Date: Sun, 8 Dec 2002 22:28:38 +0000 From: Bruce Cran <bruce@cran.org.uk> To: "Cliff L. Biffle" <cbiffle@safety.net> Cc: current@freebsd.org Subject: Re: ATACD issues slowly coming back... Message-ID: <20021208222838.GA3422@fourtytwo.gamesoc> In-Reply-To: <200212061435.45583.cbiffle@safety.net> References: <200212061435.45583.cbiffle@safety.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, Dec 06, 2002 at 02:35:45PM -0700, Cliff L. Biffle wrote: > I mentioned earlier on the list that the ATA issues I'd been having with 4.7 > had disappeared since installing 5.0. They're still much less frequent -- > i.e. I can burn CDs now -- but I just got one of the old messages and wanted > to submit it for your perusal. > > cliff50 kernel: acd0: READ_BIG - MEDIUM ERROR asc=0x11 ascq=0x00 error=0x00 > > I'm not well versed on the ATA command set, but that looks like a command to > the drive failing to execute. Is this likely to be the drive, the > controller, both, or is it impossible to tell from the data set? > > This is on the aforementioned potentially-buggy Apollo KT133 chipset, but it's > never reported these errors on my hard disk (knock on wood), only the CD > drive. They -are- on separate channels. > > I'll see what information I can collect here, but suggestions on where to look > are appreciated. I reported a panic under -CURRENT a few days ago (see ATA/ATAPI related panic). When I tested out DP1 I never saw any ATAPI problems, but since trying DP2 I have, once more, seen errors. I've got a A7V-266E motherboard with a KT266A chipset. The solution in my case was to tell the BIOS I didn't have any ATAPI drives. FreeBSD then found everything properly, without any problems - I think the BIOS was maybe configuring the master for UDMA33 and the slave for PIO4, whereas FreeBSD seems to prefer them both as PIO4, but I really don't know. I kept seeing READ_CAPACITY and READ_BIG timeouts, along with data overrun errors. I think it probably is the controller, because even with no drives connected, my computer paused for a good few seconds while trying to probe the io/irq - it found the primary (which only has a hard drive on it) immediately, but paused far too long trying to find the secondary. -- Bruce Cran To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20021208222838.GA3422>