Hello,
I am trying to build OpenCV 4.8.1 on Linux 22.04 on Nvidia 4070 card. The machine has Xeon processor. While enabling CUDA==On, I get the following errors:
/home/csverma/Downloads/opencv-4.8.1/modules/core/include/opencv2/core/hal/intrin_sse.hpp:1439:1: note: in expansion of macro ‘OPENCV_HAL_IMPL_SSE_SHIFT_OP’
1439 | OPENCV_HAL_IMPL_SSE_SHIFT_OP(v_uint64x2, v_int64x2, epi64, v_srai_epi64)
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
/home/csverma/Downloads/opencv_contrib/modules/xphoto/src/grayworld_white_balance.cpp:332:26: note: candidate expects 1 argument, 2 provided
332 | v_sR1 = v_shr(v_mul(v_sR1, v_gainR), 16);
| ~~~~~~~~~~~~~~~~~~~~~~~^
/home/csverma/Downloads/opencv_contrib/modules/xphoto/src/grayworld_white_balance.cpp:333:26: error: no matching function for call to ‘v_shr(cv::hal_baseline::simd128_cpp::v_uint32, int)’
333 | v_sR2 = v_shr(v_mul(v_sR2, v_gainR), 16);
| ~~~~~~~~~~~~~~~~~~~~~~~^
In file included from /home/csverma/Downloads/opencv-4.8.1/modules/core/include/opencv2/core/hal/intrin.hpp:217,