DMPlexCreateOverlap -> DMPlexCreateOverlapLabel; doc fixes

Merge branch 'haplav/feature-dmaddlabel-reference' into 'master'

DMAddLabel: reference the label

See merge request petsc/petsc!2064

DMPlexDistributeLabels: use DMRemoveLabelBySelf()

DMPlexDistributeLabels: DMLabelDestroy() should no more be called here

DMAddLabel: Reference the label.

This is a clear case of referencing one PetscObject from another.

Merge branch 'master' into 'barry/2019-08-22/fix-fblaslapack-for-nag'

# Conflicts:
# lib/petsc/conf/test

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

SF: Check validity of graph in debug mode

See merge request petsc/petsc!1983

plex: Fix voting scheme in point sf creation when rebalancing

We previously used a two-round system when using random shifts to
compute point ownership. This, however, could result in incorrect
resu

plex: Fix voting scheme in point sf creation when rebalancing

We previously used a two-round system when using random shifts to
compute point ownership. This, however, could result in incorrect
results if the vote from two different ranks happened to coincide. At
which point the second entry being compared acts a tie-breaker with
the MPI standard requiring that the minimum value wins.

Consider an SF:

PetscSF Object: 2 MPI processes
type: basic
sort=rank-order
[0] Number of roots=1, leaves=1, remote ranks=1
[0] 0 <- (0,0)
[1] Number of roots=1, leaves=1, remote ranks=1
[1] 0 <- (0,0)

If we reduce with MAXLOC where the leafdata is:

Rank 0:

(1, 5)

Rank 1:

(1, 4)

We end up with

(1, 4) in the rootdata on rank 0.

Now to determine the correct rank, we do a subsequence reduce with
MAXLOC where now the leaf data is:

Rank 0:

(1, 0)

Rank 1:

(1, 1)

And end up with

(1, 0) in the rootdata on rank 0.

We therefore construct the incorrect entry:

(0, 4)

Rather than either (0, 5) or (1, 4).

The solve this, do the reduction in one go on a 3-tuple of integers,
carrying both the rank and index we will select.

show more ...

Merge branch 'jed/deprecate-dmgetdefaultsection' [PR #1969]

* jed/deprecate-dmgetdefaultsection:
DMGetDefaultSection et al: add PETSC_DEPRECATED_FUNCTION
These functions were deprecated in v3.9, but

Merge branch 'jed/deprecate-dmgetdefaultsection' [PR #1969]

* jed/deprecate-dmgetdefaultsection:
DMGetDefaultSection et al: add PETSC_DEPRECATED_FUNCTION
These functions were deprecated in v3.9, but no PETSC_DEPRECATED_FUNCTION was added at that time so many uses crept back into the source code. Here we mark it deprecated and fix all existing uses in PETSc.

DMGetLocalSection/DMSetLocalSection: add explicit names for DMGetSection/DMSetSection
These names are for symmetry with DMGetLocalVector() versus DMGetGlobalVector().

DMGetSection: replace all internal usage with DMGetLocalSection
Same for DMSetSection -> DMSetLocalSection. We retain the interface for DMGetSection/DMSetSection without deprecation warnings (for now).

show more ...

DMGetSection: replace all internal usage with DMGetLocalSection

Same for DMSetSection -> DMSetLocalSection. We retain the interface
for DMGetSection/DMSetSection without deprecation warnings (for n

DMGetSection: replace all internal usage with DMGetLocalSection

Same for DMSetSection -> DMSetLocalSection. We retain the interface
for DMGetSection/DMSetSection without deprecation warnings (for now).

Suggested-by: Barry Smith <bsmith@mcs.anl.gov>

show more ...

Merge branch 'dalcinl/fix-headers' [PR #1897]

* dalcinl/fix-headers:
Minor fixes to headers.

Fix a bunch of PetscValidPointer() -> PetscValidXXXPointer()

Merge branch 'knepley/fix-plex-distribute-overlap' [PR #1858]

* knepley/fix-plex-distribute-overlap:
Doc: State that arguments originalVec and newVec take local vectors in DMPlexDistributeField()
Pl

Merge branch 'knepley/fix-plex-distribute-overlap' [PR #1858]

* knepley/fix-plex-distribute-overlap:
Doc: State that arguments originalVec and newVec take local vectors in DMPlexDistributeField()
Plex: Fix error in overlap distribution
When remapping SF, must allow for the case when the leaf ordering is not canonical
Plex ex36: TEst for overlapping field redistribution

show more ...

Merge branch 'gropp/doc-fixes-1' [PR #1838]

Plex: Fix allocation size

Plex: Fix error in overlap distribution - When remapping SF, must allow for the case when the leaf ordering is not canonical

Doc: State that it takes a local vector

Merge from doc-fixes branch

Fixes for doctext update

Fix manual page formating. Only single return parameter

Commit-type: documentation
Reported-by: Pierre Jolivet <pierre.jolivet@enseeiht.fr>

Merged in barry/update-collective-on (pull request #1744)

Update the use of Collective on in the manual pages to reflect the new style

Update the use of Collective on in the manual pages to reflect the new style

Commit-type: style-fix, documentation
Thanks-to: Patrick Sanan <patrick.sanan@gmail.com>

Merged in psanan/man-pages-remove-keywords (pull request #1717)

Man pages: Remove .keywords fields

Approved-by: BarryFSmith <bsmith@mcs.anl.gov>

Man pages: remove .keywords: fields

This field is recognized by doctext from Sowing (http://wgropp.cs.illinois.edu/projects/software/sowing/doctext/node20.htm#Node20),
but it doesn't seem to be unif

Man pages: remove .keywords: fields

This field is recognized by doctext from Sowing (http://wgropp.cs.illinois.edu/projects/software/sowing/doctext/node20.htm#Node20),
but it doesn't seem to be uniformly used or maintained.

Thus, remove all .keywords: fields, and a following blank line, if present.

This is accomplished with GNU sed (gsed on OS X), with the following commands.
*Warning* that this type of command can corrupt a .git directory,
so be cautious in reusing or modifying these commands. They first look
for and delete matching lines with a following line consisting of only whitespace,
and then delete any remaining matching lines.

find src/ -type f -exec gsed -i '/keywords:/ {N; /\n\s*$/d}' {} +
find src/ -type f -exec gsed -i '/keywords:/d' {} +
find include/ -type f -exec gsed -i '/keywords:/ {N; /\n\s*$/d}' {} +
find include/ -type f -exec gsed -i '/keywords:/d' {} +

Hints on the sed command obtained from: https://unix.stackexchange.com/questions/100754/how-to-delete-a-specific-line-and-the-following-blank-line-using-gnu-sed

show more ...

Merged in stefano_zampini/dmplex-speedup (pull request #1665)

Last speedups to the PLEX code with large number of MPI processes

Approved-by: Matthew Knepley <knepley@gmail.com>