cuda: dropping some cuda deps for 0.19.0#3
Open
d-goldin wants to merge 10 commits into
Open
Conversation
Co-authored-by: Luna Nova <git@lunnova.dev>
- Bumping triton to a newer version, the older one didn't work for me with 0.17 - Drops quarck-kernels and cuteDSL from dependencies. From what I can tell those are only needed for FA4 and would also require some nvidia blobs. We are at FA2 right now, so this shouldn't remove any functionality that was present before - Adding NCCL to wrapper args, for better UX
From what I can tell nvidia-cudnn-frontend, was just added to pin transitive dependnecies of flashinfer which doesn't affect us in nixpkgs right now. Cubins was added to get some AOT compiled kernels, primarily for blackwell. JIT should still work fine and we can with time update the flashinfer package as well. See: - vllm-project/vllm#36719 - vllm-project/vllm#37233
13 tasks
CertainLach
force-pushed
the
push-lklxouywkrnv
branch
6 times, most recently
from
dc73532 to
ba2de00
Compare
This file contains hidden or 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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
From what I can tell nvidia-cudnn-frontend, was just added to pin transitive dependnecies of flashinfer which doesn't affect us in nixpkgs right now.
Cubins was added to get some AOT compiled kernels, primarily for blackwell. JIT should still work fine and we can with time update the flashinfer package as well.
See: