[diffusion] chore: revise process title

[mickqian]

after this PR, the setting of --num-gpus=4 --ulysses-degree=2 --enable-cfg-parallel would look like this in nvidia-smi:

|    3   N/A  N/A         3356502      C   sgl_diffusion::scheduler_U0_C0         6148MiB |
|    4   N/A  N/A         3356503      C   sgl_diffusion::scheduler_U1_C0         6148MiB |
|    5   N/A  N/A         3356504      C   sgl_diffusion::scheduler_U0_C1         6148MiB |
|    6   N/A  N/A         3356505      C   sgl_diffusion::scheduler_U1_C1         6148MiB |

Motivation

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 clarity and debuggability of GPU worker processes within the diffusion module by revising their naming convention. The change ensures that process titles are set after the distributed environment is fully initialized and incorporates detailed information about the active model parallel groups, allowing for easier identification and monitoring of individual processes in a distributed setup.

Highlights

• Process Title Refactoring: The logic for setting the process title (setproctitle) in gpu_worker.py has been refactored and moved to occur after the distributed environment initialization.
• Dynamic Process Naming: The process title now dynamically includes suffixes based on various initialized model parallel groups (Tensor Parallel, Ulysses Parallel, Ring Parallel, and Classifier-Free Guidance Parallel) and their respective ranks, providing more descriptive process identification.
• Distributed State Imports: New imports for distributed parallel state functions (e.g., get_tp_rank, get_ulysses_parallel_rank, get_classifier_free_guidance_rank) have been added to support the dynamic process naming.

🧠 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/managers/gpu_worker.py
  • Added imports for distributed parallel state functions such as get_tp_rank, get_tp_world_size, model_parallel_is_initialized, get_classifier_free_guidance_rank, get_classifier_free_guidance_world_size, get_ring_parallel_rank, get_ring_parallel_world_size, get_ulysses_parallel_rank, and get_ulysses_parallel_world_size.
  • Removed the setproctitle call from its original position before distributed environment initialization.
  • Re-inserted the setproctitle call after the maybe_init_distributed_environment_and_model_parallel function call.
  • Modified the setproctitle logic to construct a dynamic suffix for the process name, incorporating ranks from Tensor Parallel (TP), Ulysses Parallel (U), Ring Parallel (R), and Classifier-Free Guidance Parallel (C) if model parallelism is initialized.
  • Implemented a fallback for the process name to sgl_diffusion::scheduler_{self.local_rank} if model parallelism is not initialized.

Activity

• No human activity (comments, reviews) has been recorded on 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 revises how the process title is set for diffusion workers, providing more detailed information about the parallelization strategy being used. The change is a good improvement for observability. I've suggested a small refactoring to make the new title-building logic more concise and maintainable.

[gemini-code-assist]

The code for building the process title suffix is a bit repetitive. You can make it more concise and maintainable by using a data-driven approach with a list comprehension. This avoids repeating the if world_size != 1 check and string concatenation logic.

            parallel_info = [
                ("TP", get_tp_world_size, get_tp_rank),
                ("U", get_ulysses_parallel_world_size, get_ulysses_parallel_rank),
                ("R", get_ring_parallel_world_size, get_ring_parallel_rank),
                ("C", get_classifier_free_guidance_world_size, get_classifier_free_guidance_rank),
            ]
            suffix = "".join([
                f"_{prefix}{rank_fn()}"
                for prefix, world_size_fn, rank_fn in parallel_info
                if world_size_fn() != 1
            ])