test: add cuda graph to comm test

[yyihuang]

📌 Description

🔍 Related Issues

🚀 Pull Request Checklist

Thank you for contributing to FlashInfer! Before we review your pull request, please make sure the following items are complete.

✅ Pre-commit Checks

• I have installed pre-commit by running pip install pre-commit (or used your preferred method).
• I have installed the hooks with pre-commit install.
• I have run the hooks manually with pre-commit run --all-files and fixed any reported issues.

If you are unsure about how to set up pre-commit, see the pre-commit documentation.

🧪 Tests

• Tests have been added or updated as needed.
• All tests are passing (unittest, etc.).

Reviewer Notes

[gemini-code-assist]

Summary of Changes

Hello @yyihuang, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

I've updated the communication test suite to include CUDA graph functionality for the trtllm_allreduce_fusion operation. This enhancement aims to validate the correctness and performance of this critical fusion pattern when executed within a CUDA graph, which is essential for optimizing deep learning workloads. The changes involve restructuring the test flow to allow for graph capture, including a warmup phase, and then replaying the captured graph to verify results.

Highlights

• CUDA Graph Integration: I've modified the _run_correctness_worker function to incorporate CUDA graph capture and replay. This change allows us to test the trtllm_allreduce_fusion operation within a CUDA graph, ensuring its compatibility and performance benefits in such an environment.
• Warmup Phase for CUDA Graph: To properly utilize CUDA graphs, I've introduced a 'warmup' phase before the graph capture. This ensures that all necessary CUDA kernels are loaded and initialized, preventing any first-run overhead from affecting the captured graph's performance.
• Test Loop Refactoring: The test logic has been refactored to move tensor initialization outside the main test loop. This is a prerequisite for effective CUDA graph capture, as graph-captured operations typically require static memory allocations and consistent input shapes.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point in your pull request via creating an issue comment (i.e. comment on the pull request page) using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in issue comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments or fill out our survey to provide feedback.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

This pull request adds CUDA graph testing to the trtllm_allreduce_fusion communication test. The overall approach is sound, but there's significant code duplication between the warmup and graph capture sections. I've provided a suggestion to refactor this by extracting the duplicated logic into a helper function, which will improve the code's maintainability.