Skip to content

Conversation

@Alexey-Rivkin
Copy link
Contributor

@Alexey-Rivkin Alexey-Rivkin commented Nov 5, 2025

What?

Upgrade CI to CUDA 13.0 for GPU testing.

Why?

Align with newer CUDA releases for CUDA 13.x support.

How?

  • Updates Docker images: Ubuntu 24.04 to 25.10-cuda13.0, Ubuntu 22.04 to nvidia/cuda:13.0.1
  • Extends GPUNETIO plugin support to CUDA 13.x (meson.build)
  • Switches Python default from nixl-cu12 to nixl-cu13
  • Updates CI configs, test scripts, and documentation

Also fix POSIX shell compatibility in common.sh to prevent OOM.

@github-actions
Copy link

github-actions bot commented Nov 5, 2025

👋 Hi Alexey-Rivkin! Thank you for contributing to ai-dynamo/nixl.

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

🚀

@Alexey-Rivkin
Copy link
Contributor Author

/build

@Alexey-Rivkin Alexey-Rivkin changed the title CI: GPU test on Ubuntu22 with CUDA13 CI: Add GPU test on Ubuntu22 with CUDA13 Nov 5, 2025
@Alexey-Rivkin
Copy link
Contributor Author

/build

@Alexey-Rivkin
Copy link
Contributor Author

/build

Some tests (e.g. gpunetio) only run on specific CUDA ver.
Adding both CUDA 12 and 13 improves coverage.

Signed-off-by: Alexey Rivkin <[email protected]>
@Alexey-Rivkin
Copy link
Contributor Author

/build

- **Operating System**: Ubuntu 22.04/24.04 LTS (recommended) or RHEL-based
- **Docker**: Version 20.10+ (for container builds)
- **Git**: For source code management
- **CUDA Toolkit**: 12.8+ (for GPU features)
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Not sure if we want to document 13.0 as a hard requirement already @aranadive WDYT?

@brminich
Copy link
Contributor

/build

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.

5 participants