Interface for BLAS library. More...

#include <complex>

Functions
void	dgemm_ (char transa, char transb, int m, int n, int k, double alpha, double A, int lda, double B, int ldb, double beta, double C, int *ldc)
	BLAS matrix multiplication. More...

void	zgemm_ (char transa, char transb, int m, int n, int k, std::complex< double > alpha, std::complex< double > A, int lda, std::complex< double > B, int ldb, std::complex< double > beta, std::complex< double > C, int *ldc)

void	dgemv_ (char trans, int m, int n, double alpha, double A, int LDA, double x, int incx, double beta, double y, int *incy)
	Matrix-Vector Multiplication. More...

void	zgemv_ (char trans, int m, int n, std::complex< double > alpha, std::complex< double > A, int LDA, std::complex< double > x, int incx, std::complex< double > beta, std::complex< double > y, int *incy)

Detailed Description

Interface for BLAS library.

Author: simon <simon@thinks>

Date: Thu May 7 22:41:17 2015

Contains declarations for BLAS functions.

Definition in file blas.h.

Function Documentation

◆ dgemm_()

void dgemm_	(	char *	transa,
		char *	transb,
		int *	m,
		int *	n,
		int *	k,
		double *	alpha,
		double *	A,
		int *	lda,
		double *	B,
		int *	ldb,
		double *	beta,
		double *	C,
		int *	ldc
	)

BLAS matrix multiplication.

Performs the operation C = alpha * (transa(A) * transb(B)) + beta * C where transa(A) is either A or A^T and trans(B) either B or B^T.

Note the difference in memory layout between arts and BLAS. arts matrices using row-major order whereas BLAS uses column major order. Therefore, passing the raw-data of the arts matrices A, B, C to BLAS will compute

C = (A^T * B^T)^T = B * A

It is therefore necessary to pass the factors to dgemm_ in reverse order.

Parameters

[in]	transa	If transa points to char 'T', the matrix A is transposed for the multiplication. Otherwise the untransposed matrix A is used.
[in]	transb	If transb points to char 'T', the matrix B is transposed for the multiplication. Otherwise the untransposed matrix B is used.
[in]	m	Pointer to an int variable containing the number of rows of A.
[in]	n	Pointer to an int variable containing the number of columns of B.
[in]	k	Pointer to an int variable containing the number of columns of A / rows of B the matrix is to be performed over.
[in]	alpha	Pointer to the factor alpha.
[in]	A	Pointer to the double array containing A.
[in]	lda	Pointer to an int variable containing the number of colums of A.
[in]	B	Pointer to the double array containing B.
[in]	ldb	Pointer to an int variabl containing the number of rows of B.
[in]	beta	Pointer to the factor beta
[in,out]	C	Pointer to the double array containing the matrix C.
[in]	ldc	Pointer to an int variable containing the number rows of C.

◆ dgemv_()

void dgemv_	(	char *	trans,
		int *	m,
		int *	n,
		double *	alpha,
		double *	A,
		int *	LDA,
		double *	x,
		int *	incx,
		double *	beta,
		double *	y,
		int *	incy
	)

Matrix-Vector Multiplication.

Perform one of the matrix-vector operations

y = alpha * (A x) + beta * y
y = alpha * (A^T x) + beta * y

where * denotes multiplication by a scalar and A x matrix-vector multiplication.

Parameters

trans	If trans points to the char 'N' or 'n', the standard matrix-vector product (A x) is computed. If trans points to 'C','c','T' or 't', the matrix-vector product (A^T x) is computed.
m	Number of rows of the matrix A.
n	Number of columns of the matrix A.
alpha	The scalar alpha.
A	Pointer to the matrix A.
LDA	The leading dimension of A.
x	Pointer to the vector x.
incx	The stride of the vector x.
beta	The scalar beta.
y	Pointer to the vector y.
incy	The stride of the vector y.

Referenced by mult().

◆ zgemm_()

void zgemm_	(	char *	transa,
		char *	transb,
		int *	m,
		int *	n,
		int *	k,
		std::complex< double > *	alpha,
		std::complex< double > *	A,
		int *	lda,
		std::complex< double > *	B,
		int *	ldb,
		std::complex< double > *	beta,
		std::complex< double > *	C,
		int *	ldc
	)

◆ zgemv_()