Alex Lyn
Alex Lyn
Hi @pennants Thx for your reporting this issue. Could you please give us more detailed information about it with recommended question format ? I think detailed info will help people...
> I'm uncertain when will #9868 be merged, so I've created this PR to initiate the review process earlier. > > @justxuewei could you please take a look at the...
> > > I'm uncertain when will #9868 be merged, so I've created this PR to initiate the review process earlier. > > > @justxuewei could you please take a...
@zvonkok FYI ^^
> This will only work for hot-plugged devices. For Confidential Containers we also need cold-plugging support. How will this work with direct attached devices? > Thx @zvonkok AFAIK, in Kata,...
@zvonkok FYI [Waiting...]
> This will only work for hot-plugged devices. For Confidential Containers we also need cold-plugging support. How will this work with direct attached devices? > @zvonkok If we combine the...
> Can we add a test for this please? > Sure, I will add tests for it later
> Can we add a test for this please? > tests added !
The test will showcase that there're two NV GPUs passed through the kata pod named ollam-0 for running ollama workload. CAUTION: it MUST be applied with several PRs and kata...