[AMD] Fix Aiter Prebuild When Releasing ROCm720 Image

[bingxche]

Motivation

Update aiter install command in rocm.dockerfile to ensure aiter prebuild kernels exists when releasing docker image.

Modifications

Accuracy Tests

The fix works, ROCm720 images now successfully build AITER kernels:
https://github.com/sgl-project/sglang/actions/runs/22856907949

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, 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 an issue where aiter prebuilt kernels were not properly included in the ROCm Docker image during its release process. By updating the installation commands in the rocm.Dockerfile, the changes ensure that the necessary kernels are correctly built and installed, thereby resolving potential runtime issues related to missing aiter components in the released images.

Highlights

• Aiter Installation Process: Modified the aiter installation commands within the rocm.Dockerfile to ensure prebuilt kernels are correctly included when building the Docker image. This involved replacing python setup.py develop with a two-step process: python setup.py build_ext --inplace followed by pip install -e . for relevant build configurations.

🧠 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

• docker/rocm.Dockerfile
  • Updated the aiter installation commands from python setup.py develop to a two-step process involving python setup.py build_ext --inplace followed by pip install -e . to ensure prebuilt kernels are included.

Activity

• No specific activity 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. ↩

[gemini-code-assist]

Code Review

This pull request correctly addresses the issue of pre-building aiter kernels by separating the build and install steps. The changes are logical and move towards modern Python packaging practices. I've included a couple of suggestions to further improve the shell commands within the Dockerfile for conciseness and efficiency.

[gemini-code-assist]

For better readability and efficiency, you can combine these two sh -c calls into a single one. This avoids spawning a new shell for each command and keeps the related logic together. By using export, the environment variables are set for all subsequent commands within the same shell.

          sh -c "export HIP_CLANG_PATH=/sgl-workspace/llvm-project/build/bin/; export GPU_ARCHS=$GPU_ARCH_LIST; PREBUILD_KERNELS=1 python setup.py build_ext --inplace && pip install -e ."; \

[gemini-code-assist]

Similarly to the above, you can combine these two sh -c calls into a single one for better readability and efficiency.

          sh -c "export GPU_ARCHS=$GPU_ARCH_LIST; PREBUILD_KERNELS=1 python setup.py build_ext --inplace && pip install -e ."; \

[bingxche]

The fix works, ROCm720 images now successfully build AITER kernels:
https://github.com/sgl-project/sglang/actions/runs/22856907949

[HaiShaw]

/tag-and-rerun-ci