Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

ninja 编译时出错 error: invalid combination of type specifiers #42440

Closed
brucmao opened this issue May 2, 2022 · 7 comments
Closed

ninja 编译时出错 error: invalid combination of type specifiers #42440

brucmao opened this issue May 2, 2022 · 7 comments
Assignees
@littletomatodonkey
Labels
status/following-up 跟进中 type/build 编译/安装问题

Comments

@brucmao
Copy link

brucmao commented May 2, 2022 •
edited
Loading

问题描述 Issue Description

[374/376] Building CXX object src\cpu\x64\CMakeFiles\dnnl_cpu_x64.dir\rnn\jit_gates_reduction.cpp.obj
[375/376] Building CXX object src\cpu\x64\CMakeFiles\dnnl_cpu_x64.dir\rnn\rnn_brgemm_utils.cpp.obj
[376/376] Linking CXX shared library src\dnnl.dll
[66/3632] Generate mkldnn.lib manually--->
已复制 1 个文件。
Microsoft (R) Library Manager Version 14.25.28614.0
Copyright (C) Microsoft Corporation. All rights reserved.

Creating library G:/workspace/Paddle/build/third_party/install/mkldnn/bin/mkldnn.lib and object G:/workspace/Paddle/build/third_party/install/mkldnn/bin/mkldnn.exp
[67/3632] Performing build step for 'extern_warpctc'
[1/4] create hidden file of ctc_entrypoint.cu
[2/4] Building NVCC (Device) object CMakeFiles/warpctc.dir/src/warpctc_generated_reduce.cu.obj
reduce.cu
nvcc warning : The 'compute_35', 'compute_37', 'sm_35', and 'sm_37' architectures are deprecated, and may be removed in a future release (Use -Wno-deprecated-gpu-targets to suppress warning).
nvcc warning : The 'compute_35', 'compute_37', 'sm_35', and 'sm_37' architectures are deprecated, and may be removed in a future release (Use -Wno-deprecated-gpu-targets to suppress warning).
reduce.cu
[3/4] Building NVCC (Device) object CMakeFiles/warpctc.dir/src/warpctc_generated_.ctc_entrypoint.cu.obj
.ctc_entrypoint.cu
nvcc warning : The 'compute_35', 'compute_37', 'sm_35', and 'sm_37' architectures are deprecated, and may be removed in a future release (Use -Wno-deprecated-gpu-targets to suppress warning).
nvcc warning : The 'compute_35', 'compute_37', 'sm_35', and 'sm_37' architectures are deprecated, and may be removed in a future release (Use -Wno-deprecated-gpu-targets to suppress warning).
.ctc_entrypoint.cu
[4/4] Linking CXX shared library warpctc.dll
[71/3632] Performing build step for 'extern_glog'
[1/9] Building CXX object CMakeFiles\glog.dir\src\windows\port.cc.obj
[2/9] Building CXX object CMakeFiles\glog.dir\src\raw_logging.cc.obj
[3/9] Building CXX object CMakeFiles\glog.dir\src\symbolize.cc.obj
[4/9] Building CXX object CMakeFiles\glog.dir\src\demangle.cc.obj
[5/9] Building CXX object CMakeFiles\glog.dir\src\utilities.cc.obj
[6/9] Building CXX object CMakeFiles\glog.dir\src\vlog_is_on.cc.obj
[7/9] Building CXX object CMakeFiles\glog.dir\src\signalhandler.cc.obj
[8/9] Building CXX object CMakeFiles\glog.dir\src\logging.cc.obj
[9/9] Linking CXX static library glog.lib
[97/3632] Building CUDA object paddle\phi\kernels\funcs\eigen\CMakeFiles\eigen_function.dir\constant.cu.obj
constant.cu
[99/3632] Building CUDA object paddle\phi\kernels\funcs\eigen\CMakeFiles\eigen_function.dir\erf.cu.obj
erf.cu
[100/3632] Building CUDA object paddle\phi\kernels\funcs\eigen\CMakeFiles\eigen_function.dir\elementwise.cu.obj
elementwise.cu
[109/3632] Building CUDA object paddle\phi\kernels\funcs\eigen\CMakeFiles\eigen_function.dir\loss.cu.obj
loss.cu
[113/3632] Building CUDA object paddle\phi\kernels\funcs\eigen\CMakeFiles\eigen_function.dir\l1_norm.cu.obj
l1_norm.cu
[126/3632] Building CUDA object paddle\phi\kernels\funcs\eigen\CMakeFiles\eigen_function.dir\scale.cu.obj
scale.cu
[130/3632] Building CUDA object paddle\phi\kernels\funcs\eigen\CMakeFiles\eigen_function.dir\sign.cu.obj
sign.cu
[149/3632] Building CUDA object paddle\phi\kernels\funcs\eigen\CMakeFiles\eigen_function.dir\reverse.cu.obj
reverse.cu
[202/3632] Building CUDA object paddle\phi\kernels\funcs\eigen\CMakeFiles\eigen_function.dir\broadcast.cu.obj
broadcast.cu
[226/3632] Building CUDA object paddle\phi\kernels\funcs\eigen\CMakeFiles\eigen_function.dir\pad.cu.obj
pad.cu
[244/3632] Building CUDA object paddle\phi\kernels\funcs\eigen\CMakeFiles\eigen_function.dir\slice.cu.obj
slice.cu
[512/3632] Performing download step (git clone) for 'extern_xxhash'
Cloning into 'extern_xxhash'...
HEAD is now at 7cc9639 Merge pull request #136 from Cyan4973/dev
[515/3632] Building CUDA object paddle\fluid\platform\CMakeFiles\profiler.dir\profiler.cu.obj
profiler.cu
[519/3632] Performing build step for 'extern_xxhash'
[1/4] Building C object CMakeFiles\xxhash.dir\G_\workspace\Paddle\build\third_party\xxhash\src\extern_xxhash\xxhash.c.obj
[2/4] Building C object CMakeFiles\xxhsum.dir\G_\workspace\Paddle\build\third_party\xxhash\src\extern_xxhash\xxhsum.c.obj
G:\workspace\Paddle\build\third_party\xxhash\src\extern_xxhash\xxhsum.c(743) : warning C5045: Compiler will insert Spectre mitigation for memory load if /Qspectre switch specified
G:\workspace\Paddle\build\third_party\xxhash\src\extern_xxhash\xxhsum.c(741) : note: index 'newBufSize' range checked by comparison on this line
G:\workspace\Paddle\build\third_party\xxhash\src\extern_xxhash\xxhsum.c(743) : note: feeds call on this line
G:\workspace\Paddle\build\third_party\xxhash\src\extern_xxhash\xxhsum.c(296) : warning C5045: Compiler will insert Spectre mitigation for memory load if /Qspectre switch specified
G:\workspace\Paddle\build\third_party\xxhash\src\extern_xxhash\xxhsum.c(295) : note: index 'benchedSize' range checked by comparison on this line
G:\workspace\Paddle\build\third_party\xxhash\src\extern_xxhash\xxhsum.c(296) : note: feeds call on this line
G:\workspace\Paddle\build\third_party\xxhash\src\extern_xxhash\xxhsum.c(990) : warning C5045: Compiler will insert Spectre mitigation for memory load if /Qspectre switch specified
G:\workspace\Paddle\build\third_party\xxhash\src\extern_xxhash\xxhsum.c(964) : note: index '' range checked by comparison on this line
G:\workspace\Paddle\build\third_party\xxhash\src\extern_xxhash\xxhsum.c(990) : note: feeds call on this line
G:\workspace\Paddle\build\third_party\xxhash\src\extern_xxhash\xxhsum.c(640) : warning C5045: Compiler will insert Spectre mitigation for memory load if /Qspectre switch specified
G:\workspace\Paddle\build\third_party\xxhash\src\extern_xxhash\xxhsum.c(639) : note: index 'fnTotal' range checked by comparison on this line
G:\workspace\Paddle\build\third_party\xxhash\src\extern_xxhash\xxhsum.c(1297) : note: feeds call on this line
[3/4] Linking C static library xxhash.lib
[4/4] Linking C executable xxhsum.exe
[561/3632] Building CUDA object paddle\phi\kernels\CMakeFiles\determinant_grad_kernel.dir\gpu\determinant_grad_kernel.cu.obj
determinant_grad_kernel.cu
[567/3632] Building CUDA object paddle\phi\kernels\CMakeFiles\full_kernel_base.dir\gpu\full_kernel.cu.obj
full_kernel.cu
[568/3632] Building CUDA object paddle\phi\kernels\CMakeFiles\conv_kernel_base.dir\gpu\conv_kernel.cu.obj
conv_kernel.cu
[570/3632] Building CUDA object paddle\phi\kernels\CMakeFiles\conv_kernel_gpudnn.dir\gpudnn\conv_kernel.cu.obj
conv_kernel.cu
[572/3632] Building CUDA object paddle\phi\kernels\CMakeFiles\conv_grad_kernel_base.dir\gpu\conv_grad_kernel.cu.obj
conv_grad_kernel.cu
[573/3632] Building CUDA object paddle\phi\kernels\CMakeFiles\conv_grad_kernel_gpudnn.dir\gpudnn\conv_grad_kernel.cu.obj
conv_grad_kernel.cu
[575/3632] Building CUDA object paddle\phi\kernels\CMakeFiles\conv_grad_grad_kernel_base.dir\gpu\conv_grad_grad_kernel.cu.obj
conv_grad_grad_kernel.cu
[576/3632] Building CUDA object paddle\phi\kernels\CMakeFiles\conv_grad_grad_kernel_gpudnn.dir\gpudnn\conv_grad_grad_kernel.cu.obj
conv_grad_grad_kernel.cu
[578/3632] Building CUDA object paddle\phi\kernels\CMakeFiles\conv_transpose_kernel_gpudnn.dir\gpudnn\conv_transpose_kernel.cu.obj
conv_transpose_kernel.cu
[583/3632] Building CUDA object paddle\phi\kernels\CMakeFiles\conv_transpose_grad_kernel_gpudnn.dir\gpudnn\conv_transpose_grad_kernel.cu.obj
conv_transpose_grad_kernel.cu
[585/3632] Building CUDA object paddle\phi\kernels\CMakeFiles\conv_transpose_kernel_base.dir\gpu\conv_transpose_kernel.cu.obj
FAILED: paddle/phi/kernels/CMakeFiles/conv_transpose_kernel_base.dir/gpu/conv_transpose_kernel.cu.obj
"C:\PROGRA~1\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\nvcc.exe" -forward-unknown-to-host-compiler -DBOOST_HAS_STATIC_ASSERT -DCUDA_TOOLKIT_ROOT_DIR=""C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6"" -DCUDA_VERSION_MAJOR="11" -DCUDA_VERSION_MINOR="6" -DCUDNN_MAJOR_VERSION="8" -DEIGEN_STRONG_INLINE=inline -DEIGEN_USE_GPU -DGOOGLE_GLOG_DLL_DECL="" -DLAPACK_FOUND -DNOMINMAX -DPADDLE_DISABLE_PROFILER -DPADDLE_DLL_EXPORT -DPADDLE_DLL_INFERENCE -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_WITH_AVX -DPADDLE_WITH_CRYPTO -DPADDLE_WITH_CUDA -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_MKLML -DPADDLE_WITH_POCKETFFT -DPADDLE_WITH_SSE3 -DTRT_PLUGIN_FP16_AVALIABLE -DUTF8PROC_STATIC -D_XKEYCHECK_H -IG:\workspace\Paddle\build -IG:\workspace\Paddle\paddle\fluid\framework\io -IG:\workspace\Paddle\build\third_party\install\zlib\include -IG:\workspace\Paddle\build\third_party\install -IG:\workspace\Paddle\build\third_party\install\gflags\include -IG:\workspace\Paddle\build\third_party\install\glog\include -IG:\workspace\Paddle\build\third_party\boost\src\extern_boost -IG:\workspace\Paddle\build\third_party\eigen3\src\extern_eigen3 -IG:\workspace\Paddle\build\third_party\threadpool\src\extern_threadpool -IG:\workspace\Paddle\build\third_party\dlpack\src\extern_dlpack\include -IG:\workspace\Paddle\build\third_party\install\xxhash\include -IG:\workspace\Paddle\build\third_party\install\warpctc\include -IG:\workspace\Paddle\build\third_party\install\utf8proc\include -IG:\workspace\Paddle\build\third_party\install\mklml\include -IG:\workspace\Paddle\build\third_party\install\mkldnn\include -IG:\workspace\Paddle\build\third_party\install\protobuf\include -IC:\Users\brucm\scoop\apps\python\3.10.4\include -IC:\Users\brucm\scoop\apps\python\current\Lib\site-packages\numpy\core\include -IG:\workspace\Paddle\build\third_party\pybind\src\extern_pybind\include -IG:\workspace\Paddle\build\third_party\install\cryptopp\include -IG:\workspace\Paddle\build\third_party\pocketfft\src -IG:\workspace\Paddle\build\third_party\dirent\src\extern_dirent\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include" -IG:\workspace\Paddle -D_WINDOWS -Xcompiler=" /W0 /GR /EHsc" -D_MWAITXINTRIN_H_INCLUDED -D__STRICT_ANSI__ -Wno-deprecated-gpu-targets -gencode arch=compute_61,code=sm_61 -w --expt-relaxed-constexpr --expt-extended-lambda -Xcompiler "/wd4244 /wd4267 /wd4819 " -Xcompiler /bigobj -Xcompiler="/arch:AVX" -Xcompiler="-MT -O2 -Ob2" -DNDEBUG -std=c++14 -MD -MT paddle\phi\kernels\CMakeFiles\conv_transpose_kernel_base.dir\gpu\conv_transpose_kernel.cu.obj -MF paddle\phi\kernels\CMakeFiles\conv_transpose_kernel_base.dir\gpu\conv_transpose_kernel.cu.obj.d -x cu -c G:\workspace\Paddle\paddle\phi\kernels\gpu\conv_transpose_kernel.cu -o paddle\phi\kernels\CMakeFiles\conv_transpose_kernel_base.dir\gpu\conv_transpose_kernel.cu.obj -Xcompiler=-Fdpaddle\phi\kernels\CMakeFiles\conv_transpose_kernel_base.dir\conv_transpose_kernel_base.pdb,-FS
C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\cub/device/dispatch/dispatch_segmented_sort.cuh(338): error: invalid combination of type specifiers

C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\cub/device/dispatch/dispatch_segmented_sort.cuh(338): error: expected an identifier

C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\cub/device/dispatch/dispatch_segmented_sort.cuh(379): error: expected a member name

3 errors detected in the compilation of "G:/workspace/Paddle/paddle/phi/kernels/gpu/conv_transpose_kernel.cu".
conv_transpose_kernel.cu
[587/3632] Building CUDA object paddle\phi\kernels\CMakeFiles\conv_transpose_grad_kernel_base.dir\gpu\conv_transpose_grad_kernel.cu.obj
FAILED: paddle/phi/kernels/CMakeFiles/conv_transpose_grad_kernel_base.dir/gpu/conv_transpose_grad_kernel.cu.obj
"C:\PROGRA~1\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\nvcc.exe" -forward-unknown-to-host-compiler -DBOOST_HAS_STATIC_ASSERT -DCUDA_TOOLKIT_ROOT_DIR=""C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6"" -DCUDA_VERSION_MAJOR="11" -DCUDA_VERSION_MINOR="6" -DCUDNN_MAJOR_VERSION="8" -DEIGEN_STRONG_INLINE=inline -DEIGEN_USE_GPU -DGOOGLE_GLOG_DLL_DECL="" -DLAPACK_FOUND -DNOMINMAX -DPADDLE_DISABLE_PROFILER -DPADDLE_DLL_EXPORT -DPADDLE_DLL_INFERENCE -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_WITH_AVX -DPADDLE_WITH_CRYPTO -DPADDLE_WITH_CUDA -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_MKLML -DPADDLE_WITH_POCKETFFT -DPADDLE_WITH_SSE3 -DTRT_PLUGIN_FP16_AVALIABLE -DUTF8PROC_STATIC -D_XKEYCHECK_H -IG:\workspace\Paddle\build -IG:\workspace\Paddle\paddle\fluid\framework\io -IG:\workspace\Paddle\build\third_party\install\zlib\include -IG:\workspace\Paddle\build\third_party\install -IG:\workspace\Paddle\build\third_party\install\gflags\include -IG:\workspace\Paddle\build\third_party\install\glog\include -IG:\workspace\Paddle\build\third_party\boost\src\extern_boost -IG:\workspace\Paddle\build\third_party\eigen3\src\extern_eigen3 -IG:\workspace\Paddle\build\third_party\threadpool\src\extern_threadpool -IG:\workspace\Paddle\build\third_party\dlpack\src\extern_dlpack\include -IG:\workspace\Paddle\build\third_party\install\xxhash\include -IG:\workspace\Paddle\build\third_party\install\warpctc\include -IG:\workspace\Paddle\build\third_party\install\utf8proc\include -IG:\workspace\Paddle\build\third_party\install\mklml\include -IG:\workspace\Paddle\build\third_party\install\mkldnn\include -IG:\workspace\Paddle\build\third_party\install\protobuf\include -IC:\Users\brucm\scoop\apps\python\3.10.4\include -IC:\Users\brucm\scoop\apps\python\current\Lib\site-packages\numpy\core\include -IG:\workspace\Paddle\build\third_party\pybind\src\extern_pybind\include -IG:\workspace\Paddle\build\third_party\install\cryptopp\include -IG:\workspace\Paddle\build\third_party\pocketfft\src -IG:\workspace\Paddle\build\third_party\dirent\src\extern_dirent\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include" -IG:\workspace\Paddle -D_WINDOWS -Xcompiler=" /W0 /GR /EHsc" -D_MWAITXINTRIN_H_INCLUDED -D__STRICT_ANSI__ -Wno-deprecated-gpu-targets -gencode arch=compute_61,code=sm_61 -w --expt-relaxed-constexpr --expt-extended-lambda -Xcompiler "/wd4244 /wd4267 /wd4819 " -Xcompiler /bigobj -Xcompiler="/arch:AVX" -Xcompiler="-MT -O2 -Ob2" -DNDEBUG -std=c++14 -MD -MT paddle\phi\kernels\CMakeFiles\conv_transpose_grad_kernel_base.dir\gpu\conv_transpose_grad_kernel.cu.obj -MF paddle\phi\kernels\CMakeFiles\conv_transpose_grad_kernel_base.dir\gpu\conv_transpose_grad_kernel.cu.obj.d -x cu -c G:\workspace\Paddle\paddle\phi\kernels\gpu\conv_transpose_grad_kernel.cu -o paddle\phi\kernels\CMakeFiles\conv_transpose_grad_kernel_base.dir\gpu\conv_transpose_grad_kernel.cu.obj -Xcompiler=-Fdpaddle\phi\kernels\CMakeFiles\conv_transpose_grad_kernel_base.dir\conv_transpose_grad_kernel_base.pdb,-FS
C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\cub/device/dispatch/dispatch_segmented_sort.cuh(338): error: invalid combination of type specifiers

C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\cub/device/dispatch/dispatch_segmented_sort.cuh(338): error: expected an identifier

C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\cub/device/dispatch/dispatch_segmented_sort.cuh(379): error: expected a member name

3 errors detected in the compilation of "G:/workspace/Paddle/paddle/phi/kernels/gpu/conv_transpose_grad_kernel.cu".
conv_transpose_grad_kernel.cu
[588/3632] Building CUDA object paddle\phi\kernels\CMakeFiles\eigh_kernel.dir\gpu\eigh_kernel.cu.obj
eigh_kernel.cu
[589/3632] Building CUDA object paddle\phi\kernels\CMakeFiles\deformable_conv_kernel.dir\gpu\deformable_conv_kernel.cu.obj
deformable_conv_kernel.cu
[590/3632] Building CUDA object paddle\phi\kernels\CMakeFiles\deformable_conv_grad_kernel.dir\gpu\deformable_conv_grad_kernel.cu.obj
deformable_conv_grad_kernel.cu
[591/3632] Building CUDA object paddle\phi\kernels\CMakeFiles\gumbel_softmax_grad_kernel.dir\gpu\gumbel_softmax_grad_kernel.cu.obj
gumbel_softmax_grad_kernel.cu
[592/3632] Building CUDA object paddle\phi\kernels\CMakeFiles\gumbel_softmax_kernel.dir\gpu\gumbel_softmax_kernel.cu.obj
FAILED: paddle/phi/kernels/CMakeFiles/gumbel_softmax_kernel.dir/gpu/gumbel_softmax_kernel.cu.obj
"C:\PROGRA~1\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\nvcc.exe" -forward-unknown-to-host-compiler -DBOOST_HAS_STATIC_ASSERT -DCUDA_TOOLKIT_ROOT_DIR=""C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6"" -DCUDA_VERSION_MAJOR="11" -DCUDA_VERSION_MINOR="6" -DCUDNN_MAJOR_VERSION="8" -DEIGEN_STRONG_INLINE=inline -DEIGEN_USE_GPU -DGOOGLE_GLOG_DLL_DECL="" -DLAPACK_FOUND -DNOMINMAX -DPADDLE_DISABLE_PROFILER -DPADDLE_DLL_EXPORT -DPADDLE_DLL_INFERENCE -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_WITH_AVX -DPADDLE_WITH_CRYPTO -DPADDLE_WITH_CUDA -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_MKLML -DPADDLE_WITH_POCKETFFT -DPADDLE_WITH_SSE3 -DTRT_PLUGIN_FP16_AVALIABLE -DUTF8PROC_STATIC -D_XKEYCHECK_H -IG:\workspace\Paddle\build -IG:\workspace\Paddle\paddle\fluid\framework\io -IG:\workspace\Paddle\build\third_party\install\zlib\include -IG:\workspace\Paddle\build\third_party\install -IG:\workspace\Paddle\build\third_party\install\gflags\include -IG:\workspace\Paddle\build\third_party\install\glog\include -IG:\workspace\Paddle\build\third_party\boost\src\extern_boost -IG:\workspace\Paddle\build\third_party\eigen3\src\extern_eigen3 -IG:\workspace\Paddle\build\third_party\threadpool\src\extern_threadpool -IG:\workspace\Paddle\build\third_party\dlpack\src\extern_dlpack\include -IG:\workspace\Paddle\build\third_party\install\xxhash\include -IG:\workspace\Paddle\build\third_party\install\warpctc\include -IG:\workspace\Paddle\build\third_party\install\utf8proc\include -IG:\workspace\Paddle\build\third_party\install\mklml\include -IG:\workspace\Paddle\build\third_party\install\mkldnn\include -IG:\workspace\Paddle\build\third_party\install\protobuf\include -IC:\Users\brucm\scoop\apps\python\3.10.4\include -IC:\Users\brucm\scoop\apps\python\current\Lib\site-packages\numpy\core\include -IG:\workspace\Paddle\build\third_party\pybind\src\extern_pybind\include -IG:\workspace\Paddle\build\third_party\install\cryptopp\include -IG:\workspace\Paddle\build\third_party\pocketfft\src -IG:\workspace\Paddle\build\third_party\dirent\src\extern_dirent\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include" -IG:\workspace\Paddle -D_WINDOWS -Xcompiler=" /W0 /GR /EHsc" -D_MWAITXINTRIN_H_INCLUDED -D__STRICT_ANSI__ -Wno-deprecated-gpu-targets -gencode arch=compute_61,code=sm_61 -w --expt-relaxed-constexpr --expt-extended-lambda -Xcompiler "/wd4244 /wd4267 /wd4819 " -Xcompiler /bigobj -Xcompiler="/arch:AVX" -Xcompiler="-MT -O2 -Ob2" -DNDEBUG -std=c++14 -MD -MT paddle\phi\kernels\CMakeFiles\gumbel_softmax_kernel.dir\gpu\gumbel_softmax_kernel.cu.obj -MF paddle\phi\kernels\CMakeFiles\gumbel_softmax_kernel.dir\gpu\gumbel_softmax_kernel.cu.obj.d -x cu -c G:\workspace\Paddle\paddle\phi\kernels\gpu\gumbel_softmax_kernel.cu -o paddle\phi\kernels\CMakeFiles\gumbel_softmax_kernel.dir\gpu\gumbel_softmax_kernel.cu.obj -Xcompiler=-Fdpaddle\phi\kernels\CMakeFiles\gumbel_softmax_kernel.dir\gumbel_softmax_kernel.pdb,-FS
C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\cub/device/dispatch/dispatch_segmented_sort.cuh(338): error: invalid combination of type specifiers

C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\cub/device/dispatch/dispatch_segmented_sort.cuh(338): error: expected an identifier

C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\cub/device/dispatch/dispatch_segmented_sort.cuh(379): error: expected a member name

3 errors detected in the compilation of "G:/workspace/Paddle/paddle/phi/kernels/gpu/gumbel_softmax_kernel.cu".
gumbel_softmax_kernel.cu
[593/3632] Building CUDA object paddle\phi\kernels\CMakeFiles\cross_entropy_kernel.dir\gpu\cross_entropy_kernel.cu.obj
FAILED: paddle/phi/kernels/CMakeFiles/cross_entropy_kernel.dir/gpu/cross_entropy_kernel.cu.obj
"C:\PROGRA~1\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\nvcc.exe" -forward-unknown-to-host-compiler -DBOOST_HAS_STATIC_ASSERT -DCUDA_TOOLKIT_ROOT_DIR=""C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6"" -DCUDA_VERSION_MAJOR="11" -DCUDA_VERSION_MINOR="6" -DCUDNN_MAJOR_VERSION="8" -DEIGEN_STRONG_INLINE=inline -DEIGEN_USE_GPU -DGOOGLE_GLOG_DLL_DECL="" -DLAPACK_FOUND -DNOMINMAX -DPADDLE_DISABLE_PROFILER -DPADDLE_DLL_EXPORT -DPADDLE_DLL_INFERENCE -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_WITH_AVX -DPADDLE_WITH_CRYPTO -DPADDLE_WITH_CUDA -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_MKLML -DPADDLE_WITH_POCKETFFT -DPADDLE_WITH_SSE3 -DTRT_PLUGIN_FP16_AVALIABLE -DUTF8PROC_STATIC -D_XKEYCHECK_H -IG:\workspace\Paddle\build -IG:\workspace\Paddle\paddle\fluid\framework\io -IG:\workspace\Paddle\build\third_party\install\zlib\include -IG:\workspace\Paddle\build\third_party\install -IG:\workspace\Paddle\build\third_party\install\gflags\include -IG:\workspace\Paddle\build\third_party\install\glog\include -IG:\workspace\Paddle\build\third_party\boost\src\extern_boost -IG:\workspace\Paddle\build\third_party\eigen3\src\extern_eigen3 -IG:\workspace\Paddle\build\third_party\threadpool\src\extern_threadpool -IG:\workspace\Paddle\build\third_party\dlpack\src\extern_dlpack\include -IG:\workspace\Paddle\build\third_party\install\xxhash\include -IG:\workspace\Paddle\build\third_party\install\warpctc\include -IG:\workspace\Paddle\build\third_party\install\utf8proc\include -IG:\workspace\Paddle\build\third_party\install\mklml\include -IG:\workspace\Paddle\build\third_party\install\mkldnn\include -IG:\workspace\Paddle\build\third_party\install\protobuf\include -IC:\Users\brucm\scoop\apps\python\3.10.4\include -IC:\Users\brucm\scoop\apps\python\current\Lib\site-packages\numpy\core\include -IG:\workspace\Paddle\build\third_party\pybind\src\extern_pybind\include -IG:\workspace\Paddle\build\third_party\install\cryptopp\include -IG:\workspace\Paddle\build\third_party\pocketfft\src -IG:\workspace\Paddle\build\third_party\dirent\src\extern_dirent\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include" -IG:\workspace\Paddle -D_WINDOWS -Xcompiler=" /W0 /GR /EHsc" -D_MWAITXINTRIN_H_INCLUDED -D__STRICT_ANSI__ -Wno-deprecated-gpu-targets -gencode arch=compute_61,code=sm_61 -w --expt-relaxed-constexpr --expt-extended-lambda -Xcompiler "/wd4244 /wd4267 /wd4819 " -Xcompiler /bigobj -Xcompiler="/arch:AVX" -Xcompiler="-MT -O2 -Ob2" -DNDEBUG -std=c++14 -MD -MT paddle\phi\kernels\CMakeFiles\cross_entropy_kernel.dir\gpu\cross_entropy_kernel.cu.obj -MF paddle\phi\kernels\CMakeFiles\cross_entropy_kernel.dir\gpu\cross_entropy_kernel.cu.obj.d -x cu -c G:\workspace\Paddle\paddle\phi\kernels\gpu\cross_entropy_kernel.cu -o paddle\phi\kernels\CMakeFiles\cross_entropy_kernel.dir\gpu\cross_entropy_kernel.cu.obj -Xcompiler=-Fdpaddle\phi\kernels\CMakeFiles\cross_entropy_kernel.dir\cross_entropy_kernel.pdb,-FS
C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\cub/device/dispatch/dispatch_segmented_sort.cuh(338): error: invalid combination of type specifiers

C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\cub/device/dispatch/dispatch_segmented_sort.cuh(338): error: expected an identifier

C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\cub/device/dispatch/dispatch_segmented_sort.cuh(379): error: expected a member name

