The real operands of gemm_split, kept by each thread from one product to the next (split_storage): mapping their pages anew took a third to a half of a complex product of order 64 to 256 on one x86-64 core (six AVX2 cores, all threads: order 128 0.52 → 0.17 ms), 15–20 % of one of order 256 on arm64. More...

#include <gemm.hpp>

Public Attributes
buffer< R >	a {}
	A's parts for a block of depth: `[Re A \| −Im A \| Im A]`, or `[Re A \| −Im A]`.

buffer< R >	b {}
	B's parts for a block of columns and of depth: `[Re B \| Im B]`, or `[[Re B, Im B], [Im B, −Re B]]`.

buffer< R >	c {}
	`[Re C \| Im C]`, a block of columns.

Detailed Description

template<typename R>
struct numerical::detail::split_operands< R >

The real operands of gemm_split, kept by each thread from one product to the next (split_storage): mapping their pages anew took a third to a half of a complex product of order 64 to 256 on one x86-64 core (six AVX2 cores, all threads: order 128 0.52 → 0.17 ms), 15–20 % of one of order 256 on arm64.

Definition at line 2415 of file gemm.hpp.