bazel-contrib / bazel-contrib/rules_cuda
cuda.toolkit's "name" should be respected
- Dominant language
- Starlark
- Stars
- 122
- Forks
- 73
- Avg merge
- 7h 34m
- Merged PRs (30d)
- 7
Description
### Observed Behavior
You can see here that `cuda.toolkit` accepts a `name` attribute but doesn't generate any name other than the one that's set explicitly in this project (used to be "local_cuda", now "cuda").
https://github.com/bazel-contrib/rules_cuda/blob/5da94965be9f145a22e0a2a8ead4b99a345b8471/MODULE.bazel#L10-L15
### Expected Behavior
This should be a field that accepts the name and creates the cuda repo in that name.
### Why it matters now?
A while ago, I filed #338 and thank you so much for fixing it.
When trying to not live on my fork anymore in our repo and use #360 after the merge here, since we use a submodule setup, we are stuck from making an atomic change to HEAD.
### Note
I understand it's not as trivial as just removing the hardcoded names to the attr, do you have some suggestions?
Contributor guide
Research direction
Start with MODULE.bazel lines 10-15 and trace how cuda.toolkit's name attribute is handled when the CUDA repository is generated. Done means the requested name is used instead of the project's hardcoded name, with the existing submodule setup still supported.
Written by the indexing model from the issue text.
Assessment
- Domain
- build-system
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 35/100