|
Tpetra parallel linear algebra
Version of the Day
|
Default implementation of local (but process-global) GEMM (dense matrix-matrix multiply), for Tpetra::MultiVector. More...
#include "TpetraCore_config.h"#include "Kokkos_ArithTraits.hpp"#include "Kokkos_Complex.hpp"#include <type_traits>Go to the source code of this file.
Namespaces | |
| Tpetra | |
| Namespace Tpetra contains the class and methods constituting the Tpetra library. | |
| Tpetra::Details | |
| Namespace for Tpetra implementation details. | |
Functions | |
| template<class ViewType1 , class ViewType2 , class ViewType3 , class CoefficientType , class IndexType = int> | |
| void | Tpetra::Details::Blas::Default::gemm (const char transA, const char transB, const CoefficientType &alpha, const ViewType1 &A, const ViewType2 &B, const CoefficientType &beta, const ViewType3 &C) |
Default implementation of dense matrix-matrix multiply on a single MPI process: C := alpha*A*B + beta*C. More... | |
Default implementation of local (but process-global) GEMM (dense matrix-matrix multiply), for Tpetra::MultiVector.
Definition in file Tpetra_Details_defaultGemm.hpp.
| void Tpetra::Details::Blas::Default::gemm | ( | const char | transA, |
| const char | transB, | ||
| const CoefficientType & | alpha, | ||
| const ViewType1 & | A, | ||
| const ViewType2 & | B, | ||
| const CoefficientType & | beta, | ||
| const ViewType3 & | C | ||
| ) |
Default implementation of dense matrix-matrix multiply on a single MPI process: C := alpha*A*B + beta*C.
| ViewType1 | Type of the first matrix input A. |
| ViewType2 | Type of the second matrix input B. |
| ViewType3 | Type of the third matrix input/output C. |
| CoefficientType | Type of the scalar coefficients alpha and beta. |
| IndexType | Type of the index used in for loops; defaults to int. |
ViewType1, ViewType2, and ViewType3 are all Kokkos::View specializations.
Definition at line 81 of file Tpetra_Details_defaultGemm.hpp.
1.8.13