Ville Pietilä fbf53fb970 [CK_TILE] Split-K autodeduction (#3351) ...

* First version of split-K autodeduction.

* Fix circular dependency and kernel construction.

* Fix tolerance calculation for bwd weight example.

* Simplify kernel construction.

* Fix kernel launching bug for split-K autodeduce.

* Add split-K autodeduction support for the two stage example.

* Fix a corner case.

* Fix clang-format.

* Fix clang-format for inc files.

* Add missing header.

* Prevent too large split-K values.

* Fix formatting.

* Add unit tests for IsSupportedArgument in grouped bwd conv.

* clang-format.

* Fix merge conflicts.

* Address feedback from code review.

* clang-format

* Fix new tests after merge.

---------

Co-authored-by: Ville Pietilä <>

[ROCm/composable_kernel commit: fc22320d78]

2025-12-10 09:30:30 +02:00

..

add_rmsnorm2d_rdquant

chore(copyright) update library wide CMakeLists.txt copyright header template (#3313)

2025-11-28 13:49:54 -08:00

atomic_add_op

chore(copyright) update library wide CMakeLists.txt copyright header template (#3313)

2025-11-28 13:49:54 -08:00

batched_gemm

[CK-Tile] Refactor base pipeline usage (#3251)

2025-12-04 11:45:49 -08:00

batched_transpose

chore(copyright) update library wide CMakeLists.txt copyright header template (#3313)

2025-11-28 13:49:54 -08:00

container

chore(copyright) update library wide CMakeLists.txt copyright header template (#3313)

2025-11-28 13:49:54 -08:00

core

chore(copyright) update library wide CMakeLists.txt copyright header template (#3313)

2025-11-28 13:49:54 -08:00

data_type

chore(copyright) update library wide CMakeLists.txt copyright header template (#3313)

2025-11-28 13:49:54 -08:00

elementwise

chore(copyright) update library wide CMakeLists.txt copyright header template (#3313)

2025-11-28 13:49:54 -08:00

epilogue

chore(copyright) update library wide CMakeLists.txt copyright header template (#3313)

2025-11-28 13:49:54 -08:00

fmha

chore(copyright) update library wide CMakeLists.txt copyright header template (#3313)

2025-11-28 13:49:54 -08:00

gemm

[CK-Tile] Refactor base pipeline usage (#3251)

2025-12-04 11:45:49 -08:00

gemm_block_scale

[CK_TILE] Support more layouts for BQuant GEMM (#3349)

2025-12-08 13:05:56 -08:00

gemm_multi_abd

[CK-Tile] Refactor base pipeline usage (#3251)

2025-12-04 11:45:49 -08:00

gemm_multi_d

[CK-Tile] Refactor base pipeline usage (#3251)

2025-12-04 11:45:49 -08:00

gemm_streamk

chore(copyright) update library wide CMakeLists.txt copyright header template (#3313)

2025-11-28 13:49:54 -08:00

gemm_tile_engine

Disable gemm_blockscale_f8 on gfx90a by default. (#3338)

2025-12-02 11:33:33 -08:00

gemm_weight_preshuffle

[CK-Tile] Refactor base pipeline usage (#3251)

2025-12-04 11:45:49 -08:00

grouped_conv

[CK_TILE] Split-K autodeduction (#3351)

2025-12-10 09:30:30 +02:00

grouped_gemm

[CK-Tile] Refactor base pipeline usage (#3251)

2025-12-04 11:45:49 -08:00

grouped_gemm_multi_d

[CK-Tile] Refactor base pipeline usage (#3251)

2025-12-04 11:45:49 -08:00

grouped_gemm_preshuffle

[CK-Tile] Refactor base pipeline usage (#3251)

2025-12-04 11:45:49 -08:00

grouped_gemm_quant

[CK Tile] Grouped GEMM aquant mode and non-persistent kernel (#3337)

2025-12-08 12:19:22 -08:00

image_to_column

chore(copyright) update library wide CMakeLists.txt copyright header template (#3313)

2025-11-28 13:49:54 -08:00

layernorm2d

chore(copyright) update library wide CMakeLists.txt copyright header template (#3313)

2025-11-28 13:49:54 -08:00

memory_copy

chore(copyright) update library wide CMakeLists.txt copyright header template (#3313)

2025-11-28 13:49:54 -08:00

moe_smoothquant

chore(copyright) update library wide CMakeLists.txt copyright header template (#3313)

2025-11-28 13:49:54 -08:00

moe_sorting

chore(copyright) update library wide CMakeLists.txt copyright header template (#3313)

2025-11-28 13:49:54 -08:00

permute

chore(copyright) update library wide CMakeLists.txt copyright header template (#3313)

2025-11-28 13:49:54 -08:00

pooling

chore(copyright) update library wide CMakeLists.txt copyright header template (#3313)

2025-11-28 13:49:54 -08:00

reduce

chore(copyright) update library wide CMakeLists.txt copyright header template (#3313)

2025-11-28 13:49:54 -08:00

rmsnorm2d

chore(copyright) update library wide CMakeLists.txt copyright header template (#3313)

2025-11-28 13:49:54 -08:00

slice_tile

chore(copyright) update library wide CMakeLists.txt copyright header template (#3313)

2025-11-28 13:49:54 -08:00

smoothquant

chore(copyright) update library wide CMakeLists.txt copyright header template (#3313)

2025-11-28 13:49:54 -08:00

topk_softmax

chore(copyright) update library wide CMakeLists.txt copyright header template (#3313)

2025-11-28 13:49:54 -08:00

utility

[CK_TILE] Generate random tensor values with multiple threads (#3324)

2025-12-09 11:02:33 +08:00

warp_gemm

[CK_TILE] Add unit test for fp4 warp gemm (#2817)

2025-12-01 13:56:48 +08:00

CMakeLists.txt

[CK_TILE] Split-K autodeduction (#3351)

2025-12-10 09:30:30 +02:00