DMPlex: Fix allocation in DMPlexMatSetClosureRefined()

Merge branch 'knepley/feature-plex-fas'

* knepley/feature-plex-fas: (56 commits)
DMPlex: Removed unused variable
DMPlex: Fix for C++ build, small fix for PetscFE
DMPlex: Removed leak from Tria

Merge branch 'knepley/feature-plex-fas'

* knepley/feature-plex-fas: (56 commits)
DMPlex: Removed unused variable
DMPlex: Fix for C++ build, small fix for PetscFE
DMPlex: Removed leak from Triangle
SNES ex12: Two working unstructured FAS solves for the unregularized p-Laplacian
DMPlex: Cleaned up some memory issues
DMPlex: Implement injection - I don't think this should be required
DMPlex: DM interpolation requires the scaling vector for FAS
DMPlex: Can now create a map from coarse to fine vertices - This is currently necessary for FAS, which I think is wrong
DMPlex: Turn off Fortran stub
DMPlex: Fix DMPlexMatSetClosureRefined() - Reject duplicate fine points - Screen out point with dof = 0 - Fix the allocation for fine points
DMPlex: My brain did not work for several days - I have fixed the interpolation code - I think I can reuse that other code for something else
DMPlex: Fix setting Plex options on automatically refined meshed
SNES ex12/31/52/62/72: Use new SNESComputeJacobian() interface
DMPlex: Remove MatStructure from DMPlexComputeJacobianFEM() - Add missing declarations
Sys Logging: Fixes for log detail - Must preserve reduction count - Output objects and memory - Output stage summary
SNES ex12: Fixing things for multilevel - Need to set PetscFE on all coarse grids - Need application context for interpolator construction
SNES ex12: Use PetscFESetUp(), and fix quadrature ref handling
SNES ex52/62: Use PetscFESetUp(), and fix quadrature ref handling
DMPlex ex3: Uses automatic section creation, PetscFESetUp(), and fix quadrature ref handling
DMPlex: Fixing interpolator construction - Fixed definition of rectangular sizes - Use PetscFERefine() - Use special MatSet
...

show more ...

Merge branch 'knepley/fix-plex-submesh-edge'

* knepley/fix-plex-submesh-edge:
DMPlex: Removed unused variable in refinement
DMPlex: Also mark hybrid cells with unsplit vertices
DMPlex: Docs fo

Merge branch 'knepley/fix-plex-submesh-edge'

* knepley/fix-plex-submesh-edge:
DMPlex: Removed unused variable in refinement
DMPlex: Also mark hybrid cells with unsplit vertices
DMPlex: Docs for preallocation
DMPlex: Fix for parallel refinement - PetscSF needs the leaves in sorted local order for it to get SFGather correct
DMPlex: Label hybrid cells with unsplit edges during submesh creation - I don't like this completely since we have to use -(cell number) to avoid aliasing vertices in the original mesh
DMPlex: Fixed submesh boundary in 3D - Correct support for unsplit edges - Correct support for hybrid faces from unsplit edges - Added checks
DMPlex: Now DMPlexCheclSymmetry() also checks multiple cone/support entries
DMPlex: Fix unsplit vertex support in submeshes - Assumed only a single split edge, which is true in 2D
DMPlex: Fixing marking of unsplit vertices - Check for presence in label was wrong (has to be exactly -1) - Check for incidence was wrong (was also getting unsplit marker)

show more ...

DMPlex: Removed unused variable in refinement

DMPlex: Also mark hybrid cells with unsplit vertices

DMPlex: Removed unused variable

Merge branch 'knepley/fix-plex-orient-parallel'

* knepley/fix-plex-orient-parallel:
DMPlex: Added braindead simple parallel orientation solver - This gathers the graph to proc 0, solves, and scatt

Merge branch 'knepley/fix-plex-orient-parallel'

* knepley/fix-plex-orient-parallel:
DMPlex: Added braindead simple parallel orientation solver - This gathers the graph to proc 0, solves, and scatters the answer - Need to store neighbor ranks and matches for each neighbor - This should be replaced with a parallel BFS
DMPlex: For 1D orientation, you have to use cone position rather than face orientation
DMPlex: Fixed some small mistakes - Initialize var - Declare at the top of the block - Added cone size call
DMPlex: Added parallel stage to DMPlexOrient() - The coarse solver is not implemented, so only works for p=2
PetscBT: Added PetscBTNegate()

show more ...

