Add a test that failed before.

diff --git a/tests/mpi/hp_choose_p_over_h.cc b/tests/mpi/hp_choose_p_over_h.cc
new file mode 100644 (file)
index 0000000..48c3ac1
--- /dev/null
+++ b/tests/mpi/hp_choose_p_over_h.cc
@@ -0,0 +1,103 @@
+// ------------------------------------------------------------------------
+//
+// SPDX-License-Identifier: LGPL-2.1-or-later
+// Copyright (C) 2019 - 2024 by the deal.II authors
+//
+// This file is part of the deal.II library.
+//
+// Part of the source code is dual licensed under Apache-2.0 WITH
+// LLVM-exception OR LGPL-2.1-or-later. Detailed license information
+// governing the source code and code contributions can be found in
+// LICENSE.md and CONTRIBUTING.md at the top level directory of deal.II.
+//
+// ------------------------------------------------------------------------
+
+
+
+// verify hp::Refinement::choose_p_over_h() on strongly distributed meshes
+
+
+#include <deal.II/distributed/tria.h>
+
+#include <deal.II/dofs/dof_handler.h>
+
+#include <deal.II/fe/fe_q.h>
+
+#include <deal.II/grid/filtered_iterator.h>
+#include <deal.II/grid/grid_generator.h>
+
+#include <deal.II/hp/fe_collection.h>
+#include <deal.II/hp/refinement.h>
+
+#include "../tests.h"
+
+
+
+void
+test()
+{
+  constexpr unsigned int dim = 2;
+
+  // Setup distributed triangulation.
+  parallel::distributed::Triangulation<dim> tr(MPI_COMM_WORLD);
+  GridGenerator::hyper_cube(tr, -1, 1);
+  tr.refine_global(1);
+
+  tr.begin_active()->set_refine_flag();
+  tr.execute_coarsening_and_refinement();
+
+  // For two dimensions in this particular scenario, each quadrant in the
+  // coordinate system will be assigned to single MPI process.
+  //
+  // See also the corresponding output file and this ASCII sketch on how p4est
+  // decided to partition this mesh with four MPI processes.
+  //
+  //  +---+---+
+  //  |   |   |
+  //  | 3 | 3 |
+  //  |   |   |
+  //  +-+-+---+
+  //  |1|1|   |
+  //  +-+-+ 2 |
+  //  |1|1|   |
+  //  +-+-+---+
+
+  // Set h-coarsen and p-refinement flags on the coarser cells.
+  DoFHandler<dim> dh(tr);
+  for (const auto &cell : dh.active_cell_iterators_on_level(1) |
+                            IteratorFilters::LocallyOwnedCell())
+    {
+      cell->set_coarsen_flag();
+      cell->set_future_fe_index(1);
+    }
+
+  // Setup FEs.
+  hp::FECollection<dim> fes;
+  for (unsigned int d = 1; d <= 2; ++d)
+    fes.push_back(FE_Q<dim>(d));
+  dh.distribute_dofs(fes);
+
+  // Check whether choose_p_over_h() finishes.
+  // (It triggered an assertion in the past.)
+  hp::Refinement::choose_p_over_h(dh);
+
+  // Check the new flags.
+  for (const auto &cell :
+       dh.active_cell_iterators() | IteratorFilters::LocallyOwnedCell())
+    deallog << cell->id().to_string()
+            << ": future_fe=" << cell->future_fe_index()
+            << " coarsen=" << cell->coarsen_flag_set() << std::endl;
+
+  deallog << "OK" << std::endl;
+}
+
+
+
+int
+main(int argc, char *argv[])
+{
+  Utilities::MPI::MPI_InitFinalize mpi_initialization(argc, argv, 1);
+  MPILogInitAll                    log;
+
+  test();
+}

diff --git a/tests/mpi/hp_choose_p_over_h.with_p4est=true.mpirun=4.output b/tests/mpi/hp_choose_p_over_h.with_p4est=true.mpirun=4.output
new file mode 100644 (file)
index 0000000..3f2cf89
--- /dev/null
+++ b/tests/mpi/hp_choose_p_over_h.with_p4est=true.mpirun=4.output
@@ -0,0 +1,18 @@
+
+DEAL:0::OK
+
+DEAL:1::0_2:00: future_fe=0 coarsen=0
+DEAL:1::0_2:01: future_fe=0 coarsen=0
+DEAL:1::0_2:02: future_fe=0 coarsen=0
+DEAL:1::0_2:03: future_fe=0 coarsen=0
+DEAL:1::OK
+
+
+DEAL:2::0_1:1: future_fe=1 coarsen=0
+DEAL:2::OK
+
+
+DEAL:3::0_1:2: future_fe=1 coarsen=0
+DEAL:3::0_1:3: future_fe=1 coarsen=0
+DEAL:3::OK
+