Skip to content

Conversation

@dorde-antic
Copy link
Contributor

Motivation

Technical Details

Test Plan

Test Result

Submission Checklist

@dorde-antic dorde-antic requested a review from Copilot November 25, 2025 15:00
Copy link
Contributor

Copilot AI left a comment

Choose a reason for hiding this comment

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

Pull request overview

This PR introduces a comprehensive test suite for Python performance utilities in the MLIR codebase. The changes add unit tests for tuning and performance runner modules, along with necessary test infrastructure.

  • Adds test coverage for tuningRunner.py module including verification mode flags and configuration selection logic
  • Adds test coverage for perfRunner.py module including layout conversions, configuration generation for conv/gemm/attention operations, and command-line parsing
  • Provides test infrastructure via conftest.py to enable testing without GPU hardware dependencies

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 2 comments.

File Description
mlir/utils/performance/tests/test_tuningRunner.py Tests for tuning runner verification modes and winning configuration selection
mlir/utils/performance/tests/test_perfRunner.py Tests for performance runner configuration generation, command-line parsing, and layout conversions
mlir/utils/performance/tests/conftest.py Test configuration providing module path setup and HIP runtime stub for GPU-less testing

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

dorde-antic and others added 3 commits November 25, 2025 16:27
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants