[CK_TILE] Restructure Tile Engine's benchmarking and profiling

[arai713]

Motivation

This PR introduces a restructure for the benchmarking and profiling aspects of CK Tile's Tile Engine, expanding on the groundwork from this previous ROCm/composable_kernel#3434 and outlined in this design document. In PR 3434, to reduce repeated code we implemented:

• Base class that centralizes common functionality and provides a default implementation (Universal GEMM)
• Child classes for GEMM variants override virtual functions to handle variant-specific behavior

This refactoring in this PR follows the same process and should greatly reduce the duplicated code present in Tile Engine and make it simpler to add in new operations, increasing scalability.

Technical Details

The files have been refactored around new base structs for benchmarks, profiling and problem descriptions. The new base structs are:

• GemmProblem
• GemmBenchmark
• GemmProfiler

Universal GEMM, Preshuffle GEMM, and Multi-D GEMM all have child classes that will inherit from these base structs overriding only what differs per variant.
All common functions across the benchmarking and profiling files have been moved into newly added common utility files under the commons/ directory. The new utility files are:

• utils.hpp: common functions for the benchmarking and profiling process
• benchmark_utils.py: common utility functions for the benchmark generation

Test Plan

I tested using the existing tests for Tile Engine.

Test Result

All tests passed.

Submission Checklist

• Look over the contributing guidelines at https://github.com/ROCm/ROCm/blob/develop/CONTRIBUTING.md#pull-requests.

[Copilot]

Pull request overview

This PR restructures the Tile Engine's benchmarking and profiling infrastructure to reduce code duplication and simplify the addition of new operations. The refactoring introduces a base class architecture with common utilities extracted into shared files, while maintaining backward compatibility with existing tests.

Changes:

• Introduced base profiler class (GemmProfiler) and common utility files to eliminate duplicated code across GEMM variants
• Restructured GEMM universal, preshuffle, and multi_d implementations to inherit from the new base classes
• Consolidated Python benchmarking utilities into a shared module accessible by all GEMM variants

Reviewed changes

Copilot reviewed 32 out of 32 changed files in this pull request and generated no comments.

Show a summary per file

File	Description
gemm_universal_profiler.hpp	New universal GEMM profiler inheriting from base GemmProfiler class
gemm_profiler.hpp	New base profiler template class providing common profiling functionality
common/utils.hpp	Common utility functions and structs (Metric, PerformanceResult, KernelInstance, Setting)
gemm_benchmark.hpp	Base problem definition and comparison utilities for GEMM operations
gemm_common.hpp	Shared kernel traits and argument parser creation
common/benchmark_utils.py	Consolidated Python utilities for running kernels and exporting results
gemm_benchmark.py	Base Python class for GEMM benchmarking with kernel discovery and execution

---

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

[cgmillette]

Can we just return the boolean value if std::is_same_v instead of wrapping it inside of a bool_constant?

[arai713]

Other commonly used functions, like get_default_stride, that call is_row_major are expecting the bool_constant return type.

[ThruptiRajLakshmanaGowda]

LGTM!

[AviralGoelAMD]

Looks good now!