From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o75BWxcV084794 for ; Thu, 5 Aug 2010 06:33:00 -0500 Received: from rcsinet10.oracle.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id DE77C1E3F7D1 for ; Thu, 5 Aug 2010 04:33:20 -0700 (PDT) Received: from rcsinet10.oracle.com (rcsinet10.oracle.com [148.87.113.121]) by cuda.sgi.com with ESMTP id a4bcAVWLXxEC0geW for ; Thu, 05 Aug 2010 04:33:20 -0700 (PDT) Date: Thu, 5 Aug 2010 07:32:40 -0400 From: Chris Mason Subject: Re: direct-io regression [Was: How to track down abysmal performance ata - raid1 - crypto - vg/lv - xfs] Message-ID: <20100805113240.GA29846@think> References: <20100804073546.GA7494@comet.dominikbrodowski.net> <20100804085039.GA11671@infradead.org> <20100804091317.GA27779@isilmar-3.linta.de> <20100804092122.GA2998@infradead.org> <20100804073546.GA7494@comet.dominikbrodowski.net> <201008041116.09822@zmi.at> <20100804102526.GB13766@isilmar-3.linta.de> <15446.1280953986@localhost> <20100805093100.GA3001@comet.dominikbrodowski.net> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: <20100805093100.GA3001@comet.dominikbrodowski.net> List-Id: XFS Filesystem from SGI List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: xfs-bounces@oss.sgi.com Errors-To: xfs-bounces@oss.sgi.com To: Dominik Brodowski , Valdis.Kletnieks@vt.edu, josef@redhat.com, Michael Monnerie , Christoph Hellwig , linux-raid@vger.kernel.org, xfs@oss.sgi.com, linux-kernel@vger.kernel.org, dm-devel@redhat.com On Thu, Aug 05, 2010 at 11:31:00AM +0200, Dominik Brodowski wrote: > Hey, > > when attempting to track down insufficient I/O performance, I found the > following reression relating to direct-io on my notebook, where an > ata device, which consists of several partitions, is combined to a lvm > volume, and one logical volume is then encrypted using dm-crypt. Test case > was the following command: > > $ dd if=/dev/mapper/vg0-root_crypt of=/dev/zero iflag=direct bs=8k count=131072 > > 2.6.34 results in ~16 MB/s, > 2.6.35 results in ~ 3.1 MB/s > > The regression was bisected down to the follwoing commit: > > commit c2c6ca417e2db7a519e6e92c82f4a933d940d076 > Author: Josef Bacik > Date: Sun May 23 11:00:55 2010 -0400 > > direct-io: do not merge logically non-contiguous requests > > ... > > How to fix this? I do not use btrfs, but ext3 (and the access was down on > the block level, not on the fs level, so this btrs-related commit should not > cause such a regression). Well, you've already bisected down to an offending if statement, that's a huge help. I'll try to reproduce this and fix it up today. But, I'm surprised your drive is doing 8K dio reads at 16MB/s, that seems a little high. -chris _______________________________________________ xfs mailing list xfs@oss.sgi.com http://oss.sgi.com/mailman/listinfo/xfs