- System: AOpen MiniPC MP965-DR - CPU/RAM: Core 2 duo T7500, 2GB ram - OS: Fedora 8 - X server: xorg 1.4.99.1-0.10.fc9 - intel_drv.so: git master - mesa/drm drivers: git master - system outputs: VGA, TMDS-1, LVDS, TV - TV: 32" Sony WEGA, NTSC 480i only Git commit "d9edd8e90588417e3d549f25132dab2f21445792" to the mesa/mesa module on January 9th, 2008, introduced a major regression on this system. After this commit, the window or fullscreen display is not updated properly. Sometimes the entire display area will be black; other times, only a rectangular subset of the display area will be properly updated. Reverting this commit fixes the problem for me.
Reassigning to Eric, as the regression is introduced by: commit d9edd8e90588417e3d549f25132dab2f21445792 Author: Eric Anholt <eric@anholt.net> Date: Wed Jan 9 13:24:29 2008 -0800 [965] Remove drawing rect upload, which is handled (better) by the kernel.
Confirming this problem, see bug #14314 for my ramblings about this and other problems I'm having. Reverting the mentioned commit makes (some) 3D applications to work properly. No idea if this is related to the fact that we can't get the TTM buffer manager initialization working (bug #13922 and http://lists.freedesktop.org/archives/xorg/2008-January/031907.html).
Once I was able to get the TTM working, this started to work also without the commit reverted. The original reporter might also want to check if DRI_MM is turned on when compiling the intel ddx driver and if not, why. Is not using TTM now obsolete, or should things still work without TTM? If the former, this bug can be closed once the original reporter checks if he can get TTM to work and if the problem disappears then.
I'd like to keep non-TTM mostly working for a few releases to give time for other OSes to catch up, but it's not a high priority since TTM is the way forward.
Closing bug. I was also able to verify that this bug is not present when everything is compiled from source, in the proper order, with the up-to-date libdrm.pc in the right place. It took a lot of testing to come to this conclusion with the git tip, though, due to other problems. I got a bunch of "3D driver claims to not support visual" errors and software-only rendering when AIGLX is disabled, and the glxgears process locks up immediately when I run it with INTEL_NO_TTM=1.
Mass version move, cvs -> git
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.