[diffusion]: Fix empty memory_checkpoints in CLI perf dump

[Ratish1]

Motivation

Fix sglang generate --perf-dump-path writing empty memory_checkpoints in the output JSON. For example: "memory_checkpoints": {}, #18928

cc: @mickqian

Modifications

• In maybe_dump_performance, copy timings["memory_snapshots"] into RequestMetrics.memory_snapshots.

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.

Review Process

1. Ping Merge Oncalls to start the PR flow. See the PR Merge Process.
2. Get approvals from CODEOWNERS and other reviewers.
3. Trigger CI tests with comments or contact authorized users to do so.
   • /tag-run-ci-label, /rerun-failed-ci, /tag-and-rerun-ci
4. After green CI and required approvals, ask Merge Oncalls to merge.

[gemini-code-assist]

Summary of Changes

Hello @Ratish1, 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 CLI performance dumping mechanism. Previously, when generating performance reports, the memory checkpoint section was left empty, leading to incomplete diagnostic data. The changes ensure that all relevant memory usage statistics are accurately captured and included in the performance dump, providing a more comprehensive view of resource utilization during execution.

Highlights

• Bug Fix: Resolved an issue where the sglang generate --perf-dump-path command was writing empty memory_checkpoints in the output JSON performance dump.
• Performance Data Accuracy: Ensured that memory snapshot data, including allocated, reserved, and peak memory usage, is correctly copied from timings into RequestMetrics.memory_snapshots within the maybe_dump_performance function.

🧠 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.

Changelog

• python/sglang/multimodal_gen/runtime/entrypoints/cli/generate.py
  • Imported the MemorySnapshot class from sglang.multimodal_gen.runtime.utils.perf_logger.
  • Added a loop to iterate through memory_snapshots in the timings_dict and populate the timings.memory_snapshots attribute of the RequestMetrics object with MemorySnapshot instances, ensuring memory data is correctly transferred.

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

The pull request effectively resolves the issue where memory_checkpoints were empty in the CLI performance dump. The changes correctly parse the dictionary representation of memory snapshots from the timings_dict and reconstruct them into MemorySnapshot objects within the RequestMetrics instance, ensuring that performance metrics are accurately captured and reported.