From: Christoph Hellwig Subject: Re: [PATCH] fsync: wait for data writeout completion before calling ->fsync Date: Thu, 3 Sep 2009 17:39:26 +0200 Message-ID: <20090903153926.GB19337@lst.de> References: <1251899966-7316-1-git-send-email-jack@suse.cz> <20090902221838.GA5311@lst.de> <20090903104704.GD5060@duck.novell.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Christoph Hellwig , linux-fsdevel@vger.kernel.org, LKML To: Jan Kara Return-path: Content-Disposition: inline In-Reply-To: <20090903104704.GD5060@duck.novell.com> Sender: linux-kernel-owner@vger.kernel.org List-Id: linux-fsdevel.vger.kernel.org On Thu, Sep 03, 2009 at 12:47:04PM +0200, Jan Kara wrote: > On Thu 03-09-09 00:18:38, Christoph Hellwig wrote: > > I think we should add this one ontop: > Agreed. Added to the series. I'll now push the whole series to linux-next > via my fs tree. Thanks. We do also need something similar in writeback_single_inode, but I'm not sure how to do that in a good way, and how fixing this now would interact with Jens' writeback changes. Probably makes sense to postpone it until 2.6.32 at least and possibly try to split data and metadata writeback fully.