Unable to Passthrough Nvidia T1000 to Linux VM

Kaac

New Member
May 7, 2023
2
0
1
Hello
I've been stuck on this issue for what feels like weeks now with no success. I will try provide as much detail as possible but please if needed ask questions where necessary!

The basis of the situation is

I recently build a Proxmox server with a Ryzen 7 5700x and Nvidia T1000 8GB and am attempting to passthrough my GPU to a linux (specifically ubuntu) Virtual machine. While I can get the GPU to show both in my proxmox shell and linux VM, when running "Nvidia-smi" on the VM I get the error of "no devices found" even though the drivers HAVE been successfully installed. Running "nvidia-settings" returns the error of "The control display is undefined; please run `nvidia-settings --help` for usage information."

There's a few reasons as to why I think this is not an error with the passthrough and something else.

1.
When running "lspci" inside of my Proxmox shell, the PCI device shows however seems to be missing lots of information regarding the model number. Inside my proxmox shell, lspci returns this when mentioning my GPU;


07:00.0 VGA compatible controller: NVIDIA Corporation Device 1ff0 (rev a1) (prog-if 00 [VGA controller]) Subsystem: Lenovo Device 1612 Flags: bus master, fast devsel, latency 0, IRQ 112, IOMMU group 16 Memory at fb000000 (32-bit, non-prefetchable) [size=16M] Memory at d0000000 (64-bit, prefetchable) [size=256M] Memory at e0000000 (64-bit, prefetchable) [size=32M] I/O ports at e000 [size=128] Expansion ROM at 000c0000 [disabled] [size=128K] Capabilities: [60] Power Management version 3 Capabilities: [68] MSI: Enable- Count=1/1 Maskable- 64bit+ Capabilities: [78] Express Legacy Endpoint, MSI 00 Capabilities: [100] Virtual Channel Capabilities: [258] L1 PM Substates Capabilities: [128] Power Budgeting <?> Capabilities: [420] Advanced Error Reporting Capabilities: [600] Vendor Specific Information: ID=0001 Rev=1 Len=024 <?> Capabilities: [900] Secondary PCI Express Capabilities: [bb0] Physical Resizable BAR Kernel driver in use: vfio-pci Kernel modules: nvidiafb, nouveau 07:00.1 Audio device: NVIDIA Corporation Device 10fa (rev a1) Subsystem: Lenovo Device 1612 Flags: bus master, fast devsel, latency 0, IRQ 113, IOMMU group 16 Memory at fc080000 (32-bit, non-prefetchable) [size=16K] Capabilities: [60] Power Management version 3 Capabilities: [68] MSI: Enable- Count=1/1 Maskable- 64bit+ Capabilities: [78] Express Endpoint, MSI 00 Capabilities: [100] Advanced Error Reporting Kernel driver in use: vfio-pci Kernel modules: snd_hda_intel


While when running the SAME command inside the Ubuntu VM, it seems to see the card much better;

00:10.0 VGA compatible controller: NVIDIA Corporation TU117GL [T1000 8GB] (rev a1) (prog-if 00 [VGA controller]) Subsystem: Lenovo Device 1612 Physical Slot: 16 Flags: bus master, fast devsel, latency 0, IRQ 11 Memory at fd000000 (32-bit, non-prefetchable) [size=16M] Memory at e0000000 (64-bit, prefetchable) [size=256M] Memory at f0000000 (64-bit, prefetchable) [size=32M] I/O ports at f000 [size=128] Expansion ROM at 000c0000 [virtual] [disabled] [size=128K] Capabilities: <access denied> Kernel driver in use: nvidia Kernel modules: nvidiafb, nouveau, nvidia_drm, nvidia 00:10.1 Audio device: NVIDIA Corporation Device 10fa (rev a1) Subsystem: Lenovo Device 1612 Physical Slot: 16 Flags: bus master, fast devsel, latency 0, IRQ 11 Memory at fe660000 (32-bit, non-prefetchable) [size=16K] Capabilities: <access denied> Kernel driver in use: snd_hda_intel Kernel modules: snd_hda_intel


2. I was reading somewhere that Debian 11 has problems specifically with this architecture of cards (no idea if true)


3. I've read many things on other posts such as this and this regarding the situation. But many return conflicting information regarding what to do.

I am rather certain my issue is not with passing through my card through Proxmox. IOMMU is enabled, as well as the drives being blacklisted, adding vfio etc. It seems to be a problem with proxmox not recognising my card.

Has anyone encountered any issue like this with a card before? Is it possible to pass it through with proxmox? I am pulling my hair out at this situation!!!!!

Any help would be BEYOND appreciated, very close to giving up on this whole build...


Build:

AMD Ryzen 7 5700X
Gigabyte B550I Aorus Pro AX
32GB ECC Memory
Nvidia T1000 8GB
4 x 10TB Seagate Ironwolf Pro's
2 x 256GB Boot NVMEs
 
Last edited:
EDIT:sent too quickly...

hi,
the PCI device shows however seems to be missing lots of information regarding the model number.
this is not really important, the db is just outdated on the base debian system. you can update this with

Code:
update-pciids

on the host

from this in the guest
Kernel driver in use: nvidia
it looks like it's correctly passed through

can you post the journal/syslog/dmesg from the guest? maybe we can see whats going on there
 
Last edited:

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!