TNL::Matrices::GinkgoOperator< Matrix > Class Template Reference

Wraps a general TNL matrix as a Ginkgo LinOp. More...

#include <TNL/Matrices/GinkgoOperator.h>

Inheritance diagram for TNL::Matrices::GinkgoOperator< Matrix >:

Collaboration diagram for TNL::Matrices::GinkgoOperator< Matrix >:

Public Member Functions
	GinkgoOperator (std::shared_ptr< const gko::Executor > exec)
	GinkgoOperator (std::shared_ptr< const gko::Executor > exec, const Matrix &matrix)

Protected Types
using	ConstViewType = typename ViewType::ConstViewType
using	ValueType = typename Matrix::RealType
using	ViewType = Containers::VectorView< ValueType, typename Matrix::DeviceType, typename Matrix::IndexType >

Protected Member Functions
void	apply_impl (const gko::LinOp *alpha, const gko::LinOp *b, const gko::LinOp *beta, gko::LinOp *x) const override
	Applies the operator on a vector: computes the expression x = alpha * op(b) + beta * x
void	apply_impl (const gko::LinOp *b, gko::LinOp *x) const override
	Applies the operator on a vector: computes the expression x = op(b)

Detailed Description

template<typename Matrix>
class TNL::Matrices::GinkgoOperator< Matrix >

Wraps a general TNL matrix as a Ginkgo LinOp.

Note

This LinOp cannot be used with preconditioners that require a CSR matrix, see https://github.com/ginkgo-project/ginkgo/issues/1094. If you have a CSR matrix in TNL already, use getGinkgoMatrixCsrView. If you have a TNL matrix in a different format, use getGinkgoMatrixCsr to do the conversion.

---

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

• src/TNL/Matrices/GinkgoOperator.h