From mboxrd@z Thu Jan 1 00:00:00 1970 From: James Bottomley Subject: Re: Possible performance regression? Date: Mon, 02 Jan 2006 10:19:06 -0600 Message-ID: <1136218746.3531.31.camel@mulgrave> References: <43B94DB9.5040505@hp.com> Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Return-path: Received: from stat9.steeleye.com ([209.192.50.41]:44188 "EHLO hancock.sc.steeleye.com") by vger.kernel.org with ESMTP id S1750799AbWABQTN (ORCPT ); Mon, 2 Jan 2006 11:19:13 -0500 In-Reply-To: <43B94DB9.5040505@hp.com> Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: "Alan D. Brunelle" Cc: linux-scsi@vger.kernel.org On Mon, 2006-01-02 at 10:58 -0500, Alan D. Brunelle wrote: > As part of HP's Open Source and Linux Organization's Performance and > Scalability Group, I've noticed what looks to be a regression in U320 > SCSI performance coming down the pike. Background: These measurements > were performed on Red Hat RHEL4 update 2 (2.6.9-based) and a "generic" > 2.6.14-based kernel; I used an HP RX4640 with 4 dual-U320 LSI Logic / > Symbios Logic 53c1030 PCI-X Fusion-MPT Dual Ultra320 SCSI (rev 08) > adapters, each with two buses containing 5 72GB/15k U320 drives (for a > total of 40 drives). The runs were done in single-user mode (to minimize > effects from other tasks), and report results in megabytes per second. I > ran tests for various block sized IOs (1KB, 2KB, 4, ...256KB per IO). This data doesn't seem to be mirrored by tiobench tests on other drivers, so it's probably fusion specific. When a similar problem was found in the adaptec driver a while back, it was due to a bug which set the TCQ depth to 1, which looks very consistent with your results, so you might try starting there. James