On Mit, 2012-04-25 at 16:41 +0200, Mariusz Bialonczyk wrote:
> On 04/20/2012 01:33 PM, Michel Dänzer wrote:
> > The thing is, nouveau uses EXA as well, so it might just work by
> > accident. To rule that out, you could try disabling acceleration for
> > nouveau (if it allows that) or radeon (Option "NoAccel").
> Bingo! :)
> From my observations it seems you was right about it.
>
> With enabled acceleration for radeon driver (default) i've got xorg
> crashing right after starting psi.
> It seems this was reproducible - I crashed xorg several times right
> after starting psi.
>
> After disabling acceleration ("NoAccel" "true") for radeon, then
> everything seems ok. I was testing it for a day and i don't have
> any xorg crash during that time (when using psi) :).
That's using the nvidia driver for the nVidia card(s)? I'm afraid that's
nothing new, as the crashing code can only run if at least one driver
uses EXA.
The idea was to test radeon with nouveau but disable acceleration for
(only) one of them. If the crash still occurs in that case, it's
probably an xserver Xinerama bug, passing pixmaps from one screen to
lower layer hooks for another screen.
--
Earthling Michel Dänzer | http://www.amd.com
Libre software enthusiast | Debian, X and DRI developer
--
To UNSUBSCRIBE, email to debian-x-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1335365658.19225.31.camel@thor.local
