Merge branch 'jed/malloc-array'

Type arguments dropped from PetscMalloc[2-7], PetscNew, and PetscNewLog,
added PetscMalloc1 for allocating typed arrays, add PetscCalloc[1-7] for
allocating cleared (

Merge branch 'jed/malloc-array'

Type arguments dropped from PetscMalloc[2-7], PetscNew, and PetscNewLog,
added PetscMalloc1 for allocating typed arrays, add PetscCalloc[1-7] for
allocating cleared (zeroed) memory.

* jed/malloc-array:
Sys: Add pointer casts from (void **) in calls to PetscMalloc1()
Sys: drop explicit type arguments from PetscNew() and PetscNewLog()
Sys: add PetscCalloc[1-7]
Sys: add PetscMalloc1 macro, array allocation without redundant types
PetscMalloc[2-7]: remove type arguments, infer from pointer type
SNESComputeJacobianDefaultColor: fix uninitialized variable

show more ...

Sys: add PetscMalloc1 macro, array allocation without redundant types

The type is inferred from the pointer return type. This patch is
automated via the following script:

git grep -l 'PetscMalloc(

Sys: add PetscMalloc1 macro, array allocation without redundant types

The type is inferred from the pointer return type. This patch is
automated via the following script:

git grep -l 'PetscMalloc(.*sizeof' src | xargs perl -pi -e 's@PetscMalloc\(([^,;]*[^,; ]) *\* *sizeof\([^,;()]+\),@PetscMalloc1($1,@'

This commit contains an additional bug-fix in csrperm.c, fixing pointer
arity. The code was introduced in 2006, but the allocation could not
have been correct at any time. This probably means that
MatDuplicate_SeqAIJPERM has never been tested.

a54129beb540034ba105796c682d589e7e1111f2
Richard Tran Mills <rmills@ornl.gov>

Added MATSEQCSRPERM support for MatDuplicate() and conversion to/from
MATSEQAIJ. Note that these changes are not quite debugged.

show more ...

PCBDDC: move creation of temporary MATIS matrix before coarse ksp is created.

This way, we can temporary set the coarse MATIS into coarse ksp.
This allows to call pc_bddc_coarse_ksp_fischer_guess <>

PCBDDC: move creation of temporary MATIS matrix before coarse ksp is created.

This way, we can temporary set the coarse MATIS into coarse ksp.
This allows to call pc_bddc_coarse_ksp_fischer_guess <> from option

show more ...

PCBDDC: remove trailing backspace

Merge branch 'stefano_zampini/pcbddc-applytranspose' into stefano_zampini/pcbddc-multilevel-improvement

PCBDDC: demand wheter or not to rebuild local scatters to the right routine

PCBDDC: remove prints. Added coarse_subassembling to debugging output

PCBDDC: added reuse of coarse subassembling information

In order to do do, I had to change the MATIS class (I dont know why there was a guard for is->mapping in matis code)

Solved a couple of bugs

PCBDDC: added reuse of coarse subassembling information

In order to do do, I had to change the MATIS class (I dont know why there was a guard for is->mapping in matis code)

Solved a couple of bugs when doing multilevel and a dbg_viewer on a single level only

show more ...

PCBDDC: coarse ranks are no longer contiguous (better memory management)

PCBDDC: remove prints

PCBDDC: KSPRICHARDON with PCBDDC is the default for BDDC coarse solver when doing multilevel

Extreme eigenvalues are computed if the coarse KSP is not preonly to accelerate convergence

Merge branch 'master' into knepley/feature-dmda-section

* master: (528 commits)
Doc: Fixed configure doc
Nightly summary: Added configure section.
TSSetErrorIfStepFails: fix __FUNCT__
TS: fi

Merge branch 'master' into knepley/feature-dmda-section

* master: (528 commits)
Doc: Fixed configure doc
Nightly summary: Added configure section.
TSSetErrorIfStepFails: fix __FUNCT__
TS: fix for TSGetRHSMats_Private() in the creation of the Arhs and Brhs matrices
protect logging of objects (which uses global variables) with PETSC_USE_LOG
error handler API had been previously changed but this example was not updated to new API.
DMPlex: Fixed bug in coordinate permutation
PetscSection: Doc fix for permutation
Bib: Added Top500 and fixed entry
fun3d: update PetscMallocValidate() usage
bugfix of MatCreateSeqAIJFromTriple() contributed by Nicolas Marsic <nicolas.marsic@montefiore.ulg.ac.be>
bib: rename ref with duplicate key
Webpage: Corrected '-dm_mat_type cusp' to 'dm_mat_type aijcusp'.
completed removing server code that was not removed in previous commit
removed PETSc webserver and configure package files that supported it
Allow calling MatGetBlockSize[s]() before matrix preallocation
SNESLINESEARCHBT: Set the norms when exiting early due to negligible step.
fix outdated PetscOptionsList
SNESQN: only monitor real part of dot product (fails with C++ complex)
Fix for multiple definitions in mat/color/interface/ custom fortran bindings
...

Conflicts:
config/builder.py
include/petscdmplex.h
src/dm/impls/da/dageometry.c
src/dm/impls/plex/examples/tests/ex4.c
src/dm/impls/plex/plex.c
src/dm/impls/plex/plexinterpolate.c

show more ...

PCBDDC: support for PCApplyTranspose completed by solving the transpose of the coarse operator in PCBDDCApplyInterfacePreconditioner.

Function pointer is now set during object creation

PCBDDC: remove unneded guards

these guards were needed in the previous version of the code when PCBDDC coarse problem was handling through the enum PCBDDCCoarseProblemType

PCBDDC: added support for solving the transposed of the substructure correction

rhs, solution and work vectors for the correction are now parameters of the function

PCBDDC: added second parameter to PCBDDCApplyInterfacePreconditioner to determine if the interface preconditioner should be applied tranposed or not

The main reason for this is to avoid code duplica

PCBDDC: added second parameter to PCBDDCApplyInterfacePreconditioner to determine if the interface preconditioner should be applied tranposed or not

The main reason for this is to avoid code duplication

show more ...

PCBDDC: simplify PCBDDCApplyInterfacePreconditioner

coarse_psi_* now always exist. For symmetric problems, they point to the same object of coarse_phi_*

PCBDDC: rename private PCBDDCSolveSaddlePoint to a more explicative PCBDDCSolveSubstructureCorrection

PCBDDC: propagate field split and neumann boundaries to coarse problem

PCBDDC: coarse matrix is structurally symmetric

Merge branch 'stefano_zampini/pcbddc-globalordering-api' into stefano_zampini/pcbddc-multilevel-improvement

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

PCBDDC: provide APIs for setting field split in global ordering

previous API moved to PCBDDCSetDofsSplittingLocal

For consistency with updated man pages, which_dof values should default to -1;
this

PCBDDC: provide APIs for setting field split in global ordering

previous API moved to PCBDDCSetDofsSplittingLocal

For consistency with updated man pages, which_dof values should default to -1;
this way, unlisted nodes in field split will belong to a different subset

show more ...

PCBDDC: provide API for global to local translation (more general than ISGlobalToLocal and memory scalable)

PCBDDC: provide API for setting Dirichlet/Neumann boundaries in global ordering

PCBDDC: fix for new reuse logic in the multilevel case