CUDA 13.2 samples update (#432)

- Added Python samples for CUDA Python 1.0 release
- Renamed top-level `Samples` directory to `cpp` to accommodate Python samples.
This commit is contained in:
Dheemanth
2026-05-13 15:13:18 -07:00
committed by GitHub
parent a4526d5229
commit aeab82ff30
2072 changed files with 15077 additions and 1970 deletions

View File

@@ -0,0 +1,67 @@
cmake_minimum_required(VERSION 3.20)
list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/Modules")
project(vectorAddMMAP LANGUAGES C CXX CUDA)
find_package(CUDAToolkit REQUIRED)
set(CMAKE_POSITION_INDEPENDENT_CODE ON)
set(CMAKE_CUDA_ARCHITECTURES 75 80 86 87 89 90 100 110 120)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
if(ENABLE_CUDA_DEBUG)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -G") # enable cuda-gdb (may significantly affect performance on some targets)
else()
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -lineinfo") # add line information to all builds for debug tools (exclusive to -G option)
endif()
# Include directories and libraries
include_directories(../../../Common)
# Source file
# Add target for vectorAddMMAP
if(CMAKE_SYSTEM_PROCESSOR STREQUAL "aarch64")
message(STATUS "Will not build sample vectorAddMMAP - not supported on aarch64")
else()
add_executable(vectorAddMMAP vectorAddMMAP.cpp multidevicealloc_memmap.cpp)
target_compile_options(vectorAddMMAP PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:--extended-lambda>)
target_compile_features(vectorAddMMAP PRIVATE cxx_std_17 cuda_std_17)
set_target_properties(vectorAddMMAP PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
target_include_directories(vectorAddMMAP PRIVATE
${CUDAToolkit_INCLUDE_DIRS}
)
target_link_libraries(vectorAddMMAP PUBLIC
CUDA::cuda_driver
)
set(CUDA_FATBIN_FILE "${CMAKE_CURRENT_BINARY_DIR}/vectorAdd_kernel64.fatbin")
set(CUDA_KERNEL_SOURCE "${CMAKE_CURRENT_SOURCE_DIR}/vectorAdd_kernel.cu")
# Construct GENCODE_FLAGS explicitly from CUDA architectures
set(GENCODE_FLAGS "")
foreach(arch ${CMAKE_CUDA_ARCHITECTURES})
list(APPEND GENCODE_FLAGS "-gencode=arch=compute_${arch},code=sm_${arch}")
endforeach()
add_custom_command(
OUTPUT ${CUDA_FATBIN_FILE}
COMMAND ${CMAKE_CUDA_COMPILER} ${INCLUDES} ${ALL_CCFLAGS} -Wno-deprecated-gpu-targets ${GENCODE_FLAGS} -o ${CUDA_FATBIN_FILE} -fatbin ${CUDA_KERNEL_SOURCE}
DEPENDS ${CUDA_KERNEL_SOURCE}
COMMENT "Building CUDA fatbin: ${CUDA_FATBIN_FILE}"
)
# Create a dummy target for fatbin generation
add_custom_target(generate_fatbin_vectorAddMMAP ALL DEPENDS ${CUDA_FATBIN_FILE})
# Ensure matrixMulDrv depends on the fatbin
add_dependencies(vectorAddMMAP generate_fatbin_vectorAddMMAP)
endif()
# Include installation configuration
include(${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/InstallSamples.cmake)
setup_samples_install()