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

Detailed Description

The AdditionalData contains data for controlling the MUMPS execution.

Definition at line 490 of file sparse_direct.h.

Classes
struct	BlockLowRank

Public Member Functions
	AdditionalData (const bool output_details=false, const bool error_statistics=false, const bool symmetric=false, const bool posdef=false, const bool blr_factorization=false, const BlockLowRank &blr=BlockLowRank())

Public Attributes
bool	output_details

bool	error_statistics

bool	symmetric

bool	posdef

bool	blr_factorization

BlockLowRank	blr

Constructor & Destructor Documentation

◆ AdditionalData()

SparseDirectMUMPS::AdditionalData::AdditionalData	(	const bool	output_details = `false`,
		const bool	error_statistics = `false`,
		const bool	symmetric = `false`,
		const bool	posdef = `false`,
		const bool	blr_factorization = `false`,
		const BlockLowRank &	blr = `BlockLowRank()`
	)

inline

Construct a new Additional Data object with the given parameters.

Definition at line 519 of file sparse_direct.h.

Member Data Documentation

◆ output_details

bool SparseDirectMUMPS::AdditionalData::output_details

If true, the MUMPS solver will print out details of the execution.

Definition at line 536 of file sparse_direct.h.

◆ error_statistics

bool SparseDirectMUMPS::AdditionalData::error_statistics

If true, the MUMPS solver will print out error statistics.

Definition at line 541 of file sparse_direct.h.

◆ symmetric

bool SparseDirectMUMPS::AdditionalData::symmetric

Use symmetric factorization.

This is only possible if the matrix is symmetric, but no checks are performed to ensure that this is actually the case. The solver will likely break or produce incorrect results if the matrix is not symmetric, and this option is set to true.

Definition at line 551 of file sparse_direct.h.

◆ posdef

bool SparseDirectMUMPS::AdditionalData::posdef

Use the positive definite factorization.

This is only possible if the matrix is symmetric and positive definite, but no checks are performed to ensure that this is actually the case. The solver will likely break or produce incorrect results if the matrix is not symmetric and positive definite, and this option is set to true.

Definition at line 561 of file sparse_direct.h.

◆ blr_factorization

bool SparseDirectMUMPS::AdditionalData::blr_factorization

Use the Block Low-Rank factorization.

Definition at line 566 of file sparse_direct.h.

◆ blr

BlockLowRank SparseDirectMUMPS::AdditionalData::blr

Data for the Block Low-Rank approximation.

Definition at line 571 of file sparse_direct.h.

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

include/deal.II/lac/sparse_direct.h