MFEM v4.7.0
Finite element discretization library
Loading...
Searching...
No Matches
mfem::Ginkgo::GinkgoIterativeSolver Class Reference

#include <ginkgo.hpp>

Inheritance diagram for mfem::Ginkgo::GinkgoIterativeSolver:
[legend]
Collaboration diagram for mfem::Ginkgo::GinkgoIterativeSolver:
[legend]

Public Member Functions

const std::shared_ptr< gko::LinOpFactory > GetFactory () const
 
void SetPrintLevel (int print_lvl)
 
int GetNumIterations () const
 
int GetConverged () const
 
real_t GetFinalNorm () const
 
virtual void SetOperator (const Operator &op)
 
virtual void Mult (const Vector &x, Vector &y) const
 
bool UsesVectorWrappers () const
 
virtual ~GinkgoIterativeSolver ()=default
 
- Public Member Functions inherited from mfem::Solver
 Solver (int s=0, bool iter_mode=false)
 Initialize a square Solver with size s.
 
 Solver (int h, int w, bool iter_mode=false)
 Initialize a Solver with height h and width w.
 
- Public Member Functions inherited from mfem::Operator
void InitTVectors (const Operator *Po, const Operator *Ri, const Operator *Pi, Vector &x, Vector &b, Vector &X, Vector &B) const
 Initializes memory for true vectors of linear system.
 
 Operator (int s=0)
 Construct a square Operator with given size s (default 0).
 
 Operator (int h, int w)
 Construct an Operator with the given height (output size) and width (input size).
 
int Height () const
 Get the height (size of output) of the Operator. Synonym with NumRows().
 
int NumRows () const
 Get the number of rows (size of output) of the Operator. Synonym with Height().
 
int Width () const
 Get the width (size of input) of the Operator. Synonym with NumCols().
 
int NumCols () const
 Get the number of columns (size of input) of the Operator. Synonym with Width().
 
virtual MemoryClass GetMemoryClass () const
 Return the MemoryClass preferred by the Operator.
 
virtual void MultTranspose (const Vector &x, Vector &y) const
 Action of the transpose operator: y=A^t(x). The default behavior in class Operator is to generate an error.
 
virtual void AddMult (const Vector &x, Vector &y, const real_t a=1.0) const
 Operator application: y+=A(x) (default) or y+=a*A(x).
 
virtual void AddMultTranspose (const Vector &x, Vector &y, const real_t a=1.0) const
 Operator transpose application: y+=A^t(x) (default) or y+=a*A^t(x).
 
virtual void ArrayMult (const Array< const Vector * > &X, Array< Vector * > &Y) const
 Operator application on a matrix: Y=A(X).
 
virtual void ArrayMultTranspose (const Array< const Vector * > &X, Array< Vector * > &Y) const
 Action of the transpose operator on a matrix: Y=A^t(X).
 
virtual void ArrayAddMult (const Array< const Vector * > &X, Array< Vector * > &Y, const real_t a=1.0) const
 Operator application on a matrix: Y+=A(X) (default) or Y+=a*A(X).
 
virtual void ArrayAddMultTranspose (const Array< const Vector * > &X, Array< Vector * > &Y, const real_t a=1.0) const
 Operator transpose application on a matrix: Y+=A^t(X) (default) or Y+=a*A^t(X).
 
virtual OperatorGetGradient (const Vector &x) const
 Evaluate the gradient operator at the point x. The default behavior in class Operator is to generate an error.
 
virtual void AssembleDiagonal (Vector &diag) const
 Computes the diagonal entries into diag. Typically, this operation only makes sense for linear Operators. In some cases, only an approximation of the diagonal is computed.
 
virtual const OperatorGetProlongation () const
 Prolongation operator from linear algebra (linear system) vectors, to input vectors for the operator. NULL means identity.
 
virtual const OperatorGetRestriction () const
 Restriction operator from input vectors for the operator to linear algebra (linear system) vectors. NULL means identity.
 
virtual const OperatorGetOutputProlongation () const
 Prolongation operator from linear algebra (linear system) vectors, to output vectors for the operator. NULL means identity.
 
virtual const OperatorGetOutputRestrictionTranspose () const
 Transpose of GetOutputRestriction, directly available in this form to facilitate matrix-free RAP-type operators.
 
virtual const OperatorGetOutputRestriction () const
 Restriction operator from output vectors for the operator to linear algebra (linear system) vectors. NULL means identity.
 
