mirror of https://github.com/ROCm/composable_kernel.git synced 2026-07-18 17:48:06 +00:00

Files

JiaLuo-CAN 39ced68ba2 [rocm-libraries] ROCm/rocm-libraries#9473 (commit f8527f1)

fix(ck_tile): CK CI Aiter test error

## Motivation

CI Aiter tests broken since May and blocking CK updating in Aiter.
Jira ID : AICK-1450
https://amd-hub.atlassian.net/browse/AICK-1450

## Technical Details

1. Include fixes from PR 9440
https://github.com/ROCm/rocm-libraries/pull/9440
2. Fix failures caused by Aiter tests changes (FLYDSL issue)
3. Fix Aiter CI test failure on GPTOSS-120b (attention sink value logic)
4. Fix failure caused by refactoring in PR4798 breaks mxfp4 MoE instance
5. Fix failure caused by PR 6526

## Test Plan

Pass CI aiter tests and regular tests.

## Test Result

<!-- Briefly summarize test outcomes. -->

## Submission Checklist

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

2026-07-17 13:12:05 +00:00

algorithm

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

arch

[rocm-libraries] ROCm/rocm-libraries#8020 (commit fa9cdd1)

2026-07-14 20:51:36 +00:00

container

[rocm-libraries] ROCm/rocm-libraries#6768 (commit 43ca43f)

2026-06-05 12:27:41 +00:00

numeric

[rocm-libraries] ROCm/rocm-libraries#8020 (commit fa9cdd1)

2026-07-14 20:51:36 +00:00

tensor

[rocm-libraries] ROCm/rocm-libraries#9473 (commit f8527f1)

2026-07-17 13:12:05 +00:00

utility

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

config.hpp

[rocm-libraries] ROCm/rocm-libraries#6768 (commit 43ca43f)

2026-06-05 12:27:41 +00:00

README.md

introducing ck_tile! (#1216 )

2024-04-15 19:27:12 -05:00

README.md

ck_tile/core

ck_tile/core contains every basic functions and structures to create a GPU kernel using ck_tile. User should only include ck_tile/core.hpp this single header to use all the functionality. Everything is under ck_tile namespace. The coding style under this folder should be similar to std (snake_case for structure/function, Camel for template types...)

algorithm/
    coordinate transform and some other reusable algorithm
arch/
    contains some basic device building block like mma, buffer addressing, etc...
container/
    contains basic container data structure, array/sequence/tuple/...
numeric/
    data type, and data type related math
tensor/
    tensor descriptors and tile level API
utility/
    other utility function for both host/device