ata - interrupt seen but task queue stalled

Tue Mar 23 00:34:46 PST 2004

At 02:49 AM 3/23/2004, Søren Schmidt wrote:

>Didn't you say it was on a ROSB4 chipset ? The ROSB4 doesn't really 
>support DMA on ATAPI devices, and has severe HW bugs that can cause 
>problems with UDMA on disks as well..

Yes, but this and sysctl's are a non-issue. The controller is set to PIO4 
and actually locked there. The drive itself doesn't do anything but PIO4, 
being a slim CD-ROM. The OS resides on an ips. ie; This is a genuinely new 
issue, and I can say with a fair degree of certainty that it's something 
that's been changed. I can't break to debug either, but the kernel 
configuration didn't change between working and non-working.

-Ketrien I. Saihr-Kesenchedra <ketrien AT error404 DOT nls DOT net>
# Maybe this world is another planet's hell. --Aldous Huxley
Feanor smiles, "We, of course, are enlightened and s/Maybe t/T" 