卡在这步了，麻烦帮忙瞅瞅，谢谢

Paddle version: 2.1.0
Paddle With CUDA: True

OS: Windows 10
Python version: 3.7.10

CUDA version: 10.2.89
cuDNN version: 7.6.5
Nvidia driver version: 462.30

• 安装方式信息：

本地编译：采用cmake-GUI配置，VS2015 UPDATE3编译
-cmake配置页面如下

• 出错片段如下

27> D:\work\build\Paddle\build\paddle\fluid\operators\eigen>"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\bin\nvcc.exe" -gencode=arch=compute_35,code=\"sm_35,compute_35\" -gencode=arch=compute_50,code=\"sm_50,compute_50\" -gencode=arch=compute_52,code=\"sm_52,compute_52\" -gencode=arch=compute_60,code=\"sm_60,compute_60\" -gencode=arch=compute_61,code=\"sm_61,compute_61\" -gencode=arch=compute_70,code=\"sm_70,compute_70\" -gencode=arch=compute_75,code=\"sm_75,compute_75\" --use-local-env 2015 -ccbin "D:\VS2015\VC\bin\x86_amd64" -x cu -ID:\work\build\Paddle\build -ID:\work\build\Paddle\paddle\fluid\framework\io -I"D:\3rdparty\TensorRT-7.0.0.11\include" -ID:\work\build\Paddle\build\third_party\install\zlib\include -ID:\work\build\Paddle\build\third_party\install -ID:\work\build\Paddle\build\third_party\install\gflags\include -ID:\work\build\Paddle\build\third_party\install\glog\include -ID:\work\build\Paddle\build\third_party\boost\src\extern_boost -ID:\work\build\Paddle\build\third_party\eigen3\src\extern_eigen3 -ID:\work\build\Paddle\build\third_party\threadpool\src\extern_threadpool -ID:\work\build\Paddle\build\third_party\dlpack\src\extern_dlpack\include -ID:\work\build\Paddle\build\third_party\install\xxhash\include -ID:\work\build\Paddle\build\third_party\install\warpctc\include -ID:\work\build\Paddle\build\third_party\install\mklml\include -ID:\work\build\Paddle\build\third_party\install\mkldnn\include -ID:\work\build\Paddle\build\third_party\install\protobuf\include -IC:\Users\Gaotiantian\anaconda3\include -I"C:\Users\Gaotiantian\anaconda3\Lib\site-packages\numpy\core\include" -ID:\work\build\Paddle\build\third_party\pybind\src\extern_pybind\include -ID:\work\build\Paddle\build\third_party\cub\src\extern_cub -ID:\work\build\Paddle\build\third_party\install\cryptopp\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" -ID:\work\build\Paddle -ID:\work\build\Paddle\build\..\paddle\fluid\framework\io -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" --keep-dir x64\Release -maxrregcount=0 --machine 64 --compile -cudart static -Wno-deprecated-gpu-targets -w --expt-relaxed-constexpr --expt-extended-lambda -O3 -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj /arch:AVX" -D_WINDOWS -D_MWAITXINTRIN_H_INCLUDED -D__STRICT_ANSI__ -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -D"CMAKE_INTDIR=\"Release\"" -D_MBCS -Xcompiler "/EHsc /W0 /nologo /O2 /FS /MT /GR" -o eigen_cu_function.dir\Release\broadcast.obj "D:\work\build\Paddle\paddle\fluid\operators\eigen\broadcast.cu" 27> nvcc fatal : A single input file is required for a non-link phase when an outputfile is specified 27>C:\Program Files (x86)\MSBuild\Microsoft.Cpp\v4.0\V140\BuildCustomizations\CUDA 10.2.targets(764,9): error MSB3721: 命令“"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\bin\nvcc.exe" -gencode=arch=compute_35,code=\"sm_35,compute_35\" -gencode=arch=compute_50,code=\"sm_50,compute_50\" -gencode=arch=compute_52,code=\"sm_52,compute_52\" -gencode=arch=compute_60,code=\"sm_60,compute_60\" -gencode=arch=compute_61,code=\"sm_61,compute_61\" -gencode=arch=compute_70,code=\"sm_70,compute_70\" -gencode=arch=compute_75,code=\"sm_75,compute_75\" --use-local-env 2015 -ccbin "D:\VS2015\VC\bin\x86_amd64" -x cu -ID:\work\build\Paddle\build -ID:\work\build\Paddle\paddle\fluid\framework\io -I"D:\3rdparty\TensorRT-7.0.0.11\include" -ID:\work\build\Paddle\build\third_party\install\zlib\include -ID:\work\build\Paddle\build\third_party\install -ID:\work\build\Paddle\build\third_party\install\gflags\include -ID:\work\build\Paddle\build\third_party\install\glog\include -ID:\work\build\Paddle\build\third_party\boost\src\extern_boost -ID:\work\build\Paddle\build\third_party\eigen3\src\extern_eigen3 -ID:\work\build\Paddle\build\third_party\threadpool\src\extern_threadpool -ID:\work\build\Paddle\build\third_party\dlpack\src\extern_dlpack\include -ID:\work\build\Paddle\build\third_party\install\xxhash\include -ID:\work\build\Paddle\build\third_party\install\warpctc\include -ID:\work\build\Paddle\build\third_party\install\mklml\include -ID:\work\build\Paddle\build\third_party\install\mkldnn\include -ID:\work\build\Paddle\build\third_party\install\protobuf\include -IC:\Users\Gaotiantian\anaconda3\include -I"C:\Users\Gaotiantian\anaconda3\Lib\site-packages\numpy\core\include" -ID:\work\build\Paddle\build\third_party\pybind\src\extern_pybind\include -ID:\work\build\Paddle\build\third_party\cub\src\extern_cub -ID:\work\build\Paddle\build\third_party\install\cryptopp\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" -ID:\work\build\Paddle -ID:\work\build\Paddle\build\..\paddle\fluid\framework\io -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" --keep-dir x64\Release -maxrregcount=0 --machine 64 --compile -cudart static -Wno-deprecated-gpu-targets -w --expt-relaxed-constexpr --expt-extended-lambda -O3 -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj /arch:AVX" -D_WINDOWS -D_MWAITXINTRIN_H_INCLUDED -D__STRICT_ANSI__ -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -D"CMAKE_INTDIR=\"Release\"" -D_MBCS -Xcompiler "/EHsc /W0 /nologo /O2 /FS /MT /GR" -o eigen_cu_function.dir\Release\broadcast.obj "D:\work\build\Paddle\paddle\fluid\operators\eigen\broadcast.cu"”已退出，返回代码为 1。 29>------ 已启动生成: 项目: extern_xxhash, 配置: Release x64 ------ 30>------ 已启动生成: 项目: framework_py_proto, 配置: Release x64 ------ 28> broadcast.cc 28>d:\work\build\paddle\build\third_party\eigen3\src\extern_eigen3\eigen\src/Core/ArithmeticSequence.h(345): fatal error C1001: 编译器中发生内部错误。 28> (编译器文件“f:\dd\vctools\compiler\cxxfe\sl\p1\c\p0gettok.c”，第 6502 行) 28> 要解决此问题，请尝试简化或更改上面所列位置附近的程序。 28> 请选择 Visual C++ 28> “帮助”菜单上的“技术支持”命令，或打开技术支持帮助文件来获得详细信息。 30> data_feed_pb2.py 30> distributed_strategy_pb2.py 30> framework_pb2.py 30> trainer_desc_pb2.py 30> __init__.py 30> 已复制 5 个文件。 30> distributed_strategy_pb2.py 30> 已复制 1 个文件。 31>------ 已启动生成: 项目: third_party, 配置: Release x64 ------ 32>------ 已启动生成: 项目: profiler, 配置: Release x64 ------ 32> Compiling CUDA source file ..\..\..\..\paddle\fluid\platform\profiler.cu... 32> 32> D:\work\build\Paddle\build\paddle\fluid\platform>"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\bin\nvcc.exe" -gencode=arch=compute_35,code=\"sm_35,compute_35\" -gencode=arch=compute_50,code=\"sm_50,compute_50\" -gencode=arch=compute_52,code=\"sm_52,compute_52\" -gencode=arch=compute_60,code=\"sm_60,compute_60\" -gencode=arch=compute_61,code=\"sm_61,compute_61\" -gencode=arch=compute_70,code=\"sm_70,compute_70\" -gencode=arch=compute_75,code=\"sm_75,compute_75\" --use-local-env 2015 -ccbin "D:\VS2015\VC\bin\x86_amd64" -x cu -ID:\work\build\Paddle\build -ID:\work\build\Paddle\paddle\fluid\framework\io -I"D:\3rdparty\TensorRT-7.0.0.11\include" -ID:\work\build\Paddle\build\third_party\install\zlib\include -ID:\work\build\Paddle\build\third_party\install -ID:\work\build\Paddle\build\third_party\install\gflags\include -ID:\work\build\Paddle\build\third_party\install\glog\include -ID:\work\build\Paddle\build\third_party\boost\src\extern_boost -ID:\work\build\Paddle\build\third_party\eigen3\src\extern_eigen3 -ID:\work\build\Paddle\build\third_party\threadpool\src\extern_threadpool -ID:\work\build\Paddle\build\third_party\dlpack\src\extern_dlpack\include -ID:\work\build\Paddle\build\third_party\install\xxhash\include -ID:\work\build\Paddle\build\third_party\install\warpctc\include -ID:\work\build\Paddle\build\third_party\install\mklml\include -ID:\work\build\Paddle\build\third_party\install\mkldnn\include -ID:\work\build\Paddle\build\third_party\install\protobuf\include -IC:\Users\Gaotiantian\anaconda3\include -I"C:\Users\Gaotiantian\anaconda3\Lib\site-packages\numpy\core\include" -ID:\work\build\Paddle\build\third_party\pybind\src\extern_pybind\include -ID:\work\build\Paddle\build\third_party\cub\src\extern_cub -ID:\work\build\Paddle\build\third_party\install\cryptopp\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" -ID:\work\build\Paddle -ID:\work\build\Paddle\build\..