From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Chinner Subject: Re: [patch 10/21] buffer heads: Support slab defrag Date: Wed, 21 May 2008 07:46:17 +1000 Message-ID: <20080520214617.GU103491721@sgi.com> References: <20080510030831.796641881@sgi.com> <20080510030916.935905242@sgi.com> <20080512002403.GP103491721@sgi.com> <20080515231045.GY155679365@sgi.com> <20080519054554.GY103491721@sgi.com> <20080520002503.GC173056135@sgi.com> <20080520065622.GA13968@2ka.mipt.ru> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: David Chinner , Christoph Lameter , akpm@linux-foundation.org, linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org, Mel Gorman , andi@firstfloor.org, Rik van Riel , Pekka Enberg , mpm@selenic.com To: Evgeniy Polyakov Return-path: Received: from relay1.sgi.com ([192.48.171.29]:42844 "EHLO relay.sgi.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1755858AbYETVqm (ORCPT ); Tue, 20 May 2008 17:46:42 -0400 Content-Disposition: inline In-Reply-To: <20080520065622.GA13968@2ka.mipt.ru> Sender: linux-fsdevel-owner@vger.kernel.org List-ID: On Tue, May 20, 2008 at 10:56:23AM +0400, Evgeniy Polyakov wrote: > On Tue, May 20, 2008 at 10:25:03AM +1000, David Chinner (dgc@sgi.com) wrote: > > + filemap_fdatawrite_range(mapping, start, end); > > > > Where [start,end] span page->index and are is large enough > > to get a substantial sized I/O to disk (say at least SWAP_CLUSTER_MAX > > pages, preferrably larger for 4k page size machines). > > Or just sync_inode(). Oh, god no. Let's not put the inode_lock right at the top of the VM page cleaning path. We don't need to modify inode state, the superblock dirty lists, etc - all we need to do is write dirty pages on a given mapping in a more efficient manner. Cheers, Dave. -- Dave Chinner Principal Engineer SGI Australian Software Group