Re: [Chromium-users] FireGL 4 hardware acceleration

[Brian P. <br...@tu...>]

Augustinas Abaravichyus wrote:
> 
> FROM: Brian Paul
> DATE: 04/25/2002 10:17:38
> SUBJECT: RE:  [Chromium-users] FireGL 4 hardware acceleration
> 
> dasilva wrote:
> >> > Brian Paul wrote:
> >> > >Let's try a little debugging first.  Try setting the LIBGL_DEBUG
> >> >environment variable to "verbose" before starting the crserver or
> >> >crappfaker.  That'll tell us if it's a libGL problem.
> 
> Thank you for the the information.  I am getting almost the same error
> as described by Jean-Claude.  The only difference in my case was the
> error with firegl23w_dri.so (not.  firegl23s_dri.so).. same version of
> the driver 1.9.7, the rest of the message was identical.  Strangely
> enough.. all opengl apps I have tried so far work fine (we have quite a
> few in-house visual apps written using SGI Performer, VTK, pure OpenGL,
> as well as tested the systems running the included demos), in fact
> WireGL works well with the same configuration.
> 
> I will submit this to ATI as well, although from past experiences found
> their linux driver support to be fairly limited.  Also I'm taking some
> time to study the chromium source code, but if anyone has any pointers
> on how to get around this issue or at least where to start (or any
> reasons why such fix would be impossilbe on the chromium side) it would
> be greatly apreciated.
> 
> Thank You for your time and the help provided so far!!!

Chromium's unique in that all the OpenGL entrypoints are used via
pointers obtained by dlsym().  Maybe that's the difference.

I've got a test program which does exactly this that I'll send
to you.

-Brian