fix: harden setup.py CUDA build guards and wheel fallback messaging - #1004
Open
Chessing234 wants to merge 3 commits into
Open
fix: harden setup.py CUDA build guards and wheel fallback messaging#1004Chessing234 wants to merge 3 commits into
Chessing234 wants to merge 3 commits into
Conversation
MAMBA_KEEP_CUDA_BUILD referenced bare_metal_version even when nvcc was unavailable, causing UnboundLocalError. Fall back to torch.version.cuda for arch flags and only compare CUDA majors when nvcc is present.
get_wheel_url() called parse(None) when torch.version.cuda was unset, producing an opaque packaging error. Raise a direct message instead.
Extend the CUDA_HOME warning so cloud notebook users know missing nvcc is expected unless they explicitly requested selective_scan_cuda.
Chessing234
force-pushed
the
fix/setup-cuda-home-guards
branch
from
August 5, 2026 10:24
e76cde4 to
fafe5fb
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
UnboundLocalErrorwhenMAMBA_KEEP_CUDA_BUILD=TRUEbutCUDA_HOME/nvcc is missing by falling back totorch.version.cudafor arch flags.packagingerrors.selective_scan_cudasource builds are slow and optional for Mamba-2/3.Test plan
CUDA_HOME is Noneandtorch.version.cuda is NonepathsMade with Cursor