linux-toradex.git/mm/oom_kill.c, branch v4.9.87

mm, oom_reaper: gather each vma to prevent leaking TLB entry

2017-12-09T21:01:47+00:00

commit 687cb0884a714ff484d038e9190edc874edcf146 upstream.

tlb_gather_mmu(&tlb, mm, 0, -1) means gathering the whole virtual memory
space.  In this case, tlb->fullmm is true.  Some archs like arm64
doesn't flush TLB when tlb->fullmm is true:

  commit 5a7862e83000 ("arm64: tlbflush: avoid flushing when fullmm == 1").

Which causes leaking of tlb entries.

Will clarifies his patch:
 "Basically, we tag each address space with an ASID (PCID on x86) which
  is resident in the TLB. This means we can elide TLB invalidation when
  pulling down a full mm because we won't ever assign that ASID to
  another mm without doing TLB invalidation elsewhere (which actually
  just nukes the whole TLB).

  I think that means that we could potentially not fault on a kernel
  uaccess, because we could hit in the TLB"

There could be a window between complete_signal() sending IPI to other
cores and all threads sharing this mm are really kicked off from cores.
In this window, the oom reaper may calls tlb_flush_mmu_tlbonly() to
flush TLB then frees pages.  However, due to the above problem, the TLB
entries are not really flushed on arm64.  Other threads are possible to
access these pages through TLB entries.  Moreover, a copy_to_user() can
also write to these pages without generating page fault, causes
use-after-free bugs.

This patch gathers each vma instead of gathering full vm space.  In this
case tlb->fullmm is not true.  The behavior of oom reaper become similar
to munmapping before do_exit, which should be safe for all archs.

Link: http://lkml.kernel.org/r/20171107095453.179940-1-wangnan0@huawei.com
Fixes: aac453635549 ("mm, oom: introduce oom reaper")
Signed-off-by: Wang Nan 
Acked-by: Michal Hocko 
Acked-by: David Rientjes 
Cc: Minchan Kim 
Cc: Will Deacon 
Cc: Bob Liu 
Cc: Ingo Molnar 
Cc: Roman Gushchin 
Cc: Konstantin Khlebnikov 
Cc: Andrea Arcangeli 
Cc: 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds 
[backported to 4.9 stable tree]
Signed-off-by: Michal Hocko 
Signed-off-by: Greg Kroah-Hartman

mm, oom_reaper: skip mm structs with mmu notifiers

2017-10-12T09:51:19+00:00

commit 4d4bbd8526a8fbeb2c090ea360211fceff952383 upstream.

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 
Reported-by: Andrea Arcangeli 
Reviewed-by: Andrea Arcangeli 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds 
Signed-off-by: Greg Kroah-Hartman

oom: print nodemask in the oom report

2016-10-08T01:46:29+00:00

We have received a hard to explain oom report from a customer.  The oom
triggered regardless there is a lot of free memory:

  PoolThread invoked oom-killer: gfp_mask=0x280da, order=0, oom_adj=0, oom_score_adj=0
  PoolThread cpuset=/ mems_allowed=0-7
  Pid: 30055, comm: PoolThread Tainted: G           E X 3.0.101-80-default #1
  Call Trace:
    dump_trace+0x75/0x300
    dump_stack+0x69/0x6f
    dump_header+0x8e/0x110
    oom_kill_process+0xa6/0x350
    out_of_memory+0x2b7/0x310
    __alloc_pages_slowpath+0x7dd/0x820
    __alloc_pages_nodemask+0x1e9/0x200
    alloc_pages_vma+0xe1/0x290
    do_anonymous_page+0x13e/0x300
    do_page_fault+0x1fd/0x4c0
    page_fault+0x25/0x30
  [...]
  active_anon:1135959151 inactive_anon:1051962 isolated_anon:0
   active_file:13093 inactive_file:222506 isolated_file:0
   unevictable:262144 dirty:2 writeback:0 unstable:0
   free:432672819 slab_reclaimable:7917 slab_unreclaimable:95308
   mapped:261139 shmem:166297 pagetables:2228282 bounce:0
  [...]
  Node 0 DMA free:15896kB min:0kB low:0kB high:0kB active_anon:0kB inactive_anon:0kB active_file:0kB inactive_file:0kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:15672kB mlocked:0kB dirty:0kB writeback:0kB mapped:0kB shmem:0kB slab_reclaimable:0kB slab_unreclaimable:0kB kernel_stack:0kB pagetables:0kB unstable:0kB bounce:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? yes
  lowmem_reserve[]: 0 2892 775542 775542
  Node 0 DMA32 free:2783784kB min:28kB low:32kB high:40kB active_anon:0kB inactive_anon:0kB active_file:0kB inactive_file:0kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:2961572kB mlocked:0kB dirty:0kB writeback:0kB mapped:0kB shmem:0kB slab_reclaimable:0kB slab_unreclaimable:0kB kernel_stack:0kB pagetables:0kB unstable:0kB bounce:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? yes
  lowmem_reserve[]: 0 0 772650 772650
  Node 0 Normal free:8120kB min:8160kB low:10200kB high:12240kB active_anon:779334960kB inactive_anon:2198744kB active_file:0kB inactive_file:180kB unevictable:131072kB isolated(anon):0kB isolated(file):0kB present:791193600kB mlocked:131072kB dirty:0kB writeback:0kB mapped:372940kB shmem:361480kB slab_reclaimable:4536kB slab_unreclaimable:68472kB kernel_stack:10104kB pagetables:1414820kB unstable:0kB bounce:0kB writeback_tmp:0kB pages_scanned:2280 all_unreclaimable? yes
  lowmem_reserve[]: 0 0 0 0
  Node 1 Normal free:476718144kB min:8192kB low:10240kB high:12288kB active_anon:307623696kB inactive_anon:283620kB active_file:10392kB inactive_file:69908kB unevictable:131072kB isolated(anon):0kB isolated(file):0kB present:794296320kB mlocked:131072kB dirty:4kB writeback:0kB mapped:257208kB shmem:189896kB slab_reclaimable:3868kB slab_unreclaimable:44756kB kernel_stack:1848kB pagetables:1369432kB unstable:0kB bounce:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no
  lowmem_reserve[]: 0 0 0 0
  Node 2 Normal free:386002452kB min:8192kB low:10240kB high:12288kB active_anon:398563752kB inactive_anon:68184kB active_file:10292kB inactive_file:29936kB unevictable:131072kB isolated(anon):0kB isolated(file):0kB present:794296320kB mlocked:131072kB dirty:0kB writeback:0kB mapped:32084kB shmem:776kB slab_reclaimable:6888kB slab_unreclaimable:60056kB kernel_stack:8208kB pagetables:1282880kB unstable:0kB bounce:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no
  lowmem_reserve[]: 0 0 0 0
  Node 3 Normal free:196406760kB min:8192kB low:10240kB high:12288kB active_anon:587445640kB inactive_anon:164396kB active_file:5716kB inactive_file:709844kB unevictable:131072kB isolated(anon):0kB isolated(file):0kB present:794296320kB mlocked:131072kB dirty:0kB writeback:0kB mapped:291776kB shmem:111416kB slab_reclaimable:5152kB slab_unreclaimable:44516kB kernel_stack:2168kB pagetables:1455956kB unstable:0kB bounce:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no
  lowmem_reserve[]: 0 0 0 0
  Node 4 Normal free:425338880kB min:8192kB low:10240kB high:12288kB active_anon:359695204kB inactive_anon:43216kB active_file:5748kB inactive_file:14772kB unevictable:131072kB isolated(anon):0kB isolated(file):0kB present:794296320kB mlocked:131072kB dirty:0kB writeback:0kB mapped:24708kB shmem:1120kB slab_reclaimable:1884kB slab_unreclaimable:41060kB kernel_stack:1856kB pagetables:1100208kB unstable:0kB bounce:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no
  lowmem_reserve[]: 0 0 0 0
  Node 5 Normal free:11140kB min:8192kB low:10240kB high:12288kB active_anon:784240872kB inactive_anon:1217164kB active_file:28kB inactive_file:48kB unevictable:131072kB isolated(anon):0kB isolated(file):0kB present:794296320kB mlocked:131072kB dirty:0kB writeback:0kB mapped:11408kB shmem:0kB slab_reclaimable:2008kB slab_unreclaimable:49220kB kernel_stack:1360kB pagetables:531600kB unstable:0kB bounce:0kB writeback_tmp:0kB pages_scanned:1202 all_unreclaimable? yes
  lowmem_reserve[]: 0 0 0 0
  Node 6 Normal free:243395332kB min:8192kB low:10240kB high:12288kB active_anon:542015544kB inactive_anon:40208kB active_file:968kB inactive_file:8484kB unevictable:131072kB isolated(anon):0kB isolated(file):0kB present:794296320kB mlocked:131072kB dirty:0kB writeback:0kB mapped:19992kB shmem:496kB slab_reclaimable:1672kB slab_unreclaimable:37052kB kernel_stack:2088kB pagetables:750264kB unstable:0kB bounce:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no
  lowmem_reserve[]: 0 0 0 0
  Node 7 Normal free:10768kB min:8192kB low:10240kB high:12288kB active_anon:784916936kB inactive_anon:192316kB active_file:19228kB inactive_file:56852kB unevictable:131072kB isolated(anon):0kB isolated(file):0kB present:794296320kB mlocked:131072kB dirty:4kB writeback:0kB mapped:34440kB shmem:4kB slab_reclaimable:5660kB slab_unreclaimable:36100kB kernel_stack:1328kB pagetables:1007968kB unstable:0kB bounce:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no
  lowmem_reserve[]: 0 0 0 0

