cvs commit: src/sys/dev/ata ata-dma.c ata-pci.c

[Nate Lawson]

On Wed, 22 Oct 2003, Kevin Oberman wrote:
> > From: Soren Schmidt <sos at spider.deepcore.dk>
> > Date: Wed, 22 Oct 2003 23:01:08 +0200 (CEST)
> > Sender: owner-cvs-all at freebsd.org
> >
> > Hmm, I've not heard of anybody else having trouble with the ich5 on
> > -current, and you did test without vinum involved ?
> >
> > -Søren
>
> Whils it's not an ICH5 (it's an ICH3, for heaven's sake), I see
> problems with CURRENT on my laptop, too. (Obviously not SATA.) This is
> probably an unrelated issue, but I thought I'd mention it in case it
> might be related.
>
> I am unable to resume from suspend. I get the "Resetting devices..."
> message for each ATA bus, but only get a single "done" if I have
> drives on both buses and no "done" when I have only one drive
> present. Looks like a lost interrupt. The drive busy LED just stays on
> until I force a power off. Last I heard Nate was seeing the same thing
> on his laptop, so I believe you may already have heard about it.
>
> The laptop is an IBM T30 (2366-92U). Other details available on
> request.
>
> This problem only exists on CURRENT when ACPI is in use. APM works
> fine, but that does not require so much interaction with the OS. The
> problem started with ATAng.

I used to have this problem but it appears to be fixed in later ATAng
commits.  I last tested it on Sunday.  There is still a panic somewhere
but I don't know if it's ATAng since I was in X and dumps no longer work
on my laptop.  I'll use a serial console and see if I can get a panic
again.

-Nate