Merge branch 'knepley/fix-distribute-check' [PR #1152]

Plex: Improved Gmsh error

Plex: Fixed argument check

Merge branch 'barry/cleanout-old-test-harness'

Plex: Set flag for uniform refinement

Plex: Must copy levels for refined/coarsened grids

Merge branch 'knepley/feature-dm-nearnullspace' [PR #1128]

* knepley/feature-dm-nearnullspace:
DM: Added interface for (near) nullspace constructors
DM: Now DMCreateMatrix() adds a (near) nullspace

Merge branch 'knepley/feature-dm-nearnullspace' [PR #1128]

* knepley/feature-dm-nearnullspace:
DM: Added interface for (near) nullspace constructors
DM: Now DMCreateMatrix() adds a (near) nullspace if we have only 1 field and the constructor set
Plex+FEM: Fix corner case where space is smaller than the proposed number of null modes
Plex+FEM: Fix another corner case in CreateRigidBody
SNES ex17: Now use rigid body modes as the near null space

show more ...

Plex: Added test for parallel partitioning

Merge branch 'knepley/feature-section-field-major' [PR #1131]

* knepley/feature-section-field-major:
PetscSection: Cleanup point iteration in SetUp()
Section: Add a point major flag to allow differe

Merge branch 'knepley/feature-section-field-major' [PR #1131]

* knepley/feature-section-field-major:
PetscSection: Cleanup point iteration in SetUp()
Section: Add a point major flag to allow different dog orderings
Does not do anything right now
Section: Added PetscSectionSetFromOptions()
Plex: When Plex creates a section automatically, call SetFromOptions
Could condition this on whether the DM has been set from options
Section: During SetUp(), change dog ordering based on the pointMajor flag
Plex ex27: Check Section dof ordering

show more ...

DMPlexView_ASCII: use calloc to prevent valgrind errors

hybsizes are accessed by all processes, but only rank 0 in the communicator has valid values

DMDA: allow users to customize the HO coordinates (as in DMPLEX)

added test

Plex+FEM: Must have all prods agree how many rigid body modes to make

Test harness does not have requires: TODO it has TODO: reason

Commit-type: bug-fix, testing-fix, style-fix

clean out the last of the old test harness artifacts

Commit-type: style-fix

autogeneration should work

the only problematic argument could be MatNullSpace which is a struct

Plex ex27: Give mesh a name

Plex ex3: Null space may have no vectors

Plex+FEM: Fixed destroy for nullspace

Merge branch 'master' into knepley/feature-dm-nearnullspace

* master: (219 commits)
petsc4py: update with VecScatter API changes
remove duplicate code
petsc4py: update to version 3.10.0
fixe

Merge branch 'master' into knepley/feature-dm-nearnullspace

* master: (219 commits)
petsc4py: update with VecScatter API changes
remove duplicate code
petsc4py: update to version 3.10.0
fixed typos in documentation
fixed typos in documentation
Fix some remaining Register()s to call InitializePackage()
Remove memory leaks of XXXRegister() is called but XXX object is never used
Add to PCTFS manual page that it does not work for complex numbers
Stray , in manual page that should be .
Only register MPI3 VecScatter if they are built
clarification of KSPSetComputeInitialGuess()
Add missing makefiles for VecScatter implementations
Changed VecScatterCreate() to take a communicator and return scatter as other first class PETSc objects do. Changed old VecScatterCreate() to VecScatterCreateWithData()
Added VecScatterSetUp() and converted implementations to use it to compute scatter information
Pull "MPI1" and "MPI3" specific code out of common VecScatter impls code This allows future implementations to use the common code also without changing the common code
Seperate VecScatter implementation code into three subdirectories seq, mpi1, and mpi3
doc: fix author list in manual wrt 6402aee
doc: fix LOCDIR wrt broken URL from MatHYPRESetPreallocation.html
ifort: ifx compile error with --with-mpi=0
changes/310.html: merged two TAO sections
...

Conflicts:
src/dm/interface/dm.c

show more ...

Merge branch 'barry/refactor-vecscatter-2'

Changed VecScatterCreate() to take a communicator and return scatter as other first class PETSc objects do.
Changed old VecScatterCreate() to VecScatterCreateWithData()

Completed basic refactorizati

Changed VecScatterCreate() to take a communicator and return scatter as other first class PETSc objects do.
Changed old VecScatterCreate() to VecScatterCreateWithData()

Completed basic refactorization of VecScatter code, ready for new implementations to be added.

(There is still some code duplication between the VecScatter MPI1 and MPI3 code but it can be fixed at anytime without requiring changes to the VecScatter API.

Part 5 of a refactorization of VecScatter to be treated as a first class PETSc object

Commit-type: style-fix

show more ...

DMPlexDistribute: fix man page

dmplexdistribute_: do not CHKFORTRANNULLOBJECT(dmDist)

This parameter is an output argument and this line causes
dm_label_tutorials-ex1_0 to fail. Also remove the special NULL
handling for the outp

dmplexdistribute_: do not CHKFORTRANNULLOBJECT(dmDist)

This parameter is an output argument and this line causes
dm_label_tutorials-ex1_0 to fail. Also remove the special NULL
handling for the output because it is now redundant.

Commit-type: bug-fix

show more ...

Merge remote-tracking branch 'origin/barry/change-fortran-null-fixup'

Removed use of hardwired -1 to represent PETSC_NULL_XXX object, since the null value is now 0.

These cases were accidently left in place when PETSC_NULL_XXX was changed from -1 to 0.

Commit-type: b

Removed use of hardwired -1 to represent PETSC_NULL_XXX object, since the null value is now 0.

These cases were accidently left in place when PETSC_NULL_XXX was changed from -1 to 0.

Commit-type: bug-fix
Reported-by: Fabian Jakub <Fabian.Jakub@physik.uni-muenchen.de>

show more ...