ATI Radeon driver 6.7.197 regression on IGP chipsets

[Xavier Bestel]

On Mon, 2008-02-11 at 17:40 -0500, Alex Deucher wrote:
> On Feb 10, 2008 12:37 PM, Frederik Himpe <fhimpe at telenet.be> wrote:
> > I have a machine with an ATI IGP 340M chipset, and since the 6.7.197
> > version, X hangs completely at startup in a loop if I don't set AGPMode 4
> > in Xorg.0.log. Looking at Debian's ati package changelog, this regression
> > has been fixed in git. Can somebody point me to a patch or something? Is
> > 6.7.198 with this fix expected soon?
> 
> You can follow the latest ati driver changes here:
> http://cgit.freedesktop.org/xorg/driver/xf86-video-ati/
> I recently reverted the default AGPMode back to whatever the bios sets
> up.  Unfortunately, AGP is a bit of a mess in general and it's hard to
> know what modes work reliably on what card/MB combinations.  It seems
> like leaving things as set up by the bios is the most reliable at the
> moment.
> 
> I hoping to make a new ati release soon.

BTW, I had lots of deadlock problems in the recent past, but it seems to
behave really better since the latest debian/experimental release (and
maybe since kernel 2.6.24).
The current version in use is 6.7.198~git20080203.f65e8dfa-1

Thanks,

	Xav