TensorProductMatrixSymmetricSum< dim, VectorizedArray< Number >, n_rows_1d > Class Template Reference

#include <deal.II/lac/tensor_product_matrix.h>

Inheritance diagram for TensorProductMatrixSymmetricSum< dim, VectorizedArray< Number >, n_rows_1d >:

Public Member Functions
	TensorProductMatrixSymmetricSum ()=default
	TensorProductMatrixSymmetricSum (const std::array< Table< 2, VectorizedArray< Number >>, dim > &mass_matrix, const std::array< Table< 2, VectorizedArray< Number >>, dim > &derivative_matrix)
	TensorProductMatrixSymmetricSum (const Table< 2, VectorizedArray< Number >> &mass_matrix, const Table< 2, VectorizedArray< Number >> &derivative_matrix)
void	reinit (const std::array< Table< 2, VectorizedArray< Number >>, dim > &mass_matrix, const std::array< Table< 2, VectorizedArray< Number >>, dim > &derivative_matrix)
void	reinit (const Table< 2, VectorizedArray< Number >> &mass_matrix, const Table< 2, VectorizedArray< Number >> &derivative_matrix)
Public Member Functions inherited from TensorProductMatrixSymmetricSumBase< dim, VectorizedArray< Number >, n_rows_1d >
unsigned int	m () const
unsigned int	n () const
void	vmult (const ArrayView< VectorizedArray< Number > > &dst, const ArrayView< const VectorizedArray< Number > > &src) const
void	apply_inverse (const ArrayView< VectorizedArray< Number > > &dst, const ArrayView< const VectorizedArray< Number > > &src) const

Private Member Functions
template<typename MatrixArray >
void	reinit_impl (MatrixArray &&mass_matrix, MatrixArray &&derivative_matrix)

Additional Inherited Members
Public Types inherited from TensorProductMatrixSymmetricSumBase< dim, VectorizedArray< Number >, n_rows_1d >
using	value_type = VectorizedArray< Number >
Static Public Attributes inherited from TensorProductMatrixSymmetricSumBase< dim, VectorizedArray< Number >, n_rows_1d >
static constexpr int	n_rows_1d_static
Protected Member Functions inherited from TensorProductMatrixSymmetricSumBase< dim, VectorizedArray< Number >, n_rows_1d >
	TensorProductMatrixSymmetricSumBase ()=default
Protected Attributes inherited from TensorProductMatrixSymmetricSumBase< dim, VectorizedArray< Number >, n_rows_1d >
std::array< Table< 2, VectorizedArray< Number > >, dim >	mass_matrix
std::array< Table< 2, VectorizedArray< Number > >, dim >	derivative_matrix
std::array< AlignedVector< VectorizedArray< Number > >, dim >	eigenvalues
std::array< Table< 2, VectorizedArray< Number > >, dim >	eigenvectors

Detailed Description

template<int dim, typename Number, int n_rows_1d>
class TensorProductMatrixSymmetricSum< dim, VectorizedArray< Number >, n_rows_1d >

This is the template specialization for VectorizedArray<Number> being the arithmetic template. For a detailed description see the main documentation of the generic TensorProductMatrixSymmetricSum class.

Author

Martin Kronbichler and Julius Witte, 2017

Definition at line 336 of file tensor_product_matrix.h.

Constructor & Destructor Documentation

TensorProductMatrixSymmetricSum() [1/3]

template<int dim, typename Number , int n_rows_1d>

TensorProductMatrixSymmetricSum< dim, VectorizedArray< Number >, n_rows_1d >::TensorProductMatrixSymmetricSum ( )

default

Default constructor.

TensorProductMatrixSymmetricSum() [2/3]

template<int dim, typename Number , int n_rows_1d>

TensorProductMatrixSymmetricSum< dim, VectorizedArray< Number >, n_rows_1d >::TensorProductMatrixSymmetricSum	(	const std::array< Table< 2, VectorizedArray< Number >>, dim > &	mass_matrix,
		const std::array< Table< 2, VectorizedArray< Number >>, dim > &	derivative_matrix
	)

Constructor that is equivalent to the empty constructor and immediately calling reinit(const std::array<Table<2,VectorizedArray<Number> >, dim>&,const std::array<Table<2,VectorizedArray<Number> >, dim>&).

TensorProductMatrixSymmetricSum() [3/3]

template<int dim, typename Number , int n_rows_1d>

TensorProductMatrixSymmetricSum< dim, VectorizedArray< Number >, n_rows_1d >::TensorProductMatrixSymmetricSum	(	const Table< 2, VectorizedArray< Number >> &	mass_matrix,
		const Table< 2, VectorizedArray< Number >> &	derivative_matrix
	)

Constructor that is equivalent to the empty constructor and immediately calling reinit(const Table<2,VectorizedArray<Number> >&,const Table<2,VectorizedArray<Number> >&).

Member Function Documentation

reinit() [1/2]

template<int dim, typename Number , int n_rows_1d>

void TensorProductMatrixSymmetricSum< dim, VectorizedArray< Number >, n_rows_1d >::reinit	(	const std::array< Table< 2, VectorizedArray< Number >>, dim > &	mass_matrix,
		const std::array< Table< 2, VectorizedArray< Number >>, dim > &	derivative_matrix
	)

Initializes the tensor product matrix by copying the arrays of 1D mass matrices mass_matrix and 1D derivative matrices derivative_matrix into its base class counterparts, respectively, and by assembling the regarding generalized eigenvalues and eigenvectors in TensorProductMatrixSymmetricSumBase::eigenvalues and TensorProductMatrixSymmetricSumBase::eigenvectors, respectively. Note that the current implementation requires each \(M_{d}\) to be symmetric and positive definite and every \(A_{d}\) to be symmetric and invertible but not necessarily positive definite.

reinit() [2/2]

template<int dim, typename Number , int n_rows_1d>

void TensorProductMatrixSymmetricSum< dim, VectorizedArray< Number >, n_rows_1d >::reinit	(	const Table< 2, VectorizedArray< Number >> &	mass_matrix,
		const Table< 2, VectorizedArray< Number >> &	derivative_matrix
	)

This function is equivalent to the previous reinit() except that we consider the same 1D mass matrix mass_matrix and the same 1D derivative matrix derivative_matrix for each tensor direction.

reinit_impl()

template<int dim, typename Number , int n_rows_1d>

template<typename MatrixArray >

void TensorProductMatrixSymmetricSum< dim, VectorizedArray< Number >, n_rows_1d >::reinit_impl ( MatrixArray &&  mass_matrix, MatrixArray &&  derivative_matrix  )

private

A generic implementation of all reinit() functions based on perfect forwarding, that allows to pass lvalue as well as rvalue arguments.

Template Parameters

MatrixArray

Has to be convertible to the underlying type of TensorProductMatrixSymmetricSumBase::mass_matrix and TensorProductMatrixSymmetricSumBase::derivative_matrix.

---

The documentation for this class was generated from the following file:

• include/deal.II/lac/tensor_product_matrix.h