New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Failed to load module "nvidia" (module does not exist, 0) #526
Comments
Tried to edit for clarity. |
i see same error message on fedora 20 after kernel upgrade to 3.12.7-300 few days back. nvidia kernel module exists in /usr/lib/modules/3.12.7-300.fc20.x86_64/kernels/drivers/video/nvidia.ko but it doesnt get loaded according to Xorg log: |
@deathkillz You can try /usr/sbin/bumblebee-nvidia --debug as root to see if the nvidia module can be compiled on fedora 20. |
@jpmgsxr I added more markdown to your first post. Your Xorg logs indicate that the xorg nvidia module cannot be found. Can you try removing bumblebee and all nvidia packages, then try to install |
only error i see is: |
@Lekensteyn Thanks for the clean up. I've removed and purged bumblebee and all nvidia packages probably about three different times. /var/log/apt/history.log
That is my current attempt. |
After a reboot, I tried a few things.
Once I run
Here's dmesg:
|
Fixed with |
Hello, find /lib/modules/$(uname -r) -name 'nvidia*.ko*'
cat /var/log/Xorg.8.log | grep -i nvidia
And if I try to load any app using optirun I get the same error as is described by this issue. |
Since i just had that error and it made me stare at my 42in 4k monitor in low resolution terminal mode for 5 hours and i stumbled on this ticket, i just want to add my comment (even if this might not be related specifically to bumblebee, i cant judge), hoping that someone else who might be in the same situation can save some time: I am not a linux config expert or anything, but to me it looks like the update of virtualbox i did at 2018-03-26 19:54:44 might have somehow added a gl-driver that was picked as the default driver which i then had to change by picking it via:
But even that did not work directly, so i apt-get purged nvidia* and virtualbox*, then installed nvidia-390, then ran nvidia-xconfig and then rebooted, and finally my eyes stopped hurting and i could stop crying. |
@zubozrout Looks like nvidia-driver-390 package in bionic (or actually xserver-xorg-video-nvidia-390) puts the the drivers in a different place. I did the following to point to the new locations: Added this to /etc/bumblebee/xorg.conf.nvidia
Also updated /etc/bumblebee/bumblebee.conf
This did the trick for me! |
Thank you very much for sharing that @kaczmarek-tomasz though it seems it still doesn't work for me :(. In any case at least I no longer have any of the missing libraries messages.
Neither of the commands above work but if I run glxgears on their own they pop up. |
After the changes of @kaczmarek-tomasz I now get
Any idea where to look further? |
@tverbele I'm guessing that the paths you have in the
So I'd locate |
Turned out I misconfigured
Any ideas? |
@tverbele Are you sure you have
|
@zubozrout I checked and unfortunately on my machine the output of these is the same. Looks like I'm not able to run glxgears with nvidia driver :( It only runs through the built-in Intel card. |
@kaczmarek-tomasz Check out this post, it seems to fix the issue: #951 (comment) (I've only tested it now and it worked) |
Edit: Ok, it probably doesn't work as I thought and I didn't notice that. You need to have Intel card enabled to run bumblebee, right? That is do "prime-select intel" first. I was on nvidia at that time which was also the reason why it worked for me. Now I am on Intel and getting this:
So it matches the replies here: #951 (comment) ... as Nvidia driver doesn't seem to load on my kernel: 4.15.0-13-generic |
In my case the nvidia_384's driver is located in /usr/lib/nvidia-384/xorg and it still causes the same error. Any ideas why it doesn't even try to load the driver located there? |
☁ ~ sudo primusrun glxgears |
@kaczmarek-tomasz it did the trick for me too! EDIT: Still having issues, the nvidia kernel module isn't loading correctly. |
issue persists with nvidia-driver-525:
|
Well, Idk if it still can help someone, but I was facing
because my nvidia module (nvidia_drv.so) was not listed in the default folders from Summing up, I've added the last part of the following, restarted the service ( before: after:
... Of course ... You need to have nvidia-driver installed (try checking Or finally, try adding manually the BusID in the |
I'm using Debian Sid, updated as of 1/17/14. When I initially installed Debian stable, I upgraded to testing. While I was using testing, I installed Bumblebee. After some time I upgraded to Sid. (Unstable.) At this point I was using the same exact kernel (which I'm still using as of this post.) as testing.
I can't pin point an exact time where things went wrong, (I haven't used Bumblebee recently.) But I tried running glxgears through Bumblebee and below is the result.
Here's some information to clarify a few things:
bumblebee.conf
xorg.conf.nvidia
/var/log/Xorg.8.log
Additional info..
So far I've completely purged the packages I installed for bumblebee. (Including primus and primus:i386.)
I ran
apt-get ---purge autoremove
andapt-get autoclean
.And then I completely reinstalled everything, following the directions given on the Bumblebee Debian wiki page.
I've worked with one or two people in #bumblebee on Freenode to try and fix the issue but nothing has worked thus far.
At this point, myself, and anyone who has helped me seemed to be stumped on the matter. If there's any other info I can provide, please let me know.
The text was updated successfully, but these errors were encountered: