Thomas Ning 66663bf2d9 Add the instance of MBlock=144 for GemmMultiplyMultiply (#1955) ...

* tempsave, not selected

* finish the feature and merge with develop

---------

Co-authored-by: aska-0096 <haocwang@amd.com>

[ROCm/composable_kernel commit: c954bd0cfa]

2025-03-07 13:44:06 -08:00

..

CMakeLists.txt

ck moe gemm implement (#1936)

2025-03-05 15:56:55 +08:00

gemm_add_add_xdl_fp16.cpp

[GEMM] gemm_universal related optimization (#1453)

2024-08-14 10:42:30 +08:00

gemm_multiply_multiply_xdl_fp8_ab_scale.cpp

Revert "[BlockScale GEMM] FP8 Blockscale GEMM optimization and ckProfiler (#1913)" (#1933)

2025-03-03 07:17:39 -08:00

gemm_multiply_multiply_xdl_fp8_bpreshuffle.cpp

[A8W8 GEMM] Optimized weight-preshuffled implementation & add quantization datatype for CK TILE rms_norm (#1862)

2025-02-20 14:00:27 -08:00

gemm_multiply_multiply_xdl_fp8.cpp

Add the instance of MBlock=144 for GemmMultiplyMultiply (#1955)

2025-03-07 13:44:06 -08:00

gemm_multiply_multiply_xdl_int8.cpp

[A8W8 GEMM] Optimized weight-preshuffled implementation & add quantization datatype for CK TILE rms_norm (#1862)

2025-02-20 14:00:27 -08:00

moe_gemm1_xdl_fp8.cpp

ck moe gemm implement (#1936)

2025-03-05 15:56:55 +08:00

moe_gemm2_xdl_fp8.cpp

ck moe gemm implement (#1936)

2025-03-05 15:56:55 +08:00