void vsip_mcopy_f_f(const vsip_mview_f* A, const vsip_mview_f* B); void vsip_mcopy_d_d(const vsip_mview_d* A, const vsip_mview_d* B); void vsip_cmcopy_f_f(const vsip_cmview_f* A, const vsip_cmview_f* B); void vsip_cmcopy_d_d(const vsip_cmview_d* A, const vsip_cmview_d* B);
These functions copy the contents of one matrix view to another, with optional type conversion. The functions handle both real and complex matrices of various precision levels:
vsip_mcopy_f_f: Copy from float matrix to float matrix
vsip_cmcopy_f_f: Copy from complex float matrix to complex float matrix
const vsip_dmview_p* A: Pointer to the source matrix view
const vsip_dmview_p* B: Pointer to the destination matrix view
vsip_mview_f *src_matrix_f; vsip_mview_f *dst_matrix_f; // Copy float matrix to float matrix vsip_mcopy_f_f(src_matrix_f, dst_matrix_f);
The source and destination matrices must have the same dimensions.
For type conversion functions, appropriate rounding or truncation is applied when converting to integer types.
When converting from higher precision to lower precision (e.g., double to float), values may be truncated or rounded.
The matrices can be views of larger matrices or blocks, allowing for copying of submatrices.