On Tue, Aug 20, 2019 at 8:25 PM David Rientjes <rientjes@xxxxxxxxxx> wrote:
>
> On Tue, 20 Aug 2019, Edward Chron wrote:
>
> > For an OOM event: print oom_score_adj value for the OOM Killed process to
> > document what the oom score adjust value was at the time the process was
> > OOM Killed. The adjustment value can be set by user code and it affects
> > the resulting oom_score so it is used to influence kill process selection.
> >
> > When eligible tasks are not printed (sysctl oom_dump_tasks = 0) printing
> > this value is the only documentation of the value for the process being
> > killed. Having this value on the Killed process message documents if a
> > miscconfiguration occurred or it can confirm that the oom_score_adj
> > value applies as expected.
> >
> > An example which illustates both misconfiguration and validation that
> > the oom_score_adj was applied as expected is:
> >
> > Aug 14 23:00:02 testserver kernel: Out of memory: Killed process 2692
> > (systemd-udevd) total-vm:1056800kB, anon-rss:1052760kB, file-rss:4kB,
> > shmem-rss:0kB oom_score_adj:1000
> >
> > The systemd-udevd is a critical system application that should have an
> > oom_score_adj of -1000. Here it was misconfigured to have a adjustment
> > of 1000 making it a highly favored OOM kill target process. The output
> > documents both the misconfiguration and the fact that the process
> > was correctly targeted by OOM due to the miconfiguration. Having
> > the oom_score_adj on the Killed message ensures that it is documented.
> >
> > Signed-off-by: Edward Chron <echron@xxxxxxxxxx>
> > Acked-by: Michal Hocko <mhocko@xxxxxxxx>
>
> Acked-by: David Rientjes <rientjes@xxxxxxxxxx>
>
> vm.oom_dump_tasks is pretty useful, however, so it's curious why you
> haven't left it enabled :/
>
> > diff --git a/mm/oom_kill.c b/mm/oom_kill.c
> > index eda2e2a0bdc6..c781f73b6cd6 100644
> > --- a/mm/oom_kill.c
> > +++ b/mm/oom_kill.c
> > @@ -884,12 +884,13 @@ static void __oom_kill_process(struct task_struct *victim, const char *message)
> > */
> > do_send_sig_info(SIGKILL, SEND_SIG_PRIV, victim, PIDTYPE_TGID);
> > mark_oom_victim(victim);
> > - pr_err("%s: Killed process %d (%s) total-vm:%lukB, anon-rss:%lukB, file-rss:%lukB, shmem-rss:%lukB\n",
> > + pr_err("%s: Killed process %d (%s) total-vm:%lukB, anon-rss:%lukB, file-rss:%lukB, shmem-rss:%lukB oom_score_adj:%ld\n",
> > message, task_pid_nr(victim), victim->comm,
> > K(victim->mm->total_vm),
> > K(get_mm_counter(victim->mm, MM_ANONPAGES)),
> > K(get_mm_counter(victim->mm, MM_FILEPAGES)),
> > - K(get_mm_counter(victim->mm, MM_SHMEMPAGES)));
> > + K(get_mm_counter(victim->mm, MM_SHMEMPAGES)),
> > + (long)victim->signal->oom_score_adj);
> > task_unlock(victim);
> >
> > /*
>
> Nit: why not just use %hd and avoid the cast to long?
Sorry I may have accidently top posted my response to this. Here is
where my response should go:
-----------------------------------------------------------------------------------------------------------------------------------
Good point, I can post this with your correction.
I will add your Acked-by: David Rientjes <rientjes@xxxxxxxxxx>
I am adding your Acked-by to the revised patch as this is what Michal
asked me to do (so I assume that is what I should do).
Should I post as a separate fix again or simply post here?
I'll post here and if you prefer a fresh submission, let me know and
I'll do that.
Thank-you for reviewing this patch.
