ERROR: Unable to load the 'nvidia-vgpu-vfio' kernel module

goanoah

New Member
Nov 11, 2022
2
1
1
Hi guys. I recently bought a dell r710 and an NVIDIA Quadro P2000. I have installed proxmox and already created some VMs. Everything works great. I also gave the graphics card to a VM as a PCI passthrough device and that also worked. But actually I want to give different VMs access to the graphics card. I read up a bit on the vGPU theme and finally almost succeeded with these instructions: https://gitlab.com/polloloco/vgpu-proxmox. Unfortunately, I always get the error "ERROR: Unable to load the 'nvidia-vgpu-vfio' kernel module" when installing the drivers. Do you have any idea what is going wrong?
Bash:
root@pve:~# lsmod | grep -i nouveau
root@pve:~#
Bash:
root@pve:~# lspci | grep -i nvidia
07:00.0 VGA compatible controller: NVIDIA Corporation GP106GL [Quadro P2000] (rev a1)
07:00.1 Audio device: NVIDIA Corporation GP106 High Definition Audio Controller (rev a1)
Bash:
root@pve:~# find /usr/lib/modules -name "*.ko" | grep -i nvidia
/usr/lib/modules/5.15.30-2-pve/kernel/drivers/i2c/busses/i2c-nvidia-gpu.ko
/usr/lib/modules/5.15.30-2-pve/kernel/drivers/net/ethernet/nvidia/forcedeth.ko
/usr/lib/modules/5.15.30-2-pve/kernel/drivers/platform/x86/nvidia-wmi-ec-backlight.ko
/usr/lib/modules/5.15.30-2-pve/kernel/drivers/video/fbdev/nvidia/nvidiafb.ko
/usr/lib/modules/5.15.30-2-pve/kernel/drivers/usb/typec/altmodes/typec_nvidia.ko
/usr/lib/modules/5.15.64-1-pve/updates/dkms/nvidia-vgpu-vfio.ko
/usr/lib/modules/5.15.64-1-pve/updates/dkms/nvidia.ko
/usr/lib/modules/5.15.64-1-pve/kernel/drivers/i2c/busses/i2c-nvidia-gpu.ko
/usr/lib/modules/5.15.64-1-pve/kernel/drivers/net/ethernet/nvidia/forcedeth.ko
/usr/lib/modules/5.15.64-1-pve/kernel/drivers/platform/x86/nvidia-wmi-ec-backlight.ko
/usr/lib/modules/5.15.64-1-pve/kernel/drivers/video/fbdev/nvidia/nvidiafb.ko
/usr/lib/modules/5.15.64-1-pve/kernel/drivers/usb/typec/altmodes/typec_nvidia.ko
Bash:
root@pve:~# dmesg | grep -i nvidia
[    7.030880] nvidia-nvlink: Nvlink Core is being initialized, major device number 234
[    7.030895] NVRM: The NVIDIA probe routine was not called for 1 device(s).
               NVRM: nouveau, rivafb, nvidiafb or rivatv
               NVRM: was loaded and obtained ownership of the NVIDIA device(s).
               NVRM: driver(s)), then try loading the NVIDIA kernel module
[    7.033114] NVRM: No NVIDIA devices probed.
[    7.033359] nvidia-nvlink: Unregistered the Nvlink Core, major device number 234
[    7.092344] [nvidia-vgpu-vfio] Unable to get symbol for nvidia_vgpu_vfio_get_ops from nvidia.ko
 
I have the exact same issue following the instructions of the page. Have you tried raising an issue on the Gitlab project you used? I checked and did not see any reference yet. Asking before doing it myself.
 
I found a solution an polloloco himself helped me :D
So the issue was, that I started with getting GPU passthrough working. That was the fault. U can either use GPU passthrough or vGPU. So if you also enabled gpu passthrough, just revert all steps and you should be fine. If that won't work, I can recommend visiting the vGPU Discord (https://discord.gg/g2K3ceYW) and ask these guys, they will help you fast :)

Good luck!
(I will update this thread soon with my solution and sry for my bad English :D)
 
Last edited:
  • Like
Reactions: TerribleSense
I found a solution an polloloco himself helped me :D
So the issue was, that I started with getting GPU passthrough working. That was the fault. U can either use GPU passthrough or vGPU. So if you also enabled gpu passthrough, just revert all steps and you should be fine. If that won't work, I can recommend visiting the vGPU Discord (https://discord.gg/g2K3ceYW) and ask these guys, they will help you fast :)

Good luck!
(I will update this thread soon with my solution and sry for my bad English :D)
Yes indeed. I too was first going down the GPU passthrough road. Still, then after doing more research trying to find the most current info I realized that I probably want to do vGPU and also found out that this is supported without patching as the P4 is on the "NVIDIA® Virtual GPU Software Supported GPUs" list here https://docs.nvidia.com/grid/gpus-supported-by-vgpu.html.

All I did to correct the "ERROR: Unable to load the 'nvidia-vgpu-vfio' kernel module." was remove "options vfio-pci ids=10de:1bb3 disable_vga=1" from /etc/modprobe.d/vfio.conf, then "update-initramfs -u", "reboot"

I followed this guide https://gitlab.com/polloloco/vgpu-proxmox made in Dec 2022 (3 weeks ago) and the only trouble I had was I had to manually update pve headers via "apt install pve-headers-$(uname -r)"

BTW I am new to homelabbing, linux, VMs, server, etc. I am about 1 month into learning (huge learning curve btw), so any mistakes, sorry in advance.
 

About

The Proxmox community has been around for many years and offers help and support for Proxmox VE, Proxmox Backup Server, and Proxmox Mail Gateway.
We think our community is one of the best thanks to people like you!

Get your subscription!

The Proxmox team works very hard to make sure you are running the best software and getting stable updates and security enhancements, as well as quick enterprise support. Tens of thousands of happy customers have a Proxmox subscription. Get yours easily in our online shop.

Buy now!