DMPlex: Docs for preallocation

DMPlex: Fix for parallel refinement
- PetscSF needs the leaves in sorted local order for it to get SFGather correct

DMPlex: Fix for C++ build, small fix for PetscFE

DMPlex: Removed leak from Triangle

DMPlex: Cleaned up some memory issues

DMPlex: Implement injection
- I don't think this should be required

DMPlex: DM interpolation requires the scaling vector for FAS

DMPlex: Can now create a map from coarse to fine vertices
- This is currently necessary for FAS, which I think is wrong

DMPlex: Label hybrid cells with unsplit edges during submesh creation
- I don't like this completely since we have to use -(cell number) to avoid aliasing vertices in the original mesh

DMPlex: Turn off Fortran stub

DMPlex: Fix DMPlexMatSetClosureRefined()
- Reject duplicate fine points
- Screen out point with dof = 0
- Fix the allocation for fine points

DMPlex: My brain did not work for several days
- I have fixed the interpolation code
- I think I can reuse that other code for something else

DMPlex: Fix setting Plex options on automatically refined meshed

Merge branch 'master' into knepley/feature-plex-fas

* master: (863 commits)
Revert "mpich: use latest mpich tarball with gcov but use '--enable-coverage' mpich configure option."
MatCreateTransp

Merge branch 'master' into knepley/feature-plex-fas

* master: (863 commits)
Revert "mpich: use latest mpich tarball with gcov but use '--enable-coverage' mpich configure option."
MatCreateTranspose: fix for lazy block size
mpich: use latest mpich tarball with gcov but use '--enable-coverage' mpich configure option.
configure: add FC_DEFINE_FLAG to petscvariables
added a citation for pounders, Stefan finally came through
update to latest netcdf fixes bug that did not allow building on recent Mac OS X
DMPlex: Fixes for CGNS - Fixed loop bounds since CGNS uses closed intervals - Bug with mixed cell types - Typo with coordinates
final updates from nightly tests
fix for countcfunction
removed outdated docs and comments on MatStructure usage removed SAME_PRECONDITIONER since no longer used removed unneed use of MatStructure in Schur complement matrix
removed MatStructure arguments from Tao
fixes for Sundials and fun3d cannot use MPIU_SUM for integer arrays
inplace ILU factorization must update state of inplace matrix to prevent infinite recursion of updating PC
fixes for BDDC by the way not having BDDC always in the testing is going to lead to trouble
the compute Jacobian functions for SNES and TS now do not take a MatStructure flag
KSPSetOperators() no longer has the MatStructure argument. The Mat objects now track that information themselves. Use KPS/PCSetReusePreconditioner() to prevent the recomputation of the preconditioner if the operator changed in the way that SAME_PRECONDITIONER did with KSPSetOperators()
tracks when Mat's have a change in nonzero structure this information can be used by PC to determine how preconditioners must be rebuilt currently only does an error check if the user said same nonzero pattern but actually changed the pattern
completed cleanup of removal of matrix pointers to compute jacobian routines
makefile: Added 'make includegraph' to create an inclusion dependency graph.
Contrib: Updated include-graph.sh script to ignore *.hh files
...

Conflicts:
src/dm/dt/interface/dt.c

show more ...

DMPlex: Remove MatStructure from DMPlexComputeJacobianFEM()
- Add missing declarations

DMPlex ex3: Uses automatic section creation, PetscFESetUp(), and fix quadrature ref handling

DMPlex: Fixing interpolator construction
- Fixed definition of rectangular sizes
- Use PetscFERefine()
- Use special MatSet

DMPlex: Made a special MatSet for interpolation
- Maybe this can be generalized