From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from relay.sgi.com (relay3.corp.sgi.com [198.149.34.15]) by oss.sgi.com (Postfix) with ESMTP id 9D3887F51 for ; Tue, 14 Apr 2015 02:27:02 -0500 (CDT) Received: from cuda.sgi.com (cuda2.sgi.com [192.48.176.25]) by relay3.corp.sgi.com (Postfix) with ESMTP id 38D19AC003 for ; Tue, 14 Apr 2015 00:26:58 -0700 (PDT) Received: from ipmail06.adl6.internode.on.net (ipmail06.adl6.internode.on.net [150.101.137.145]) by cuda.sgi.com with ESMTP id KmIASRdMdWt5IGHU for ; Tue, 14 Apr 2015 00:26:55 -0700 (PDT) Received: from disappointment.disaster.area ([192.168.1.110] helo=disappointment) by dastard with esmtp (Exim 4.80) (envelope-from ) id 1YhvFE-0000qY-S2 for xfs@oss.sgi.com; Tue, 14 Apr 2015 17:26:52 +1000 Received: from dave by disappointment with local (Exim 4.82_1-5b7a7c0-XX) (envelope-from ) id 1YhvFE-0000VR-R4 for xfs@oss.sgi.com; Tue, 14 Apr 2015 17:26:52 +1000 From: Dave Chinner Subject: [PATCH 0/8 v2] xfs: fix direct IO completion issues Date: Tue, 14 Apr 2015 17:26:43 +1000 Message-Id: <1428996411-1507-1-git-send-email-david@fromorbit.com> List-Id: XFS Filesystem from SGI List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Errors-To: xfs-bounces@oss.sgi.com Sender: xfs-bounces@oss.sgi.com To: xfs@oss.sgi.com Hi folks, This is the second version of the series I first posted here: http://oss.sgi.com/archives/xfs/2015-04/msg00097.html I've reworked and refactored the first two patches in the series, breaking them out into simpler, more self contained patches. The main change from the previous version is that I dropped the pre-allocation of the append transaction to avoid the nesting problems it could cause. The only other major change is that the ioend is only allocated when a mapping is either for an unwritten region or spans EOF. hence if the mapping is an overwrite within EOF we do not allocate an ioend, we do not defer completion to the DIO workqueue and we do nothing in the completion function as there is nothing to do. Comments, thoughts? -Dave. _______________________________________________ xfs mailing list xfs@oss.sgi.com http://oss.sgi.com/mailman/listinfo/xfs