From: Vladimir Davydov <vdavydov@virtuozzo.com>
To: Johannes Weiner <hannes@cmpxchg.org>
Cc: Andrew Morton <akpm@linux-foundation.org>,
Michal Hocko <mhocko@suse.cz>,
linux-mm@kvack.org, cgroups@vger.kernel.org,
linux-kernel@vger.kernel.org, kernel-team@fb.com
Subject: Re: [PATCH] mm: memcontrol: clean up alloc, online, offline, free functions fix
Date: Thu, 17 Dec 2015 11:39:31 +0300 [thread overview]
Message-ID: <20151217083931.GM28521@esperanza> (raw)
In-Reply-To: <1450312460-27582-1-git-send-email-hannes@cmpxchg.org>
On Wed, Dec 16, 2015 at 07:34:20PM -0500, Johannes Weiner wrote:
> Fixlets based on review feedback from Vladimir:
>
> 1. The memcg_create_mutex is to stabilize a cgroup's hereditary
> settings that are not allowed to change once the cgroup has
> children: kmem accounting and hierarchy mode. However, the cleanup
> patch moves inheritance of these settings from onlining time to
> allocation time, before the new child will show up in the parent's
> list of children, and this opens a race window where the parent can
> change a setting that has been passed on to a new child already.
>
> That being said, this rule for kmem and hierarchy mode is somewhat
> gratuitous: there is no strong reason why these configurations
> shouldn't exist, and the outcome of a race is not harmful. It's
> also unlikely that somebody will even trigger this race because we
> don't expect anybody to flip-flop either settings while creating
> child groups. So instead of readding complexity to close an
> unlikely race window that doesn't do any harm, simply remove the
> now pointless mutex as a follow-up cleanup.
>
> 2. Kmem initialization consists of several steps that are undone in
> both css_offline() and css_free(). However, if css allocation fails
> later on then css_offline() is never called and we don't properly
> free the kmem state. Let css_free() detect this and call kmem
> offlining itself.
>
> 3. Children in !use_hierarchy mode would inherit the OOM killer
> setting from their physical parent rather than the logical parent,
> rootmemcg. This is silly, but no reason to change the semantics as
> part of this cleanup patch, so restore it.
>
> Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Vladimir Davydov <vdavydov@virtuozzo.com>
--
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>
prev parent reply other threads:[~2015-12-17 8:39 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-12-17 0:34 Johannes Weiner
2015-12-17 8:39 ` Vladimir Davydov [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=20151217083931.GM28521@esperanza \
--to=vdavydov@virtuozzo.com \
--cc=akpm@linux-foundation.org \
--cc=cgroups@vger.kernel.org \
--cc=hannes@cmpxchg.org \
--cc=kernel-team@fb.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mhocko@suse.cz \
/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