Merge branch 'stefanozampini/revert-for-glvis-parallel' into 'master'

revert check same comm on DMView

See merge request petsc/petsc!2039

Knepley/is separate section

Merge branch 'maint'

MSMPI-32: compile fixes

Reported-by: Mohammad Asghar <mohammad.asghar@lusas.com>

PetscViewerGLVIS: minor checks

petscintview and friends now indeed do print ranks

now petscint and friends dont print ranks if size is 1

Merge branch 'wence/feature/sf-check-graph-valid' into 'master'

SF: Check validity of graph in debug mode

See merge request petsc/petsc!1983

ISSortRemoveDups(): deal with the now immutable layout

PetscLayout: forbid calling setup twice with different sizes

* PetscLayout objects should be considered immutable, especially because they can be shared.
* Make use of PetscLayoutCreateFromSizes().

PetscLayout: forbid calling setup twice with different sizes

* PetscLayout objects should be considered immutable, especially because they can be shared.
* Make use of PetscLayoutCreateFromSizes().
* setupcalled is now the single indicator that PetscLayout is ready.
* Insure the rest of variables (e.g. map->rstart) are calculated even if n, N and range are set.

show more ...

Add PetscLayoutCreateFrom{Sizes,Ranges}.

Remove PetscLayoutSetUp_SizesFromRanges_Private.
Update MatLoad_AIJ_HDF5.
ISGeneralSetIndices: Use PetscLayoutCreateFromSizes(), preserve blocksize.

Make IS sizes fully given by PetscLayout.

* Remove n,N from IS_Stride, which were just bypassing PetscLayout ones
* Remove getsize, getlocalsize from _ISOps
* ISGetSize, ISGetLocalSize now take size

Make IS sizes fully given by PetscLayout.

* Remove n,N from IS_Stride, which were just bypassing PetscLayout ones
* Remove getsize, getlocalsize from _ISOps
* ISGetSize, ISGetLocalSize now take sizes right from the PetscLayout
* Deal with now immutable PetscLayout using PetscLayoutCreateFromSizes()
* Assure that blocksize is preserved in ISGeneralSetIndices, ISStrideSetStride

show more ...

PetscViewerHDF5Load: fix manpage (map is inout)

PetscViewerHDF5ReadSizes_Private: optionally do PetscLayoutSetUp()

PetscLayoutDuplicate: don't copy range if it doesn't exist

PetscLayoutCreate: manpage fixes

SF: Stricter check for valid inputs in PetscSFCompose

An SF might not encode a true star forest (leaves could be repeated).
Composition of SFs does, however, rely on unique leaves, so check that in

SF: Stricter check for valid inputs in PetscSFCompose

An SF might not encode a true star forest (leaves could be repeated).
Composition of SFs does, however, rely on unique leaves, so check that in
debug mode.

show more ...

SF: Check validity of graph in debug mode

Forbid negative leaf indices in ilocal, negative root indices in
iremote, and invalid ranks in iremote. Since sometimes an SF is used
just to carry informat

SF: Check validity of graph in debug mode

Forbid negative leaf indices in ilocal, negative root indices in
iremote, and invalid ranks in iremote. Since sometimes an SF is used
just to carry information (e.g. DMLocatePoints), do the check when
we're setting up rather than when setting the graph.

show more ...

Merge branch 'barry/2019-09-08/therefor' into 'master'

Docs only, Fix spelling of therefor

See merge request petsc/petsc!2023

Merge branch 'barry/2019-08-21/add-matfdcoloringsetvalues' into 'master'

Added MatFDColoringSetValues() takes matrix in compressed color form (as often...

See merge request petsc/petsc!1999

Merge branch 'knepley/fix-doc-section' into 'master'

Doc: Reclassified Section routines, added some docs, fixed some docs

See merge request petsc/petsc!1981

Docs only, Fix spelling of therefor

Commit-type: style-fix
Reported-by: Pierre Jolivet <Pierre.Jolivet@enseeiht.fr>

Merge branch 'wence/fix/sf-compose-non-identity' into 'master'

Fix PetscSFCompose when sfA has non-identity permutation of the local space

See merge request petsc/petsc!1979

Section+Doc: Fixes Suggested-by: "Lawrence Mitchell" <lawrence.mitchell@durham.ac.uk>

Doc: Reclassified Section routines, added some docs, fixed some docs

Also sets isperm = PETSC_TRUE.