Reference documentation for deal.II version GIT 5dcc62ab57 2022-07-04 21:05:02+00:00
\(\newcommand{\dealvcentcolon}{\mathrel{\mathop{:}}}\) \(\newcommand{\dealcoloneq}{\dealvcentcolon\mathrel{\mkern-1.2mu}=}\) \(\newcommand{\jump}[1]{\left[\!\left[ #1 \right]\!\right]}\) \(\newcommand{\average}[1]{\left\{\!\left\{ #1 \right\}\!\right\}}\)
Enumerations
Utilities::MPI::internal::Tags Namespace Reference

Enumerations

enum  enumeration : std::uint16_t {
  mpi_some_to_some = 300 , compute_point_to_point_communication_pattern , exchange_cell_data_to_ghosts , triangulation_communicate_locally_moved_vertices ,
  exchange_cell_data_request , exchange_cell_data_reply , mg_transfer_fill_copy_indices , sparsity_tools_distribute_sparsity_pattern ,
  dictionary_reinit , consensus_algorithm_payload_get_requesters , fe_tools_extrapolate , fe_tools_extrapolate_end = fe_tools_extrapolate + 10 ,
  consensus_algorithm_nbx_answer_request , consensus_algorithm_nbx_process_deliver , consensus_algorithm_pex_answer_request , consensus_algorithm_pex_process_deliver ,
  fully_distributed_create , triangulation_base_fill_level_ghost_owners , grid_tools_compute_local_to_global_vertex_index_map , grid_tools_compute_local_to_global_vertex_index_map2 ,
  particle_handler_send_recv_particles_setup , particle_handler_send_recv_particles_cache_setup , particle_handler_send_recv_particles_send , scalapack_copy_to ,
  scalapack_copy_to2 , scalapack_copy_from , process_grid_constructor , partitioner_import_start ,
  partitioner_import_end = partitioner_import_start + 200 , partitioner_export_start , partitioner_export_end = partitioner_export_start + 200 , noncontiguous_partitioner_update_ghost_values_start ,
  noncontiguous_partitioner_update_ghost_values_end , compute_union , remote_point_evaluation , fine_dof_handler_view_reinit ,
  distributed_compute_point_locations , affine_constraints_make_consistent_in_parallel_0 , affine_constraints_make_consistent_in_parallel_1
}
 

Detailed Description

This enum holds all MPI tags used in point to point MPI communications inside the deal.II library.

We keep these tags in a central location so that they are unique within the library. Otherwise, communication that receives packages might pick up packets from a different algorithm. This is especially true if MPI_ANY_SOURCE is used.

The list of MPI functions that use an MPI tag is:

Enumeration Type Documentation

◆ enumeration

The enum with the tags.

Enumerator
mpi_some_to_some 

Utilities::MPI::some_to_some()

compute_point_to_point_communication_pattern 

Utilities::MPI::compute_point_to_point_communication_pattern()

exchange_cell_data_to_ghosts 

GridTools::exchange_cell_data_to_ghosts():

triangulation_communicate_locally_moved_vertices 

Triangulation<dim, spacedim>::communicate_locally_moved_vertices()

exchange_cell_data_request 

grid_tools.h: exchange_cell_ghosts()

exchange_cell_data_reply 

grid_tools.h: exchange_cell_ghosts()

mg_transfer_fill_copy_indices 

mg_transfer_internal.cc: fill_copy_indices()

sparsity_tools_distribute_sparsity_pattern 

SparsityTools::sparsity_tools_distribute_sparsity_pattern()

dictionary_reinit 

Dictionary::reinit()

consensus_algorithm_payload_get_requesters 

ConsensusAlgorithms::Payload::get_requesters()

fe_tools_extrapolate 

FETools::extrapolate()

fe_tools_extrapolate_end 

FETools::extrapolate(), allocate space for 10 rounds:

consensus_algorithm_nbx_answer_request 

ConsensusAlgorithms::NBX::process.

consensus_algorithm_nbx_process_deliver 

ConsensusAlgorithms::NBX::process.

consensus_algorithm_pex_answer_request 

ConsensusAlgorithms::PEX::process.

consensus_algorithm_pex_process_deliver 

ConsensusAlgorithms::PEX::process.

fully_distributed_create 

TriangulationDescription::Utilities::create_description_from_triangulation()

triangulation_base_fill_level_ghost_owners 

TriangulationBase<dim, spacedim>::fill_level_ghost_owners()

grid_tools_compute_local_to_global_vertex_index_map 

GridTools::compute_local_to_global_vertex_index_map.

grid_tools_compute_local_to_global_vertex_index_map2 

GridTools::compute_local_to_global_vertex_index_map second tag.

particle_handler_send_recv_particles_setup 

ParticleHandler<dim, spacedim>::send_recv_particles.

particle_handler_send_recv_particles_cache_setup 

ParticleHandler<dim, spacedim>::send_recv_particles.

particle_handler_send_recv_particles_send 

ParticleHandler<dim, spacedim>::send_recv_particles.

scalapack_copy_to 

ScaLAPACKMatrix<NumberType>::copy_to.

scalapack_copy_to2 

ScaLAPACKMatrix<NumberType>::copy_to.

scalapack_copy_from 

ScaLAPACKMatrix<NumberType>::copy_from.

process_grid_constructor 

ProcessGrid::ProcessGrid.

partitioner_import_start 

200 tags for Partitioner::import_from_ghosted_array_start

partitioner_import_end 
partitioner_export_start 

200 tags for Partitioner::export_to_ghosted_array_start

partitioner_export_end 
noncontiguous_partitioner_update_ghost_values_start 

NoncontiguousPartitioner::update_values.

noncontiguous_partitioner_update_ghost_values_end 
compute_union 
remote_point_evaluation 
fine_dof_handler_view_reinit 
distributed_compute_point_locations 
affine_constraints_make_consistent_in_parallel_0 
affine_constraints_make_consistent_in_parallel_1 

Definition at line 51 of file mpi_tags.h.