3 errors detected in the compilation of "G:/workspace/Paddle/paddle/phi/kernels/gpu/cross_entropy_kernel.cu".
cross_entropy_kernel.cu
[594/3632] Building CUDA object paddle\phi\kernels\CMakeFiles\reduce_sum_kernel_base.dir\gpu\reduce_sum_kernel.cu.obj
FAILED: paddle/phi/kernels/CMakeFiles/reduce_sum_kernel_base.dir/gpu/reduce_sum_kernel.cu.obj
"C:\PROGRA~1\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\nvcc.exe" -forward-unknown-to-host-compiler -DBOOST_HAS_STATIC_ASSERT -DCUDA_TOOLKIT_ROOT_DIR=""C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6"" -DCUDA_VERSION_MAJOR="11" -DCUDA_VERSION_MINOR="6" -DCUDNN_MAJOR_VERSION="8" -DEIGEN_STRONG_INLINE=inline -DEIGEN_USE_GPU -DGOOGLE_GLOG_DLL_DECL="" -DLAPACK_FOUND -DNOMINMAX -DPADDLE_DISABLE_PROFILER -DPADDLE_DLL_EXPORT -DPADDLE_DLL_INFERENCE -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_WITH_AVX -DPADDLE_WITH_CRYPTO -DPADDLE_WITH_CUDA -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_MKLML -DPADDLE_WITH_POCKETFFT -DPADDLE_WITH_SSE3 -DTRT_PLUGIN_FP16_AVALIABLE -DUTF8PROC_STATIC -D_XKEYCHECK_H -IG:\workspace\Paddle\build -IG:\workspace\Paddle\paddle\fluid\framework\io -IG:\workspace\Paddle\build\third_party\install\zlib\include -IG:\workspace\Paddle\build\third_party\install -IG:\workspace\Paddle\build\third_party\install\gflags\include -IG:\workspace\Paddle\build\third_party\install\glog\include -IG:\workspace\Paddle\build\third_party\boost\src\extern_boost -IG:\workspace\Paddle\build\third_party\eigen3\src\extern_eigen3 -IG:\workspace\Paddle\build\third_party\threadpool\src\extern_threadpool -IG:\workspace\Paddle\build\third_party\dlpack\src\extern_dlpack\include -IG:\workspace\Paddle\build\third_party\install\xxhash\include -IG:\workspace\Paddle\build\third_party\install\warpctc\include -IG:\workspace\Paddle\build\third_party\install\utf8proc\include -IG:\workspace\Paddle\build\third_party\install\mklml\include -IG:\workspace\Paddle\build\third_party\install\mkldnn\include -IG:\workspace\Paddle\build\third_party\install\protobuf\include -IC:\Users\brucm\scoop\apps\python\3.10.4\include -IC:\Users\brucm\scoop\apps\python\current\Lib\site-packages\numpy\core\include -IG:\workspace\Paddle\build\third_party\pybind\src\extern_pybind\include -IG:\workspace\Paddle\build\third_party\install\cryptopp\include -IG:\workspace\Paddle\build\third_party\pocketfft\src -IG:\workspace\Paddle\build\third_party\dirent\src\extern_dirent\include -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include" -IG:\workspace\Paddle -D_WINDOWS -Xcompiler=" /W0 /GR /EHsc" -D_MWAITXINTRIN_H_INCLUDED -D__STRICT_ANSI__ -Wno-deprecated-gpu-targets -gencode arch=compute_61,code=sm_61 -w --expt-relaxed-constexpr --expt-extended-lambda -Xcompiler "/wd4244 /wd4267 /wd4819 " -Xcompiler /bigobj -Xcompiler="/arch:AVX" -Xcompiler="-MT -O2 -Ob2" -DNDEBUG -std=c++14 -MD -MT paddle\phi\kernels\CMakeFiles\reduce_sum_kernel_base.dir\gpu\reduce_sum_kernel.cu.obj -MF paddle\phi\kernels\CMakeFiles\reduce_sum_kernel_base.dir\gpu\reduce_sum_kernel.cu.obj.d -x cu -c G:\workspace\Paddle\paddle\phi\kernels\gpu\reduce_sum_kernel.cu -o paddle\phi\kernels\CMakeFiles\reduce_sum_kernel_base.dir\gpu\reduce_sum_kernel.cu.obj -Xcompiler=-Fdpaddle\phi\kernels\CMakeFiles\reduce_sum_kernel_base.dir\reduce_sum_kernel_base.pdb,-FS
C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\cub/device/dispatch/dispatch_segmented_sort.cuh(338): error: invalid combination of type specifiers

C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\cub/device/dispatch/dispatch_segmented_sort.cuh(338): error: expected an identifier

C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\cub/device/dispatch/dispatch_segmented_sort.cuh(379): error: expected a member name

3 errors detected in the compilation of "G:/workspace/Paddle/paddle/phi/kernels/gpu/reduce_sum_kernel.cu".
reduce_sum_kernel.cu
ninja: build stopped: subcommand failed.

版本&环境信息 Version & Environment Information

版本&环境信息

Paddle version: None
Paddle With CUDA: None

OS: Windows 10
Python version: 3.10.4

CUDA version: 11.6.124
Build cuda_11.6.r11.6/compiler.31057947_0
cuDNN version: None.None.None
Nvidia driver version: 512.15

1.PaddlePaddle 版本 2.3rc
2. 处理器名称 QuadCore Intel Core i7-7700HQ, 3600 MHz (36 x 100) ； 指令集 x86, x86-64, MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2, AVX, AVX2, FMA, AES
3. 显示适配器 nVIDIA GeForce GTX 1060 with Max-Q Design (Dell);CUDA 11.6 ;cuDNN v8.4.0
4. win10 10.0.19044.1682
5. Python 3.10.4
6. Visual Studio 2019 Developer Command Prompt v16.5.4
@brucmao brucmao added status/new-issue 新建 type/build 编译/安装问题 labels May 2, 2022
@paddle-bot-old
Copy link

paddle-bot-old bot commented May 2, 2022

您好，我们已经收到了您的问题，会安排技术人员尽快解答您的问题，请耐心等待。请您再次检查是否提供了清晰的问题描述、复现代码、环境&版本、报错信息等。同时，您也可以通过查看官网API文档、常见问题、历史Issue、AI社区来寻求解答。祝您生活愉快～

Hi! We've received your issue and please be patient to get responded. We will arrange technicians to answer your questions as soon as possible. Please make sure that you have posted enough message to demo your request. You may also check out the API，FAQ，Github Issue and AI community to get the answer.Have a nice day!

@littletomatodonkey
Copy link
Contributor

你好，编译的话，暂时建议vs2017，https://www.paddlepaddle.org.cn/documentation/docs/zh/install/compile/windows-compile.html#benjibianyiguocheng

此外，如果不是自己开发的新op，如果希望体验最新的功能，可以从官网上pip install develop的whl包哈

@brucmao
Copy link
Author

brucmao commented May 3, 2022

pip 只支持cuda11.2的版本，我CUDA 11.6.124 没法用pip安装

@littletomatodonkey
Copy link
Contributor

目前最高只支持到11.2

@brucmao
Copy link
Author

brucmao commented May 4, 2022 •
edited
Loading

换了 vs2017 ninja编译还是报错

[713/2287] Building CXX object paddle\fluid\operators\CMakeFiles\cholesky_op.dir\cholesky_op.cc.obj
Unknown compiler version - please run the configure tests and report the results
C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\thrust/detail/config/cpp_dialect.h:118: warning: Thrust requires at least MSVC 2019 (19.20/16.0/14.20). MSVC 2017 is deprecated but still supported. MSVC 2017 support will be removed in a future release. Define THRUST_IGNORE_DEPRECATED_CPP_DIALECT to suppress this message.
[714/2287] Building CXX object paddle\fluid\operators\CMakeFiles\clip_by_norm_op.dir\clip_by_norm_op.cc.obj
C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\thrust/detail/config/cpp_dialect.h:118: warning: Thrust requires at least MSVC 2019 (19.20/16.0/14.20). MSVC 2017 is deprecated but still supported. MSVC 2017 support will be removed in a future release. Define THRUST_IGNORE_DEPRECATED_CPP_DIALECT to suppress this message.
Unknown compiler version - please run the configure tests and report the results
[715/2287] Building CUDA object paddle\fluid\operators\CMakeFiles\cast_op.dir\cast_op.cu.obj
C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\thrust/detail/config/cpp_dialect.h:118: warning: Thrust requires at least MSVC 2019 (19.20/16.0/14.20). MSVC 2017 is deprecated but still supported. MSVC 2017 support will be removed in a future release. Define THRUST_IGNORE_DEPRECATED_CPP_DIALECT to suppress this message.
Unknown compiler version - please run the configure tests and report the results
c:\program files\nvidia gpu computing toolkit\cuda\v11.6\include\cub\util_cpp_dialect.cuh:129: warning: CUB requires at least MSVC 2019 (19.20/16.0/14.20). MSVC 2017 is deprecated but still supported. MSVC 2017 support will be removed in a future release. Define CUB_IGNORE_DEPRECATED_CPP_DIALECT to suppress this message.
cast_op.cu
C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\thrust/detail/config/cpp_dialect.h:118: warning: Thrust requires at least MSVC 2019 (19.20/16.0/14.20). MSVC 2017 is deprecated but still supported. MSVC 2017 support will be removed in a future release. Define THRUST_IGNORE_DEPRECATED_CPP_DIALECT to suppress this message.
Unknown compiler version - please run the configure tests and report the results
c:\program files\nvidia gpu computing toolkit\cuda\v11.6\include\cub\util_cpp_dialect.cuh:129: warning: CUB requires at least MSVC 2019 (19.20/16.0/14.20). MSVC 2017 is deprecated but still supported. MSVC 2017 support will be removed in a future release. Define CUB_IGNORE_DEPRECATED_CPP_DIALECT to suppress this message.
[716/2287] Building CXX object paddle\fluid\operators\CMakeFiles\clip_op.dir\clip_op.cc.obj
C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\thrust/detail/config/cpp_dialect.h:118: warning: Thrust requires at least MSVC 2019 (19.20/16.0/14.20). MSVC 2017 is deprecated but still supported. MSVC 2017 support will be removed in a future release. Define THRUST_IGNORE_DEPRECATED_CPP_DIALECT to suppress this message.
Unknown compiler version - please run the configure tests and report the results
[717/2287] Building CXX object paddle\fluid\operators\CMakeFiles\coalesce_tensor_op.dir\coalesce_tensor_op.cc.obj
Unknown compiler version - please run the configure tests and report the results
C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\thrust/detail/config/cpp_dialect.h:118: warning: Thrust requires at least MSVC 2019 (19.20/16.0/14.20). MSVC 2017 is deprecated but still supported. MSVC 2017 support will be removed in a future release. Define THRUST_IGNORE_DEPRECATED_CPP_DIALECT to suppress this message.
[718/2287] Building CUDA object paddle\fluid\operators\CMakeFiles\broadcast_tensors_op.dir\broadcast_tensors_op.cu.obj
FAILED: paddle/fluid/operators/CMakeFiles/broadcast_tensors_op.dir/broadcast_tensors_op.cu.obj
"C:\PROGRA~1\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\nvcc.exe" -forward-unknown-to-host-compiler -DBOOST_HAS_STATIC_ASSERT -DCUDA_TOOLKIT_ROOT_DIR=""C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6"" -DCUDA_VERSION_MAJOR="11" -DCUDA_VERSION_MINOR="6" -DCUDNN_MAJOR_VERSION="8" -DEIGEN_STRONG_INLINE=inline -DEIGEN_USE_GPU -DGOOGLE_GLOG_DLL_DECL="" -DLAPACK_FOUND -DPADDLE_DISABLE_PROFILER -DPADDLE_DLL_EXPORT -DPADDLE_DLL_INFERENCE -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_WITH_AVX -DPADDLE_WITH_CRYPTO -DPADDLE_WITH_CUDA -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_MKLML -DPADDLE_WITH_POCKETFFT -DPADDLE_WITH_SSE3 -DTRT_PLUGIN_FP16_AVALIABLE -DUTF8PROC_STATIC -D_XKEYCHECK_H -IG:\workspace\Paddle\build -IG:\workspace\Paddle\paddle\fluid\framework\io -IG:\workspace\Paddle\build\third_party\install\zlib\include -IG:\workspace\Paddle\build\third_party\install -IG:\workspace\Paddle\build\third_party\install\gflags\include -IG:\workspace\Paddle\build\third_party\install\glog\include -IG:\workspace\Paddle\build\third_party\boost\src\extern_boost -IG:\workspace\Paddle\build\third_party\eigen3\src\extern_eigen3 -IG:\workspace\Paddle\build\third_party\threadpool\src\extern_threadpool -IG:\workspace\Paddle\build\third_party\dlpack\src\extern_dlpack\include -IG:\workspace\Paddle\build\third_party\install\xxhash\include -IG:\workspace\Paddle\build\third_party\install\warpctc\include -IG:\workspace\Paddle\build\third_party\install\utf8proc\include -IG:\workspace\Paddle\build\third_party\install\mklml\include -IG:\workspace\Paddle\build\third_party\install\mkldnn\include -IG:\workspace\Paddle\build\third_party\install\protobuf\include -IC:\Users\brucm\scoop\apps\python\3.10.4\include -IC:\Users\brucm\scoop\apps\python\current\Lib\site-packages\numpy\core\include -IG:\workspace\Paddle\build\third_party\pybind\src\extern_pybind\include -IG:\workspace\Paddle\build\third_party\install\cryptopp\include -IG:\workspace\Paddle\build\third_party\pocketfft\src -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include" -IG:\workspace\Paddle -D_WINDOWS -Xcompiler=" /W0 /GR /EHsc" -D_MWAITXINTRIN_H_INCLUDED -D__STRICT_ANSI__ -Wno-deprecated-gpu-targets -gencode arch=compute_61,code=sm_61 -w --expt-relaxed-constexpr --expt-extended-lambda -Xcompiler "/wd4244 /wd4267 /wd4819 " -Xcompiler /bigobj -Xcompiler="/arch:AVX" -Xcompiler="-MT -O2 -Ob2" -DNDEBUG -std=c++14 -MD -MT paddle\fluid\operators\CMakeFiles\broadcast_tensors_op.dir\broadcast_tensors_op.cu.obj -MF paddle\fluid\operators\CMakeFiles\broadcast_tensors_op.dir\broadcast_tensors_op.cu.obj.d -x cu -c G:\workspace\Paddle\paddle\fluid\operators\broadcast_tensors_op.cu -o paddle\fluid\operators\CMakeFiles\broadcast_tensors_op.dir\broadcast_tensors_op.cu.obj -Xcompiler=-Fdpaddle\fluid\operators\CMakeFiles\broadcast_tensors_op.dir\broadcast_tensors_op.pdb,-FS
C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\thrust/detail/config/cpp_dialect.h:118: warning: Thrust requires at least MSVC 2019 (19.20/16.0/14.20). MSVC 2017 is deprecated but still supported. MSVC 2017 support will be removed in a future release. Define THRUST_IGNORE_DEPRECATED_CPP_DIALECT to suppress this message.
Unknown compiler version - please run the configure tests and report the results
c:\program files\nvidia gpu computing toolkit\cuda\v11.6\include\cub\util_cpp_dialect.cuh:129: warning: CUB requires at least MSVC 2019 (19.20/16.0/14.20). MSVC 2017 is deprecated but still supported. MSVC 2017 support will be removed in a future release. Define CUB_IGNORE_DEPRECATED_CPP_DIALECT to suppress this message.
C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\cub/device/dispatch/dispatch_segmented_sort.cuh(338): error: invalid combination of type specifiers

C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\cub/device/dispatch/dispatch_segmented_sort.cuh(338): error: expected an identifier

C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\cub/device/dispatch/dispatch_segmented_sort.cuh(379): error: expected a member name

3 errors detected in the compilation of "G:/workspace/Paddle/paddle/fluid/operators/broadcast_tensors_op.cu".
broadcast_tensors_op.cu
[719/2287] Building CXX object paddle\fluid\operators\CMakeFiles\conv_op.dir\conv_op.cu.cc.obj
C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\thrust/detail/config/cpp_dialect.h:118: warning: Thrust requires at least MSVC 2019 (19.20/16.0/14.20). MSVC 2017 is deprecated but still supported. MSVC 2017 support will be removed in a future release. Define THRUST_IGNORE_DEPRECATED_CPP_DIALECT to suppress this message.
Unknown compiler version - please run the configure tests and report the results
[720/2287] Building CXX object paddle\fluid\operators\CMakeFiles\conj_op.dir\conj_op.cc.obj
C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\thrust/detail/config/cpp_dialect.h:118: warning: Thrust requires at least MSVC 2019 (19.20/16.0/14.20). MSVC 2017 is deprecated but still supported. MSVC 2017 support will be removed in a future release. Define THRUST_IGNORE_DEPRECATED_CPP_DIALECT to suppress this message.
Unknown compiler version - please run the configure tests and report the results
[721/2287] Building CUDA object paddle\fluid\operators\CMakeFiles\cholesky_op.dir\cholesky_op.cu.obj
C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\thrust/detail/config/cpp_dialect.h:118: warning: Thrust requires at least MSVC 2019 (19.20/16.0/14.20). MSVC 2017 is deprecated but still supported. MSVC 2017 support will be removed in a future release. Define THRUST_IGNORE_DEPRECATED_CPP_DIALECT to suppress this message.
c:\program files\nvidia gpu computing toolkit\cuda\v11.6\include\cub\util_cpp_dialect.cuh:129: warning: CUB requires at least MSVC 2019 (19.20/16.0/14.20). MSVC 2017 is deprecated but still supported. MSVC 2017 support will be removed in a future release. Define CUB_IGNORE_DEPRECATED_CPP_DIALECT to suppress this message.
Unknown compiler version - please run the configure tests and report the results
cholesky_op.cu
C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\thrust/detail/config/cpp_dialect.h:118: warning: Thrust requires at least MSVC 2019 (19.20/16.0/14.20). MSVC 2017 is deprecated but still supported. MSVC 2017 support will be removed in a future release. Define THRUST_IGNORE_DEPRECATED_CPP_DIALECT to suppress this message.
c:\program files\nvidia gpu computing toolkit\cuda\v11.6\include\cub\util_cpp_dialect.cuh:129: warning: CUB requires at least MSVC 2019 (19.20/16.0/14.20). MSVC 2017 is deprecated but still supported. MSVC 2017 support will be removed in a future release. Define CUB_IGNORE_DEPRECATED_CPP_DIALECT to suppress this message.
Unknown compiler version - please run the configure tests and report the results
[722/2287] Building CXX object paddle\fluid\operators\CMakeFiles\conv_op.dir\conv_op.cc.obj
C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\thrust/detail/config/cpp_dialect.h:118: warning: Thrust requires at least MSVC 2019 (19.20/16.0/14.20). MSVC 2017 is deprecated but still supported. MSVC 2017 support will be removed in a future release. Define THRUST_IGNORE_DEPRECATED_CPP_DIALECT to suppress this message.
Unknown compiler version - please run the configure tests and report the results
[723/2287] Building CUDA object paddle\fluid\operators\CMakeFiles\center_loss_op.dir\center_loss_op.cu.obj
C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\thrust/detail/config/cpp_dialect.h:118: warning: Thrust requires at least MSVC 2019 (19.20/16.0/14.20). MSVC 2017 is deprecated but still supported. MSVC 2017 support will be removed in a future release. Define THRUST_IGNORE_DEPRECATED_CPP_DIALECT to suppress this message.
Unknown compiler version - please run the configure tests and report the results
c:\program files\nvidia gpu computing toolkit\cuda\v11.6\include\cub\util_cpp_dialect.cuh:129: warning: CUB requires at least MSVC 2019 (19.20/16.0/14.20). MSVC 2017 is deprecated but still supported. MSVC 2017 support will be removed in a future release. Define CUB_IGNORE_DEPRECATED_CPP_DIALECT to suppress this message.
center_loss_op.cu
C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\thrust/detail/config/cpp_dialect.h:118: warning: Thrust requires at least MSVC 2019 (19.20/16.0/14.20). MSVC 2017 is deprecated but still supported. MSVC 2017 support will be removed in a future release. Define THRUST_IGNORE_DEPRECATED_CPP_DIALECT to suppress this message.
Unknown compiler version - please run the configure tests and report the results
c:\program files\nvidia gpu computing toolkit\cuda\v11.6\include\cub\util_cpp_dialect.cuh:129: warning: CUB requires at least MSVC 2019 (19.20/16.0/14.20). MSVC 2017 is deprecated but still supported. MSVC 2017 support will be removed in a future release. Define CUB_IGNORE_DEPRECATED_CPP_DIALECT to suppress this message.
[724/2287] Building CUDA object paddle\fluid\operators\CMakeFiles\clip_by_norm_op.dir\clip_by_norm_op.cu.obj
FAILED: paddle/fluid/operators/CMakeFiles/clip_by_norm_op.dir/clip_by_norm_op.cu.obj
"C:\PROGRA~1\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\nvcc.exe" -forward-unknown-to-host-compiler -DBOOST_HAS_STATIC_ASSERT -DCUDA_TOOLKIT_ROOT_DIR=""C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6"" -DCUDA_VERSION_MAJOR="11" -DCUDA_VERSION_MINOR="6" -DCUDNN_MAJOR_VERSION="8" -DEIGEN_STRONG_INLINE=inline -DEIGEN_USE_GPU -DGOOGLE_GLOG_DLL_DECL="" -DLAPACK_FOUND -DPADDLE_DISABLE_PROFILER -DPADDLE_DLL_EXPORT -DPADDLE_DLL_INFERENCE -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_WITH_AVX -DPADDLE_WITH_CRYPTO -DPADDLE_WITH_CUDA -DPADDLE_WITH_MKLDNN -DPADDLE_WITH_MKLML -DPADDLE_WITH_POCKETFFT -DPADDLE_WITH_SSE3 -DTRT_PLUGIN_FP16_AVALIABLE -DUTF8PROC_STATIC -D_XKEYCHECK_H -IG:\workspace\Paddle\build -IG:\workspace\Paddle\paddle\fluid\framework\io -IG:\workspace\Paddle\build\third_party\install\zlib\include -IG:\workspace\Paddle\build\third_party\install -IG:\workspace\Paddle\build\third_party\install\gflags\include -IG:\workspace\Paddle\build\third_party\install\glog\include -IG:\workspace\Paddle\build\third_party\boost\src\extern_boost -IG:\workspace\Paddle\build\third_party\eigen3\src\extern_eigen3 -IG:\workspace\Paddle\build\third_party\threadpool\src\extern_threadpool -IG:\workspace\Paddle\build\third_party\dlpack\src\extern_dlpack\include -IG:\workspace\Paddle\build\third_party\install\xxhash\include -IG:\workspace\Paddle\build\third_party\install\warpctc\include -IG:\workspace\Paddle\build\third_party\install\utf8proc\include -IG:\workspace\Paddle\build\third_party\install\mklml\include -IG:\workspace\Paddle\build\third_party\install\mkldnn\include -IG:\workspace\Paddle\build\third_party\install\protobuf\include -IC:\Users\brucm\scoop\apps\python\3.10.4\include -IC:\Users\brucm\scoop\apps\python\current\Lib\site-packages\numpy\core\include -IG:\workspace\Paddle\build\third_party\pybind\src\extern_pybind\include -IG:\workspace\Paddle\build\third_party\install\cryptopp\include -IG:\workspace\Paddle\build\third_party\pocketfft\src -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include" -IG:\workspace\Paddle -D_WINDOWS -Xcompiler=" /W0 /GR /EHsc" -D_MWAITXINTRIN_H_INCLUDED -D__STRICT_ANSI__ -Wno-deprecated-gpu-targets -gencode arch=compute_61,code=sm_61 -w --expt-relaxed-constexpr --expt-extended-lambda -Xcompiler "/wd4244 /wd4267 /wd4819 " -Xcompiler /bigobj -Xcompiler="/arch:AVX" -Xcompiler="-MT -O2 -Ob2" -DNDEBUG -std=c++14 -MD -MT paddle\fluid\operators\CMakeFiles\clip_by_norm_op.dir\clip_by_norm_op.cu.obj -MF paddle\fluid\operators\CMakeFiles\clip_by_norm_op.dir\clip_by_norm_op.cu.obj.d -x cu -c G:\workspace\Paddle\paddle\fluid\operators\clip_by_norm_op.cu -o paddle\fluid\operators\CMakeFiles\clip_by_norm_op.dir\clip_by_norm_op.cu.obj -Xcompiler=-Fdpaddle\fluid\operators\CMakeFiles\clip_by_norm_op.dir\clip_by_norm_op.pdb,-FS
C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\thrust/detail/config/cpp_dialect.h:118: warning: Thrust requires at least MSVC 2019 (19.20/16.0/14.20). MSVC 2017 is deprecated but still supported. MSVC 2017 support will be removed in a future release. Define THRUST_IGNORE_DEPRECATED_CPP_DIALECT to suppress this message.
Unknown compiler version - please run the configure tests and report the results
c:\program files\nvidia gpu computing toolkit\cuda\v11.6\include\cub\util_cpp_dialect.cuh:129: warning: CUB requires at least MSVC 2019 (19.20/16.0/14.20). MSVC 2017 is deprecated but still supported. MSVC 2017 support will be removed in a future release. Define CUB_IGNORE_DEPRECATED_CPP_DIALECT to suppress this message.
C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\cub/device/dispatch/dispatch_segmented_sort.cuh(338): error: invalid combination of type specifiers

C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\cub/device/dispatch/dispatch_segmented_sort.cuh(338): error: expected an identifier

C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\cub/device/dispatch/dispatch_segmented_sort.cuh(379): error: expected a member name

3 errors detected in the compilation of "G:/workspace/Paddle/paddle/fluid/operators/clip_by_norm_op.cu".
clip_by_norm_op.cu
[725/2287] Building CUDA object paddle\fluid\operators\CMakeFiles\class_center_sample_op.dir\class_center_sample_op.cu.obj
c:\program files\nvidia gpu computing toolkit\cuda\v11.6\include\cub\util_cpp_dialect.cuh:129: warning: CUB requires at least MSVC 2019 (19.20/16.0/14.20). MSVC 2017 is deprecated but still supported. MSVC 2017 support will be removed in a future release. Define CUB_IGNORE_DEPRECATED_CPP_DIALECT to suppress this message.
C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\thrust/detail/config/cpp_dialect.h:118: warning: Thrust requires at least MSVC 2019 (19.20/16.0/14.20). MSVC 2017 is deprecated but still supported. MSVC 2017 support will be removed in a future release. Define THRUST_IGNORE_DEPRECATED_CPP_DIALECT to suppress this message.
Unknown compiler version - please run the configure tests and report the results
class_center_sample_op.cu
c:\program files\nvidia gpu computing toolkit\cuda\v11.6\include\cub\util_cpp_dialect.cuh:129: warning: CUB requires at least MSVC 2019 (19.20/16.0/14.20). MSVC 2017 is deprecated but still supported. MSVC 2017 support will be removed in a future release. Define CUB_IGNORE_DEPRECATED_CPP_DIALECT to suppress this message.
C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\thrust/detail/config/cpp_dialect.h:118: warning: Thrust requires at least MSVC 2019 (19.20/16.0/14.20). MSVC 2017 is deprecated but still supported. MSVC 2017 support will be removed in a future release. Define THRUST_IGNORE_DEPRECATED_CPP_DIALECT to suppress this message.
Unknown compiler version - please run the configure tests and report the results
[726/2287] Building CUDA object paddle\fluid\operators\CMakeFiles\conj_op.dir\conj_op.cu.obj
C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\thrust/detail/config/cpp_dialect.h:118: warning: Thrust requires at least MSVC 2019 (19.20/16.0/14.20). MSVC 2017 is deprecated but still supported. MSVC 2017 support will be removed in a future release. Define THRUST_IGNORE_DEPRECATED_CPP_DIALECT to suppress this message.
Unknown compiler version - please run the configure tests and report the results
conj_op.cu
C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\thrust/detail/config/cpp_dialect.h:118: warning: Thrust requires at least MSVC 2019 (19.20/16.0/14.20). MSVC 2017 is deprecated but still supported. MSVC 2017 support will be removed in a future release. Define THRUST_IGNORE_DEPRECATED_CPP_DIALECT to suppress this message.
Unknown compiler version - please run the configure tests and report the results
[727/2287] Building CUDA object paddle\fluid\operators\CMakeFiles\clip_op.dir\clip_op.cu.obj
C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\thrust/detail/config/cpp_dialect.h:118: warning: Thrust requires at least MSVC 2019 (19.20/16.0/14.20). MSVC 2017 is deprecated but still supported. MSVC 2017 support will be removed in a future release. Define THRUST_IGNORE_DEPRECATED_CPP_DIALECT to suppress this message.
Unknown compiler version - please run the configure tests and report the results
c:\program files\nvidia gpu computing toolkit\cuda\v11.6\include\cub\util_cpp_dialect.cuh:129: warning: CUB requires at least MSVC 2019 (19.20/16.0/14.20). MSVC 2017 is deprecated but still supported. MSVC 2017 support will be removed in a future release. Define CUB_IGNORE_DEPRECATED_CPP_DIALECT to suppress this message.
clip_op.cu
C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/include\thrust/detail/config/cpp_dialect.h:118: warning: Thrust requires at least MSVC 2019 (19.20/16.0/14.20). MSVC 2017 is deprecated but still supported. MSVC 2017 support will be removed in a future release. Define THRUST_IGNORE_DEPRECATED_CPP_DIALECT to suppress this message.
Unknown compiler version - please run the configure tests and report the results
c:\program files\nvidia gpu computing toolkit\cuda\v11.6\include\cub\util_cpp_dialect.cuh:129: warning: CUB requires at least MSVC 2019 (19.20/16.0/14.20). MSVC 2017 is deprecated but still supported. MSVC 2017 support will be removed in a future release. Define CUB_IGNORE_DEPRECATED_CPP_DIALECT to suppress this message.
ninja: build stopped: subcommand failed.

@Ligoml Ligoml added status/following-up 跟进中 and removed status/new-issue 新建 labels May 5, 2022
@zhwesky2010
Copy link
Contributor

zhwesky2010 commented May 5, 2022 •
edited
Loading

@brucmao 你好，目前还未支持到CUDA11.6，正式发版只发布到CUDA11.2，CUDA11.3-11.5 目前还在内测中，应该也是可以编过。这里看起来是CUDA11.6编译还有语法错误，可以使用CUDA11.2-11.5试试。
太新的CUDA一般会不太稳定或者有计算精度风险，推荐下次安装次新的cuda/cudnn/python版本即可。

@paddle-bot paddle-bot bot closed this as completed May 9, 2023
@paddle-bot
Copy link

paddle-bot bot commented May 9, 2023

Since you haven't replied for more than a year, we have closed this issue/pr.
If the problem is not solved or there is a follow-up one, please reopen it at any time and we will continue to follow up.
由于您超过一年未回复，我们将关闭这个issue/pr。
若问题未解决或有后续问题，请随时重新打开，我们会继续跟进。

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@littletomatodonkey littletomatodonkey

Labels
status/following-up 跟进中 type/build 编译/安装问题
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
4 participants
@brucmao @littletomatodonkey @Ligoml @zhwesky2010