[sgl-kernel] Opt per_token_quant_fp8 with warp reduce #8130
Conversation
yuan-luo
requested review from
zhyncs,
ispobock,
HandH1998,
BBuf,
yizhang2077,
merrymercy,
yinfan98 and
HaiShaw
as code owners
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Summary of Changes
Hello @yuan-luo, 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 focuses on a significant performance optimization for the per_token_quant_fp8 CUDA kernel, which is critical for efficient FP8 quantization in the SGLang framework. By refactoring the kernel to utilize advanced CUDA features like warp-level parallelism and the cute tensor library, it achieves notable speed improvements across various batch and sequence lengths.
Highlights
- Performance Optimization: I've re-implemented the
per_token_quant_fp8_kernelto leverage warp-level reductions andcutetensors for improved efficiency, resulting in up to 10% speedup as demonstrated by the provided benchmarks. - Kernel Architecture Refinement: The kernel now transitions from block-level reduction (
cub::block::block_reduce) to warp-level reduction (warpReduceMax) for calculating the maximum value per token, enabling more granular and efficient parallelization. - CUDA
cuteTensor Integration: I've adopted thecutetensor library for managing global memory access within the kernel, providing a more expressive and potentially optimized way to handle tensor operations. - Optimized Kernel Launch Configuration: The kernel launch parameters have been adjusted to align with the new warp-centric design, processing multiple tokens per CTA (specifically, 8 tokens per 256-thread CTA).
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 is currently in preview and 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 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
-
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. ↩
There was a problem hiding this comment.
Code Review
This pull request refactors the per_token_quant_fp8 CUDA kernel to improve performance by processing multiple tokens per thread block, with each token handled by a single warp. This is a solid optimization strategy. The review identified a critical race condition in the use of shared memory for the scaling factor, which would lead to incorrect quantization results. A medium-severity issue with an unnecessary const_cast was also found.
yuan-luo
force-pushed
the
per_token_quant_fp8
branch
from
7d3e37f to
8974703
Compare
Motivation
Optimize per_token_quant_fp8 kernel with warp reduce and cute tensor. Obtained 5% speedup in large seq len.
Modifications
Checklist