Skip to content

Fix for CUDA 13.0 and WSL builds#825

Open
ademeure wants to merge 2 commits into
karpathy:masterfrom
ademeure:cuda13_fix
Open

Fix for CUDA 13.0 and WSL builds#825
ademeure wants to merge 2 commits into
karpathy:masterfrom
ademeure:cuda13_fix

Conversation

@ademeure

Copy link
Copy Markdown
Contributor

Support CUDA 13.0 by replacing deprecated cudaMemAdvise function arguments.

Fix default build for WSL by disabling Multi-GPU if there is only 1 GPU present (this is faster for other platforms with a single GPU as well since it avoids the overhead of redundant NCCL calls).

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant