CDRW causes Thinkpad T41 to panic

[Daniel Eischen]

On Sun, 11 Jul 2004, Bruce M Simpson wrote:

> On Wed, May 26, 2004 at 05:21:14PM -1000, Kent Hauser wrote:
> > ad0: 35293MB <HTS548040MAT00> [71707/16/63] at ata0-master UDMA100
> > ata1-slave: FAILURE - ATAPI_IDENTIFY no interrupt
> > ata1-slave: FAILURE - ATAPI_IDENTIFY no interrupt
> > acd0: CDRW <UJDA755zDVD/CDRW> at ata1-master UDMA33
> > Memory modified after free 0xc44d8a00(508) val=1ff01ff @ 0xc44d8a00
> 
> Again, this looks VERY, VERY similar to the behaviour which I observed
> and documented on two separate IBM T41 systems this week.

Same here.  If you set request->retries to 1 instead of 2 in
src/sys/dev/ata/ata/foo it seems to work.  I think only one of
the retries needs to change, but I forget which one it is.

I believe there is a problem in allocation/deallocation in
the retry.  Something is getting deallocated but still being
referenced.  I thought it was the request (ata_alloc_request()),
but after debugging it a bit, I think I figured eliminated that
as a possibility.  I stopped there when I ran out of time.

Has anyone reported this problem to Soren?  I don't think
he reads -current any longer (too much spam).

-- 
Dan Eischen