From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1760881AbZEZXng (ORCPT ); Tue, 26 May 2009 19:43:36 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1758802AbZEZXn1 (ORCPT ); Tue, 26 May 2009 19:43:27 -0400 Received: from smtp1.linux-foundation.org ([140.211.169.13]:34173 "EHLO smtp1.linux-foundation.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1758850AbZEZXn0 (ORCPT ); Tue, 26 May 2009 19:43:26 -0400 Date: Tue, 26 May 2009 16:42:52 -0700 From: Andrew Morton To: Wu Fengguang Cc: hifumi.hisashi@oss.ntt.co.jp, linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org, kosaki.motohiro@jp.fujitsu.com, linux-mm@kvack.org, jens.axboe@oracle.com Subject: Re: [PATCH] readahead:add blk_run_backing_dev Message-Id: <20090526164252.0741b392.akpm@linux-foundation.org> In-Reply-To: <20090522023323.GA10864@localhost> References: <6.0.0.20.2.20090518183752.0581fdc0@172.19.0.2> <20090518175259.GL4140@kernel.dk> <20090520025123.GB8186@localhost> <6.0.0.20.2.20090521145005.06f81fe0@172.19.0.2> <20090522010538.GB6010@localhost> <6.0.0.20.2.20090522102551.0705aea0@172.19.0.2> <20090522023323.GA10864@localhost> X-Mailer: Sylpheed version 2.2.4 (GTK+ 2.8.20; i486-pc-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Fri, 22 May 2009 10:33:23 +0800 Wu Fengguang wrote: > > I tested above patch, and I got same performance number. > > I wonder why if (PageUptodate(page)) check is there... > > Thanks! This is an interesting micro timing behavior that > demands some research work. The above check is to confirm if it's > the PageUptodate() case that makes the difference. So why that case > happens so frequently so as to impact the performance? Will it also > happen in NFS? > > The problem is readahead IO pipeline is not running smoothly, which is > undesirable and not well understood for now. The patch causes a remarkably large performance increase. A 9% reduction in time for a linear read? I'd be surprised if the workload even consumed 9% of a CPU, so where on earth has the kernel gone to? Have you been able to reproduce this in your testing? Thanks.