Does the current script support cc on H200s ?

[jorgeantonio21]

I am currently trying to set up a node with 8xH200s with NVIDIA + TDX cc, following the official docs:

https://docs.nvidia.com/cc-deployment-guide-tdx.pdf

However, I can't make it work past the OS host administrator part II, as the system does not recognize any of the GPU devices:

$ sudo python3 ./nvidia_gpu_tools.py --gpu-name=H100 --query-cc-mode

[moconnor725]

What does "lspci -d 10de:" return?

[pjaroszynski]

If it's H200 then the name parameter needs to match --gpu-name H200. But that doesn't really scale to 8 GPUs anyway, you want something like for g in $(seq 0 7); do sudo python3 ./nvidia_gpu_tools.py --gpu $g --query-cc-mode; done etc.

[jorgeantonio21]

@moconnor725 this is what "lspci -d 10de:" return:

$ lspci -d 10de:
19:00.0 3D controller: NVIDIA Corporation GH100 [H200 SXM 141GB] (rev a1)
3b:00.0 3D controller: NVIDIA Corporation GH100 [H200 SXM 141GB] (rev a1)
4c:00.0 3D controller: NVIDIA Corporation GH100 [H200 SXM 141GB] (rev a1)
5d:00.0 3D controller: NVIDIA Corporation GH100 [H200 SXM 141GB] (rev a1)
83:00.0 Bridge: NVIDIA Corporation GH100 [H100 NVSwitch] (rev a1)
84:00.0 Bridge: NVIDIA Corporation GH100 [H100 NVSwitch] (rev a1)
85:00.0 Bridge: NVIDIA Corporation GH100 [H100 NVSwitch] (rev a1)
86:00.0 Bridge: NVIDIA Corporation GH100 [H100 NVSwitch] (rev a1)
9b:00.0 3D controller: NVIDIA Corporation GH100 [H200 SXM 141GB] (rev a1)
bb:00.0 3D controller: NVIDIA Corporation GH100 [H200 SXM 141GB] (rev a1)
cb:00.0 3D controller: NVIDIA Corporation GH100 [H200 SXM 141GB] (rev a1)
db:00.0 3D controller: NVIDIA Corporation GH100 [H200 SXM 141GB] (rev a1)