From: Chris Mason Subject: Re: [PATCH 0/3] Ext3 latency improvement patches Date: Mon, 30 Mar 2009 07:44:22 -0400 Message-ID: <1238413462.30488.0.camel@think.oraclecorp.com> References: <1238185471-31152-1-git-send-email-tytso@mit.edu> <1238187031.27455.212.camel@think.oraclecorp.com> <1238187818.27455.217.camel@think.oraclecorp.com> <20090327213052.GC5176@mit.edu> <20090330112330.GA11357@skywalker> Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Cc: Theodore Tso , Ric Wheeler , Linux Kernel Developers List , Ext4 Developers List , jack@suse.cz To: "Aneesh Kumar K.V" Return-path: Received: from acsinet12.oracle.com ([141.146.126.234]:49248 "EHLO acsinet12.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750754AbZC3Low (ORCPT ); Mon, 30 Mar 2009 07:44:52 -0400 In-Reply-To: <20090330112330.GA11357@skywalker> Sender: linux-ext4-owner@vger.kernel.org List-ID: On Mon, 2009-03-30 at 16:53 +0530, Aneesh Kumar K.V wrote: > On Fri, Mar 27, 2009 at 05:30:52PM -0400, Theodore Tso wrote: > > On Fri, Mar 27, 2009 at 05:03:38PM -0400, Chris Mason wrote: > > > > Ric had asked me about a test program that would show the worst case > > > > ext3 behavior. So I've modified your ext3 program a little. It now > > > > creates a 8G file and forks off another proc to do random IO to that > > > > file. > > > > > > > > > > My understanding of ext4 delalloc is that once blocks are allocated to > > > file, we go back to data=ordered. > > > > Yes, that's correct. > > > > > Ext4 is going pretty slowly for this fsync test (slower than ext3), it > > > looks like we're going for a very long time in > > > jbd2_journal_commit_transaction -> write_cache_pages. > > > > One of the things that we can do to optimize this case for ext4 (and > > ext3) is that if block has already been written out to disk once, we > > don't have to flush it to disk a second time. So if we add a new > > buffer_head flag which can distinguish between blocks that have been > > newly allocated (and not yet been flushed to disk) versus blocks that > > have already been flushed to disk at least once, we wouldn't need to > > force I/O for blocks in the latter case. > > write_cache_pages should only look at pages which are marked dirty right > ?. So why are we writing these pages again and again ? The test program is constantly creating new dirty pages to random offsets on the disk ;) -chris