From: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
To: mhocko@suse.com
Cc: linux-mm@kvack.org, rientjes@google.com
Subject: Re: [PATCH 1/2] mm,oom_reaper: Show trace of unable to reap victim thread.
Date: Tue, 20 Mar 2018 21:52:33 +0900 [thread overview]
Message-ID: <201803202152.HED82804.QFOHLMVFFtOOJS@I-love.SAKURA.ne.jp> (raw)
In-Reply-To: <20180320122818.GL23100@dhcp22.suse.cz>
Michal Hocko wrote:
> On Tue 20-03-18 20:57:55, Tetsuo Handa wrote:
> > I found that it is not difficult to hit "oom_reaper: unable to reap pid:"
> > messages if the victim thread is doing copy_process(). Since I noticed
> > that it is likely helpful to show trace of unable to reap victim thread
> > for finding locations which should use killable wait, this patch does so.
> >
> > [ 226.608508] oom_reaper: unable to reap pid:9261 (a.out)
> > [ 226.611971] a.out D13056 9261 6927 0x00100084
> > [ 226.615879] Call Trace:
> > [ 226.617926] ? __schedule+0x25f/0x780
> > [ 226.620559] schedule+0x2d/0x80
> > [ 226.623356] rwsem_down_write_failed+0x2bb/0x440
> > [ 226.626426] ? rwsem_down_write_failed+0x55/0x440
> > [ 226.629458] ? anon_vma_fork+0x124/0x150
> > [ 226.632679] call_rwsem_down_write_failed+0x13/0x20
> > [ 226.635884] down_write+0x49/0x60
> > [ 226.638867] ? copy_process.part.41+0x12f2/0x1fe0
> > [ 226.642042] copy_process.part.41+0x12f2/0x1fe0 /* i_mmap_lock_write() in dup_mmap() */
> > [ 226.645087] ? _do_fork+0xe6/0x560
> > [ 226.647991] _do_fork+0xe6/0x560
> > [ 226.650495] ? syscall_trace_enter+0x1a9/0x240
> > [ 226.653443] ? retint_user+0x18/0x18
> > [ 226.656601] ? page_fault+0x2f/0x50
> > [ 226.659159] ? trace_hardirqs_on_caller+0x11f/0x1b0
> > [ 226.662399] do_syscall_64+0x74/0x230
> > [ 226.664989] entry_SYSCALL_64_after_hwframe+0x42/0xb7
>
> A single stack trace in the changelog would be sufficient IMHO.
> Appart from that. What do you expect users will do about this trace?
> Sure they will see a path which holds mmap_sem, we will see a bug report
> but we can hardly do anything about that. We simply cannot drop the lock
> from that path in 99% of situations. So _why_ do we want to add more
> information to the log?
This case is blocked at i_mmap_lock_write(). If we can add error handling path
there, we can replace i_mmap_lock_write() with i_mmap_lock_write_killable() and
bail out soon. This patch helps finding such locations.
next prev parent reply other threads:[~2018-03-20 12:52 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-03-20 11:57 [PATCH 1/2] mm,oom_reaper: Show trace of unable to reap victim thread Tetsuo Handa
2018-03-20 11:57 ` [PATCH 2/2] mm,oom_reaper: Correct MAX_OOM_REAP_RETRIES'th attempt Tetsuo Handa
2018-03-20 12:12 ` Michal Hocko
2018-03-20 12:37 ` Tetsuo Handa
2018-03-20 12:47 ` [PATCH 2/2] mm,oom_reaper: Check for MMF_OOM_SKIP before complain Tetsuo Handa
2018-03-20 20:49 ` David Rientjes
2018-03-22 10:46 ` [PATCH] " Tetsuo Handa
2018-03-22 11:59 ` Michal Hocko
2018-03-20 13:24 ` [PATCH 2/2] mm,oom_reaper: Correct MAX_OOM_REAP_RETRIES'th attempt Michal Hocko
2018-03-20 13:37 ` Tetsuo Handa
2018-03-20 13:44 ` Michal Hocko
2018-03-20 12:28 ` [PATCH 1/2] mm,oom_reaper: Show trace of unable to reap victim thread Michal Hocko
2018-03-20 12:52 ` Tetsuo Handa [this message]
2018-03-20 13:19 ` Michal Hocko
2018-03-20 13:30 ` Tetsuo Handa
2018-03-20 13:34 ` Michal Hocko
2018-03-20 13:50 ` Tetsuo Handa
2018-03-20 14:10 ` Michal Hocko
2018-03-20 14:20 ` Tetsuo Handa
2018-03-20 20:46 ` David Rientjes
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=201803202152.HED82804.QFOHLMVFFtOOJS@I-love.SAKURA.ne.jp \
--to=penguin-kernel@i-love.sakura.ne.jp \
--cc=linux-mm@kvack.org \
--cc=mhocko@suse.com \
--cc=rientjes@google.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).