From: Nick Piggin Subject: Re: [PATCH RFC] ext3 data=guarded v3 Date: Fri, 17 Apr 2009 04:09:01 +1000 Message-ID: <200904170409.02222.nickpiggin@yahoo.com.au> References: <1239816159-6868-1-git-send-email-chris.mason@oracle.com> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Cc: Chris Mason , Jan Kara , "Theodore Ts'o" , Linux Kernel Developers List , Ext4 Developers List To: Linus Torvalds , linux-fsdevel@vger.kernel.org Return-path: Received: from smtp115.mail.mud.yahoo.com ([209.191.84.164]:43521 "HELO smtp115.mail.mud.yahoo.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with SMTP id S1755559AbZDPSJJ (ORCPT ); Thu, 16 Apr 2009 14:09:09 -0400 In-Reply-To: Content-Disposition: inline Sender: linux-ext4-owner@vger.kernel.org List-ID: On Thursday 16 April 2009 06:35:23 Linus Torvalds wrote: > > On Wed, 15 Apr 2009, Chris Mason wrote: > > > > Here is another version of the data=guarded work for ext3. The main > > difference between this code and yesterday's is the guarded writepage > > function now sends any newly allocated block through the old data=ordered code. > > I'm inclined to apply the first two patches as infrastructure, since they > seem to make sense regardless of data=ordered. The ability to get a > callback when IO ends sounds like something that a number of cases might > find intriguing, and it's obviously how the actual IO has worked > internally anyway. > > Comments? I think also fscache really should be using proper read completed notifications rather than hijacking the page lock waitqueue for this.