From mboxrd@z Thu Jan 1 00:00:00 1970 From: Petr Vandrovec Subject: Re: Resets on sil3124 & sil3726 PMP Date: Thu, 06 Sep 2007 02:52:48 -0700 Message-ID: <46DFCDF0.6050107@vc.cvut.cz> References: <46D2595B.1030409@sauce.co.nz> <46D2792B.401@vc.cvut.cz> <46DBCCE9.9080806@gmail.com> <46DBDA84.3050505@vc.cvut.cz> <46DC030E.9070704@gmail.com> <46DCB709.902@vc.cvut.cz> <46DE9C20.9030007@vc.cvut.cz> <20070905213006.GH22695@curie-int.orbis-terrarum.net> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from mailgw.cvut.cz ([147.32.3.235]:44722 "EHLO mailgw.cvut.cz" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753439AbXIFJwz (ORCPT ); Thu, 6 Sep 2007 05:52:55 -0400 In-Reply-To: <20070905213006.GH22695@curie-int.orbis-terrarum.net> Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: "Robin H. Johnson" Cc: linux-ide@vger.kernel.org, Tejun Heo , Richard Scobie Robin H. Johnson wrote: > On Wed, Sep 05, 2007 at 05:08:00AM -0700, Petr Vandrovec wrote: >>> 3124-2 (norco 4618): >>> NCQ: >>> 1TB alone: 82.30, 82.43 >>> 1TB+1TB: 68.36+68.25 >>> noNCQ: >>> 1TB alone: 82.39, 82.29 >>> 1TB+1TB: 70.33+70.32, 69.47+70.01 >>> Unfortunately that enclosure has only two slots used. I'll try to not >>> forget bring two more disks tomorrow to get it fully populated. >> In case you are still follwing this thread, 3124-2 with 4 1TB drives >> delivers 58.5MBps for each drive with NCQ enabled, and 60MBps with NCQ >> disabled - total 235-240MBps read speed. I would like to see 320MBps, but >> apparently today is not my lucky day... > Just to confirm, how were you testing the multiple drive combination? Concurrent hdparm -t, like hdparm -t /dev/sdd & hdparm -t /dev/sde & hdparm -t /dev/sdf & hdparm -t /dev/sdg & sleep 20 (and from hdparm output & visually confirmed that all activity LEDs go on & off simultaneously) So NCQ should not be able to show any (definitely not huge) benefits, but there should not be any reason to not get 4x 1 drive speed as long as neither PMP link nor PCI-X limits are hit. Petr