public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Benjamin LaHaise <bcrl@kvack.org>
To: Sebastian Ott <sebott@linux.vnet.ibm.com>
Cc: Anatol Pomozov <anatol.pomozov@gmail.com>,
	linux-aio@kvack.org, linux-kernel@vger.kernel.org
Subject: Re: hanging aio process
Date: Tue, 20 May 2014 10:26:46 -0400	[thread overview]
Message-ID: <20140520142646.GG2915@kvack.org> (raw)
In-Reply-To: <alpine.LFD.2.11.1405201508450.1606@denkbrett>

On Tue, May 20, 2014 at 03:16:47PM +0200, Sebastian Ott wrote:
> On Tue, 20 May 2014, Sebastian Ott wrote:
> > On Mon, 19 May 2014, Benjamin LaHaise wrote:
> > > It is entirely possible the bug isn't 
> > > caused by the referenced commit, as the commit you're pointing to merely 
> > > makes io_destroy() syscall wait for all aio outstanding to complete 
> > > before returning.
> > 
> > I cannot reproduce this when I revert said commit (on top of 14186fe). If
> > that matters - the arch is s390.

> Hm, ok - maybe that commit is really just highlighting a refcounting bug.
> I just compared traces for a good and a few bad cases. The good case:
...

> (4 fio workers, free_ioctx_reqs is called 4 times)


> One of the bad cases:
....
> (1 fio worker in D state, free_ioctx_reqs is called 3 times)

This would seem to indicate that the problem is not with Anatol's change, 
and the hang is a consequence of the AIO not completing.  Can you trace 
calls to aio_complete() in addition to free_ioctx_reqs() to see if a 
completion is happening in the failed case?  If aio_complete() is only 
getting called 3 times, the problem is not in the aio layer.

		-ben


> Regards,
> Sebastian
> 
> > > 
> > > > git bisect points to:
> > > > 	commit e02ba72aabfade4c9cd6e3263e9b57bf890ad25c
> > > > 	Author: Anatol Pomozov <anatol.pomozov@gmail.com>
> > > > 	Date:   Tue Apr 15 11:31:33 2014 -0700
> > > > 
> > > > 	    aio: block io_destroy() until all context requests are completed
> > > > 
> > > > 
> > > > The fio workers are on the wait_for_completion in sys_io_destroy.
> > > > 
> > > > Regards,
> > > > Sebastian
> > > > [global]
> > > > blocksize=4K
> > > > size=256M
> > > > rw=randrw
> > > > verify=md5
> > > > iodepth=32
> > > > ioengine=libaio
> > > > direct=1
> > > > end_fsync=1
> > > > 
> > > > [file1]
> > > > filename=/dev/scma
> > > > 
> > > > [file2]
> > > > filename=/dev/scmbw
> > > > 
> > > > [file3]
> > > > filename=/dev/scmc
> > > > 
> > > > [file4]
> > > > filename=/dev/scmx
> > > 
> > > 
> > > -- 
> > > "Thought is the essence of where you are now."
> > > 
> > > 
> > 

-- 
"Thought is the essence of where you are now."

  reply	other threads:[~2014-05-20 14:26 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-05-19 17:38 hanging aio process Sebastian Ott
2014-05-19 18:08 ` Benjamin LaHaise
2014-05-20  8:08   ` Sebastian Ott
2014-05-20 13:16     ` Sebastian Ott
2014-05-20 14:26       ` Benjamin LaHaise [this message]
2014-05-20 16:12       ` Anatol Pomozov
2014-05-20 18:09         ` Sebastian Ott
2014-05-20 18:39           ` Anatol Pomozov
2014-05-21  8:07             ` Sebastian Ott
2014-05-21  8:48               ` Sebastian Ott
2014-05-21 13:20                 ` Benjamin LaHaise
2014-05-21 14:12                   ` Sebastian Ott
2014-05-21 14:18                 ` Anatol Pomozov
2014-05-21 14:57                   ` Sebastian Ott
2014-06-04  5:24                     ` Gu Zheng
2014-06-11  8:47                       ` Ming Lei
2014-06-04 13:58                   ` [PATCH] percpu-refcount: fix usage of this_cpu_ops (was Re: hanging aio process) Sebastian Ott
2014-06-04 16:11                     ` Tejun Heo
2014-06-04 16:56                       ` Sebastian Ott
2014-06-04 17:02                         ` Tejun Heo
2014-06-04 16:41                     ` Tejun Heo
2014-06-04 17:17                       ` Christoph Lameter
2014-05-19 18:31 ` hanging aio process Anatol Pomozov

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=20140520142646.GG2915@kvack.org \
    --to=bcrl@kvack.org \
    --cc=anatol.pomozov@gmail.com \
    --cc=linux-aio@kvack.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=sebott@linux.vnet.ibm.com \
    /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