So all nodes but Node 0 have a lot of free memory which should suggest
that there is an available memory especially when mems_allowed=0-7.  One
could speculate that a massive process has managed to terminate and free
up a lot of memory while racing with the above allocation request.
Although this is highly unlikely it cannot be ruled out.

A further debugging, however shown that the faulting process had
mempolicy (not cpuset) to bind to Node 0.  We cannot see that
information from the report though.  mems_allowed turned out to be more
confusing than really helpful.

Fix this by always priting the nodemask.  It is either mempolicy mask
(and non-null) or the one defined by the cpusets.  The new output for
the above oom report would be

  PoolThread invoked oom-killer: gfp_mask=0x280da(GFP_HIGHUSER_MOVABLE|__GFP_ZERO), nodemask=0, order=0, oom_adj=0, oom_score_adj=0

This patch doesn't touch show_mem and the node filtering based on the
cpuset node mask because mempolicy is always a subset of cpusets and
seeing the full cpuset oom context might be helpful for tunning more
specific mempolicies inside cpusets (e.g.  when they turn out to be too
restrictive).  To prevent from ugly ifdefs the mask is printed even for
!NUMA configurations but this should be OK (a single node will be
printed).

Link: http://lkml.kernel.org/r/20160930214146.28600-1-mhocko@kernel.org
Signed-off-by: Michal Hocko 
Reported-by: Sellami Abdelkader 
Acked-by: Vlastimil Babka 
Cc: David Rientjes 
Cc: Sellami Abdelkader 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

mm: don't emit warning from pagefault_out_of_memory()

2016-10-08T01:46:29+00:00

Commit c32b3cbe0d06 ("oom, PM: make OOM detection in the freezer path
raceless") inserted a WARN_ON() into pagefault_out_of_memory() in order
to warn when we raced with disabling the OOM killer.

Now, patch "oom, suspend: fix oom_killer_disable vs.  pm suspend
properly" introduced a timeout for oom_killer_disable().  Even if we
raced with disabling the OOM killer and the system is OOM livelocked,
the OOM killer will be enabled eventually (in 20 seconds by default) and
the OOM livelock will be solved.  Therefore, we no longer need to warn
when we raced with disabling the OOM killer.

Link: http://lkml.kernel.org/r/1473442120-7246-1-git-send-email-penguin-kernel@I-love.SAKURA.ne.jp
Signed-off-by: Tetsuo Handa 
Acked-by: Michal Hocko 
Cc: David Rientjes 
Cc: Johannes Weiner 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

oom: warn if we go OOM for higher order and compaction is disabled

2016-10-08T01:46:28+00:00

Since the lumpy reclaim is gone there is no source of higher order pages
if CONFIG_COMPACTION=n except for the order-0 pages reclaim which is
unreliable for that purpose to say the least.  Hitting an OOM for
!costly higher order requests is therefore all not that hard to imagine.
We are trying hard to not invoke OOM killer as much as possible but
there is simply no reliable way to detect whether more reclaim retries
make sense.

Disabling COMPACTION is not widespread but it seems that some users
might have disable the feature without realizing full consequences
(mostly along with disabling THP because compaction used to be THP
mainly thing).  This patch just adds a note if the OOM killer was
triggered by higher order request with compaction disabled.  This will
help us identifying possible misconfiguration right from the oom report
which is easier than to always keep in mind that somebody might have
disabled COMPACTION without a good reason.