\paddle\fluid\framework\io -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" --keep-dir x64\Release -maxrregcount=0 --machine 64 --compile -cudart static -Wno-deprecated-gpu-targets -w --expt-relaxed-constexpr --expt-extended-lambda -O3 -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj /arch:AVX" -D_WINDOWS -D_MWAITXINTRIN_H_INCLUDED -D__STRICT_ANSI__ -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -DWIN32 -D_WINDOWS -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -D_MBCS -Xcompiler "/EHsc /W0 /nologo /O2 /FS /MT /GR" -o profiler.dir\Release\/profiler.cu.obj "D:\work\build\Paddle\paddle\fluid\platform\profiler.cu" 32> nvcc fatal : A single input file is required for a non-link phase when an outputfile is specified 32>C:\Program Files (x86)\MSBuild\Microsoft.Cpp\v4.0\V140\BuildCustomizations\CUDA 10.2.targets(764,9): error MSB3721: 命令“"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\bin\nvcc.exe" -gencode=arch=compute_35,code=\"sm_35,compute_35\" -gencode=arch=compute_50,code=\"sm_50,compute_50\" -gencode=arch=compute_52,code=\"sm_52,compute_52\" -gencode=arch=compute_60,code=\"sm_60,compute_60\" -gencode=arch=compute_61,code=\"sm_61,compute_61\" -gencode=arch=compute_70,code=\"sm_70,compute_70\" -gencode=arch=compute_75,code=\"sm_75,compute_75\" --use-local-env 2015 -ccbin "D:\VS2015\VC\bin\x86_amd64" -x cu -ID:\work\build\Paddle\build -ID:\work\build\Paddle\paddle\fluid\framework\io -I"D:\3rdparty\TensorRT-7.0.0.11\include" -ID:\work\build\Paddle\build\third_party\install\zlib\include -ID:\work\build\Paddle\build\third_party\install -ID:\work\build\Paddle\build\third_party\install\gflags\include -ID:\work\build\Paddle\build\third_party\install\glog\include -ID:\work\build\Paddle\build\third_party\boost\src\extern_boost -ID:\work\build\Paddle\build\third_party\eigen3\src\extern_eigen3 -ID:\work\build\Paddle\build\third_party\threadpool\src\extern_threadpool -ID:\work\build\Paddle\build\third_party\dlpack\src\extern_dlpack\include -ID:\work\build\Paddle\build\third_party\install\xxhash\include -ID:\work\build\Paddle\build\third_party\install\warpctc\include -ID:\work\build\Paddle\build\third_party\install\mklml\include -ID:\work\build\Paddle\build\third_party\install\mkldnn\include -ID:\work\build\Paddle\build\third_party\install\protobuf\include -IC:\Users\Gaotiantian\anaconda3\include -I"C:\Users\Gaotiantian\anaconda3\Lib\site-packages\numpy\core\include" -ID:\work\build\Paddle\build\third_party\pybind\src\extern_pybind\include -ID:\work\build\Paddle\build\third_party\cub\src\extern_cub -ID:\work\build\Paddle\build\third_party\install\cryptopp\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" -ID:\work\build\Paddle -ID:\work\build\Paddle\build\..\paddle\fluid\framework\io -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" --keep-dir x64\Release -maxrregcount=0 --machine 64 --compile -cudart static -Wno-deprecated-gpu-targets -w --expt-relaxed-constexpr --expt-extended-lambda -O3 -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj /arch:AVX" -D_WINDOWS -D_MWAITXINTRIN_H_INCLUDED -D__STRICT_ANSI__ -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -DWIN32 -D_WINDOWS -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -D_MBCS -Xcompiler "/EHsc /W0 /nologo /O2 /FS /MT /GR" -o profiler.dir\Release\/profiler.cu.obj "D:\work\build\Paddle\paddle\fluid\platform\profiler.cu"”已退出，返回代码为 1。 33>------ 已启动生成: 项目: device_context, 配置: Release x64 ------ 33> device_context.cc 33> Unknown compiler version - please run the configure tests and report the results 33>d:\work\build\paddle\build\third_party\eigen3\src\extern_eigen3\eigen\src/Core/ArithmeticSequence.h(345): fatal error C1001: 编译器中发生内部错误。 33> (编译器文件“f:\dd\vctools\compiler\cxxfe\sl\p1\c\p0gettok.c”，第 6502 行) 33> 要解决此问题，请尝试简化或更改上面所列位置附近的程序。 33> 请选择 Visual C++ 33> “帮助”菜单上的“技术支持”命令，或打开技术支持帮助文件来获得详细信息。 33> init.cc 33> Unknown compiler version - please run the configure tests and report the results 33>d:\work\build\paddle\build\third_party\eigen3\src\extern_eigen3\eigen\src/Core/ArithmeticSequence.h(345): fatal error C1001: 编译器中发生内部错误。 33> (编译器文件“f:\dd\vctools\compiler\cxxfe\sl\p1\c\p0gettok.c”，第 6502 行) 33> 要解决此问题，请尝试简化或更改上面所列位置附近的程序。 33> 请选择 Visual C++ 33> “帮助”菜单上的“技术支持”命令，或打开技术支持帮助文件来获得详细信息。 33> 正在生成代码... 34>------ 已启动生成: 项目: memcpy, 配置: Release x64 ------ 35>------ 已启动生成: 项目: data_type, 配置: Release x64 ------ 36>------ 已启动生成: 项目: blas, 配置: Release x64 ------ 37>------ 已启动生成: 项目: no_need_buffer_vars_inference, 配置: Release x64 ------ 38>------ 已启动生成: 项目: shape_inference, 配置: Release x64 ------ 39>------ 已启动生成: 项目: op_kernel_type, 配置: Release x64 ------ 40>------ 已启动生成: 项目: im2col, 配置: Release x64 ------ 41>------ 已启动生成: 项目: depthwise_conv, 配置: Release x64 ------ 42>------ 已启动生成: 项目: vol2col, 配置: Release x64 ------ 43>------ 已启动生成: 项目: pooling, 配置: Release x64 ------ 44>------ 已启动生成: 项目: prelu, 配置: Release x64 ------ 45>------ 已启动生成: 项目: matrix_bit_code, 配置: Release x64 ------ 46>------ 已启动生成: 项目: bert_encoder_functor, 配置: Release x64 ------ 47>------ 已启动生成: 项目: matrix_inverse, 配置: Release x64 ------ 48>------ 已启动生成: 项目: concat_and_split, 配置: Release x64 ------ 49>------ 已启动生成: 项目: sample_prob, 配置: Release x64 ------ 36> blas.cc 43> Compiling CUDA source file ..\..\..\..\..\paddle\fluid\operators\math\pooling.cu... 35> data_type.cc 37> no_need_buffer_vars_inference.cc 41> Compiling CUDA source file ..\..\..\..\..\paddle\fluid\operators\math\depthwise_conv.cu... 38> shape_inference.cc 42> Compiling CUDA source file ..\..\..\..\..\paddle\fluid\operators\math\vol2col.cu... 40> Compiling CUDA source file ..\..\..\..\..\paddle\fluid\operators\math\im2col.cu... 39> op_kernel_type.cc 34> memcpy.cc 44> Compiling CUDA source file ..\..\..\..\..\paddle\fluid\operators\math\prelu.cu... 48> Compiling CUDA source file ..\..\..\..\..\paddle\fluid\operators\math\concat_and_split.cu... 47> matrix_inverse.cc 49> Compiling CUDA source file ..\..\..\..\..\paddle\fluid\operators\math\sample_prob.cu... 45> matrix_bit_code.cc 46> Compiling CUDA source file ..\..\..\..\..\paddle\fluid\operators\math\bert_encoder_functor.cu... 38> Unknown compiler version - please run the configure tests and report the results 37> Unknown compiler version - please run the configure tests and report the results 40> 40> D:\work\build\Paddle\build\paddle\fluid\operators\math>"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\bin\nvcc.exe" -gencode=arch=compute_35,code=\"sm_35,compute_35\" -gencode=arch=compute_50,code=\"sm_50,compute_50\" -gencode=arch=compute_52,code=\"sm_52,compute_52\" -gencode=arch=compute_60,code=\"sm_60,compute_60\" -gencode=arch=compute_61,code=\"sm_61,compute_61\" -gencode=arch=compute_70,code=\"sm_70,compute_70\" -gencode=arch=compute_75,code=\"sm_75,compute_75\" --use-local-env 2015 -ccbin "D:\VS2015\VC\bin\x86_amd64" -x cu -ID:\work\build\Paddle\build -ID:\work\build\Paddle\paddle\fluid\framework\io -I"D:\3rdparty\TensorRT-7.0.0.11\include" -ID:\work\build\Paddle\build\third_party\install\zlib\include -ID:\work\build\Paddle\build\third_party\install -ID:\work\build\Paddle\build\third_party\install\gflags\include -ID:\work\build\Paddle\build\third_party\install\glog\include -ID:\work\build\Paddle\build\third_party\boost\src\extern_boost -ID:\work\build\Paddle\build\third_party\eigen3\src\extern_eigen3 -ID:\work\build\Paddle\build\third_party\threadpool\src\extern_threadpool -ID:\work\build\Paddle\build\third_party\dlpack\src\extern_dlpack\include -ID:\work\build\Paddle\build\third_party\install\xxhash\include -ID:\work\build\Paddle\build\third_party\install\warpctc\include -ID:\work\build\Paddle\build\third_party\install\mklml\include -ID:\work\build\Paddle\build\third_party\install\mkldnn\include -ID:\work\build\Paddle\build\third_party\install\protobuf\include -IC:\Users\Gaotiantian\anaconda3\include -I"C:\Users\Gaotiantian\anaconda3\Lib\site-packages\numpy\core\include" -ID:\work\build\Paddle\build\third_party\pybind\src\extern_pybind\include -ID:\work\build\Paddle\build\third_party\cub\src\extern_cub -ID:\work\build\Paddle\build\third_party\install\cryptopp\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" -ID:\work\build\Paddle -ID:\work\build\Paddle\build\..\paddle\fluid\framework\io -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" --keep-dir x64\Release -maxrregcount=0 --machine 64 --compile -cudart static -Wno-deprecated-gpu-targets -w --expt-relaxed-constexpr --expt-extended-lambda -O3 -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj /arch:AVX" -D_WINDOWS -D_MWAITXINTRIN_H_INCLUDED -D__STRICT_ANSI__ -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -DWIN32 -D_WINDOWS -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -D_MBCS -Xcompiler "/EHsc /W0 /nologo /O2 /FS /MT /GR" -o im2col.dir\Release\/im2col.cu.obj "D:\work\build\Paddle\paddle\fluid\operators\math\im2col.cu" 40> nvcc fatal : A single input file is required for a non-link phase when an outputfile is specified 42> 36> Unknown compiler version - please run the configure tests and report the results 42> D:\work\build\Paddle\build\paddle\fluid\operators\math>"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\bin\nvcc.exe" -gencode=arch=compute_35,code=\"sm_35,compute_35\" -gencode=arch=compute_50,code=\"sm_50,compute_50\" -gencode=arch=compute_52,code=\"sm_52,compute_52\" -gencode=arch=compute_60,code=\"sm_60,compute_60\" -gencode=arch=compute_61,code=\"sm_61,compute_61\" -gencode=arch=compute_70,code=\"sm_70,compute_70\" -gencode=arch=compute_75,code=\"sm_75,compute_75\" --use-local-env 2015 -ccbin "D:\VS2015\VC\bin\x86_amd64" -x cu -ID:\work\build\Paddle\build -ID:\work\build\Paddle\paddle\fluid\framework\io -I"D:\3rdparty\TensorRT-7.0.0.11\include" -ID:\work\build\Paddle\build\third_party\install\zlib\include -ID:\work\build\Paddle\build\third_party\install -ID:\work\build\Paddle\build\third_party\install\gflags\include -ID:\work\build\Paddle\build\third_party\install\glog\include -ID:\work\build\Paddle\build\third_party\boost\src\extern_boost -ID:\work\build\Paddle\build\third_party\eigen3\src\extern_eigen3 -ID:\work\build\Paddle\build\third_party\threadpool\src\extern_threadpool -ID:\work\build\Paddle\build\third_party\dlpack\src\extern_dlpack\include -ID:\work\build\Paddle\build\third_party\install\xxhash\include -ID:\work\build\Paddle\build\third_party\install\warpctc\include -ID:\work\build\Paddle\build\third_party\install\mklml\include -ID:\work\build\Paddle\build\third_party\install\mkldnn\include -ID:\work\build\Paddle\build\third_party\install\protobuf\include -IC:\Users\Gaotiantian\anaconda3\include -I"C:\Users\Gaotiantian\anaconda3\Lib\site-packages\numpy\core\include" -ID:\work\build\Paddle\build\third_party\pybind\src\extern_pybind\include -ID:\work\build\Paddle\build\third_party\cub\src\extern_cub -ID:\work\build\Paddle\build\third_party\install\cryptopp\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" -ID:\work\build\Paddle -ID:\work\build\Paddle\build\..\paddle\fluid\framework\io -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" --keep-dir x64\Release -maxrregcount=0 --machine 64 --compile -cudart static -Wno-deprecated-gpu-targets -w --expt-relaxed-constexpr --expt-extended-lambda -O3 -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj /arch:AVX" -D_WINDOWS -D_MWAITXINTRIN_H_INCLUDED -D__STRICT_ANSI__ -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -DWIN32 -D_WINDOWS -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -D_MBCS -Xcompiler "/EHsc /W0 /nologo /O2 /FS /MT /GR" -o vol2col.dir\Release\/vol2col.cu.obj "D:\work\build\Paddle\paddle\fluid\operators\math\vol2col.cu" 40>C:\Program Files (x86)\MSBuild\Microsoft.Cpp\v4.0\V140\BuildCustomizations\CUDA 10.2.targets(764,9): error MSB3721: 命令“"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\bin\nvcc.exe" -gencode=arch=compute_35,code=\"sm_35,compute_35\" -gencode=arch=compute_50,code=\"sm_50,compute_50\" -gencode=arch=compute_52,code=\"sm_52,compute_52\" -gencode=arch=compute_60,code=\"sm_60,compute_60\" -gencode=arch=compute_61,code=\"sm_61,compute_61\" -gencode=arch=compute_70,code=\"sm_70,compute_70\" -gencode=arch=compute_75,code=\"sm_75,compute_75\" --use-local-env 2015 -ccbin "D:\VS2015\VC\bin\x86_amd64" -x cu -ID:\work\build\Paddle\build -ID:\work\build\Paddle\paddle\fluid\framework\io -I"D:\3rdparty\TensorRT-7.0.0.11\include" -ID:\work\build\Paddle\build\third_party\install\zlib\include -ID:\work\build\Paddle\build\third_party\install -ID:\work\build\Paddle\build\third_party\install\gflags\include -ID:\work\build\Paddle\build\third_party\install\glog\include -ID:\work\build\Paddle\build\third_party\boost\src\extern_boost -ID:\work\build\Paddle\build\third_party\eigen3\src\extern_eigen3 -ID:\work\build\Paddle\build\third_party\threadpool\src\extern_threadpool -ID:\work\build\Paddle\build\third_party\dlpack\src\extern_dlpack\include -ID:\work\build\Paddle\build\third_party\install\xxhash\include -ID:\work\build\Paddle\build\third_party\install\warpctc\include -ID:\work\build\Paddle\build\third_party\install\mklml\include -ID:\work\build\Paddle\build\third_party\install\mkldnn\include -ID:\work\build\Paddle\build\third_party\install\protobuf\include -IC:\Users\Gaotiantian\anaconda3\include -I"C:\Users\Gaotiantian\anaconda3\Lib\site-packages\numpy\core\include" -ID:\work\build\Paddle\build\third_party\pybind\src\extern_pybind\include -ID:\work\build\Paddle\build\third_party\cub\src\extern_cub -ID:\work\build\Paddle\build\third_party\install\cryptopp\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" -ID:\work\build\Paddle -ID:\work\build\Paddle\build\..