml.kernel.org/r/20180802003201.817-3-guro@fb.com Signed-off-by: Roman Gushchin <guro@fb.com> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Acked-by: David Rientjes <rientjes@google.com> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: David Rientjes <rientjes@google.com> Cc: Tejun Heo <tj@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-08-22 12:53:50 +08:00
static DEFINE_RATELIMIT_STATE(oom_rs, DEFAULT_RATELIMIT_INTERVAL,
DEFAULT_RATELIMIT_BURST);
/*
* If the task is already exiting, don't alarm the sysadmin or kill
* its children or threads, just give it access to memory reserves
* so it can die quickly
*/
mm, oom: remove 'prefer children over parent' heuristic Since the start of the git history of Linux, the kernel after selecting the worst process to be oom-killed, prefer to kill its child (if the child does not share mm with the parent). Later it was changed to prefer to kill a child who is worst. If the parent is still the worst then the parent will be killed. This heuristic assumes that the children did less work than their parent and by killing one of them, the work lost will be less. However this is very workload dependent. If there is a workload which can benefit from this heuristic, can use oom_score_adj to prefer children to be killed before the parent. The select_bad_process() has already selected the worst process in the system/memcg. There is no need to recheck the badness of its children and hoping to find a worse candidate. That's a lot of unneeded racy work. Also the heuristic is dangerous because it make fork bomb like workloads to recover much later because we constantly pick and kill processes which are not memory hogs. So, let's remove this whole heuristic. [akpm@linux-foundation.org: coding-style fixes] Link: http://lkml.kernel.org/r/20190121215850.221745-2-shakeelb@google.com Signed-off-by: Shakeel Butt <shakeelb@google.com> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: Roman Gushchin <guro@fb.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: David Rientjes <rientjes@google.com> Cc: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-03-06 07:46:12 +08:00
task_lock(victim);
if (task_will_free_mem(victim)) {
mark_oom_victim(victim);
wake_oom_reaper(victim);
task_unlock(victim);
put_task_struct(victim);
mm, oom: refactor oom_kill_process() Patch series "introduce memory.oom.group", v2. This is a tiny implementation of cgroup-aware OOM killer, which adds an ability to kill a cgroup as a single unit and so guarantee the integrity of the workload. Although it has only a limited functionality in comparison to what now resides in the mm tree (it doesn't change the victim task selection algorithm, doesn't look at memory stas on cgroup level, etc), it's also much simpler and more straightforward. So, hopefully, we can avoid having long debates here, as we had with the full implementation. As it doesn't prevent any futher development, and implements an useful and complete feature, it looks as a sane way forward. This patch (of 2): oom_kill_process() consists of two logical parts: the first one is responsible for considering task's children as a potential victim and printing the debug information. The second half is responsible for sending SIGKILL to all tasks sharing the mm struct with the given victim. This commit splits oom_kill_process() with an intention to re-use the the second half: __oom_kill_process(). The cgroup-aware OOM killer will kill multiple tasks belonging to the victim cgroup. We don't need to print the debug information for the each task, as well as play with task selection (considering task's children), so we can't use the existing oom_kill_process(). Link: http://lkml.kernel.org/r/20171130152824.1591-2-guro@fb.com Link: http://lkml.kernel.org/r/20180802003201.817-3-guro@fb.com Signed-off-by: Roman Gushchin <guro@fb.com> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Acked-by: David Rientjes <rientjes@google.com> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: David Rientjes <rientjes@google.com> Cc: Tejun Heo <tj@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-08-22 12:53:50 +08:00
return;
}
mm, oom: remove 'prefer children over parent' heuristic Since the start of the git history of Linux, the kernel after selecting the worst process to be oom-killed, prefer to kill its child (if the child does not share mm with the parent). Later it was changed to prefer to kill a child who is worst. If the parent is still the worst then the parent will be killed. This heuristic assumes that the children did less work than their parent and by killing one of them, the work lost will be less. However this is very workload dependent. If there is a workload which can benefit from this heuristic, can use oom_score_adj to prefer children to be killed before the parent. The select_bad_process() has already selected the worst process in the system/memcg. There is no need to recheck the badness of its children and hoping to find a worse candidate. That's a lot of unneeded racy work. Also the heuristic is dangerous because it make fork bomb like workloads to recover much later because we constantly pick and kill processes which are not memory hogs. So, let's remove this whole heuristic. [akpm@linux-foundation.org: coding-style fixes] Link: http://lkml.kernel.org/r/20190121215850.221745-2-shakeelb@google.com Signed-off-by: Shakeel Butt <shakeelb@google.com> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: Roman Gushchin <guro@fb.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: David Rientjes <rientjes@google.com> Cc: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-03-06 07:46:12 +08:00
task_unlock(victim);
mm, oom: refactor oom_kill_process() Patch series "introduce memory.oom.group", v2. This is a tiny implementation of cgroup-aware OOM killer, which adds an ability to kill a cgroup as a single unit and so guarantee the integrity of the workload. Although it has only a limited functionality in comparison to what now resides in the mm tree (it doesn't change the victim task selection algorithm, doesn't look at memory stas on cgroup level, etc), it's also much simpler and more straightforward. So, hopefully, we can avoid having long debates here, as we had with the full implementation. As it doesn't prevent any futher development, and implements an useful and complete feature, it looks as a sane way forward. This patch (of 2): oom_kill_process() consists of two logical parts: the first one is responsible for considering task's children as a potential victim and printing the debug information. The second half is responsible for sending SIGKILL to all tasks sharing the mm struct with the given victim. This commit splits oom_kill_process() with an intention to re-use the the second half: __oom_kill_process(). The cgroup-aware OOM killer will kill multiple tasks belonging to the victim cgroup. We don't need to print the debug information for the each task, as well as play with task selection (considering task's children), so we can't use the existing oom_kill_process(). Link: http://lkml.kernel.org/r/20171130152824.1591-2-guro@fb.com Link: http://lkml.kernel.org/r/20180802003201.817-3-guro@fb.com Signed-off-by: Roman Gushchin <guro@fb.com> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Acked-by: David Rientjes <rientjes@google.com> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: David Rientjes <rientjes@google.com> Cc: Tejun Heo <tj@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-08-22 12:53:50 +08:00
if (__ratelimit(&oom_rs))
mm, oom: remove 'prefer children over parent' heuristic Since the start of the git history of Linux, the kernel after selecting the worst process to be oom-killed, prefer to kill its child (if the child does not share mm with the parent). Later it was changed to prefer to kill a child who is worst. If the parent is still the worst then the parent will be killed. This heuristic assumes that the children did less work than their parent and by killing one of them, the work lost will be less. However this is very workload dependent. If there is a workload which can benefit from this heuristic, can use oom_score_adj to prefer children to be killed before the parent. The select_bad_process() has already selected the worst process in the system/memcg. There is no need to recheck the badness of its children and hoping to find a worse candidate. That's a lot of unneeded racy work. Also the heuristic is dangerous because it make fork bomb like workloads to recover much later because we constantly pick and kill processes which are not memory hogs. So, let's remove this whole heuristic. [akpm@linux-foundation.org: coding-style fixes] Link: http://lkml.kernel.org/r/20190121215850.221745-2-shakeelb@google.com Signed-off-by: Shakeel Butt <shakeelb@google.com> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: Roman Gushchin <guro@fb.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: David Rientjes <rientjes@google.com> Cc: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-03-06 07:46:12 +08:00
dump_header(oc, victim);
mm, oom: refactor oom_kill_process() Patch series "introduce memory.oom.group", v2. This is a tiny implementation of cgroup-aware OOM killer, which adds an ability to kill a cgroup as a single unit and so guarantee the integrity of the workload. Although it has only a limited functionality in comparison to what now resides in the mm tree (it doesn't change the victim task selection algorithm, doesn't look at memory stas on cgroup level, etc), it's also much simpler and more straightforward. So, hopefully, we can avoid having long debates here, as we had with the full implementation. As it doesn't prevent any futher development, and implements an useful and complete feature, it looks as a sane way forward. This patch (of 2): oom_kill_process() consists of two logical parts: the first one is responsible for considering task's children as a potential victim and printing the debug information. The second half is responsible for sending SIGKILL to all tasks sharing the mm struct with the given victim. This commit splits oom_kill_process() with an intention to re-use the the second half: __oom_kill_process(). The cgroup-aware OOM killer will kill multiple tasks belonging to the victim cgroup. We don't need to print the debug information for the each task, as well as play with task selection (considering task's children), so we can't use the existing oom_kill_process(). Link: http://lkml.kernel.org/r/20171130152824.1591-2-guro@fb.com Link: http://lkml.kernel.org/r/20180802003201.817-3-guro@fb.com Signed-off-by: Roman Gushchin <guro@fb.com> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Acked-by: David Rientjes <rientjes@google.com> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: David Rientjes <rientjes@google.com> Cc: Tejun Heo <tj@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-08-22 12:53:50 +08:00
mm, oom: introduce memory.oom.group For some workloads an intervention from the OOM killer can be painful. Killing a random task can bring the workload into an inconsistent state. Historically, there are two common solutions for this problem: 1) enabling panic_on_oom, 2) using a userspace daemon to monitor OOMs and kill all outstanding processes. Both approaches have their downsides: rebooting on each OOM is an obvious waste of capacity, and handling all in userspace is tricky and requires a userspace agent, which will monitor all cgroups for OOMs. In most cases an in-kernel after-OOM cleaning-up mechanism can eliminate the necessity of enabling panic_on_oom. Also, it can simplify the cgroup management for userspace applications. This commit introduces a new knob for cgroup v2 memory controller: memory.oom.group. The knob determines whether the cgroup should be treated as an indivisible workload by the OOM killer. If set, all tasks belonging to the cgroup or to its descendants (if the memory cgroup is not a leaf cgroup) are killed together or not at all. To determine which cgroup has to be killed, we do traverse the cgroup hierarchy from the victim task's cgroup up to the OOMing cgroup (or root) and looking for the highest-level cgroup with memory.oom.group set. Tasks with the OOM protection (oom_score_adj set to -1000) are treated as an exception and are never killed. This patch doesn't change the OOM victim selection algorithm. Link: http://lkml.kernel.org/r/20180802003201.817-4-guro@fb.com Signed-off-by: Roman Gushchin <guro@fb.com> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: David Rientjes <rientjes@google.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Tejun Heo <tj@kernel.org> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-08-22 12:53:54 +08:00
/*
* Do we need to kill the entire memory cgroup?
