2009/5/6 Francois Gouget fgouget@free.fr:
On Wed, 6 May 2009, Henri Verbeet wrote: [...]
I also *still* think us working around AMD's bugs is the wrong approach. Of course it would also help if AMD took these kind of things seriously, or at least replied to my posts (http://www.phoronix.com/forums/showpost.php?p=63527&postcount=107).
Do we have a more formal way of reporting bugs to AMD/ATI? Something like a Bugzilla? Concerning this issue (the driver reporting a number of vec4s instead of the number of uniforms), do we have a way of checking independently of AMD? If so that would strengthen our bug report.
I think they're aware of that particular issue because Stefan talked privately to one of their developers, but unless things changed recently http://ati.cchtml.com/ is probably the closest to a bug tracker there is, and I'm not completely sure they read that regularly.
But hey, I got a quick answer to my bug report to the Intel developpers so it's not always bad. Unfortunately now I'm the one that's slow to answer as I have not had time to test their proposed fix. But I have not forgotten...
Mesa (that includes e.g. the intel and radeon drivers) is generally responsive to bug reports, as is NVIDIA. It's really just AMD and Apple that are problematic wrt. getting bugs fixed.