Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S266027AbVBFEZs (ORCPT ); Sat, 5 Feb 2005 23:25:48 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S272721AbVBFEZs (ORCPT ); Sat, 5 Feb 2005 23:25:48 -0500 Received: from bl5-232-195.dsl.telepac.pt ([82.154.232.195]:42639 "EHLO puma-vgertech.no-ip.com") by vger.kernel.org with ESMTP id S266027AbVBFEZg (ORCPT ); Sat, 5 Feb 2005 23:25:36 -0500 Message-ID: <42059C46.1080406@vgertech.com> Date: Sun, 06 Feb 2005 04:25:42 +0000 From: Nuno Silva User-Agent: Debian Thunderbird 1.0 (X11/20050116) X-Accept-Language: en-us, en MIME-Version: 1.0 To: Andrew Morton Cc: Lincoln Dale , Ian.Godin@lowrydigital.com, linux-kernel@vger.kernel.org Subject: Re: Drive performance bottleneck References: <5.1.0.14.2.20050205094038.05323240@171.71.163.14> <20050204150728.6a697e0e.akpm@osdl.org> In-Reply-To: <20050204150728.6a697e0e.akpm@osdl.org> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1273 Lines: 36 Andrew Morton wrote: > Lincoln Dale wrote: > >>sg_dd uses a window into a kernel DMA window. as such, two of the four >>memory acccesses are cut out (1. DMA from HBA to RAM, 2. userspace >>accessing data). >>1.6Gbps / 2 = 800MB/s -- or roughly what Ian was seeing with sg_dd. > > > Right. That's a fancy way of saying "cheating" ;) > > But from the oprofile output it appears to me that there is plenty of CPU > capacity left over. Maybe I'm misreading it due to oprofile adding in the > SMP factor (25% CPU on a 4-way means we've exhausted CPU capacity). sg_dd is lying or /dev/sg* is broken. Try to do that sg_dd test in any single drive and you'll get 20 times the performance you're supposed to achieve: puma:/tmp/dd# time sg_dd if=/dev/sg1 of=/dev/null bs=64k count=1000000 time=1 Reducing read to 64 blocks per loop time to transfer data was 69.784784 secs, 939.12 MB/sec 1000000+0 records in 1000000+0 records out This is a single sata drive. I'm lucky, am I not? ;-) Regards, Nuno Silva - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/