mirror of
https://github.com/ROCm/composable_kernel.git
synced 2026-03-25 09:37:42 +00:00
8f27f65d44
[CK_TILE] MX GEMM non-preshuffled RCR layout ## Motivation Implements a GEMM with MX scaling for fp4 and fp8 in non-preshuffled layouts using async pipeline. ## Technical Details <!-- Explain the changes along with any relevant GitHub links. --> ## Test Plan <!-- Explain any relevant testing done to verify this PR. --> ## 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.
37 lines
1.1 KiB
CMake
37 lines
1.1 KiB
CMake
# Copyright (c) Advanced Micro Devices, Inc., or its affiliates.
|
|
# SPDX-License-Identifier: MIT
|
|
|
|
include_directories(AFTER
|
|
${CMAKE_CURRENT_LIST_DIR}
|
|
)
|
|
|
|
add_subdirectory(01_fmha)
|
|
add_subdirectory(02_layernorm2d)
|
|
add_subdirectory(03_gemm)
|
|
add_subdirectory(04_img2col)
|
|
add_subdirectory(05_reduce)
|
|
add_subdirectory(06_permute)
|
|
add_subdirectory(09_topk_softmax)
|
|
add_subdirectory(10_rmsnorm2d)
|
|
add_subdirectory(11_add_rmsnorm2d_rdquant)
|
|
add_subdirectory(12_smoothquant)
|
|
add_subdirectory(13_moe_sorting)
|
|
add_subdirectory(14_moe_smoothquant)
|
|
add_subdirectory(15_fused_moe)
|
|
add_subdirectory(16_batched_gemm)
|
|
add_subdirectory(17_grouped_gemm)
|
|
add_subdirectory(18_flatmm)
|
|
add_subdirectory(19_gemm_multi_d)
|
|
add_subdirectory(20_grouped_convolution)
|
|
add_subdirectory(21_elementwise)
|
|
add_subdirectory(22_gemm_multi_abd)
|
|
add_subdirectory(35_batched_transpose)
|
|
add_subdirectory(36_pooling)
|
|
add_subdirectory(38_block_scale_gemm)
|
|
add_subdirectory(40_streamk_gemm)
|
|
add_subdirectory(41_batched_contraction)
|
|
add_subdirectory(42_mx_gemm)
|
|
add_subdirectory(50_sparse_attn)
|
|
add_subdirectory(51_tile_distr_enc_reg_map)
|
|
|