I have very poor steam game performance using primusrun %command%, as poor as when primusrun is not used. This is after the bug that caused steam unusable for three months has been fixed recently.
I checked the log and there is sth like this when the game is launched
Attempting to create OpenGL context: v3.3 (Core profile)...success.
Attempting to create OpenGL context: v3.3 (Core profile)...success.
DumpContextInfo: OpenGL vendor Intel Open Source Technology Center
DumpContextInfo: OpenGL renderer Mesa DRI Intel(R) Haswell Mobile
DumpContextInfo: Using OpenGL context version 3.3
DumpContextInfo: Context supports GLSL version 3.30
I doubt that nvidia is used at all since it shows intel with opengl?
But the command “primusrun glxspheres” works fine.
I don’t have Optimus graphics hardware, so can only offer limited advice, but hopefully others who do use bumblebee can advise further. Anyway, to start with, what do the following report?
The following results were obtained when I have the steam game running with “primusrun %command%” and a “primusrun glxspheres” session running in the background. So it seems nvidia is not running at all?
systemctl status bumblebeed
**●** bumblebeed.service - Bumblebee C Daemon
Loaded: loaded (/usr/lib/systemd/system/bumblebeed.service; enabled; vendor preset: disabled)
Active: **active (running)** since Sun 2016-10-23 12:18:05 CST; 38min ago
Main PID: 1411 (bumblebeed)
CGroup: /system.slice/bumblebeed.service
└─1411 /usr/sbin/bumblebeed
Oct 23 12:18:05 LENSU systemd[1]: Started Bumblebee C Daemon.
Oct 23 12:18:09 LENSU.suse bumblebeed[1411]: 25.129322] [INFO]/usr/sbin/bumblebeed 3.2.1 started
optirun --status
Bumblebee status: Ready (3.2.1). X inactive. Discrete video card is off.
Please also post the content of /etc/modprobe.d/50-bbswitch.conf
Though I am using prime-select now, I had had a problem with bumblebee, that looks similar to yours.
The Nvidia card got detected, but was not initialized correctly on boot up, because bbswitch turned its power off
before this could be done.
sudo systemctl status bumblebeed
● bumblebeed.service - Bumblebee C Daemon
Loaded: loaded (/usr/lib/systemd/system/bumblebeed.service; enabled; vendor preset: disabled)
Active: activating (auto-restart) (Result: exit-code) since Sun 2016-10-30 22:27:05 GMT; 24s ago
Process: 16390 ExecStart=/usr/sbin/bumblebeed (code=exited, status=1/FAILURE)
Main PID: 16390 (code=exited, status=1/FAILURE)
Oct 30 22:27:05 LENOVO systemd[1]: bumblebeed.service: Unit entered failed state.
Oct 30 22:27:05 LENOVO systemd[1]: bumblebeed.service: Failed with result 'exit-code'.
and
sudo optirun --status
[11946.962811] [ERROR]The Bumblebee daemon has not been started yet or the socket path /var/run/bumblebee.socket was incorrect.
[11946.962831] [ERROR]Could not connect to bumblebee daemon - is it running?
[11946.962836] [WARN]The Bumblebee server was not available.
Segmentation fault
Where do I report that bumblebeed broke in Tumbleweed over a week ago, and is still broken?
The fix for this was to login in as root on tty1 (Alt Ctrl F1)
killall lightdm
systemctl stop bumblebeed
cd Nvidia
./NVIDIA-Linux-x86_64-367.57.run
And re-building that fixed bumblebeed
My thought is that previously - and I may be wronf - that bumblebeed has been running, but the nvidia driver hasn’t been working properly. Previously I was getting about 25fps on optirun glxspheres, but it did run
this time I got errors with bumblebeed not running, check it’s installed or something similar.
hence trying to start the service via systemctl and then seeing the above mentioned errors
Is the default fix if you have glx / optirun errrors just to kill your DM, stop bumblebeed (if it’s even running, anyway) and rebuild the Nvidia drivers?