add MatDestroySubMatrices(); bugfix

Merge branch 'barry/remove-funct'

Merge branch 'barry/fix-matreuse'

Add more error checks for use of MatReuse; change MatTranspose() to use MatReuse as used by other functions

Time: 2.5 hours
Reported-by: Stefano Zampini <stefano.zampini@gmail.com>, Eric Chamberlan

Add more error checks for use of MatReuse; change MatTranspose() to use MatReuse as used by other functions

Time: 2.5 hours
Reported-by: Stefano Zampini <stefano.zampini@gmail.com>, Eric Chamberland <Eric.Chamberland@giref.ulaval.ca>

show more ...

Merge branch 'barry/half-precision'

Remove the use and definition of __FUNCT__ throughout the code

Since all modern C/C++ compilers provide this functionality we no longer need to provide it manually in PETSc
Time: 1.5 hours
Thanks-t

Remove the use and definition of __FUNCT__ throughout the code

Since all modern C/C++ compilers provide this functionality we no longer need to provide it manually in PETSc
Time: 1.5 hours
Thanks-to: Andreas Mang <andreas@ices.utexas.edu>

show more ...

Support for half-precision using __fp16.

Requires repository trunk of gcc and ARM systems
Configure with --with-precision=__fp16 --download-f2cblaslapack --with-fc=0 --with-cxx=0
Many standard PETSc

Support for half-precision using __fp16.

Requires repository trunk of gcc and ARM systems
Configure with --with-precision=__fp16 --download-f2cblaslapack --with-fc=0 --with-cxx=0
Many standard PETSc tests fail due to lack of precision

Funded-by: ECP
Project: fp16
Time: 24 hours

show more ...

Merge branch 'stefano_zampini/feature-pcbddc-saddlepoint'

Conflicts:
src/mat/impls/aij/mpi/mpiaij.c
src/mat/impls/is/matis.c

Modified ex23.c for the API change of the function CheckMat

Merge branch 'master' into tisaac/dmplex-feature-real-to-ref

I need updated PetscFE interfaces in what I'm going to do

* master: (603 commits)
parmetis: fix typo in testsuite
Added PETSC_HAVE_A

Merge branch 'master' into tisaac/dmplex-feature-real-to-ref

I need updated PetscFE interfaces in what I'm going to do

* master: (603 commits)
parmetis: fix typo in testsuite
Added PETSC_HAVE_ATTRIBUTEALIGNED flag to makefile
PCSetUp_SVD() used incorrect matrix for matrix size in parallel, also need to skip LAPACK call if matrix size is zero
strumpack: update to latest version 1.0.4
Remove valgrind warning about MPI_Allreduce() with uninitialized data
sed for test output that has atol convergence with quad precision
updated ex1_nest tests
fixed incorrect key value
fix: bad merge a69d34a
test: move runex78 to TESTEXAMPLES_C_NOTSINGLE
build: fix warnings
sys: update test to produce same result - irrespective of debug/opt build - and presence of ~/.petscrc
sys: src/sys/examples/tests/ex30.c -> src/sys/examples/tests/ex32.c to avoid conflict in master
Sed output, since quad precision reduces below atol on these tests
remove automatically generated matrix names from test output files since can be different
Fixes for MatMPIXXXSetPreallocation(), they were not reentrant, since they kept the information used for matrix-vector products
GAMG: fix PCView_GAMG() format - and update output/ex56_0.out
superlu_dist: 64-bit-indices fix https://github.com/xiaoyeli/superlu_dist/issues/4
Add SNESSetDivergenceTolerance / -snes_divergence_tolerance.
fixed unit read in Plex
...

show more ...

Merge branch 'stefano_zampini/feature-convert-matnest-matis' into stefano_zampini/feature-pcbddc-saddlepoint

I have to merge it here, since there's an API addition for MatIS in stefano_zampini/featu

Merge branch 'stefano_zampini/feature-convert-matnest-matis' into stefano_zampini/feature-pcbddc-saddlepoint

I have to merge it here, since there's an API addition for MatIS in stefano_zampini/feature-pcbddc-saddlepoint
that prevents me to smoothly merge stefano_zampini/feature-convert-matnest-matis into next.

This will also facilitate merging stefano_zampini/feature-pcbddc-saddlepoint into master

Conflicts:
src/docs/website/documentation/changes/dev.html

show more ...

