feat: add pdl for trtllm-gen attn

[yyihuang]

📌 Description

Following #1446.

🔍 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!

This pull request introduces support for Programmatic Dependent Launch (PDL) within the TensorRT-LLM generated attention kernels. This enhancement allows for finer-grained control over kernel launch dependencies, potentially improving performance and resource utilization for attention operations, especially on newer NVIDIA architectures. The changes involve modifying the underlying CUDA kernel interfaces, updating the Python API wrappers for both decode and prefill operations, and adding robust test coverage for the new functionality.

Highlights

• Core Kernel Parameter: The core CUDA kernel launchers for TensorRT-LLM paged attention (trtllm_paged_attention_launcher, trtllm_paged_attention_decode, and trtllm_paged_attention_context) now accept an enable_pdl boolean parameter.
• Python API Integration: The enable_pdl parameter has been integrated into the Python wrappers for both decode (flashinfer/decode.py) and prefill (flashinfer/prefill.py) attention functions, allowing users to control this behavior from the Python API.
• Automatic PDL Detection: Logic has been added to automatically detect if the current device supports Programmatic Dependent Launch (PDL) via device_support_pdl. If the enable_pdl parameter is not explicitly set, it will default based on this detection.
• Expanded Test Coverage: Comprehensive unit tests have been expanded to cover various scenarios for the new enable_pdl parameter, including True, False, and None values, ensuring robust functionality across different configurations.

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 adds support for Programmatic Dependent Launch (PDL) to the trtllm-gen attention kernels. The changes involve plumbing an enable_pdl flag through various function layers, from Python wrappers down to the C++/CUDA implementation. The implementation appears correct and consistent, and the tests have been updated to cover the new functionality. My feedback focuses on improving the documentation for the new enable_pdl parameter to ensure clarity and accuracy.