DiFfRG::FEOutput< dim, VectorType >::SerialMirrorDoFs Struct Reference

Serial mirror of the solution DoFHandler, plus the map from its dof numbering to ours. More...

Public Attributes
std::unique_ptr< DoFHandler< safe_dim > >	dof_handler
std::vector< types::global_dof_index >	to_mirror
boost::signals2::scoped_connection	connection
bool	stale = true

Detailed Description

template<uint dim, typename VectorType>
struct DiFfRG::FEOutput< dim, VectorType >::SerialMirrorDoFs

Serial mirror of the solution DoFHandler, plus the map from its dof numbering to ours.

The numbering genuinely differs: a DoFHandler on a parallel::shared::Triangulation renumbers so that each rank owns a contiguous range, so global index i means different nodes in the two handlers as soon as there is more than one rank. (At one rank they coincide, which is exactly why single-rank distributed runs looked fine.) The permutation is built cell by cell – the two triangulations are structurally identical, so cells match by level and index.

Member Data Documentation

connection

template<uint dim, typename VectorType >

boost::signals2::scoped_connection DiFfRG::FEOutput< dim, VectorType >::SerialMirrorDoFs::connection

dof_handler

template<uint dim, typename VectorType >

std::unique_ptr<DoFHandler<safe_dim> > DiFfRG::FEOutput< dim, VectorType >::SerialMirrorDoFs::dof_handler

stale

template<uint dim, typename VectorType >

bool DiFfRG::FEOutput< dim, VectorType >::SerialMirrorDoFs::stale = true

to_mirror

template<uint dim, typename VectorType >

std::vector<types::global_dof_index> DiFfRG::FEOutput< dim, VectorType >::SerialMirrorDoFs::to_mirror

---

The documentation for this struct was generated from the following file:

• /home/runner/work/DiFfRG_current/DiFfRG_current/DiFfRG/include/DiFfRG/discretization/data/fe_output.hh