[Bugfix][CI/CD][CPU] Fix CPU CI tests #20383
Conversation
|
👋 Hi! Thank you for contributing to the vLLM project. 💬 Join our developer Slack at https://slack.vllm.ai to discuss your PR in #pr-reviews, coordinate on features in #feat- channels, or join special interest groups in #sig- channels. Just a reminder: PRs would not trigger full CI run by default. Instead, it would only run Once the PR is approved and ready to go, your PR reviewer(s) can run CI to test the changes comprehensively before merging. To run CI, PR reviewers can either: Add 🚀 |
There was a problem hiding this comment.
Summary of Changes
Hello @bigPYJ1151, 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 resolves a critical bug that was causing CPU CI tests to fail. The issue stemmed from an operation registration (apply_w8a8_block_fp8_linear) that utilized type signatures unsupported by torch 2.6 on CPU. The fix involves preventing this specific operation from being registered when running on a CPU platform, thereby ensuring compatibility and allowing CPU tests to pass.
Highlights
- Bug Fix: CPU Compatibility: Addressed a bug causing CPU CI test failures by conditionally registering a custom operation (
apply_w8a8_block_fp8_linear). This operation's type signatures were incompatible withtorch 2.6on CPU platforms. - Conditional Operation Registration: The
direct_register_custom_opcall forapply_w8a8_block_fp8_linearis now guarded by anif not current_platform.is_cpu():check, ensuring it is only registered on non-CPU platforms.
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 in your pull request via creating an issue comment (i.e. comment on the pull request page) using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands.
| 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 issue 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 is currently in preview and 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 to provide feedback.
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
-
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. ↩
There was a problem hiding this comment.
Code Review
This pull request fixes a CI failure on CPU platforms by preventing a GPU-specific custom operator, apply_w8a8_block_fp8_linear, from being registered in CPU-only environments. The change is correct and directly addresses the issue. I've added one suggestion to make the platform check more specific, which will improve future maintainability.
| mutates_args=[], | ||
| fake_impl=apply_w8a8_block_fp8_linear_fake, | ||
| ) | ||
| if not current_platform.is_cpu(): |
There was a problem hiding this comment.
The condition not current_platform.is_cpu() is correct for the immediate problem, but it could be more specific. The apply_w8a8_block_fp8_linear function appears to be implemented only for CUDA and ROCm platforms, as it relies on GPU-specific kernels (Triton, CUTLASS, or AITer).
Using a more specific check like current_platform.is_cuda_alike() would be more robust. This would prevent the operator from being incorrectly registered on other future non-CPU platforms (e.g., TPU, HPU) where it might not be supported, improving future maintainability.
| if not current_platform.is_cpu(): | |
| if current_platform.is_cuda_alike(): |
|
The test still fails. PTAL |
|
@DarkLight1337 I suppose #20381 can solve the failure. Let's wait it and try again. |
Signed-off-by: jiang1.li <jiang1.li@intel.com>
Signed-off-by: jiang1.li <jiang1.li@intel.com>
Signed-off-by: jiang1.li <jiang1.li@intel.com>
Signed-off-by: jiang1.li <jiang1.li@intel.com>
Signed-off-by: jiang1.li <jiang1.li@intel.com> Signed-off-by: avigny <47987522+avigny@users.noreply.github.com>
Signed-off-by: jiang1.li <jiang1.li@intel.com>
Signed-off-by: jiang1.li <jiang1.li@intel.com> Signed-off-by: Jinzhen Lin <linjinzhen@hotmail.com>
Signed-off-by: jiang1.li <jiang1.li@intel.com>
Essential Elements of an Effective PR Description Checklist
supported_models.mdandexamplesfor a new model.Purpose
Test Plan
Test Result
(Optional) Documentation Update