On Tue 12-07-16 22:29:23, Tetsuo Handa wrote:
> We can revert commit e2fe14564d3316d1 ("oom_reaper: close race with
> exiting task") because oom_has_pending_mm() which will return true until
> exit_oom_mm() is called after OOM victim's mm is reclaimed by __mmput()
> or oom_reap_task() can close that race.
>
> Signed-off-by: Tetsuo Handa <penguin-kernel@xxxxxxxxxxxxxxxxxxx>
Acked-by: Michal Hocko <mhocko@xxxxxxxx>
> ---
> mm/oom_kill.c | 29 ++++-------------------------
> 1 file changed, 4 insertions(+), 25 deletions(-)
>
> diff --git a/mm/oom_kill.c b/mm/oom_kill.c
> index fab0bec..232c1ce 100644
> --- a/mm/oom_kill.c
> +++ b/mm/oom_kill.c
> @@ -476,28 +476,9 @@ static bool __oom_reap_task(struct task_struct *tsk, struct mm_struct *mm)
> struct vm_area_struct *vma;
> struct zap_details details = {.check_swap_entries = true,
> .ignore_dirty = true};
> - bool ret = true;
>
> - /*
> - * We have to make sure to not race with the victim exit path
> - * and cause premature new oom victim selection:
> - * __oom_reap_task exit_mm
> - * mmget_not_zero
> - * mmput
> - * atomic_dec_and_test
> - * exit_oom_victim
> - * [...]
> - * out_of_memory
> - * select_bad_process
> - * # no TIF_MEMDIE task selects new victim
> - * unmap_page_range # frees some memory
> - */
> - mutex_lock(&oom_lock);
> -
> - if (!down_read_trylock(&mm->mmap_sem)) {
> - ret = false;
> - goto unlock_oom;
> - }
> + if (!down_read_trylock(&mm->mmap_sem))
> + return false;
>
> /*
> * increase mm_users only after we know we will reap something so
> @@ -506,7 +487,7 @@ static bool __oom_reap_task(struct task_struct *tsk, struct mm_struct *mm)
> */
> if (!mmget_not_zero(mm)) {
> up_read(&mm->mmap_sem);
> - goto unlock_oom;
> + return true;
> }
>
> tlb_gather_mmu(&tlb, mm, 0, -1);
> @@ -554,9 +535,7 @@ static bool __oom_reap_task(struct task_struct *tsk, struct mm_struct *mm)
> * put the oom_reaper out of the way.
> */
> mmput_async(mm);
> -unlock_oom:
> - mutex_unlock(&oom_lock);
> - return ret;
> + return true;
> }
>
> #define MAX_OOM_REAP_RETRIES 10
> --
> 1.8.3.1
>
--
Michal Hocko
SUSE Labs
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>