Red Hat Bugzilla – Bug 373761
EXA causes Nautilus to use random colors instead of set preference if monochromatic
Last modified: 2007-11-30 17:12:21 EST
Description of problem:
After enabling EXA via the AccelMethod xorg.conf option, Nautilus uses random
colors for its backgrounds, even for the desktop background (if I set an empty
wallpaper and choose "Solid Color").
However, if I choose to use a gradient, and both the "start" and "end" colors
are the same, this does not occur. (It then properly uses that color.)
This leads me to believe, at first glance, that it's something to do with how
EXA accelerates the solid-fill stuff.
I'm quite certain that it is nothing else which might cause this, as merely
toggling EXA (via the AccelMethod option in my xorg.conf) then restarting X (via
a reboot or init 3 then init 5) creates this error with nothing else
intentionally modified or changed.
Version-Release number of selected component (if applicable):
Steps to Reproduce:
1. Install Fedora 8, using the "intel" video driver.
2. Change the AccelMethod option in /etc/X11/xorg.conf to "exa" instead of the
3. Restart X somehow.
The random colors seen in the screenshot.
It should be using the proper colors set in my preferences.
My hardware is the built-in G965 graphics on a DG965WH desktop motherboard,
using an E6600 processor (x86_64).
00:00.0 Host bridge: Intel Corporation 82P965/G965 Memory Controller Hub (rev 02)
00:02.0 VGA compatible controller: Intel Corporation 82G965 Integrated Graphics
Controller (rev 02)
00:02.1 Display controller: Intel Corporation 82G965 Integrated Graphics
Controller (rev 02)
If you need me to attach other logs or data, please don't hesitate to ask. :)
Created attachment 253491 [details]
Xorg.0.log with EXA enabled
Created attachment 253511 [details]
Xorg.0.log with XAA
I believe that this problem is the same as
The difference being that I was using the greedy Heuristic.
If not it would then be related to.
I definitely ran acrossed the bug in my testing and now have it resolved on my
Yup, that's the same. Thanks, Jon! :]
*** This bug has been marked as a duplicate of 340471 ***