A basic GPU question

Incus
,
1

Hello,

Here is my gpu information and want to add that device but I havent done it.

GPU:NUMA node: 0
Driver: panthor (6.18.0-edge-rockchip64)
DRM:
ID: 0
Card: card0 (226:0)
Render: renderD128 (226:128)

I have tried, ```incus config device add llamacpp agpu gpu gputype=mdev=0```

Error: Failed to start device “agpu”: Failed to detect requested GPU device
Try incus info --show-log llamacpp for more info

But can start the device, anyone has idea how to add that kind of gpu to the vm?

2

Is that GPU its own PCI device?
If not, you won’t be able to pass it to a VM.

The output above suggests that this is an embedded GPU on an ARM board, so very unlikely to be connected internally over PCI.

3

Humm, thanks.

Regards.