Re: can't oom-kill zap the victim's memory?

[Michal Hocko]

On Tue 06-10-15 20:45:02, Oleg Nesterov wrote:
[...]
> And I was going to make V1 which avoids queue_work/kthread and zaps the
> memory in oom_kill_process() context.
> 
> But this can't work because we need to increment ->mm_users to avoid
> the race with exit_mmap/etc. And this means that we need mmput() after
> that, and as we recently discussed it can deadlock if mm_users goes
> to zero, we can't do exit_mmap/etc in oom_kill_process().

Right. I hoped we could rely on mm_count just to pin mm but that is not
sufficient because exit_mmap doesn't rely on mmap_sem so we do not have
any synchronization there. Unfortunate. This means that we indeed have
to do it asynchronously. Maybe we can come up with some trickery but
let's do it later. I do agree that going with a kernel thread for now
would be easier. Sorry about misleading you, I should have realized that
mmput from the oom killing path is dangerous.

-- 
Michal Hocko
SUSE Labs
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/