Message-ID: <48DC3A4E.7010602@sgi.com>
Date: Fri, 26 Sep 2008 11:26:38 +1000
From: Lachlan McIlroy <lachlan@sgi.com>
Reply-To: lachlan@sgi.com
User-Agent: Thunderbird 2.0.0.16 (X11/20080707)
MIME-Version: 1.0
To: Peter Leckie <pleckie@sgi.com>
CC: xfs@oss.sgi.com, xfs-dev@sgi.com
Subject: Re: [PATCH v2] Use atomic_t and wait_event to track dquot pincount
References: <48D9C1DD.6030607@sgi.com> <48D9EB8F.1070104@sgi.com> <48D9EF6E.8010505@sgi.com> <20080924074604.GK5448@disturbed> <48D9F718.4010905@sgi.com> <20080925010318.GB27997@disturbed> <48DB4F3F.8040307@sgi.com> <20080926003401.GG27997@disturbed> <48DC3638.3050601@sgi.com>
In-Reply-To: <48DC3638.3050601@sgi.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit

Peter Leckie wrote:
> Dave Chinner wrote:
>> but it doesn't fix the underlying problem that was causing the
>> spurious wakeups, which is the fact that xfs_qm_dqflush() is not
>> obeying non-blocking flush directions. The patch below should fix
>> that. Can you please test it before you add your patch?
>>   
> Yeah I already had this idea I  just have not posted a patch because 
> Lachlan though
> it might introduce a deadlock.
I suggested some changes a while back to make tail pushing non-blocking
and Dave thought it might cause a deadlock.

http://oss.sgi.com/archives/xfs/2008-07/msg00472.html

I actually did hit a deadlock with this change but could not figure out
why.  It may have been the same issue Pete is trying to fix here.

> If you think this is a good Idea I will 
> update my patch
> to  be non blocking.
>