part of nmblas library, sgemm function implementation Подробнее...

Подробное описание

part of nmblas library, sgemm function implementation

single precision matrix multiplication implementation

C = alpha*A*B + beta*C

see blas\cblas docs for detailed description

limitations: M, N, K, ld* must be all even, all matrix rows must be even aligned

limitations: row major (C order, not Fortran) only

limitations: TransA is not supported yet

Аргументы

[in]	TransA	transpose first multiplicand (A) flag
[in]	TransB	transpose second multiplicand (B) flag
[in]	M	number of rows of the matrix A and C
[in]	N	number of columns of the matrix B and C
[in]	K	number of rows of the matrix B and of columns of the matrix C
[in]	alpha	specifies the scalar alpha
[in]	A	pointer to the first multiplicand
[in]	lda	A stride, first dimension of A holder
[in]	B	pointer to the second multiplicand
[in]	ldb	B stride, first dimension of B holder
[in]	beta	specifies the scalar beta
[in,out]	C	pointer to accumulator matrix
[in]	ldc	C stride, first dimension of C holder