[diffusion] fix: respect dist_timeout option

[mickqian]

fixes #15792

Motivation

Respect --dist-timeout and set the default timeout to 3600 seconds

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.

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 @mickqian, 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 robustness of distributed operations by allowing users to configure the timeout for torch.distributed processes. It integrates a new dist_timeout option across the distributed environment initialization, ensuring that the specified timeout is respected. A default timeout of one hour is now applied, which should help prevent premature disconnections in idle distributed setups, and the command-line interface documentation has been updated for clarity.

Highlights

• Configurable Distributed Timeout: Introduced a new dist_timeout parameter to allow explicit configuration of the timeout for torch.distributed operations.
• Default Timeout Value: Set a default dist_timeout of 3600 seconds (1 hour) in ServerArgs to mitigate 'Connection closed by peer' errors during periods of inactivity.
• Improved Documentation: Updated the help message for the --dist-timeout CLI argument to provide clearer guidance on its purpose and when to adjust it.

🧠 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/distributed/parallel_state.py
  • Added a timeout parameter to init_distributed_environment to configure the torch.distributed.init_process_group timeout.
  • Modified maybe_init_distributed_environment_and_model_parallel to accept and pass a dist_timeout parameter to the underlying distributed initialization function.
  • Updated logging statements to include the timeout value for better debugging.
• python/sglang/multimodal_gen/runtime/managers/gpu_worker.py
  • Modified init_device_and_model to pass the self.server_args.dist_timeout to the distributed environment initialization.
• python/sglang/multimodal_gen/runtime/server_args.py
  • Changed the default value of dist_timeout from None to 3600 seconds (1 hour).
  • Updated the help message for the --dist-timeout CLI argument to explain its purpose in preventing 'Connection closed by peer' errors and its default value.

Activity

• No specific activity (comments, reviews, or progress updates) has been recorded for this pull request yet.

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. ↩

[mickqian]

/tag-and-rerun-ci

[gemini-code-assist]

Code Review

This pull request introduces a dist_timeout option to configure the timeout for torch.distributed operations. This is a useful feature to prevent 'Connection closed by peer' errors in services that might be idle for long periods. The changes are well-implemented, propagating the new parameter from server arguments down to the distributed initialization. I have one minor suggestion regarding code style to improve maintainability.

[gemini-code-assist]

According to PEP 8, imports should be at the top of the file. For better code style and to avoid repeated imports, it's recommended to move this import datetime statement to the top of the file with other imports. You can remove these lines and add import datetime at the top of the file.

References

1. PEP 8 recommends that imports are always put at the top of the file, just after any module comments and docstrings, and before module globals and constants. This improves readability and makes it easier to track dependencies. ^(link)