* Or even one of the ancestor memory cgroups?
* Check this out before killing the victim task.
*/
oom_group = mem_cgroup_get_oom_group(victim, oc->memcg);
mm, oom: remove 'prefer children over parent' heuristic Since the start of the git history of Linux, the kernel after selecting the worst process to be oom-killed, prefer to kill its child (if the child does not share mm with the parent). Later it was changed to prefer to kill a child who is worst. If the parent is still the worst then the parent will be killed. This heuristic assumes that the children did less work than their parent and by killing one of them, the work lost will be less. However this is very workload dependent. If there is a workload which can benefit from this heuristic, can use oom_score_adj to prefer children to be killed before the parent. The select_bad_process() has already selected the worst process in the system/memcg. There is no need to recheck the badness of its children and hoping to find a worse candidate. That's a lot of unneeded racy work. Also the heuristic is dangerous because it make fork bomb like workloads to recover much later because we constantly pick and kill processes which are not memory hogs. So, let's remove this whole heuristic. [akpm@linux-foundation.org: coding-style fixes] Link: http://lkml.kernel.org/r/20190121215850.221745-2-shakeelb@google.com Signed-off-by: Shakeel Butt <shakeelb@google.com> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: Roman Gushchin <guro@fb.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: David Rientjes <rientjes@google.com> Cc: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-03-06 07:46:12 +08:00
__oom_kill_process(victim, message);
mm, oom: introduce memory.oom.group For some workloads an intervention from the OOM killer can be painful. Killing a random task can bring the workload into an inconsistent state. Historically, there are two common solutions for this problem: 1) enabling panic_on_oom, 2) using a userspace daemon to monitor OOMs and kill all outstanding processes. Both approaches have their downsides: rebooting on each OOM is an obvious waste of capacity, and handling all in userspace is tricky and requires a userspace agent, which will monitor all cgroups for OOMs. In most cases an in-kernel after-OOM cleaning-up mechanism can eliminate the necessity of enabling panic_on_oom. Also, it can simplify the cgroup management for userspace applications. This commit introduces a new knob for cgroup v2 memory controller: memory.oom.group. The knob determines whether the cgroup should be treated as an indivisible workload by the OOM killer. If set, all tasks belonging to the cgroup or to its descendants (if the memory cgroup is not a leaf cgroup) are killed together or not at all. To determine which cgroup has to be killed, we do traverse the cgroup hierarchy from the victim task's cgroup up to the OOMing cgroup (or root) and looking for the highest-level cgroup with memory.oom.group set. Tasks with the OOM protection (oom_score_adj set to -1000) are treated as an exception and are never killed. This patch doesn't change the OOM victim selection algorithm. Link: http://lkml.kernel.org/r/20180802003201.817-4-guro@fb.com Signed-off-by: Roman Gushchin <guro@fb.com> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: David Rientjes <rientjes@google.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Tejun Heo <tj@kernel.org> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-08-22 12:53:54 +08:00
/*
* If necessary, kill all tasks in the selected memory cgroup.
*/
if (oom_group) {
mem_cgroup_print_oom_group(oom_group);
mm, oom: remove 'prefer children over parent' heuristic Since the start of the git history of Linux, the kernel after selecting the worst process to be oom-killed, prefer to kill its child (if the child does not share mm with the parent). Later it was changed to prefer to kill a child who is worst. If the parent is still the worst then the parent will be killed. This heuristic assumes that the children did less work than their parent and by killing one of them, the work lost will be less. However this is very workload dependent. If there is a workload which can benefit from this heuristic, can use oom_score_adj to prefer children to be killed before the parent. The select_bad_process() has already selected the worst process in the system/memcg. There is no need to recheck the badness of its children and hoping to find a worse candidate. That's a lot of unneeded racy work. Also the heuristic is dangerous because it make fork bomb like workloads to recover much later because we constantly pick and kill processes which are not memory hogs. So, let's remove this whole heuristic. [akpm@linux-foundation.org: coding-style fixes] Link: http://lkml.kernel.org/r/20190121215850.221745-2-shakeelb@google.com Signed-off-by: Shakeel Butt <shakeelb@google.com> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: Roman Gushchin <guro@fb.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: David Rientjes <rientjes@google.com> Cc: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-03-06 07:46:12 +08:00
mem_cgroup_scan_tasks(oom_group, oom_kill_memcg_member,
(void*)message);
mm, oom: introduce memory.oom.group For some workloads an intervention from the OOM killer can be painful. Killing a random task can bring the workload into an inconsistent state. Historically, there are two common solutions for this problem: 1) enabling panic_on_oom, 2) using a userspace daemon to monitor OOMs and kill all outstanding processes. Both approaches have their downsides: rebooting on each OOM is an obvious waste of capacity, and handling all in userspace is tricky and requires a userspace agent, which will monitor all cgroups for OOMs. In most cases an in-kernel after-OOM cleaning-up mechanism can eliminate the necessity of enabling panic_on_oom. Also, it can simplify the cgroup management for userspace applications. This commit introduces a new knob for cgroup v2 memory controller: memory.oom.group. The knob determines whether the cgroup should be treated as an indivisible workload by the OOM killer. If set, all tasks belonging to the cgroup or to its descendants (if the memory cgroup is not a leaf cgroup) are killed together or not at all. To determine which cgroup has to be killed, we do traverse the cgroup hierarchy from the victim task's cgroup up to the OOMing cgroup (or root) and looking for the highest-level cgroup with memory.oom.group set. Tasks with the OOM protection (oom_score_adj set to -1000) are treated as an exception and are never killed. This patch doesn't change the OOM victim selection algorithm. Link: http://lkml.kernel.org/r/20180802003201.817-4-guro@fb.com Signed-off-by: Roman Gushchin <guro@fb.com> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: David Rientjes <rientjes@google.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Tejun Heo <tj@kernel.org> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-08-22 12:53:54 +08:00
mem_cgroup_put(oom_group);
}
mm, oom: refactor oom_kill_process() Patch series "introduce memory.oom.group", v2. This is a tiny implementation of cgroup-aware OOM killer, which adds an ability to kill a cgroup as a single unit and so guarantee the integrity of the workload. Although it has only a limited functionality in comparison to what now resides in the mm tree (it doesn't change the victim task selection algorithm, doesn't look at memory stas on cgroup level, etc), it's also much simpler and more straightforward. So, hopefully, we can avoid having long debates here, as we had with the full implementation. As it doesn't prevent any futher development, and implements an useful and complete feature, it looks as a sane way forward. This patch (of 2): oom_kill_process() consists of two logical parts: the first one is responsible for considering task's children as a potential victim and printing the debug information. The second half is responsible for sending SIGKILL to all tasks sharing the mm struct with the given victim. This commit splits oom_kill_process() with an intention to re-use the the second half: __oom_kill_process(). The cgroup-aware OOM killer will kill multiple tasks belonging to the victim cgroup. We don't need to print the debug information for the each task, as well as play with task selection (considering task's children), so we can't use the existing oom_kill_process(). Link: http://lkml.kernel.org/r/20171130152824.1591-2-guro@fb.com Link: http://lkml.kernel.org/r/20180802003201.817-3-guro@fb.com Signed-off-by: Roman Gushchin <guro@fb.com> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Acked-by: David Rientjes <rientjes@google.com> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: David Rientjes <rientjes@google.com> Cc: Tejun Heo <tj@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-08-22 12:53:50 +08:00
}
/*
* Determines whether the kernel must panic because of the panic_on_oom sysctl.
*/
static void check_panic_on_oom(struct oom_control *oc)
{
if (likely(!sysctl_panic_on_oom))
return;
if (sysctl_panic_on_oom != 2) {
/*
* panic_on_oom == 1 only affects CONSTRAINT_NONE, the kernel
* does not panic for cpuset, mempolicy, or memcg allocation
* failures.
*/
if (oc->constraint != CONSTRAINT_NONE)
return;
}
/* Do not panic for oom kills triggered by sysrq */
if (is_sysrq_oom(oc))
return;
dump_header(oc, NULL);
panic("Out of memory: %s panic_on_oom is enabled\n",
sysctl_panic_on_oom == 2 ? "compulsory" : "system-wide");
}
static BLOCKING_NOTIFIER_HEAD(oom_notify_list);
int register_oom_notifier(struct notifier_block *nb)
{
return blocking_notifier_chain_register(&oom_notify_list, nb);
}
EXPORT_SYMBOL_GPL(register_oom_notifier);
int unregister_oom_notifier(struct notifier_block *nb)
{
return blocking_notifier_chain_unregister(&oom_notify_list, nb);
}
EXPORT_SYMBOL_GPL(unregister_oom_notifier);
/**
* out_of_memory - kill the "best" process when we run out of memory
* @oc: pointer to struct oom_control
*
* If we run out of memory, we have the choice between either
* killing a random task (bad), letting the system crash (worse)
* OR try to be smart about which process to kill. Note that we
* don't have to be perfect here, we just have to be good.
*/
bool out_of_memory(struct oom_control *oc)
{
unsigned long freed = 0;
if (oom_killer_disabled)
return false;
if (!is_memcg_oom(oc)) {
blocking_notifier_call_chain(&oom_notify_list, 0, &freed);
if (freed > 0)
/* Got some memory back in the last second. */
return true;
}
oom: give current access to memory reserves if it has been killed It's possible to livelock the page allocator if a thread has mm->mmap_sem and fails to make forward progress because the oom killer selects another thread sharing the same ->mm to kill that cannot exit until the semaphore is dropped. The oom killer will not kill multiple tasks at the same time; each oom killed task must exit before another task may be killed. Thus, if one thread is holding mm->mmap_sem and cannot allocate memory, all threads sharing the same ->mm are blocked from exiting as well. In the oom kill case, that means the thread holding mm->mmap_sem will never free additional memory since it cannot get access to memory reserves and the thread that depends on it with access to memory reserves cannot exit because it cannot acquire the semaphore. Thus, the page allocators livelocks. When the oom killer is called and current happens to have a pending SIGKILL, this patch automatically gives it access to memory reserves and returns. Upon returning to the page allocator, its allocation will hopefully succeed so it can quickly exit and free its memory. If not, the page allocator will fail the allocation if it is not __GFP_NOFAIL. Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Signed-off-by: David Rientjes <rientjes@google.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 08:18:48 +08:00
/*
mm, oom: allow exiting threads to have access to memory reserves Exiting threads, those with PF_EXITING set, can pagefault and require memory before they can make forward progress. This happens, for instance, when a process must fault task->robust_list, a userspace structure, before detaching its memory. These threads also aren't guaranteed to get access to memory reserves unless oom killed or killed from userspace. The oom killer won't grant memory reserves if other threads are also exiting other than current and stalling at the same point. This prevents needlessly killing processes when others are already exiting. Instead of special casing all the possible situations between PF_EXITING getting set and a thread detaching its mm where it may allocate memory, which probably wouldn't get updated when a change is made to the exit path, the solution is to give all exiting threads access to memory reserves if they call the oom killer. This allows them to quickly allocate, detach its mm, and free the memory it represents. Summary of Luigi's bug report: : He had an oom condition where threads were faulting on task->robust_list : and repeatedly called the oom killer but it would defer killing a thread : because it saw other PF_EXITING threads. This can happen anytime we need : to allocate memory after setting PF_EXITING and before detaching our mm; : if there are other threads in the same state then the oom killer won't do : anything unless one of them happens to be killed from userspace. : : So instead of only deferring for PF_EXITING and !task->robust_list, it's : better to just give them access to memory reserves to prevent a potential : livelock so that any other faults that may be introduced in the future in : the exit path don't cause the same problem (and hopefully we don't allow : too many of those!). Signed-off-by: David Rientjes <rientjes@google.com> Acked-by: Minchan Kim <minchan@kernel.org> Tested-by: Luigi Semenzato <semenzato@google.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-12-12 08:01:30 +08:00
* If current has a pending SIGKILL or is exiting, then automatically
* select it. The goal is to allow it to allocate so that it may
* quickly exit and free its memory.
oom: give current access to memory reserves if it has been killed It's possible to livelock the page allocator if a thread has mm->mmap_sem and fails to make forward progress because the oom killer selects another thread sharing the same ->mm to kill that cannot exit until the semaphore is dropped. The oom killer will not kill multiple tasks at the same time; each oom killed task must exit before another task may be killed. Thus, if one thread is holding mm->mmap_sem and cannot allocate memory, all threads sharing the same ->mm are blocked from exiting as well. In the oom kill case, that means the thread holding mm->mmap_sem will never free additional memory since it cannot get access to memory reserves and the thread that depends on it with access to memory reserves cannot exit because it cannot acquire the semaphore. Thus, the page allocators livelocks. When the oom killer is called and current happens to have a pending SIGKILL, this patch automatically gives it access to memory reserves and returns. Upon returning to the page allocator, its allocation will hopefully succeed so it can quickly exit and free its memory. If not, the page allocator will fail the allocation if it is not __GFP_NOFAIL. Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Signed-off-by: David Rientjes <rientjes@google.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 08:18:48 +08:00
*/
if (task_will_free_mem(current)) {
mark_oom_victim(current);
wake_oom_reaper(current);
return true;
oom: give current access to memory reserves if it has been killed It's possible to livelock the page allocator if a thread has mm->mmap_sem and fails to make forward progress because the oom killer selects another thread sharing the same ->mm to kill that cannot exit until the semaphore is dropped. The oom killer will not kill multiple tasks at the same time; each oom killed task must exit before another task may be killed. Thus, if one thread is holding mm->mmap_sem and cannot allocate memory, all threads sharing the same ->mm are blocked from exiting as well. In the oom kill case, that means the thread holding mm->mmap_sem will never free additional memory since it cannot get access to memory reserves and the thread that depends on it with access to memory reserves cannot exit because it cannot acquire the semaphore. Thus, the page allocators livelocks. When the oom killer is called and current happens to have a pending SIGKILL, this patch automatically gives it access to memory reserves and returns. Upon returning to the page allocator, its allocation will hopefully succeed so it can quickly exit and free its memory. If not, the page allocator will fail the allocation if it is not __GFP_NOFAIL. Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Signed-off-by: David Rientjes <rientjes@google.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 08:18:48 +08:00
}
mm, oom: move GFP_NOFS check to out_of_memory __alloc_pages_may_oom is the central place to decide when the out_of_memory should be invoked. This is a good approach for most checks there because they are page allocator specific and the allocation fails right after for all of them. The notable exception is GFP_NOFS context which is faking did_some_progress and keep the page allocator looping even though there couldn't have been any progress from the OOM killer. This patch doesn't change this behavior because we are not ready to allow those allocation requests to fail yet (and maybe we will face the reality that we will never manage to safely fail these request). Instead __GFP_FS check is moved down to out_of_memory and prevent from OOM victim selection there. There are two reasons for that - OOM notifiers might release some memory even from this context as none of the registered notifier seems to be FS related - this might help a dying thread to get an access to memory reserves and move on which will make the behavior more consistent with the case when the task gets killed from a different context. Keep a comment in __alloc_pages_may_oom to make sure we do not forget how GFP_NOFS is special and that we really want to do something about it. Note to the current oom_notifier users: The observable difference for you is that oom notifiers cannot depend on any fs locks because we could deadlock. Not that this would be allowed today because that would just lockup machine in most of the cases and ruling out the OOM killer along the way. Another difference is that callbacks might be invoked sooner now because GFP_NOFS is a weaker reclaim context and so there could be reclaimable memory which is just not reachable now. That would require GFP_NOFS only loads which are really rare and more importantly the observable result would be dropping of reconstructible object and potential performance drop which is not such a big deal when we are struggling to fulfill other important allocation requests. Signed-off-by: Michal Hocko <mhocko@suse.com> Cc: Raushaniya Maksudova <rmaksudova@parallels.com> Cc: Michael S. Tsirkin <mst@redhat.com> Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: David Rientjes <rientjes@google.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Daniel Vetter <daniel.vetter@intel.com> Cc: Oleg Nesterov <oleg@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-20 08:13:09 +08:00
/*
* The OOM killer does not compensate for IO-less reclaim.
* pagefault_out_of_memory lost its gfp context so we have to
* make sure exclude 0 mask - all other users should have at least
memcg, oom: don't require __GFP_FS when invoking memcg OOM killer Masoud Sharbiani noticed that commit 29ef680ae7c21110 ("memcg, oom: move out_of_memory back to the charge path") broke memcg OOM called from __xfs_filemap_fault() path. It turned out that try_charge() is retrying forever without making forward progress because mem_cgroup_oom(GFP_NOFS) cannot invoke the OOM killer due to commit 3da88fb3bacfaa33 ("mm, oom: move GFP_NOFS check to out_of_memory"). Allowing forced charge due to being unable to invoke memcg OOM killer will lead to global OOM situation. Also, just returning -ENOMEM will be risky because OOM path is lost and some paths (e.g. get_user_pages()) will leak -ENOMEM. Therefore, invoking memcg OOM killer (despite GFP_NOFS) will be the only choice we can choose for now. Until 29ef680ae7c21110, we were able to invoke memcg OOM killer when GFP_KERNEL reclaim failed [1]. But since 29ef680ae7c21110, we need to invoke memcg OOM killer when GFP_NOFS reclaim failed [2]. Although in the past we did invoke memcg OOM killer for GFP_NOFS [3], we might get pre-mature memcg OOM reports due to this patch. [1] leaker invoked oom-killer: gfp_mask=0x6200ca(GFP_HIGHUSER_MOVABLE), nodemask=(null), order=0, oom_score_adj=0 CPU: 0 PID: 2746 Comm: leaker Not tainted 4.18.0+ #19 Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS 6.00 04/13/2018 Call Trace: dump_stack+0x63/0x88 dump_header+0x67/0x27a ? mem_cgroup_scan_tasks+0x91/0xf0 oom_kill_process+0x210/0x410 out_of_memory+0x10a/0x2c0 mem_cgroup_out_of_memory+0x46/0x80 mem_cgroup_oom_synchronize+0x2e4/0x310 ? high_work_func+0x20/0x20 pagefault_out_of_memory+0x31/0x76 mm_fault_error+0x55/0x115 ? handle_mm_fault+0xfd/0x220 __do_page_fault+0x433/0x4e0 do_page_fault+0x22/0x30 ? page_fault+0x8/0x30 page_fault+0x1e/0x30 RIP: 0033:0x4009f0 Code: 03 00 00 00 e8 71 fd ff ff 48 83 f8 ff 49 89 c6 74 74 48 89 c6 bf c0 0c 40 00 31 c0 e8 69 fd ff ff 45 85 ff 7e 21 31 c9 66 90 <41> 0f be 14 0e 01 d3 f7 c1 ff 0f 00 00 75 05 41 c6 04 0e 2a 48 83 RSP: 002b:00007ffe29ae96f0 EFLAGS: 00010206 RAX: 000000000000001b RBX: 0000000000000000 RCX: 0000000001ce1000 RDX: 0000000000000000 RSI: 000000007fffffe5 RDI: 0000000000000000 RBP: 000000000000000c R08: 0000000000000000 R09: 00007f94be09220d R10: 0000000000000002 R11: 0000000000000246 R12: 00000000000186a0 R13: 0000000000000003 R14: 00007f949d845000 R15: 0000000002800000 Task in /leaker killed as a result of limit of /leaker memory: usage 524288kB, limit 524288kB, failcnt 158965 memory+swap: usage 0kB, limit 9007199254740988kB, failcnt 0 kmem: usage 2016kB, limit 9007199254740988kB, failcnt 0 Memory cgroup stats for /leaker: cache:844KB rss:521136KB rss_huge:0KB shmem:0KB mapped_file:0KB dirty:132KB writeback:0KB inactive_anon:0KB active_anon:521224KB inactive_file:1012KB active_file:8KB unevictable:0KB Memory cgroup out of memory: Kill process 2746 (leaker) score 998 or sacrifice child Killed process 2746 (leaker) total-vm:536704kB, anon-rss:521176kB, file-rss:1208kB, shmem-rss:0kB oom_reaper: reaped process 2746 (leaker), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kB [2] leaker invoked oom-killer: gfp_mask=0x600040(GFP_NOFS), nodemask=(null), order=0, oom_score_adj=0 CPU: 1 PID: 2746 Comm: leaker Not tainted 4.18.0+ #20 Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS 6.00 04/13/2018 Call Trace: dump_stack+0x63/0x88 dump_header+0x67/0x27a ? mem_cgroup_scan_tasks+0x91/0xf0 oom_kill_process+0x210/0x410 out_of_memory+0x109/0x2d0 mem_cgroup_out_of_memory+0x46/0x80 try_charge+0x58d/0x650 ? __radix_tree_replace+0x81/0x100 mem_cgroup_try_charge+0x7a/0x100 __add_to_page_cache_locked+0x92/0x180 add_to_page_cache_lru+0x4d/0xf0 iomap_readpages_actor+0xde/0x1b0 ? iomap_zero_range_actor+0x1d0/0x1d0 iomap_apply+0xaf/0x130 iomap_readpages+0x9f/0x150 ? iomap_zero_range_actor+0x1d0/0x1d0 xfs_vm_readpages+0x18/0x20 [xfs] read_pages+0x60/0x140 __do_page_cache_readahead+0x193/0x1b0 ondemand_readahead+0x16d/0x2c0 page_cache_async_readahead+0x9a/0xd0 filemap_fault+0x403/0x620 ? alloc_set_pte+0x12c/0x540 ? _cond_resched+0x14/0x30 __xfs_filemap_fault+0x66/0x180 [xfs] xfs_filemap_fault+0x27/0x30 [xfs] __do_fault+0x19/0x40 __handle_mm_fault+0x8e8/0xb60 handle_mm_fault+0xfd/0x220 __do_page_fault+0x238/0x4e0 do_page_fault+0x22/0x30 ? page_fault+0x8/0x30 page_fault+0x1e/0x30 RIP: 0033:0x4009f0 Code: 03 00 00 00 e8 71 fd ff ff 48 83 f8 ff 49 89 c6 74 74 48 89 c6 bf c0 0c 40 00 31 c0 e8 69 fd ff ff 45 85 ff 7e 21 31 c9 66 90 <41> 0f be 14 0e 01 d3 f7 c1 ff 0f 00 00 75 05 41 c6 04 0e 2a 48 83 RSP: 002b:00007ffda45c9290 EFLAGS: 00010206 RAX: 000000000000001b RBX: 0000000000000000 RCX: 0000000001a1e000 RDX: 0000000000000000 RSI: 000000007fffffe5 RDI: 0000000000000000 RBP: 000000000000000c R08: 0000000000000000 R09: 00007f6d061ff20d R10: 0000000000000002 R11: 0000000000000246 R12: 00000000000186a0 R13: 0000000000000003 R14: 00007f6ce59b2000 R15: 0000000002800000 Task in /leaker killed as a result of limit of /leaker memory: usage 524288kB, limit 524288kB, failcnt 7221 memory+swap: usage 0kB, limit 9007199254740988kB, failcnt 0 kmem: usage 1944kB, limit 9007199254740988kB, failcnt 0 Memory cgroup stats for /leaker: cache:3632KB rss:518232KB rss_huge:0KB shmem:0KB mapped_file:0KB dirty:0KB writeback:0KB inactive_anon:0KB active_anon:518408KB inactive_file:3908KB active_file:12KB unevictable:0KB Memory cgroup out of memory: Kill process 2746 (leaker) score 992 or sacrifice child Killed process 2746 (leaker) total-vm:536704kB, anon-rss:518264kB, file-rss:1188kB, shmem-rss:0kB oom_reaper: reaped process 2746 (leaker), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kB [3] leaker invoked oom-killer: gfp_mask=0x50, order=0, oom_score_adj=0 leaker cpuset=/ mems_allowed=0 CPU: 1 PID: 3206 Comm: leaker Not tainted 3.10.0-957.27.2.el7.x86_64 #1 Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS 6.00 04/13/2018 Call Trace: [<ffffffffaf364147>] dump_stack+0x19/0x1b [<ffffffffaf35eb6a>] dump_header+0x90/0x229 [<ffffffffaedbb456>] ? find_lock_task_mm+0x56/0xc0 [<ffffffffaee32a38>] ? try_get_mem_cgroup_from_mm+0x28/0x60 [<ffffffffaedbb904>] oom_kill_process+0x254/0x3d0 [<ffffffffaee36c36>] mem_cgroup_oom_synchronize+0x546/0x570 [<ffffffffaee360b0>] ? mem_cgroup_charge_common+0xc0/0xc0 [<ffffffffaedbc194>] pagefault_out_of_memory+0x14/0x90 [<ffffffffaf35d072>] mm_fault_error+0x6a/0x157 [<ffffffffaf3717c8>] __do_page_fault+0x3c8/0x4f0 [<ffffffffaf371925>] do_page_fault+0x35/0x90 [<ffffffffaf36d768>] page_fault+0x28/0x30 Task in /leaker killed as a result of limit of /leaker memory: usage 524288kB, limit 524288kB, failcnt 20628 memory+swap: usage 524288kB, limit 9007199254740988kB, failcnt 0 kmem: usage 0kB, limit 9007199254740988kB, failcnt 0 Memory cgroup stats for /leaker: cache:840KB rss:523448KB rss_huge:0KB mapped_file:0KB swap:0KB inactive_anon:0KB active_anon:523448KB inactive_file:464KB active_file:376KB unevictable:0KB Memory cgroup out of memory: Kill process 3206 (leaker) score 970 or sacrifice child Killed process 3206 (leaker) total-vm:536692kB, anon-rss:523304kB, file-rss:412kB, shmem-rss:0kB Bisected by Masoud Sharbiani. Link: http://lkml.kernel.org/r/cbe54ed1-b6ba-a056-8899-2dc42526371d@i-love.sakura.ne.jp Fixes: 3da88fb3bacfaa33 ("mm, oom: move GFP_NOFS check to out_of_memory") [necessary after 29ef680ae7c21110] Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Reported-by: Masoud Sharbiani <msharbiani@apple.com> Tested-by: Masoud Sharbiani <msharbiani@apple.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: David Rientjes <rientjes@google.com> Cc: <stable@vger.kernel.org> [4.19+] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-09-24 06:37:08 +08:00
* ___GFP_DIRECT_RECLAIM to get here. But mem_cgroup_oom() has to
* invoke the OOM killer even if it is a GFP_NOFS allocation.
mm, oom: move GFP_NOFS check to out_of_memory __alloc_pages_may_oom is the central place to decide when the out_of_memory should be invoked. This is a good approach for most checks there because they are page allocator specific and the allocation fails right after for all of them. The notable exception is GFP_NOFS context which is faking did_some_progress and keep the page allocator looping even though there couldn't have been any progress from the OOM killer. This patch doesn't change this behavior because we are not ready to allow those allocation requests to fail yet (and maybe we will face the reality that we will never manage to safely fail these request). Instead __GFP_FS check is moved down to out_of_memory and prevent from OOM victim selection there. There are two reasons for that - OOM notifiers might release some memory even from this context as none of the registered notifier seems to be FS related - this might help a dying thread to get an access to memory reserves and move on which will make the behavior more consistent with the case when the task gets killed from a different context. Keep a comment in __alloc_pages_may_oom to make sure we do not forget how GFP_NOFS is special and that we really want to do something about it. Note to the current oom_notifier users: The observable difference for you is that oom notifiers cannot depend on any fs locks because we could deadlock. Not that this would be allowed today because that would just lockup machine in most of the cases and ruling out the OOM killer along the way. Another difference is that callbacks might be invoked sooner now because GFP_NOFS is a weaker reclaim context and so there could be reclaimable memory which is just not reachable now. That would require GFP_NOFS only loads which are really rare and more importantly the observable result would be dropping of reconstructible object and potential performance drop which is not such a big deal when we are struggling to fulfill other important allocation requests. Signed-off-by: Michal Hocko <mhocko@suse.com> Cc: Raushaniya Maksudova <rmaksudova@parallels.com> Cc: Michael S. Tsirkin <mst@redhat.com> Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: David Rientjes <rientjes@google.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Daniel Vetter <daniel.vetter@intel.com> Cc: Oleg Nesterov <oleg@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-20 08:13:09 +08:00
*/
memcg, oom: don't require __GFP_FS when invoking memcg OOM killer Masoud Sharbiani noticed that commit 29ef680ae7c21110 ("memcg, oom: move out_of_memory back to the charge path") broke memcg OOM called from __xfs_filemap_fault() path. It turned out that try_charge() is retrying forever without making forward progress because mem_cgroup_oom(GFP_NOFS) cannot invoke the OOM killer due to commit 3da88fb3bacfaa33 ("mm, oom: move GFP_NOFS check to out_of_memory"). Allowing forced charge due to being unable to invoke memcg OOM killer will lead to global OOM situation. Also, just returning -ENOMEM will be risky because OOM path is lost and some paths (e.g. get_user_pages()) will leak -ENOMEM. Therefore, invoking memcg OOM killer (despite GFP_NOFS) will be the only choice we can choose for now. Until 29ef680ae7c21110, we were able to invoke memcg OOM killer when GFP_KERNEL reclaim failed [1]. But since 29ef680ae7c21110, we need to invoke memcg OOM killer when GFP_NOFS reclaim failed [2]. Although in the past we did invoke memcg OOM killer for GFP_NOFS [3], we might get pre-mature memcg OOM reports due to this patch. [1] leaker invoked oom-killer: gfp_mask=0x6200ca(GFP_HIGHUSER_MOVABLE), nodemask=(null), order=0, oom_score_adj=0 CPU: 0 PID: 2746 Comm: leaker Not tainted 4.18.0+ #19 Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS 6.00 04/13/2018 Call Trace: dump_stack+0x63/0x88 dump_header+0x67/0x27a ? mem_cgroup_scan_tasks+0x91/0xf0 oom_kill_process+0x210/0x410 out_of_memory+0x10a/0x2c0 mem_cgroup_out_of_memory+0x46/0x80 mem_cgroup_oom_synchronize+0x2e4/0x310 ? high_work_func+0x20/0x20 pagefault_out_of_memory+0x31/0x76 mm_fault_error+0x55/0x115 ? handle_mm_fault+0xfd/0x220 __do_page_fault+0x433/0x4e0 do_page_fault+0x22/0x30 ? page_fault+0x8/0x30 page_fault+0x1e/0x30 RIP: 0033:0x4009f0 Code: 03 00 00 00 e8 71 fd ff ff 48 83 f8 ff 49 89 c6 74 74 48 89 c6 bf c0 0c 40 00 31 c0 e8 69 fd ff ff 45 85 ff 7e 21 31 c9 66 90 <41> 0f be 14 0e 01 d3 f7 c1 ff 0f 00 00 75 05 41 c6 04 0e 2a 48 83 RSP: 002b:00007ffe29ae96f0 EFLAGS: 00010206 RAX: 000000000000001b RBX: 0000000000000000 RCX: 0000000001ce1000 RDX: 0000000000000000 RSI: 000000007fffffe5 RDI: 0000000000000000 RBP: 000000000000000c R08: 0000000000000000 R09: 00007f94be09220d R10: 0000000000000002 R11: 0000000000000246 R12: 00000000000186a0 R13: 0000000000000003 R14: 00007f949d845000 R15: 0000000002800000 Task in /leaker killed as a result of limit of /leaker memory: usage 524288kB, limit 524288kB, failcnt 158965 memory+swap: usage 0kB, limit 9007199254740988kB, failcnt 0 kmem: usage 2016kB, limit 9007199254740988kB, failcnt 0 Memory cgroup stats for /leaker: cache:844KB rss:521136KB rss_huge:0KB shmem:0KB mapped_file:0KB dirty:132KB writeback:0KB inactive_anon:0KB active_anon:521224KB inactive_file:1012KB active_file:8KB unevictable:0KB Memory cgroup out of memory: Kill process 2746 (leaker) score 998 or sacrifice child Killed process 2746 (leaker) total-vm:536704kB, anon-rss:521176kB, file-rss:1208kB, shmem-rss:0kB oom_reaper: reaped process 2746 (leaker), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kB [2] leaker invoked oom-killer: gfp_mask=0x600040(GFP_NOFS), nodemask=(null), order=0, oom_score_adj=0 CPU: 1 PID: 2746 Comm: leaker Not tainted 4.18.0+ #20 Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS 6.00 04/13/2018 Call Trace: dump_stack+0x63/0x88 dump_header+0x67/0x27a ? mem_cgroup_scan_tasks+0x91/0xf0 oom_kill_process+0x210/0x410 out_of_memory+0x109/0x2d0 mem_cgroup_out_of_memory+0x46/0x80 try_charge+0x58d/0x650 ? __radix_tree_replace+0x81/0x100 mem_cgroup_try_charge+0x7a/0x100 __add_to_page_cache_locked+0x92/0x180 add_to_page_cache_lru+0x4d/0xf0 iomap_readpages_actor+0xde/0x1b0 ? iomap_zero_range_actor+0x1d0/0x1d0 iomap_apply+0xaf/0x130 iomap_readpages+0x9f/0x150 ? iomap_zero_range_actor+0x1d0/0x1d0 xfs_vm_readpages+0x18/0x20 [xfs] read_pages+0x60/0x140 __do_page_cache_readahead+0x193/0x1b0 ondemand_readahead+0x16d/0x2c0 page_cache_async_readahead+0x9a/0xd0 filemap_fault+0x403/0x620 ? alloc_set_pte+0x12c/0x540 ? _cond_resched+0x14/0x30 __xfs_filemap_fault+0x66/0x180 [xfs] xfs_filemap_fault+0x27/0x30 [xfs] __do_fault+0x19/0x40 __handle_mm_fault+0x8e8/0xb60 handle_mm_fault+0xfd/0x220 __do_page_fault+0x238/0x4e0 do_page_fault+0x22/0x30 ? page_fault+0x8/0x30 page_fault+0x1e/0x30 RIP: 0033:0x4009f0 Code: 03 00 00 00 e8 71 fd ff ff 48 83 f8 ff 49 89 c6 74 74 48 89 c6 bf c0 0c 40 00 31 c0 e8 69 fd ff ff 45 85 ff 7e 21 31 c9 66 90 <41> 0f be 14 0e 01 d3 f7 c1 ff 0f 00 00 75 05 41 c6 04 0e 2a 48 83 RSP: 002b:00007ffda45c9290 EFLAGS: 00010206 RAX: 000000000000001b RBX: 0000000000000000 RCX: 0000000001a1e000 RDX: 0000000000000000 RSI: 000000007fffffe5 RDI: 0000000000000000 RBP: 000000000000000c R08: 0000000000000000 R09: 00007f6d061ff20d R10: 0000000000000002 R11: 0000000000000246 R12: 00000000000186a0 R13: 0000000000000003 R14: 00007f6ce59b2000 R15: 0000000002800000 Task in /leaker killed as a result of limit of /leaker memory: usage 524288kB, limit 524288kB, failcnt 7221 memory+swap: usage 0kB, limit 9007199254740988kB, failcnt 0 kmem: usage 1944kB, limit 9007199254740988kB, failcnt 0 Memory cgroup stats for /leaker: cache:3632KB rss:518232KB rss_huge:0KB shmem:0KB mapped_file:0KB dirty:0KB writeback:0KB inactive_anon:0KB active_anon:518408KB inactive_file:3908KB active_file:12KB unevictable:0KB Memory cgroup out of memory: Kill process 2746 (leaker) score 992 or sacrifice child Killed process 2746 (leaker) total-vm:536704kB, anon-rss:518264kB, file-rss:1188kB, shmem-rss:0kB oom_reaper: reaped process 2746 (leaker), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kB [3] leaker invoked oom-killer: gfp_mask=0x50, order=0, oom_score_adj=0 leaker cpuset=/ mems_allowed=0 CPU: 1 PID: 3206 Comm: leaker Not tainted 3.10.0-957.27.2.el7.x86_64 #1 Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS 6.00 04/13/2018 Call Trace: [<ffffffffaf364147>] dump_stack+0x19/0x1b [<ffffffffaf35eb6a>] dump_header+0x90/0x229 [<ffffffffaedbb456>] ? find_lock_task_mm+0x56/0xc0 [<ffffffffaee32a38>] ? try_get_mem_cgroup_from_mm+0x28/0x60 [<ffffffffaedbb904>] oom_kill_process+0x254/0x3d0 [<ffffffffaee36c36>] mem_cgroup_oom_synchronize+0x546/0x570 [<ffffffffaee360b0>] ? mem_cgroup_charge_common+0xc0/0xc0 [<ffffffffaedbc194>] pagefault_out_of_memory+0x14/0x90 [<ffffffffaf35d072>] mm_fault_error+0x6a/0x157 [<ffffffffaf3717c8>] __do_page_fault+0x3c8/0x4f0 [<ffffffffaf371925>] do_page_fault+0x35/0x90 [<ffffffffaf36d768>] page_fault+0x28/0x30 Task in /leaker killed as a result of limit of /leaker memory: usage 524288kB, limit 524288kB, failcnt 20628 memory+swap: usage 524288kB, limit 9007199254740988kB, failcnt 0 kmem: usage 0kB, limit 9007199254740988kB, failcnt 0 Memory cgroup stats for /leaker: cache:840KB rss:523448KB rss_huge:0KB mapped_file:0KB swap:0KB inactive_anon:0KB active_anon:523448KB inactive_file:464KB active_file:376KB unevictable:0KB Memory cgroup out of memory: Kill process 3206 (leaker) score 970 or sacrifice child Killed process 3206 (leaker) total-vm:536692kB, anon-rss:523304kB, file-rss:412kB, shmem-rss:0kB Bisected by Masoud Sharbiani. Link: http://lkml.kernel.org/r/cbe54ed1-b6ba-a056-8899-2dc42526371d@i-love.sakura.ne.jp Fixes: 3da88fb3bacfaa33 ("mm, oom: move GFP_NOFS check to out_of_memory") [necessary after 29ef680ae7c21110] Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Reported-by: Masoud Sharbiani <msharbiani@apple.com> Tested-by: Masoud Sharbiani <msharbiani@apple.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: David Rientjes <rientjes@google.com> Cc: <stable@vger.kernel.org> [4.19+] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-09-24 06:37:08 +08:00
if (oc->gfp_mask && !(oc->gfp_mask & __GFP_FS) && !is_memcg_oom(oc))
mm, oom: move GFP_NOFS check to out_of_memory __alloc_pages_may_oom is the central place to decide when the out_of_memory should be invoked. This is a good approach for most checks there because they are page allocator specific and the allocation fails right after for all of them. The notable exception is GFP_NOFS context which is faking did_some_progress and keep the page allocator looping even though there couldn't have been any progress from the OOM killer. This patch doesn't change this behavior because we are not ready to allow those allocation requests to fail yet (and maybe we will face the reality that we will never manage to safely fail these request). Instead __GFP_FS check is moved down to out_of_memory and prevent from OOM victim selection there. There are two reasons for that - OOM notifiers might release some memory even from this context as none of the registered notifier seems to be FS related - this might help a dying thread to get an access to memory reserves and move on which will make the behavior more consistent with the case when the task gets killed from a different context. Keep a comment in __alloc_pages_may_oom to make sure we do not forget how GFP_NOFS is special and that we really want to do something about it. Note to the current oom_notifier users: The observable difference for you is that oom notifiers cannot depend on any fs locks because we could deadlock. Not that this would be allowed today because that would just lockup machine in most of the cases and ruling out the OOM killer along the way. Another difference is that callbacks might be invoked sooner now because GFP_NOFS is a weaker reclaim context and so there could be reclaimable memory which is just not reachable now. That would require GFP_NOFS only loads which are really rare and more importantly the observable result would be dropping of reconstructible object and potential performance drop which is not such a big deal when we are struggling to fulfill other important allocation requests. Signed-off-by: Michal Hocko <mhocko@suse.com> Cc: Raushaniya Maksudova <rmaksudova@parallels.com> Cc: Michael S. Tsirkin <mst@redhat.com> Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: David Rientjes <rientjes@google.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Daniel Vetter <daniel.vetter@intel.com> Cc: Oleg Nesterov <oleg@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-20 08:13:09 +08:00
return true;
/*
* Check if there were limitations on the allocation (only relevant for
* NUMA and memcg) that may require different handling.
