generic-library/opencv
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Implement internal HAL for GEMM and matrix decompositions

Browse Source
This commit is contained in:
Vladislav Sovrasov
2016-06-03 10:38:30 +03:00
parent e1ba4399e8
commit a2d0cc878c
13 changed files with 1028 additions and 76 deletions
Download Patch File Download Diff File Expand all files Collapse all files

26
modules/core/include/opencv2/core/hal/hal.hpp
View File

@@ -49,17 +49,6 @@
#include "opencv2/core/cvstd.hpp"
#include "opencv2/core/hal/interface.h"
//! @cond IGNORED
#define CALL_HAL(name, fun, ...) \
int res = fun(__VA_ARGS__); \
if (res == CV_HAL_ERROR_OK) \
return; \
else if (res != CV_HAL_ERROR_NOT_IMPLEMENTED) \
CV_Error_(cv::Error::StsInternal, \
("HAL implementation " CVAUX_STR(name) " ==> " CVAUX_STR(fun) " returned %d (0x%08x)", res, res));
//! @endcond
namespace cv { namespace hal {
//! @addtogroup core_hal_functions
@@ -75,6 +64,21 @@ CV_EXPORTS int LU32f(float* A, size_t astep, int m, float* b, size_t bstep, int
CV_EXPORTS int LU64f(double* A, size_t astep, int m, double* b, size_t bstep, int n);
CV_EXPORTS bool Cholesky32f(float* A, size_t astep, int m, float* b, size_t bstep, int n);
CV_EXPORTS bool Cholesky64f(double* A, size_t astep, int m, double* b, size_t bstep, int n);
CV_EXPORTS void SVD32f(float* At, size_t astep, float* W, float* U, size_t ustep, float* Vt, size_t vstep, int m, int n, int flags);
CV_EXPORTS void SVD64f(double* At, size_t astep, double* W, double* U, size_t ustep, double* Vt, size_t vstep, int m, int n, int flags);
CV_EXPORTS void gemm32f(const float* src1, size_t src1_step, const float* src2, size_t src2_step,
float alpha, const float* src3, size_t src3_step, float beta, float* dst, size_t dst_step,
int m_a, int n_a, int n_d, int flags);
CV_EXPORTS void gemm64f(const double* src1, size_t src1_step, const double* src2, size_t src2_step,
double alpha, const double* src3, size_t src3_step, double beta, double* dst, size_t dst_step,
int m_a, int n_a, int n_d, int flags);
CV_EXPORTS void gemm32fc(const float* src1, size_t src1_step, const float* src2, size_t src2_step,
float alpha, const float* src3, size_t src3_step, float beta, float* dst, size_t dst_step,
int m_a, int n_a, int n_d, int flags);
CV_EXPORTS void gemm64fc(const double* src1, size_t src1_step, const double* src2, size_t src2_step,
double alpha, const double* src3, size_t src3_step, double beta, double* dst, size_t dst_step,
int m_a, int n_a, int n_d, int flags);
CV_EXPORTS int normL1_(const uchar* a, const uchar* b, int n);
CV_EXPORTS float normL1_(const float* a, const float* b, int n);

15
modules/core/include/opencv2/core/hal/interface.h
View File

@@ -158,6 +158,21 @@ typedef signed char schar;
#define CV_HAL_DFT_IS_INPLACE 1024
//! @}
//! @name SVD flags
//! @{
#define CV_HAL_SVD_NO_UV 1
#define CV_HAL_SVD_SHORT_UV 2
#define CV_HAL_SVD_MODIFY_A 4
#define CV_HAL_SVD_FULL_UV 8
//! @}
//! @name Gemm flags
//! @{
#define CV_HAL_GEMM_1_T 1
#define CV_HAL_GEMM_2_T 2
#define CV_HAL_GEMM_3_T 4
//! @}
//! @}
#endif

2
modules/core/include/opencv2/core/matx.hpp
View File

@@ -438,7 +438,7 @@ template<typename _Tp, int m> struct Matx_DetOp
return p;
for( int i = 0; i < m; i++ )
p *= temp(i, i);
return 1./p;
return p;
}
};