On Thu 21-04-22 15:05:33, Nico Pache wrote:
> The MMF_OOM_SKIP bit is used to indicate weather a mm_struct can not be
> invalided or has already been invalided. exit_mmap currently calls
> __oom_reap_task_mm unconditionally despite the fact that the oom reaper
> may have already called this.
>
> Add a check for the MMF_OOM_SKIP bit being set in exit_mmap to avoid
> unnessary calls to the invalidate code.
Why do we care about this?
> A slight race can occur on the MMF_OOM_SKIP bit that will still allow
> this to run twice. My testing has shown an ~66% decrease in double calls
> to _oom_reap_task_mm.
>
> Fixes: 27ae357fa82b ("mm, oom: fix concurrent munlock and oom reaper unmap, v3")
I do not see this would be fixing anything.
> Cc: David Rientjes <rientjes@xxxxxxxxxx>
> Cc: Michal Hocko <mhocko@xxxxxxxx>
> Cc: Andrea Arcangeli <aarcange@xxxxxxxxxx>
> Signed-off-by: Nico Pache <npache@xxxxxxxxxx>
> ---
> mm/mmap.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/mm/mmap.c b/mm/mmap.c
> index a2968669fd4e..b867f408dacd 100644
> --- a/mm/mmap.c
> +++ b/mm/mmap.c
> @@ -3113,7 +3113,8 @@ void exit_mmap(struct mm_struct *mm)
> /* mm's last user has gone, and its about to be pulled down */
> mmu_notifier_release(mm);
>
> - if (unlikely(mm_is_oom_victim(mm))) {
> + if (unlikely(mm_is_oom_victim(mm)) &&
> + !test_bit(MMF_OOM_SKIP, &mm->flags)) {
> /*
> * Manually reap the mm to free as much memory as possible.
> * Then, as the oom reaper does, set MMF_OOM_SKIP to disregard
> --
> 2.35.1
--
Michal Hocko
SUSE Labs
