[1721] in Release_7.7_team
One step forward, one step back: Open GL problems on Suns
daemon@ATHENA.MIT.EDU (Bill Cattey)
Tue Mar 23 15:20:52 1999
Date: Tue, 23 Mar 1999 15:20:41 -0500 (EST)
From: Bill Cattey <wdc@MIT.EDU>
To: Alex T Prengel <alexp@MIT.EDU>
Cc: ajfox@MIT.EDU, bugs@MIT.EDU, nschmidt@MIT.EDU, release-team@MIT.EDU
In-Reply-To: <199903231655.LAA10160@allegretto>
I met with Abby after the release team meeting and we played a bit with
the openGL tests.
We successfully ran HelloUniverse both on my Creator3D (ffb) and on Greg
Hudson's Creator Elite3D (afb). This puts us nearly back to square one.
:-(
Clarification:
ALL Ultra 5 and 10's have an m64 video display.
The reason I asked you to grep for it in dmesg was to find out whether
it was a new Ultra-5 (4M of memory) or an old Ultra-5 (2M of memory)
At that point, the relevant bit of information is whether the machine
has an afb (Creator Elite3D) or and
ffb (Creator 3D).
Unfortunately it is becomming clear that the bug is not so clearly in
one type or another of frame buffer.
There is yet another set of patches and a new version of OpenGL
available from sun. My suggested plan of attack here is this:
In Parallel:
Continue to isolate the fault down to a size Sun can reproduce.
Test with the latest set of patches and pray the bug goes away.
Does this sound reasonable?
-wdc