void FormLinearSystem (const Array< int > &ess_tdof_list, Vector &x, Vector &b, Operator *&A, Vector &X, Vector &B, int copy_interior=0)
 Form a constrained linear system using a matrix-free approach.
 
void FormRectangularLinearSystem (const Array< int > &trial_tdof_list, const Array< int > &test_tdof_list, Vector &x, Vector &b, Operator *&A, Vector &X, Vector &B)
 Form a column-constrained linear system using a matrix-free approach.
 
virtual void RecoverFEMSolution (const Vector &X, const Vector &b, Vector &x)
 Reconstruct a solution vector x (e.g. a GridFunction) from the solution X of a constrained linear system obtained from Operator::FormLinearSystem() or Operator::FormRectangularLinearSystem().
 
void FormSystemOperator (const Array< int > &ess_tdof_list, Operator *&A)
 Return in A a parallel (on truedofs) version of this square operator.
 
void FormRectangularSystemOperator (const Array< int > &trial_tdof_list, const Array< int > &test_tdof_list, Operator *&A)
 Return in A a parallel (on truedofs) version of this rectangular operator (including constraints).
 
void FormDiscreteOperator (Operator *&A)
 Return in A a parallel (on truedofs) version of this rectangular operator.
 
void PrintMatlab (std::ostream &out, int n, int m=0) const
 Prints operator with input size n and output size m in Matlab format.
 
virtual void PrintMatlab (std::ostream &out) const
 Prints operator in Matlab format.
 
virtual ~Operator ()
 Virtual destructor.
 
Type GetType () const
 Return the type ID of the Operator class.
 

Protected Member Functions

 GinkgoIterativeSolver (GinkgoExecutor &exec, bool use_implicit_res_norm)
 
void update_stop_factory ()
 
- Protected Member Functions inherited from mfem::Operator
void FormConstrainedSystemOperator (const Array< int > &ess_tdof_list, ConstrainedOperator *&Aout)
 see FormSystemOperator()
 
void FormRectangularConstrainedSystemOperator (const Array< int > &trial_tdof_list, const Array< int > &test_tdof_list, RectangularConstrainedOperator *&Aout)
 see FormRectangularSystemOperator()
 
OperatorSetupRAP (const Operator *Pi, const Operator *Po)
 Returns RAP Operator of this, using input/output Prolongation matrices Pi corresponds to "P", Po corresponds to "Rt".
 

Protected Attributes

bool use_implicit_res_norm
 
int print_level
 
int max_iter
 
real_t rel_tol
 
real_t abs_tol
 
real_t final_norm
 
int final_iter
 
int converged
 
std::shared_ptr< gko::LinOpFactory > solver_gen
 
std::shared_ptr< gko::LinOp > solver
 
std::shared_ptr< gko::stop::ResidualNorm<>::Factory > rel_criterion
 
std::shared_ptr< gko::stop::ResidualNorm<>::Factory > abs_criterion
 
std::shared_ptr< gko::stop::ImplicitResidualNorm<>::Factory > imp_rel_criterion
 
std::shared_ptr< gko::stop::ImplicitResidualNorm<>::Factory > imp_abs_criterion
 
std::shared_ptr< gko::log::Convergence<> > convergence_logger
 
std::shared_ptr< ResidualLogger<> > residual_logger
 
std::shared_ptr< gko::stop::Combined::Factory > combined_factory
 
std::shared_ptr< gko::Executor > executor
 
bool needs_wrapped_vecs
 
bool sub_op_needs_wrapped_vecs
 
- Protected Attributes inherited from mfem::Operator
int height
 Dimension of the output / number of rows in the matrix.
 
int width
 Dimension of the input / number of columns in the matrix.
 

Additional Inherited Members

- Public Types inherited from mfem::Operator
enum  DiagonalPolicy { DIAG_ZERO , DIAG_ONE , DIAG_KEEP }
 Defines operator diagonal policy upon elimination of rows and/or columns. More...
 
enum  Type {
  ANY_TYPE , MFEM_SPARSEMAT , Hypre_ParCSR , PETSC_MATAIJ ,
  PETSC_MATIS , PETSC_MATSHELL , PETSC_MATNEST , PETSC_MATHYPRE ,
  PETSC_MATGENERIC , Complex_Operator , MFEM_ComplexSparseMat , Complex_Hypre_ParCSR ,
  Complex_DenseMat , MFEM_Block_Matrix , MFEM_Block_Operator
}
 Enumeration defining IDs for some classes derived from Operator. More...
 
