some uses of PetscTryMethod() actually require that the

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

Merge branch 'pr426/dmay/dmay/pctelescope-mods/master'

Merge branch 'barry/fix-mg-coarse-redundant'

Merge branch 'jed/fix-ml-visibility'

a couple of fixes suggested by clang analyzer
add missing error checking
minor formatting fixes

a few more set but not used variables detected by clang static analyzer

Time: . 1 hours

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

Time: 1 hour

PCBDDC: fixes from static analyzer nightly build

PCTelescope - fix for static analysis warnings:
* I think that this commit will remove the last warning about indexing a null pointer

Because PCMGSetLevels() called PCRedundantGetKSP() which forced one redundant PC
per process (i.e. made -mg_coarse_pc_redundant_number <n> not valid)
I have added PCFactorSetShiftType_Redundant() tha

Because PCMGSetLevels() called PCRedundantGetKSP() which forced one redundant PC
per process (i.e. made -mg_coarse_pc_redundant_number <n> not valid)
I have added PCFactorSetShiftType_Redundant() that allows setting the shift type
but delaying the creation of the inner KSP of PCREDUNDANT until after the user
has the chance to set the redundant_number

Reported-by: Dave May <dave.mayhem23@gmail.com>

show more ...

PCTelescope - Fixes reported by static analysis:
* Initialize variables which are otherwise set inside if (isActiveRank(sred->psubcomm)){} statements

few more clang static analyzer fixes

Merge branch 'pr421/alex/feature-veccuda'

* pr421/alex/feature-veccuda
The rationale of this pull request is to have GPU-enabled vectors purely based on CUDA,
and with the possibility of placing a u

Merge branch 'pr421/alex/feature-veccuda'

* pr421/alex/feature-veccuda
The rationale of this pull request is to have GPU-enabled vectors purely based on CUDA,
and with the possibility of placing a user-provided array on the GPU side.

Conflicts:
src/vec/vec/impls/seq/seqcusp/cuspvecimpl.h
src/vec/vec/impls/seq/seqviennacl/viennaclvecimpl.h

show more ...

PCMG man page: fix typo in options name

Merge branch 'barry/monitor-fromoptions'

Use VECCUDA with MATAIJCUSPARSE

PCML: fix MatDestroy_ML to use MatShell correctly and compatibly with visibility

MatShell interprets MATOP_DESTROY as a destructor for the context, but
the ML interface bypassed that mechanism, inve

PCML: fix MatDestroy_ML to use MatShell correctly and compatibly with visibility

MatShell interprets MATOP_DESTROY as a destructor for the context, but
the ML interface bypassed that mechanism, inverting control to call
MatDestroy_Shell statically. However, that function is not exported by
libpetscksp.so (relevant for single-library=0), so we currently get a
link-time failure. This commit switches to using MatShell and
MATOP_DESTROY in the usual way, so MatDestroy_Shell calls MatDestroy_ML
instead of vice-versa.

show more ...

Merge branch 'pr426/dmay/dmay/pctelescope-mods/master'

PCBDDC: fixes for complex builds

PCBDDC: if adaptive selection has not been requested at the coarse level, we need to use zerorows option for deluxe

PCBDDC: use MatZeroRowsColumnsIS

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

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

Merge branch 'barry/add-dmshellcreaterestriction'

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 ...