fix typo
Reported-by: Febrian Setianto <febrian.setianto@gmail.com>

Man pages: add newlines after "Notes:"

This allows for proper formatting from sowing.

On OS X (using gsed, not the default BSD sed), from the PETSc root directory:

find src include -type f \(

Man pages: add newlines after "Notes:"

This allows for proper formatting from sowing.

On OS X (using gsed, not the default BSD sed), from the PETSc root directory:

find src include -type f \( -name "*.c" -or -name "*.h" -or -name "*.cxx" \) | xargs gsed -i 's/Notes\s*:\s*\(\w.*\)/Notes:\n \1/'

This adds a newline and 4 spaces whenever "Notes:" is followed by any "word" character, in any .c, .h, or .cxx file in src/ or include/

show more ...

Man pages: capitalize "notes:" to "Notes:"

On OS X (using gsed, not default BSD sed), from PETSc root directory,

find src include -type f \( -name "*.h" -or -name "*.c" \) | xargs gsed -i 's/no

Man pages: capitalize "notes:" to "Notes:"

On OS X (using gsed, not default BSD sed), from PETSc root directory,

find src include -type f \( -name "*.h" -or -name "*.c" \) | xargs gsed -i 's/notes:/Notes:/g'

show more ...

Merge remote-tracking branch 'origin/knepley/fix-fe-vector-spaces'

* origin/knepley/fix-fe-vector-spaces: (21 commits)
SNES ex12: Updated test output
Plex ex8: Updated test output
Plex ex7: Up

Merge remote-tracking branch 'origin/knepley/fix-fe-vector-spaces'

* origin/knepley/fix-fe-vector-spaces: (21 commits)
SNES ex12: Updated test output
Plex ex8: Updated test output
Plex ex7: Updated test output
Plex ex5: Updated test output
Plex ex4: Updated test output
Plex ex3: Updated test output
Plex ex20: Updated test output
Plex ex19: Updated test output
Plex ex18: Updated test output
Plex ex17: Updated test output
Plex ex16: Updated test output
Plex ex13: Updated test output
Plex ex12: Updated test output
Plex ex1: Updated test output
DualSpace: Interaction of branches
FAS: Added viewing during coarse correction
Plex+FEM: Fixes for reorg of vector spaces which I missed - Fixed DMPlexComputeInterpolatorNested() field offsets - Fixed DMPlexComputeInjector()
PetscFE: Fixed viewers - Added PetscDualSpace viewer - Can view FE without quadrature
PC: Added some options and documentation for BJacobi, ASM, Richardson
Plex: Fixes for ASCII viewer
...

show more ...

Merge branch 'master' into knepley/feature-pc-patch

* master: (440 commits)
revert sources to dev - reset PETSC_VERSION_RELEASE - reset changes/dev.html
release: set v3.9 strings
fix grep us

Merge branch 'master' into knepley/feature-pc-patch

* master: (440 commits)
revert sources to dev - reset PETSC_VERSION_RELEASE - reset changes/dev.html
release: set v3.9 strings
fix grep usage on solaris [use egrep]
opensolaris: fix errors
Update timer reporting and minor reporting fix
Remove redundant if statement. Found by @jedbrown
Fixed problem with order of developer photos on website.
superlu_dist: use correct flag in example [PETSC_HAVE_SUPERLU -> PETSC_HAVE_SUPERLU_DIST]
Bib: PERMON ref
fix for osx
doc: more author fixes
doc: more author fixes
Italicized "orderings" in MPI process placement discussion.
Added Richard to the user manual author list.
Added Richard to the list of active developers on the website.
add string check for MS-MPI and IMPI timeout errors
opensolaris does not have multiple grep utils - but default one does not support -F. So use fgrep instead
Standardizing timeout checks
Report wall clock time of total tests
Use time instead of date %s
...

Conflicts:
include/petscpc.h
src/ksp/ksp/interface/itfunc.c

show more ...

PCFieldSplit: Fix tabbing in view

Merge remote-tracking branch 'origin/dalcinl/mat-getset-ops'

PCPatch: Correct indentation in PCView output

Deprecate DMSet/GetDefaultSection() for DMSet/GetSection()

Revert manual page change per Matt's preference

Commit-type: style-fix
Thanks-to: Matthew Knepley

PCPatch: Small fixes
- ObjectView() should not push the stack
- Fixed behavior of sub_mat_type on Reset
- Fixed nsubspaces on Reset

BoomerAMG, unlike 2 other hypre preconditioners, does not require its own MPI_Comm obtained via MPI_Comm_dup()

Presumably BoomerAMG must use the communictor provided in the HYPRE matrix passed in

M

BoomerAMG, unlike 2 other hypre preconditioners, does not require its own MPI_Comm obtained via MPI_Comm_dup()

Presumably BoomerAMG must use the communictor provided in the HYPRE matrix passed in

MPI implementations have a limited number of MPI_Communicators that can co-exist, hence one should minimize the
creation MPI_Communicators and never create any unnecessarily.

Commit-type: portability-fix
Reported-by: Fande Kong <fande.kong@inl.gov>

show more ...

Merge remote-tracking branch 'origin/jed/fix-analyzer-bugs'

Add MatGetOperation() and fixes to MATSHELL

PCPatch: Fixed memory leak

PCPatch: Fixed initialization

PCPatch: Looks like Q1-Q0 Vanka for Stokes might work
- Now ComputeOperator() also gets the patch number (changes SetComputeOperator())
- Replaced ViewFromOptions() with flags to avoid string compari

PCPatch: Looks like Q1-Q0 Vanka for Stokes might work
- Now ComputeOperator() also gets the patch number (changes SetComputeOperator())
- Replaced ViewFromOptions() with flags to avoid string comparisons in tight loops
- Fixed loop bounds for closures
- Added -ignore_dim to ignore points of a given dimension on the patch boundary, rather than patch proper
- cellNodeMap is copied rather than being a borrowed reference
- Allow setting DiscretizationInfo from a combined DM
- CompleteCellPatch() now takes PC instead of DM
- GetPointsDofs() now takes PC instead of imll
- Also make point lists for patches, in addition to cell lists. We need this for Section indexing
- Put in SF support
- Create GlobalSection for each patch that can be passed to patch assembly routine
- Change mat prefix to pc_patch_sub_
- Added Plex operator construction
- Reformatting

show more ...

Doc: Improve documentation for debugging options

Merge branch 'karlrupp/feature-remove-cusp'

* karlrupp/feature-remove-cusp:
Removes VECCUSP vectors and AIJCUSP matrices, because they provide no added benefit over VECCUDA and AIJCUSPARSE.
Also rem

Merge branch 'karlrupp/feature-remove-cusp'

* karlrupp/feature-remove-cusp:
Removes VECCUSP vectors and AIJCUSP matrices, because they provide no added benefit over VECCUDA and AIJCUSPARSE.
Also removes the CUSP-based preconditioners for now.
SACUSP will be re-added later, after the new release is out.

show more ...

GPUs: Allow ViennaCL and VECCUDA to be used concurrently.

This also gets rid of the weird veccuda.py.

Merge branch 'stefano_zampini/fix-matis-dmda-l2g'

CUSP: Removed VECCUSP and AIJCUSP as well as preconditioners.

SA-AMG from CUSP will be re-added in a follow-up commit,
made to work with AIJCUSPARSE and VECCUDA.

Merge branch 'master' into knepley/feature-pc-patch

* master: (33 commits)
cleanup very ugly PCGAMG manual page
minor clarifications for VecScatterCreate() manual page
gcov: PETSC_ARCH is requ

Merge branch 'master' into knepley/feature-pc-patch

* master: (33 commits)
cleanup very ugly PCGAMG manual page
minor clarifications for VecScatterCreate() manual page
gcov: PETSC_ARCH is required onlyfor -run_gcov - and not for -merge_gcov [where it doesn't exist]
make clearer in help statements what --with-external-packages-dir is for.
Website: Added news entry for PETSc User Meeting 2018 on main page.
KSPGUESSPOD: exclude negative of zero eigenvalues from the energy computation
KSPGUESSFISCHER: use PetscInfo for KSP class
CUDA: Further test output fixes for mat_tests-ex2
Turn off elemental test for parmetis running since this does not work on batch systems.
3.8 Changes: KSPCHEBYSHEV
CUDA: Fixed most test errors due to MatMultTranspose*
KSPFGMRES: add KSP_NORM_NONE / PC_RIGHT as supported
KSPSetSupportedNorm man page: remove old note about default support for KSP_NORM_NONE
Dev manual: copyedit
DMPlexDistributeLabels: Propagate the output flag
cuda/cusp: update minimum supported versions of cuda to 7.5 and cusp to 0.5.0
cuda: remove configure option -with-cuda-arch. [use CUDAFLAGS instead]
projected gradient code fixed to obey recent convergence test changed
Merge conflicts resolved after rebase
norm update bug fixed in GPGC, test output reverted
...

show more ...

cleanup very ugly PCGAMG manual page

Commit-type: documentation

PCBDDC: yet another recipe

Same as before, but save one factorization

PC: Added some options and documentation for BJacobi, ASM, Richardson