Date: Wed, 11 Feb 2015 17:30:30 -0500
From: Tejun Heo <tj@kernel.org>
To: Konstantin Khlebnikov <koct9i@gmail.com>
Cc: Greg Thelen <gthelen@google.com>,
        Konstantin Khlebnikov <khlebnikov@yandex-team.ru>,
        Johannes Weiner <hannes@cmpxchg.org>, Michal Hocko <mhocko@suse.cz>,
        Cgroups <cgroups@vger.kernel.org>,
        "linux-mm@kvack.org" <linux-mm@kvack.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        Jan Kara <jack@suse.cz>, Dave Chinner <david@fromorbit.com>,
        Jens Axboe <axboe@kernel.dk>, Christoph Hellwig <hch@infradead.org>,
        Li Zefan <lizefan@huawei.com>, Hugh Dickins <hughd@google.com>
Subject: Re: [RFC] Making memcg track ownership per address_space or anon_vma
Message-ID: <20150211223030.GB12728@htj.duckdns.org>
References: <CAHH2K0bxvc34u1PugVQsSfxXhmN8qU6KRpiCWwOVBa6BPqMDOg@mail.gmail.com>
 <20150207143839.GA9926@htj.dyndns.org>
 <20150211021906.GA21356@htj.duckdns.org>
 <CAHH2K0aHM=jmzbgkSCdFX0NxWbHBcVXqi3EAr0MS-gE3Txk93w@mail.gmail.com>
 <20150211203359.GF21356@htj.duckdns.org>
 <CALYGNiMm2VajBx0Y+XtLJ8860JS-GHfuSXQrBt32Wt0K7QpH0A@mail.gmail.com>
 <20150211214650.GA11920@htj.duckdns.org>
 <CALYGNiPX89HsgUS8BrJvL_jW-EU95xezc7uPf=0Pm72qiUwp7A@mail.gmail.com>
 <20150211220530.GA12728@htj.duckdns.org>
 <CALYGNiMgpU51vNr186x6h-uh_9NqaTqZ_a2L60XG0STozy=30g@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <CALYGNiMgpU51vNr186x6h-uh_9NqaTqZ_a2L60XG0STozy=30g@mail.gmail.com>
User-Agent: Mutt/1.5.23 (2014-03-12)
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1131
Lines: 28

Hello,

On Thu, Feb 12, 2015 at 02:15:29AM +0400, Konstantin Khlebnikov wrote:
> Well, ok. Even if shared writes are rare whey should be handled somehow
> without relying on kupdate-like writeback. If memcg has a lot of dirty pages

This only works iff we consider those cases to be marginal enough to
be handle them in a pretty ghetto way.

> but their inodes are accidentially belong to wrong wb queues when tasks in
> that memcg shouldn't stuck in balance-dirty-pages until somebody outside
> acidentially writes this data. That's all what I wanted to say.

But, right, yeah, corner cases around this could be nasty if writeout
interval is set really high.  I don't think it matters for the default
5s interval at all.  Maybe what we need is queueing a delayed per-wb
work w/ the default writeout interval when dirtying a foreign inode.
I'll think more about it.

Thanks.

-- 
tejun
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/