mirror of
https://github.com/Kitware/CMake.git
synced 2026-04-28 01:49:23 -05:00
Brad King
d1b48bfabd
CUDA 11.6 added the `nvcc -arch=native` flag to automatically compile for the host GPUs' architectures. Add support for specifying this special `native` value in `CMAKE_CUDA_ARCHITECTURES` and `CUDA_ARCHITECTURES`. During the compiler ABI detection step, detect the native architectures so we can pass them explicitly when using Clang or older versions of nvcc. Fixes: #22375
7 lines
395 B
CMake
7 lines
395 B
CMake
set(CMAKE_CUDA_ARCHITECTURES "native")
|
|
enable_language(CUDA)
|
|
message(STATUS "CMAKE_CUDA_ARCHITECTURES='${CMAKE_CUDA_ARCHITECTURES}'")
|
|
message(STATUS "CMAKE_CUDA_ARCHITECTURES_ALL='${CMAKE_CUDA_ARCHITECTURES_ALL}'")
|
|
message(STATUS "CMAKE_CUDA_ARCHITECTURES_ALL_MAJOR='${CMAKE_CUDA_ARCHITECTURES_ALL_MAJOR}'")
|
|
message(STATUS "CMAKE_CUDA_ARCHITECTURES_NATIVE='${CMAKE_CUDA_ARCHITECTURES_NATIVE}'")
|