From owner-freebsd-scsi Fri Dec 17 3:40:41 1999 Delivered-To: freebsd-scsi@freebsd.org Received: from fuchs.omnilink.de (hase.omnilink.de [194.64.25.65]) by hub.freebsd.org (Postfix) with ESMTP id C199C14C41 for ; Fri, 17 Dec 1999 03:40:33 -0800 (PST) (envelope-from ob@omnilink.net) Received: by fuchs.omnilink.de; id MAA21657; Fri, 17 Dec 1999 12:38:26 +0100 (MET) Received: from sondermuell.omnilink.de(194.64.25.218) by fuchs.omnilink.de via smap (V4.2) id xma021653; Fri, 17 Dec 99 12:37:52 +0100 Message-ID: <008101bf4883$e275cd70$da1940c2@omnilink.de> Reply-To: "Oliver Blasnik" From: "Oliver Blasnik" To: Subject: Results of CRD5440-Checks Date: Fri, 17 Dec 1999 12:43:06 +0100 Organization: Omnilink ISC MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.00.2314.1300 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300 Sender: owner-freebsd-scsi@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hi there, here are the results of my bugtracking of that weird = tagged-queueing-problem (if anyone want's to know it). It took about one = day to come up with this. Environment:=20 System 1: P133, ASUS-MB (some oooold) System 2: PII/366, ASUS P2B-S, Scsi-Bios 2.11 (latest) System 3: dual PII/450, ASUS P2B-DS, Scsi-Bios 2.11 (latest) Two CRD5440, different Manufacturing Date, 2 logical Partitions Adaptec 2940UW, latest Bios (1.34.x) =20 Results: The problems do occure on BOTH scsi-controllers (onboard 7890 and 2940) = in single-host AND dual-host-mode of the CRD. They do NOT occure on = System 1, but System 2 and 3 (the faster machines). There is no difference between FreeBSD 3.1R, 3.3R and 3.4RC. After changing cam_xpt.c to disable TQ (or disable TQ directly on the = CRD's), the systems were stable again. There was no real difference = between setting a fixed value of 4 tqs/disk or dynamic-mode (fixed did a = little bit longer...). The problem ONLY comes up after accessing both logical CRD-partitions = (da0 and da1 in FreeBSD) in a high-load-state (40 dd's for one, 40 dd's = for the other drive to fill up the command cache). If just accessing one = partition, the system worked well. So, resulting fact is -> this controller does NOT work correctly, if = tagged queuing enabled with more than one active logical partition. I = switched off TQ at controller-level to let the kernel unchanged. = Possibly this information should go to some type of hcl. I will post a full-featuerd report to crd-tech and hope to get a result. Cu, Oliver --=20 __ OMNILINK Internet Service Center GmbH / \ Hahnstrasse 70, 60528 Frankfurt __\ /_________ Tel.: (0 69)66 44 10 Fax: (0 69)66 44 11 99 O M N I L I N K http://www.omnilink.net To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message