From: Johannes Weiner <hannes@cmpxchg.org>
To: Roman Gushchin <guro@fb.com>
Cc: linux-mm@kvack.org, Michal Hocko <mhocko@kernel.org>,
Vladimir Davydov <vdavydov.dev@gmail.com>,
Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>,
David Rientjes <rientjes@google.com>,
Andrew Morton <akpm@linux-foundation.org>,
Tejun Heo <tj@kernel.org>,
kernel-team@fb.com, cgroups@vger.kernel.org,
linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [v10 4/6] mm, oom: introduce memory.oom_group
Date: Wed, 4 Oct 2017 15:37:00 -0400 [thread overview]
Message-ID: <20171004193700.GD1501@cmpxchg.org> (raw)
In-Reply-To: <20171004154638.710-5-guro@fb.com>
On Wed, Oct 04, 2017 at 04:46:36PM +0100, Roman Gushchin wrote:
> The cgroup-aware OOM killer treats leaf memory cgroups as memory
> consumption entities and performs the victim selection by comparing
> them based on their memory footprint. Then it kills the biggest task
> inside the selected memory cgroup.
>
> But there are workloads, which are not tolerant to a such behavior.
> Killing a random task may leave the workload in a broken state.
>
> To solve this problem, memory.oom_group knob is introduced.
> It will define, whether a memory group should be treated as an
> indivisible memory consumer, compared by total memory consumption
> with other memory consumers (leaf memory cgroups and other memory
> cgroups with memory.oom_group set), and whether all belonging tasks
> should be killed if the cgroup is selected.
>
> If set on memcg A, it means that in case of system-wide OOM or
> memcg-wide OOM scoped to A or any ancestor cgroup, all tasks,
> belonging to the sub-tree of A will be killed. If OOM event is
> scoped to a descendant cgroup (A/B, for example), only tasks in
> that cgroup can be affected. OOM killer will never touch any tasks
> outside of the scope of the OOM event.
>
> Also, tasks with oom_score_adj set to -1000 will not be killed.
>
> The default value is 0.
>
> Signed-off-by: Roman Gushchin <guro@fb.com>
> Cc: Michal Hocko <mhocko@kernel.org>
> Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
> Cc: Johannes Weiner <hannes@cmpxchg.org>
> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
> Cc: David Rientjes <rientjes@google.com>
> Cc: Andrew Morton <akpm@linux-foundation.org>
> Cc: Tejun Heo <tj@kernel.org>
> Cc: kernel-team@fb.com
> Cc: cgroups@vger.kernel.org
> Cc: linux-doc@vger.kernel.org
> Cc: linux-kernel@vger.kernel.org
> Cc: linux-mm@kvack.org
Those semantics make sense to me and the code looks good.
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
WARNING: multiple messages have this Message-ID (diff)
From: Johannes Weiner <hannes@cmpxchg.org>
To: Roman Gushchin <guro@fb.com>
Cc: linux-mm@kvack.org, Michal Hocko <mhocko@kernel.org>,
Vladimir Davydov <vdavydov.dev@gmail.com>,
Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>,
David Rientjes <rientjes@google.com>,
Andrew Morton <akpm@linux-foundation.org>,
Tejun Heo <tj@kernel.org>,
kernel-team@fb.com, cgroups@vger.kernel.org,
linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [v10 4/6] mm, oom: introduce memory.oom_group
Date: Wed, 4 Oct 2017 15:37:00 -0400 [thread overview]
Message-ID: <20171004193700.GD1501@cmpxchg.org> (raw)
In-Reply-To: <20171004154638.710-5-guro@fb.com>
On Wed, Oct 04, 2017 at 04:46:36PM +0100, Roman Gushchin wrote:
> The cgroup-aware OOM killer treats leaf memory cgroups as memory
> consumption entities and performs the victim selection by comparing
> them based on their memory footprint. Then it kills the biggest task
> inside the selected memory cgroup.
>
> But there are workloads, which are not tolerant to a such behavior.
> Killing a random task may leave the workload in a broken state.
>
> To solve this problem, memory.oom_group knob is introduced.
> It will define, whether a memory group should be treated as an
> indivisible memory consumer, compared by total memory consumption
> with other memory consumers (leaf memory cgroups and other memory
> cgroups with memory.oom_group set), and whether all belonging tasks
> should be killed if the cgroup is selected.
>
> If set on memcg A, it means that in case of system-wide OOM or
> memcg-wide OOM scoped to A or any ancestor cgroup, all tasks,
> belonging to the sub-tree of A will be killed. If OOM event is
> scoped to a descendant cgroup (A/B, for example), only tasks in
> that cgroup can be affected. OOM killer will never touch any tasks
> outside of the scope of the OOM event.
>
> Also, tasks with oom_score_adj set to -1000 will not be killed.
>
> The default value is 0.
>
> Signed-off-by: Roman Gushchin <guro@fb.com>
> Cc: Michal Hocko <mhocko@kernel.org>
> Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
> Cc: Johannes Weiner <hannes@cmpxchg.org>
> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
> Cc: David Rientjes <rientjes@google.com>
> Cc: Andrew Morton <akpm@linux-foundation.org>
> Cc: Tejun Heo <tj@kernel.org>
> Cc: kernel-team@fb.com
> Cc: cgroups@vger.kernel.org
> Cc: linux-doc@vger.kernel.org
> Cc: linux-kernel@vger.kernel.org
> Cc: linux-mm@kvack.org
Those semantics make sense to me and the code looks good.
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
next prev parent reply other threads:[~2017-10-04 19:37 UTC|newest]
Thread overview: 99+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-10-04 15:46 [v10 0/6] cgroup-aware OOM killer Roman Gushchin
2017-10-04 15:46 ` Roman Gushchin
2017-10-04 15:46 ` [v10 1/6] mm, oom: refactor the oom_kill_process() function Roman Gushchin
2017-10-04 15:46 ` Roman Gushchin
2017-10-04 15:46 ` Roman Gushchin
[not found] ` <20171004154638.710-2-guro-b10kYP2dOMg@public.gmane.org>
2017-10-04 19:14 ` Johannes Weiner
2017-10-04 19:14 ` Johannes Weiner
2017-10-04 19:14 ` Johannes Weiner
2017-10-04 15:46 ` [v10 2/6] mm: implement mem_cgroup_scan_tasks() for the root memory cgroup Roman Gushchin
2017-10-04 15:46 ` Roman Gushchin
2017-10-04 19:15 ` Johannes Weiner
2017-10-04 19:15 ` Johannes Weiner
2017-10-04 20:10 ` David Rientjes
2017-10-04 20:10 ` David Rientjes
2017-10-04 15:46 ` [v10 3/6] mm, oom: cgroup-aware OOM killer Roman Gushchin
2017-10-04 15:46 ` Roman Gushchin
2017-10-04 19:27 ` Johannes Weiner
2017-10-04 19:27 ` Johannes Weiner
2017-10-04 19:51 ` Roman Gushchin
2017-10-04 19:51 ` Roman Gushchin
2017-10-04 19:51 ` Roman Gushchin
2017-10-04 20:17 ` David Rientjes
2017-10-04 20:17 ` David Rientjes
[not found] ` <alpine.DEB.2.10.1710041316120.67374-X6Q0R45D7oAcqpCFd4KODRPsWskHk0ljAL8bYrjMMd8@public.gmane.org>
2017-10-04 20:22 ` Roman Gushchin
2017-10-04 20:22 ` Roman Gushchin
2017-10-04 20:22 ` Roman Gushchin
2017-10-04 20:31 ` Johannes Weiner
2017-10-04 20:31 ` Johannes Weiner
2017-10-05 11:14 ` Michal Hocko
2017-10-05 11:14 ` Michal Hocko
2017-10-04 19:48 ` Shakeel Butt
2017-10-04 19:48 ` Shakeel Butt
2017-10-04 20:15 ` Roman Gushchin
2017-10-04 20:15 ` Roman Gushchin
2017-10-04 21:24 ` Shakeel Butt
2017-10-04 21:24 ` Shakeel Butt
[not found] ` <CALvZod45ObeQwq-pKeqyLe2bNwfKAr0majCbNfqPOEJL+AeiNw-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2017-10-05 10:27 ` Roman Gushchin
2017-10-05 10:27 ` Roman Gushchin
2017-10-05 10:27 ` Roman Gushchin
[not found] ` <20171005102707.GA12982-2xczL/1GIl5a1dPMsufgnw2O0Ztt9esIQQ4Iyu8u01E@public.gmane.org>
2017-10-05 11:12 ` Michal Hocko
2017-10-05 11:12 ` Michal Hocko
2017-10-05 11:12 ` Michal Hocko
2017-10-05 11:45 ` Roman Gushchin
2017-10-05 11:45 ` Roman Gushchin
[not found] ` <20171004154638.710-4-guro-b10kYP2dOMg@public.gmane.org>
2017-10-04 20:27 ` David Rientjes
2017-10-04 20:27 ` David Rientjes
2017-10-04 20:27 ` David Rientjes
2017-10-04 20:41 ` Johannes Weiner
2017-10-04 20:41 ` Johannes Weiner
2017-10-05 8:40 ` David Rientjes
2017-10-05 8:40 ` David Rientjes
2017-10-05 10:27 ` Johannes Weiner
2017-10-05 10:27 ` Johannes Weiner
2017-10-05 21:53 ` David Rientjes
2017-10-05 21:53 ` David Rientjes
2017-10-05 10:44 ` Roman Gushchin
2017-10-05 10:44 ` Roman Gushchin
2017-10-05 22:02 ` David Rientjes
2017-10-05 22:02 ` David Rientjes
2017-10-06 5:43 ` Michal Hocko
2017-10-06 5:43 ` Michal Hocko
2017-10-05 11:40 ` Michal Hocko
2017-10-05 11:40 ` Michal Hocko
2017-10-04 15:46 ` [v10 4/6] mm, oom: introduce memory.oom_group Roman Gushchin
2017-10-04 15:46 ` Roman Gushchin
2017-10-04 15:46 ` Roman Gushchin
2017-10-04 19:37 ` Johannes Weiner [this message]
2017-10-04 19:37 ` Johannes Weiner
2017-10-05 12:06 ` Michal Hocko
2017-10-05 12:06 ` Michal Hocko
2017-10-05 12:32 ` Roman Gushchin
2017-10-05 12:32 ` Roman Gushchin
2017-10-05 12:32 ` Roman Gushchin
2017-10-05 12:58 ` Michal Hocko
2017-10-05 12:58 ` Michal Hocko
2017-10-04 15:46 ` [v10 5/6] mm, oom: add cgroup v2 mount option for cgroup-aware OOM killer Roman Gushchin
2017-10-04 15:46 ` Roman Gushchin
[not found] ` <20171004154638.710-6-guro-b10kYP2dOMg@public.gmane.org>
2017-10-04 20:04 ` Johannes Weiner
2017-10-04 20:04 ` Johannes Weiner
2017-10-04 20:04 ` Johannes Weiner
[not found] ` <20171004200453.GE1501-druUgvl0LCNAfugRpC6u6w@public.gmane.org>
2017-10-05 13:14 ` Michal Hocko
2017-10-05 13:14 ` Michal Hocko
2017-10-05 13:14 ` Michal Hocko
2017-10-05 13:41 ` Roman Gushchin
2017-10-05 13:41 ` Roman Gushchin
2017-10-05 13:41 ` Roman Gushchin
2017-10-05 14:10 ` Michal Hocko
2017-10-05 14:10 ` Michal Hocko
2017-10-05 14:54 ` Johannes Weiner
2017-10-05 14:54 ` Johannes Weiner
2017-10-05 16:40 ` Michal Hocko
2017-10-05 16:40 ` Michal Hocko
2017-10-05 15:51 ` Tejun Heo
2017-10-05 15:51 ` Tejun Heo
2017-10-04 15:46 ` [v10 6/6] mm, oom, docs: describe the " Roman Gushchin
2017-10-04 15:46 ` Roman Gushchin
[not found] ` <20171004154638.710-7-guro-b10kYP2dOMg@public.gmane.org>
2017-10-04 20:08 ` Johannes Weiner
2017-10-04 20:08 ` Johannes Weiner
2017-10-04 20:08 ` Johannes Weiner
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=20171004193700.GD1501@cmpxchg.org \
--to=hannes@cmpxchg.org \
--cc=akpm@linux-foundation.org \
--cc=cgroups@vger.kernel.org \
--cc=guro@fb.com \
--cc=kernel-team@fb.com \
--cc=linux-doc@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mhocko@kernel.org \
--cc=penguin-kernel@I-love.SAKURA.ne.jp \
--cc=rientjes@google.com \
--cc=tj@kernel.org \
--cc=vdavydov.dev@gmail.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.