From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1758276AbXGDKqN (ORCPT ); Wed, 4 Jul 2007 06:46:13 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1757059AbXGDKp4 (ORCPT ); Wed, 4 Jul 2007 06:45:56 -0400 Received: from cantor2.suse.de ([195.135.220.15]:60469 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754330AbXGDKpz (ORCPT ); Wed, 4 Jul 2007 06:45:55 -0400 To: Dan Williams Cc: neilb@suse.de, raziebe@gmail.com, akpm@linux-foundation.org, davidsen@tmr.com, linux-kernel@vger.kernel.org, linux-raid@vger.kernel.org Subject: Re: [RFC PATCH 0/2] raid5: 65% sequential-write performance improvement, stripe-queue take2 References: <20070703231838.17500.34505.stgit@dwillia2-linux.ch.intel.com> From: Andi Kleen Date: 04 Jul 2007 13:41:26 +0200 In-Reply-To: <20070703231838.17500.34505.stgit@dwillia2-linux.ch.intel.com> Message-ID: User-Agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.3 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Dan Williams writes: > The write performance numbers are better than I expected and would seem > to address the concerns raised in the thread "Odd (slow) RAID > performance"[2]. The read performance drop was not expected. However, > the numbers suggest some additional changes to be made to the queuing > model. Have you considered supporting copy-xor in MD for non accelerated RAID? I've been looking at fixing the old dubious slow crufty x86 SSE XOR functions. One thing I discovered is that it seems fairly pointless to make them slower with cache avoidance when most of the data is copied before anyways. I think much more advantage could be gotten by supporting copy-xor because XORing during a copy should be nearly free. On the other hand ext3 write() also uses a cache avoiding copy now and for the XOR it would need to load the data from memory again. Perhaps this could be also optimized somehow (e.g. setting a flag somewhere and using a normal copy for the RAID-5 case) -Andi