From owner-freebsd-bugs Fri Mar 3 11:40: 7 2000 Delivered-To: freebsd-bugs@freebsd.org Received: from freefall.freebsd.org (freefall.FreeBSD.ORG [204.216.27.21]) by hub.freebsd.org (Postfix) with ESMTP id 9083E37B58A for ; Fri, 3 Mar 2000 11:40:01 -0800 (PST) (envelope-from gnats@FreeBSD.org) Received: (from gnats@localhost) by freefall.freebsd.org (8.9.3/8.9.2) id LAA52282; Fri, 3 Mar 2000 11:40:01 -0800 (PST) (envelope-from gnats@FreeBSD.org) Received: from feral.com (feral.com [192.67.166.1]) by hub.freebsd.org (Postfix) with ESMTP id 06E0737B6EB for ; Fri, 3 Mar 2000 11:30:48 -0800 (PST) (envelope-from mjacob@lorq.feral.com) Received: from lorq.feral.com (lorq [192.67.166.93]) by feral.com (8.9.3/8.9.3) with ESMTP id LAA32465 for ; Fri, 3 Mar 2000 11:30:46 -0800 Received: (from mjacob@localhost) by lorq.feral.com (8.9.3/8.9.3) id LAA01066; Fri, 3 Mar 2000 11:30:46 -0800 (PST) (envelope-from mjacob) Message-Id: <200003031930.LAA01066@lorq.feral.com> Date: Fri, 3 Mar 2000 11:30:46 -0800 (PST) From: Matthew Jacob Reply-To: mjacob@freebsd.org To: FreeBSD-gnats-submit@freebsd.org X-Send-Pr-Version: 3.2 Subject: kern/17153: Qlogic Ultra2/Ultra3 cards seem to write very slow to IBM disks Sender: owner-freebsd-bugs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org >Number: 17153 >Category: kern >Synopsis: Qlogic Ultra3 cards seem to write very slow to IBM disks >Confidential: no >Severity: serious >Priority: medium >Responsible: freebsd-bugs >State: open >Quarter: >Keywords: >Date-Required: >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Fri Mar 3 11:40:01 PST 2000 >Closed-Date: >Last-Modified: >Originator: Matthew Jacob >Release: FreeBSD 4.0-CURRENT i386 >Organization: Feral Software >Environment: -current as of today. I have a Qlogic 1280 dual Ultra2 controller and a Qlogic 12160 dual Ultra3 card- this behaviour is the same for both - and have 3 Ultra3 drives (shown in their Ultra2 invocations): da8 at isp1 bus 0 target 4 lun 0 da8: Fixed Direct Access SCSI-2 device da8: 80.000MB/s transfers (40.000MHz, offset 8, 16bit), Tagged Queueing Enabled da8: 8683MB (17783249 512 byte sectors: 255H 63S/T 1106C) da9 at isp1 bus 1 target 0 lun 0 da9: Fixed Direct Access SCSI-3 device da9: 80.000MB/s transfers (40.000MHz, offset 8, 16bit), Tagged Queueing Enabled da9: 17501MB (35843670 512 byte sectors: 255H 63S/T 2231C) da10 at isp1 bus 1 target 1 lun 0 da10: Fixed Direct Access SCSI-3 device da10: 80.000MB/s transfers (40.000MHz, offset 8, 16bit), Tagged Queueing Enabled da10: 17501MB (35843670 512 byte sectors: 255H 63S/T 2231C) Tyan 200Mhz Pentium. >Description: The read data rate is quite respectable for these drives: lorq.feral.com > root lmdd if=/dev/rda8 bs=1024k count=50 50.00 MB in 4.18 seconds (11.9530 MB/sec) lorq.feral.com > root lmdd if=/dev/rda9 bs=1024k count=50 50.00 MB in 1.90 seconds (26.3304 MB/sec) The write data rate for the Quantum okay: lorq.feral.com > root lmdd of=/dev/rda8 bs=1024k count=50 50.00 MB in 3.88 seconds (12.8946 MB/sec) The write data rate for the IBM totally sucks: lorq.feral.com > root lmdd of=/dev/rda9 bs=1024k count=50 50.00 MB in 6.80 seconds (7.3521 MB/sec) Write Cache Enable is not on for the quantum. Setting it for the IBM doesn't seem to make a difference. [ I'm filing this bug for myself to remember to look into in a couple of weeks ] >How-To-Repeat: See above. >Fix: >Release-Note: >Audit-Trail: >Unformatted: To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-bugs" in the body of the message