\paddle\fluid\framework\io -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" --keep-dir x64\Release -maxrregcount=0 --machine 64 --compile -cudart static -Wno-deprecated-gpu-targets -w --expt-relaxed-constexpr --expt-extended-lambda -O3 -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj /arch:AVX" -D_WINDOWS -D_MWAITXINTRIN_H_INCLUDED -D__STRICT_ANSI__ -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -DWIN32 -D_WINDOWS -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -D_MBCS -Xcompiler "/EHsc /W0 /nologo /O2 /FS /MT /GR" -o im2col.dir\Release\/im2col.cu.obj "D:\work\build\Paddle\paddle\fluid\operators\math\im2col.cu"”已退出，返回代码为 1。 42> nvcc fatal : A single input file is required for a non-link phase when an outputfile is specified 42>C:\Program Files (x86)\MSBuild\Microsoft.Cpp\v4.0\V140\BuildCustomizations\CUDA 10.2.targets(764,9): error MSB3721: 命令“"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\bin\nvcc.exe" -gencode=arch=compute_35,code=\"sm_35,compute_35\" -gencode=arch=compute_50,code=\"sm_50,compute_50\" -gencode=arch=compute_52,code=\"sm_52,compute_52\" -gencode=arch=compute_60,code=\"sm_60,compute_60\" -gencode=arch=compute_61,code=\"sm_61,compute_61\" -gencode=arch=compute_70,code=\"sm_70,compute_70\" -gencode=arch=compute_75,code=\"sm_75,compute_75\" --use-local-env 2015 -ccbin "D:\VS2015\VC\bin\x86_amd64" -x cu -ID:\work\build\Paddle\build -ID:\work\build\Paddle\paddle\fluid\framework\io -I"D:\3rdparty\TensorRT-7.0.0.11\include" -ID:\work\build\Paddle\build\third_party\install\zlib\include -ID:\work\build\Paddle\build\third_party\install -ID:\work\build\Paddle\build\third_party\install\gflags\include -ID:\work\build\Paddle\build\third_party\install\glog\include -ID:\work\build\Paddle\build\third_party\boost\src\extern_boost -ID:\work\build\Paddle\build\third_party\eigen3\src\extern_eigen3 -ID:\work\build\Paddle\build\third_party\threadpool\src\extern_threadpool -ID:\work\build\Paddle\build\third_party\dlpack\src\extern_dlpack\include -ID:\work\build\Paddle\build\third_party\install\xxhash\include -ID:\work\build\Paddle\build\third_party\install\warpctc\include -ID:\work\build\Paddle\build\third_party\install\mklml\include -ID:\work\build\Paddle\build\third_party\install\mkldnn\include -ID:\work\build\Paddle\build\third_party\install\protobuf\include -IC:\Users\Gaotiantian\anaconda3\include -I"C:\Users\Gaotiantian\anaconda3\Lib\site-packages\numpy\core\include" -ID:\work\build\Paddle\build\third_party\pybind\src\extern_pybind\include -ID:\work\build\Paddle\build\third_party\cub\src\extern_cub -ID:\work\build\Paddle\build\third_party\install\cryptopp\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" -ID:\work\build\Paddle -ID:\work\build\Paddle\build\..\paddle\fluid\framework\io -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" --keep-dir x64\Release -maxrregcount=0 --machine 64 --compile -cudart static -Wno-deprecated-gpu-targets -w --expt-relaxed-constexpr --expt-extended-lambda -O3 -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj /arch:AVX" -D_WINDOWS -D_MWAITXINTRIN_H_INCLUDED -D__STRICT_ANSI__ -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -DWIN32 -D_WINDOWS -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -D_MBCS -Xcompiler "/EHsc /W0 /nologo /O2 /FS /MT /GR" -o vol2col.dir\Release\/vol2col.cu.obj "D:\work\build\Paddle\paddle\fluid\operators\math\vol2col.cu"”已退出，返回代码为 1。 50>------ 已启动生成: 项目: cross_entropy, 配置: Release x64 ------ 41> 41> D:\work\build\Paddle\build\paddle\fluid\operators\math>"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\bin\nvcc.exe" -gencode=arch=compute_35,code=\"sm_35,compute_35\" -gencode=arch=compute_50,code=\"sm_50,compute_50\" -gencode=arch=compute_52,code=\"sm_52,compute_52\" -gencode=arch=compute_60,code=\"sm_60,compute_60\" -gencode=arch=compute_61,code=\"sm_61,compute_61\" -gencode=arch=compute_70,code=\"sm_70,compute_70\" -gencode=arch=compute_75,code=\"sm_75,compute_75\" --use-local-env 2015 -ccbin "D:\VS2015\VC\bin\x86_amd64" -x cu -ID:\work\build\Paddle\build -ID:\work\build\Paddle\paddle\fluid\framework\io -I"D:\3rdparty\TensorRT-7.0.0.11\include" -ID:\work\build\Paddle\build\third_party\install\zlib\include -ID:\work\build\Paddle\build\third_party\install -ID:\work\build\Paddle\build\third_party\install\gflags\include -ID:\work\build\Paddle\build\third_party\install\glog\include -ID:\work\build\Paddle\build\third_party\boost\src\extern_boost -ID:\work\build\Paddle\build\third_party\eigen3\src\extern_eigen3 -ID:\work\build\Paddle\build\third_party\threadpool\src\extern_threadpool -ID:\work\build\Paddle\build\third_party\dlpack\src\extern_dlpack\include -ID:\work\build\Paddle\build\third_party\install\xxhash\include -ID:\work\build\Paddle\build\third_party\install\warpctc\include -ID:\work\build\Paddle\build\third_party\install\mklml\include -ID:\work\build\Paddle\build\third_party\install\mkldnn\include -ID:\work\build\Paddle\build\third_party\install\protobuf\include -IC:\Users\Gaotiantian\anaconda3\include -I"C:\Users\Gaotiantian\anaconda3\Lib\site-packages\numpy\core\include" -ID:\work\build\Paddle\build\third_party\pybind\src\extern_pybind\include -ID:\work\build\Paddle\build\third_party\cub\src\extern_cub -ID:\work\build\Paddle\build\third_party\install\cryptopp\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" -ID:\work\build\Paddle -ID:\work\build\Paddle\build\..\paddle\fluid\framework\io -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" --keep-dir x64\Release -maxrregcount=0 --machine 64 --compile -cudart static -Wno-deprecated-gpu-targets -w --expt-relaxed-constexpr --expt-extended-lambda -O3 -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj /arch:AVX" -D_WINDOWS -D_MWAITXINTRIN_H_INCLUDED -D__STRICT_ANSI__ -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -DWIN32 -D_WINDOWS -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -D_MBCS -Xcompiler "/EHsc /W0 /nologo /O2 /FS /MT /GR" -o depthwise_conv.dir\Release\depthwise_conv.obj "D:\work\build\Paddle\paddle\fluid\operators\math\depthwise_conv.cu" 41> nvcc fatal : A single input file is required for a non-link phase when an outputfile is specified 41>C:\Program Files (x86)\MSBuild\Microsoft.Cpp\v4.0\V140\BuildCustomizations\CUDA 10.2.targets(764,9): error MSB3721: 命令“"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\bin\nvcc.exe" -gencode=arch=compute_35,code=\"sm_35,compute_35\" -gencode=arch=compute_50,code=\"sm_50,compute_50\" -gencode=arch=compute_52,code=\"sm_52,compute_52\" -gencode=arch=compute_60,code=\"sm_60,compute_60\" -gencode=arch=compute_61,code=\"sm_61,compute_61\" -gencode=arch=compute_70,code=\"sm_70,compute_70\" -gencode=arch=compute_75,code=\"sm_75,compute_75\" --use-local-env 2015 -ccbin "D:\VS2015\VC\bin\x86_amd64" -x cu -ID:\work\build\Paddle\build -ID:\work\build\Paddle\paddle\fluid\framework\io -I"D:\3rdparty\TensorRT-7.0.0.11\include" -ID:\work\build\Paddle\build\third_party\install\zlib\include -ID:\work\build\Paddle\build\third_party\install -ID:\work\build\Paddle\build\third_party\install\gflags\include -ID:\work\build\Paddle\build\third_party\install\glog\include -ID:\work\build\Paddle\build\third_party\boost\src\extern_boost -ID:\work\build\Paddle\build\third_party\eigen3\src\extern_eigen3 -ID:\work\build\Paddle\build\third_party\threadpool\src\extern_threadpool -ID:\work\build\Paddle\build\third_party\dlpack\src\extern_dlpack\include -ID:\work\build\Paddle\build\third_party\install\xxhash\include -ID:\work\build\Paddle\build\third_party\install\warpctc\include -ID:\work\build\Paddle\build\third_party\install\mklml\include -ID:\work\build\Paddle\build\third_party\install\mkldnn\include -ID:\work\build\Paddle\build\third_party\install\protobuf\include -IC:\Users\Gaotiantian\anaconda3\include -I"C:\Users\Gaotiantian\anaconda3\Lib\site-packages\numpy\core\include" -ID:\work\build\Paddle\build\third_party\pybind\src\extern_pybind\include -ID:\work\build\Paddle\build\third_party\cub\src\extern_cub -ID:\work\build\Paddle\build\third_party\install\cryptopp\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" -ID:\work\build\Paddle -ID:\work\build\Paddle\build\..