Skip to content

Update to use CUDA runtime APIs that work in CUDA 12 and 13#2004

Merged
rapids-bot[bot] merged 4 commits intorapidsai:branch-25.10from
robertmaynard:support_cuda_13
Aug 4, 2025
Merged

Update to use CUDA runtime APIs that work in CUDA 12 and 13#2004
rapids-bot[bot] merged 4 commits intorapidsai:branch-25.10from
robertmaynard:support_cuda_13

Conversation

@robertmaynard
Copy link
Contributor

Description

CUDA 13 removes the cudaMemAdvise and cudaMemPrefetchAsync function signatures used by RMM. To continue to support CUDA 12 and 13 we move over to the v2 APIs when in CUDA 12, which are now the unversioned APIs in 13.

Checklist

  • I am familiar with the Contributing Guidelines.
  • New or existing tests cover these changes.
  • The documentation is up to date with these changes.

@robertmaynard
Copy link
Contributor Author

/merge

@robertmaynard
Copy link
Contributor Author

robertmaynard commented Aug 4, 2025

Working on changes so that we use a version of cudaMemPrefetchAsync that works with the CUDA 12.0 runtime/driver

@rapids-bot rapids-bot bot merged commit b6bb091 into rapidsai:branch-25.10 Aug 4, 2025
50 checks passed
@robertmaynard robertmaynard deleted the support_cuda_13 branch August 4, 2025 16:57
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

feature request New feature or request non-breaking Non-breaking change

Projects

Status: Done

Development

Successfully merging this pull request may close these issues.

2 participants