Skip to content

mm/oom_kill: ensure we actually kill all tasks sharing the same mm

Oleg Nesterov lkml 2 versions – In review

Get series with b4
b4 am [email protected]

Versions newest first · linked by author and title

  1. v2 Latest3 patches 23 repliesIn review
  2. v13 patches 11 repliesSuperseded

Patches in v2 No applies recorded yet

#Subject and filesSizeRepliesApplied
1/3mm/oom_kill: remove the wrong fatal_signal_pending() check in oom_kill_process()mm/oom_kill.c+0 -418 replies
2/3mm/oom_kill: cleanup the "kill sharing same memory" loopmm/oom_kill.c+11 -72 replies
3/3mm/oom_kill: fix the wrong task->mm == mm checks in oom_kill_process()mm/oom_kill.c+14 -23 replies

Reviews trailers per patch

All 3 patches have a Reviewed-by or Acked-by.

PatchDavid RientjesTetsuo HandaMichal Hocko
1/3A-b A-b A-b
2/3A-b A-b
3/3A-b A-b

Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes. Scroll the table sideways for all 3 reviewers.

Read the v2 thread

Cover letter [PATCH -mm v2 0/3] mm/oom_kill: ensure we actually kill all tasks sharing the same mm
Quoted 9 lines · Oleg Nesterov

Let me send some initial changes which imo makes sense regardless, but if we want to zap the victim's memory we need to ensure that all tasks which share this ->mm were actually killed (see 3/3).

Please review, this series is simple but only compile tested.

Andrew, this is on top of linux-mmotm.git + the recent mm-oom-remove- task_lock-protecting-comm-printing.patch from David.

Please consider v2 based on the comments from Tetsuo and David (thanks a lot!).

Oleg.