\paddle\fluid\framework\io -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" --keep-dir x64\Release -maxrregcount=0 --machine 64 --compile -cudart static -Wno-deprecated-gpu-targets -w --expt-relaxed-constexpr --expt-extended-lambda -O3 -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj /arch:AVX" -D_WINDOWS -D_MWAITXINTRIN_H_INCLUDED -D__STRICT_ANSI__ -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -DWIN32 -D_WINDOWS -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -D_MBCS -Xcompiler "/EHsc /W0 /nologo /O2 /FS /MT /GR" -o depthwise_conv.dir\Release\depthwise_conv.obj "D:\work\build\Paddle\paddle\fluid\operators\math\depthwise_conv.cu"”已退出，返回代码为 1。 48> 44> 44> D:\work\build\Paddle\build\paddle\fluid\operators\math>"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\bin\nvcc.exe" -gencode=arch=compute_35,code=\"sm_35,compute_35\" -gencode=arch=compute_50,code=\"sm_50,compute_50\" -gencode=arch=compute_52,code=\"sm_52,compute_52\" -gencode=arch=compute_60,code=\"sm_60,compute_60\" -gencode=arch=compute_61,code=\"sm_61,compute_61\" -gencode=arch=compute_70,code=\"sm_70,compute_70\" -gencode=arch=compute_75,code=\"sm_75,compute_75\" --use-local-env 2015 -ccbin "D:\VS2015\VC\bin\x86_amd64" -x cu -ID:\work\build\Paddle\build -ID:\work\build\Paddle\paddle\fluid\framework\io -I"D:\3rdparty\TensorRT-7.0.0.11\include" -ID:\work\build\Paddle\build\third_party\install\zlib\include -ID:\work\build\Paddle\build\third_party\install -ID:\work\build\Paddle\build\third_party\install\gflags\include -ID:\work\build\Paddle\build\third_party\install\glog\include -ID:\work\build\Paddle\build\third_party\boost\src\extern_boost -ID:\work\build\Paddle\build\third_party\eigen3\src\extern_eigen3 -ID:\work\build\Paddle\build\third_party\threadpool\src\extern_threadpool -ID:\work\build\Paddle\build\third_party\dlpack\src\extern_dlpack\include -ID:\work\build\Paddle\build\third_party\install\xxhash\include -ID:\work\build\Paddle\build\third_party\install\warpctc\include -ID:\work\build\Paddle\build\third_party\install\mklml\include -ID:\work\build\Paddle\build\third_party\install\mkldnn\include -ID:\work\build\Paddle\build\third_party\install\protobuf\include -IC:\Users\Gaotiantian\anaconda3\include -I"C:\Users\Gaotiantian\anaconda3\Lib\site-packages\numpy\core\include" -ID:\work\build\Paddle\build\third_party\pybind\src\extern_pybind\include -ID:\work\build\Paddle\build\third_party\cub\src\extern_cub -ID:\work\build\Paddle\build\third_party\install\cryptopp\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" -ID:\work\build\Paddle -ID:\work\build\Paddle\build\..\paddle\fluid\framework\io -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" --keep-dir x64\Release -maxrregcount=0 --machine 64 --compile -cudart static -Wno-deprecated-gpu-targets -w --expt-relaxed-constexpr --expt-extended-lambda -O3 -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj /arch:AVX" -D_WINDOWS -D_MWAITXINTRIN_H_INCLUDED -D__STRICT_ANSI__ -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -DWIN32 -D_WINDOWS -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -D_MBCS -Xcompiler "/EHsc /W0 /nologo /O2 /FS /MT /GR" -o prelu.dir\Release\prelu.obj "D:\work\build\Paddle\paddle\fluid\operators\math\prelu.cu" 48> D:\work\build\Paddle\build\paddle\fluid\operators\math>"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\bin\nvcc.exe" -gencode=arch=compute_35,code=\"sm_35,compute_35\" -gencode=arch=compute_50,code=\"sm_50,compute_50\" -gencode=arch=compute_52,code=\"sm_52,compute_52\" -gencode=arch=compute_60,code=\"sm_60,compute_60\" -gencode=arch=compute_61,code=\"sm_61,compute_61\" -gencode=arch=compute_70,code=\"sm_70,compute_70\" -gencode=arch=compute_75,code=\"sm_75,compute_75\" --use-local-env 2015 -ccbin "D:\VS2015\VC\bin\x86_amd64" -x cu -ID:\work\build\Paddle\build -ID:\work\build\Paddle\paddle\fluid\framework\io -I"D:\3rdparty\TensorRT-7.0.0.11\include" -ID:\work\build\Paddle\build\third_party\install\zlib\include -ID:\work\build\Paddle\build\third_party\install -ID:\work\build\Paddle\build\third_party\install\gflags\include -ID:\work\build\Paddle\build\third_party\install\glog\include -ID:\work\build\Paddle\build\third_party\boost\src\extern_boost -ID:\work\build\Paddle\build\third_party\eigen3\src\extern_eigen3 -ID:\work\build\Paddle\build\third_party\threadpool\src\extern_threadpool -ID:\work\build\Paddle\build\third_party\dlpack\src\extern_dlpack\include -ID:\work\build\Paddle\build\third_party\install\xxhash\include -ID:\work\build\Paddle\build\third_party\install\warpctc\include -ID:\work\build\Paddle\build\third_party\install\mklml\include -ID:\work\build\Paddle\build\third_party\install\mkldnn\include -ID:\work\build\Paddle\build\third_party\install\protobuf\include -IC:\Users\Gaotiantian\anaconda3\include -I"C:\Users\Gaotiantian\anaconda3\Lib\site-packages\numpy\core\include" -ID:\work\build\Paddle\build\third_party\pybind\src\extern_pybind\include -ID:\work\build\Paddle\build\third_party\cub\src\extern_cub -ID:\work\build\Paddle\build\third_party\install\cryptopp\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" -ID:\work\build\Paddle -ID:\work\build\Paddle\build\..\paddle\fluid\framework\io -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" --keep-dir x64\Release -maxrregcount=0 --machine 64 --compile -cudart static -Wno-deprecated-gpu-targets -w --expt-relaxed-constexpr --expt-extended-lambda -O3 -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj /arch:AVX" -D_WINDOWS -D_MWAITXINTRIN_H_INCLUDED -D__STRICT_ANSI__ -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -DWIN32 -D_WINDOWS -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -D_MBCS -Xcompiler "/EHsc /W0 /nologo /O2 /FS /MT /GR" -o concat_and_split.dir\Release\/concat_and_split.cu.obj "D:\work\build\Paddle\paddle\fluid\operators\math\concat_and_split.cu" 44> nvcc fatal : A single input file is required for a non-link phase when an outputfile is specified 48> nvcc fatal : A single input file is required for a non-link phase when an outputfile is specified 44>C:\Program Files (x86)\MSBuild\Microsoft.Cpp\v4.0\V140\BuildCustomizations\CUDA 10.2.targets(764,9): error MSB3721: 命令“"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\bin\nvcc.exe" -gencode=arch=compute_35,code=\"sm_35,compute_35\" -gencode=arch=compute_50,code=\"sm_50,compute_50\" -gencode=arch=compute_52,code=\"sm_52,compute_52\" -gencode=arch=compute_60,code=\"sm_60,compute_60\" -gencode=arch=compute_61,code=\"sm_61,compute_61\" -gencode=arch=compute_70,code=\"sm_70,compute_70\" -gencode=arch=compute_75,code=\"sm_75,compute_75\" --use-local-env 2015 -ccbin "D:\VS2015\VC\bin\x86_amd64" -x cu -ID:\work\build\Paddle\build -ID:\work\build\Paddle\paddle\fluid\framework\io -I"D:\3rdparty\TensorRT-7.0.0.11\include" -ID:\work\build\Paddle\build\third_party\install\zlib\include -ID:\work\build\Paddle\build\third_party\install -ID:\work\build\Paddle\build\third_party\install\gflags\include -ID:\work\build\Paddle\build\third_party\install\glog\include -ID:\work\build\Paddle\build\third_party\boost\src\extern_boost -ID:\work\build\Paddle\build\third_party\eigen3\src\extern_eigen3 -ID:\work\build\Paddle\build\third_party\threadpool\src\extern_threadpool -ID:\work\build\Paddle\build\third_party\dlpack\src\extern_dlpack\include -ID:\work\build\Paddle\build\third_party\install\xxhash\include -ID:\work\build\Paddle\build\third_party\install\warpctc\include -ID:\work\build\Paddle\build\third_party\install\mklml\include -ID:\work\build\Paddle\build\third_party\install\mkldnn\include -ID:\work\build\Paddle\build\third_party\install\protobuf\include -IC:\Users\Gaotiantian\anaconda3\include -I"C:\Users\Gaotiantian\anaconda3\Lib\site-packages\numpy\core\include" -ID:\work\build\Paddle\build\third_party\pybind\src\extern_pybind\include -ID:\work\build\Paddle\build\third_party\cub\src\extern_cub -ID:\work\build\Paddle\build\third_party\install\cryptopp\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" -ID:\work\build\Paddle -ID:\work\build\Paddle\build\..\paddle\fluid\framework\io -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" --keep-dir x64\Release -maxrregcount=0 --machine 64 --compile -cudart static -Wno-deprecated-gpu-targets -w --expt-relaxed-constexpr --expt-extended-lambda -O3 -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj /arch:AVX" -D_WINDOWS -D_MWAITXINTRIN_H_INCLUDED -D__STRICT_ANSI__ -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -DWIN32 -D_WINDOWS -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -D_MBCS -Xcompiler "/EHsc /W0 /nologo /O2 /FS /MT /GR" -o prelu.dir\Release\prelu.obj "D:\work\build\Paddle\paddle\fluid\operators\math\prelu.cu"”已退出，返回代码为 1。 48>C:\Program Files (x86)\MSBuild\Microsoft.Cpp\v4.0\V140\BuildCustomizations\CUDA 10.2.targets(764,9): error MSB3721: 命令“"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\bin\nvcc.exe" -gencode=arch=compute_35,code=\"sm_35,compute_35\" -gencode=arch=compute_50,code=\"sm_50,compute_50\" -gencode=arch=compute_52,code=\"sm_52,compute_52\" -gencode=arch=compute_60,code=\"sm_60,compute_60\" -gencode=arch=compute_61,code=\"sm_61,compute_61\" -gencode=arch=compute_70,code=\"sm_70,compute_70\" -gencode=arch=compute_75,code=\"sm_75,compute_75\" --use-local-env 2015 -ccbin "D:\VS2015\VC\bin\x86_amd64" -x cu -ID:\work\build\Paddle\build -ID:\work\build\Paddle\paddle\fluid\framework\io -I"D:\3rdparty\TensorRT-7.0.0.11\include" -ID:\work\build\Paddle\build\third_party\install\zlib\include -ID:\work\build\Paddle\build\third_party\install -ID:\work\build\Paddle\build\third_party\install\gflags\include -ID:\work\build\Paddle\build\third_party\install\glog\include -ID:\work\build\Paddle\build\third_party\boost\src\extern_boost -ID:\work\build\Paddle\build\third_party\eigen3\src\extern_eigen3 -ID:\work\build\Paddle\build\third_party\threadpool\src\extern_threadpool -ID:\work\build\Paddle\build\third_party\dlpack\src\extern_dlpack\include -ID:\work\build\Paddle\build\third_party\install\xxhash\include -ID:\work\build\Paddle\build\third_party\install\warpctc\include -ID:\work\build\Paddle\build\third_party\install\mklml\include -ID:\work\build\Paddle\build\third_party\install\mkldnn\include -ID:\work\build\Paddle\build\third_party\install\protobuf\include -IC:\Users\Gaotiantian\anaconda3\include -I"C:\Users\Gaotiantian\anaconda3\Lib\site-packages\numpy\core\include" -ID:\work\build\Paddle\build\third_party\pybind\src\extern_pybind\include -ID:\work\build\Paddle\build\third_party\cub\src\extern_cub -ID:\work\build\Paddle\build\third_party\install\cryptopp\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" -ID:\work\build\Paddle -ID:\work\build\Paddle\build\..\paddle\fluid\framework\io -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" --keep-dir x64\Release -maxrregcount=0 --machine 64 --compile -cudart static -Wno-deprecated-gpu-targets -w --expt-relaxed-constexpr --expt-extended-lambda -O3 -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj /arch:AVX" -D_WINDOWS -D_MWAITXINTRIN_H_INCLUDED -D__STRICT_ANSI__ -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -DWIN32 -D_WINDOWS -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -D_MBCS -Xcompiler "/EHsc /W0 /nologo /O2 /FS /MT /GR" -o concat_and_split.dir\Release\/concat_and_split.cu.obj "D:\work\build\Paddle\paddle\fluid\operators\math\concat_and_split.cu"”已退出，返回代码为 1。 47>d:\work\build\paddle\build\third_party\eigen3\src\extern_eigen3\eigen\src/Core/ArithmeticSequence.h(345): fatal error C1001: 编译器中发生内部错误。 43> 49> 43> D:\work\build\Paddle\build\paddle\fluid\operators\math>"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\bin\nvcc.exe" -gencode=arch=compute_35,code=\"sm_35,compute_35\" -gencode=arch=compute_50,code=\"sm_50,compute_50\" -gencode=arch=compute_52,code=\"sm_52,compute_52\" -gencode=arch=compute_60,code=\"sm_60,compute_60\" -gencode=arch=compute_61,code=\"sm_61,compute_61\" -gencode=arch=compute_70,code=\"sm_70,compute_70\" -gencode=arch=compute_75,code=\"sm_75,compute_75\" --use-local-env 2015 -ccbin "D:\VS2015\VC\bin\x86_amd64" -x cu -ID:\work\build\Paddle\build -ID:\work\build\Paddle\paddle\fluid\framework\io -I"D:\3rdparty\TensorRT-7.0.0.11\include" -ID:\work\build\Paddle\build\third_party\install\zlib\include -ID:\work\build\Paddle\build\third_party\install -ID:\work\build\Paddle\build\third_party\install\gflags\include -ID:\work\build\Paddle\build\third_party\install\glog\include -ID:\work\build\Paddle\build\third_party\boost\src\extern_boost -ID:\work\build\Paddle\build\third_party\eigen3\src\extern_eigen3 -ID:\work\build\Paddle\build\third_party\threadpool\src\extern_threadpool -ID:\work\build\Paddle\build\third_party\dlpack\src\extern_dlpack\include -ID:\work\build\Paddle\build\third_party\install\xxhash\include -ID:\work\build\Paddle\build\third_party\install\warpctc\include -ID:\work\build\Paddle\build\third_party\install\mklml\include -ID:\work\build\Paddle\build\third_party\install\mkldnn\include -ID:\work\build\Paddle\build\third_party\install\protobuf\include -IC:\Users\Gaotiantian\anaconda3\include -I"C:\Users\Gaotiantian\anaconda3\Lib\site-packages\numpy\core\include" -ID:\work\build\Paddle\build\third_party\pybind\src\extern_pybind\include -ID:\work\build\Paddle\build\third_party\cub\src\extern_cub -ID:\work\build\Paddle\build\third_party\install\cryptopp\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" -ID:\work\build\Paddle -ID:\work\build\Paddle\build\..