linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Vivek Goyal <vgoyal@redhat.com>
To: Tejun Heo <tj@kernel.org>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	avi@redhat.com, nate@cpanel.net, cl@linux-foundation.org,
	oleg@redhat.com, axboe@kernel.dk, linux-kernel@vger.kernel.org
Subject: Re: [PATCHSET] block, mempool, percpu: implement percpu mempool and fix blkcg percpu alloc deadlock
Date: Mon, 16 Jan 2012 10:28:53 -0500	[thread overview]
Message-ID: <20120116152853.GB9129@redhat.com> (raw)
In-Reply-To: <20120105012842.GQ31746@google.com>

On Wed, Jan 04, 2012 at 05:28:42PM -0800, Tejun Heo wrote:
> On Tue, Dec 27, 2011 at 04:41:02PM -0800, Tejun Heo wrote:
> > Hello,
> > 
> > On Wed, Dec 28, 2011 at 09:14:02AM +0900, KAMEZAWA Hiroyuki wrote:
> > > > This is essentially more specialized form of the mempool approach.  It
> > > > doesn't seem any simpler to me while being less generic.  I don't see
> > > > what the upside would be.
> > > 
> > > Hm, but this never causes -ENOMEM error, at all.
> > 
> > Ooh, I missed the part it falls back to the global counter if percpu
> > counters aren't allocated yet.  Yeah, this is an interesting approach.
> > I'll think more about it.
> 
> I've been staring at the blkcg stats code and commit logs and am
> wondering whether we can just scrap percpu counters there.  It seems
> the reason why it was introduced in the first place is to avoid
> stats->lock, which BTW is extremely heavy handed for gathering stats,
> overhead in fast paths and I think there can be easier ways to avoid
> stats->lock.

Tejun, 

If you get rid of stats->lock, and also don't use per cpu data strucutres
how would we avoid races in case of concurrent update (IO submission?).

Thanks
Vivek

  reply	other threads:[~2012-01-16 15:29 UTC|newest]

Thread overview: 51+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-12-22 21:45 [PATCHSET] block, mempool, percpu: implement percpu mempool and fix blkcg percpu alloc deadlock Tejun Heo
2011-12-22 21:45 ` [PATCH 1/7] mempool: fix and document synchronization and memory barrier usage Tejun Heo
2011-12-22 21:45 ` [PATCH 2/7] mempool: drop unnecessary and incorrect BUG_ON() from mempool_destroy() Tejun Heo
2011-12-22 21:45 ` [PATCH 3/7] mempool: fix first round failure behavior Tejun Heo
2011-12-22 21:45 ` [PATCH 4/7] mempool: factor out mempool_fill() Tejun Heo
2011-12-22 21:45 ` [PATCH 5/7] mempool: separate out __mempool_create() Tejun Heo
2011-12-22 21:45 ` [PATCH 6/7] mempool, percpu: implement percpu mempool Tejun Heo
2011-12-22 21:45 ` [PATCH 7/7] block: fix deadlock through percpu allocation in blk-cgroup Tejun Heo
2011-12-23  1:00   ` Vivek Goyal
2011-12-23 22:54     ` Tejun Heo
2011-12-22 21:59 ` [PATCHSET] block, mempool, percpu: implement percpu mempool and fix blkcg percpu alloc deadlock Andrew Morton
2011-12-22 22:09   ` Tejun Heo
2011-12-22 22:20     ` Andrew Morton
2011-12-22 22:41       ` Tejun Heo
2011-12-22 22:54         ` Andrew Morton
2011-12-22 23:00           ` Tejun Heo
2011-12-22 23:16             ` Andrew Morton
2011-12-22 23:24               ` Tejun Heo
2011-12-22 23:41                 ` Andrew Morton
2011-12-22 23:54                   ` Tejun Heo
2011-12-23  1:14                     ` Andrew Morton
2011-12-23 15:17                       ` Vivek Goyal
2011-12-27 18:34                       ` Tejun Heo
2011-12-27 21:20                         ` Andrew Morton
2011-12-27 21:44                           ` Tejun Heo
2011-12-27 21:58                             ` Andrew Morton
2011-12-27 22:22                               ` Tejun Heo
2011-12-23  1:21                   ` Vivek Goyal
2011-12-23  1:38                     ` Andrew Morton
2011-12-23  2:54                       ` Vivek Goyal
2011-12-23  3:11                         ` Andrew Morton
2011-12-23 14:58                           ` Vivek Goyal
2011-12-27 21:25                             ` Andrew Morton
2011-12-27 22:07                               ` Tejun Heo
2011-12-27 22:21                                 ` Andrew Morton
2011-12-27 22:30                                   ` Tejun Heo
2012-01-16 15:26                                     ` Vivek Goyal
2011-12-23  1:40       ` Vivek Goyal
2011-12-23  1:58         ` Andrew Morton
2011-12-23  2:56           ` Vivek Goyal
2011-12-26  6:05             ` KAMEZAWA Hiroyuki
2011-12-27 17:52               ` Tejun Heo
2011-12-28  0:14                 ` KAMEZAWA Hiroyuki
2011-12-28  0:41                   ` Tejun Heo
2012-01-05  1:28                     ` Tejun Heo
2012-01-16 15:28                       ` Vivek Goyal [this message]
2012-02-09 23:58                       ` Tejun Heo
2012-02-10 16:26                         ` Vivek Goyal
2012-02-13 22:31                           ` Tejun Heo
2012-02-15 15:43                             ` Vivek Goyal
2011-12-23 14:46           ` Vivek Goyal

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=20120116152853.GB9129@redhat.com \
    --to=vgoyal@redhat.com \
    --cc=akpm@linux-foundation.org \
    --cc=avi@redhat.com \
    --cc=axboe@kernel.dk \
    --cc=cl@linux-foundation.org \
    --cc=kamezawa.hiroyu@jp.fujitsu.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=nate@cpanel.net \
    --cc=oleg@redhat.com \
    --cc=tj@kernel.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;
as well as URLs for NNTP newsgroup(s).