[cig-commits] r18459 - short/3D/PyLith/trunk

[brad at geodynamics.org]

Author: brad
Date: 2011-05-24 17:05:19 -0700 (Tue, 24 May 2011)
New Revision: 18459

Modified:
   short/3D/PyLith/trunk/TODO
Log:
Updated TODO.

Modified: short/3D/PyLith/trunk/TODO
===================================================================
--- short/3D/PyLith/trunk/TODO	2011-05-24 23:37:01 UTC (rev 18458)
+++ short/3D/PyLith/trunk/TODO	2011-05-25 00:05:19 UTC (rev 18459)
@@ -7,6 +7,10 @@
   (1) cygwin
       + packaging
 
+----------------------------------------
+BUGS
+----------------------------------------
+
 * Friction
   rate and state friction - sliding test
   predictor/corrector scheme?
@@ -18,6 +22,20 @@
 
     Check refinement in parallel thoroughly.
 
+  BUG: multiple faults and refinement?
+
+    cd benchmarks/static/performance/solvertest
+    pylith hex8.cfg \
+      --mesh_generator.refiner=pylith.topology.RefineUniform \
+      --nodes=2
+
+    src/cig/pylith/libsrc/pylith/topology/RefineVol8Face4Edges2.cc:505: void ALE::RefineVol8Face4Edges2::overlapAddNewVertices(const ALE::Obj<ALE::IMesh<int, double> >&, const ALE::MeshOrder&, const ALE::Obj<ALE::IMesh<int, double> >&, const ALE::MeshOrder&): Assertion `remotePoint >= 0' failed.
+
+
+----------------------------------------
+MISSING FEATURES
+----------------------------------------
+
 * Output to HDF5 files. [BRAD and MATT]
 
   (2) Add time dataset to vertex_fields and cell_fields. [BRAD]
@@ -26,8 +44,6 @@
 
 * Field split.
 
-  Need to add the ability to add null space (subduction/step03).
-
     Add flag to material [default is false] for creating null vector
     When setting up solver, create null vector, and pass it to KSP
 
@@ -46,7 +62,6 @@
     + Are material models consistent with governing equations discussion?
 
 * Fault preconditioner [BRAD]
-    + Add code make null space for a given material and add to KSP
 
   FaultCohesiveLagrange::calcPreconditioner() [need unit test]