Greetings all,
Im not sure where to put this and I could be mistaken but I recently setup my machine for GPU Passthrough and the documentation here mentions kvm_intel but I believe if you’re pure team red (other than the card(s) you are passing through) you would want kvm_amd no? At least that’s how I have things setup and appear to be running OK.
If I haven’t made a mistake - I don’t know if its worth looking to update the documentation is all.
@Gorship Hi, well it defines the purpose of the document (A.2), this section is for computational gpu setup with Nvidia GPU’s in vm’s.
And it does state for AMD or Intel, perhaps there is some expectation that the reading audience understands this part already?
"This article describes how to assign an NVIDIA GPU graphics card on the host machine to a virtualized guest. "
Which is still being done as a full amd system for the host and a secondary gpu being passed to the VM.
if there is a secondary document for amd systems that I have not seen then that’s fine - was just a suggestion to add a quick line or a note about “kvm_amd” . As there is the only one section in virtualization for GPU passthrough that I can see.
edit: if its a bad idea then fine - just would have saved me some time and perhaps others.
@Gorship A.2 states “This article deals with a set of instructions based on V100/T1000 NVIDIA cards, and is meant for GPU computation purposes only.”
You can make the change and then push your edit?
https://en.opensuse.org/Portal:Documentation
https://github.com/openSUSE/doc-o-o
Ill take a look, I’m a touch hesitant as I’m hardly a github expert.
However I may recognizing your point now and seeing its a specific use case being outlined and not a generic methodology.
@Gorship You also just create an issue for clarification of the section?