Fix overlap scheduler not take effect when outputing logprobs

[fzyzcjy]

Motivation

about +30% speed

Modifications

Accuracy Tests

Benchmarking and Profiling

Checklist

• Format your code according to the Format code with pre-commit.
• Add unit tests according to the Run and add unit tests.
• Update documentation according to Write documentations.
• Provide accuracy and speed benchmark results according to Test the accuracy and Benchmark the speed.
• Follow the SGLang code style guidance.
• Work with maintainers to merge your PR. See the PR Merge Process

[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 resolves a performance bottleneck in the overlap scheduler that occurred specifically when log probabilities were being outputted. By ensuring the correct logprob tensors are copied to the CPU and properly integrating the return_logprob flag throughout the batch processing, the system can now leverage the overlap scheduler effectively, resulting in a significant speedup for such operations.

Highlights

• Corrected Logprob Handling: Ensured that the copy_to_cpu method correctly processes and transfers next_token_logprobs to the CPU when return_logprob is enabled, rather than next_token_logits.
• Parameter Enforcement: Modified the copy_to_cpu method signature to explicitly require the return_logprob boolean parameter, removing its default value.
• Scheduler Integration: Updated the run_batch and launch_batch_sample_if_needed functions in the scheduler to pass the return_logprob flag from the current batch to the copy_to_cpu method.
• Performance Improvement: This fix addresses an issue where the overlap scheduler was not effective when outputting logprobs, leading to an estimated 30% speed improvement.

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. ↩

[fzyzcjy]

/tag-and-rerun-ci

[gemini-code-assist]

Code Review

This pull request aims to fix an issue where the overlap scheduler was not effective when log probabilities were requested, which is a good performance enhancement. The changes correctly pass the return_logprob flag to the copy_to_cpu method and make this parameter mandatory, improving code clarity and correctness. However, the implementation of copy_to_cpu is incomplete, as it fails to copy all necessary logprob-related tensors. I've provided a suggestion to address this, which will ensure the feature works correctly and robustly.