[ROCm] Skip test_batch_axis_sharding_jvp

[AratiGanesh]

Motivation

The test_batch_axis_sharding_jvp13 test (qr with complex64) fails on ROCm devices due to numerical precision issues in rocSolver as of ROCm 7.2.

Technical Details

Added a conditional skip in tests/linalg_sharding_test.py for test_batch_axis_sharding_jvp that only skips when:
Running on ROCm devices (jtu.is_device_rocm())
Function is lax.linalg.qr (fun_and_shapes[0] is lax.linalg.qr)
Dtype is np.complex64 (dtype == np.complex64)
This targets only the failing test case (test_batch_axis_sharding_jvp13). All other test cases continue to run normally.

Test Result

The test is now skipped on ROCm for the specific failing case, preventing test failures while preserving coverage for other variants.

[gemini-code-assist]

Summary of Changes

Hello @AratiGanesh, 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 introduces a targeted conditional skip for a specific test case within the linalg_sharding_test.py suite. The change prevents a known numerical precision failure on ROCm devices when performing qr decomposition with complex64 data types, ensuring the test suite remains stable while acknowledging an underlying issue in rocSolver.

Highlights

• Conditional Test Skipping: Implemented a conditional skip for the test_batch_axis_sharding_jvp test case when running on ROCm devices, specifically for lax.linalg.qr with np.complex64 dtype.
• Addressing Numerical Precision Issues: The skip addresses numerical precision issues observed in rocSolver as of ROCm 7.2 for qr decomposition with complex64 inputs, which caused test_batch_axis_sharding_jvp13 to fail.
• Targeted Fix: The conditional skip is highly targeted to only the failing test case, ensuring that all other variants of the test continue to execute normally and maintain coverage.

🧠 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

• tests/linalg_sharding_test.py
  • Added a conditional self.skipTest statement within the test_batch_axis_sharding_jvp method.

Activity

• No human activity (comments, reviews, etc.) 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 introduces a conditional skip for a specific test case (test_batch_axis_sharding_jvp) on ROCm devices. The skip is triggered when lax.linalg.qr is used with np.complex64 due to known numerical precision issues in rocSolver as of ROCm 7.2. This is a pragmatic approach to prevent test failures while retaining test coverage for other configurations. The inclusion of a TODO comment is helpful for tracking the re-enablement of the test once the underlying issue is resolved.

[gemini-code-assist]

The skip message currently hardcodes the test case name "test_batch_axis_sharding_jvp13". This name is derived from the parameterization order and might become inaccurate if the test parameters or their order change in the future. To improve robustness and clarity, consider dynamically including the specific function and dtype that trigger the skip in the message.

Suggested change

	self.skipTest("test_batch_axis_sharding_jvp13 not supported on ROCm due to rocSolver issue")
	self.skipTest(f"lax.linalg.qr with np.complex64 not supported on ROCm due to rocSolver issue (fun={fun_and_shapes[0].__name__}, dtype={dtype})")