Skip to content

Conversation

@msanft
Copy link
Member

@msanft msanft commented Dec 18, 2025

Since the gpu just deploy target relies on the resourcegen tool, the verdict made in 1 was false. The GPU target is only used in development, so hardcoding the platforms should be fine.

Since the `gpu` just deploy target relies on the resourcegen tool, the verdict made in [1] was false. The GPU target is only used in development, so hardcoding the platforms should be fine.

[1]: #2030 (comment)
@msanft msanft requested a review from burgerdev December 18, 2025 17:33
@msanft msanft added the no changelog PRs not listed in the release notes label Dec 18, 2025
@msanft msanft requested a review from katexochen as a code owner December 18, 2025 17:33
Copy link
Member

@burgerdev burgerdev left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The other commit lgtm. Maybe it makes sense to separate them?

Comment on lines +81 to +84
case platforms.MetalQEMUTDXGPU:
deviceURI = "nvidia.com/GB100_B200"
case platforms.MetalQEMUSNPGPU:
deviceURI = "nvidia.com/GH100_H100_PCIE"
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I still believe that this is not the right place to configure this. Rather, this should be modelled similar to the tcb-specs.json and be an input to resourcegen.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

How would that work with the just targets? Not saying that it isn't possible, I just wonder what the dataflow would look like.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

While not ideal, we could configure this with a justfile.env var for now, since we're already doing it for the nodeinstaller target config:

contrast/justfile

Lines 338 to 339 in fb4de83

sed -i 's/^default_platform=.*/default_platform="Metal-QEMU-TDX"/' justfile.env
sed -i 's/^node_installer_target_conf_type=.*/node_installer_target_conf_type="k3s"/' justfile.env

Eventually, I'd prefer this to be configured in the cluster instead of in the repo, though.

@msanft
Copy link
Member Author

msanft commented Dec 19, 2025

Maybe it makes sense to separate them?

Done in #2042. We should have that and #2035 merged ASAP so we have a working CI again with the new operator version.

@msanft msanft changed the title Various GPU fixes kuberesource: template GPU per platform Dec 19, 2025
Base automatically changed from msanft/nix/cleanups to main December 19, 2025 13:54
@msanft
Copy link
Member Author

msanft commented Jan 9, 2026

Obsolete with #2070

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

no changelog PRs not listed in the release notes

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants