From: Vivek Goyal <vgoyal@redhat.com>
To: Tejun Heo <tj@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>,
axboe@kernel.dk, hughd@google.com, avi@redhat.com,
nate@cpanel.net, cl@linux-foundation.org,
linux-kernel@vger.kernel.org, dpshah@google.com,
ctalbott@google.com, rni@google.com
Subject: Re: [PATCHSET] mempool, percpu, blkcg: fix percpu stat allocation and remove stats_lock
Date: Thu, 8 Mar 2012 15:16:16 -0500 [thread overview]
Message-ID: <20120308201616.GD22922@redhat.com> (raw)
In-Reply-To: <20120308180833.GA25508@google.com>
On Thu, Mar 08, 2012 at 10:08:33AM -0800, Tejun Heo wrote:
[..]
> > Tejun, I noticed that in UP case, once in a while cgroup removal is
> > hanging. Looks like it is hung in cgroup_rmdir() somewhere. I will debug
> > more to find out what is happening. May be blkcg->refcount issue.
>
> It's probably from something forgetting to put cgroup and pre_destroy
> waiting for it. Such bugs would have been masked before but now show
> up as stalls during rmdir.
I am not sure what is happening here yet. What I have noticed that
somebody is holding a reference on blkg->refcnt and that's why css->refcnt
is not zero hence rmdir is hanging.
I susect it is cfqq refcount on blkg which is not released till cfqq is
reclaimed.
Looking at the code, in general it seems to be a problem. If a task
issues bunch of IO, changes the cgroup and does not issue IO any more
for some time, that means old cfqq will still be linked to task's
cic and still be holding reference to blkg and one can't remove the
cgroup.
We had this disucssion in the past. So looks like to get rid of this
problem, you will have to drop old cic->cfqq association during
cgroup change to avoid hanging rmdir.
Thanks
Vivek
next prev parent reply other threads:[~2012-03-08 20:16 UTC|newest]
Thread overview: 56+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-02-23 22:30 [PATCHSET] mempool, percpu, blkcg: fix percpu stat allocation and remove stats_lock Tejun Heo
2012-02-23 22:30 ` [PATCH 1/8] mempool: factor out mempool_fill() Tejun Heo
2012-02-23 22:30 ` [PATCH 2/8] mempool: separate out __mempool_create() Tejun Heo
2012-02-23 22:30 ` [PATCH 3/8] mempool, percpu: implement percpu mempool Tejun Heo
2012-02-23 22:30 ` [PATCH 4/8] block: fix deadlock through percpu allocation in blk-cgroup Tejun Heo
2012-02-23 22:30 ` [PATCH 5/8] blkcg: don't use percpu for merged stats Tejun Heo
2012-02-23 22:30 ` [PATCH 6/8] blkcg: simplify stat reset Tejun Heo
2012-02-23 22:30 ` [PATCH 7/8] blkcg: restructure blkio_get_stat() Tejun Heo
2012-02-23 22:30 ` [PATCH 8/8] blkcg: remove blkio_group->stats_lock Tejun Heo
2012-02-23 22:43 ` [PATCHSET] mempool, percpu, blkcg: fix percpu stat allocation and remove stats_lock Andrew Morton
2012-02-23 23:01 ` Tejun Heo
2012-02-23 23:12 ` Tejun Heo
2012-02-23 23:22 ` Andrew Morton
2012-02-23 23:24 ` Tejun Heo
2012-02-24 14:20 ` Vivek Goyal
2012-02-25 21:44 ` Tejun Heo
2012-02-27 3:11 ` Vivek Goyal
2012-02-27 9:11 ` Tejun Heo
2012-02-27 19:43 ` Vivek Goyal
2012-02-29 17:36 ` Vivek Goyal
2012-03-05 22:13 ` Tejun Heo
2012-03-06 21:09 ` Vivek Goyal
2012-03-06 21:20 ` Andrew Morton
2012-03-06 21:34 ` Vivek Goyal
2012-03-06 21:55 ` Andrew Morton
2012-03-07 14:55 ` Vivek Goyal
2012-03-07 17:05 ` Tejun Heo
2012-03-07 19:13 ` Vivek Goyal
2012-03-07 19:22 ` Tejun Heo
2012-03-07 19:42 ` Vivek Goyal
2012-03-07 22:56 ` Tejun Heo
2012-03-07 23:08 ` Andrew Morton
2012-03-07 23:15 ` Tejun Heo
2012-03-07 23:05 ` Andrew Morton
2012-03-08 17:57 ` Vivek Goyal
2012-03-08 18:08 ` Tejun Heo
2012-03-08 18:11 ` Tejun Heo
2012-03-08 18:22 ` Vivek Goyal
2012-03-08 18:27 ` Tejun Heo
2012-03-15 16:48 ` Vivek Goyal
2012-03-15 16:59 ` Tejun Heo
2012-03-20 11:50 ` Jens Axboe
2012-03-08 20:16 ` Vivek Goyal [this message]
2012-03-08 20:33 ` Vivek Goyal
2012-03-08 20:35 ` Tejun Heo
2012-03-08 19:06 ` Andrew Morton
2012-02-25 3:44 ` Vivek Goyal
2012-02-25 21:46 ` Tejun Heo
2012-02-25 22:21 ` Tejun Heo
2012-02-27 14:25 ` Vivek Goyal
2012-02-27 14:40 ` Vivek Goyal
2012-03-05 17:45 ` Tejun Heo
2012-02-27 18:22 ` Vivek Goyal
2012-02-29 19:03 ` Vivek Goyal
2012-03-05 17:20 ` Tejun Heo
2012-03-05 18:03 ` 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=20120308201616.GD22922@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=ctalbott@google.com \
--cc=dpshah@google.com \
--cc=hughd@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=nate@cpanel.net \
--cc=rni@google.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).