Skip to content

Conversation

@romerojosh
Copy link
Collaborator

This PR makes a few changes to enable compilation against CUDA 13.0:

  • Unversioned driver function pointers are no longer available so I've updated to using versioned symbols. Additionally, the cudaGetDriverEntryPoint API is now marked deprecated so I've updated to using cudaGetDriverEntryPointByVersion.
  • The CCCL include structure within the CUDA Toolkit has changed (see: https://github.com/NVIDIA/cccl/releases/tag/v3.0.0). Updating CMakeLists.txt to reflect these changes.
  • CUDA 13 removes support for Volta generation GPUs. Reming CC70 from the default CC list when compiling with CUDA 13+.

@romerojosh
Copy link
Collaborator Author

/build

@github-actions
Copy link

🚀 Build workflow triggered! View run

@github-actions
Copy link

❌ Build workflow failed! View run

@romerojosh
Copy link
Collaborator Author

/build

@github-actions
Copy link

🚀 Build workflow triggered! View run

@github-actions
Copy link

✅ Build workflow passed! View run

@romerojosh
Copy link
Collaborator Author

/build

@romerojosh
Copy link
Collaborator Author

/build

@github-actions
Copy link

🚀 Build workflow triggered! View run

@github-actions
Copy link

✅ Build workflow passed! View run

@romerojosh
Copy link
Collaborator Author

/build

@github-actions
Copy link

github-actions bot commented Oct 6, 2025

🚀 Build workflow triggered! View run

@github-actions
Copy link

github-actions bot commented Oct 6, 2025

✅ Build workflow passed! View run

@romerojosh romerojosh merged commit 9c24af4 into main Oct 8, 2025
4 checks passed
@romerojosh romerojosh deleted the cuda_13_updates branch October 8, 2025 16:23
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.

2 participants