- Public Attributes inherited from mfem::Solver
bool iterative_mode
 If true, use the second argument of Mult() as an initial guess.
 

Detailed Description

This class forms the base class for all of Ginkgo's iterative solvers. It is not intended to be used directly by MFEM applications. The various derived classes only take the additional data that is specific to them and solve the given linear system. The entire collection of solvers that Ginkgo implements is available at the Ginkgo documentation and manual pages, https://ginkgo-project.github.io/ginkgo/doc/develop.

Definition at line 682 of file ginkgo.hpp.

Constructor & Destructor Documentation

◆ ~GinkgoIterativeSolver()

virtual mfem::Ginkgo::GinkgoIterativeSolver::~GinkgoIterativeSolver ( )
virtualdefault

Destructor.

◆ GinkgoIterativeSolver()

mfem::Ginkgo::GinkgoIterativeSolver::GinkgoIterativeSolver ( GinkgoExecutor & exec,
bool use_implicit_res_norm )
protected

Constructor.

The exec defines the paradigm where the solution is computed. use_implicit_res_norm is for internal use by the derived classes for specific Ginkgo solvers; it indicates whether the solver makes an implicit residual norm estimate available for convergence checking. Each derived class automatically sets the correct value when calling this base class constructor.

Definition at line 342 of file ginkgo.cpp.

Member Function Documentation

◆ GetConverged()

int mfem::Ginkgo::GinkgoIterativeSolver::GetConverged ( ) const
inline

Definition at line 697 of file ginkgo.hpp.

◆ GetFactory()

const std::shared_ptr< gko::LinOpFactory > mfem::Ginkgo::GinkgoIterativeSolver::GetFactory ( ) const
inline

Return a pointer to the LinOpFactory that will generate the solver with the parameters set through the specific constructor.

Definition at line 689 of file ginkgo.hpp.

◆ GetFinalNorm()

real_t mfem::Ginkgo::GinkgoIterativeSolver::GetFinalNorm ( ) const
inline

Definition at line 698 of file ginkgo.hpp.

◆ GetNumIterations()

int mfem::Ginkgo::GinkgoIterativeSolver::GetNumIterations ( ) const
inline

Definition at line 696 of file ginkgo.hpp.

◆ Mult()

void mfem::Ginkgo::GinkgoIterativeSolver::Mult ( const Vector & x,
Vector & y ) const
virtual

Solve the linear system Ax=y. Dependent on the information provided by derived classes one of Ginkgo's linear solvers is chosen.

Implements mfem::Operator.

Definition at line 508 of file ginkgo.cpp.

◆ SetOperator()

void mfem::Ginkgo::GinkgoIterativeSolver::SetOperator ( const Operator & op)
virtual

If the Operator is a SparseMatrix, set up a Ginkgo Csr matrix to use its data directly. If the Operator is not a matrix, create an OperatorWrapper for it and store.

Implements mfem::Solver.

Definition at line 640 of file ginkgo.cpp.

◆ SetPrintLevel()

void mfem::Ginkgo::GinkgoIterativeSolver::SetPrintLevel ( int print_lvl)
inline

Definition at line 694 of file ginkgo.hpp.

◆ update_stop_factory()

void mfem::Ginkgo::GinkgoIterativeSolver::update_stop_factory ( )
protected

Rebuild the Ginkgo stopping criterion factory with the latest values of rel_tol, abs_tol, and max_iter.

Definition at line 360 of file ginkgo.cpp.

◆ UsesVectorWrappers()

bool mfem::Ginkgo::GinkgoIterativeSolver::UsesVectorWrappers ( ) const
inline

Return whether this GinkgoIterativeSolver object will use VectorWrapper types for input and output vectors. Note that Mult() will automatically create these wrappers if needed.

Definition at line 718 of file ginkgo.hpp.

Member Data Documentation

◆ abs_criterion

std::shared_ptr<gko::stop::ResidualNorm<>::Factory> mfem::Ginkgo::GinkgoIterativeSolver::abs_criterion
protected

The residual criterion object that controls the reduction of the residual based on an absolute tolerance.

Definition at line 774 of file ginkgo.hpp.

◆ abs_tol

real_t mfem::Ginkgo::GinkgoIterativeSolver::abs_tol
protected

Definition at line 747 of file ginkgo.hpp.

◆ combined_factory

std::shared_ptr<gko::stop::Combined::Factory> mfem::Ginkgo::GinkgoIterativeSolver::combined_factory
protected

The Ginkgo combined factory object is used to create a combined stopping criterion to be passed to the solver.

