fix get max_q_len in page prefill plan

[ZhuJiaqi9905]

📌 Description

the plan method in BatchPrefillWithPagedKVCacheWrapper will compute max query len by using qo_indptr_host. However, previous code only get the maximum value of qo_indptr, which is incorrect.

[gemini-code-assist]

Summary of Changes

Hello @ZhuJiaqi9905, 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 BatchPrefillWithPagedKVCacheWrapper's plan method, ensuring that the _max_q_len is accurately determined. The fix prevents potential miscalculations in query length, which is crucial for the correct operation of the prefill plan, thereby improving the reliability of the system's batch processing capabilities.

Highlights

• Corrected max_q_len calculation: The method for calculating _max_q_len in BatchPrefillWithPagedKVCacheWrapper has been fixed. Previously, it incorrectly used the maximum value of qo_indptr_host; now, it correctly computes the maximum query length by taking the maximum difference between consecutive elements of qo_indptr_host.

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 correctly fixes the calculation of max_q_len by using the differences in qo_indptr rather than its maximum value. This is a good correction. I've added one comment regarding an edge case for empty batches that could lead to a runtime error.

[gemini-code-assist]

This line correctly calculates the maximum query length. However, it can raise a RuntimeError if qo_indptr_host contains only one element (i.e., when batch_size is 0), because max() would be called on an empty tensor. This can be handled by checking if batch_size > 0.

self._max_q_len = (qo_indptr_host[1:] - qo_indptr_host[:-1]).max().item() if batch_size > 0 else 0