public inbox for linux-xfs@vger.kernel.org
 help / color / mirror / Atom feed
From: Christoph Hellwig <hch@infradead.org>
To: Dave Chinner <david@fromorbit.com>
Cc: Christoph Hellwig <hch@infradead.org>,
	"xfs@oss.sgi.com" <xfs@oss.sgi.com>,
	Stefan Priebe - Profihost AG <s.priebe@profihost.ag>
Subject: Re: [xfs-masters] xfs deadlock in stable kernel 3.0.4
Date: Thu, 22 Sep 2011 18:01:51 -0400	[thread overview]
Message-ID: <20110922220151.GA21701@infradead.org> (raw)
In-Reply-To: <20110922214956.GX15688@dastard>

On Fri, Sep 23, 2011 at 07:49:56AM +1000, Dave Chinner wrote:
> On Thu, Sep 22, 2011 at 10:14:57AM -0400, Christoph Hellwig wrote:
> >         By default, a wq guarantees non-reentrance only on the same
> > 	CPU.  A work item may not be executed concurrently on the same
> > 	CPU by multiple workers but is allowed to be executed
> > 	concurrently on multiple CPUs.  This flag makes sure
> > 	non-reentrance is enforced across all CPUs.  Work items queued
> > 	to a non-reentrant wq are guaranteed to be executed by at most
> > 	one worker system-wide at any given time.
> > 
> > So this still seems to preferable for the ail workqueue, and should be
> > able to replace the XFS_AIL_PUSHING_BIT protections.
> 
> No, we can't. WQ_NON_REENTRANT only protects against concurrency on
> the same CPU, not across all CPUs - it still allows concurrent
> per-CPU work processing on the same work item.

Non concurrently for a given work_struct on the same CPU is the default,
WQ_NON_REENTRANT extents that to not beeing exectuted concurrently at
all.  Check the documentation above again, or the code - just look
for the only occurance of WQ_NON_REENTRANT in kernel/workqueue.c and
the surronuding code (e.g. find_worker_executing_work and the
current_work field in struct worker)

> However, we want only a *single* AIL worker instance executing per
> filesystem, not per-cpu per filesystem. Concurrent per-filesystem
> workers will simply bash on the AIL lock trying to walk the AIL at
> the same time, and this is precisely the issue the single AIL worker
> setup is avoiding. The XFS_AIL_PUSHING_BIT is what enforces the
> single per-filesystem push worker running at any time.

I think that's exactly what WQ_NON_REENTRANT is intended for.

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

  reply	other threads:[~2011-09-22 22:01 UTC|newest]

Thread overview: 49+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-09-10 12:23 xfs deadlock in stable kernel 3.0.4 Stefan Priebe
2011-09-12 15:21 ` Christoph Hellwig
2011-09-12 16:46   ` Stefan Priebe
2011-09-12 20:05     ` Christoph Hellwig
2011-09-13  6:04       ` Stefan Priebe - Profihost AG
2011-09-13 19:31         ` Stefan Priebe - Profihost AG
2011-09-13 20:50         ` Christoph Hellwig
2011-09-13 21:52           ` [xfs-masters] " Alex Elder
2011-09-13 21:58             ` Alex Elder
2011-09-13 22:26               ` Christoph Hellwig
2011-09-14  7:26           ` Stefan Priebe - Profihost AG
2011-09-14  7:48             ` Stefan Priebe - Profihost AG
2011-09-14  8:49               ` Stefan Priebe - Profihost AG
2011-09-14 14:30                 ` Christoph Hellwig
2011-09-14 14:30               ` Christoph Hellwig
2011-09-14 16:06                 ` Stefan Priebe - Profihost AG
2011-09-18  9:14                 ` Stefan Priebe - Profihost AG
2011-09-18 20:04                   ` Christoph Hellwig
2011-09-19 10:54                     ` Stefan Priebe - Profihost AG
2011-09-18 23:02                   ` Dave Chinner
2011-09-20  0:47                     ` Stefan Priebe
2011-09-20  1:01                       ` Stefan Priebe
2011-09-20 10:09                     ` Stefan Priebe - Profihost AG
2011-09-20 16:02                       ` Christoph Hellwig
2011-09-20 17:23                         ` Stefan Priebe - Profihost AG
2011-09-20 17:24                           ` Christoph Hellwig
2011-09-20 17:35                             ` Stefan Priebe - Profihost AG
2011-09-20 22:30                               ` Christoph Hellwig
2011-09-21  2:11                                 ` [xfs-masters] " Dave Chinner
2011-09-21  7:40                                   ` Stefan Priebe - Profihost AG
2011-09-21 11:42                                     ` Dave Chinner
2011-09-21 11:55                                       ` Stefan Priebe - Profihost AG
2011-09-21 12:26                                       ` Christoph Hellwig
2011-09-21 13:42                                         ` Stefan Priebe
2011-09-21 16:48                                         ` Stefan Priebe - Profihost AG
2011-09-21 17:26                                           ` Stefan Priebe - Profihost AG
2011-09-21 19:01                                         ` Stefan Priebe - Profihost AG
2011-09-21 23:07                                         ` Dave Chinner
2011-09-22 14:14                                           ` Christoph Hellwig
2011-09-22 21:49                                             ` Dave Chinner
2011-09-22 22:01                                               ` Christoph Hellwig [this message]
2011-09-23  5:28                                                 ` Stefan Priebe - Profihost AG
2011-09-22  0:53                                       ` Dave Chinner
2011-09-22  5:27                                         ` Stefan Priebe - Profihost AG
2011-09-22  7:52                                           ` Stefan Priebe - Profihost AG
2011-09-21  7:36                                 ` Stefan Priebe - Profihost AG
2011-09-21 11:39                                   ` Christoph Hellwig
2011-09-21 13:39                                     ` Stefan Priebe
2011-09-21 14:17                                       ` Christoph Hellwig

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=20110922220151.GA21701@infradead.org \
    --to=hch@infradead.org \
    --cc=david@fromorbit.com \
    --cc=s.priebe@profihost.ag \
    --cc=xfs@oss.sgi.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