Detailed Description

Utilities for the BLAS/LAPACK (and cuBLAS/cuSOLVER) interface.

Concepts
concept	nda::blas_lapack::BlasArray
	BLAS/LAPACK compatible array type.
concept	nda::blas_lapack::BlasArrayCplx
	BLAS/LAPACK compatible array type with complex value type.
concept	nda::blas_lapack::BlasArrayFor
	BLAS/LAPACK compatible array type that has the same value type as the reference array type and a compatible address space.
concept	nda::blas_lapack::BlasArrayOrConj
	BLAS/LAPACK compatible array or conjugate lazy expression type.
concept	nda::blas_lapack::BlasArrayOrConjFor
	BLAS/LAPACK compatible array or conjugate lazy expression type that has the same value type as the reference array type and a compatible address space.
concept	nda::blas_lapack::BlasArrayReal
	BLAS/LAPACK compatible array type with real value type.
concept	nda::blas_lapack::BlasArrayRealFor
	BLAS/LAPACK compatible array type that has a compatible floating-point value type and address space with the reference array type.
concept	nda::blas_lapack::PivotArrayFor
	BLAS/LAPACK compatible pivot array type that has a compatible address space with the reference array type.

Typedefs
using	nda::dcomplex = std::complex<double>
	Alias for std::complex<double> type.
template<MemoryArray A>
using	nda::blas_lapack::vector_fp_t = vector<get_fp_t<A>, heap<mem::get_addr_space<A>>>
	Alias for an nda::vector with the same address space as the given type and its value type determined by nda::get_fp_t<A>.
template<MemoryArray A>
using	nda::blas_lapack::vector_value_t = vector<get_value_t<A>, heap<mem::get_addr_space<A>>>
	Alias for an nda::vector with the same value type and address space as the given type.

Functions
template<Array A> requires (MemoryArrayOfRank<A, 3> or MemoryArrayOfRank<A, 2>)
auto	nda::blas_lapack::batch_ptrs (A &&a)
	Given a 2- or 3-dimensional array get an array of pointers to each of the submatrices/subvectors indexed by the slowest varying dimension.
template<Array A> requires (MemoryArray<A> or is_conj_array_expr<A>)
MemoryArray decltype(auto)	nda::blas_lapack::get_array (A &&a)
	Get the underlying array of a conjugate lazy expression or return the array itself in case it is an nda::MemoryArray.
template<MemoryArray A> requires (get_rank<A> == 1 or get_rank<A> == 2)
int	nda::blas_lapack::get_ld (A const &a)
	Get the leading dimension of an nda::MemoryArray with rank 1 or 2 for BLAS/LAPACK calls.
template<MemoryArray A> requires (get_rank<A> == 1 or get_rank<A> == 2)
int	nda::blas_lapack::get_ncols (A const &a)
	Get the number of columns of an nda::MemoryArray with rank 1 or 2 for BLAS/LAPACK calls.
template<typename A> requires (is_regular_or_view_v<A> and get_rank<A> == 1)
void	nda::blas_lapack::resize_or_check_work_buffer (A &a, long min_size)
	Resize or check the size of a 1D array/view.

Variables
template<Array A>
static constexpr char	nda::blas_lapack::get_op
	Variable template that determines the BLAS matrix operation tag ('N','T','C') based on the given boolean flags for conjugation and transposition.
template<Array... As>
static constexpr bool	nda::blas_lapack::has_C_layout
	Constexpr variable that is true if all given nda::Array types have nda::C_layout.
template<Array... As>
static constexpr bool	nda::blas_lapack::has_F_layout
	Constexpr variable that is true if all given nda::Array types have nda::F_layout.
template<typename A>
static constexpr bool	nda::blas_lapack::is_conj_array_expr = false
	Constexpr variable that is true if the given type is a conjugate lazy expression.
template<MemoryArray A>
static constexpr bool	nda::blas_lapack::is_conj_array_expr< expr_call< detail::conj_f, A > > = true
	Specialization of nda::blas_lapack::is_conj_array_expr for the conjugate lazy expressions.

Typedef Documentation

◆ vector_fp_t

template<MemoryArray A>

using nda::blas_lapack::vector_fp_t = vector<get_fp_t<A>, heap<mem::get_addr_space<A>>>

#include <nda/blas/tools.hpp>

Alias for an nda::vector with the same address space as the given type and its value type determined by nda::get_fp_t<A>.

Template Parameters

A	nda::MemoryArray type.

Definition at line 170 of file tools.hpp.

