linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
* [PATCH] mm: oom_kill: add trace logs in process_mrelease() system call
@ 2021-08-16  9:39 Charan Teja Reddy
  2021-08-16 14:19 ` Shakeel Butt
  0 siblings, 1 reply; 2+ messages in thread
From: Charan Teja Reddy @ 2021-08-16  9:39 UTC (permalink / raw)
  To: akpm, rientjes, shakeelb, surenb, mhocko, hannes
  Cc: vinmenon, linux-mm, linux-kernel, Charan Teja Reddy

The process_mrelease() system call[1] is used to release the memory of
a dying process from the context of the caller, which is similar to and
uses the functions of the oom reaper logic. There exists trace logs for
a process when reaped by the oom reaper. Just extend the same to when
done by the process_mrelease() system call.

[1] https://lore.kernel.org/linux-mm/20210809185259.405936-1-surenb@google.com/

Signed-off-by: Charan Teja Reddy <charante@codeaurora.org>
---
 mm/oom_kill.c | 14 +++++++++++++-
 1 file changed, 13 insertions(+), 1 deletion(-)

diff --git a/mm/oom_kill.c b/mm/oom_kill.c
index 431d38c..8f4020a 100644
--- a/mm/oom_kill.c
+++ b/mm/oom_kill.c
@@ -941,7 +941,6 @@ static void __oom_kill_process(struct task_struct *victim, const char *message)
 	mmdrop(mm);
 	put_task_struct(victim);
 }
-#undef K
 
 /*
  * Kill provided task unless it's secured by setting
@@ -1199,8 +1198,18 @@ SYSCALL_DEFINE2(process_mrelease, int, pidfd, unsigned int, flags)
 		ret = -EINTR;
 		goto drop_mm;
 	}
+	trace_start_task_reaping(task->pid);
 	if (!__oom_reap_task_mm(mm))
 		ret = -EAGAIN;
+	if (!ret) {
+		pr_info("process_mrelease: reaped process %d (%s), now anon-rss:%lukB, file-rss:%lukB, shmem-rss:%lukB oom_score_adj:%d\n",
+			task_pid_nr(task), task->comm,
+			K(get_mm_counter(mm, MM_ANONPAGES)),
+			K(get_mm_counter(mm, MM_FILEPAGES)),
+			K(get_mm_counter(mm, MM_SHMEMPAGES)),
+			task->signal->oom_score_adj);
+	}
+	trace_finish_task_reaping(task->pid);
 	mmap_read_unlock(mm);
 
 drop_mm:
@@ -1208,9 +1217,12 @@ SYSCALL_DEFINE2(process_mrelease, int, pidfd, unsigned int, flags)
 put_task:
 	put_task_struct(task);
 put_pid:
+	if (ret && ret != -EAGAIN)
+		trace_skip_task_reaping(pid_vnr(pid));
 	put_pid(pid);
 	return ret;
 #else
 	return -ENOSYS;
 #endif /* CONFIG_MMU */
 }
+#undef K
-- 
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a
member of the Code Aurora Forum, hosted by The Linux Foundation



^ permalink raw reply related	[flat|nested] 2+ messages in thread

* Re: [PATCH] mm: oom_kill: add trace logs in process_mrelease() system call
  2021-08-16  9:39 [PATCH] mm: oom_kill: add trace logs in process_mrelease() system call Charan Teja Reddy
@ 2021-08-16 14:19 ` Shakeel Butt
  0 siblings, 0 replies; 2+ messages in thread
From: Shakeel Butt @ 2021-08-16 14:19 UTC (permalink / raw)
  To: Charan Teja Reddy
  Cc: Andrew Morton, David Rientjes, Suren Baghdasaryan, Michal Hocko,
	Johannes Weiner, vinmenon, Linux MM, LKML

On Mon, Aug 16, 2021 at 2:39 AM Charan Teja Reddy
<charante@codeaurora.org> wrote:
>
> The process_mrelease() system call[1] is used to release the memory of
> a dying process from the context of the caller, which is similar to and
> uses the functions of the oom reaper logic. There exists trace logs for
> a process when reaped by the oom reaper. Just extend the same to when
> done by the process_mrelease() system call.
>
> [1] https://lore.kernel.org/linux-mm/20210809185259.405936-1-surenb@google.com/
>
> Signed-off-by: Charan Teja Reddy <charante@codeaurora.org>
> ---
>  mm/oom_kill.c | 14 +++++++++++++-
>  1 file changed, 13 insertions(+), 1 deletion(-)
>
> diff --git a/mm/oom_kill.c b/mm/oom_kill.c
> index 431d38c..8f4020a 100644
> --- a/mm/oom_kill.c
> +++ b/mm/oom_kill.c
> @@ -941,7 +941,6 @@ static void __oom_kill_process(struct task_struct *victim, const char *message)
>         mmdrop(mm);
>         put_task_struct(victim);
>  }
> -#undef K
>
>  /*
>   * Kill provided task unless it's secured by setting
> @@ -1199,8 +1198,18 @@ SYSCALL_DEFINE2(process_mrelease, int, pidfd, unsigned int, flags)
>                 ret = -EINTR;
>                 goto drop_mm;
>         }
> +       trace_start_task_reaping(task->pid);
>         if (!__oom_reap_task_mm(mm))
>                 ret = -EAGAIN;
> +       if (!ret) {
> +               pr_info("process_mrelease: reaped process %d (%s), now anon-rss:%lukB, file-rss:%lukB, shmem-rss:%lukB oom_score_adj:%d\n",
> +                       task_pid_nr(task), task->comm,
> +                       K(get_mm_counter(mm, MM_ANONPAGES)),
> +                       K(get_mm_counter(mm, MM_FILEPAGES)),
> +                       K(get_mm_counter(mm, MM_SHMEMPAGES)),
> +                       task->signal->oom_score_adj);

Tracing is good but I don't think we want this line to be flooding the
kernel logs. I would recommend either simply removing it or adding
another trace for this, so only folks interested in this info can get
it.


^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2021-08-16 14:19 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-08-16  9:39 [PATCH] mm: oom_kill: add trace logs in process_mrelease() system call Charan Teja Reddy
2021-08-16 14:19 ` Shakeel Butt

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).