support add_gemma_rms_norm

[RuixuanZhang06]

No description provided.

[gemini-code-assist]

Summary of Changes

Hello @RuixuanZhang06, 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 enhances the sgl_kernel_npu library by integrating Gemma RMS Normalization. It provides an optimized Triton kernel that efficiently handles the combined operation of adding a residual to a hidden state and then applying root mean square normalization, specifically tailored for NPU acceleration. This addition is crucial for supporting models that utilize Gemma's normalization scheme.

Highlights

• New Gemma RMS Norm Support: Introduced add_gemma_rms_norm_kernel and its Python wrapper add_gemma_rms_norm to support Gemma-specific RMS Normalization.
• Triton Kernel for NPU: The new normalization logic is implemented as a Triton JIT kernel, add_gemma_rms_norm_kernel, designed for efficient execution on NPU devices.
• Combined Add and Norm Operation: The kernel performs an element-wise addition of hidden_state and residual before applying the RMS normalization, returning both the normalized output and the intermediate added value.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

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 by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

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 pull request 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 introduces a new Triton kernel add_gemma_rms_norm to support RMS normalization as used in Gemma models. The kernel implementation appears functionally correct. However, the changes are missing crucial unit tests to validate the new functionality. Additionally, both the kernel and its wrapper function lack docstrings, which impacts code readability and maintainability. I've also pointed out a magic number that should be a named constant and a minor style issue. Addressing these points will significantly improve the quality of this contribution.