batmat::ops Namespace Reference

Namespaces
namespace	detail

Gathering elements from memory
template<class T, class AbiT, class I, class AbiI, class M>
datapar::simd< T, AbiT >	gather (const T *p, datapar::simd< I, AbiI > idx, M mask)
	Gathers elements from memory at the addresses specified by idx, which should be an integer SIMD vector, and returns them in a SIMD vector of type datapar::simd<T, AbiT>.

Lane-wise rotations of SIMD vectors
template<int S, class F, class Abi>
datapar::simd< F, Abi >	rotl (datapar::simd< F, Abi > x)
	Rotates the elements of x by s positions to the left.
template<int S, class F, class Abi>
datapar::simd< F, Abi >	rotr (datapar::simd< F, Abi > x)
	Rotate the elements of x to the right by S positions.
template<int S, class F, class Abi>
datapar::simd< F, Abi >	shiftl (datapar::simd< F, Abi > x)
	Shift the elements of x to the left by S positions, shifting in zeros.
template<int S, class F, class Abi>
datapar::simd< F, Abi >	shiftr (datapar::simd< F, Abi > x)
	Shift the elements of x to the right by S positions, shifting in zeros.
template<class F, class Abi>
datapar::simd< F, Abi >	rot (datapar::simd< F, Abi > x, int s)
	Rotate the elements of x to the right by s positions.

Inverse square root
template<std::floating_point T>
T	rsqrt (T x)
	Inverse square root.
template<class T, class Abi>
datapar::simd< T, Abi >	rsqrt (datapar::simd< T, Abi > x)
	Inverse square root.

Transposition
template<index_t R, index_t C, class T>
void	transpose_dyn (const T *pa, index_t lda, T *pb, index_t ldb, index_t d=R)
	Transposes the R × C matrix at pa with leading dimension lda, writing the result to pb with leading dimension ldb, writing only the d first columns of the result.
template<index_t R, index_t C, class T>
void	transpose (const T *pa, index_t lda, T *pb, index_t ldb)
	Transposes the R × C matrix at pa with leading dimension lda, writing the result to pb with leading dimension ldb.

Functions
template<class T>
T	cneg (T x, T signs)
	Conditionally negates the sign bit of x, depending on signs, which should contain only ±0 (i.e.

Variables
template<class T>
constexpr index_t	RowsRegTranspose = 8
template<class T>
constexpr index_t	ColsRegTranspose = 8
template<>
constexpr index_t	RowsRegTranspose< double > = 4
template<>
constexpr index_t	ColsRegTranspose< double > = 4

Function Documentation

cneg()

template<class T>

T batmat::ops::detail::cneg	(	T	x,
		T	signs )

Conditionally negates the sign bit of x, depending on signs, which should contain only ±0 (i.e.

only the sign bit of an IEEE-754 floating point number).

Definition at line 42 of file cneg.hpp.

Variable Documentation

RowsRegTranspose

template<class T>

index_t batmat::ops::RowsRegTranspose = 8

inlineconstexpr

Definition at line 23 of file avx-512.hpp.

ColsRegTranspose

template<class T>

index_t batmat::ops::ColsRegTranspose = 8

inlineconstexpr

Definition at line 25 of file avx-512.hpp.

RowsRegTranspose< double >

template<>

index_t batmat::ops::RowsRegTranspose< double > = 4

inlineconstexpr

Definition at line 29 of file avx-512.hpp.

ColsRegTranspose< double >

template<>

index_t batmat::ops::ColsRegTranspose< double > = 4

inlineconstexpr

Definition at line 31 of file avx-512.hpp.