FastDeploy 在CUDA 12(Windows) 上构建失败:nvcc fatal : Unsupported gpu architecture 'compute

在CUDA 12(Windows) 上构建失败:nvcc fatal : Unsupported gpu architecture 'compute_35'

Open justghostof opened this issue 1 year ago • 6 comments

环境

【FastDeploy版本】：73f42e36beb959717f8a901251a12b0ff162e8d3
【编译命令】:

cmake .. -G "Visual Studio 17 2022" -A x64 -DENABLE_ORT_BACKEND=ON -DENABLE_PADDLE_BACKEND=ON -DENABLE_OPENVINO_BACKEND=ON -DENABLE_TRT_BACKEND=ON -DENABLE_VISION=ON -DENABLE_TEXT=ON -DWITH_GPU=ON -DTRT_DIRECTORY="C:\Program Files\NVIDIA GPU Computing Toolkit\TensorRT\8.6.1" -DCUDA_DIRECTORY="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1" -DCMAKE_INSTALL_PREFIX="C:\Users\MWX\FastDeploy\compiled_fastdeploy"
msbuild fastdeploy.sln /m /p:Configuration=Release /p:Platform=x64

【系统平台】:Windows x64(Windows11)
【硬件】： Nvidia GPU 3070TI Laptop， CUDA 12.1 CUDNN 8.9.2
【编译语言】： C++

问题日志及出现问题的操作流程

附上详细的问题日志有助于快速定位分析
【编译失败】

> msbuild fastdeploy.sln /m /p:Configuration=Release /p:Platform=x64
...
 23>CudaBuildCore:
         nvcc fatal   : Unsupported gpu architecture 'compute_35'
    23>C:\Program Files\Microsoft Visual Studio\2022\Community\MSBuild\Microsoft\VC\v170\BuildCustomizations\CUDA 12.1.
       targets(799,9): error MSB3721: 命令“"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\bin\nvcc.exe"  --use
       -local-env -ccbin "C:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\bin\HostX64
       \x64" -x cu   -IC:\Users\MWX\FastDeploy\. -IC:\Users\MWX\FastDeploy\build -IC:\Users\MWX\FastDeploy\third_party\
       eigen -IC:\Users\MWX\FastDeploy\build\third_libs\install\onnxruntime\include -IC:\Users\MWX\FastDeploy\build\thi
       rd_libs\install\paddle_inference -IC:\Users\MWX\FastDeploy\build\third_libs\install\openvino\runtime\include -IC
       :\Users\MWX\FastDeploy\build\third_libs\install\openvino\runtime\include\ie -I"C:\Program Files\NVIDIA GPU Compu
       ting Toolkit\CUDA\v12.1\include" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\TensorRT\8.6.1\include" -IC:\U
       sers\MWX\FastDeploy\.\fastdeploy\runtime\backends\tensorrt\common -I"C:\Users\MWX\FastDeploy\third_party\yaml-cp
       p\include" -IC:\Users\MWX\FastDeploy\build\third_libs\install\fast_tokenizer\include -IC:\Users\MWX\FastDeploy\b
       uild\third_libs\install\fast_tokenizer\third_party\include -IC:\Users\MWX\FastDeploy\build\third_libs\install\pa
       ddle2onnx\include -IC:\Users\MWX\FastDeploy\build\third_libs\install\opencv\build\include -IC:\Users\MWX\FastDep
       loy\build\third_libs\install\opencv\build\include\opencv -IC:\Users\MWX\FastDeploy\build\third_libs\install\open
       vino\runtime\3rdparty\tbb\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\include"     --kee
       p-dir x64\Release  -maxrregcount=0   --machine 64 --compile -cudart static -gencode arch=compute_35,code=sm_35 -
       gencode arch=compute_50,code=sm_50 -gencode arch=compute_52,code=sm_52 -gencode arch=compute_60,code=sm_60 -genc
       ode arch=compute_61,code=sm_61 -gencode arch=compute_70,code=sm_70 -gencode arch=compute_75,code=sm_75 -gencode
       arch=compute_80,code=sm_80 -gencode arch=compute_86,code=sm_86 -w --expt-relaxed-constexpr --expt-extended-lambd
       a -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj -Ob2 /wd4251"   -D_WINDOWS -DNDEBUG -DYAML_CPP_DLL -DFASTDEP
       LOY_LIB -DCMAKE_BUILD_TYPE=Release -DEIGEN_STRONG_INLINE=inline -DENABLE_ORT_BACKEND -DENABLE_PADDLE_BACKEND -DE
       NABLE_OPENVINO_BACKEND -DWITH_GPU -DENABLE_NVJPEG -DENABLE_TRT_BACKEND -DENABLE_VISION -DENABLE_TEXT -DENABLE_PA
       DDLE2ONNX -D__TBB_NO_IMPLICIT_LINKAGE=1 -D"CMAKE_INTDIR=\"Release\"" -Dfastdeploy_EXPORTS -D"__REL_FILE__=\"fast
       deploy/vision/common/processors/normalize_and_permute.cu\"" -D_WINDLL -D_MBCS -DWIN32 -D_WINDOWS -DNDEBUG -DYAML
       _CPP_DLL -DFASTDEPLOY_LIB -DCMAKE_BUILD_TYPE=Release -DEIGEN_STRONG_INLINE=inline -DENABLE_ORT_BACKEND -DENABLE_
       PADDLE_BACKEND -DENABLE_OPENVINO_BACKEND -DWITH_GPU -DENABLE_NVJPEG -DENABLE_TRT_BACKEND -DENABLE_VISION -DENABL
       E_TEXT -DENABLE_PADDLE2ONNX -D__TBB_NO_IMPLICIT_LINKAGE=1 -D"CMAKE_INTDIR=\"Release\"" -Dfastdeploy_EXPORTS -Xco
       mpiler "/EHsc /W0 /nologo /O2 /FS   /MD /GR" -Xcompiler "/Fdfastdeploy.dir\Release\vc143.pdb" -o fastdeploy.dir\
       Release\/fastdeploy/vision/common/processors/normalize_and_permute.cu.obj "C:\Users\MWX\FastDeploy\fastdeploy\vi
       sion\common\processors\normalize_and_permute.cu"”已退出，返回代码为 1。 [C:\Users\MWX\FastDeploy\build\fastdeploy.vcxproj]
    23>已完成生成项目“C:\Users\MWX\FastDeploy\build\fastdeploy.vcxproj”(CudaBuildCore 个目标)的操作 - 失败。
    23>CudaBuildCore:
         nvcc fatal   : Unsupported gpu architecture 'compute_35'
    23>CudaBuildCore:
         nvcc fatal   : Unsupported gpu architecture 'compute_35'
    23>C:\Program Files\Microsoft Visual Studio\2022\Community\MSBuild\Microsoft\VC\v170\BuildCustomizations\CUDA 12.1.
       targets(799,9): error MSB3721: 命令“"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\bin\nvcc.exe"  --use
       -local-env -ccbin "C:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\bin\HostX64
       \x64" -x cu   -IC:\Users\MWX\FastDeploy\. -IC:\Users\MWX\FastDeploy\build -IC:\Users\MWX\FastDeploy\third_party\
       eigen -IC:\Users\MWX\FastDeploy\build\third_libs\install\onnxruntime\include -IC:\Users\MWX\FastDeploy\build\thi
       rd_libs\install\paddle_inference -IC:\Users\MWX\FastDeploy\build\third_libs\install\openvino\runtime\include -IC
       :\Users\MWX\FastDeploy\build\third_libs\install\openvino\runtime\include\ie -I"C:\Program Files\NVIDIA GPU Compu
       ting Toolkit\CUDA\v12.1\include" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\TensorRT\8.6.1\include" -IC:\U
       sers\MWX\FastDeploy\.\fastdeploy\runtime\backends\tensorrt\common -I"C:\Users\MWX\FastDeploy\third_party\yaml-cp
       p\include" -IC:\Users\MWX\FastDeploy\build\third_libs\install\fast_tokenizer\include -IC:\Users\MWX\FastDeploy\b
       uild\third_libs\install\fast_tokenizer\third_party\include -IC:\Users\MWX\FastDeploy\build\third_libs\install\pa
       ddle2onnx\include -IC:\Users\MWX\FastDeploy\build\third_libs\install\opencv\build\include -IC:\Users\MWX\FastDep
       loy\build\third_libs\install\opencv\build\include\opencv -IC:\Users\MWX\FastDeploy\build\third_libs\install\open
       vino\runtime\3rdparty\tbb\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\include"     --kee
       p-dir x64\Release  -maxrregcount=0   --machine 64 --compile -cudart static -gencode arch=compute_35,code=sm_35 -
       gencode arch=compute_50,code=sm_50 -gencode arch=compute_52,code=sm_52 -gencode arch=compute_60,code=sm_60 -genc
       ode arch=compute_61,code=sm_61 -gencode arch=compute_70,code=sm_70 -gencode arch=compute_75,code=sm_75 -gencode
       arch=compute_80,code=sm_80 -gencode arch=compute_86,code=sm_86 -w --expt-relaxed-constexpr --expt-extended-lambd
       a -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj -Ob2 /wd4251"   -D_WINDOWS -DNDEBUG -DYAML_CPP_DLL -DFASTDEP
       LOY_LIB -DCMAKE_BUILD_TYPE=Release -DEIGEN_STRONG_INLINE=inline -DENABLE_ORT_BACKEND -DENABLE_PADDLE_BACKEND -DE
       NABLE_OPENVINO_BACKEND -DWITH_GPU -DENABLE_NVJPEG -DENABLE_TRT_BACKEND -DENABLE_VISION -DENABLE_TEXT -DENABLE_PA
       DDLE2ONNX -D__TBB_NO_IMPLICIT_LINKAGE=1 -D"CMAKE_INTDIR=\"Release\"" -Dfastdeploy_EXPORTS -D"__REL_FILE__=\"fast
       deploy/runtime/backends/common/cuda/adaptive_pool2d_kernel.cu\"" -D_WINDLL -D_MBCS -DWIN32 -D_WINDOWS -DNDEBUG -
       DYAML_CPP_DLL -DFASTDEPLOY_LIB -DCMAKE_BUILD_TYPE=Release -DEIGEN_STRONG_INLINE=inline -DENABLE_ORT_BACKEND -DEN
       ABLE_PADDLE_BACKEND -DENABLE_OPENVINO_BACKEND -DWITH_GPU -DENABLE_NVJPEG -DENABLE_TRT_BACKEND -DENABLE_VISION -D
       ENABLE_TEXT -DENABLE_PADDLE2ONNX -D__TBB_NO_IMPLICIT_LINKAGE=1 -D"CMAKE_INTDIR=\"Release\"" -Dfastdeploy_EXPORTS
        -Xcompiler "/EHsc /W0 /nologo /O2 /FS   /MD /GR" -Xcompiler "/Fdfastdeploy.dir\Release\vc143.pdb" -o fastdeploy
       .dir\Release\adaptive_pool2d_kernel.obj "C:\Users\MWX\FastDeploy\fastdeploy\runtime\backends\common\cuda\adaptiv
       e_pool2d_kernel.cu"”已退出，返回代码为 1。 [C:\Users\MWX\FastDeploy\build\fastdeploy.vcxproj]
    23>C:\Program Files\Microsoft Visual Studio\2022\Community\MSBuild\Microsoft\VC\v170\BuildCustomizations\CUDA 12.1.
       targets(799,9): error MSB3721: 命令“"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\bin\nvcc.exe"  --use
       -local-env -ccbin "C:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\bin\HostX64
       \x64" -x cu   -IC:\Users\MWX\FastDeploy\. -IC:\Users\MWX\FastDeploy\build -IC:\Users\MWX\FastDeploy\third_party\
       eigen -IC:\Users\MWX\FastDeploy\build\third_libs\install\onnxruntime\include -IC:\Users\MWX\FastDeploy\build\thi
       rd_libs\install\paddle_inference -IC:\Users\MWX\FastDeploy\build\third_libs\install\openvino\runtime\include -IC
       :\Users\MWX\FastDeploy\build\third_libs\install\openvino\runtime\include\ie -I"C:\Program Files\NVIDIA GPU Compu
       ting Toolkit\CUDA\v12.1\include" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\TensorRT\8.6.1\include" -IC:\U
       sers\MWX\FastDeploy\.\fastdeploy\runtime\backends\tensorrt\common -I"C:\Users\MWX\FastDeploy\third_party\yaml-cp
       p\include" -IC:\Users\MWX\FastDeploy\build\third_libs\install\fast_tokenizer\include -IC:\Users\MWX\FastDeploy\b
       uild\third_libs\install\fast_tokenizer\third_party\include -IC:\Users\MWX\FastDeploy\build\third_libs\install\pa
       ddle2onnx\include -IC:\Users\MWX\FastDeploy\build\third_libs\install\opencv\build\include -IC:\Users\MWX\FastDep
       loy\build\third_libs\install\opencv\build\include\opencv -IC:\Users\MWX\FastDeploy\build\third_libs\install\open
       vino\runtime\3rdparty\tbb\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\include"     --kee
       p-dir x64\Release  -maxrregcount=0   --machine 64 --compile -cudart static -gencode arch=compute_35,code=sm_35 -
       gencode arch=compute_50,code=sm_50 -gencode arch=compute_52,code=sm_52 -gencode arch=compute_60,code=sm_60 -genc
       ode arch=compute_61,code=sm_61 -gencode arch=compute_70,code=sm_70 -gencode arch=compute_75,code=sm_75 -gencode
       arch=compute_80,code=sm_80 -gencode arch=compute_86,code=sm_86 -w --expt-relaxed-constexpr --expt-extended-lambd
       a -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj -Ob2 /wd4251"   -D_WINDOWS -DNDEBUG -DYAML_CPP_DLL -DFASTDEP
       LOY_LIB -DCMAKE_BUILD_TYPE=Release -DEIGEN_STRONG_INLINE=inline -DENABLE_ORT_BACKEND -DENABLE_PADDLE_BACKEND -DE
       NABLE_OPENVINO_BACKEND -DWITH_GPU -DENABLE_NVJPEG -DENABLE_TRT_BACKEND -DENABLE_VISION -DENABLE_TEXT -DENABLE_PA
       DDLE2ONNX -D__TBB_NO_IMPLICIT_LINKAGE=1 -D"CMAKE_INTDIR=\"Release\"" -Dfastdeploy_EXPORTS -D"__REL_FILE__=\"fast
       deploy/vision/common/processors/normalize.cu\"" -D_WINDLL -D_MBCS -DWIN32 -D_WINDOWS -DNDEBUG -DYAML_CPP_DLL -DF
       ASTDEPLOY_LIB -DCMAKE_BUILD_TYPE=Release -DEIGEN_STRONG_INLINE=inline -DENABLE_ORT_BACKEND -DENABLE_PADDLE_BACKE
       ND -DENABLE_OPENVINO_BACKEND -DWITH_GPU -DENABLE_NVJPEG -DENABLE_TRT_BACKEND -DENABLE_VISION -DENABLE_TEXT -DENA
       BLE_PADDLE2ONNX -D__TBB_NO_IMPLICIT_LINKAGE=1 -D"CMAKE_INTDIR=\"Release\"" -Dfastdeploy_EXPORTS -Xcompiler "/EHs
       c /W0 /nologo /O2 /FS   /MD /GR" -Xcompiler "/Fdfastdeploy.dir\Release\vc143.pdb" -o fastdeploy.dir\Release\/fas
       tdeploy/vision/common/processors/normalize.cu.obj "C:\Users\MWX\FastDeploy\fastdeploy\vision\common\processors\n
       ormalize.cu"”已退出，返回代码为 1。 [C:\Users\MWX\FastDeploy\build\fastdeploy.vcxproj]
    23>已完成生成项目“C:\Users\MWX\FastDeploy\build\fastdeploy.vcxproj”(CudaBuildCore 个目标)的操作 - 失败。
    23>已完成生成项目“C:\Users\MWX\FastDeploy\build\fastdeploy.vcxproj”(CudaBuildCore 个目标)的操作 - 失败。
    23>CudaBuildCore:
         cmd.exe /C "C:\Users\MWX\AppData\Local\Temp\tmp3f74f4dfd91c406496900b81aaa10ee5.cmd"
         "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\bin\nvcc.exe"  --use-local-env -ccbin "C:\Program Fi
         les\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\bin\HostX64\x64" -x cu   -IC:\Users\MWX\F
         astDeploy\. -IC:\Users\MWX\FastDeploy\build -IC:\Users\MWX\FastDeploy\third_party\eigen -IC:\Users\MWX\FastDep
         loy\build\third_libs\install\onnxruntime\include -IC:\Users\MWX\FastDeploy\build\third_libs\install\paddle_inf
         erence -IC:\Users\MWX\FastDeploy\build\third_libs\install\openvino\runtime\include -IC:\Users\MWX\FastDeploy\b
         uild\third_libs\install\openvino\runtime\include\ie -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.
         1\include" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\TensorRT\8.6.1\include" -IC:\Users\MWX\FastDeploy\
         .\fastdeploy\runtime\backends\tensorrt\common -I"C:\Users\MWX\FastDeploy\third_party\yaml-cpp\include" -IC:\Us
         ers\MWX\FastDeploy\build\third_libs\install\fast_tokenizer\include -IC:\Users\MWX\FastDeploy\build\third_libs\
         install\fast_tokenizer\third_party\include -IC:\Users\MWX\FastDeploy\build\third_libs\install\paddle2onnx\incl
         ude -IC:\Users\MWX\FastDeploy\build\third_libs\install\opencv\build\include -IC:\Users\MWX\FastDeploy\build\th
         ird_libs\install\opencv\build\include\opencv -IC:\Users\MWX\FastDeploy\build\third_libs\install\openvino\runti
         me\3rdparty\tbb\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\include"     --keep-dir x6
         4\Release  -maxrregcount=0   --machine 64 --compile -cudart static -gencode arch=compute_35,code=sm_35 -gencod
         e arch=compute_50,code=sm_50 -gencode arch=compute_52,code=sm_52 -gencode arch=compute_60,code=sm_60 -gencode
         arch=compute_61,code=sm_61 -gencode arch=compute_70,code=sm_70 -gencode arch=compute_75,code=sm_75 -gencode ar
         ch=compute_80,code=sm_80 -gencode arch=compute_86,code=sm_86 -w --expt-relaxed-constexpr --expt-extended-lambd
         a -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj -Ob2 /wd4251"   -D_WINDOWS -DNDEBUG -DYAML_CPP_DLL -DFASTD
         EPLOY_LIB -DCMAKE_BUILD_TYPE=Release -DEIGEN_STRONG_INLINE=inline -DENABLE_ORT_BACKEND -DENABLE_PADDLE_BACKEND
          -DENABLE_OPENVINO_BACKEND -DWITH_GPU -DENABLE_NVJPEG -DENABLE_TRT_BACKEND -DENABLE_VISION -DENABLE_TEXT -DENA
         BLE_PADDLE2ONNX -D__TBB_NO_IMPLICIT_LINKAGE=1 -D"CMAKE_INTDIR=\"Release\"" -Dfastdeploy_EXPORTS -D"__REL_FILE_
         _=\"fastdeploy/vision/utils/yolo_preprocess.cu\"" -D_WINDLL -D_MBCS -DWIN32 -D_WINDOWS -DNDEBUG -DYAML_CPP_DLL
          -DFASTDEPLOY_LIB -DCMAKE_BUILD_TYPE=Release -DEIGEN_STRONG_INLINE=inline -DENABLE_ORT_BACKEND -DENABLE_PADDLE
         _BACKEND -DENABLE_OPENVINO_BACKEND -DWITH_GPU -DENABLE_NVJPEG -DENABLE_TRT_BACKEND -DENABLE_VISION -DENABLE_TE
         XT -DENABLE_PADDLE2ONNX -D__TBB_NO_IMPLICIT_LINKAGE=1 -D"CMAKE_INTDIR=\"Release\"" -Dfastdeploy_EXPORTS -Xcomp
         iler "/EHsc /W0 /nologo /O2 /FS   /MD /GR" -Xcompiler "/Fdfastdeploy.dir\Release\vc143.pdb" -o fastdeploy.dir\
         Release\yolo_preprocess.obj "C:\Users\MWX\FastDeploy\fastdeploy\vision\utils\yolo_preprocess.cu"
    23>CudaBuildCore:
         cmd.exe /C "C:\Users\MWX\AppData\Local\Temp\tmp1a6ff40ea2a641169ed293d39d93eee7.cmd"
         "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\bin\nvcc.exe"  --use-local-env -ccbin "C:\Program Fi
         les\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\bin\HostX64\x64" -x cu   -IC:\Users\MWX\F
         astDeploy\. -IC:\Users\MWX\FastDeploy\build -IC:\Users\MWX\FastDeploy\third_party\eigen -IC:\Users\MWX\FastDep
         loy\build\third_libs\install\onnxruntime\include -IC:\Users\MWX\FastDeploy\build\third_libs\install\paddle_inf
         erence -IC:\Users\MWX\FastDeploy\build\third_libs\install\openvino\runtime\include -IC:\Users\MWX\FastDeploy\b
         uild\third_libs\install\openvino\runtime\include\ie -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.
         1\include" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\TensorRT\8.6.1\include" -IC:\Users\MWX\FastDeploy\
         .\fastdeploy\runtime\backends\tensorrt\common -I"C:\Users\MWX\FastDeploy\third_party\yaml-cpp\include" -IC:\Us
         ers\MWX\FastDeploy\build\third_libs\install\fast_tokenizer\include -IC:\Users\MWX\FastDeploy\build\third_libs\
         install\fast_tokenizer\third_party\include -IC:\Users\MWX\FastDeploy\build\third_libs\install\paddle2onnx\incl
         ude -IC:\Users\MWX\FastDeploy\build\third_libs\install\opencv\build\include -IC:\Users\MWX\FastDeploy\build\th
         ird_libs\install\opencv\build\include\opencv -IC:\Users\MWX\FastDeploy\build\third_libs\install\openvino\runti
         me\3rdparty\tbb\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\include"     --keep-dir x6
         4\Release  -maxrregcount=0   --machine 64 --compile -cudart static -gencode arch=compute_35,code=sm_35 -gencod
         e arch=compute_50,code=sm_50 -gencode arch=compute_52,code=sm_52 -gencode arch=compute_60,code=sm_60 -gencode
         arch=compute_61,code=sm_61 -gencode arch=compute_70,code=sm_70 -gencode arch=compute_75,code=sm_75 -gencode ar
         ch=compute_80,code=sm_80 -gencode arch=compute_86,code=sm_86 -w --expt-relaxed-constexpr --expt-extended-lambd
         a -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj -Ob2 /wd4251"   -D_WINDOWS -DNDEBUG -DYAML_CPP_DLL -DFASTD
         EPLOY_LIB -DCMAKE_BUILD_TYPE=Release -DEIGEN_STRONG_INLINE=inline -DENABLE_ORT_BACKEND -DENABLE_PADDLE_BACKEND
          -DENABLE_OPENVINO_BACKEND -DWITH_GPU -DENABLE_NVJPEG -DENABLE_TRT_BACKEND -DENABLE_VISION -DENABLE_TEXT -DENA
         BLE_PADDLE2ONNX -D__TBB_NO_IMPLICIT_LINKAGE=1 -D"CMAKE_INTDIR=\"Release\"" -Dfastdeploy_EXPORTS -D"__REL_FILE_
         _=\"fastdeploy/function/cuda_cast.cu\"" -D_WINDLL -D_MBCS -DWIN32 -D_WINDOWS -DNDEBUG -DYAML_CPP_DLL -DFASTDEP
         LOY_LIB -DCMAKE_BUILD_TYPE=Release -DEIGEN_STRONG_INLINE=inline -DENABLE_ORT_BACKEND -DENABLE_PADDLE_BACKEND -
         DENABLE_OPENVINO_BACKEND -DWITH_GPU -DENABLE_NVJPEG -DENABLE_TRT_BACKEND -DENABLE_VISION -DENABLE_TEXT -DENABL
         E_PADDLE2ONNX -D__TBB_NO_IMPLICIT_LINKAGE=1 -D"CMAKE_INTDIR=\"Release\"" -Dfastdeploy_EXPORTS -Xcompiler "/EHs
         c /W0 /nologo /O2 /FS   /MD /GR" -Xcompiler "/Fdfastdeploy.dir\Release\vc143.pdb" -o fastdeploy.dir\Release\cu
         da_cast.obj "C:\Users\MWX\FastDeploy\fastdeploy\function\cuda_cast.cu"
    23>CudaBuildCore:

         C:\Users\MWX\FastDeploy\build>"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\bin\nvcc.exe"  --use-l
         ocal-env -ccbin "C:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\bin\HostX64
         \x64" -x cu   -IC:\Users\MWX\FastDeploy\. -IC:\Users\MWX\FastDeploy\build -IC:\Users\MWX\FastDeploy\third_part
         y\eigen -IC:\Users\MWX\FastDeploy\build\third_libs\install\onnxruntime\include -IC:\Users\MWX\FastDeploy\build
         \third_libs\install\paddle_inference -IC:\Users\MWX\FastDeploy\build\third_libs\install\openvino\runtime\inclu
         de -IC:\Users\MWX\FastDeploy\build\third_libs\install\openvino\runtime\include\ie -I"C:\Program Files\NVIDIA G
         PU Computing Toolkit\CUDA\v12.1\include" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\TensorRT\8.6.1\inclu
         de" -IC:\Users\MWX\FastDeploy\.\fastdeploy\runtime\backends\tensorrt\common -I"C:\Users\MWX\FastDeploy\third_p
         arty\yaml-cpp\include" -IC:\Users\MWX\FastDeploy\build\third_libs\install\fast_tokenizer\include -IC:\Users\MW
         X\FastDeploy\build\third_libs\install\fast_tokenizer\third_party\include -IC:\Users\MWX\FastDeploy\build\third
         _libs\install\paddle2onnx\include -IC:\Users\MWX\FastDeploy\build\third_libs\install\opencv\build\include -IC:
         \Users\MWX\FastDeploy\build\third_libs\install\opencv\build\include\opencv -IC:\Users\MWX\FastDeploy\build\thi
         rd_libs\install\openvino\runtime\3rdparty\tbb\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v1
         2.1\include"     --keep-dir x64\Release  -maxrregcount=0   --machine 64 --compile -cudart static -gencode arch
         =compute_35,code=sm_35 -gencode arch=compute_50,code=sm_50 -gencode arch=compute_52,code=sm_52 -gencode arch=c
         ompute_60,code=sm_60 -gencode arch=compute_61,code=sm_61 -gencode arch=compute_70,code=sm_70 -gencode arch=com
         pute_75,code=sm_75 -gencode arch=compute_80,code=sm_80 -gencode arch=compute_86,code=sm_86 -w --expt-relaxed-c
         onstexpr --expt-extended-lambda -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj -Ob2 /wd4251"   -D_WINDOWS -
         DNDEBUG -DYAML_CPP_DLL -DFASTDEPLOY_LIB -DCMAKE_BUILD_TYPE=Release -DEIGEN_STRONG_INLINE=inline -DENABLE_ORT_B
         ACKEND -DENABLE_PADDLE_BACKEND -DENABLE_OPENVINO_BACKEND -DWITH_GPU -DENABLE_NVJPEG -DENABLE_TRT_BACKEND -DENA
         BLE_VISION -DENABLE_TEXT -DENABLE_PADDLE2ONNX -D__TBB_NO_IMPLICIT_LINKAGE=1 -D"CMAKE_INTDIR=\"Release\"" -Dfas
         tdeploy_EXPORTS -D"__REL_FILE__=\"fastdeploy/vision/utils/yolo_preprocess.cu\"" -D_WINDLL -D_MBCS -DWIN32 -D_W
         INDOWS -DNDEBUG -DYAML_CPP_DLL -DFASTDEPLOY_LIB -DCMAKE_BUILD_TYPE=Release -DEIGEN_STRONG_INLINE=inline -DENAB
         LE_ORT_BACKEND -DENABLE_PADDLE_BACKEND -DENABLE_OPENVINO_BACKEND -DWITH_GPU -DENABLE_NVJPEG -DENABLE_TRT_BACKE
         ND -DENABLE_VISION -DENABLE_TEXT -DENABLE_PADDLE2ONNX -D__TBB_NO_IMPLICIT_LINKAGE=1 -D"CMAKE_INTDIR=\"Release\
         "" -Dfastdeploy_EXPORTS -Xcompiler "/EHsc /W0 /nologo /O2 /FS   /MD /GR" -Xcompiler "/Fdfastdeploy.dir\Release
         \vc143.pdb" -o fastdeploy.dir\Release\yolo_preprocess.obj "C:\Users\MWX\FastDeploy\fastdeploy\vision\utils\yol
         o_preprocess.cu"
         nvcc fatal   : Unsupported gpu architecture 'compute_35'
    23>C:\Program Files\Microsoft Visual Studio\2022\Community\MSBuild\Microsoft\VC\v170\BuildCustomizations\CUDA 12.1.
       targets(799,9): error MSB3721: 命令“"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\bin\nvcc.exe"  --use
       -local-env -ccbin "C:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\bin\HostX64
       \x64" -x cu   -IC:\Users\MWX\FastDeploy\. -IC:\Users\MWX\FastDeploy\build -IC:\Users\MWX\FastDeploy\third_party\
       eigen -IC:\Users\MWX\FastDeploy\build\third_libs\install\onnxruntime\include -IC:\Users\MWX\FastDeploy\build\thi
       rd_libs\install\paddle_inference -IC:\Users\MWX\FastDeploy\build\third_libs\install\openvino\runtime\include -IC
       :\Users\MWX\FastDeploy\build\third_libs\install\openvino\runtime\include\ie -I"C:\Program Files\NVIDIA GPU Compu
       ting Toolkit\CUDA\v12.1\include" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\TensorRT\8.6.1\include" -IC:\U
       sers\MWX\FastDeploy\.\fastdeploy\runtime\backends\tensorrt\common -I"C:\Users\MWX\FastDeploy\third_party\yaml-cp
       p\include" -IC:\Users\MWX\FastDeploy\build\third_libs\install\fast_tokenizer\include -IC:\Users\MWX\FastDeploy\b
       uild\third_libs\install\fast_tokenizer\third_party\include -IC:\Users\MWX\FastDeploy\build\third_libs\install\pa
       ddle2onnx\include -IC:\Users\MWX\FastDeploy\build\third_libs\install\opencv\build\include -IC:\Users\MWX\FastDep
       loy\build\third_libs\install\opencv\build\include\opencv -IC:\Users\MWX\FastDeploy\build\third_libs\install\open
       vino\runtime\3rdparty\tbb\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\include"     --kee
       p-dir x64\Release  -maxrregcount=0   --machine 64 --compile -cudart static -gencode arch=compute_35,code=sm_35 -
       gencode arch=compute_50,code=sm_50 -gencode arch=compute_52,code=sm_52 -gencode arch=compute_60,code=sm_60 -genc
       ode arch=compute_61,code=sm_61 -gencode arch=compute_70,code=sm_70 -gencode arch=compute_75,code=sm_75 -gencode
       arch=compute_80,code=sm_80 -gencode arch=compute_86,code=sm_86 -w --expt-relaxed-constexpr --expt-extended-lambd
       a -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj -Ob2 /wd4251"   -D_WINDOWS -DNDEBUG -DYAML_CPP_DLL -DFASTDEP
       LOY_LIB -DCMAKE_BUILD_TYPE=Release -DEIGEN_STRONG_INLINE=inline -DENABLE_ORT_BACKEND -DENABLE_PADDLE_BACKEND -DE
       NABLE_OPENVINO_BACKEND -DWITH_GPU -DENABLE_NVJPEG -DENABLE_TRT_BACKEND -DENABLE_VISION -DENABLE_TEXT -DENABLE_PA
       DDLE2ONNX -D__TBB_NO_IMPLICIT_LINKAGE=1 -D"CMAKE_INTDIR=\"Release\"" -Dfastdeploy_EXPORTS -D"__REL_FILE__=\"fast
       deploy/vision/utils/yolo_preprocess.cu\"" -D_WINDLL -D_MBCS -DWIN32 -D_WINDOWS -DNDEBUG -DYAML_CPP_DLL -DFASTDEP
       LOY_LIB -DCMAKE_BUILD_TYPE=Release -DEIGEN_STRONG_INLINE=inline -DENABLE_ORT_BACKEND -DENABLE_PADDLE_BACKEND -DE
       NABLE_OPENVINO_BACKEND -DWITH_GPU -DENABLE_NVJPEG -DENABLE_TRT_BACKEND -DENABLE_VISION -DENABLE_TEXT -DENABLE_PA
       DDLE2ONNX -D__TBB_NO_IMPLICIT_LINKAGE=1 -D"CMAKE_INTDIR=\"Release\"" -Dfastdeploy_EXPORTS -Xcompiler "/EHsc /W0
       /nologo /O2 /FS   /MD /GR" -Xcompiler "/Fdfastdeploy.dir\Release\vc143.pdb" -o fastdeploy.dir\Release\yolo_prepr
       ocess.obj "C:\Users\MWX\FastDeploy\fastdeploy\vision\utils\yolo_preprocess.cu"”已退出，返回代码为 1。 [C:\Users\MWX\FastDe
       ploy\build\fastdeploy.vcxproj]
    23>已完成生成项目“C:\Users\MWX\FastDeploy\build\fastdeploy.vcxproj”(CudaBuildCore 个目标)的操作 - 失败。
    23>CudaBuildCore:

         C:\Users\MWX\FastDeploy\build>"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\bin\nvcc.exe"  --use-l
         ocal-env -ccbin "C:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\bin\HostX64
         \x64" -x cu   -IC:\Users\MWX\FastDeploy\. -IC:\Users\MWX\FastDeploy\build -IC:\Users\MWX\FastDeploy\third_part
         y\eigen -IC:\Users\MWX\FastDeploy\build\third_libs\install\onnxruntime\include -IC:\Users\MWX\FastDeploy\build
         \third_libs\install\paddle_inference -IC:\Users\MWX\FastDeploy\build\third_libs\install\openvino\runtime\inclu
         de -IC:\Users\MWX\FastDeploy\build\third_libs\install\openvino\runtime\include\ie -I"C:\Program Files\NVIDIA G
         PU Computing Toolkit\CUDA\v12.1\include" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\TensorRT\8.6.1\inclu
         de" -IC:\Users\MWX\FastDeploy\.\fastdeploy\runtime\backends\tensorrt\common -I"C:\Users\MWX\FastDeploy\third_p
         arty\yaml-cpp\include" -IC:\Users\MWX\FastDeploy\build\third_libs\install\fast_tokenizer\include -IC:\Users\MW
         X\FastDeploy\build\third_libs\install\fast_tokenizer\third_party\include -IC:\Users\MWX\FastDeploy\build\third
         _libs\install\paddle2onnx\include -IC:\Users\MWX\FastDeploy\build\third_libs\install\opencv\build\include -IC:
         \Users\MWX\FastDeploy\build\third_libs\install\opencv\build\include\opencv -IC:\Users\MWX\FastDeploy\build\thi
         rd_libs\install\openvino\runtime\3rdparty\tbb\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v1
         2.1\include"     --keep-dir x64\Release  -maxrregcount=0   --machine 64 --compile -cudart static -gencode arch
         =compute_35,code=sm_35 -gencode arch=compute_50,code=sm_50 -gencode arch=compute_52,code=sm_52 -gencode arch=c
         ompute_60,code=sm_60 -gencode arch=compute_61,code=sm_61 -gencode arch=compute_70,code=sm_70 -gencode arch=com
         pute_75,code=sm_75 -gencode arch=compute_80,code=sm_80 -gencode arch=compute_86,code=sm_86 -w --expt-relaxed-c
         onstexpr --expt-extended-lambda -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj -Ob2 /wd4251"   -D_WINDOWS -
         DNDEBUG -DYAML_CPP_DLL -DFASTDEPLOY_LIB -DCMAKE_BUILD_TYPE=Release -DEIGEN_STRONG_INLINE=inline -DENABLE_ORT_B
         ACKEND -DENABLE_PADDLE_BACKEND -DENABLE_OPENVINO_BACKEND -DWITH_GPU -DENABLE_NVJPEG -DENABLE_TRT_BACKEND -DENA
         BLE_VISION -DENABLE_TEXT -DENABLE_PADDLE2ONNX -D__TBB_NO_IMPLICIT_LINKAGE=1 -D"CMAKE_INTDIR=\"Release\"" -Dfas
         tdeploy_EXPORTS -D"__REL_FILE__=\"fastdeploy/function/cuda_cast.cu\"" -D_WINDLL -D_MBCS -DWIN32 -D_WINDOWS -DN
         DEBUG -DYAML_CPP_DLL -DFASTDEPLOY_LIB -DCMAKE_BUILD_TYPE=Release -DEIGEN_STRONG_INLINE=inline -DENABLE_ORT_BAC
         KEND -DENABLE_PADDLE_BACKEND -DENABLE_OPENVINO_BACKEND -DWITH_GPU -DENABLE_NVJPEG -DENABLE_TRT_BACKEND -DENABL
         E_VISION -DENABLE_TEXT -DENABLE_PADDLE2ONNX -D__TBB_NO_IMPLICIT_LINKAGE=1 -D"CMAKE_INTDIR=\"Release\"" -Dfastd
         eploy_EXPORTS -Xcompiler "/EHsc /W0 /nologo /O2 /FS   /MD /GR" -Xcompiler "/Fdfastdeploy.dir\Release\vc143.pdb
         " -o fastdeploy.dir\Release\cuda_cast.obj "C:\Users\MWX\FastDeploy\fastdeploy\function\cuda_cast.cu"
         nvcc fatal   : Unsupported gpu architecture 'compute_35'
    23>C:\Program Files\Microsoft Visual Studio\2022\Community\MSBuild\Microsoft\VC\v170\BuildCustomizations\CUDA 12.1.
       targets(799,9): error MSB3721: 命令“"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\bin\nvcc.exe"  --use
       -local-env -ccbin "C:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\bin\HostX64
       \x64" -x cu   -IC:\Users\MWX\FastDeploy\. -IC:\Users\MWX\FastDeploy\build -IC:\Users\MWX\FastDeploy\third_party\
       eigen -IC:\Users\MWX\FastDeploy\build\third_libs\install\onnxruntime\include -IC:\Users\MWX\FastDeploy\build\thi
       rd_libs\install\paddle_inference -IC:\Users\MWX\FastDeploy\build\third_libs\install\openvino\runtime\include -IC
       :\Users\MWX\FastDeploy\build\third_libs\install\openvino\runtime\include\ie -I"C:\Program Files\NVIDIA GPU Compu
       ting Toolkit\CUDA\v12.1\include" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\TensorRT\8.6.1\include" -IC:\U
       sers\MWX\FastDeploy\.\fastdeploy\runtime\backends\tensorrt\common -I"C:\Users\MWX\FastDeploy\third_party\yaml-cp
       p\include" -IC:\Users\MWX\FastDeploy\build\third_libs\install\fast_tokenizer\include -IC:\Users\MWX\FastDeploy\b
       uild\third_libs\install\fast_tokenizer\third_party\include -IC:\Users\MWX\FastDeploy\build\third_libs\install\pa
       ddle2onnx\include -IC:\Users\MWX\FastDeploy\build\third_libs\install\opencv\build\include -IC:\Users\MWX\FastDep
       loy\build\third_libs\install\opencv\build\include\opencv -IC:\Users\MWX\FastDeploy\build\third_libs\install\open
       vino\runtime\3rdparty\tbb\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\include"     --kee
       p-dir x64\Release  -maxrregcount=0   --machine 64 --compile -cudart static -gencode arch=compute_35,code=sm_35 -
       gencode arch=compute_50,code=sm_50 -gencode arch=compute_52,code=sm_52 -gencode arch=compute_60,code=sm_60 -genc
       ode arch=compute_61,code=sm_61 -gencode arch=compute_70,code=sm_70 -gencode arch=compute_75,code=sm_75 -gencode
       arch=compute_80,code=sm_80 -gencode arch=compute_86,code=sm_86 -w --expt-relaxed-constexpr --expt-extended-lambd
       a -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj -Ob2 /wd4251"   -D_WINDOWS -DNDEBUG -DYAML_CPP_DLL -DFASTDEP
       LOY_LIB -DCMAKE_BUILD_TYPE=Release -DEIGEN_STRONG_INLINE=inline -DENABLE_ORT_BACKEND -DENABLE_PADDLE_BACKEND -DE
       NABLE_OPENVINO_BACKEND -DWITH_GPU -DENABLE_NVJPEG -DENABLE_TRT_BACKEND -DENABLE_VISION -DENABLE_TEXT -DENABLE_PA
       DDLE2ONNX -D__TBB_NO_IMPLICIT_LINKAGE=1 -D"CMAKE_INTDIR=\"Release\"" -Dfastdeploy_EXPORTS -D"__REL_FILE__=\"fast
       deploy/function/cuda_cast.cu\"" -D_WINDLL -D_MBCS -DWIN32 -D_WINDOWS -DNDEBUG -DYAML_CPP_DLL -DFASTDEPLOY_LIB -D
       CMAKE_BUILD_TYPE=Release -DEIGEN_STRONG_INLINE=inline -DENABLE_ORT_BACKEND -DENABLE_PADDLE_BACKEND -DENABLE_OPEN
       VINO_BACKEND -DWITH_GPU -DENABLE_NVJPEG -DENABLE_TRT_BACKEND -DENABLE_VISION -DENABLE_TEXT -DENABLE_PADDLE2ONNX
       -D__TBB_NO_IMPLICIT_LINKAGE=1 -D"CMAKE_INTDIR=\"Release\"" -Dfastdeploy_EXPORTS -Xcompiler "/EHsc /W0 /nologo /O
       2 /FS   /MD /GR" -Xcompiler "/Fdfastdeploy.dir\Release\vc143.pdb" -o fastdeploy.dir\Release\cuda_cast.obj "C:\Us
       ers\MWX\FastDeploy\fastdeploy\function\cuda_cast.cu"”已退出，返回代码为 1。 [C:\Users\MWX\FastDeploy\build\fastdeploy.vcxp
       roj]
    23>已完成生成项目“C:\Users\MWX\FastDeploy\build\fastdeploy.vcxproj”(CudaBuildCore 个目标)的操作 - 失败。
    23>已完成生成项目“C:\Users\MWX\FastDeploy\build\fastdeploy.vcxproj”(默认目标)的操作 - 失败。
     4>已完成生成项目“C:\Users\MWX\FastDeploy\build\fastdeploy.vcxproj.metaproj”(默认目标)的操作 - 失败。
     3>已完成生成项目“C:\Users\MWX\FastDeploy\build\copy_yaml_include.vcxproj.metaproj”(默认目标)的操作 - 失败。
     2>已完成生成项目“C:\Users\MWX\FastDeploy\build\ALL_BUILD.vcxproj.metaproj”(默认目标)的操作 - 失败。
     1>已完成生成项目“C:\Users\MWX\FastDeploy\build\fastdeploy.sln”(默认目标)的操作 - 失败。

生成失败。

       “C:\Users\MWX\FastDeploy\build\fastdeploy.sln”(默认目标) (1) ->
       “C:\Users\MWX\FastDeploy\build\ALL_BUILD.vcxproj.metaproj”(默认目标) (2) ->
       “C:\Users\MWX\FastDeploy\build\copy_yaml_include.vcxproj.metaproj”(默认目标) (3) ->
       “C:\Users\MWX\FastDeploy\build\fastdeploy.vcxproj.metaproj”(默认目标) (4) ->
       “C:\Users\MWX\FastDeploy\build\fastdeploy.vcxproj”(默认目标) (23) ->
       “C:\Users\MWX\FastDeploy\build\fastdeploy.vcxproj”(CudaBuildCore 目标) (23:4) ->
       (CudaBuildCore 目标) ->
         C:\Program Files\Microsoft Visual Studio\2022\Community\MSBuild\Microsoft\VC\v170\BuildCustomizations\CUDA 12.
       1.targets(799,9): error MSB3721: 命令“"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\bin\nvcc.exe"  --u
       se-local-env -ccbin "C:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\bin\HostX
       64\x64" -x cu   -IC:\Users\MWX\FastDeploy\. -IC:\Users\MWX\FastDeploy\build -IC:\Users\MWX\FastDeploy\third_part
       y\eigen -IC:\Users\MWX\FastDeploy\build\third_libs\install\onnxruntime\include -IC:\Users\MWX\FastDeploy\build\t
       hird_libs\install\paddle_inference -IC:\Users\MWX\FastDeploy\build\third_libs\install\openvino\runtime\include -
       IC:\Users\MWX\FastDeploy\build\third_libs\install\openvino\runtime\include\ie -I"C:\Program Files\NVIDIA GPU Com
       puting Toolkit\CUDA\v12.1\include" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\TensorRT\8.6.1\include" -IC:
       \Users\MWX\FastDeploy\.\fastdeploy\runtime\backends\tensorrt\common -I"C:\Users\MWX\FastDeploy\third_party\yaml-
       cpp\include" -IC:\Users\MWX\FastDeploy\build\third_libs\install\fast_tokenizer\include -IC:\Users\MWX\FastDeploy
       \build\third_libs\install\fast_tokenizer\third_party\include -IC:\Users\MWX\FastDeploy\build\third_libs\install\
       paddle2onnx\include -IC:\Users\MWX\FastDeploy\build\third_libs\install\opencv\build\include -IC:\Users\MWX\FastD
       eploy\build\third_libs\install\opencv\build\include\opencv -IC:\Users\MWX\FastDeploy\build\third_libs\install\op
       envino\runtime\3rdparty\tbb\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\include"     --k
       eep-dir x64\Release  -maxrregcount=0   --machine 64 --compile -cudart static -gencode arch=compute_35,code=sm_35
        -gencode arch=compute_50,code=sm_50 -gencode arch=compute_52,code=sm_52 -gencode arch=compute_60,code=sm_60 -ge
       ncode arch=compute_61,code=sm_61 -gencode arch=compute_70,code=sm_70 -gencode arch=compute_75,code=sm_75 -gencod
       e arch=compute_80,code=sm_80 -gencode arch=compute_86,code=sm_86 -w --expt-relaxed-constexpr --expt-extended-lam
       bda -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj -Ob2 /wd4251"   -D_WINDOWS -DNDEBUG -DYAML_CPP_DLL -DFASTD
       EPLOY_LIB -DCMAKE_BUILD_TYPE=Release -DEIGEN_STRONG_INLINE=inline -DENABLE_ORT_BACKEND -DENABLE_PADDLE_BACKEND -
       DENABLE_OPENVINO_BACKEND -DWITH_GPU -DENABLE_NVJPEG -DENABLE_TRT_BACKEND -DENABLE_VISION -DENABLE_TEXT -DENABLE_
       PADDLE2ONNX -D__TBB_NO_IMPLICIT_LINKAGE=1 -D"CMAKE_INTDIR=\"Release\"" -Dfastdeploy_EXPORTS -D"__REL_FILE__=\"fa
       stdeploy/vision/common/processors/normalize_and_permute.cu\"" -D_WINDLL -D_MBCS -DWIN32 -D_WINDOWS -DNDEBUG -DYA
       ML_CPP_DLL -DFASTDEPLOY_LIB -DCMAKE_BUILD_TYPE=Release -DEIGEN_STRONG_INLINE=inline -DENABLE_ORT_BACKEND -DENABL
       E_PADDLE_BACKEND -DENABLE_OPENVINO_BACKEND -DWITH_GPU -DENABLE_NVJPEG -DENABLE_TRT_BACKEND -DENABLE_VISION -DENA
       BLE_TEXT -DENABLE_PADDLE2ONNX -D__TBB_NO_IMPLICIT_LINKAGE=1 -D"CMAKE_INTDIR=\"Release\"" -Dfastdeploy_EXPORTS -X
       compiler "/EHsc /W0 /nologo /O2 /FS   /MD /GR" -Xcompiler "/Fdfastdeploy.dir\Release\vc143.pdb" -o fastdeploy.di
       r\Release\/fastdeploy/vision/common/processors/normalize_and_permute.cu.obj "C:\Users\MWX\FastDeploy\fastdeploy\
       vision\common\processors\normalize_and_permute.cu"”已退出，返回代码为 1。 [C:\Users\MWX\FastDeploy\build\fastdeploy.vcxpro
       j]


       “C:\Users\MWX\FastDeploy\build\fastdeploy.sln”(默认目标) (1) ->
       “C:\Users\MWX\FastDeploy\build\ALL_BUILD.vcxproj.metaproj”(默认目标) (2) ->
       “C:\Users\MWX\FastDeploy\build\copy_yaml_include.vcxproj.metaproj”(默认目标) (3) ->
       “C:\Users\MWX\FastDeploy\build\fastdeploy.vcxproj.metaproj”(默认目标) (4) ->
       “C:\Users\MWX\FastDeploy\build\fastdeploy.vcxproj”(默认目标) (23) ->
       “C:\Users\MWX\FastDeploy\build\fastdeploy.vcxproj”(CudaBuildCore 目标) (23:3) ->
         C:\Program Files\Microsoft Visual Studio\2022\Community\MSBuild\Microsoft\VC\v170\BuildCustomizations\CUDA 12.
       1.targets(799,9): error MSB3721: 命令“"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\bin\nvcc.exe"  --u
       se-local-env -ccbin "C:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\bin\HostX
       64\x64" -x cu   -IC:\Users\MWX\FastDeploy\. -IC:\Users\MWX\FastDeploy\build -IC:\Users\MWX\FastDeploy\third_part
       y\eigen -IC:\Users\MWX\FastDeploy\build\third_libs\install\onnxruntime\include -IC:\Users\MWX\FastDeploy\build\t
       hird_libs\install\paddle_inference -IC:\Users\MWX\FastDeploy\build\third_libs\install\openvino\runtime\include -
       IC:\Users\MWX\FastDeploy\build\third_libs\install\openvino\runtime\include\ie -I"C:\Program Files\NVIDIA GPU Com
       puting Toolkit\CUDA\v12.1\include" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\TensorRT\8.6.1\include" -IC:
       \Users\MWX\FastDeploy\.\fastdeploy\runtime\backends\tensorrt\common -I"C:\Users\MWX\FastDeploy\third_party\yaml-
       cpp\include" -IC:\Users\MWX\FastDeploy\build\third_libs\install\fast_tokenizer\include -IC:\Users\MWX\FastDeploy
       \build\third_libs\install\fast_tokenizer\third_party\include -IC:\Users\MWX\FastDeploy\build\third_libs\install\
       paddle2onnx\include -IC:\Users\MWX\FastDeploy\build\third_libs\install\opencv\build\include -IC:\Users\MWX\FastD
       eploy\build\third_libs\install\opencv\build\include\opencv -IC:\Users\MWX\FastDeploy\build\third_libs\install\op
       envino\runtime\3rdparty\tbb\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\include"     --k
       eep-dir x64\Release  -maxrregcount=0   --machine 64 --compile -cudart static -gencode arch=compute_35,code=sm_35
        -gencode arch=compute_50,code=sm_50 -gencode arch=compute_52,code=sm_52 -gencode arch=compute_60,code=sm_60 -ge
       ncode arch=compute_61,code=sm_61 -gencode arch=compute_70,code=sm_70 -gencode arch=compute_75,code=sm_75 -gencod
       e arch=compute_80,code=sm_80 -gencode arch=compute_86,code=sm_86 -w --expt-relaxed-constexpr --expt-extended-lam
       bda -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj -Ob2 /wd4251"   -D_WINDOWS -DNDEBUG -DYAML_CPP_DLL -DFASTD
       EPLOY_LIB -DCMAKE_BUILD_TYPE=Release -DEIGEN_STRONG_INLINE=inline -DENABLE_ORT_BACKEND -DENABLE_PADDLE_BACKEND -
       DENABLE_OPENVINO_BACKEND -DWITH_GPU -DENABLE_NVJPEG -DENABLE_TRT_BACKEND -DENABLE_VISION -DENABLE_TEXT -DENABLE_
       PADDLE2ONNX -D__TBB_NO_IMPLICIT_LINKAGE=1 -D"CMAKE_INTDIR=\"Release\"" -Dfastdeploy_EXPORTS -D"__REL_FILE__=\"fa
       stdeploy/runtime/backends/common/cuda/adaptive_pool2d_kernel.cu\"" -D_WINDLL -D_MBCS -DWIN32 -D_WINDOWS -DNDEBUG
        -DYAML_CPP_DLL -DFASTDEPLOY_LIB -DCMAKE_BUILD_TYPE=Release -DEIGEN_STRONG_INLINE=inline -DENABLE_ORT_BACKEND -D
       ENABLE_PADDLE_BACKEND -DENABLE_OPENVINO_BACKEND -DWITH_GPU -DENABLE_NVJPEG -DENABLE_TRT_BACKEND -DENABLE_VISION
       -DENABLE_TEXT -DENABLE_PADDLE2ONNX -D__TBB_NO_IMPLICIT_LINKAGE=1 -D"CMAKE_INTDIR=\"Release\"" -Dfastdeploy_EXPOR
       TS -Xcompiler "/EHsc /W0 /nologo /O2 /FS   /MD /GR" -Xcompiler "/Fdfastdeploy.dir\Release\vc143.pdb" -o fastdepl
       oy.dir\Release\adaptive_pool2d_kernel.obj "C:\Users\MWX\FastDeploy\fastdeploy\runtime\backends\common\cuda\adapt
       ive_pool2d_kernel.cu"”已退出，返回代码为 1。 [C:\Users\MWX\FastDeploy\build\fastdeploy.vcxproj]


       “C:\Users\MWX\FastDeploy\build\fastdeploy.sln”(默认目标) (1) ->
       “C:\Users\MWX\FastDeploy\build\ALL_BUILD.vcxproj.metaproj”(默认目标) (2) ->
       “C:\Users\MWX\FastDeploy\build\copy_yaml_include.vcxproj.metaproj”(默认目标) (3) ->
       “C:\Users\MWX\FastDeploy\build\fastdeploy.vcxproj.metaproj”(默认目标) (4) ->
       “C:\Users\MWX\FastDeploy\build\fastdeploy.vcxproj”(默认目标) (23) ->
       “C:\Users\MWX\FastDeploy\build\fastdeploy.vcxproj”(CudaBuildCore 目标) (23:5) ->
         C:\Program Files\Microsoft Visual Studio\2022\Community\MSBuild\Microsoft\VC\v170\BuildCustomizations\CUDA 12.
       1.targets(799,9): error MSB3721: 命令“"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\bin\nvcc.exe"  --u
       se-local-env -ccbin "C:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\bin\HostX
       64\x64" -x cu   -IC:\Users\MWX\FastDeploy\. -IC:\Users\MWX\FastDeploy\build -IC:\Users\MWX\FastDeploy\third_part
       y\eigen -IC:\Users\MWX\FastDeploy\build\third_libs\install\onnxruntime\include -IC:\Users\MWX\FastDeploy\build\t
       hird_libs\install\paddle_inference -IC:\Users\MWX\FastDeploy\build\third_libs\install\openvino\runtime\include -
       IC:\Users\MWX\FastDeploy\build\third_libs\install\openvino\runtime\include\ie -I"C:\Program Files\NVIDIA GPU Com
       puting Toolkit\CUDA\v12.1\include" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\TensorRT\8.6.1\include" -IC:
       \Users\MWX\FastDeploy\.\fastdeploy\runtime\backends\tensorrt\common -I"C:\Users\MWX\FastDeploy\third_party\yaml-
       cpp\include" -IC:\Users\MWX\FastDeploy\build\third_libs\install\fast_tokenizer\include -IC:\Users\MWX\FastDeploy
       \build\third_libs\install\fast_tokenizer\third_party\include -IC:\Users\MWX\FastDeploy\build\third_libs\install\
       paddle2onnx\include -IC:\Users\MWX\FastDeploy\build\third_libs\install\opencv\build\include -IC:\Users\MWX\FastD
       eploy\build\third_libs\install\opencv\build\include\opencv -IC:\Users\MWX\FastDeploy\build\third_libs\install\op
       envino\runtime\3rdparty\tbb\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\include"     --k
       eep-dir x64\Release  -maxrregcount=0   --machine 64 --compile -cudart static -gencode arch=compute_35,code=sm_35
        -gencode arch=compute_50,code=sm_50 -gencode arch=compute_52,code=sm_52 -gencode arch=compute_60,code=sm_60 -ge
       ncode arch=compute_61,code=sm_61 -gencode arch=compute_70,code=sm_70 -gencode arch=compute_75,code=sm_75 -gencod
       e arch=compute_80,code=sm_80 -gencode arch=compute_86,code=sm_86 -w --expt-relaxed-constexpr --expt-extended-lam
       bda -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj -Ob2 /wd4251"   -D_WINDOWS -DNDEBUG -DYAML_CPP_DLL -DFASTD
       EPLOY_LIB -DCMAKE_BUILD_TYPE=Release -DEIGEN_STRONG_INLINE=inline -DENABLE_ORT_BACKEND -DENABLE_PADDLE_BACKEND -
       DENABLE_OPENVINO_BACKEND -DWITH_GPU -DENABLE_NVJPEG -DENABLE_TRT_BACKEND -DENABLE_VISION -DENABLE_TEXT -DENABLE_
       PADDLE2ONNX -D__TBB_NO_IMPLICIT_LINKAGE=1 -D"CMAKE_INTDIR=\"Release\"" -Dfastdeploy_EXPORTS -D"__REL_FILE__=\"fa
       stdeploy/vision/common/processors/normalize.cu\"" -D_WINDLL -D_MBCS -DWIN32 -D_WINDOWS -DNDEBUG -DYAML_CPP_DLL -
       DFASTDEPLOY_LIB -DCMAKE_BUILD_TYPE=Release -DEIGEN_STRONG_INLINE=inline -DENABLE_ORT_BACKEND -DENABLE_PADDLE_BAC
       KEND -DENABLE_OPENVINO_BACKEND -DWITH_GPU -DENABLE_NVJPEG -DENABLE_TRT_BACKEND -DENABLE_VISION -DENABLE_TEXT -DE
       NABLE_PADDLE2ONNX -D__TBB_NO_IMPLICIT_LINKAGE=1 -D"CMAKE_INTDIR=\"Release\"" -Dfastdeploy_EXPORTS -Xcompiler "/E
       Hsc /W0 /nologo /O2 /FS   /MD /GR" -Xcompiler "/Fdfastdeploy.dir\Release\vc143.pdb" -o fastdeploy.dir\Release\/f
       astdeploy/vision/common/processors/normalize.cu.obj "C:\Users\MWX\FastDeploy\fastdeploy\vision\common\processors
       \normalize.cu"”已退出，返回代码为 1。 [C:\Users\MWX\FastDeploy\build\fastdeploy.vcxproj]


       “C:\Users\MWX\FastDeploy\build\fastdeploy.sln”(默认目标) (1) ->
       “C:\Users\MWX\FastDeploy\build\ALL_BUILD.vcxproj.metaproj”(默认目标) (2) ->
       “C:\Users\MWX\FastDeploy\build\copy_yaml_include.vcxproj.metaproj”(默认目标) (3) ->
       “C:\Users\MWX\FastDeploy\build\fastdeploy.vcxproj.metaproj”(默认目标) (4) ->
       “C:\Users\MWX\FastDeploy\build\fastdeploy.vcxproj”(默认目标) (23) ->
       “C:\Users\MWX\FastDeploy\build\fastdeploy.vcxproj”(CudaBuildCore 目标) (23:6) ->
         C:\Program Files\Microsoft Visual Studio\2022\Community\MSBuild\Microsoft\VC\v170\BuildCustomizations\CUDA 12.
       1.targets(799,9): error MSB3721: 命令“"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\bin\nvcc.exe"  --u
       se-local-env -ccbin "C:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\bin\HostX
       64\x64" -x cu   -IC:\Users\MWX\FastDeploy\. -IC:\Users\MWX\FastDeploy\build -IC:\Users\MWX\FastDeploy\third_part
       y\eigen -IC:\Users\MWX\FastDeploy\build\third_libs\install\onnxruntime\include -IC:\Users\MWX\FastDeploy\build\t
       hird_libs\install\paddle_inference -IC:\Users\MWX\FastDeploy\build\third_libs\install\openvino\runtime\include -
       IC:\Users\MWX\FastDeploy\build\third_libs\install\openvino\runtime\include\ie -I"C:\Program Files\NVIDIA GPU Com
       puting Toolkit\CUDA\v12.1\include" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\TensorRT\8.6.1\include" -IC:
       \Users\MWX\FastDeploy\.\fastdeploy\runtime\backends\tensorrt\common -I"C:\Users\MWX\FastDeploy\third_party\yaml-
       cpp\include" -IC:\Users\MWX\FastDeploy\build\third_libs\install\fast_tokenizer\include -IC:\Users\MWX\FastDeploy
       \build\third_libs\install\fast_tokenizer\third_party\include -IC:\Users\MWX\FastDeploy\build\third_libs\install\
       paddle2onnx\include -IC:\Users\MWX\FastDeploy\build\third_libs\install\opencv\build\include -IC:\Users\MWX\FastD
       eploy\build\third_libs\install\opencv\build\include\opencv -IC:\Users\MWX\FastDeploy\build\third_libs\install\op
       envino\runtime\3rdparty\tbb\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\include"     --k
       eep-dir x64\Release  -maxrregcount=0   --machine 64 --compile -cudart static -gencode arch=compute_35,code=sm_35
        -gencode arch=compute_50,code=sm_50 -gencode arch=compute_52,code=sm_52 -gencode arch=compute_60,code=sm_60 -ge
       ncode arch=compute_61,code=sm_61 -gencode arch=compute_70,code=sm_70 -gencode arch=compute_75,code=sm_75 -gencod
       e arch=compute_80,code=sm_80 -gencode arch=compute_86,code=sm_86 -w --expt-relaxed-constexpr --expt-extended-lam
       bda -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj -Ob2 /wd4251"   -D_WINDOWS -DNDEBUG -DYAML_CPP_DLL -DFASTD
       EPLOY_LIB -DCMAKE_BUILD_TYPE=Release -DEIGEN_STRONG_INLINE=inline -DENABLE_ORT_BACKEND -DENABLE_PADDLE_BACKEND -
       DENABLE_OPENVINO_BACKEND -DWITH_GPU -DENABLE_NVJPEG -DENABLE_TRT_BACKEND -DENABLE_VISION -DENABLE_TEXT -DENABLE_
       PADDLE2ONNX -D__TBB_NO_IMPLICIT_LINKAGE=1 -D"CMAKE_INTDIR=\"Release\"" -Dfastdeploy_EXPORTS -D"__REL_FILE__=\"fa
       stdeploy/vision/utils/yolo_preprocess.cu\"" -D_WINDLL -D_MBCS -DWIN32 -D_WINDOWS -DNDEBUG -DYAML_CPP_DLL -DFASTD
       EPLOY_LIB -DCMAKE_BUILD_TYPE=Release -DEIGEN_STRONG_INLINE=inline -DENABLE_ORT_BACKEND -DENABLE_PADDLE_BACKEND -
       DENABLE_OPENVINO_BACKEND -DWITH_GPU -DENABLE_NVJPEG -DENABLE_TRT_BACKEND -DENABLE_VISION -DENABLE_TEXT -DENABLE_
       PADDLE2ONNX -D__TBB_NO_IMPLICIT_LINKAGE=1 -D"CMAKE_INTDIR=\"Release\"" -Dfastdeploy_EXPORTS -Xcompiler "/EHsc /W
       0 /nologo /O2 /FS   /MD /GR" -Xcompiler "/Fdfastdeploy.dir\Release\vc143.pdb" -o fastdeploy.dir\Release\yolo_pre
       process.obj "C:\Users\MWX\FastDeploy\fastdeploy\vision\utils\yolo_preprocess.cu"”已退出，返回代码为 1。 [C:\Users\MWX\Fast
       Deploy\build\fastdeploy.vcxproj]


       “C:\Users\MWX\FastDeploy\build\fastdeploy.sln”(默认目标) (1) ->
       “C:\Users\MWX\FastDeploy\build\ALL_BUILD.vcxproj.metaproj”(默认目标) (2) ->
       “C:\Users\MWX\FastDeploy\build\copy_yaml_include.vcxproj.metaproj”(默认目标) (3) ->
       “C:\Users\MWX\FastDeploy\build\fastdeploy.vcxproj.metaproj”(默认目标) (4) ->
       “C:\Users\MWX\FastDeploy\build\fastdeploy.vcxproj”(默认目标) (23) ->
       “C:\Users\MWX\FastDeploy\build\fastdeploy.vcxproj”(CudaBuildCore 目标) (23:2) ->
         C:\Program Files\Microsoft Visual Studio\2022\Community\MSBuild\Microsoft\VC\v170\BuildCustomizations\CUDA 12.
       1.targets(799,9): error MSB3721: 命令“"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\bin\nvcc.exe"  --u
       se-local-env -ccbin "C:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\bin\HostX
       64\x64" -x cu   -IC:\Users\MWX\FastDeploy\. -IC:\Users\MWX\FastDeploy\build -IC:\Users\MWX\FastDeploy\third_part
       y\eigen -IC:\Users\MWX\FastDeploy\build\third_libs\install\onnxruntime\include -IC:\Users\MWX\FastDeploy\build\t
       hird_libs\install\paddle_inference -IC:\Users\MWX\FastDeploy\build\third_libs\install\openvino\runtime\include -
       IC:\Users\MWX\FastDeploy\build\third_libs\install\openvino\runtime\include\ie -I"C:\Program Files\NVIDIA GPU Com
       puting Toolkit\CUDA\v12.1\include" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\TensorRT\8.6.1\include" -IC:
       \Users\MWX\FastDeploy\.\fastdeploy\runtime\backends\tensorrt\common -I"C:\Users\MWX\FastDeploy\third_party\yaml-
       cpp\include" -IC:\Users\MWX\FastDeploy\build\third_libs\install\fast_tokenizer\include -IC:\Users\MWX\FastDeploy
       \build\third_libs\install\fast_tokenizer\third_party\include -IC:\Users\MWX\FastDeploy\build\third_libs\install\
       paddle2onnx\include -IC:\Users\MWX\FastDeploy\build\third_libs\install\opencv\build\include -IC:\Users\MWX\FastD
       eploy\build\third_libs\install\opencv\build\include\opencv -IC:\Users\MWX\FastDeploy\build\third_libs\install\op
       envino\runtime\3rdparty\tbb\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\include"     --k
       eep-dir x64\Release  -maxrregcount=0   --machine 64 --compile -cudart static -gencode arch=compute_35,code=sm_35
        -gencode arch=compute_50,code=sm_50 -gencode arch=compute_52,code=sm_52 -gencode arch=compute_60,code=sm_60 -ge
       ncode arch=compute_61,code=sm_61 -gencode arch=compute_70,code=sm_70 -gencode arch=compute_75,code=sm_75 -gencod
       e arch=compute_80,code=sm_80 -gencode arch=compute_86,code=sm_86 -w --expt-relaxed-constexpr --expt-extended-lam
       bda -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj -Ob2 /wd4251"   -D_WINDOWS -DNDEBUG -DYAML_CPP_DLL -DFASTD
       EPLOY_LIB -DCMAKE_BUILD_TYPE=Release -DEIGEN_STRONG_INLINE=inline -DENABLE_ORT_BACKEND -DENABLE_PADDLE_BACKEND -
       DENABLE_OPENVINO_BACKEND -DWITH_GPU -DENABLE_NVJPEG -DENABLE_TRT_BACKEND -DENABLE_VISION -DENABLE_TEXT -DENABLE_
       PADDLE2ONNX -D__TBB_NO_IMPLICIT_LINKAGE=1 -D"CMAKE_INTDIR=\"Release\"" -Dfastdeploy_EXPORTS -D"__REL_FILE__=\"fa
       stdeploy/function/cuda_cast.cu\"" -D_WINDLL -D_MBCS -DWIN32 -D_WINDOWS -DNDEBUG -DYAML_CPP_DLL -DFASTDEPLOY_LIB
       -DCMAKE_BUILD_TYPE=Release -DEIGEN_STRONG_INLINE=inline -DENABLE_ORT_BACKEND -DENABLE_PADDLE_BACKEND -DENABLE_OP
       ENVINO_BACKEND -DWITH_GPU -DENABLE_NVJPEG -DENABLE_TRT_BACKEND -DENABLE_VISION -DENABLE_TEXT -DENABLE_PADDLE2ONN
       X -D__TBB_NO_IMPLICIT_LINKAGE=1 -D"CMAKE_INTDIR=\"Release\"" -Dfastdeploy_EXPORTS -Xcompiler "/EHsc /W0 /nologo
       /O2 /FS   /MD /GR" -Xcompiler "/Fdfastdeploy.dir\Release\vc143.pdb" -o fastdeploy.dir\Release\cuda_cast.obj "C:\
       Users\MWX\FastDeploy\fastdeploy\function\cuda_cast.cu"”已退出，返回代码为 1。 [C:\Users\MWX\FastDeploy\build\fastdeploy.vc
       xproj]

    0 个警告
    5 个错误

已用时间 00:00:02.01

Jun 09 '23 08:06 justghostof

我编译时也遇到。请到cmake文件夹修改cuda.cmake文件，把支持老显卡的算力代号 35删除.我的修改成这样了：

if(NOT WITH_GPU)
  return()
endif()

# This is to eliminate the CMP0104 warnings from cmake 3.18+.
# Instead of setting CUDA_ARCHITECTURES, we will set CMAKE_CUDA_FLAGS.
set(CMAKE_CUDA_ARCHITECTURES OFF)

if(BUILD_ON_JETSON)
  set(fd_known_gpu_archs "53 62 72")
  set(fd_known_gpu_archs10 "53 62 72")
else()
  message("Using New Release Strategy - All Arches Packge")
  set(fd_known_gpu_archs "70 75 80 86")
  set(fd_known_gpu_archs10 "70 75")
  set(fd_known_gpu_archs11 "70 75 80")
endif()

######################################################################################
# A function for automatic detection of GPUs installed  (if autodetection is enabled)
# Usage:
#   detect_installed_gpus(out_variable)
function(detect_installed_gpus out_variable)
  if(NOT CUDA_gpu_detect_output)
    set(cufile ${PROJECT_BINARY_DIR}/detect_cuda_archs.cu)

    file(
      WRITE ${cufile}
      ""
      "#include \"stdio.h\"\n"
      "#include \"cuda.h\"\n"
      "#include \"cuda_runtime.h\"\n"
      "int main() {\n"
      "  int count = 0;\n"
      "  if (cudaSuccess != cudaGetDeviceCount(&count)) return -1;\n"
      "  if (count == 0) return -1;\n"
      "  for (int device = 0; device < count; ++device) {\n"
      "    cudaDeviceProp prop;\n"
      "    if (cudaSuccess == cudaGetDeviceProperties(&prop, device))\n"
      "      printf(\"%d.%d \", prop.major, prop.minor);\n"
      "  }\n"
      "  return 0;\n"
      "}\n")

    execute_process(
      COMMAND "${CMAKE_CUDA_COMPILER}" "--run" "${cufile}"
      WORKING_DIRECTORY "${PROJECT_BINARY_DIR}/CMakeFiles/"
      RESULT_VARIABLE nvcc_res
      OUTPUT_VARIABLE nvcc_out
      ERROR_QUIET OUTPUT_STRIP_TRAILING_WHITESPACE)

    if(nvcc_res EQUAL 0)
      # only keep the last line of nvcc_out
      string(REGEX REPLACE ";" "\\\\;" nvcc_out "${nvcc_out}")
      string(REGEX REPLACE "\n" ";" nvcc_out "${nvcc_out}")
      list(GET nvcc_out -1 nvcc_out)
      string(REPLACE "2.1" "2.1(2.0)" nvcc_out "${nvcc_out}")
      set(CUDA_gpu_detect_output
          ${nvcc_out}
          CACHE INTERNAL
                "Returned GPU architetures from detect_installed_gpus tool"
                FORCE)
    endif()
  endif()

  if(NOT CUDA_gpu_detect_output)
    message(
      STATUS
        "Automatic GPU detection failed. Building for all known architectures.")
    set(${out_variable}
        ${fd_known_gpu_archs}
        PARENT_SCOPE)
  else()
    set(${out_variable}
        ${CUDA_gpu_detect_output}
        PARENT_SCOPE)
  endif()
endfunction()

########################################################################
# Function for selecting GPU arch flags for nvcc based on CUDA_ARCH_NAME
# Usage:
#   select_nvcc_arch_flags(out_variable)
function(select_nvcc_arch_flags out_variable)
  # List of arch names
  set(archs_names
      "Kepler"
      "Maxwell"
      "Pascal"
      "Volta"
      "Turing"
      "Ampere"
      "All"
      "Manual")
  set(archs_name_default "All")
  list(APPEND archs_names "Auto")

  # set CUDA_ARCH_NAME strings (so it will be seen as dropbox in CMake-Gui)
  set(CUDA_ARCH_NAME
      ${archs_name_default}
      CACHE STRING "Select target NVIDIA GPU achitecture.")
  set_property(CACHE CUDA_ARCH_NAME PROPERTY STRINGS "" ${archs_names})
  mark_as_advanced(CUDA_ARCH_NAME)

  # verify CUDA_ARCH_NAME value
  if(NOT ";${archs_names};" MATCHES ";${CUDA_ARCH_NAME};")
    string(REPLACE ";" ", " archs_names "${archs_names}")
    message(
      FATAL_ERROR "Only ${archs_names} architectures names are supported.")
  endif()

  if(${CUDA_ARCH_NAME} STREQUAL "Manual")
    set(CUDA_ARCH_BIN
        ${fd_known_gpu_archs}
        CACHE
          STRING
          "Specify 'real' GPU architectures to build binaries for, BIN(PTX) format is supported"
    )
    set(CUDA_ARCH_PTX
        ""
        CACHE
          STRING
          "Specify 'virtual' PTX architectures to build PTX intermediate code for"
    )
    mark_as_advanced(CUDA_ARCH_BIN CUDA_ARCH_PTX)
  else()
    unset(CUDA_ARCH_BIN CACHE)
    unset(CUDA_ARCH_PTX CACHE)
  endif()

  if(${CUDA_ARCH_NAME} STREQUAL "Maxwell")
    if(BUILD_ON_JETSON)
      set(cuda_arch_bin "53")
    else()
      set(cuda_arch_bin "50")
    endif()
  elseif(${CUDA_ARCH_NAME} STREQUAL "Pascal")
    if(BUILD_ON_JETSON)
      set(cuda_arch_bin "62")
    else()
      set(cuda_arch_bin "60 61")
    endif()
  elseif(${CUDA_ARCH_NAME} STREQUAL "Volta")
    if(BUILD_ON_JETSON)
      set(cuda_arch_bin "72")
    else()
      set(cuda_arch_bin "70")
    endif()
  elseif(${CUDA_ARCH_NAME} STREQUAL "Turing")
    set(cuda_arch_bin "75")
  elseif(${CUDA_ARCH_NAME} STREQUAL "Ampere")
    if(${CMAKE_CUDA_COMPILER_VERSION} LESS 11.1) # CUDA 11.0
      set(cuda_arch_bin "80")
    elseif(${CMAKE_CUDA_COMPILER_VERSION} LESS 12.0) # CUDA 11.1+
      set(cuda_arch_bin "80 86")
    endif()
  elseif(${CUDA_ARCH_NAME} STREQUAL "All")
    set(cuda_arch_bin ${fd_known_gpu_archs})
  elseif(${CUDA_ARCH_NAME} STREQUAL "Auto")
    message(
      STATUS
        "WARNING: This is just a warning for publishing release.
      You are building GPU version without supporting different architectures.
      So the wheel package may fail on other GPU architectures.
      You can add -DCUDA_ARCH_NAME=All in cmake command
      to get a full wheel package to resolve this warning.
      While, this version will still work on local GPU architecture.")
    detect_installed_gpus(cuda_arch_bin)
  else() # (${CUDA_ARCH_NAME} STREQUAL "Manual")
    set(cuda_arch_bin ${CUDA_ARCH_BIN})
  endif()

  if(NEW_RELEASE_JIT)
    set(cuda_arch_ptx "${cuda_arch_ptx}${cuda_arch_bin}")
    set(cuda_arch_bin "")
  endif()

  # remove dots and convert to lists
  string(REGEX REPLACE "\\." "" cuda_arch_bin "${cuda_arch_bin}")
  string(REGEX REPLACE "\\." "" cuda_arch_ptx "${cuda_arch_ptx}")
  string(REGEX MATCHALL "[0-9()]+" cuda_arch_bin "${cuda_arch_bin}")
  string(REGEX MATCHALL "[0-9]+" cuda_arch_ptx "${cuda_arch_ptx}")

  list(REMOVE_DUPLICATES cuda_arch_bin)
  list(REMOVE_DUPLICATES cuda_arch_ptx)

  set(nvcc_flags "")
  set(nvcc_archs_readable "")

  # Tell NVCC to add binaries for the specified GPUs
  foreach(arch ${cuda_arch_bin})
    if(arch MATCHES "([0-9]+)\\(([0-9]+)\\)")
      # User explicitly specified PTX for the concrete BIN
      string(APPEND nvcc_flags
             " -gencode arch=compute_${CMAKE_MATCH_2},code=sm_${CMAKE_MATCH_1}")
      string(APPEND nvcc_archs_readable " sm_${CMAKE_MATCH_1}")
    else()
      # User didn't explicitly specify PTX for the concrete BIN, we assume PTX=BIN
      string(APPEND nvcc_flags " -gencode arch=compute_${arch},code=sm_${arch}")
      string(APPEND nvcc_archs_readable " sm_${arch}")
    endif()
  endforeach()

  # Tell NVCC to add PTX intermediate code for the specified architectures
  foreach(arch ${cuda_arch_ptx})
    string(APPEND nvcc_flags
           " -gencode arch=compute_${arch},code=compute_${arch}")
    string(APPEND nvcc_archs_readable " compute_${arch}")
  endforeach()

  string(REPLACE ";" " " nvcc_archs_readable "${nvcc_archs_readable}")
  set(${out_variable}
      ${nvcc_flags}
      PARENT_SCOPE)
  set(${out_variable}_readable
      ${nvcc_archs_readable}
      PARENT_SCOPE)
endfunction()

message(STATUS "CUDA detected: " ${CMAKE_CUDA_COMPILER_VERSION})
if(${CMAKE_CUDA_COMPILER_VERSION} LESS 11.0) # CUDA 10.x
  set(fd_known_gpu_archs ${fd_known_gpu_archs10})
  set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -D_MWAITXINTRIN_H_INCLUDED")
  set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -D__STRICT_ANSI__")
  set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
elseif(${CMAKE_CUDA_COMPILER_VERSION} LESS 11.2) # CUDA 11.0/11.1
  set(fd_known_gpu_archs ${fd_known_gpu_archs11})
  set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -D_MWAITXINTRIN_H_INCLUDED")
  set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -D__STRICT_ANSI__")
  set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
elseif(${CMAKE_CUDA_COMPILER_VERSION} LESS 12.0) # CUDA 11.2+
  set(fd_known_gpu_archs "${fd_known_gpu_archs11} 86")
  set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -D_MWAITXINTRIN_H_INCLUDED")
  set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -D__STRICT_ANSI__")
  set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
endif()

# setting nvcc arch flags
select_nvcc_arch_flags(NVCC_FLAGS_EXTRA)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} ${NVCC_FLAGS_EXTRA}")
message(STATUS "NVCC_FLAGS_EXTRA: ${NVCC_FLAGS_EXTRA}")

# Set C++14 support
set(CUDA_PROPAGATE_HOST_FLAGS OFF)
# Release/Debug flags set by cmake. Such as -O3 -g -DNDEBUG etc.
# So, don't set these flags here.
set(CMAKE_CUDA_STANDARD 11)

# (Note) For windows, if delete /W[1-4], /W1 will be added defaultly and conflic with -w
# So replace /W[1-4] with /W0
if(WIN32)
  string(REGEX REPLACE "/W[1-4]" " /W0 " CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS}")
endif()
# in cuda9, suppress cuda warning on eigen
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -w")
# Set :expt-relaxed-constexpr to suppress Eigen warnings
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} --expt-relaxed-constexpr")
# Set :expt-extended-lambda to enable HOSTDEVICE annotation on lambdas
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} --expt-extended-lambda")

if(WIN32)
  set(CMAKE_CUDA_FLAGS
      "${CMAKE_CUDA_FLAGS} -Xcompiler \"/wd4244 /wd4267 /wd4819 \"")
  set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Xcompiler /bigobj")
  if(MSVC_STATIC_CRT)
    foreach(flag_var
            CMAKE_CUDA_FLAGS CMAKE_CUDA_FLAGS_DEBUG CMAKE_CUDA_FLAGS_RELEASE
            CMAKE_CUDA_FLAGS_MINSIZEREL CMAKE_CUDA_FLAGS_RELWITHDEBINFO)
      if(${flag_var} MATCHES "-MD")
        string(REGEX REPLACE "-MD" "-MT" ${flag_var} "${${flag_var}}")
      endif()
    endforeach()
  endif()
endif()

mark_as_advanced(CUDA_BUILD_CUBIN CUDA_BUILD_EMULATION CUDA_VERBOSE_BUILD)
mark_as_advanced(CUDA_SDK_ROOT_DIR CUDA_SEPARABLE_COMPILATION)

Jun 10 '23 03:06 weiweijeff

我编译时也遇到。请到cmake文件夹修改cuda.cmake文件，把支持老显卡的算力代号 35删除.我的修改成这样了：

if(NOT WITH_GPU)
  return()
endif()

# This is to eliminate the CMP0104 warnings from cmake 3.18+.
# Instead of setting CUDA_ARCHITECTURES, we will set CMAKE_CUDA_FLAGS.
set(CMAKE_CUDA_ARCHITECTURES OFF)

if(BUILD_ON_JETSON)
  set(fd_known_gpu_archs "53 62 72")
  set(fd_known_gpu_archs10 "53 62 72")
else()
  message("Using New Release Strategy - All Arches Packge")
  set(fd_known_gpu_archs "70 75 80 86")
  set(fd_known_gpu_archs10 "70 75")
  set(fd_known_gpu_archs11 "70 75 80")
endif()

