NASP/cuda-samples
47
0
Fork 0
mirror of https://github.com/NVIDIA/cuda-samples.git synced 2025-07-01 20:20:29 +08:00
Code Issues Packages Projects Releases Wiki Activity

Update gencode arguments for separate kernel fatbin builds

Browse Source
This commit is contained in:
Rob Armstrong 2025-03-26 09:28:37 -07:00
parent 06d72496c2
commit e214cd29aa
8 changed files with 38 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
Samples/0_Introduction/matrixMulDrv/CMakeLists.txt
View File

@ -38,6 +38,12 @@ target_link_libraries(matrixMulDrv PUBLIC
set(CUDA_FATBIN_FILE "${CMAKE_CURRENT_BINARY_DIR}/matrixMul_kernel64.fatbin")
set(CUDA_KERNEL_SOURCE "${CMAKE_CURRENT_SOURCE_DIR}/matrixMul_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}

6
Samples/0_Introduction/simpleDrvRuntime/CMakeLists.txt
View File

@ -40,6 +40,12 @@ target_link_libraries(simpleDrvRuntime PUBLIC
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}

6
Samples/0_Introduction/simpleTextureDrv/CMakeLists.txt
View File

@ -37,6 +37,12 @@ target_link_libraries(simpleTextureDrv PUBLIC
set(CUDA_FATBIN_FILE "${CMAKE_CURRENT_BINARY_DIR}/simpleTexture_kernel64.fatbin")
set(CUDA_KERNEL_SOURCE "${CMAKE_CURRENT_SOURCE_DIR}/simpleTexture_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}

6
Samples/0_Introduction/vectorAddDrv/CMakeLists.txt
View File

@ -37,6 +37,12 @@ target_link_libraries(vectorAddDrv PUBLIC
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}

6
Samples/0_Introduction/vectorAddMMAP/CMakeLists.txt
View File

@ -40,6 +40,12 @@ else()
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}

6
Samples/2_Concepts_and_Techniques/threadMigration/CMakeLists.txt
View File

@ -42,6 +42,12 @@ endif()
set(CUDA_FATBIN_FILE "${CMAKE_CURRENT_BINARY_DIR}/threadMigration_kernel64.fatbin")
set(CUDA_KERNEL_SOURCE "${CMAKE_CURRENT_SOURCE_DIR}/threadMigration_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}

2
Samples/3_CUDA_Features/memMapIPCDrv/CMakeLists.txt
View File

@ -44,7 +44,7 @@ set(CUDA_KERNEL_SOURCE "${CMAKE_CURRENT_SOURCE_DIR}/memMapIpc_kernel.cu")
add_custom_command(
OUTPUT ${CUDA_PTX_FILE}
COMMAND ${CMAKE_CUDA_COMPILER} ${INCLUDES} ${ALL_CCFLAGS} -Wno-deprecated-gpu-targets ${GENCODE_FLAGS} -o ${CUDA_PTX_FILE} -ptx ${CUDA_KERNEL_SOURCE}
COMMAND ${CMAKE_CUDA_COMPILER} ${INCLUDES} ${ALL_CCFLAGS} -Wno-deprecated-gpu-targets -o ${CUDA_PTX_FILE} -ptx ${CUDA_KERNEL_SOURCE}
DEPENDS ${CUDA_KERNEL_SOURCE}
COMMENT "Building CUDA PTX: ${CUDA_PTX_FILE}"
)

2
Samples/3_CUDA_Features/ptxjit/CMakeLists.txt
View File

@ -41,7 +41,7 @@ set(CUDA_KERNEL_SOURCE "${CMAKE_CURRENT_SOURCE_DIR}/ptxjit_kernel.cu")
add_custom_command(
OUTPUT ${CUDA_PTX_FILE}
COMMAND ${CMAKE_CUDA_COMPILER} ${INCLUDES} ${ALL_CCFLAGS} -Wno-deprecated-gpu-targets ${GENCODE_FLAGS} -o ${CUDA_PTX_FILE} -ptx ${CUDA_KERNEL_SOURCE}
COMMAND ${CMAKE_CUDA_COMPILER} ${INCLUDES} ${ALL_CCFLAGS} -Wno-deprecated-gpu-targets -o ${CUDA_PTX_FILE} -ptx ${CUDA_KERNEL_SOURCE}
DEPENDS ${CUDA_KERNEL_SOURCE}
COMMENT "Building CUDA PTX: ${CUDA_PTX_FILE}"
)