mirror of
https://github.com/ROCm/composable_kernel.git
synced 2026-05-11 17:00:18 +00:00
* add DeviceGemmXdl * update script * fix naming issue * fix comment * output HostTensorDescriptor * rename * padded GEMM for fwd v4r4r4 nhwc * refactor * refactor * refactor * adding ckProfiler * adding ckProfiler * refactor * fix tuning parameter bug * add more gemm instances * add more fp16 GEMM instances * fix profiler driver * fix bug in tuning parameter * add fp32 gemm instances * small fix * refactor * rename * refactor gemm profiler; adding DeviceConv and conv profiler * refactor * fix * add conv profiler * refactor * adding more GEMM and Conv instance * Create README.md Add build instruction for ckProfiler * Create README.md Add Readme for gemm_xdl example * Update README.md Remove build instruction from top most folder * Update README.md * clean up
51 lines
2.6 KiB
CMake
51 lines
2.6 KiB
CMake
include_directories(BEFORE
|
|
include
|
|
${PROJECT_SOURCE_DIR}/host/host_tensor/include
|
|
${PROJECT_SOURCE_DIR}/device/include
|
|
${PROJECT_SOURCE_DIR}/device_operation/include
|
|
${PROJECT_SOURCE_DIR}/profiler/include
|
|
${PROJECT_SOURCE_DIR}/composable_kernel/include
|
|
${PROJECT_SOURCE_DIR}/composable_kernel/include/utility
|
|
${PROJECT_SOURCE_DIR}/composable_kernel/include/tensor_description
|
|
${PROJECT_SOURCE_DIR}/composable_kernel/include/tensor_operation
|
|
${PROJECT_SOURCE_DIR}/composable_kernel/include/problem_transform
|
|
${PROJECT_SOURCE_DIR}/external/rocm/include
|
|
)
|
|
|
|
# device_gemm_instance
|
|
set(DEVICE_GEMM_INSTANCE_SOURCE
|
|
${PROJECT_SOURCE_DIR}/device_operation/device_gemm_xdl_instance_f32_f32_f32_mk_kn_mn.cpp;
|
|
${PROJECT_SOURCE_DIR}/device_operation/device_gemm_xdl_instance_f32_f32_f32_mk_nk_mn.cpp;
|
|
${PROJECT_SOURCE_DIR}/device_operation/device_gemm_xdl_instance_f32_f32_f32_km_kn_mn.cpp;
|
|
${PROJECT_SOURCE_DIR}/device_operation/device_gemm_xdl_instance_f32_f32_f32_km_nk_mn.cpp;
|
|
${PROJECT_SOURCE_DIR}/device_operation/device_gemm_xdl_instance_f16_f16_f16_mk_kn_mn.cpp;
|
|
${PROJECT_SOURCE_DIR}/device_operation/device_gemm_xdl_instance_f16_f16_f16_mk_nk_mn.cpp;
|
|
${PROJECT_SOURCE_DIR}/device_operation/device_gemm_xdl_instance_f16_f16_f16_km_kn_mn.cpp;
|
|
${PROJECT_SOURCE_DIR}/device_operation/device_gemm_xdl_instance_f16_f16_f16_km_nk_mn.cpp;
|
|
)
|
|
|
|
add_library(device_gemm_instance SHARED ${DEVICE_GEMM_INSTANCE_SOURCE})
|
|
target_include_directories(device_gemm_instance SYSTEM PUBLIC $<BUILD_INTERFACE:${HALF_INCLUDE_DIR}>)
|
|
target_compile_features(device_gemm_instance PUBLIC)
|
|
set_target_properties(device_gemm_instance PROPERTIES POSITION_INDEPENDENT_CODE ON)
|
|
install(TARGETS device_gemm_instance LIBRARY DESTINATION lib)
|
|
|
|
# device_conv_instance
|
|
set(DEVICE_CONV_INSTANCE_SOURCE
|
|
${PROJECT_SOURCE_DIR}/device_operation/device_conv_xdl_instance_f32_f32_f32_nhwc_kyxc_nhwk.cpp;
|
|
${PROJECT_SOURCE_DIR}/device_operation/device_conv_xdl_instance_f16_f16_f16_nhwc_kyxc_nhwk.cpp;
|
|
)
|
|
|
|
add_library(device_conv_instance SHARED ${DEVICE_CONV_INSTANCE_SOURCE})
|
|
target_include_directories(device_conv_instance SYSTEM PUBLIC $<BUILD_INTERFACE:${HALF_INCLUDE_DIR}>)
|
|
target_compile_features(device_conv_instance PUBLIC)
|
|
set_target_properties(device_conv_instance PROPERTIES POSITION_INDEPENDENT_CODE ON)
|
|
install(TARGETS device_conv_instance LIBRARY DESTINATION lib)
|
|
|
|
# ck_profiler
|
|
set(PROFILER_SOURCE profiler.cpp gemm_profiler.cpp conv_profiler.cpp)
|
|
add_executable(ckProfiler ${PROFILER_SOURCE})
|
|
|
|
target_link_libraries(ckProfiler PRIVATE host_tensor)
|
|
target_link_libraries(ckProfiler PRIVATE device_gemm_instance device_conv_instance)
|