|
Discontinuous Galerkin Library
#include "dg/algorithm.h"
|
MPI matrix classes. More...
Go to the source code of this file.
Classes | |
| struct | dg::MPISparseBlockMat< Vector, LocalMatrixInner, LocalMatrixOuter > |
| Distributed memory Sparse block matrix class, asynchronous communication. More... | |
| struct | dg::MPIDistMat< Vector, LocalMatrixInner, LocalMatrixOuter > |
| Distributed memory matrix class, asynchronous communication. More... | |
| struct | dg::TensorTraits< MPISparseBlockMat< V, LI, LO > > |
| struct | dg::TensorTraits< MPIDistMat< V, LI, LO > > |
Namespaces | |
| namespace | dg |
| This is the namespace for all functions and classes defined and used by the discontinuous Galerkin library. | |
| namespace | dg::blas2 |
| BLAS Level 2 routines. | |
| namespace | dg::blas2::detail |
Functions | |
| template<class MatrixType , class ContainerType1 , class ContainerType2 > | |
| void | dg::blas2::symv (MatrixType &&M, const ContainerType1 &x, ContainerType2 &y) |
| \( y = M x\) | |
| template<class FunctorType , class MatrixType , class ContainerType1 , class ContainerType2 > | |
| void | dg::blas2::stencil (FunctorType f, MatrixType &&M, const ContainerType1 &x, ContainerType2 &y) |
| \( F(M, x, y)\) | |
| template<class MatrixType , class ContainerType1 , class ContainerType2 > | |
| void | dg::blas2::symv (get_value_type< ContainerType1 > alpha, MatrixType &&M, const ContainerType1 &x, get_value_type< ContainerType1 > beta, ContainerType2 &y) |
| \( y = \alpha M x + \beta y\) | |
| template<class Stencil , class ContainerType , class ... ContainerTypes> | |
| void | dg::blas2::detail::doParallelFor (SharedVectorTag, Stencil f, unsigned N, ContainerType &&x, ContainerTypes &&... xs) |
| template<class real_type , class ConversionPolicyRows , class ConversionPolicyCols > | |
| auto | dg::make_mpi_sparseblockmat (const EllSparseBlockMat< real_type, thrust::host_vector > &src, const ConversionPolicyRows &g_rows, const ConversionPolicyCols &g_cols) |
Split given EllSparseBlockMat into computation and communication part. | |
MPI matrix classes.