mm, oom: fix missing tlb_finish_mmu() in __oom_reap_task_mm().
authorTetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>
Tue, 4 Sep 2018 22:45:37 +0000 (15:45 -0700)
committerLinus Torvalds <torvalds@linux-foundation.org>
Tue, 4 Sep 2018 23:45:02 +0000 (16:45 -0700)
Commit 93065ac753e4 ("mm, oom: distinguish blockable mode for mmu
notifiers") has added an ability to skip over vmas with blockable mmu
notifiers. This however didn't call tlb_finish_mmu as it should.

As a result inc_tlb_flush_pending has been called without its pairing
dec_tlb_flush_pending and all callers mm_tlb_flush_pending would flush
even though this is not really needed.  This alone is not harmful and it
seems there shouldn't be any such callers for oom victims at all but
there is no real reason to skip tlb_finish_mmu on early skip either so
call it.

[mhocko@suse.com: new changelog]
Link: http://lkml.kernel.org/r/b752d1d5-81ad-7a35-2394-7870641be51c@i-love.sakura.ne.jp
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>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm/oom_kill.c

index 95fbbc46f68f32fcbc3731afcee857fd80a8575b..f10aa5360616ea3247b6e508ed28ace237d70ad9 100644 (file)
@@ -522,6 +522,7 @@ bool __oom_reap_task_mm(struct mm_struct *mm)
 
                        tlb_gather_mmu(&tlb, mm, start, end);
                        if (mmu_notifier_invalidate_range_start_nonblock(mm, start, end)) {
+                               tlb_finish_mmu(&tlb, start, end);
                                ret = false;
                                continue;
                        }