I have been having issues with bumblebee after a kernel package upgrade. It started in 12.04 and has been carried on into 13.04.
Basically it is outlined as follows; after a clean install of ubuntu (no matter the release), i install bumblebee and it works well out of the box. When the kernel is upgraded (e.g. from 3.8.0-19 to 3.8.0-22), After some time, bumblebee breaks with the following error in /var/log/syslog:
kernel: [ 4.135691] bbswitch: Unknown symbol mcount (err 0)
bumblebeed[1259]: Module bbswitch could not be loaded (timeout?)
bumblebeed[1259]: No switching method available. The dedicated card will always be on.
bumblebeed[1259]: /usr/sbin/bumblebeed 3.2.1 started
The output of bumblebeed --v:
[ 3326.529062] [DEBUG]Found card: 01:00.0 (discrete)
[ 3326.529106] [DEBUG]Found card: 00:02.0 (integrated)
[ 3326.529121] [DEBUG]Reading file: /etc/bumblebee/bumblebee.conf
[ 3326.529739] [INFO]Configured driver: nvidia
[ 3326.529767] [DEBUG]Skipping auto-detection, using configured driver 'nvidia'
[ 3326.529986] [DEBUG]Process /sbin/modprobe started, PID 4986.
[ 3326.530101] [DEBUG]Hiding stderr for execution of /sbin/modprobe
[ 3326.533252] [DEBUG]SIGCHILD received, but wait failed with No child processes
[ 3326.533715] [INFO]Loading driver bbswitch (module bbswitch)
[ 3326.533961] [DEBUG]Process modprobe started, PID 4987.
ERROR: could not insert 'bbswitch': Unknown symbol in module, or unknown parameter (see dmesg)
[ 3326.537523] [DEBUG]Process with PID 4987 returned code 1
[ 3326.537888] [ERROR]Module bbswitch could not be loaded (timeout?)
[ 3326.537936] [DEBUG]bbswitch is not available, perhaps you need to insmod it?
[ 3326.537993] [INFO]Skipping switcheroo PM method because it is not explicitly selected in the configuration.
[ 3326.538022] [WARN]No switching method available. The dedicated card will always be on.
[ 3326.538047] [DEBUG]Active configuration:
[ 3326.538078] [DEBUG] bumblebeed config file: /etc/bumblebee/bumblebee.conf
[ 3326.538100] [DEBUG] X display: :8
[ 3326.538122] [DEBUG] LD_LIBRARY_PATH: /usr/lib/nvidia-304:/usr/lib32/nvidia-304
[ 3326.538150] [DEBUG] Socket path: /var/run/bumblebee.socket
[ 3326.538172] [DEBUG] pidfile: /var/run/bumblebeed.pid
[ 3326.538207] [DEBUG] xorg.conf file: /etc/bumblebee/xorg.conf.nvidia
[ 3326.538238] [DEBUG] xorg.conf.d dir: /etc/bumblebee/xorg.conf.d
[ 3326.538260] [DEBUG] ModulePath: /usr/lib/nvidia-304/xorg,/usr/lib/xorg/modules
[ 3326.538282] [DEBUG] GID name: bumblebee
[ 3326.538306] [DEBUG] Power method: auto
[ 3326.538328] [DEBUG] Stop X on exit: 1
[ 3326.538354] [DEBUG] Driver: nvidia
[ 3326.538375] [DEBUG] Driver module: nvidia-304
[ 3326.538402] [DEBUG] Card shutdown state: 1
[ 3326.538635] [DEBUG]Process /sbin/modprobe started, PID 4988.
[ 3326.538722] [DEBUG]Hiding stderr for execution of /sbin/modprobe
[ 3326.541954] [DEBUG]SIGCHILD received, but wait failed with No child processes
[ 3326.542076] [DEBUG]Configuration test passed.
[ 3326.542196] [ERROR]Daemon already running, pid 1292
The output of dmesg | grep bbswitch:
bbswitch: Unknown symbol mcount (err 0)
The output of dkms bbswitch status:
bbswitch, 0.7, 3.8.0-19-generic, x86_64: installed
bbswitch, 0.7, 3.8.0-22-generic, x86_64: installed
This issue has recently been posted here but no solution has been posted. Also it is different from this issue.
Currently, the only workaround i have found so far is redoing a clean install of ubuntu, upgrading to the lastest kernel version and then installing bumblebee. After this it works fine, until the next kernel bumblebee upgrade where it will break again and i have to redo the installation. This is far from optimal as i have to do this approximately every 2 weeks :(.
Some info:
Ubuntu 13.04; kernel version 3.8.0-22-generic
Clevo W150HR, i7 2630QM, NV GT555M, 8GB RAM, SSD+HDD
Anybody any ideas on how to fix this?
Edit: It turns out to be an issue with an outdated version of gcc.
