Radeon 9000 Pro corruption problem on PPC when DRI is enabled

[Michel Dänzer]

On Wed, 2006-08-30 at 19:50 -0400, Ari Entlich wrote:
> On Mon, 2006-08-28 at 10:14 +0200, Michel Dänzer wrote:
> > 
> > If disabling colour tiling and the AGP options doesn't make a
> > difference, I'm not sure what's up; can you put up a corresponding log
> > file in that case?
> 
> Hmm... That didn't help :-/.  Get the log file at
> http://home.twcny.rr.com/fandaswebsite/xorg_log_corruption . I can
> reliably reproduce this corruption by going to a text VT and back to X.
> However, I think it can also happen if the session get sufficiently old
> (not positive though).

Hmm. I've actually seen something like this; I thought it was caused by
my experimental changes to the AccelDFS code, but maybe not. I never saw
anything like this when I was testing my original exa-damagetrack
changes though. As you have a reliable way to reproduce it, you're in a
good position to do some regression testing. :) I'd suggest a git bisect
in the xserver tree, but you may want to first try xf86-video-ati 6.6.0
as well to rule out a driver regression that may have snuck in at some
point.


-- 
Earthling Michel Dänzer           |          http://tungstengraphics.com
Libre software enthusiast         |          Debian, X and DRI developer