I got a new pc, and am having troubles getting my extra graphics cards to be detected bys PyTorch and nvidia-smi. I've looked around and can't find articles with the same issue
I'm using Ubuntu 22.04
I have a 4060 ti and two Quadro GP100s connected by a NVLink bridge
nvidia-smi detects my 4060 just fine and it outputs graphics, but not the quadros. lspci detects the quadros though, so they seem to be registered by the system.
The nvidia driver installed is 560.35.03, which Nvidia says supports the GP100s.
Does anyone know how to fix this? Is this somehow due to the NVlink bridge? Is there other software I need to install?
I set up my drivers automatically with
sudo ubuntu-drivers install
and I installed the cuda toolkit with
sudo apt install nvidia-cuda-toolkit
Below is the output of nvidia-smi
+-----------------------------------------------------------------------------------------+ | NVIDIA-SMI 560.35.03 Driver Version: 560.35.03 CUDA Version: 12.6 | |-----------------------------------------+------------------------+----------------------+ | GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC | | Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. | | | | MIG M. | |=========================================+========================+======================| | 0 NVIDIA GeForce RTX 4060 Ti Off | 00000000:01:00.0 On | N/A | | 0% 38C P8 10W / 160W | 446MiB / 8188MiB | 14% Default | | | | N/A |+-----------------------------------------+------------------------+----------------------++-----------------------------------------------------------------------------------------+ | Processes: | | GPU GI CI PID Type Process name GPU Memory | | ID ID Usage | |=========================================================================================| | 0 N/A N/A 1675 G /usr/lib/xorg/Xorg 294MiB | | 0 N/A N/A 1952 G /usr/bin/gnome-shell 97MiB |+-----------------------------------------------------------------------------------------+