batmat::linalg::micro_kernels::syomv Namespace Reference

Classes
struct	KernelConfig

Functions
template<class T, class Abi, KernelConfig Conf, index_t RowsReg, StorageOrder OA, StorageOrder OB, StorageOrder OD>
void	syomv_microkernel (const uview< const T, Abi, OA > A, const uview< const T, Abi, OB > B, const uview< T, Abi, OD > D, const index_t l0, const index_t k) noexcept
	Symmetric off-diagonal block multiply. Single register block.
template<class T, class Abi, KernelConfig Conf, StorageOrder OA, StorageOrder OB, StorageOrder OD>
void	syomv_register (const view< const T, Abi, OA > A, const view< const T, Abi, OB > B, const view< T, Abi, OD > D) noexcept
	Generalized matrix multiplication D = C ± A⁽ᵀ⁾ B⁽ᵀ⁾. Using register blocking.

Variables
template<class T, class Abi, KernelConfig Conf, StorageOrder OA, StorageOrder OB, StorageOrder OD>
const constinit auto	syomv_lut
template<class T, class Abi>
constexpr index_t	RowsReg
	Register block size of the matrix-matrix multiplication micro-kernels.

---

Class Documentation

batmat::linalg::micro_kernels::syomv::KernelConfig

struct batmat::linalg::micro_kernels::syomv::KernelConfig

Class Members
bool	negate = false
MatrixStructure	struc_A = MatrixStructure::LowerTriangular

Function Documentation

syomv_microkernel()

template<class T, class Abi, KernelConfig Conf, index_t RowsReg, StorageOrder OA, StorageOrder OB, StorageOrder OD>

void batmat::linalg::micro_kernels::syomv::syomv_microkernel ( uview< const T, Abi, OA > A, uview< const T, Abi, OB > B, uview< T, Abi, OD > D, index_t l0, index_t k )

noexcept

Symmetric off-diagonal block multiply. Single register block.

Definition at line 24 of file syomv.tpp.

syomv_register()

template<class T, class Abi, KernelConfig Conf, StorageOrder OA, StorageOrder OB, StorageOrder OD>

void batmat::linalg::micro_kernels::syomv::syomv_register ( view< const T, Abi, OA > A, view< const T, Abi, OB > B, view< T, Abi, OD > D )

noexcept

Generalized matrix multiplication D = C ± A⁽ᵀ⁾ B⁽ᵀ⁾. Using register blocking.

Definition at line 63 of file syomv.tpp.

Variable Documentation

syomv_lut

template<class T, class Abi, KernelConfig Conf, StorageOrder OA, StorageOrder OB, StorageOrder OD>

const constinit auto batmat::linalg::micro_kernels::syomv::syomv_lut

inlineconstinit

Initial value:

                                  =
make_1d_lut<RowsReg<T, Abi>>([]<index_t Row>(index_constant<Row>) {
    return syomv_microkernel<T, Abi, Conf, Row + 1, OA, OB, OD>;
})

Definition at line 15 of file syomv.tpp.

RowsReg

template<class T, class Abi>

index_t batmat::linalg::micro_kernels::gemm::RowsReg

inlineconstexpr

Register block size of the matrix-matrix multiplication micro-kernels.

AVX-512 has 32 vector registers, we use 25 registers for a 5×5 accumulator block of matrix C (leaving some registers for loading A and B):

AVX2 has 16 vector registers, we use 9 registers for a 3×3 accumulator block of matrix C (leaving some registers for loading A and B):

Note

A block size of 4×4 is slightly faster than 3×3 for large matrices, because the even block size results in full cache lines being consumed. For small matrices, 3×3 is faster because it does not spill any registers in the micro-kernels. 2×2 is slower than 3×3 for both small and large matrices (tested using GCC 15.1 on an i7-10750H).

Assumes that the platform has at least 16 vector registers, we use 9 registers for a 3×3 accumulator block of matrix C (leaving some registers for loading A and B):

NEON has 32 vector registers, we use 16 registers for a 4×4 accumulator block of matrix C (leaving plenty of registers for loading A and B):

Definition at line 13 of file avx-512.hpp.