CMatrixInverse.hpp File Reference

Dense small-matrix inversion via Gauss-Jordan elimination, shared between the host (CSysMatrix::MatrixInverse) and device (CSysPreconditionerGPU.cu) implementations. More...

#include <cmath>
#include "../code_config.hpp"

Go to the source code of this file.

Namespaces
namespace	SU2_LinAlg

Macros
#define	A(I, J)   matrix[(I)*nVar + (J)]
#define	M(I, J)   inverse[(I)*nVar + (J)]

Functions
template<class ScalarType >
void	SU2_LinAlg::RegularizePivot (ScalarType &pivot)
	Regularize a pivot that is too small to prevent divide-by-zero, on host and device this needs to clamp to the same value so that the two produce the same factors.
template<class ScalarType >
void	SU2_LinAlg::MatrixInverse (unsigned long nVar, ScalarType *matrix, ScalarType *inverse)
	Invert the nVar by nVar dense matrix matrix into inverse via Gauss-Jordan elimination with partial pivoting on the diagonal.

Detailed Description

Dense small-matrix inversion via Gauss-Jordan elimination, shared between the host (CSysMatrix::MatrixInverse) and device (CSysPreconditionerGPU.cu) implementations.

Author

F. Palacios, A. Bueno, T. Economon, P. Gomes

Version

8.5.0 "Harrier"

SU2 Project Website: https://su2code.github.io

The SU2 Project is maintained by the SU2 Foundation (http://su2foundation.org)

Copyright 2012-2026, SU2 Contributors (cf. AUTHORS.md)

SU2 is free software; you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation; either version 2.1 of the License, or (at your option) any later version.

SU2 is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.

You should have received a copy of the GNU Lesser General Public License along with SU2. If not, see http://www.gnu.org/licenses/.

Macro Definition Documentation

A

#define A	(		I,
			J
	)		matrix[(I)*nVar + (J)]

M

#define M	(		I,
			J
	)		inverse[(I)*nVar + (J)]