◆ vector_value_t

template<MemoryArray A>

using nda::blas_lapack::vector_value_t = vector<get_value_t<A>, heap<mem::get_addr_space<A>>>

#include <nda/blas/tools.hpp>

Alias for an nda::vector with the same value type and address space as the given type.

Template Parameters

A	nda::MemoryArray type.

Definition at line 161 of file tools.hpp.

Function Documentation

◆ batch_ptrs()

template<Array A>
requires (MemoryArrayOfRank<A, 3> or MemoryArrayOfRank<A, 2>)

auto nda::blas_lapack::batch_ptrs ( A && a )

#include <nda/blas/tools.hpp>

Given a 2- or 3-dimensional array get an array of pointers to each of the submatrices/subvectors indexed by the slowest varying dimension.

Template Parameters

A	nda::MemoryArray of rank 2 or 3.

Parameters

a	Input array.

Returns: nda::vector of pointers to each submatrix/subvector.

Definition at line 182 of file tools.hpp.

◆ get_array()

template<Array A>
requires (MemoryArray<A> or is_conj_array_expr<A>)

MemoryArray decltype(auto) nda::blas_lapack::get_array ( A && a )

#include <nda/blas/tools.hpp>

Get the underlying array of a conjugate lazy expression or return the array itself in case it is an nda::MemoryArray.

Template Parameters

A	nda::Array type.

Parameters

a	Conjugate expression or array/view.

Returns: nda::MemoryArray object.

Definition at line 68 of file tools.hpp.

◆ get_ld()

template<MemoryArray A>
requires (get_rank<A> == 1 or get_rank<A> == 2)

int nda::blas_lapack::get_ld ( A const & a )

#include <nda/blas/tools.hpp>

Get the leading dimension of an nda::MemoryArray with rank 1 or 2 for BLAS/LAPACK calls.

The leading dimension is the stride between two consecutive columns (rows) of a matrix in Fortran (C) layout. For 1-dimensional arrays, we simply return the size of the array.

Template Parameters

A	nda::MemoryArray type.

Parameters

a	nda::MemoryArray object.

Returns: Leading dimension for BLAS/LAPACK calls.

Definition at line 128 of file tools.hpp.

◆ get_ncols()

template<MemoryArray A>
requires (get_rank<A> == 1 or get_rank<A> == 2)

int nda::blas_lapack::get_ncols ( A const & a )

#include <nda/blas/tools.hpp>

Get the number of columns of an nda::MemoryArray with rank 1 or 2 for BLAS/LAPACK calls.

The number of columns corresponds to the extent of the second (first) dimension of a matrix in Fortran (C) layout. For 1-dimensional arrays, we return 1.

Template Parameters

A	nda::MemoryArray type.

Parameters

a	nda::MemoryArray object.

Returns: Number of columns for BLAS/LAPACK calls.

Definition at line 148 of file tools.hpp.

◆ resize_or_check_work_buffer()

template<typename A>
requires (is_regular_or_view_v<A> and get_rank<A> == 1)

void nda::blas_lapack::resize_or_check_work_buffer	(	A &	a,
		long	min_size )

#include <nda/blas/tools.hpp>

Resize or check the size of a 1D array/view.

This function is similar to nda::resize_or_check_if_view except that

it only works for 1D arrays/views,
it does not resize or throw an error if the size is too big and
it expects that the memory is contiguous.

Template Parameters

A	Type of the object.

Parameters

a	Object to resize or check.
min_size	Minimum size.

Definition at line 207 of file tools.hpp.

Variable Documentation

◆ get_op

template<Array A>

char nda::blas_lapack::get_op

staticconstexpr

#include <nda/blas/tools.hpp>

Initial value:

                             = []() {
  auto constexpr conj      = is_conj_array_expr<A>;
  auto constexpr transpose = has_C_layout<A>;
  static_assert(!(conj and not transpose), "Error in nda::blas_lapack::get_op: Cannot use conjugate operation alone in BLAS operations");
  if constexpr (conj and transpose)
    return 'C';
  else if constexpr (transpose)
    return 'T';
  else 
    return 'N';
}()

Variable template that determines the BLAS matrix operation tag ('N','T','C') based on the given boolean flags for conjugation and transposition.

Template Parameters

conj	Boolean flag for conjugation.
transpose	Boolean flag for transposition.

Definition at line 104 of file tools.hpp.

Detailed Description

Concepts

Typedefs

Functions

Variables

Typedef Documentation

◆ vector_fp_t

◆ vector_value_t

Function Documentation

◆ batch_ptrs()

◆ get_array()

◆ get_ld()

◆ get_ncols()

◆ resize_or_check_work_buffer()

Variable Documentation

◆ get_op