mirror of
https://github.com/ROCm/composable_kernel.git
synced 2026-07-12 10:08:01 +00:00
2a73eb3bc0
* tf32:bf16x3:use bf16x3 emulate tf32 gemm * change blockwiseGemm to demo bf16x3 * temp push * self review * self review * fix multi-device compile error * bug fix * code refactor * limit to gfx950 * enhance gemm gfx942 threshold * lower change from blockwise to warpwise * refact codes * refact codes * error fix * change threshold * bug fix * fix threshold error * change host reference implement to same as device * bug fix * bug fix * code refact * fix clang-format fail * code refine
46 lines
2.4 KiB
CMake
46 lines
2.4 KiB
CMake
add_custom_target(example_grouped_gemm_xdl)
|
|
add_example_executable(example_grouped_gemm_xdl_fp32 grouped_gemm_xdl_fp32.cpp)
|
|
add_example_dependencies(example_grouped_gemm_xdl example_grouped_gemm_xdl_fp32)
|
|
|
|
add_example_executable(example_grouped_gemm_xdl_fp16 grouped_gemm_xdl_fp16.cpp)
|
|
add_example_dependencies(example_grouped_gemm_xdl example_grouped_gemm_xdl_fp16)
|
|
|
|
add_example_executable(example_grouped_gemm_multiple_d_dl_fp16 grouped_gemm_multiple_d_dl_fp16.cpp)
|
|
add_example_dependencies(example_grouped_gemm_xdl example_grouped_gemm_multiple_d_dl_fp16)
|
|
|
|
add_example_executable(example_grouped_gemm_xdl_splitk_fp16 grouped_gemm_xdl_splitk_fp16.cpp)
|
|
add_example_dependencies(example_grouped_gemm_xdl example_grouped_gemm_xdl_splitk_fp16)
|
|
|
|
add_example_executable(example_grouped_gemm_xdl_fixed_nk_fp16 grouped_gemm_xdl_fixed_nk_fp16.cpp)
|
|
add_example_dependencies(example_grouped_gemm_xdl example_grouped_gemm_xdl_fixed_nk_fp16)
|
|
|
|
add_example_executable(example_grouped_gemm_xdl_fixed_nk_bias_fp16 grouped_gemm_xdl_fixed_nk_bias_fp16.cpp)
|
|
add_example_dependencies(example_grouped_gemm_xdl example_grouped_gemm_xdl_fixed_nk_bias_fp16)
|
|
|
|
add_example_executable(example_grouped_gemm_xdl_bf16 grouped_gemm_xdl_bf16.cpp)
|
|
add_example_dependencies(example_grouped_gemm_xdl example_grouped_gemm_xdl_bf16)
|
|
|
|
add_example_executable(example_grouped_gemm_xdl_int8 grouped_gemm_xdl_int8.cpp)
|
|
add_example_dependencies(example_grouped_gemm_xdl example_grouped_gemm_xdl_int8)
|
|
|
|
add_example_executable(example_grouped_gemm_xdl_fixed_nk_fp16_fp8 grouped_gemm_xdl_fixed_nk_fp16_fp8.cpp)
|
|
add_example_dependencies(example_grouped_gemm_xdl example_grouped_gemm_xdl_fixed_nk_fp16_fp8)
|
|
|
|
add_example_executable(example_grouped_gemm_multiple_d_xdl_fp16 grouped_gemm_multiple_d_xdl_fp16.cpp)
|
|
add_example_dependencies(example_grouped_gemm_xdl example_grouped_gemm_multiple_d_xdl_fp16)
|
|
|
|
if(USE_BITINT_EXTENSION_INT4)
|
|
add_example_executable(example_grouped_gemm_xdl_int4 grouped_gemm_xdl_int4.cpp)
|
|
add_example_dependencies(example_grouped_gemm_xdl example_grouped_gemm_xdl_int4)
|
|
endif()
|
|
|
|
list(APPEND gpu_list_tf32 gfx942 gfx950)
|
|
set(target 0)
|
|
foreach(gpu IN LISTS GPU_TARGETS)
|
|
if(gpu IN_LIST gpu_list_tf32 AND target EQUAL 0)
|
|
add_example_executable(example_grouped_gemm_xdl_fp32_tf32 grouped_gemm_xdl_fp32_tf32.cpp)
|
|
add_example_dependencies(example_grouped_gemm_xdl example_grouped_gemm_xdl_fp32_tf32)
|
|
set(target 1)
|
|
endif()
|
|
endforeach()
|