Skip to content

Conversation

@michal-shalev
Copy link
Contributor

@michal-shalev michal-shalev commented Sep 22, 2025

What?

Replace two-step GPU transfer request creation with a single createGpuXferReq API that takes descriptor lists directly and returns both transfer request and GPU request handles.

Why?

Simplify the API by reducing GPU transfer request creation from two separate function calls to a single call.

How?

Updated createGpuXferReq to substitute calls to createXferReq + createGpuXferReq. Removed the old createGpuXferReq method and updated tests accordingly.

@github-actions
Copy link

👋 Hi michal-shalev! Thank you for contributing to ai-dynamo/nixl.

Your PR reviewers will review your contribution then trigger the CI to test your changes.

🚀

Signed-off-by: Michal Shalev <mshalev@nvidia.com>
@michal-shalev michal-shalev force-pushed the nixl-single-step-gpu-xfer-creation branch from 58ce4b6 to d5060d0 Compare September 22, 2025 11:34
@michal-shalev michal-shalev changed the title DEVICE/API: Simplify createGpuXferReq to single-step API DEVICE/API: Simplify createGpuXferReq to single-step API - WIP Oct 12, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant