From owner-freebsd-hackers Wed Feb 20 11:53:29 2002 Delivered-To: freebsd-hackers@freebsd.org Received: from freebsd.dk (fw-rl0.freebsd.dk [212.242.86.114]) by hub.freebsd.org (Postfix) with ESMTP id 782B537B400; Wed, 20 Feb 2002 11:53:24 -0800 (PST) Received: (from sos@localhost) by freebsd.dk (8.11.6/8.11.6) id g1KJrKq77437; Wed, 20 Feb 2002 20:53:20 +0100 (CET) (envelope-from sos) From: Søren Schmidt Message-Id: <200202201953.g1KJrKq77437@freebsd.dk> Subject: Re: Serverworks ATA controller & data corruption In-Reply-To: <15475.50753.252494.269972@grasshopper.cs.duke.edu> To: Andrew Gallatin Date: Wed, 20 Feb 2002 20:53:20 +0100 (CET) Cc: sos@FreeBSD.ORG, freebsd-hackers@FreeBSD.ORG Reply-To: sos@freebsd.dk X-Mailer: ELM [version 2.4ME+ PL94b (25)] MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset=ISO-8859-1 Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG It seems Andrew Gallatin wrote: > I have a few machines with the following ata controller: > > atapci0@pci0:15:1: class=0x01018a card=0x00000000 chip=0x02111166 rev=0x00 hdr=0x00 > They're dual-boot FreeeBSD/linux boxes. After loosing 2 filesystems in > linux, I did a web search and I found that linux has a problem with > data corruption when using these controllers & disabled IDE DMA in linux. > > Thinking that FreeBSD was immune (these boxes spend 90% of their time > in FreeBSD doing driver development & crashing frequently, haven't > lost a fs yet), I continued to run the disk at UDMA33 in FreeBSD. > > However, I began to notice very minor data corruption when I attempted > to buildworld them from 4.4-RELEASE to 4.5-STABLE. By minor, I mean > syntax errors in files caused by 4 characters moved a few lines up or > down in a file, things like that. I concluded that FreeBSD might not > be immune after all and switched to PIO. After checking out a new src > tree, the problem seems to be gone and I'm able to buildworld. > > Do you think it would be wise to disable DMA by default on these chips? Hmm, the problem is known, but belived to be fixed *IF* your BIOS setup things the right way. I've newer seen the problem on my ASUS CUR-DLS, but I have several reports of TYAN's (forgot the model#) that fails all over. I have not verified if ASUS has done some HW trickery or if its just a BIOS matter. However the Serverworks ROSB4 chips is not one I would recommend using, if you need serious ATA support on such a board, install a Promise TX2 or later or a HPT370 or later ... -Søren To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message