From owner-freebsd-hardware@FreeBSD.ORG Wed Jul 2 09:10:01 2003 Return-Path: Delivered-To: freebsd-hardware@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id D18CC37B401 for ; Wed, 2 Jul 2003 09:10:01 -0700 (PDT) Received: from smtp0.adl1.internode.on.net (smtp0.adl1.internode.on.net [203.16.214.194]) by mx1.FreeBSD.org (Postfix) with ESMTP id 900A343F75 for ; Wed, 2 Jul 2003 09:10:00 -0700 (PDT) (envelope-from smckay@internode.on.net) Received: from dungeon.home (ppp188-212.lns1.bne1.internode.on.net [150.101.188.212])h62G9wUl064372; Thu, 3 Jul 2003 01:39:58 +0930 (CST) Received: from dungeon.home (localhost [127.0.0.1]) by dungeon.home (8.12.8p1/8.11.6) with ESMTP id h62G9vqY012229; Thu, 3 Jul 2003 02:09:58 +1000 (EST) (envelope-from mckay) Message-Id: <200307021609.h62G9vqY012229@dungeon.home> To: freebsd-hardware@freebsd.org References: <200306171554.h5HFs2DQ041575@mail.synology.com> <200306181052.h5IAqTu2008960@dungeon.home> <200306231412.h5NECn7K006239@dungeon.home> In-Reply-To: <200306231412.h5NECn7K006239@dungeon.home> from Stephen McKay at "Tue, 24 Jun 2003 00:12:49 +1000" Date: Thu, 03 Jul 2003 02:09:57 +1000 From: Stephen McKay cc: Stephen McKay Subject: Re: ATA READ command timeout (and worse) X-BeenThere: freebsd-hardware@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: General discussion of FreeBSD hardware List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 02 Jul 2003 16:10:02 -0000 On Tuesday, 24th June 2003, Stephen McKay wrote: >Woo hoo! First bug found. I'll see if I can find more. :-) A status update on my SiI 0680 troubles in FreeBSD 4.7: I was running one disk on the M/B IDE controller and 3 disks on the SiI0680. When I thrashed 2 disks, one on each channel of the SiI, I got command timeouts. It all looked very bad. I changed cables, considered replacing the power supply, and even fiddled the driver code. But when I moved another disk off the SiI onto the M/B and put both problem disks on the same channel, all my problems went away. So far, in the only-use-one-channel configuration, I've done about a dozen runs of my exerciser test without a failure. Previously there was a failure (and sometimes a panic) during every test run. At the same time, Søren has committed some fixes and improvements to -current and says that with the better docs he now has for the SiI0680 he can better detect which channel is posting an interrupt since they share one on this card. He says he can't duplicate my problem, but I'm assuming this extra check will do the trick. So, the summary is: - As of very recently, SiI 0680 on -current should be OK (I've not tried it though). - Only use one channel of the SiI 0680 in 4.7/4.8/-stable unless you want occasional horrible failures. - There should be a fix in -stable soon. Cheers, Stephen.