On Mon, 15 Nov 2010, Saurabh Srivastava wrote:
I applied the patch but as you guessed before it didnt solve the
problem as patch pertains to another possible bug in AMR/parallel
..However I have aptly modified ex13 to reproduce the error I
complained about. Essentially I added added a few lines to remesh
the region and call reint() function.
Maybe now its trivial for you to trace the bug...
Wait, wait, wait - the bug here is that you're completely wiping out
the old mesh! EquationSystems::reinit() will project solutions to
the new DoF data on Mesh-A from the old DoF data on Mesh-A before the
previous refinement or coarsening (with at most one step of either on
each element). If you need to wipe out your whole mesh and regenerate
a new one from scratch, then the system vectors are going to be out of
sync with the mesh, the old indexing information for those vectors is
going to be gone, and the System isn't going to know what to do with
any of that.
I'll cc: this (along with your ex13.C) to libmesh-users in case anyone
else has a better idea, but I think the only safe way to do this would
be to generate new System and EquationSystems objects as well, then
manually hand a MeshFunction on the old EquationSystems to
System::project_vector on the new objects.
---
Roy
------------------------------------------------------------------------------
Centralized Desktop Delivery: Dell and VMware Reference Architecture
Simplifying enterprise desktop deployment and management using
Dell EqualLogic storage and VMware View: A highly scalable, end-to-end
client virtualization framework. Read more!
http://p.sf.net/sfu/dell-eql-dev2dev
_______________________________________________
Libmesh-users mailing list
Libmesh-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/libmesh-users
