ci: use cuda from pip #63
Merged
+1
−1
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
In the GPU CI, we're currently using CUDA from the
nvidia/cuda:12.2.2-cudnn8-devel-ubuntu22.04Docker image. JAX tends to stay pretty bleeding-edge in terms of CUDA requirements though, so our choices are to manually update the image each time the installation breaks, or we just install the CUDA binaries from pip. The latter feels more robust to me, so we'll do that for now, but the tradeoff is that the CI takes an extra ~1 minute.Also, that this means we're compiling jax-finufft against CUDA 12.2 and running it against the 12.3 libraries, but CUDA has good forward-compatibility guarantees so this shouldn't be a problem.