Hyper-V P2000 GPU Passthrough to Ubuntu - nvidia-smi returns 'device not found'


I’ve been trying for days to get GPU passthrough to work on my system. This is the closest I have come so far but this last piece has me stumped.

I’m Running Ubuntu 22.04 inside Hyper-V on Windows 2022 Server with a Quadro P2000. Using various resources including Setting up Discrete Device Assignment with a GPU I managed to unmount the GPU on the host and get it showing in the VM:

I was able to install the latest NVIDIA-Linux-x86_64-550.54.14.run driver and now lspci-v returns:

d40b:00:00.0 VGA compatible controller: NVIDIA Corporation GP106GL [Quadro P2000] (rev a1) (prog-if 00 [VGA controller])
        Subsystem: NVIDIA Corporation GP106GL [Quadro P2000]
        Physical Slot: 18037621
        Flags: bus master, fast devsel, latency 0, NUMA node 0
        Memory at 41000000 (32-bit, non-prefetchable) [size=16M]
        Memory at fe0000000 (64-bit, prefetchable) [size=256M]
        Memory at ff0000000 (64-bit, prefetchable) [size=32M]
        Expansion ROM at 000c0000 [virtual] [disabled] [size=128K]
        Capabilities: <access denied>
        Kernel driver in use: nvidia
        Kernel modules: nvidiafb, nouveau, nvidia_drm, nvidia

I’m attaching a bug report in the hopes that someone more knowledgeable than me can make sense of it.

Thanks so much.

nvidia-bug-report.log.gz (90.0 KB)