BLAS-like Extensions#

oneAPI Math Kernel Library DPC++ provides additional routines to extend the functionality of the BLAS routines. These include routines to compute many independent vector-vector and matrix-matrix operations.

The following table lists the BLAS-like extensions with their descriptions.

Routines

Description

axpy_batch

Computes groups of vector-scalar products added to a vector.

gemm_batch

Computes groups of matrix-matrix products with general matrices.

trsm_batch

Solves a triangular matrix equation for a group of matrices.

gemmt

Computes a matrix-matrix product with general matrices, but updates only the upper or lower triangular part of the result matrix.

gemm_bias

Computes a matrix-matrix product using general integer matrices with bias

imatcopy

Computes an in-place matrix transposition or copy.

omatcopy

Computes an out-of-place matrix transposition or copy.

omatcopy2

Computes a two-strided out-of-place matrix transposition or copy.

omatadd

Computes scaled matrix addition with possibly transposed arguments.

imatcopy_batch

Computes groups of in-place matrix transposition or copy operations.

omatcopy_batch

Computes groups of out-of-place matrix transposition or copy operations.

omatadd_batch

Computes groups of scaled matrix additions.

Parent topic: BLAS Routines