*/
oc->constraint = constrained_alloc(oc);
if (oc->constraint != CONSTRAINT_MEMORY_POLICY)
oc->nodemask = NULL;
check_panic_on_oom(oc);
if (!is_memcg_oom(oc) && sysctl_oom_kill_allocating_task &&
oom: decouple mems_allowed from oom_unkillable_task Commit ef08e3b4981a ("[PATCH] cpusets: confine oom_killer to mem_exclusive cpuset") introduces a heuristic where a potential oom-killer victim is skipped if the intersection of the potential victim and the current (the process triggered the oom) is empty based on the reason that killing such victim most probably will not help the current allocating process. However the commit 7887a3da753e ("[PATCH] oom: cpuset hint") changed the heuristic to just decrease the oom_badness scores of such potential victim based on the reason that the cpuset of such processes might have changed and previously they may have allocated memory on mems where the current allocating process can allocate from. Unintentionally 7887a3da753e ("[PATCH] oom: cpuset hint") introduced a side effect as the oom_badness is also exposed to the user space through /proc/[pid]/oom_score, so, readers with different cpusets can read different oom_score of the same process. Later, commit 6cf86ac6f36b ("oom: filter tasks not sharing the same cpuset") fixed the side effect introduced by 7887a3da753e by moving the cpuset intersection back to only oom-killer context and out of oom_badness. However the combination of ab290adbaf8f ("oom: make oom_unkillable_task() helper function") and 26ebc984913b ("oom: /proc/<pid>/oom_score treat kernel thread honestly") unintentionally brought back the cpuset intersection check into the oom_badness calculation function. Other than doing cpuset/mempolicy intersection from oom_badness, the memcg oom context is also doing cpuset/mempolicy intersection which is quite wrong and is caught by syzcaller with the following report: kasan: CONFIG_KASAN_INLINE enabled kasan: GPF could be caused by NULL-ptr deref or user memory access general protection fault: 0000 [#1] PREEMPT SMP KASAN CPU: 0 PID: 28426 Comm: syz-executor.5 Not tainted 5.2.0-rc3-next-20190607 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:__read_once_size include/linux/compiler.h:194 [inline] RIP: 0010:has_intersects_mems_allowed mm/oom_kill.c:84 [inline] RIP: 0010:oom_unkillable_task mm/oom_kill.c:168 [inline] RIP: 0010:oom_unkillable_task+0x180/0x400 mm/oom_kill.c:155 Code: c1 ea 03 80 3c 02 00 0f 85 80 02 00 00 4c 8b a3 10 07 00 00 48 b8 00 00 00 00 00 fc ff df 4d 8d 74 24 10 4c 89 f2 48 c1 ea 03 <80> 3c 02 00 0f 85 67 02 00 00 49 8b 44 24 10 4c 8d a0 68 fa ff ff RSP: 0018:ffff888000127490 EFLAGS: 00010a03 RAX: dffffc0000000000 RBX: ffff8880a4cd5438 RCX: ffffffff818dae9c RDX: 100000000c3cc602 RSI: ffffffff818dac8d RDI: 0000000000000001 RBP: ffff8880001274d0 R08: ffff888000086180 R09: ffffed1015d26be0 R10: ffffed1015d26bdf R11: ffff8880ae935efb R12: 8000000061e63007 R13: 0000000000000000 R14: 8000000061e63017 R15: 1ffff11000024ea6 FS: 00005555561f5940(0000) GS:ffff8880ae800000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000607304 CR3: 000000009237e000 CR4: 00000000001426f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000600 Call Trace: oom_evaluate_task+0x49/0x520 mm/oom_kill.c:321 mem_cgroup_scan_tasks+0xcc/0x180 mm/memcontrol.c:1169 select_bad_process mm/oom_kill.c:374 [inline] out_of_memory mm/oom_kill.c:1088 [inline] out_of_memory+0x6b2/0x1280 mm/oom_kill.c:1035 mem_cgroup_out_of_memory+0x1ca/0x230 mm/memcontrol.c:1573 mem_cgroup_oom mm/memcontrol.c:1905 [inline] try_charge+0xfbe/0x1480 mm/memcontrol.c:2468 mem_cgroup_try_charge+0x24d/0x5e0 mm/memcontrol.c:6073 mem_cgroup_try_charge_delay+0x1f/0xa0 mm/memcontrol.c:6088 do_huge_pmd_wp_page_fallback+0x24f/0x1680 mm/huge_memory.c:1201 do_huge_pmd_wp_page+0x7fc/0x2160 mm/huge_memory.c:1359 wp_huge_pmd mm/memory.c:3793 [inline] __handle_mm_fault+0x164c/0x3eb0 mm/memory.c:4006 handle_mm_fault+0x3b7/0xa90 mm/memory.c:4053 do_user_addr_fault arch/x86/mm/fault.c:1455 [inline] __do_page_fault+0x5ef/0xda0 arch/x86/mm/fault.c:1521 do_page_fault+0x71/0x57d arch/x86/mm/fault.c:1552 page_fault+0x1e/0x30 arch/x86/entry/entry_64.S:1156 RIP: 0033:0x400590 Code: 06 e9 49 01 00 00 48 8b 44 24 10 48 0b 44 24 28 75 1f 48 8b 14 24 48 8b 7c 24 20 be 04 00 00 00 e8 f5 56 00 00 48 8b 74 24 08 <89> 06 e9 1e 01 00 00 48 8b 44 24 08 48 8b 14 24 be 04 00 00 00 8b RSP: 002b:00007fff7bc49780 EFLAGS: 00010206 RAX: 0000000000000001 RBX: 0000000000760000 RCX: 0000000000000000 RDX: 0000000000000000 RSI: 000000002000cffc RDI: 0000000000000001 RBP: fffffffffffffffe R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000075 R11: 0000000000000246 R12: 0000000000760008 R13: 00000000004c55f2 R14: 0000000000000000 R15: 00007fff7bc499b0 Modules linked in: ---[ end trace a65689219582ffff ]--- RIP: 0010:__read_once_size include/linux/compiler.h:194 [inline] RIP: 0010:has_intersects_mems_allowed mm/oom_kill.c:84 [inline] RIP: 0010:oom_unkillable_task mm/oom_kill.c:168 [inline] RIP: 0010:oom_unkillable_task+0x180/0x400 mm/oom_kill.c:155 Code: c1 ea 03 80 3c 02 00 0f 85 80 02 00 00 4c 8b a3 10 07 00 00 48 b8 00 00 00 00 00 fc ff df 4d 8d 74 24 10 4c 89 f2 48 c1 ea 03 <80> 3c 02 00 0f 85 67 02 00 00 49 8b 44 24 10 4c 8d a0 68 fa ff ff RSP: 0018:ffff888000127490 EFLAGS: 00010a03 RAX: dffffc0000000000 RBX: ffff8880a4cd5438 RCX: ffffffff818dae9c RDX: 100000000c3cc602 RSI: ffffffff818dac8d RDI: 0000000000000001 RBP: ffff8880001274d0 R08: ffff888000086180 R09: ffffed1015d26be0 R10: ffffed1015d26bdf R11: ffff8880ae935efb R12: 8000000061e63007 R13: 0000000000000000 R14: 8000000061e63017 R15: 1ffff11000024ea6 FS: 00005555561f5940(0000) GS:ffff8880ae800000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000001b2f823000 CR3: 000000009237e000 CR4: 00000000001426f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000600 The fix is to decouple the cpuset/mempolicy intersection check from oom_unkillable_task() and make sure cpuset/mempolicy intersection check is only done in the global oom context. [shakeelb@google.com: change function name and update comment] Link: http://lkml.kernel.org/r/20190628152421.198994-3-shakeelb@google.com Link: http://lkml.kernel.org/r/20190624212631.87212-3-shakeelb@google.com Signed-off-by: Shakeel Butt <shakeelb@google.com> Reported-by: syzbot+d0fc9d3c166bc5e4a94b@syzkaller.appspotmail.com Acked-by: Roman Gushchin <guro@fb.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Nick Piggin <npiggin@suse.de> Cc: Paul Jackson <pj@sgi.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-07-12 12:00:31 +08:00
current->mm && !oom_unkillable_task(current) &&
oom_cpuset_eligible(current, oc) &&
current->signal->oom_score_adj != OOM_SCORE_ADJ_MIN) {
get_task_struct(current);
oc->chosen = current;
oom_kill_process(oc, "Out of memory (oom_kill_allocating_task)");
return true;
}
select_bad_process(oc);
mm: memcontrol: print proper OOM header when no eligible victim left When the memcg OOM killer runs out of killable tasks, it currently prints a WARN with no further OOM context. This has caused some user confusion. Warnings indicate a kernel problem. In a reported case, however, the situation was triggered by a nonsensical memcg configuration (hard limit set to 0). But without any VM context this wasn't obvious from the report, and it took some back and forth on the mailing list to identify what is actually a trivial issue. Handle this OOM condition like we handle it in the global OOM killer: dump the full OOM context and tell the user we ran out of tasks. This way the user can identify misconfigurations easily by themselves and rectify the problem - without having to go through the hassle of running into an obscure but unsettling warning, finding the appropriate kernel mailing list and waiting for a kernel developer to remote-analyze that the memcg configuration caused this. If users cannot make sense of why the OOM killer was triggered or why it failed, they will still report it to the mailing list, we know that from experience. So in case there is an actual kernel bug causing this, kernel developers will very likely hear about it. Link: http://lkml.kernel.org/r/20180821160406.22578-1-hannes@cmpxchg.org Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-09-05 06:45:34 +08:00
/* Found nothing?!?! */
if (!oc->chosen) {
dump_header(oc, NULL);
mm: memcontrol: print proper OOM header when no eligible victim left When the memcg OOM killer runs out of killable tasks, it currently prints a WARN with no further OOM context. This has caused some user confusion. Warnings indicate a kernel problem. In a reported case, however, the situation was triggered by a nonsensical memcg configuration (hard limit set to 0). But without any VM context this wasn't obvious from the report, and it took some back and forth on the mailing list to identify what is actually a trivial issue. Handle this OOM condition like we handle it in the global OOM killer: dump the full OOM context and tell the user we ran out of tasks. This way the user can identify misconfigurations easily by themselves and rectify the problem - without having to go through the hassle of running into an obscure but unsettling warning, finding the appropriate kernel mailing list and waiting for a kernel developer to remote-analyze that the memcg configuration caused this. If users cannot make sense of why the OOM killer was triggered or why it failed, they will still report it to the mailing list, we know that from experience. So in case there is an actual kernel bug causing this, kernel developers will very likely hear about it. Link: http://lkml.kernel.org/r/20180821160406.22578-1-hannes@cmpxchg.org Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-09-05 06:45:34 +08:00
pr_warn("Out of memory and no killable processes...\n");
/*
* If we got here due to an actual allocation at the
* system level, we cannot survive this and will enter
* an endless loop in the allocator. Bail out now.
*/
if (!is_sysrq_oom(oc) && !is_memcg_oom(oc))
panic("System is deadlocked on memory\n");
}
if (oc->chosen && oc->chosen != (void *)-1UL)
oom_kill_process(oc, !is_memcg_oom(oc) ? "Out of memory" :
"Memory cgroup out of memory");
return !!oc->chosen;
oom, PM: make OOM detection in the freezer path raceless Commit 5695be142e20 ("OOM, PM: OOM killed task shouldn't escape PM suspend") has left a race window when OOM killer manages to note_oom_kill after freeze_processes checks the counter. The race window is quite small and really unlikely and partial solution deemed sufficient at the time of submission. Tejun wasn't happy about this partial solution though and insisted on a full solution. That requires the full OOM and freezer's task freezing exclusion, though. This is done by this patch which introduces oom_sem RW lock and turns oom_killer_disable() into a full OOM barrier. oom_killer_disabled check is moved from the allocation path to the OOM level and we take oom_sem for reading for both the check and the whole OOM invocation. oom_killer_disable() takes oom_sem for writing so it waits for all currently running OOM killer invocations. Then it disable all the further OOMs by setting oom_killer_disabled and checks for any oom victims. Victims are counted via mark_tsk_oom_victim resp. unmark_oom_victim. The last victim wakes up all waiters enqueued by oom_killer_disable(). Therefore this function acts as the full OOM barrier. The page fault path is covered now as well although it was assumed to be safe before. As per Tejun, "We used to have freezing points deep in file system code which may be reacheable from page fault." so it would be better and more robust to not rely on freezing points here. Same applies to the memcg OOM killer. out_of_memory tells the caller whether the OOM was allowed to trigger and the callers are supposed to handle the situation. The page allocation path simply fails the allocation same as before. The page fault path will retry the fault (more on that later) and Sysrq OOM trigger will simply complain to the log. Normally there wouldn't be any unfrozen user tasks after try_to_freeze_tasks so the function will not block. But if there was an OOM killer racing with try_to_freeze_tasks and the OOM victim didn't finish yet then we have to wait for it. This should complete in a finite time, though, because - the victim cannot loop in the page fault handler (it would die on the way out from the exception) - it cannot loop in the page allocator because all the further allocation would fail and __GFP_NOFAIL allocations are not acceptable at this stage - it shouldn't be blocked on any locks held by frozen tasks (try_to_freeze expects lockless context) and kernel threads and work queues are not frozen yet Signed-off-by: Michal Hocko <mhocko@suse.cz> Suggested-by: Tejun Heo <tj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-02-12 07:26:24 +08:00
}
/*
* The pagefault handler calls here because it is out of memory, so kill a
* memory-hogging task. If oom_lock is held by somebody else, a parallel oom
* killing is already in progress so do nothing.
*/
void pagefault_out_of_memory(void)
{
struct oom_control oc = {
.zonelist = NULL,
.nodemask = NULL,
.memcg = NULL,
.gfp_mask = 0,
.order = 0,
};
if (mem_cgroup_oom_synchronize(true))
return;
mm: memcg: do not trap chargers with full callstack on OOM The memcg OOM handling is incredibly fragile and can deadlock. When a task fails to charge memory, it invokes the OOM killer and loops right there in the charge code until it succeeds. Comparably, any other task that enters the charge path at this point will go to a waitqueue right then and there and sleep until the OOM situation is resolved. The problem is that these tasks may hold filesystem locks and the mmap_sem; locks that the selected OOM victim may need to exit. For example, in one reported case, the task invoking the OOM killer was about to charge a page cache page during a write(), which holds the i_mutex. The OOM killer selected a task that was just entering truncate() and trying to acquire the i_mutex: OOM invoking task: mem_cgroup_handle_oom+0x241/0x3b0 mem_cgroup_cache_charge+0xbe/0xe0 add_to_page_cache_locked+0x4c/0x140 add_to_page_cache_lru+0x22/0x50 grab_cache_page_write_begin+0x8b/0xe0 ext3_write_begin+0x88/0x270 generic_file_buffered_write+0x116/0x290 __generic_file_aio_write+0x27c/0x480 generic_file_aio_write+0x76/0xf0 # takes ->i_mutex do_sync_write+0xea/0x130 vfs_write+0xf3/0x1f0 sys_write+0x51/0x90 system_call_fastpath+0x18/0x1d OOM kill victim: do_truncate+0x58/0xa0 # takes i_mutex do_last+0x250/0xa30 path_openat+0xd7/0x440 do_filp_open+0x49/0xa0 do_sys_open+0x106/0x240 sys_open+0x20/0x30 system_call_fastpath+0x18/0x1d The OOM handling task will retry the charge indefinitely while the OOM killed task is not releasing any resources. A similar scenario can happen when the kernel OOM killer for a memcg is disabled and a userspace task is in charge of resolving OOM situations. In this case, ALL tasks that enter the OOM path will be made to sleep on the OOM waitqueue and wait for userspace to free resources or increase the group's limit. But a userspace OOM handler is prone to deadlock itself on the locks held by the waiting tasks. For example one of the sleeping tasks may be stuck in a brk() call with the mmap_sem held for writing but the userspace handler, in order to pick an optimal victim, may need to read files from /proc/<pid>, which tries to acquire the same mmap_sem for reading and deadlocks. This patch changes the way tasks behave after detecting a memcg OOM and makes sure nobody loops or sleeps with locks held: 1. When OOMing in a user fault, invoke the OOM killer and restart the fault instead of looping on the charge attempt. This way, the OOM victim can not get stuck on locks the looping task may hold. 2. When OOMing in a user fault but somebody else is handling it (either the kernel OOM killer or a userspace handler), don't go to sleep in the charge context. Instead, remember the OOMing memcg in the task struct and then fully unwind the page fault stack with -ENOMEM. pagefault_out_of_memory() will then call back into the memcg code to check if the -ENOMEM came from the memcg, and then either put the task to sleep on the memcg's OOM waitqueue or just restart the fault. The OOM victim can no longer get stuck on any lock a sleeping task may hold. Debugged by Michal Hocko. Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> Reported-by: azurIt <azurit@pobox.sk> Acked-by: Michal Hocko <mhocko@suse.cz> Cc: David Rientjes <rientjes@google.com> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-09-13 06:13:44 +08:00
if (!mutex_trylock(&oom_lock))
return;
out_of_memory(&oc);
mutex_unlock(&oom_lock);
}