linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Michal Hocko <mhocko@suse.cz>
To: linux-mm@kvack.org
Cc: Greg Thelen <gthelen@google.com>,
	Glauber Costa <glommer@gmail.com>, Mel Gorman <mgorman@suse.de>,
	Andrew Morton <akpm@linux-foundation.org>,
	Johannes Weiner <hannes@cmpxchg.org>,
	KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>,
	David Rientjes <rientjes@google.com>,
	Rik van Riel <riel@redhat.com>, Joern Engel <joern@logfs.org>,
	Hugh Dickins <hughd@google.com>,
	LKML <linux-kernel@vger.kernel.org>
Subject: Re: user defined OOM policies
Date: Tue, 19 Nov 2013 14:40:07 +0100	[thread overview]
Message-ID: <20131119134007.GD20655@dhcp22.suse.cz> (raw)
In-Reply-To: <20131119131400.GC20655@dhcp22.suse.cz>

On Tue 19-11-13 14:14:00, Michal Hocko wrote:
[...]
> We have basically ended up with 3 options AFAIR:
> 	1) allow memcg approach (memcg.oom_control) on the root level
>            for both OOM notification and blocking OOM killer and handle
>            the situation from the userspace same as we can for other
> 	   memcgs.

This looks like a straightforward approach as the similar thing is done
on the local (memcg) level. There are several problems though.
Running userspace from within OOM context is terribly hard to do
right. This is true even in the memcg case and we strongly discurage
users from doing that. The global case has nothing like outside of OOM
context though. So any hang would blocking the whole machine. Even
if the oom killer is careful and locks in all the resources it would
have hard time to query the current system state (existing processes
and their states) without any allocation.  There are certain ways to
workaround these issues - e.g. give the killer access to memory reserves
- but this all looks scary and fragile.

> 	2) allow modules to hook into OOM killer path and take the
> 	   appropriate action.

This already exists actually. There is oom_notify_list callchain and
{un}register_oom_notifier that allow modules to hook into oom and
skip the global OOM if some memory is freed. There are currently only
s390 and powerpc which seem to abuse it for something that looks like a
shrinker except it is done in OOM path...

I think the interface should be changed if something like this would be
used in practice. There is a lot of information lost on the way. I would
basically expect to get everything that out_of_memory gets.

> 	3) create a generic filtering mechanism which could be
> 	   controlled from the userspace by a set of rules (e.g.
> 	   something analogous to packet filtering).

This looks generic enough but I have no idea about the complexity.
-- 
Michal Hocko
SUSE Labs

--
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>

  reply	other threads:[~2013-11-19 13:40 UTC|newest]

Thread overview: 65+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-11-19 13:14 user defined OOM policies Michal Hocko
2013-11-19 13:40 ` Michal Hocko [this message]
2013-11-20  8:02   ` David Rientjes
2013-11-20 15:22     ` Michal Hocko
2013-11-20 17:14       ` Luigi Semenzato
2013-11-21  3:36         ` David Rientjes
2013-11-21  7:03           ` Luigi Semenzato
2013-11-22 18:08             ` Johannes Weiner
2013-11-28 11:36               ` Michal Hocko
2013-11-26  1:29             ` David Rientjes
2013-11-28 11:42               ` Michal Hocko
2013-12-02 23:09                 ` David Rientjes
2013-11-21  3:33       ` David Rientjes
2013-11-28 11:54         ` Michal Hocko
2013-12-02 23:07           ` David Rientjes
2013-12-04  5:19             ` [patch 1/8] fork: collapse copy_flags into copy_process David Rientjes
2013-12-04  5:19               ` [patch 2/8] mm, mempolicy: rename slab_node for clarity David Rientjes
2013-12-04 15:21                 ` Christoph Lameter
2013-12-04  5:20               ` [patch 3/8] mm, mempolicy: remove per-process flag David Rientjes
2013-12-04 15:24                 ` Christoph Lameter
2013-12-05  0:53                   ` David Rientjes
2013-12-05 19:05                     ` Christoph Lameter
2013-12-05 23:53                       ` David Rientjes
2013-12-06 14:46                         ` Christoph Lameter
2013-12-04  5:20               ` [patch 4/8] mm, memcg: add tunable for oom reserves David Rientjes
2013-12-04  5:20               ` [patch 5/8] res_counter: remove interface for locked charging and uncharging David Rientjes
2013-12-04  5:20               ` [patch 6/8] res_counter: add interface for maximum nofail charge David Rientjes
2013-12-04  5:20               ` [patch 7/8] mm, memcg: allow processes handling oom notifications to access reserves David Rientjes
2013-12-04  5:45                 ` Johannes Weiner
2013-12-05  1:49                   ` David Rientjes
2013-12-05  2:50                     ` Tejun Heo
2013-12-05 23:49                       ` David Rientjes
2013-12-06 17:34                         ` Johannes Weiner
2013-12-07 16:38                           ` Tim Hockin
2013-12-07 17:40                             ` Johannes Weiner
2013-12-07 18:12                               ` Tim Hockin
2013-12-07 19:06                                 ` Johannes Weiner
2013-12-07 21:04                                   ` Tim Hockin
2013-12-06 19:01                         ` Tejun Heo
2013-12-09 20:10                           ` David Rientjes
2013-12-09 22:37                             ` Johannes Weiner
2013-12-10 21:50                             ` Tejun Heo
2013-12-10 23:55                               ` David Rientjes
2013-12-11  9:49                                 ` Mel Gorman
2013-12-11 12:42                                 ` Tejun Heo
2013-12-12  5:37                                   ` Tim Hockin
2013-12-12 14:21                                     ` Tejun Heo
2013-12-12 16:32                                       ` Michal Hocko
2013-12-12 16:37                                         ` Tejun Heo
2013-12-12 18:42                                       ` Tim Hockin
2013-12-12 19:23                                         ` Tejun Heo
2013-12-13  0:23                                           ` Tim Hockin
2013-12-13 11:47                                             ` Tejun Heo
2013-12-04  5:20               ` [patch 8/8] mm, memcg: add memcg oom reserve documentation David Rientjes
2013-11-20 17:25     ` user defined OOM policies Vladimir Murzin
2013-11-20 17:21   ` Vladimir Murzin
2013-11-20 17:33     ` Michal Hocko
2013-11-21  3:38       ` David Rientjes
2013-11-21 17:13         ` Michal Hocko
2013-11-26  1:36           ` David Rientjes
2013-11-22  7:28       ` Vladimir Murzin
2013-11-22 13:18         ` Michal Hocko
2013-11-20  7:50 ` David Rientjes
2013-11-22  0:19 ` Jörn Engel
2013-11-26  1:31   ` 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=20131119134007.GD20655@dhcp22.suse.cz \
    --to=mhocko@suse.cz \
    --cc=akpm@linux-foundation.org \
    --cc=glommer@gmail.com \
    --cc=gthelen@google.com \
    --cc=hannes@cmpxchg.org \
    --cc=hughd@google.com \
    --cc=joern@logfs.org \
    --cc=kosaki.motohiro@jp.fujitsu.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mgorman@suse.de \
    --cc=riel@redhat.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).