\paddle\fluid\framework\io -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" --keep-dir x64\Release -maxrregcount=0 --machine 64 --compile -cudart static -Wno-deprecated-gpu-targets -w --expt-relaxed-constexpr --expt-extended-lambda -O3 -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj /arch:AVX" -D_WINDOWS -D_MWAITXINTRIN_H_INCLUDED -D__STRICT_ANSI__ -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -DWIN32 -D_WINDOWS -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -D_MBCS -Xcompiler "/EHsc /W0 /nologo /O2 /FS /MT /GR" -o pooling.dir\Release\/pooling.cu.obj "D:\work\build\Paddle\paddle\fluid\operators\math\pooling.cu" 43> nvcc fatal : A single input file is required for a non-link phase when an outputfile is specified 49> D:\work\build\Paddle\build\paddle\fluid\operators\math>"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\bin\nvcc.exe" -gencode=arch=compute_35,code=\"sm_35,compute_35\" -gencode=arch=compute_50,code=\"sm_50,compute_50\" -gencode=arch=compute_52,code=\"sm_52,compute_52\" -gencode=arch=compute_60,code=\"sm_60,compute_60\" -gencode=arch=compute_61,code=\"sm_61,compute_61\" -gencode=arch=compute_70,code=\"sm_70,compute_70\" -gencode=arch=compute_75,code=\"sm_75,compute_75\" --use-local-env 2015 -ccbin "D:\VS2015\VC\bin\x86_amd64" -x cu -ID:\work\build\Paddle\build -ID:\work\build\Paddle\paddle\fluid\framework\io -I"D:\3rdparty\TensorRT-7.0.0.11\include" -ID:\work\build\Paddle\build\third_party\install\zlib\include -ID:\work\build\Paddle\build\third_party\install -ID:\work\build\Paddle\build\third_party\install\gflags\include -ID:\work\build\Paddle\build\third_party\install\glog\include -ID:\work\build\Paddle\build\third_party\boost\src\extern_boost -ID:\work\build\Paddle\build\third_party\eigen3\src\extern_eigen3 -ID:\work\build\Paddle\build\third_party\threadpool\src\extern_threadpool -ID:\work\build\Paddle\build\third_party\dlpack\src\extern_dlpack\include -ID:\work\build\Paddle\build\third_party\install\xxhash\include -ID:\work\build\Paddle\build\third_party\install\warpctc\include -ID:\work\build\Paddle\build\third_party\install\mklml\include -ID:\work\build\Paddle\build\third_party\install\mkldnn\include -ID:\work\build\Paddle\build\third_party\install\protobuf\include -IC:\Users\Gaotiantian\anaconda3\include -I"C:\Users\Gaotiantian\anaconda3\Lib\site-packages\numpy\core\include" -ID:\work\build\Paddle\build\third_party\pybind\src\extern_pybind\include -ID:\work\build\Paddle\build\third_party\cub\src\extern_cub -ID:\work\build\Paddle\build\third_party\install\cryptopp\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" -ID:\work\build\Paddle -ID:\work\build\Paddle\build\..\paddle\fluid\framework\io -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" --keep-dir x64\Release -maxrregcount=0 --machine 64 --compile -cudart static -Wno-deprecated-gpu-targets -w --expt-relaxed-constexpr --expt-extended-lambda -O3 -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj /arch:AVX" -D_WINDOWS -D_MWAITXINTRIN_H_INCLUDED -D__STRICT_ANSI__ -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -DWIN32 -D_WINDOWS -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -D_MBCS -Xcompiler "/EHsc /W0 /nologo /O2 /FS /MT /GR" -o sample_prob.dir\Release\/sample_prob.cu.obj "D:\work\build\Paddle\paddle\fluid\operators\math\sample_prob.cu" 49> nvcc fatal : A single input file is required for a non-link phase when an outputfile is specified 49>C:\Program Files (x86)\MSBuild\Microsoft.Cpp\v4.0\V140\BuildCustomizations\CUDA 10.2.targets(764,9): error MSB3721: 命令“"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\bin\nvcc.exe" -gencode=arch=compute_35,code=\"sm_35,compute_35\" -gencode=arch=compute_50,code=\"sm_50,compute_50\" -gencode=arch=compute_52,code=\"sm_52,compute_52\" -gencode=arch=compute_60,code=\"sm_60,compute_60\" -gencode=arch=compute_61,code=\"sm_61,compute_61\" -gencode=arch=compute_70,code=\"sm_70,compute_70\" -gencode=arch=compute_75,code=\"sm_75,compute_75\" --use-local-env 2015 -ccbin "D:\VS2015\VC\bin\x86_amd64" -x cu -ID:\work\build\Paddle\build -ID:\work\build\Paddle\paddle\fluid\framework\io -I"D:\3rdparty\TensorRT-7.0.0.11\include" -ID:\work\build\Paddle\build\third_party\install\zlib\include -ID:\work\build\Paddle\build\third_party\install -ID:\work\build\Paddle\build\third_party\install\gflags\include -ID:\work\build\Paddle\build\third_party\install\glog\include -ID:\work\build\Paddle\build\third_party\boost\src\extern_boost -ID:\work\build\Paddle\build\third_party\eigen3\src\extern_eigen3 -ID:\work\build\Paddle\build\third_party\threadpool\src\extern_threadpool -ID:\work\build\Paddle\build\third_party\dlpack\src\extern_dlpack\include -ID:\work\build\Paddle\build\third_party\install\xxhash\include -ID:\work\build\Paddle\build\third_party\install\warpctc\include -ID:\work\build\Paddle\build\third_party\install\mklml\include -ID:\work\build\Paddle\build\third_party\install\mkldnn\include -ID:\work\build\Paddle\build\third_party\install\protobuf\include -IC:\Users\Gaotiantian\anaconda3\include -I"C:\Users\Gaotiantian\anaconda3\Lib\site-packages\numpy\core\include" -ID:\work\build\Paddle\build\third_party\pybind\src\extern_pybind\include -ID:\work\build\Paddle\build\third_party\cub\src\extern_cub -ID:\work\build\Paddle\build\third_party\install\cryptopp\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" -ID:\work\build\Paddle -ID:\work\build\Paddle\build\..\paddle\fluid\framework\io -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" --keep-dir x64\Release -maxrregcount=0 --machine 64 --compile -cudart static -Wno-deprecated-gpu-targets -w --expt-relaxed-constexpr --expt-extended-lambda -O3 -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj /arch:AVX" -D_WINDOWS -D_MWAITXINTRIN_H_INCLUDED -D__STRICT_ANSI__ -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -DWIN32 -D_WINDOWS -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -D_MBCS -Xcompiler "/EHsc /W0 /nologo /O2 /FS /MT /GR" -o sample_prob.dir\Release\/sample_prob.cu.obj "D:\work\build\Paddle\paddle\fluid\operators\math\sample_prob.cu"”已退出，返回代码为 1。 43>C:\Program Files (x86)\MSBuild\Microsoft.Cpp\v4.0\V140\BuildCustomizations\CUDA 10.2.targets(764,9): error MSB3721: 命令“"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\bin\nvcc.exe" -gencode=arch=compute_35,code=\"sm_35,compute_35\" -gencode=arch=compute_50,code=\"sm_50,compute_50\" -gencode=arch=compute_52,code=\"sm_52,compute_52\" -gencode=arch=compute_60,code=\"sm_60,compute_60\" -gencode=arch=compute_61,code=\"sm_61,compute_61\" -gencode=arch=compute_70,code=\"sm_70,compute_70\" -gencode=arch=compute_75,code=\"sm_75,compute_75\" --use-local-env 2015 -ccbin "D:\VS2015\VC\bin\x86_amd64" -x cu -ID:\work\build\Paddle\build -ID:\work\build\Paddle\paddle\fluid\framework\io -I"D:\3rdparty\TensorRT-7.0.0.11\include" -ID:\work\build\Paddle\build\third_party\install\zlib\include -ID:\work\build\Paddle\build\third_party\install -ID:\work\build\Paddle\build\third_party\install\gflags\include -ID:\work\build\Paddle\build\third_party\install\glog\include -ID:\work\build\Paddle\build\third_party\boost\src\extern_boost -ID:\work\build\Paddle\build\third_party\eigen3\src\extern_eigen3 -ID:\work\build\Paddle\build\third_party\threadpool\src\extern_threadpool -ID:\work\build\Paddle\build\third_party\dlpack\src\extern_dlpack\include -ID:\work\build\Paddle\build\third_party\install\xxhash\include -ID:\work\build\Paddle\build\third_party\install\warpctc\include -ID:\work\build\Paddle\build\third_party\install\mklml\include -ID:\work\build\Paddle\build\third_party\install\mkldnn\include -ID:\work\build\Paddle\build\third_party\install\protobuf\include -IC:\Users\Gaotiantian\anaconda3\include -I"C:\Users\Gaotiantian\anaconda3\Lib\site-packages\numpy\core\include" -ID:\work\build\Paddle\build\third_party\pybind\src\extern_pybind\include -ID:\work\build\Paddle\build\third_party\cub\src\extern_cub -ID:\work\build\Paddle\build\third_party\install\cryptopp\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" -ID:\work\build\Paddle -ID:\work\build\Paddle\build\..\paddle\fluid\framework\io -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" --keep-dir x64\Release -maxrregcount=0 --machine 64 --compile -cudart static -Wno-deprecated-gpu-targets -w --expt-relaxed-constexpr --expt-extended-lambda -O3 -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj /arch:AVX" -D_WINDOWS -D_MWAITXINTRIN_H_INCLUDED -D__STRICT_ANSI__ -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -DWIN32 -D_WINDOWS -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -D_MBCS -Xcompiler "/EHsc /W0 /nologo /O2 /FS /MT /GR" -o pooling.dir\Release\/pooling.cu.obj "D:\work\build\Paddle\paddle\fluid\operators\math\pooling.cu"”已退出，返回代码为 1。 46> 46> D:\work\build\Paddle\build\paddle\fluid\operators\math>"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\bin\nvcc.exe" -gencode=arch=compute_35,code=\"sm_35,compute_35\" -gencode=arch=compute_50,code=\"sm_50,compute_50\" -gencode=arch=compute_52,code=\"sm_52,compute_52\" -gencode=arch=compute_60,code=\"sm_60,compute_60\" -gencode=arch=compute_61,code=\"sm_61,compute_61\" -gencode=arch=compute_70,code=\"sm_70,compute_70\" -gencode=arch=compute_75,code=\"sm_75,compute_75\" --use-local-env 2015 -ccbin "D:\VS2015\VC\bin\x86_amd64" -x cu -ID:\work\build\Paddle\build -ID:\work\build\Paddle\paddle\fluid\framework\io -I"D:\3rdparty\TensorRT-7.0.0.11\include" -ID:\work\build\Paddle\build\third_party\install\zlib\include -ID:\work\build\Paddle\build\third_party\install -ID:\work\build\Paddle\build\third_party\install\gflags\include -ID:\work\build\Paddle\build\third_party\install\glog\include -ID:\work\build\Paddle\build\third_party\boost\src\extern_boost -ID:\work\build\Paddle\build\third_party\eigen3\src\extern_eigen3 -ID:\work\build\Paddle\build\third_party\threadpool\src\extern_threadpool -ID:\work\build\Paddle\build\third_party\dlpack\src\extern_dlpack\include -ID:\work\build\Paddle\build\third_party\install\xxhash\include -ID:\work\build\Paddle\build\third_party\install\warpctc\include -ID:\work\build\Paddle\build\third_party\install\mklml\include -ID:\work\build\Paddle\build\third_party\install\mkldnn\include -ID:\work\build\Paddle\build\third_party\install\protobuf\include -IC:\Users\Gaotiantian\anaconda3\include -I"C:\Users\Gaotiantian\anaconda3\Lib\site-packages\numpy\core\include" -ID:\work\build\Paddle\build\third_party\pybind\src\extern_pybind\include -ID:\work\build\Paddle\build\third_party\cub\src\extern_cub -ID:\work\build\Paddle\build\third_party\install\cryptopp\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" -ID:\work\build\Paddle -ID:\work\build\Paddle\build\..\paddle\fluid\framework\io -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" --keep-dir x64\Release -maxrregcount=0 --machine 64 --compile -cudart static -Wno-deprecated-gpu-targets -w --expt-relaxed-constexpr --expt-extended-lambda -O3 -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj /arch:AVX" -D_WINDOWS -D_MWAITXINTRIN_H_INCLUDED -D__STRICT_ANSI__ -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -DWIN32 -D_WINDOWS -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -D_MBCS -Xcompiler "/EHsc /W0 /nologo /O2 /FS /MT /GR" -o bert_encoder_functor.dir\Release\bert_encoder_functor.obj "D:\work\build\Paddle\paddle\fluid\operators\math\bert_encoder_functor.cu" 46> nvcc fatal : A single input file is required for a non-link phase when an outputfile is specified 46>C:\Program Files (x86)\MSBuild\Microsoft.Cpp\v4.0\V140\BuildCustomizations\CUDA 10.2.targets(764,9): error MSB3721: 命令“"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\bin\nvcc.exe" -gencode=arch=compute_35,code=\"sm_35,compute_35\" -gencode=arch=compute_50,code=\"sm_50,compute_50\" -gencode=arch=compute_52,code=\"sm_52,compute_52\" -gencode=arch=compute_60,code=\"sm_60,compute_60\" -gencode=arch=compute_61,code=\"sm_61,compute_61\" -gencode=arch=compute_70,code=\"sm_70,compute_70\" -gencode=arch=compute_75,code=\"sm_75,compute_75\" --use-local-env 2015 -ccbin "D:\VS2015\VC\bin\x86_amd64" -x cu -ID:\work\build\Paddle\build -ID:\work\build\Paddle\paddle\fluid\framework\io -I"D:\3rdparty\TensorRT-7.0.0.11\include" -ID:\work\build\Paddle\build\third_party\install\zlib\include -ID:\work\build\Paddle\build\third_party\install -ID:\work\build\Paddle\build\third_party\install\gflags\include -ID:\work\build\Paddle\build\third_party\install\glog\include -ID:\work\build\Paddle\build\third_party\boost\src\extern_boost -ID:\work\build\Paddle\build\third_party\eigen3\src\extern_eigen3 -ID:\work\build\Paddle\build\third_party\threadpool\src\extern_threadpool -ID:\work\build\Paddle\build\third_party\dlpack\src\extern_dlpack\include -ID:\work\build\Paddle\build\third_party\install\xxhash\include -ID:\work\build\Paddle\build\third_party\install\warpctc\include -ID:\work\build\Paddle\build\third_party\install\mklml\include -ID:\work\build\Paddle\build\third_party\install\mkldnn\include -ID:\work\build\Paddle\build\third_party\install\protobuf\include -IC:\Users\Gaotiantian\anaconda3\include -I"C:\Users\Gaotiantian\anaconda3\Lib\site-packages\numpy\core\include" -ID:\work\build\Paddle\build\third_party\pybind\src\extern_pybind\include -ID:\work\build\Paddle\build\third_party\cub\src\extern_cub -ID:\work\build\Paddle\build\third_party\install\cryptopp\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" -ID:\work\build\Paddle -ID:\work\build\Paddle\build\..\paddle\fluid\framework\io -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\include" --keep-dir x64\Release -maxrregcount=0 --machine 64 --compile -cudart static -Wno-deprecated-gpu-targets -w --expt-relaxed-constexpr --expt-extended-lambda -O3 -Xcompiler="/EHsc /wd4244 /wd4267 /wd4819 /bigobj /arch:AVX" -D_WINDOWS -D_MWAITXINTRIN_H_INCLUDED -D__STRICT_ANSI__ -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -DWIN32 -D_WINDOWS -DNDEBUG -DTRT_PLUGIN_FP16_AVALIABLE -D"CUDA_VERSION_MAJOR=\"10\"" -D"CUDA_VERSION_MINOR=\"2\"" -D"CUDA_TOOLKIT_ROOT_DIR=\"C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.2\"" -D"CUDNN_MAJOR_VERSION=\"7\"" -DPADDLE_WITH_TENSORRT -DGOOGLE_GLOG_DLL_DECL= -DBOOST_HAS_STATIC_ASSERT -DEIGEN_STRONG_INLINE=inline -DPADDLE_WITH_MKLML -DLAPACK_FOUND -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_CRYPTO -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_DISABLE_PROFILER -DPADDLE_WITH_AVX -DPADDLE_WITH_SSE3 -D_XKEYCHECK_H -DPADDLE_DLL_INFERENCE -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_ON_INFERENCE -DPADDLE_DLL_EXPORT -D"CMAKE_INTDIR=\"Release\"" -D_MBCS -Xcompiler "/EHsc /W0 /nologo /O2 /FS /MT /GR" -o bert_encoder_functor.dir\Release\bert_encoder_functor.obj "D:\work\build\Paddle\paddle\fluid\operators\math\bert_encoder_functor.cu"”已退出，返回代码为 1。