Link: http://lkml.kernel.org/r/20160830111632.GD23963@dhcp22.suse.cz
Signed-off-by: Michal Hocko 
Cc: David Rientjes 
Cc: Johannes Weiner 
Cc: Tetsuo Handa 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

oom, oom_reaper: allow to reap mm shared by the kthreads

2016-10-08T01:46:28+00:00

oom reaper was skipped for an mm which is shared with the kernel thread
(aka use_mm()).  The primary concern was that such a kthread might want
to read from the userspace memory and see zero page as a result of the
oom reaper action.  This is no longer a problem after "mm: make sure
that kthreads will not refault oom reaped memory" because any attempt to
fault in when the MMF_UNSTABLE is set will result in SIGBUS and so the
target user should see an error.  This means that we can finally allow
oom reaper also to tasks which share their mm with kthreads.

Link: http://lkml.kernel.org/r/1472119394-11342-10-git-send-email-mhocko@kernel.org
Signed-off-by: Michal Hocko 
Cc: Tetsuo Handa 
Cc: Oleg Nesterov 
Cc: David Rientjes 
Cc: Vladimir Davydov 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

mm: make sure that kthreads will not refault oom reaped memory

2016-10-08T01:46:28+00:00

There are only few use_mm() users in the kernel right now.  Most of them
write to the target memory but vhost driver relies on
copy_from_user/get_user from a kernel thread context.  This makes it
impossible to reap the memory of an oom victim which shares the mm with
the vhost kernel thread because it could see a zero page unexpectedly
and theoretically make an incorrect decision visible outside of the
killed task context.

To quote Michael S. Tsirkin:
: Getting an error from __get_user and friends is handled gracefully.
: Getting zero instead of a real value will cause userspace
: memory corruption.

The vhost kernel thread is bound to an open fd of the vhost device which
is not tight to the mm owner life cycle in general.  The device fd can
be inherited or passed over to another process which means that we
really have to be careful about unexpected memory corruption because
unlike for normal oom victims the result will be visible outside of the
oom victim context.

Make sure that no kthread context (users of use_mm) can ever see
corrupted data because of the oom reaper and hook into the page fault
path by checking MMF_UNSTABLE mm flag.  __oom_reap_task_mm will set the
flag before it starts unmapping the address space while the flag is
checked after the page fault has been handled.  If the flag is set then
SIGBUS is triggered so any g-u-p user will get a error code.

Regular tasks do not need this protection because all which share the mm
are killed when the mm is reaped and so the corruption will not outlive
them.

This patch shouldn't have any visible effect at this moment because the
OOM killer doesn't invoke oom reaper for tasks with mm shared with
kthreads yet.

Link: http://lkml.kernel.org/r/1472119394-11342-9-git-send-email-mhocko@kernel.org
Signed-off-by: Michal Hocko 
Acked-by: "Michael S. Tsirkin" 
Cc: Tetsuo Handa 
Cc: Oleg Nesterov 
Cc: David Rientjes 
Cc: Vladimir Davydov 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

mm, oom: enforce exit_oom_victim on current task

2016-10-08T01:46:28+00:00

There are no users of exit_oom_victim on !current task anymore so enforce
the API to always work on the current.

Link: http://lkml.kernel.org/r/1472119394-11342-8-git-send-email-mhocko@kernel.org
Signed-off-by: Tetsuo Handa 
Signed-off-by: Michal Hocko 
Cc: Oleg Nesterov 
Cc: David Rientjes 
Cc: Vladimir Davydov 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

oom, suspend: fix oom_killer_disable vs. pm suspend properly

2016-10-08T01:46:27+00:00

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 
Cc: Tetsuo Handa 
Cc: Oleg Nesterov 
Cc: David Rientjes 
Cc: Vladimir Davydov 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

mm, oom: get rid of signal_struct::oom_victims

2016-10-08T01:46:27+00:00

After "oom: keep mm of the killed task available" we can safely detect
an oom victim by checking task->signal->oom_mm so we do not need the
signal_struct counter anymore so let's get rid of it.

This alone wouldn't be sufficient for nommu archs because
exit_oom_victim doesn't hide the process from the oom killer anymore.
We can, however, mark the mm with a MMF flag in __mmput.  We can reuse
MMF_OOM_REAPED and rename it to a more generic MMF_OOM_SKIP.

Link: http://lkml.kernel.org/r/1472119394-11342-6-git-send-email-mhocko@kernel.org
Signed-off-by: Michal Hocko 
Cc: Tetsuo Handa 
Cc: Oleg Nesterov 
Cc: David Rientjes 
Cc: Vladimir Davydov 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds