All of lore.kernel.org
 help / color / mirror / Atom feed
From: Greg Banks <gnb@sgi.com>
To: raven@themaw.net
Cc: autofs mailing list <autofs@linux.kernel.org>,
	Thomas Moestl <moestl@ibr.cs.tu-bs.de>,
	nfs@lists.sourceforge.net,
	Kernel Mailing List <linux-kernel@vger.kernel.org>
Subject: Re: Re: umount() and NFS races in 2.4.26
Date: Sun, 11 Jul 2004 01:25:49 +1000	[thread overview]
Message-ID: <20040710152549.GD21121@sgi.com> (raw)
In-Reply-To: <Pine.LNX.4.58.0407101419210.1378@donald.themaw.net>

On Sat, Jul 10, 2004 at 02:57:46PM +0800, raven@themaw.net wrote:
> On Thu, 8 Jul 2004, Thomas Moestl wrote:
> > I believe that I have found two problems:
> > 
> > - The NFS async unlink code (fs/nfs/unlink.c) does keep a dentry for
> >   later asynchronous processing, but the mount point is unbusied via
> >   path_release() once sys_unlink() returns (fs/namei.c). [...]

This used to be a bug.  It was fixed in 2.4.26 with

http://linux.bkbits.net:8080/linux-2.4/diffs/fs/nfs/dir.c@1.13

What happens now is that the dentry and its inode are cleaned up
when the async unlink task is deleted in nfs_put_super() between
the first and second calls to invalidate_inodes() in kill_super().

> > - There is a SMP race between the shrink_dcache_parent() (fs/dcache.c)
> >   called from kill_super() and prune_dache() called via
> >   shrink_dache_memory() (called by kswapd), as follows: [...]

Your scenario sounds plausible and might explain at least some 
of the autofs unmount races we've been seeing.

> >   In the attached patch, I have used a semaphore to serialize purging
> >   accesses to the dentry_unused list. [...]

Can we see the patch please?

Greg.
-- 
Greg Banks, R&D Software Engineer, SGI Australian Software Group.
I don't speak for SGI.

WARNING: multiple messages have this Message-ID (diff)
From: Greg Banks <gnb@sgi.com>
To: raven@themaw.net
Cc: Thomas Moestl <moestl@ibr.cs.tu-bs.de>,
	autofs mailing list <autofs@linux.kernel.org>,
	nfs@lists.sourceforge.net,
	Kernel Mailing List <linux-kernel@vger.kernel.org>
Subject: Re: [autofs] Re: umount() and NFS races in 2.4.26
Date: Sun, 11 Jul 2004 01:25:49 +1000	[thread overview]
Message-ID: <20040710152549.GD21121@sgi.com> (raw)
In-Reply-To: <Pine.LNX.4.58.0407101419210.1378@donald.themaw.net>

On Sat, Jul 10, 2004 at 02:57:46PM +0800, raven@themaw.net wrote:
> On Thu, 8 Jul 2004, Thomas Moestl wrote:
> > I believe that I have found two problems:
> > 
> > - The NFS async unlink code (fs/nfs/unlink.c) does keep a dentry for
> >   later asynchronous processing, but the mount point is unbusied via
> >   path_release() once sys_unlink() returns (fs/namei.c). [...]

This used to be a bug.  It was fixed in 2.4.26 with

http://linux.bkbits.net:8080/linux-2.4/diffs/fs/nfs/dir.c@1.13

What happens now is that the dentry and its inode are cleaned up
when the async unlink task is deleted in nfs_put_super() between
the first and second calls to invalidate_inodes() in kill_super().

> > - There is a SMP race between the shrink_dcache_parent() (fs/dcache.c)
> >   called from kill_super() and prune_dache() called via
> >   shrink_dache_memory() (called by kswapd), as follows: [...]

Your scenario sounds plausible and might explain at least some 
of the autofs unmount races we've been seeing.

> >   In the attached patch, I have used a semaphore to serialize purging
> >   accesses to the dentry_unused list. [...]

Can we see the patch please?

Greg.
-- 
Greg Banks, R&D Software Engineer, SGI Australian Software Group.
I don't speak for SGI.

  reply	other threads:[~2004-07-10 15:25 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-07-08 18:07 umount() and NFS races in 2.4.26 Thomas Moestl
2004-07-09 14:32 ` Marcelo Tosatti
2004-07-10 21:38   ` Trond Myklebust
2004-07-11  0:32     ` viro
2004-07-10  6:57 ` raven
2004-07-10 15:25   ` Greg Banks [this message]
2004-07-10 15:25     ` [autofs] " Greg Banks
2004-07-10 18:25     ` Thomas Moestl
2004-07-10 18:19   ` Thomas Moestl
2004-07-10 19:25     ` raven
2004-07-10 19:25       ` raven
2004-07-10 19:50       ` Thomas Moestl
2004-07-11 10:16     ` raven
2004-07-11 10:16       ` raven

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=20040710152549.GD21121@sgi.com \
    --to=gnb@sgi.com \
    --cc=autofs@linux.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=moestl@ibr.cs.tu-bs.de \
    --cc=nfs@lists.sourceforge.net \
    --cc=raven@themaw.net \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.