[CK_TILE] Stream-K Gemm Example for fp8 and bf8

[arai713]

Proposed changes

This PR is expanding the supported datatypes for Stream-K Gemm by adding examples for fp8 and bf8. Previously examples for fp16 and bf 16 were added through this PR. Currently these examples only support atomic reduction. Unit tests for fp8 and bf8 have been added through a test suite.

Checklist

Please put an x into the boxes that apply. You can also fill these out after creating the PR. If you're not sure, please don't hesitate to ask.

• I have added tests relevant to the introduced functionality, and the unit tests are passing locally
• I have added the test to REGRESSION_TESTS list defined at the top of CMakeLists.txt in tests/CMakeLists.txt, IF the test takes more than 30 seconds to run.
• I have added inline documentation which enables the maintainers with understanding the motivation
• I have removed the stale documentation which is no longer relevant after this pull request
• (If this change is user-facing) I have added release notes which provide the end users with a brief summary of the improvement from this pull request
• I have run clang-format on all changed files
• Any dependent changes have been merged

[cgmillette]

Need to consolidate new fp8 tests into smoke tests with existing test harness.

[ecamartins]

LGTM! Great job on refactoring the example and tests :)

[cgmillette]

This looks great! Approved pending CI!

[Copilot]

Pull Request Overview

This PR extends Stream-K GEMM support to fp8 and bf8 data types, building on the existing fp16 and bf16 implementation. The changes introduce type configuration headers, test suite infrastructure, and example code updates to handle the new 8-bit floating-point formats with atomic reduction support.

• Adds fp8 and bf8 type definitions and layout configurations
• Implements smoke tests covering multiple layout combinations (RRR, RCR, CRR, CCR) for both data types
• Updates the basic example to support fp8 and bf8 with appropriate command-line interface changes

Reviewed Changes

Copilot reviewed 17 out of 17 changed files in this pull request and generated 4 comments.

Show a summary per file

File	Description
test/ck_tile/gemm_streamk/test_gemm_streamk_types_fp8.hpp	Defines F8Layouts struct and macros for fp8 test configurations
test/ck_tile/gemm_streamk/test_gemm_streamk_types_bf8.hpp	Defines BF8Layouts struct and macros for bf8 test configurations
test/ck_tile/gemm_streamk/test_gemm_streamk_types.hpp	Adds F8 and BF8 type aliases and includes new type headers
test/ck_tile/gemm_streamk/smoke_tests/*.cpp	Eight new smoke test files covering fp8/bf8 with different layouts
test/ck_tile/gemm_streamk/CMakeLists.txt	Adds compile options for OCP FP8 support and registers new test files
example/ck_tile/40_streamk_gemm/streamk_gemm_basic.cpp	Adds fp8 and bf8 case handling in the example runner
example/ck_tile/40_streamk_gemm/gemm_utils.hpp	Adds DataTypeTraits specializations and updates documentation
example/ck_tile/40_streamk_gemm/README.md	Updates documentation to reflect fp8/bf8 support
example/ck_tile/40_streamk_gemm/CMakeLists.txt	Adds OCP FP8 compile options for the example
example/ck_tile/CMakeLists.txt	Adds trailing newline

---

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

[cgmillette]

lgtm!