From: Jeff Moyer Subject: Re: [PATCH 0/4 v3] ext3/4: enhance fsync performance when using CFQ Date: Thu, 15 Apr 2010 09:13:51 -0400 Message-ID: References: <1271279826-30294-1-git-send-email-jmoyer@redhat.com> <20100415103345.GY27497@kernel.dk> <20100415130854.GA27497@kernel.dk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: linux-kernel@vger.kernel.org, linux-ext4@vger.kernel.org, vgoyal@redhat.com To: Jens Axboe Return-path: Received: from mx1.redhat.com ([209.132.183.28]:29074 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753486Ab0DONN5 (ORCPT ); Thu, 15 Apr 2010 09:13:57 -0400 In-Reply-To: <20100415130854.GA27497@kernel.dk> (Jens Axboe's message of "Thu, 15 Apr 2010 15:08:55 +0200") Sender: linux-ext4-owner@vger.kernel.org List-ID: Jens Axboe writes: >> > This is starting to look better. Can you share what tests you did? I >> > tried reproducing with fs_mark last time and could not. >> >> Did you use the fs_mark command line I (think I) had posted? What >> storage were you using? > > No, I didn't see any references to example command lines. I tested on a > few single disks, rotating and SSD. I expected the single spinning disk > to show the problem to some extent at least, but there was no difference > observed with 64kb blocks. Boy, I'm really slipping. Try this one: ./fs_mark -S 1 -D 100 -N 1000 -d /mnt/test/fs_mark -s 65536 -t 1 -w 4096 Cheers, Jeff