FETIDP: fix for benign subspace trick. Now FETI-DP works with the benign subspace trick

FETIDP: fix for change of basis

FETIDP with the benign subspace still does not properly work;
the eigenvalues of the preconditioned operators are the same as BDDC, but the solution from FETI-DP is w

FETIDP: fix for change of basis

FETIDP with the benign subspace still does not properly work;
the eigenvalues of the preconditioned operators are the same as BDDC, but the solution from FETI-DP is wrong on the pressures (modulo a subdomain constant, so it should be easy to fix)

show more ...

Merge branch 'master' into stefano_zampini/feature-pcbddc-saddlepoint

Conflicts:
src/ksp/pc/impls/bddc/bddc.c
src/ksp/pc/impls/bddc/bddcprivate.c
src/mat/impls/aij/seq/mkl_pardiso/mkl_pardiso.c

Merge branch 'stefano_zampini/fix-chebyshev-rhslocked' into maint

Merge branch 'stefano_zampini/fix-bddcnullspace'

Merge branch 'stefano_zampini/fix-chebyshev-rhslocked'

Added PCPreSolveChangeRHS as suggested by Barry.

the method is declared as intern in kspimpl.h to not introduce a new API
Implemented as a composable function to not change pcimpl.h

PCBDDC: remove PCBDDCSetNullSpace. Nullspace info is now local and should be attached to the local Neumann matrix

PCBDDC; start removing code for bddcnullspace

Merge branch 'master' into stefano_zampini/feature-pcbddc-saddlepoint

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

PCBDDC: WIP added support for mixed boundary conditions (works for Darcy). Need to check if it works for Stokes.

Multilevel extension does not currently work with mixed BC or a mix of benign and non

PCBDDC: WIP added support for mixed boundary conditions (works for Darcy). Need to check if it works for Stokes.

Multilevel extension does not currently work with mixed BC or a mix of benign and non-benign subdomains. Need to pass the B0 information to coarser levels anyway, because we need to properly compute the no-net-flux condition at coarser levels.

show more ...

PCBDDC: use exact BDDC without dirichlet trick when computing the initial guess

PCBDDC: add pcis->n and pcbddc->benign_n to the statistics

raise error if adaptive selection is incompatible with pointwise scaling

some uses of PetscTryMethod() actually require that the

object is correct and the function returns an appropriate value.
Change those calls to PetscUseMethod().

Fixes for issues found with clang static analyzer,
mostly set but unused variables

Time: 1 hour

PCBDDC: fixes from static analyzer nightly build

PCBDDC: added KSPPIPECGRR to the list of methods support the exact dirichlet trick

PCBDDC: optimizations for the multilevel BDDC with benign subspace and CHEBYSHEV coarse solvers

Most of the times, the computation of the benign correction is not needed.
Added benign_have_null, ben

PCBDDC: optimizations for the multilevel BDDC with benign subspace and CHEBYSHEV coarse solvers

Most of the times, the computation of the benign correction is not needed.
Added benign_have_null, benign_need_correction, so that we can decide at command line to not compute the benign correction for Stokes

show more ...

PCBDDC: improve PCView_BDDC

Added control for the explicit change of variables for pressure dofs

PCBDDC: add runtime options to control dirichlet dofs elimination in presolve

Most the time removing dir dofs does not have an impact on the global convergence

PCBDDC: remove faster deluxe switch

PCBDDC: remove the horrible logic around the remap of the coarse problem

the code now is more readable, maintanable. In case of multilevel, it is also faster, since one shift is avoided.
when the co

PCBDDC: remove the horrible logic around the remap of the coarse problem

the code now is more readable, maintanable. In case of multilevel, it is also faster, since one shift is avoided.
when the coarse matrix has to be assembled, it still uses MatISSubassemble followed by MatISGetMPIXAIJ

fixed some bugs in MatISSubassemble and MatISGetSubassemblingPattern, with preliminary checks added

show more ...

PCBDDC: allow processes to pass 0 nvtxs and void arrays

PCBDDC: make change_primal_sub a member of sub_schurs

we need it during adaptive selection
sub_schurs->change and sub_schurs->chane_primal_sub are created when needed

PCBDDC: Added support for adaptive selection of constraints with pointwise scalings