Message ID | 1560853257-14934-1-git-send-email-penguin-kernel@I-love.SAKURA.ne.jp (mailing list archive) |
---|---|
State | New, archived |
Headers | show |
Series | mm: oom: Remove thread group leader check in oom_evaluate_task(). | expand |
On Tue 18-06-19 19:20:57, Tetsuo Handa wrote: > Since mem_cgroup_scan_tasks() uses CSS_TASK_ITER_PROCS, only thread group > leaders will be scanned (unless dying leaders with live threads). Thus, > commit d49ad9355420c743 ("mm, oom: prefer thread group leaders for display > purposes") makes little sense. This can be folded into mm-memcontrol-use-css_task_iter_procs-at-mem_cgroup_scan_tasks.patch right? > Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> > Cc: David Rientjes <rientjes@google.com> > Cc: Johannes Weiner <hannes@cmpxchg.org> > Cc: Michal Hocko <mhocko@suse.cz> > Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> > Cc: Greg Thelen <gthelen@google.com> > --- > mm/oom_kill.c | 3 --- > 1 file changed, 3 deletions(-) > > diff --git a/mm/oom_kill.c b/mm/oom_kill.c > index 32abc7a..09a5116 100644 > --- a/mm/oom_kill.c > +++ b/mm/oom_kill.c > @@ -348,9 +348,6 @@ static int oom_evaluate_task(struct task_struct *task, void *arg) > if (!points || points < oc->chosen_points) > goto next; > > - /* Prefer thread group leaders for display purposes */ > - if (points == oc->chosen_points && thread_group_leader(oc->chosen)) > - goto next; > select: > if (oc->chosen) > put_task_struct(oc->chosen); > -- > 1.8.3.1
On 2019/06/18 21:14, Michal Hocko wrote: > On Tue 18-06-19 19:20:57, Tetsuo Handa wrote: >> Since mem_cgroup_scan_tasks() uses CSS_TASK_ITER_PROCS, only thread group >> leaders will be scanned (unless dying leaders with live threads). Thus, >> commit d49ad9355420c743 ("mm, oom: prefer thread group leaders for display >> purposes") makes little sense. > > This can be folded into mm-memcontrol-use-css_task_iter_procs-at-mem_cgroup_scan_tasks.patch > right? Yes, if we want to do so.
diff --git a/mm/oom_kill.c b/mm/oom_kill.c index 32abc7a..09a5116 100644 --- a/mm/oom_kill.c +++ b/mm/oom_kill.c @@ -348,9 +348,6 @@ static int oom_evaluate_task(struct task_struct *task, void *arg) if (!points || points < oc->chosen_points) goto next; - /* Prefer thread group leaders for display purposes */ - if (points == oc->chosen_points && thread_group_leader(oc->chosen)) - goto next; select: if (oc->chosen) put_task_struct(oc->chosen);
Since mem_cgroup_scan_tasks() uses CSS_TASK_ITER_PROCS, only thread group leaders will be scanned (unless dying leaders with live threads). Thus, commit d49ad9355420c743 ("mm, oom: prefer thread group leaders for display purposes") makes little sense. Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Michal Hocko <mhocko@suse.cz> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Greg Thelen <gthelen@google.com> --- mm/oom_kill.c | 3 --- 1 file changed, 3 deletions(-)