mirror of
https://github.com/ROCm/composable_kernel.git
synced 2026-05-12 17:26:00 +00:00
* support dynamic tensor descriptor * use buffer load OOB feature for padding case * add navi support * add int8x4 inference kernel Co-authored-by: Chao Liu <chao@ixt-rack-81.local.lan> Co-authored-by: Jing Zhang <jizhan@amd.com>
81 lines
3.6 KiB
CMake
81 lines
3.6 KiB
CMake
cmake_minimum_required(VERSION 2.8.3)
|
|
project(modular_convolution)
|
|
|
|
#c++
|
|
enable_language(CXX)
|
|
set(CMAKE_CXX_STANDARD 17)
|
|
set(CMAKE_CXX_STANDARD_REQUIRED ON)
|
|
set(CMAKE_CXX_EXTENSIONS OFF)
|
|
message("CMAKE_CXX_COMPILER_ID: ${CMAKE_CXX_COMPILER_ID}")
|
|
|
|
#boost
|
|
find_package(Boost REQUIRED)
|
|
|
|
message("Boost_INCLUDE_DIRS: ${Boost_INCLUDE_DIRS}")
|
|
message("Boost_LIBRARY_DIRS: ${Boost_LIBRARY_DIRS}")
|
|
|
|
include_directories(BEFORE ${Boost_INCLUDE_DIRS})
|
|
link_directories(${Boost_LIBRARY_DIRS})
|
|
|
|
#OpenMP
|
|
if(CMAKE_CXX_COMPILER_ID MATCHES "Clang")
|
|
# workaround issue hipcc in rocm3.5 cannot find openmp
|
|
set(OpenMP_CXX "${CMAKE_CXX_COMPILER}")
|
|
set(OpenMP_CXX_FLAGS "-fopenmp=libomp -Wno-unused-command-line-argument")
|
|
set(OpenMP_CXX_LIB_NAMES "libomp" "libgomp" "libiomp5")
|
|
set(OpenMP_libomp_LIBRARY ${OpenMP_CXX_LIB_NAMES})
|
|
set(OpenMP_libgomp_LIBRARY ${OpenMP_CXX_LIB_NAMES})
|
|
set(OpenMP_libiomp5_LIBRARY ${OpenMP_CXX_LIB_NAMES})
|
|
else()
|
|
find_package(OpenMP REQUIRED)
|
|
endif()
|
|
|
|
message("OpenMP_CXX_LIB_NAMES: ${OpenMP_CXX_LIB_NAMES}")
|
|
message("OpenMP_gomp_LIBRARY: ${OpenMP_gomp_LIBRARY}")
|
|
message("OpenMP_pthread_LIBRARY: ${OpenMP_pthread_LIBRARY}")
|
|
message("OpenMP_CXX_FLAGS: ${OpenMP_CXX_FLAGS}")
|
|
|
|
set (CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} ${OpenMP_CXX_FLAGS}")
|
|
link_libraries(${OpenMP_gomp_LIBRARY})
|
|
link_libraries(${OpenMP_pthread_LIBRARY})
|
|
|
|
#GPU backend
|
|
if(DEVICE_BACKEND STREQUAL "AMD")
|
|
find_package(HIP REQUIRED)
|
|
elseif(DEVICE_BACKEND STREQUAL "NVIDIA")
|
|
enable_language(CUDA)
|
|
endif()
|
|
|
|
#
|
|
include_directories(BEFORE
|
|
${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/kernel_algorithm
|
|
${PROJECT_SOURCE_DIR}/composable_kernel/include/driver
|
|
${PROJECT_SOURCE_DIR}/external/half/include
|
|
${PROJECT_SOURCE_DIR}/driver/include
|
|
${PROJECT_BINARY_DIR}/composable_kernel/include/utility
|
|
)
|
|
|
|
if(DEVICE_BACKEND STREQUAL "AMD")
|
|
include_directories(BEFORE
|
|
${PROJECT_SOURCE_DIR}/external/rocm/include
|
|
)
|
|
endif()
|
|
|
|
if(DEVICE_BACKEND STREQUAL "AMD")
|
|
configure_file("${PROJECT_SOURCE_DIR}/composable_kernel/include/utility/config.amd.hpp.in" "${PROJECT_BINARY_DIR}/composable_kernel/include/utility/config.hpp")
|
|
configure_file("${PROJECT_SOURCE_DIR}/composable_kernel/include/utility/float_type.amd.hpp.in" "${PROJECT_BINARY_DIR}/composable_kernel/include/utility/float_type.hpp")
|
|
configure_file("${PROJECT_SOURCE_DIR}/composable_kernel/include/utility/in_memory_operation.amd.hpp.in" "${PROJECT_BINARY_DIR}/composable_kernel/include/utility/in_memory_operation.hpp")
|
|
configure_file("${PROJECT_SOURCE_DIR}/composable_kernel/include/utility/synchronization.amd.hpp.in" "${PROJECT_BINARY_DIR}/composable_kernel/include/utility/synchronization.hpp")
|
|
elseif(DEVICE_BACKEND STREQUAL "NVIDIA")
|
|
configure_file("${PROJECT_SOURCE_DIR}/composable_kernel/include/utility/config.nvidia.hpp.in" "${PROJECT_BINARY_DIR}/composable_kernel/include/utility/config.hpp")
|
|
configure_file("${PROJECT_SOURCE_DIR}/composable_kernel/include/utility/float_type.nvidia.hpp.in" "${PROJECT_BINARY_DIR}/composable_kernel/include/utility/float_type.hpp")
|
|
configure_file("${PROJECT_SOURCE_DIR}/composable_kernel/include/utility/in_memory_operation.nvidia.hpp.in" "${PROJECT_BINARY_DIR}/composable_kernel/include/utility/in_memory_operation.hpp")
|
|
configure_file("${PROJECT_SOURCE_DIR}/composable_kernel/include/utility/synchronization.nvidia.hpp.in" "${PROJECT_BINARY_DIR}/composable_kernel/include/utility/synchronization.hpp")
|
|
endif()
|
|
|
|
add_subdirectory(driver)
|