Add error check for PetscInt overflow in creating MATDENSE

Time: .1 hours
Reported-by: Amirhossein Molavi Tabrizi <a.molavitabrizi@northeastern.edu>

PCBDDC: added support for Nedelec elements using the discrete gradient

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

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

Merge branch 'hzhang/matpackage-type'

set solvertype for inplace MatxxxFactor()

Fix null output from PCFactorGetMatSolverPackage();
Reported-by: Jose E. Roman <jroman@dsic.upv.es>

Fixes found by clang static analyzer
mostly set but unused variables plus a couple of true bugs

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

Merged in dalcinl/draw-null (pull request #434)

PetscDraw: Proper implementation of PETSC_DRAW_NULL

PetscDraw: Proper implementation of PETSC_DRAW_NULL

* Remove special casing of PETSC_DRAW_NULL in the draw interface
* Rework handling of PETSC_DRAW_NULL in Draw{Axis|LG|SP|BAR|HG}
* Some 'extern' -

PetscDraw: Proper implementation of PETSC_DRAW_NULL

* Remove special casing of PETSC_DRAW_NULL in the draw interface
* Rework handling of PETSC_DRAW_NULL in Draw{Axis|LG|SP|BAR|HG}
* Some 'extern' -> 'PETSC_INTERN' fixes
* Add missing error checking

show more ...

Merge branch 'master' into knepley/feature-plex-handle-func-failure

* master: (378 commits)
SNES ex5: Fix coordinate determination for empty subdomains
fixed petsc.bib entry
p4est: fix syntax

Merge branch 'master' into knepley/feature-plex-handle-func-failure

* master: (378 commits)
SNES ex5: Fix coordinate determination for empty subdomains
fixed petsc.bib entry
p4est: fix syntax error with error message
SNES ex5: Correct Jacobian handling on patches
SNES ex5: Stupid complex fixes
DMP4est: fix transfer of labels to parents
petsc4py: update to latest snapshot. Fixes da example breakage in petsc4py test suite
Plex: Correct setting of options on refined or coarsened meshes
Plex: Add info for grid hashing
Plex: Add logging in interpolator creation
Plex: Correct use of tabs in ASCII viewer
DM: Shorten long event name
Plex ex12: Create boundary label automatically at all levels of mesh hierarchy
Plex ex12: Change PetscViewerSetFormat --> PetscViewerPushFormat - Fix order of SetMode and SetFilename
Plex ex12: Change all functions to static
Since DMDA with periodic boundary conditions does support P0 interpolation remove checks that error out under those conditions
ascem-io.py and amanzi.py need to handle self.hdf5.include being None correctly
fix MPI Uni to recycle communicators so it won't run out like with the petsc4py tests also fix check for location of python shared library on mac to check all cases each time
Switch back to Sherry's version since she accepted by pull request
PetscDraw: Update help messages for -draw_save* options
...

Conflicts:
src/dm/impls/plex/plexfem.c

show more ...

Merge branch 'stefano_zampini/feature-zerorowscolumns-seqdense'

Merged petsc/petsc into master

Merge branch 'barry/fix-private-functions'

Merge remote-tracking branch 'origin/knepley/fix-plex-hdf5-label' into tisaac/dmforest

* origin/knepley/fix-plex-hdf5-label: (50 commits)
Plex: Fixed HDF5 output for DMLabel - The ISAllGather() wa

Merge remote-tracking branch 'origin/knepley/fix-plex-hdf5-label' into tisaac/dmforest

* origin/knepley/fix-plex-hdf5-label: (50 commits)
Plex: Fixed HDF5 output for DMLabel - The ISAllGather() was being called with an IS that had COMM_SELF - Also, we needed to protect for empty index arrays
superlu_dist: install target should first make sure prefix/[include,lib] dirs exist - similar to superlu
dm: add DMRemoveLabel() fortran interface
nightlybuilds: allocate more time to run arch-linux-xsdk-dbg build...
Remove superfluous information messages
Cannot have rank 1 print before parallel print since it may end up before or after the print from process 0. There change to have rank 0 print the sequential matrix
Bib: Added ref
Plex: Update ex12_5 with correct level-2 overlap
DMLabel: Add missing cleanup for DMLabelGather()
mpiuni: check for "long_long" before checking for "__int64" [as "typedef __int64" gives error with icc] Note: cannot use 'int64_t' as 'typedef unsigned int64_t foo' gives eror with icc
Add filter for solaris linker scope warning to master since that is used in nightly tests of next
Previous commit did not correctly handle downloading from --with-packages-dir
Use special error code of 10 for output list of needed downloaded package tarballs otherwise the outer configure.py script converts the error code to an error message we do not want.
Have ./configure return a nonzero exit code when --with-packages-dir detects missing packages so that calling scripts know this is occurred simplify some code
Completed support for --with-packages-dir=directory that allows --download-xxx features to be cleanly used behind firewalls etc also updated to docs to indicate this new feature
Adding supporting for --download-xxx to reported needed downloads when off line or behind a firewall and using local copies when available with the --with-packages-dir=directory option
Fix typos in linesearch impls:armijo, morethuente, owarmijo.
PetscDraw: Explicitly save draw in KSP/SNES/TS monitors
PetscViewerDraw: Propagate prefix from viewer to draw contexts
PetscDraw: Explicitly save draw in {Vec|Mat|DMDA}View() routines
...

show more ...