Skip to content

docs: improve CI and testing documentation#21202

Merged
merrymercy merged 16 commits intomainfrom
lianmin/simplify-ci
Mar 23, 2026
Merged

docs: improve CI and testing documentation#21202
merrymercy merged 16 commits intomainfrom
lianmin/simplify-ci

Conversation

@merrymercy
Copy link
Copy Markdown
Contributor

Summary

This PR is primarily focused on improving documentation for CI and tests: clearer guidance in test/README.md, small doc/index and root README touch-ups, and related workflow readability updates where test suites are named or described.

Additional changes (supporting / cleanup)

  • CI workflow adjustments (e.g. AMD PR test workflows) and script path moves under scripts/ci/amd/.
  • Test runner and registration tweaks, removal of unused SRT experiment/compare helpers, and a few small fixes (e.g. CPU test suite / mock-related follow-ups).

Reviewers: the largest user-facing doc delta is test/README.md; start there for the testing/CI story.

Made with Cursor

merrymercy and others added 8 commits March 23, 2026 00:40
Use register_fake_if_exists for sgl_kernel ops in fp8_kernel.py and
fp8_utils.py, and replace the simple MagicMock stub in
test_serving_embedding.py with a MetaPathFinder that properly handles
all sgl_kernel.* submodule imports.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
@gemini-code-assist
Copy link
Copy Markdown
Contributor

Summary of Changes

Hello, 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 primarily focuses on enhancing the clarity and structure of the project's continuous integration and testing documentation. It introduces a comprehensive overview of the CI pipeline, test organization, and local execution instructions, making it easier for developers to understand and contribute to the testing framework. Alongside these documentation improvements, the PR also refactors various test-related scripts and removes obsolete utilities, contributing to a cleaner and more maintainable codebase.

Highlights

  • Documentation Improvement: Significantly improved documentation for CI and testing, including a detailed three-stage CI pipeline, folder organization, and local test execution guidance in test/README.md.
  • CI/Test Infrastructure Refactoring: Refactored CI-related scripts and test files, moving AMD-specific CI scripts to a dedicated directory (scripts/ci/amd/) and consolidating test execution logic.
  • Test Runner Enhancements: Enhanced test runner and registration mechanisms, including clearer definitions for CI registry fields and an updated sgl_kernel mocking strategy for CPU-only test environments.
  • Cleanup of Obsolete Utilities: Removed several unused experiment and comparison helper scripts (experiment_runner.py, compare.py, parse_results.py, run_suite_nightly.py), streamlining the test infrastructure.
  • Core Documentation Updates: Updated core documentation (README.md, docs/index.rst) to reflect new NVIDIA GPU support (5090) and a new blog post.

🧠 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.

Ignored Files
  • Ignored by pattern: .github/workflows/** (2)
    • .github/workflows/pr-test-amd-rocm720.yml
    • .github/workflows/pr-test-amd.yml
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.

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.

Copy link
Copy Markdown
Contributor

@gemini-code-assist gemini-code-assist Bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Code Review

This pull request significantly improves the documentation for the CI and testing systems, making it much clearer and more comprehensive for developers. The refactoring of the test infrastructure, including the move to a unified registry-based system, the consolidation of test runners, and more robust mocking for CPU-only tests, are all excellent changes that enhance maintainability and reliability. I've found one issue in the updated testing documentation that could cause CI to misreport test results, which I've detailed in a specific comment.

Comment thread test/README.md Outdated
- Expand test/README.md with per-commit CUDA/CPU/AMD suite tables and
  guidance for choosing a suite; clarify est_time is in seconds.
- Add stage-b-test-4-gpu-b200 to CUDA PER_COMMIT_SUITES and reorder
  entries to match documentation.

Made-with: Cursor
@merrymercy
Copy link
Copy Markdown
Contributor Author

/tag-and-rerun-ci

- hisparse_memory_pool: import sgl_kernel.kvcacheio only when is_cuda or is_hip
- write-sglang-test SKILL: link test/README.md, clarify GPU suite descriptions

Made-with: Cursor
….toml

Pin torchaudio==2.10.0+xpu in the Dockerfile pip install step to prevent
a newer version (2.11.0+xpu) from being installed, which causes a
version mismatch when pip install later resolves pyproject_xpu.toml
dependencies.

Made-with: Cursor
pytest.main() returns an exit code but does not exit the process,
causing CI to misreport test failures. Wrap all bare pytest.main()
calls in sys.exit() and add import sys where missing. Also fix the
documentation in test/README.md and skill files.

Made-with: Cursor
@merrymercy merrymercy merged commit 27ac831 into main Mar 23, 2026
41 of 75 checks passed
@merrymercy merrymercy deleted the lianmin/simplify-ci branch March 23, 2026 17:48
adityavaid pushed a commit to adityavaid/sglang that referenced this pull request Mar 24, 2026
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
0-693 pushed a commit to 0-693/sglang that referenced this pull request Mar 25, 2026
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
JustinTong0323 pushed a commit to JustinTong0323/sglang that referenced this pull request Apr 7, 2026
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
yhyang201 pushed a commit to yhyang201/sglang that referenced this pull request Apr 22, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant