Last update killed my Beryl [Solved]

Message

grooveman · Post by **grooveman** » Wed Apr 25, 2007 2:07 am

Hi.

I had beryl working for a while now. It was nice and spiffy.

I did an 'emerge -uD world' (after syncing my portage tree) last night, and that seems to have killed it.

Here is what was merged:

Code: Select all

These are the packages that would be merged, in order:

Calculating world dependencies  ..... ..... ..... .... done!
[ebuild     U ] sys-apps/ed-0.4 [0.2-r6]
[ebuild     U ] x11-libs/libXfixes-4.0.3 [4.0.1]
[ebuild     U ] x11-proto/inputproto-1.4.1 [1.3.2]
[ebuild     U ] media-gfx/imagemagick-6.3.3 [6.3.0.5-r1]
[ebuild     U ] x11-proto/randrproto-1.2.1 [1.1.2]
[ebuild     U ] sys-process/psmisc-22.3 [22.2]
[ebuild     U ] x11-libs/libXi-1.1.0 [1.0.1]
[ebuild     U ] x11-libs/libXrandr-1.2.1 [1.1.1]
[ebuild     U ] x11-libs/libXcomposite-0.3.1 [0.3]
[ebuild     U ] x11-libs/libXdamage-1.1.1 [1.0.3]
[ebuild     U ] x11-wm/fluxbox-1.0_rc3 [0.9.15.1-r2]
[ebuild     U ] dev-libs/dbus-glib-0.73 [0.72]
[ebuild     U ] app-cdr/k3b-1.0.1 [1.0] LINGUAS="-ar% -fa% -fi% -gl% -ka% -rw% -

When beryl starts, things work for about 1 second, then the screen locks. I still have a working mouse cursor, but I can't click anything on my KDE bar (like the big 'K' or anything on the kicker). The screen will likewise not accept any regular keyboard input.

Weirdly enough, I can spin the cube, and I can ctr+alt+F# to get a VT. Once at an ANSI terminal, I kill the beryl process, and my backup wm (kwin) kicks in, and things work as you would expect sans beryl.

I have since remerged all these beryl components:

Code: Select all

#Beryl Core
x11-wm/beryl
x11-wm/beryl-core
x11-plugins/beryl-plugins
x11-plugins/beryl-dbus
x11-misc/beryl-manager
x11-misc/beryl-settings
x11-misc/beryl-settings-bindings
x11-apps/xlsclients
x11-proto/damageproto

#Window Decorators
x11-wm/emerald
x11-themes/emerald-themes
x11-wm/heliodor
x11-wm/aquamarine
=x11-drivers/nvidia-drivers-1.0.9631*

It did not fix anything. None of the packages I had updated caused any writes to any files in /etc, so nothing changed in terms of configuration.

I do have this in my /var/log/messages: but I don't know if it is significant to this problem:

Code: Select all

Apr 24 15:38:22 chris64 beryl[32469]: segfault at 0000000000000150 rip 000000000040dfdd rsp 00007fff32b46050 error 4

I'm on a 64bit AMD single core compile.

Any ideas?

Thanks.

G

Tyir · Post by **Tyir** » Wed Apr 25, 2007 4:19 am

http://forums.gentoo.org/viewtopic-t-55 ... beryl.html

(in short, recompile xorg-server)

Koraktor · Post by **Koraktor** » Wed Apr 25, 2007 8:45 am

If you use XGL, you have to rebuild xgl (and xorg-server?).
Did cost me a few hours of testing.

SpikeXtrem · Post by **SpikeXtrem** » Fri Apr 27, 2007 1:02 am

I had the same problem. I updated xorg-server to 1.3.0 and xorg-x11 to 7.2

Now it works fine

Progman3K · Post by **Progman3K** » Fri Apr 27, 2007 9:57 pm

Hi,
I'm having the same problem you originally started out with and I've tried recompiling the same modules but it doesn't change anything.
The e-builds don't even seem to compile anything, but end very quickly saying the original instance was un-merged and all is OK...

What are the flags you are supplying to emerge?

grooveman · Post by **grooveman** » Sun Apr 29, 2007 1:40 pm

remerging xorg-server did it for me!

Thanks!

G