80525dfcde
Through the new `-foffload-via-llvm` flag, CUDA kernels can now be lowered to the LLVM/Offload API. On the Clang side, this is simply done by using the OpenMP offload toolchain and emitting calls to `llvm*` functions to orchestrate the kernel launch rather than `cuda*` functions. These `llvm*` functions are implemented on top of the existing LLVM/Offload API. As we are about to redefine the Offload API, this wil help us in the design process as a second offload language. We do not support any CUDA APIs yet, however, we could: https://www.osti.gov/servlets/purl/1892137 For proper host execution we need to resurrect/rebase https://tianshilei.me/wp-content/uploads/2021/12/llpp-2021.pdf (which was designed for debugging). ``` ❯❯❯ cat test.cu extern "C" { void *llvm_omp_target_alloc_shared(size_t Size, int DeviceNum); void llvm_omp_target_free_shared(void *DevicePtr, int DeviceNum); } __global__ void square(int *A) { *A = 42; } int main(int argc, char **argv) { int DevNo = 0; int *Ptr = reinterpret_cast<int *>(llvm_omp_target_alloc_shared(4, DevNo)); *Ptr = 7; printf("Ptr %p, *Ptr %i\n", Ptr, *Ptr); square<<<1, 1>>>(Ptr); printf("Ptr %p, *Ptr %i\n", Ptr, *Ptr); llvm_omp_target_free_shared(Ptr, DevNo); } ❯❯❯ clang++ test.cu -O3 -o test123 -foffload-via-llvm --offload-arch=native ❯❯❯ llvm-objdump --offloading test123 test123: file format elf64-x86-64 OFFLOADING IMAGE [0]: kind elf arch gfx90a triple amdgcn-amd-amdhsa producer openmp ❯❯❯ LIBOMPTARGET_INFO=16 ./test123 Ptr 0x155448ac8000, *Ptr 7 Ptr 0x155448ac8000, *Ptr 42 ```
72 lines
1.7 KiB
CMake
72 lines
1.7 KiB
CMake
message(STATUS "Building offloading runtime library libomptarget.")
|
|
|
|
if(LIBOMP_STANDALONE)
|
|
set(LIBOMP ${LIBOMP_STANDALONE})
|
|
else()
|
|
set(LIBOMP omp)
|
|
endif()
|
|
|
|
add_llvm_library(omptarget
|
|
SHARED
|
|
|
|
device.cpp
|
|
interface.cpp
|
|
omptarget.cpp
|
|
OffloadRTL.cpp
|
|
LegacyAPI.cpp
|
|
PluginManager.cpp
|
|
DeviceImage.cpp
|
|
|
|
OpenMP/API.cpp
|
|
OpenMP/Mapping.cpp
|
|
OpenMP/InteropAPI.cpp
|
|
OpenMP/OMPT/Callback.cpp
|
|
|
|
KernelLanguage/API.cpp
|
|
|
|
ADDITIONAL_HEADER_DIRS
|
|
${LIBOMPTARGET_INCLUDE_DIR}
|
|
${LIBOMPTARGET_BINARY_INCLUDE_DIR}
|
|
|
|
LINK_COMPONENTS
|
|
FrontendOpenMP
|
|
Support
|
|
Object
|
|
|
|
LINK_LIBS
|
|
PUBLIC
|
|
${LIBOMP}
|
|
|
|
NO_INSTALL_RPATH
|
|
BUILDTREE_ONLY
|
|
)
|
|
target_include_directories(omptarget PRIVATE
|
|
${LIBOMPTARGET_INCLUDE_DIR} ${LIBOMPTARGET_BINARY_INCLUDE_DIR}
|
|
)
|
|
|
|
if (LIBOMP_HAVE_VERSION_SCRIPT_FLAG)
|
|
target_link_libraries(omptarget PRIVATE
|
|
"-Wl,--version-script=${CMAKE_CURRENT_SOURCE_DIR}/exports")
|
|
endif()
|
|
|
|
# Define the TARGET_NAME and DEBUG_PREFIX.
|
|
target_compile_definitions(omptarget PRIVATE
|
|
TARGET_NAME=omptarget
|
|
DEBUG_PREFIX="omptarget"
|
|
)
|
|
|
|
foreach(plugin IN LISTS LIBOMPTARGET_PLUGINS_TO_BUILD)
|
|
target_link_libraries(omptarget PRIVATE omptarget.rtl.${plugin})
|
|
endforeach()
|
|
|
|
target_compile_options(omptarget PRIVATE ${offload_compile_flags})
|
|
target_link_options(omptarget PRIVATE ${offload_link_flags})
|
|
|
|
# libomptarget.so needs to be aware of where the plugins live as they
|
|
# are now separated in the build directory.
|
|
set_target_properties(omptarget PROPERTIES
|
|
POSITION_INDEPENDENT_CODE ON
|
|
INSTALL_RPATH "$ORIGIN"
|
|
BUILD_RPATH "$ORIGIN:${CMAKE_CURRENT_BINARY_DIR}/..")
|
|
install(TARGETS omptarget LIBRARY COMPONENT omptarget DESTINATION "${OFFLOAD_INSTALL_LIBDIR}")
|