Mesh

The mesh object represents the mesh on a given processor/MPI rank. There is one mesh per processor. The Mesh object owns all MeshBlocks on a given processor.

Looping over MeshBlocks

MeshBlocks are stored in a variable Mesh::block_list, which is an object of type

using BlockList_t = std::vector<std::shared_ptr<MeshBlock>>;

and so to get the predicted time step for each mesh block, you can call:

for (auto &pmb : pmesh->block_list) {
    std::cout << pmb->NewDt() << std::endl;
}

where pmesh is a pointer to a Mesh object. This paradigm may appear, for example, in an application driver.

Mesh Functions Controlling Evolulution and Diagnostics

Some mesh functions are stored as std::function members and are called by the EvolutionDriver. They may be overridden by the application by reassigning the default values of the std::function member.

• PreStepUserWorkInLoop(Mesh*, ParameterInput*, SimTime const&) is called to perform mesh-wide work before the time-integration advance (the default is a no-op)

• PostStepUserWorkInLoop(Mesh*, ParameterInput*, SimTime const&) is called to perform mesh-wide work after the time-integration advance (the default is a no-op)

• PreStepUserDiagnosticsInLoop(Mesh*, ParameterInput*, SimTime const&) is called to perform diagnostic calculations and/or edits before the time-integration advance. The default behavior calls to each package’s (StateDesrcriptor’s) PreStepDiagnostics method which, in turn, delegates to a std::function member that defaults to a no-op.

• PostStepUserDiagnosticsInLoop(Mesh*, ParameterInput*, SimTime const&) is called to perform diagnostic calculations and/or edits after the time-integration advance. The default behavior calls to each package’s (StateDesrcriptor’s) PreStepDiagnostics method which, in turn, delegates to a std::function member that defaults to a no-op.

• UserMeshWorkBeforeOutput(Mesh*, ParameterInput*, SimTime const&) is called to perform mesh-wide work immediately before writing an output (the default is a no-op). The most likely use case is to fill derived fields with updated values before writing them out to disk (or passing them to Ascent for in-situ analysis).

Multi-grid Grids Stored in Mesh

If the parameter parthenon/mesh/multigrid is set to true, the Mesh constructor and AMR routines populate both std::vector<LogicalLocMap_t> Mesh::gmg_grid_locs and std::vector<BlockList_t> gmg_block_lists, where each entry into the vectors describes one level of the of the geometric multi-grid (GMG) mesh. For refined meshes, each GMG level only includes blocks that are at a given logical level (starting from the finest logical level on the grid and including both internal and leaf nodes in the refinement tree) as well as leaf blocks on the next coarser level that are neighbors to finer blocks, which implies that below the root grid level the blocks may not cover the entire mesh. For levels above the root grid, blocks may change shape so that they only cover the domain of the root grid. Note that leaf blocks may be contained in multiple blocklists, and the lists all point to the same block (not a separate copy). To be explicit, when parthenon/mesh/multigrid is set to true blocks corresponding to all internal nodes of the refinement tree are created, in addition to the leaf node blocks that are normally created.

GMG Implementation Note: The reason for including two levels in the GMG block lists is for dealing with accurately setting the boundary values of the fine blocks. Convergence can be poor or non-exististent if the fine-coarse boundaries of a given level are not self-consistently updated (since the boundary prolongation from the coarse grid to the fine grid also depends on interior values of the fine grid that are being updated by a smoothing operation). This means that each smoothing step, boundary communication must occur between all blocks corresponding to all internal and leaf nodes at a given level in the tree and with all leaf nodes at the next coarser level which abut blocks at the current level. Therefore, the GMG block lists contain blocks at two levels, but smoothing operations etc. should usually only occur on the subset of those blocks that are at the fine level.

To work with these GMG levels, MeshData objects containing these blocks can be recovered from a Mesh pointer using

auto &md = pmesh->gmg_mesh_data[level].GetOrAdd(level, "base", partition_idx);

This MeshData will include blocks at the current level and possibly some blocks at the next coarser level. Often, one will only want to operate on blocks on the finer level (the coarser blocks are required mainly for boundary communication). To make packs containing only a subset of blocks from a GMG MeshData pointer md, one would use

int nblocks = md->NumBlocks();
std::vector<bool> include_block(nblocks, true);
for (int b = 0; b < nblocks; ++b)
  include_block[b] =
      (md->grid.logical_level == md->GetBlockData(b)->GetBlockPointer()->loc.level());

auto desc = parthenon::MakePackDescriptor<in, out>(md.get());
auto pack = desc.GetPack(md.get(), include_block);

In addition to creating the LogicalLocation and block lists for the GMG levels, Mesh fills neighbor arrays in MeshBlock for intra- and inter-GMG block list communication (i.e. boundary communication and internal prolongation/restriction, respectively). Communication within and between GMG levels can be done by calling boundary communication routines with the boundary tags gmg_same, gmg_restrict_send, gmg_restrict_recv, gmg_prolongate_send, gmg_prolongate_recv (see Boundary Communication Tasks).

Parthenon Forest Implementation Details

Parthenon implements AMR via a forest of octrees type approach.

• A LogicalLocation is defined by the index of the octree location is contained in combined with the refinement level and location of the block within the octree. The location within the tree can be converted to a unique Morton number, and the combined Morton number and tree index can provide a unique index for the location. We use this fact to make LogicalLocation hashable so that it can be used with std::map and other hashed containers.

• Unlike in the Athena++ MeshBlockTree implementation that was historically used by Parthenon, the entire logical index space of the