[cig-commits] [commit] master: Leave a TODO. (4f595a4)

cig_noreply at geodynamics.org cig_noreply at geodynamics.org
Tue Jun 24 13:13:54 PDT 2014


Repository : https://github.com/geodynamics/aspect

On branch  : master
Link       : https://github.com/geodynamics/aspect/compare/6b7c137fa297caa534f0c498bb594d05ed012a95...dd3794be780d4879f029fcef1efb957c61665109

>---------------------------------------------------------------

commit 4f595a4c640b8b4b595e0952ab1abe436683dc7f
Author: Wolfgang Bangerth <bangerth at math.tamu.edu>
Date:   Sun Jun 22 18:27:46 2014 -0500

    Leave a TODO.


>---------------------------------------------------------------

4f595a4c640b8b4b595e0952ab1abe436683dc7f
 source/simulator/core.cc | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/source/simulator/core.cc b/source/simulator/core.cc
index e2656a7..7dba2db 100644
--- a/source/simulator/core.cc
+++ b/source/simulator/core.cc
@@ -1265,9 +1265,17 @@ namespace aspect
 
         std::vector<LinearAlgebra::Vector *> system_tmp (1);
         system_tmp[0] = &distributed_mesh_vertices;
+
         freesurface_trans->interpolate (system_tmp);
 
-	free_surface->mesh_constraints.distribute (distributed_mesh_vertices);
+//TODO: as explained in the documentation of SolutionTransfer, after interpolate()
+// we may have a non-conforming solution at hanging nodes. so we would like to 
+// call distribute() as in this line:	
+//	free_surface->mesh_constraints.distribute (distributed_mesh_vertices);
+// This doesn't seem to work, however, and leads to an invalid mesh vertex
+// field. Maybe because free_surface->mesh_constraints can only be applied to
+// the vertex increments field and not the vertex field!?
+	
         free_surface->mesh_vertices = distributed_mesh_vertices;
       }
 



More information about the CIG-COMMITS mailing list