Definition at line 806 of file ginkgo.hpp.

◆ converged

int mfem::Ginkgo::GinkgoIterativeSolver::converged
mutableprotected

Definition at line 750 of file ginkgo.hpp.

◆ convergence_logger

std::shared_ptr<gko::log::Convergence<> > mfem::Ginkgo::GinkgoIterativeSolver::convergence_logger
mutableprotected

The Ginkgo convergence logger used to check for convergence and other solver data if needed.

Definition at line 794 of file ginkgo.hpp.

◆ executor

std::shared_ptr<gko::Executor> mfem::Ginkgo::GinkgoIterativeSolver::executor
protected

The execution paradigm in Ginkgo. The choices are between gko::OmpExecutor, gko::CudaExecutor and gko::ReferenceExecutor and more details can be found in Ginkgo's documentation.

Definition at line 813 of file ginkgo.hpp.

◆ final_iter

int mfem::Ginkgo::GinkgoIterativeSolver::final_iter
mutableprotected

Definition at line 749 of file ginkgo.hpp.

◆ final_norm

real_t mfem::Ginkgo::GinkgoIterativeSolver::final_norm
mutableprotected

Definition at line 748 of file ginkgo.hpp.

◆ imp_abs_criterion

std::shared_ptr<gko::stop::ImplicitResidualNorm<>::Factory> mfem::Ginkgo::GinkgoIterativeSolver::imp_abs_criterion
protected

The implicit residual criterion object that controls the reduction of the residual based on an absolute tolerance, based on an implicit residual norm value.

Definition at line 788 of file ginkgo.hpp.

◆ imp_rel_criterion

std::shared_ptr<gko::stop::ImplicitResidualNorm<>::Factory> mfem::Ginkgo::GinkgoIterativeSolver::imp_rel_criterion
protected

The implicit residual criterion object that controls the reduction of the residual relative to the initial residual, based on an implicit residual norm value.

Definition at line 781 of file ginkgo.hpp.

◆ max_iter

int mfem::Ginkgo::GinkgoIterativeSolver::max_iter
protected

Definition at line 745 of file ginkgo.hpp.

◆ needs_wrapped_vecs

bool mfem::Ginkgo::GinkgoIterativeSolver::needs_wrapped_vecs
protected

Whether or not we need to use VectorWrapper types with this solver.

Definition at line 818 of file ginkgo.hpp.

◆ print_level

int mfem::Ginkgo::GinkgoIterativeSolver::print_level
protected

Definition at line 744 of file ginkgo.hpp.

◆ rel_criterion

std::shared_ptr<gko::stop::ResidualNorm<>::Factory> mfem::Ginkgo::GinkgoIterativeSolver::rel_criterion
protected

The residual criterion object that controls the reduction of the residual relative to the initial residual.

Definition at line 767 of file ginkgo.hpp.

◆ rel_tol

real_t mfem::Ginkgo::GinkgoIterativeSolver::rel_tol
protected

Definition at line 746 of file ginkgo.hpp.

◆ residual_logger

std::shared_ptr<ResidualLogger<> > mfem::Ginkgo::GinkgoIterativeSolver::residual_logger
mutableprotected

The residual logger object used to check for convergence and other solver data if needed.

Definition at line 800 of file ginkgo.hpp.

◆ solver

std::shared_ptr<gko::LinOp> mfem::Ginkgo::GinkgoIterativeSolver::solver
protected

The Ginkgo solver object, generated for a specific operator.

Definition at line 760 of file ginkgo.hpp.

◆ solver_gen

std::shared_ptr<gko::LinOpFactory> mfem::Ginkgo::GinkgoIterativeSolver::solver_gen
protected

The Ginkgo solver factory object, to generate specific solvers.

Definition at line 755 of file ginkgo.hpp.

◆ sub_op_needs_wrapped_vecs

bool mfem::Ginkgo::GinkgoIterativeSolver::sub_op_needs_wrapped_vecs
protected

Whether or not we need to use VectorWrapper types with the preconditioner or an inner solver. This value is set upon creation of the GinkgoIterativeSolver object and should never change.

Definition at line 825 of file ginkgo.hpp.

◆ use_implicit_res_norm

bool mfem::Ginkgo::GinkgoIterativeSolver::use_implicit_res_norm
protected

Definition at line 743 of file ginkgo.hpp.


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