Bug 20527 - [G965 non-GEM] systray in KDE 4 completely broken
[G965 non-GEM] systray in KDE 4 completely broken
Product: xorg
Classification: Unclassified
Component: Driver/intel
x86-64 (AMD64) Linux (All)
: high major
Assigned To: Eric Anholt
Xorg Project Team
: regression
Depends on:
Blocks: 20276
  Show dependency treegraph
Reported: 2009-03-07 12:25 UTC by walch.martin
Modified: 2009-03-10 12:08 UTC (History)
0 users

See Also:
i915 platform:
i915 features:

broken icons in systray (1.47 KB, image/png)
2009-03-07 12:25 UTC, walch.martin
no flags Details
dmesg (28.83 KB, text/plain)
2009-03-07 12:34 UTC, walch.martin
no flags Details
/var/log/Xorg.0.log (109.14 KB, text/plain)
2009-03-07 12:35 UTC, walch.martin
no flags Details
/etc/X11/xorg.conf (2.04 KB, text/plain)
2009-03-07 12:36 UTC, walch.martin
no flags Details

Note You need to log in before you can comment on or make changes to this bug.
Description walch.martin 2009-03-07 12:25:55 UTC
Created attachment 23642 [details]
broken icons in systray

If you know KDE 4, you probably know about the broken systray icons: the background is garbled.

With current xf86-video-intel 2.6.3, the systray also does not show any icons. It switches between not showing anything and showing garbage (without distinguishing between fore- and background).

This was fine with 2.5.1. I bisected it to 6c0ca1676bf60529dd331cc739abdf68fa9e918d

System environment: 
 -- chipset: G965
 -- system architecture: 64-bit
 -- xf86-video-intel: 2.6.3
 -- xserver: 1.5.3
 -- mesa: 7.2
 -- libdrm: 2.4.5
 -- kernel: 2.6.27-gentoo-r8
 -- Linux distribution: Gentoo
 -- Machine or mobo model: Intel DG965SS
 -- Display connector: VGA/DE-15
more precisely:

The mainboard:

uname -a:
Linux tacticalops 2.6.27-gentoo-r8 #10 SMP Sun Mar 1 17:18:33 CET 2009 x86_64 Intel(R) Core(TM)2 CPU 4300 @ 1.80GHz GenuineIntel GNU/Linux
Comment 1 walch.martin 2009-03-07 12:34:31 UTC
Created attachment 23643 [details]
Comment 2 walch.martin 2009-03-07 12:35:34 UTC
Created attachment 23644 [details]
Comment 3 walch.martin 2009-03-07 12:36:12 UTC
Created attachment 23645 [details]
Comment 4 Gordon Jin 2009-03-07 17:34:07 UTC
Thanks for the bisect. So it's caused by:

author	Eric Anholt <eric@anholt.net>	2009-01-21 23:52:02 (GMT)
committer	Eric Anholt <eric@anholt.net>	2009-01-27 22:33:43 (GMT)
commit	6c0ca1676bf60529dd331cc739abdf68fa9e918d (patch)
Don't forget the new state bos in check_aperture.
They're tiny so it shouldn't have been a problem, but play it safe. This is another <5% loss on top of the previously reported value, bringing the whole series to about 8%. 
Comment 5 Eric Anholt 2009-03-09 22:05:13 UTC
Might check with:

commit fb6e00f40f713a87c760fc7603159ed11ea9b0d5
Author: Eric Anholt <eric@anholt.net>
Date:   Mon Mar 9 20:06:30 2009 -0700

    Fix XV with non-GEM kernels by allocating a larger fake bufmgr.
Comment 6 walch.martin 2009-03-10 02:57:07 UTC
After syncing to git master, it's working fine. :)

(I do not set the bug to resolved/fixed, because I do not know your policies for doing that)
Comment 7 Eric Anholt 2009-03-10 12:08:50 UTC
The policy is, if it's fixed upstream, go ahead and mark it fixed :)

Thanks for testing.  I'm pretty sure it was that commit.