Ginkgo Generated from branch based on main. Ginkgo version 1.9.0
A numerical linear algebra library targeting many-core architectures
Loading...
Searching...
No Matches
assembly.hpp
1// SPDX-FileCopyrightText: 2017 - 2024 The Ginkgo authors
2//
3// SPDX-License-Identifier: BSD-3-Clause
4
5#ifndef GKO_PUBLIC_CORE_DISTRIBUTED_ASSEMBLY_HPP_
6#define GKO_PUBLIC_CORE_DISTRIBUTED_ASSEMBLY_HPP_
7
8
9#include <ginkgo/config.hpp>
10
11
12#if GINKGO_BUILD_MPI
13
14
15#include <ginkgo/core/base/device_matrix_data.hpp>
16#include <ginkgo/core/base/mpi.hpp>
17#include <ginkgo/core/base/range.hpp>
18
19
20namespace gko {
21namespace experimental {
22namespace distributed {
23
24
25template <typename LocalIndexType, typename GlobalIndexType>
26class Partition;
27
28
44template <typename ValueType, typename LocalIndexType, typename GlobalIndexType>
49
50
51} // namespace distributed
52} // namespace experimental
53} // namespace gko
54
55
56#endif // GINKGO_BUILD_MPI
57#endif // GKO_PUBLIC_CORE_DISTRIBUTED_ASSEMBLY_HPP_
This type is a device-side equivalent to matrix_data.
Definition device_matrix_data.hpp:36
Represents a partition of a range of indices [0, size) into a disjoint set of parts.
Definition vector.hpp:36
A thin wrapper of MPI_Comm that supports most MPI calls.
Definition mpi.hpp:408
This class is used for function parameters in the place of raw pointers.
Definition utils_helper.hpp:41
device_matrix_data< ValueType, GlobalIndexType > assemble_rows_from_neighbors(mpi::communicator comm, const device_matrix_data< ValueType, GlobalIndexType > &input, ptr_param< const Partition< LocalIndexType, GlobalIndexType > > partition)
Assembles device_matrix_data entries owned by this MPI rank from other ranks and communicates entries...
The Ginkgo namespace.
Definition abstract_factory.hpp:20