On 2019/01/09 20:03, Michal Hocko wrote: > Tetsuo, > can you confirm that these two patches are fixing the issue you have > reported please? > My patch fixes the issue better than your "[PATCH 2/2] memcg: do not report racy no-eligible OOM tasks" does. You can post "[PATCH 1/2] mm, oom: marks all killed tasks as oom victims" based on a report that we needlessly select more OOM victims because MMF_OOM_SKIP is quickly set by the OOM reaper. In fact, updating oom_reap_task() / exit_mmap() to use mutex_lock(&oom_lock); set_bit(MMF_OOM_SKIP, &mm->flags); mutex_unlock(&oom_lock); will mostly close the race as well.
