[UCX] Bump Ver + Add CUDA#13039
Merged
Merged
Conversation
Collaborator
Author
Contributor
I would double check with AMDGPU.jl folks. As I thought UCX.jl might be leveraged in a backend for Distributed.jl (here). Maybe @jpsamaroo knows updates or if this is a good idea. |
Merged
Collaborator
Author
|
@giordano Can we merge this? No one else is actively using UCX. Distributed nor AMDGPU do and UCX.jl has not been updated in 5 years. If updates need to be made down the road for this recipe to work for others then we can make those changes then. |
giordano
reviewed
Feb 4, 2026
Collaborator
Author
|
@giordano fixed! |
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Updates to 1.20 and updates the script to use more modern binary builder functionality.
I have removed ROCm support. I do not think anyone was using that.
Technically, I think all of the CUDA things could be build dependencies but this is just easier for now. I want something that for sure works. In pricniple UCX is smart enough to be built with everything enabled and silently turn off what it cannot find at runtime.
Figuring this out was too much of a headache right now and I'm pretty sure what I've done here should be good enough. There are builds for each CUDA (x86 only) as well as an ARM and x86 CPU-only build.