bugfix: Fix stream handling in cutedsl gemm

[fzyzcjy]

📌 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 @fzyzcjy, 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!

This pull request addresses a critical bug in the cutedsl GEMM implementation by correcting the CUDA stream used for operations. Previously, the default stream was implicitly used, which could lead to incorrect execution contexts or synchronization issues. The change ensures that GEMM operations now correctly utilize the currently active CUDA stream, improving reliability and correctness.

Highlights

• CUDA Stream Correction: The core change involves updating the stream acquisition method in flashinfer/cute_dsl/blockscaled_gemm.py from cutlass_torch.default_stream() to cutlass_torch.current_stream(). This ensures that all subsequent CUDA operations within the GEMM kernel are properly synchronized and executed on the intended stream, resolving potential concurrency issues.

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. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

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 correctly fixes a critical issue in CUDA stream handling within the cutedsl GEMM implementation. By switching from cutlass_torch.default_stream() to cutlass_torch.current_stream(), the code now properly respects the user's active CUDA stream context. This is a crucial change that prevents potential race conditions and synchronization errors when using non-default streams. The fix is accurate and improves the robustness of the code.

[gemini-code-assist]

Using default_stream() here was a critical bug. It forces the operation onto the default CUDA stream, ignoring any stream context the user might have set (e.g., via with torch.cuda.stream(s):). This can lead to race conditions and incorrect execution order. The change to current_stream() correctly resolves this issue by respecting the active stream context.