mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Michal Hocko <mhocko@suse.cz>
To: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Johannes Weiner <jweiner@redhat.com>,
	linux-mm@kvack.org, Balbir Singh <bsingharora@gmail.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH RFC] memcg: fix drain_all_stock crash
Date: Wed, 10 Aug 2011 09:29:47 +0200	[thread overview]
Message-ID: <20110810072947.GB12560@tiehlicka.suse.cz> (raw)
In-Reply-To: <20110810085437.ed023651.kamezawa.hiroyu@jp.fujitsu.com>

On Wed 10-08-11 08:54:37, KAMEZAWA Hiroyuki wrote:
> On Tue, 9 Aug 2011 13:46:42 +0200
> Michal Hocko <mhocko@suse.cz> wrote:
> 
> > On Tue 09-08-11 19:07:25, KAMEZAWA Hiroyuki wrote:
> > > On Tue, 9 Aug 2011 12:09:44 +0200
> > > Michal Hocko <mhocko@suse.cz> wrote:
> > > 
> > > > On Tue 09-08-11 18:53:13, KAMEZAWA Hiroyuki wrote:
> > > > > On Tue, 9 Aug 2011 11:45:03 +0200
> > > > > Michal Hocko <mhocko@suse.cz> wrote:
> > > > > 
> > > > > > On Tue 09-08-11 18:32:16, KAMEZAWA Hiroyuki wrote:
> > > > > > > On Tue, 9 Aug 2011 11:31:50 +0200
> > > > > > > Michal Hocko <mhocko@suse.cz> wrote:
> > > > > > > 
> > > > > > > > What do you think about the half backed patch bellow? I didn't manage to
> > > > > > > > test it yet but I guess it should help. I hate asymmetry of drain_lock
> > > > > > > > locking (it is acquired somewhere else than it is released which is
> > > > > > > > not). I will think about a nicer way how to do it.
> > > > > > > > Maybe I should also split the rcu part in a separate patch.
> > > > > > > > 
> > > > > > > > What do you think?
> > > > > > > 
> > > > > > > 
> > > > > > > I'd like to revert 8521fc50 first and consider total design change
> > > > > > > rather than ad-hoc fix.
> > > > > > 
> > > > > > Agreed. Revert should go into 3.0 stable as well. Although the global
> > > > > > mutex is buggy we have that behavior for a long time without any reports.
> > > > > > We should address it but it can wait for 3.2.
> > > > 
> > > > I will send the revert request to Linus.
> > > > 
> > > > > What "buggy" means here ? "problematic" or "cause OOps ?"
> > > > 
> > > > I have described that in an earlier email. Consider pathological case
> > > > when CPU0 wants to async. drain a memcg which has a lot of cached charges while
> > > > CPU1 is already draining so it holds the mutex. CPU0 backs off so it has
> > > > to reclaim although we could prevent from it by getting rid of cached
> > > > charges. This is not critical though.
> > > > 
> > > 
> > > That problem should be fixed by background reclaim.
> > 
> > How? Do you plan to rework locking or the charge caching completely?
> > 
> 
> From your description, the problem is not the lock itself but a task
> may go into _unnecessary_ direct-reclaim even if there are remaining
> chages on per-cpu stocks, which cause latency.

The problem is partly the lock because it prevents from parallel async
reclaimers. This is too restrictive. If we are going to rely on async
draining we will have to above problem.

> 
> In (all) my automatic background reclaim tests, no direct reclaim happens
> if background reclaim is enabled. 

Yes, I haven't seen this problem yet but I guess that there is non 0
chance that there is a workload which triggers this.

> And as I said before, we may be able to add a flag not to cache
> more. It's set by some condition ....as usage is near to the limit.
> 
> Thanks,
> -Kame

Thanks
-- 
Michal Hocko
SUSE Labs
SUSE LINUX s.r.o.
Lihovarska 1060/12
190 00 Praha 9    
Czech Republic

      reply	other threads:[~2011-08-10  7:30 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <cover.1311338634.git.mhocko@suse.cz>
     [not found] ` <2f17df54db6661c39a05669d08a9e6257435b898.1311338634.git.mhocko@suse.cz>
2011-07-25  1:16   ` [PATCH 1/4] memcg: do not try to drain per-cpu caches without pages KAMEZAWA Hiroyuki
2011-08-17 19:49     ` [patch] memcg: pin execution to current cpu while draining stock Johannes Weiner
2011-08-18  0:24       ` KAMEZAWA Hiroyuki
2011-08-18  5:56       ` Michal Hocko
     [not found] ` <a9244082ba28c4c2e4a6997311d5493bdaa117e9.1311338634.git.mhocko@suse.cz>
2011-07-25  1:18   ` [PATCH 4/4] memcg: get rid of percpu_charge_mutex lock KAMEZAWA Hiroyuki
2011-08-08 18:47   ` Johannes Weiner
2011-08-08 21:47     ` Michal Hocko
2011-08-08 23:19       ` Johannes Weiner
2011-08-09  7:26         ` Michal Hocko
2011-08-09  9:31           ` [PATCH RFC] memcg: fix drain_all_stock crash Michal Hocko
2011-08-09  9:32             ` KAMEZAWA Hiroyuki
2011-08-09  9:45               ` Michal Hocko
2011-08-09  9:53                 ` KAMEZAWA Hiroyuki
2011-08-09 10:09                   ` Michal Hocko
2011-08-09 10:07                     ` KAMEZAWA Hiroyuki
2011-08-09 11:46                       ` Michal Hocko
2011-08-09 23:54                         ` KAMEZAWA Hiroyuki
2011-08-10  7:29                           ` Michal Hocko [this message]

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=20110810072947.GB12560@tiehlicka.suse.cz \
    --to=mhocko@suse.cz \
    --cc=akpm@linux-foundation.org \
    --cc=bsingharora@gmail.com \
    --cc=jweiner@redhat.com \
    --cc=kamezawa.hiroyu@jp.fujitsu.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    /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

all inboxes | Powered by JetHome®