You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Operating System / Platform => Windows 11 Pro 64 Bit 10.0.22631
Compiler => Visual Studio 2022 MSVC LTSC v17.8.9
CL version => 19.38.33136
C++ standard => 17
Cmake version => 3.26.4
CUDA version => 11.8.89
cuDNN version => 8.9.7
NVCUVID version => 12.1.14
Detailed description
I want to build the opencv package with CUDA support, now I using anaconda build with the edited recipe, but the I went the issue about this opencv_contrib. It seems that the type of the parameter of this nppiMeanStdDevGetBufferHostSize_8u_C1R function is worng with cannot convert argument 2 from 'int *' to 'size_t *. It may be the problem of source code, I think it is not a question.
Just like the following log shows:
Could any one can help me to figure out this.
The log of the error
[556/2115] Building CXX object modules\cudaarithm\CMakeFiles\opencv_cudaarithm.dir\src\reductions.cpp.obj
FAILED: modules/cudaarithm/CMakeFiles/opencv_cudaarithm.dir/src/reductions.cpp.obj
"C:\PROGRA~1\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.38.33130\bin\Hostx64\x64\cl.exe" /nologo /TP -DCVAPI_EXPORTS -D_USE_MATH_DEFINES -D_VARIADIC_MAX=10 -D_WIN32_WINNT=0x0601 -D__OPENCV_BUILD=1 -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D__STDC_LIMIT_MACROS -I%SRC_DIR%\build\3rdparty\ippicv\ippicv_win\icv\include -I%SRC_DIR%\build\3rdparty\ippicv\ippicv_win\iw\include -I%SRC_DIR%\build -I%SRC_DIR%\opencv_contrib\modules\cudaarithm\include -I%SRC_DIR%\build\modules\cudaarithm -I%SRC_DIR%\opencv_contrib\modules\cudev\include -I%SRC_DIR%\modules\core\include -external:I%PREFIX%\Library\include -external:I%BUILD_PREFIX%\include -external:I%PREFIX%\Library\include\eigen3 -external:W0 -D_CRT_USE_C_COMPLEX_H /DWIN32 /D_WINDOWS /W4 /GR /D _CRT_SECURE_NO_DEPRECATE /D _CRT_NONSTDC_NO_DEPRECATE /D _SCL_SECURE_NO_WARNINGS /Gy /bigobj /Oi /fp:fast /FS /EHa /wd4127 /wd4251 /wd4324 /wd4275 /wd4512 /wd4589 /wd4819 /wd4127 /wd4324 /wd4512 /O2 /Ob2 /DNDEBUG -std:c++17 -MD /showIncludes /Fomodules\cudaarithm\CMakeFiles\opencv_cudaarithm.dir\src\reductions.cpp.obj /Fdlib\opencv_cudaarithm490.pdb /FS -c %SRC_DIR%\opencv_contrib\modules\cudaarithm\src\reductions.cpp
%SRC_DIR%\opencv_contrib\modules\cudaarithm\src\reductions.cpp(159): error C2664: 'NppStatus nppiMeanStdDevGetBufferHostSize_8u_C1R(NppiSize,size_t *)': cannot convert argument 2 from 'int *' to 'size_t *'
%SRC_DIR%\opencv_contrib\modules\cudaarithm\src\reductions.cpp(159): note: Types pointed to are unrelated; conversion requires reinterpret_cast, C-style cast or parenthesized function-style cast
%BUILD_PREFIX%\include\nppi_statistics_functions.h(5350): note: see declaration of 'nppiMeanStdDevGetBufferHostSize_8u_C1R'
%SRC_DIR%\opencv_contrib\modules\cudaarithm\src\reductions.cpp(159): note: while trying to match the argument list '(NppiSize, int *)'
%SRC_DIR%\opencv_contrib\modules\cudaarithm\src\reductions.cpp(161): error C2664: 'NppStatus nppiMeanStdDevGetBufferHostSize_32f_C1R(NppiSize,size_t *)': cannot convert argument 2 from 'int *' to 'size_t *'
%SRC_DIR%\opencv_contrib\modules\cudaarithm\src\reductions.cpp(161): note: Types pointed to are unrelated; conversion requires reinterpret_cast, C-style cast or parenthesized function-style cast
%BUILD_PREFIX%\include\nppi_statistics_functions.h(5401): note: see declaration of 'nppiMeanStdDevGetBufferHostSize_32f_C1R'
%SRC_DIR%\opencv_contrib\modules\cudaarithm\src\reductions.cpp(161): note: while trying to match the argument list '(NppiSize, int *)'
%SRC_DIR%\opencv_contrib\modules\cudaarithm\src\reductions.cpp(235): error C2664: 'NppStatus nppiMeanStdDevGetBufferHostSize_8u_C1MR(NppiSize,size_t *)': cannot convert argument 2 from 'int *' to 'size_t *'
%SRC_DIR%\opencv_contrib\modules\cudaarithm\src\reductions.cpp(235): note: Types pointed to are unrelated; conversion requires reinterpret_cast, C-style cast or parenthesized function-style cast
%BUILD_PREFIX%\include\nppi_statistics_functions.h(5418): note: see declaration of 'nppiMeanStdDevGetBufferHostSize_8u_C1MR'
%SRC_DIR%\opencv_contrib\modules\cudaarithm\src\reductions.cpp(235): note: while trying to match the argument list '(NppiSize, int *)'
%SRC_DIR%\opencv_contrib\modules\cudaarithm\src\reductions.cpp(237): error C2664: 'NppStatus nppiMeanStdDevGetBufferHostSize_32f_C1MR(NppiSize,size_t *)': cannot convert argument 2 from 'int *' to 'size_t *'
%SRC_DIR%\opencv_contrib\modules\cudaarithm\src\reductions.cpp(237): note: Types pointed to are unrelated; conversion requires reinterpret_cast, C-style cast or parenthesized function-style cast
%BUILD_PREFIX%\include\nppi_statistics_functions.h(5469): note: see declaration of 'nppiMeanStdDevGetBufferHostSize_32f_C1MR'
%SRC_DIR%\opencv_contrib\modules\cudaarithm\src\reductions.cpp(237): note: while trying to match the argument list '(NppiSize, int *)'
I found the problem, the conda build downloads wrong version of libnpp of 12.2.5.30-0, which introduces a breakable change
, changed the type of the parameaters from int * to size_t *, so the opencv don't supports the CUDA 12 well. It is actually as a issue.
System information (version)
Detailed description
I want to build the opencv package with CUDA support, now I using anaconda build with the edited recipe, but the I went the issue about this opencv_contrib. It seems that the type of the parameter of this
nppiMeanStdDevGetBufferHostSize_8u_C1R
function is worng withcannot convert argument 2 from 'int *' to 'size_t *
. It may be the problem of source code, I think it is not a question.Just like the following log shows:
Could any one can help me to figure out this.
The log of the error
Build configration
Steps to reproduce
Download the source code from github acrhive of this version
build it with these cmake command
Issue submission checklist
forum.opencv.org, Stack Overflow, etc and have not found any solution
The text was updated successfully, but these errors were encountered: