MATIS: remove mapping member of MATIS. Use layout mappings instead

Pass the reuse preconditioner flag down to children of PCCOMPOSITE.

PCHYPRE: add option for setting the projection frequency on the compatible subspace in case of zero-conductivity regions

PCHYPRE: remove components in the kernel in case of zero conductivity regions

PCHYPRE: in case of zero conductivities in the whole domain, we need to pass null to HYPRE_AMSSetBetaPoisson

Merge remote-tracking branch 'petsc/master' into pcgasm-increaseoverlap

added PCCompositeGetNumberPC() to provide number of PC in a composite PC

Provided-by: Andrew Cramer <a.cramer@uq.edu.au>

fixing EM petsc params

added PREONLY default for coarse grid solver in GAMG

Fix incorrect indices in PCShell Fortran wrappers.

Some indices overlapped, and in some cases, the size of the allocated
function point array was too small.

remove nonexistent -pc_mg_monitor from documentation

PCBDDC: fixes when reusing the preconditioner with adaptive selection

PC: Fixed propagation of DM by PCCOMPOSITE
Reported-by: Lawrence Mitchell <lawrence.mitchell@imperial.ac.uk>

PCBDDC: fix typos in the specification of MPI operation

Merge remote-tracking branch 'petsc/master' into mat-increaseoverlap

Merge branch 'stefano_zampini/pcbddc-adaptive-selection'

PCBDDC: update call to PCBDDCNullSpaceAssembleCorrection

disable propagation of null space to coarse levels since it is untested

PCBDDC: forgot to add the header file

PCBDDC: Fix for nullspace

the nullspace branch was not tested. This fixes a bug

Merge branch 'stefano_zampini/pcbddc-adaptive-selection'

Conflicts:
src/ksp/pc/impls/bddc/bddcprivate.c
src/ksp/pc/impls/bddc/bddcscalingbasic.c

Merged in PR312: karpeev/ksp-pcgasm-overhaul.

PCBDDC: better to use size_active_schur as a guard

PCBDDC: solve some errors in subschurs appearing when not restricting the work on active processes

PCBDDC: workaround when reusing the coarse solver with different sizes

With constraints adaptivity in place, it could happen that coarse problems generated in successive calls can have different siz

PCBDDC: workaround when reusing the coarse solver with different sizes

With constraints adaptivity in place, it could happen that coarse problems generated in successive calls can have different sizes
Since PCBDCD does not have a reset method so far, lets destroy the object.

show more ...

Custom Fortran stub pcfieldsplitgetsubksp_() did not free array containing the KSPs

Reported-by: Mark Adams <mfadams@lbl.gov>