From mboxrd@z Thu Jan  1 00:00:00 1970
From: Dave Chinner <david@fromorbit.com>
Subject: Re: [RESEND PATCH] writeback: Judge bdi->dev when set worker desc in
  bdi_writeback_workfn.
Date: Thu, 26 Sep 2013 08:43:51 +1000
Message-ID: <20130925224351.GJ26872@dastard>
References: <201309100811409182210@gmail.com>
 <2013092516084285960311@gmail.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: tj <tj@kernel.org>, axboe <axboe@kernel.dk>,
	linux-fsdevel <linux-fsdevel@vger.kernel.org>
To: majianpeng <majianpeng@gmail.com>
Return-path: <linux-fsdevel-owner@vger.kernel.org>
Received: from ipmail07.adl2.internode.on.net ([150.101.137.131]:41463 "EHLO
	ipmail07.adl2.internode.on.net" rhost-flags-OK-OK-OK-OK)
	by vger.kernel.org with ESMTP id S1753653Ab3IYWn5 (ORCPT
	<rfc822;linux-fsdevel@vger.kernel.org>);
	Wed, 25 Sep 2013 18:43:57 -0400
Content-Disposition: inline
In-Reply-To: <2013092516084285960311@gmail.com>
Sender: linux-fsdevel-owner@vger.kernel.org
List-ID: <linux-fsdevel.vger.kernel.org>

On Wed, Sep 25, 2013 at 04:08:54PM +0800, majianpeng wrote:
> Hi,
> 	How about this patch?
> This bug can easily to reproduce.
> dd if=/dev/zero of=/dev/sdb bs=64k
> For a while, remove the disk. At my machine, it at most 100% occured.

...
> >
> >This bug introduced by commit ef3b101925f2170c.
> >I think the situation is like:
> >remove disk                     flush_thread_work
> >bdi_destroy()
> > bdi_unregister()
> >    bdi->dev = NULL
> >                                bdi_writeback_workfn()

You're just papering over the larger problem, in that the writeback
work is running concurrently with the bdi_unregister() function that
is tearing the bdi down. You should try to fix the underlying race
condition, as documented in bdi_destroy.

Cheers,

Dave.
-- 
Dave Chinner
david@fromorbit.com