LU matrix decomposition

Functions
int	hal_ni_LU32f (float *src1, size_t src1_step, int m, float *src2, size_t src2_step, int n, int *info)
int	hal_ni_LU64f (double *src1, size_t src1_step, int m, double *src2, size_t src2_step, int n, int *info)

Detailed Description

Performs \(LU\) decomposition of square matrix \(A=P*L*U\) (where \(P\) is permutation matrix) and solves matrix equation \(A*X=B\). Function returns the \(sign\) of permutation \(P\) via parameter info.

Parameters

src1	pointer to input matrix \(A\) stored in row major order. After finish of work src1 contains at least \(U\) part of \(LU\) decomposition which is appropriate for determainant calculation: \(det(A)=sign*\prod_{j=1}^{M}a_{jj}\).
src1_step	number of bytes between two consequent rows of matrix \(A\).
m	size of square matrix \(A\).
src2	pointer to \(M\times N\) matrix \(B\) which is the right-hand side of system \(A*X=B\). \(B\) stored in row major order. If src2 is null pointer only \(LU\) decomposition will be performed. After finish of work src2 contains solution \(X\) of system \(A*X=B\).
src2_step	number of bytes between two consequent rows of matrix \(B\).
n	number of right-hand vectors in \(M\times N\) matrix \(B\).
info	indicates success of decomposition. If *info is equals to zero decomposition failed, othervise *info is equals to \(sign\).

Function Documentation

hal_ni_LU32f()

int hal_ni_LU32f ( float *  src1, size_t  src1_step, int  m, float *  src2, size_t  src2_step, int  n, int *  info  )

inline

#include <core/src/hal_replacement.hpp>

hal_ni_LU64f()

int hal_ni_LU64f ( double *  src1, size_t  src1_step, int  m, double *  src2, size_t  src2_step, int  n, int *  info  )

inline

#include <core/src/hal_replacement.hpp>