######################################################################################
# A function for automatic detection of GPUs installed  (if autodetection is enabled)
# Usage:
#   detect_installed_gpus(out_variable)
function(detect_installed_gpus out_variable)
  if(NOT CUDA_gpu_detect_output)
    set(cufile ${PROJECT_BINARY_DIR}/detect_cuda_archs.cu)

    file(
      WRITE ${cufile}
      ""
      "#include \"stdio.h\"\n"
      "#include \"cuda.h\"\n"
      "#include \"cuda_runtime.h\"\n"
      "int main() {\n"
      "  int count = 0;\n"
      "  if (cudaSuccess != cudaGetDeviceCount(&count)) return -1;\n"
      "  if (count == 0) return -1;\n"
      "  for (int device = 0; device < count; ++device) {\n"
      "    cudaDeviceProp prop;\n"
      "    if (cudaSuccess == cudaGetDeviceProperties(&prop, device))\n"
      "      printf(\"%d.%d \", prop.major, prop.minor);\n"
      "  }\n"
      "  return 0;\n"
      "}\n")

    execute_process(
      COMMAND "${CMAKE_CUDA_COMPILER}" "--run" "${cufile}"
      WORKING_DIRECTORY "${PROJECT_BINARY_DIR}/CMakeFiles/"
      RESULT_VARIABLE nvcc_res
      OUTPUT_VARIABLE nvcc_out
      ERROR_QUIET OUTPUT_STRIP_TRAILING_WHITESPACE)

    if(nvcc_res EQUAL 0)
      # only keep the last line of nvcc_out
      string(REGEX REPLACE ";" "\\\\;" nvcc_out "${nvcc_out}")
      string(REGEX REPLACE "\n" ";" nvcc_out "${nvcc_out}")
      list(GET nvcc_out -1 nvcc_out)
      string(REPLACE "2.1" "2.1(2.0)" nvcc_out "${nvcc_out}")
      set(CUDA_gpu_detect_output
          ${nvcc_out}
          CACHE INTERNAL
                "Returned GPU architetures from detect_installed_gpus tool"
                FORCE)
    endif()
  endif()

  if(NOT CUDA_gpu_detect_output)
    message(
      STATUS
        "Automatic GPU detection failed. Building for all known architectures.")
    set(${out_variable}
        ${fd_known_gpu_archs}
        PARENT_SCOPE)
  else()
    set(${out_variable}
        ${CUDA_gpu_detect_output}
        PARENT_SCOPE)
  endif()
endfunction()

########################################################################
# Function for selecting GPU arch flags for nvcc based on CUDA_ARCH_NAME
# Usage:
#   select_nvcc_arch_flags(out_variable)
function(select_nvcc_arch_flags out_variable)
  # List of arch names
  set(archs_names
      "Kepler"
      "Maxwell"
      "Pascal"
      "Volta"
      "Turing"
      "Ampere"
      "All"
      "Manual")
  set(archs_name_default "All")
  list(APPEND archs_names "Auto")

  # set CUDA_ARCH_NAME strings (so it will be seen as dropbox in CMake-Gui)
  set(CUDA_ARCH_NAME
      ${archs_name_default}
      CACHE STRING "Select target NVIDIA GPU achitecture.")
  set_property(CACHE CUDA_ARCH_NAME PROPERTY STRINGS "" ${archs_names})
  mark_as_advanced(CUDA_ARCH_NAME)

  # verify CUDA_ARCH_NAME value
  if(NOT ";${archs_names};" MATCHES ";${CUDA_ARCH_NAME};")
    string(REPLACE ";" ", " archs_names "${archs_names}")
    message(
      FATAL_ERROR "Only ${archs_names} architectures names are supported.")
  endif()

  if(${CUDA_ARCH_NAME} STREQUAL "Manual")
    set(CUDA_ARCH_BIN
        ${fd_known_gpu_archs}
        CACHE
          STRING
          "Specify 'real' GPU architectures to build binaries for, BIN(PTX) format is supported"
    )
    set(CUDA_ARCH_PTX
        ""
        CACHE
          STRING
          "Specify 'virtual' PTX architectures to build PTX intermediate code for"
    )
    mark_as_advanced(CUDA_ARCH_BIN CUDA_ARCH_PTX)
  else()
    unset(CUDA_ARCH_BIN CACHE)
    unset(CUDA_ARCH_PTX CACHE)
  endif()

  if(${CUDA_ARCH_NAME} STREQUAL "Maxwell")
    if(BUILD_ON_JETSON)
      set(cuda_arch_bin "53")
    else()
      set(cuda_arch_bin "50")
    endif()
  elseif(${CUDA_ARCH_NAME} STREQUAL "Pascal")
    if(BUILD_ON_JETSON)
      set(cuda_arch_bin "62")
    else()
      set(cuda_arch_bin "60 61")
    endif()
  elseif(${CUDA_ARCH_NAME} STREQUAL "Volta")
    if(BUILD_ON_JETSON)
      set(cuda_arch_bin "72")
    else()
      set(cuda_arch_bin "70")
    endif()
  elseif(${CUDA_ARCH_NAME} STREQUAL "Turing")
    set(cuda_arch_bin "75")
  elseif(${CUDA_ARCH_NAME} STREQUAL "Ampere")
    if(${CMAKE_CUDA_COMPILER_VERSION} LESS 11.1) # CUDA 11.0
      set(cuda_arch_bin "80")
    elseif(${CMAKE_CUDA_COMPILER_VERSION} LESS 12.0) # CUDA 11.1+
      set(cuda_arch_bin "80 86")
    endif()
  elseif(${CUDA_ARCH_NAME} STREQUAL "All")
    set(cuda_arch_bin ${fd_known_gpu_archs})
  elseif(${CUDA_ARCH_NAME} STREQUAL "Auto")
    message(
      STATUS
        "WARNING: This is just a warning for publishing release.
      You are building GPU version without supporting different architectures.
      So the wheel package may fail on other GPU architectures.
      You can add -DCUDA_ARCH_NAME=All in cmake command
      to get a full wheel package to resolve this warning.
      While, this version will still work on local GPU architecture.")
    detect_installed_gpus(cuda_arch_bin)
  else() # (${CUDA_ARCH_NAME} STREQUAL "Manual")
    set(cuda_arch_bin ${CUDA_ARCH_BIN})
  endif()

  if(NEW_RELEASE_JIT)
    set(cuda_arch_ptx "${cuda_arch_ptx}${cuda_arch_bin}")
    set(cuda_arch_bin "")
  endif()

  # remove dots and convert to lists
  string(REGEX REPLACE "\\." "" cuda_arch_bin "${cuda_arch_bin}")
  string(REGEX REPLACE "\\." "" cuda_arch_ptx "${cuda_arch_ptx}")
  string(REGEX MATCHALL "[0-9()]+" cuda_arch_bin "${cuda_arch_bin}")
  string(REGEX MATCHALL "[0-9]+" cuda_arch_ptx "${cuda_arch_ptx}")

  list(REMOVE_DUPLICATES cuda_arch_bin)
  list(REMOVE_DUPLICATES cuda_arch_ptx)

  set(nvcc_flags "")
  set(nvcc_archs_readable "")

  # Tell NVCC to add binaries for the specified GPUs
  foreach(arch ${cuda_arch_bin})
    if(arch MATCHES "([0-9]+)\\(([0-9]+)\\)")
      # User explicitly specified PTX for the concrete BIN
      string(APPEND nvcc_flags
             " -gencode arch=compute_${CMAKE_MATCH_2},code=sm_${CMAKE_MATCH_1}")
      string(APPEND nvcc_archs_readable " sm_${CMAKE_MATCH_1}")
    else()
      # User didn't explicitly specify PTX for the concrete BIN, we assume PTX=BIN
      string(APPEND nvcc_flags " -gencode arch=compute_${arch},code=sm_${arch}")
      string(APPEND nvcc_archs_readable " sm_${arch}")
    endif()
  endforeach()

  # Tell NVCC to add PTX intermediate code for the specified architectures
  foreach(arch ${cuda_arch_ptx})
    string(APPEND nvcc_flags
           " -gencode arch=compute_${arch},code=compute_${arch}")
    string(APPEND nvcc_archs_readable " compute_${arch}")
  endforeach()

  string(REPLACE ";" " " nvcc_archs_readable "${nvcc_archs_readable}")
  set(${out_variable}
      ${nvcc_flags}
      PARENT_SCOPE)
  set(${out_variable}_readable
      ${nvcc_archs_readable}
      PARENT_SCOPE)
endfunction()

message(STATUS "CUDA detected: " ${CMAKE_CUDA_COMPILER_VERSION})
if(${CMAKE_CUDA_COMPILER_VERSION} LESS 11.0) # CUDA 10.x
  set(fd_known_gpu_archs ${fd_known_gpu_archs10})
  set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -D_MWAITXINTRIN_H_INCLUDED")
  set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -D__STRICT_ANSI__")
  set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
elseif(${CMAKE_CUDA_COMPILER_VERSION} LESS 11.2) # CUDA 11.0/11.1
  set(fd_known_gpu_archs ${fd_known_gpu_archs11})
  set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -D_MWAITXINTRIN_H_INCLUDED")
  set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -D__STRICT_ANSI__")
  set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
elseif(${CMAKE_CUDA_COMPILER_VERSION} LESS 12.0) # CUDA 11.2+
  set(fd_known_gpu_archs "${fd_known_gpu_archs11} 86")
  set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -D_MWAITXINTRIN_H_INCLUDED")
  set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -D__STRICT_ANSI__")
  set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
endif()

# setting nvcc arch flags
select_nvcc_arch_flags(NVCC_FLAGS_EXTRA)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} ${NVCC_FLAGS_EXTRA}")
message(STATUS "NVCC_FLAGS_EXTRA: ${NVCC_FLAGS_EXTRA}")

# Set C++14 support
set(CUDA_PROPAGATE_HOST_FLAGS OFF)
# Release/Debug flags set by cmake. Such as -O3 -g -DNDEBUG etc.
# So, don't set these flags here.
set(CMAKE_CUDA_STANDARD 11)

# (Note) For windows, if delete /W[1-4], /W1 will be added defaultly and conflic with -w
# So replace /W[1-4] with /W0
if(WIN32)
  string(REGEX REPLACE "/W[1-4]" " /W0 " CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS}")
endif()
# in cuda9, suppress cuda warning on eigen
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -w")
# Set :expt-relaxed-constexpr to suppress Eigen warnings
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} --expt-relaxed-constexpr")
# Set :expt-extended-lambda to enable HOSTDEVICE annotation on lambdas
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} --expt-extended-lambda")

if(WIN32)
  set(CMAKE_CUDA_FLAGS
      "${CMAKE_CUDA_FLAGS} -Xcompiler \"/wd4244 /wd4267 /wd4819 \"")
  set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Xcompiler /bigobj")
  if(MSVC_STATIC_CRT)
    foreach(flag_var
            CMAKE_CUDA_FLAGS CMAKE_CUDA_FLAGS_DEBUG CMAKE_CUDA_FLAGS_RELEASE
            CMAKE_CUDA_FLAGS_MINSIZEREL CMAKE_CUDA_FLAGS_RELWITHDEBINFO)
      if(${flag_var} MATCHES "-MD")
        string(REGEX REPLACE "-MD" "-MT" ${flag_var} "${${flag_var}}")
      endif()
    endforeach()
  endif()
endif()

mark_as_advanced(CUDA_BUILD_CUBIN CUDA_BUILD_EMULATION CUDA_VERBOSE_BUILD)
mark_as_advanced(CUDA_SDK_ROOT_DIR CUDA_SEPARABLE_COMPILATION)

thx，我稍后去试试

Jun 10 '23 04:06 justghostof

我也遇到了,根据上述方案,已解决

Aug 07 '23 02:08 lym169

Helpful! Problem solved! Thx a lot!

Apr 30 '24 12:04 LateLinux

为什么我打不开cuda.cmake文件啊

May 06 '24 07:05 smalie2222

Windows用notepad，Ubuntu用nano，其他linux用touch

linux下注意文件权限。

在 2024-05-06 15:29:32，"smalie2222" @.***> 写道：

为什么我打不开cuda.cmake文件啊

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: @.***>

Jun 07 '24 03:06 LateLinux

FastDeploy FastDeploy copied to clipboard

在CUDA 12(Windows) 上构建失败:nvcc fatal : Unsupported gpu architecture 'compute_35'

环境

问题日志及出现问题的操作流程

FastDeploy
FastDeploy copied to clipboard