Re: [patch v2] mm, oom: fix concurrent munlock and oom reaper unmap

[David Rientjes <rientjes@google.com>]

On Wed, 25 Apr 2018, Tetsuo Handa wrote:

> > One of the reasons that I extracted __oom_reap_task_mm() out of the new 
> > oom_reap_task_mm() is to avoid the checks that would be unnecessary when 
> > called from exit_mmap().  In this case, we can ignore the 
> > mm_has_blockable_invalidate_notifiers() check because exit_mmap() has 
> > already done mmu_notifier_release().  So I don't think there's a concern 
> > about __oom_reap_task_mm() blocking while holding oom_lock.  Unless you 
> > are referring to something else?
> 
> Oh, mmu_notifier_release() made mm_has_blockable_invalidate_notifiers() == false. OK.
> 
> But I want comments why it is safe; I will probably miss that dependency
> when we move that code next time.
> 

Ok, makes sense.  I'll send a v3 to update the comment.