Need to dely attachment of null spaces from IS to matrices inside fieldsplit until after the matrices are available

Reported-by: Lawrence Mitchell <lawrence.mitchell@imperial.ac.uk>

Merge branch 'master' of bitbucket.org:petsc/petsc

Merge branch 'master' into mesgarnejad/feature-plex-globaltonatural

* master: (111 commits)
add filters for website tests that removes warnings about MPI type tag mismatches due to const
remove

Merge branch 'master' into mesgarnejad/feature-plex-globaltonatural

* master: (111 commits)
add filters for website tests that removes warnings about MPI type tag mismatches due to const
remove Peter from blame-email list.
don't warn about MPI const long * versus long long *
Configure: Added GMP an MPRF - These allow arbitrary precision arithmetic
SNES ex69: SolCx finally works - Fixed order of parameters - Made different versions of pointwise functions
SNES ex69: Fixed bug where SolKx and SolCx have different numbers of parameters
SNES ex69: Small bug fixes - Initialize coordinates for SolCx - Fixed declaration length
I believe this call on line 357: ierr = DMSetFromOptions(dau); should actually set `dak`. The DM `dau` is already set on line 349.
SNESSolve_KSPOnly() did not compute function at new solution unless monitoring is turned on but SNESSolve_FAS() internals assume that each subsnes solve it calls does compute the new function value so broke when -fas_levels_snes_monitor was not turned on but -fas_levels_snes_type kspmonitor was used
Have SNESVIGetInactiveSet return the *actual* current inactive set.
One now passes the nullspace in the matrix for a linear system as the first Amat argument to KSPSetOperators, SNESSetOperators, TSSetIJacobian() not the second pmat argument
Remove change documentation of Removed MatMFFDAddNullSpace() just use MatSetNullSpace() as it is now backported in maint (36.html)
MatMFFD doesn't need to have its own null space management cherry-picked from 39601f4 (barry/remove-matmffdaddnullspace/master)
Removed trailing '\\' in the authors line.
add note to PetscOptionsInsertFile() that it should not be used for PETSc wide options
Add Fortran stub for SNESSetUpdate()
improve manual page slightly for clarity
PCGASM.
Basic PCGASM description added to manual.
add DMDAVecGet/RestoreArrayReadF90()
...

show more ...

Merge branch 'maint'

Merge branch 'barry/maint/mv-matnullspace-to-mat' into maint

One now passes the nullspace in the matrix for a linear system as the first Amat argument to KSPSetOperators, SNESSetOperators, TSSetIJacobian() not the second pmat argument

This allows nesting of K

One now passes the nullspace in the matrix for a linear system as the first Amat argument to KSPSetOperators, SNESSetOperators, TSSetIJacobian() not the second pmat argument

This allows nesting of KSP/PC where the matrix is singular but the matrix passed to construct the preconditioner (the pmat) is not singular

Reported-by: Stephan Kramer <s.kramer@imperial.ac.uk>

show more ...

Merge remote-tracking branch 'petsc/master' into mat-increaseoverlap

Merge branch 'master' into knepley/solkx

* master: (6933 commits)
Bib: Added reference
Includegraph: Updated for compatibility with new folder layout.
Add SNESLineSearchReason to fortran inclu

Merge branch 'master' into knepley/solkx

* master: (6933 commits)
Bib: Added reference
Includegraph: Updated for compatibility with new folder layout.
Add SNESLineSearchReason to fortran includes Also truncate names to fit 32 char fortran limit
fix compile warnings /usr/home/balay/petsc.clone-2/src/vec/vec/interface/vector.c:1944: warning: division by zero in '1.0e+0 / 0.' /usr/home/balay/petsc.clone-2/src/ksp/ksp/interface/itfunc.c:508: warning: 'vec_rhs' may be used uninitialized in this function
updated output for new trust region initial size
initial size of the trust region is set as a percent of the norm of the initial guess, it should not be related to the initial norm of the function (which is kind of nuts).
The default bounds for SNESVISetVariableBounds() in the manual page were reversed from correct values
Bib: Added refs
Bib: Update MPICH webpage
Plex: Forgot to close file
removed nonexistent ex46f from makefile test rule
Revert "fixed bad merge into master"
fixed bad merge into master
fix memory leak in mkl_pardiso fix some formatting in mkl_pardiso code
mv runex111 from TESTEXAMPLES_C to TESTEXAMPLES_DATAFILESPATH
Mat+Doc: More Fortran docs
Plex ex3: Run tests with the correct number of field components
Bib: Added TetGen cite
parmetis: make sure 'ldd libparmetis.so' points to the correct libmetis.so that its linked with.
small fix as reported by the nightly test
...

Conflicts:
config/builder.py

show more ...

Merged petsc/petsc into master

Merge branch 'master' into tisaac/dmforest

directory layout changed again: merging to switching branches back and forth
doesn't cause reconfigure to have a hard time

* master: (357 commits)
remov

Merge branch 'master' into tisaac/dmforest

directory layout changed again: merging to switching branches back and forth
doesn't cause reconfigure to have a hard time

* master: (357 commits)
removed nonexistent ex46f from makefile test rule
Revert "fixed bad merge into master"
fixed bad merge into master
fix memory leak in mkl_pardiso fix some formatting in mkl_pardiso code
mv runex111 from TESTEXAMPLES_C to TESTEXAMPLES_DATAFILESPATH
Mat+Doc: More Fortran docs
Bib: Added TetGen cite
parmetis: make sure 'ldd libparmetis.so' points to the correct libmetis.so that its linked with.
small fix as reported by the nightly test
add a test for MatPtAP()
small cleanup; add tests
cleanup MatPtAPSymbolic_MPIAIJ_MPIAIJ()
Fix nonzerostate tracking in all MATMPI types.
update location of test matrices in faq.html file
unused variable still declared
Fix for MatLoad_MPIDense() from sparsely stored matrix also fix for MatLoad_MPIDense_DenseInFile() when matrix has not had its local sizes preset by user
Preserve the B nonzerostate during disassembly.
TSEvent: Fix arch-mswin nightly errors due to declaring variables in the middle of a function.
Fix error in parallel loading of dense format matrix
removed KSPSetNullSpace() since mat already carries that information and having two locations for it is confusing, unnecessary and prone to error added MatSetTransposeNullSpace() for providing the null space of the transpose of the matrix and added support in KSPSolve() to remove this null space from the right handside of the linear system thus making the system consistent.
...

show more ...

Merge branch 'barry/propagate-pcsetup-failures'

Infrastructure that allows failures in PCSetUp(), PCApply(), MatMult() etc due to, for example a zero pivot or function evaluation outside its domain, to propagate up and become a KSP_DIVERGED inste

Infrastructure that allows failures in PCSetUp(), PCApply(), MatMult() etc due to, for example a zero pivot or function evaluation outside its domain, to propagate up and become a KSP_DIVERGED instead of generating an error that stops the program. In response to Issue 96. This includes failures in MatCreateSNESMF() applications due to domain errors. The mechanism to propagate some errors is by setting Info or Nan into the output vector and using the norm or inner product reductions in SNES or KSP to propagate the error condition to all processes and then handling them immediately after the norm or inner product.

This allows, for example, ODE integrators to try again with a smaller time-step if the PCSetUp failed instead of requiring a complete restart of the run with other options.

Currently some error conditions, such as function domain error in a line search may not get progated up using the correct SNESConvergedReason

See src/snes/examples/tests/ex69.c for the handling of several conditions

show more ...

Merge branch 'master' into knepley/feature-composable-callbacks

* master: (151 commits)
PetscMalloc: Fix handling of zero-sized allocations
add note to DMGetLocal/GlobalVector() about intended s

Merge branch 'master' into knepley/feature-composable-callbacks

* master: (151 commits)
PetscMalloc: Fix handling of zero-sized allocations
add note to DMGetLocal/GlobalVector() about intended short-time usage of the vectors.
sf: use patched openmpi tarball that has a potential fix for sf example hang
SNES problems that set VI bounds must use a VI solver since the other solvers ignore the bounds
pounders viewer did not cast PetscReal to double in printf nor use %D instead of %d
fix ts ex20adj
fix ts theta
fixed another variable declaration in the middle of a code block
example main() had PetscInt return type and argument that should be int
DMDAVTKWriteAll_VTS: Fix handling of higher-dimension coordinates
fixed example that declared variables in the middle of a function
C programs do not return PetscInt nor do they take an argument that is PetscInt, fixed example that did
You cannot declare variables in the middle of functions in PETSc.
add ierr checking to PetscOptionsBegin/End() in example
configure: add error check if user specifies PETSC_ARCH=-dbg [i.e starts with a '-']
scale norm of vector difference by norm of vector to insure it is small enough to consider the two vectors equal
missing PetscViewerDestroy() added
MatNestGetISs: use custom fortranstub to handle PETSC_NULL_OBJECT
update location of open-mx website on FAQ page
If the user sets bounds and calls a solver that does not support them, throw an error.
...

show more ...

Merge branch 'pr192/Fande-Kong/scalable-matincreaseoverlap/master' of https://bitbucket.org/petsc/petsc

Merged petsc/petsc into master

Merge branch 'master' into stefano_zampini/pcbddc-adaptive-selection

Conflicts:
src/ksp/pc/impls/bddc/bddcprivate.c
src/ksp/pc/impls/bddc/bddcscalingbasic.c
src/mat/impls/aij/mpi/mumps/mumps.c
s

Merge branch 'master' into stefano_zampini/pcbddc-adaptive-selection

Conflicts:
src/ksp/pc/impls/bddc/bddcprivate.c
src/ksp/pc/impls/bddc/bddcscalingbasic.c
src/mat/impls/aij/mpi/mumps/mumps.c
src/mat/impls/is/matis.h

show more ...

Merged master into semihozmen/ksp-fcg-eigen-comp-support

Merge branch 'master' into pr192/Fande-Kong/scalable-matincreaseoverlap/master

Needed to merge master to update to new calling sequence PetscCommBuildTwoSided

Merged petsc/petsc into master

Merge branch 'barry/fix-directory-layout'

updated PETSc directory layout to match standard packaging strategies

include/petsc finclude,private,mpiuni
lib/petsc conf
bin/petsc*

Merge branch 'master' into knepley/feature-viewer-partial-read

* master: (48 commits)
ViennaCL: Fixes memory leak when destructing an aijviennacl matrix.
MatLoad_MPIDense() always allocated arra

Merge branch 'master' into knepley/feature-viewer-partial-read

* master: (48 commits)
ViennaCL: Fixes memory leak when destructing an aijviennacl matrix.
MatLoad_MPIDense() always allocated array space even if it was previously provided
Clarified lumping options for PCFieldSplitSchurPreType SELFP.
fixed up PetscInfo in GAMG
TSErrorWeightedNorm[2|Infinity]: API changes
MatLoad_SeqDense() always allocated new space for the data even if the user already provided it
Fixed DMCompositeGetAccessArray() to support readonly vectors
Fix obviously-wrong aspects of documentation on SNESVINEWTONRSLS and SNESVINEWTONSSLS.
fixed example to compile - demonstrates a bug
var name 'len' was F90 key word
For user provided Jacobian functions they should always assembly into the second matrix, the one from which the preconditioner is built because the first may be used as matrix-free with -snes_mf_operator
Suggest --with-64-bit-indices if users mistakenly use --with-64bit-indices
Fix configure logging of integer size.
missing underscore in #define of vecgetarrayread translation of fortran stub for VECGETARRAYREAD
updated test output files to reflect differences introduced previously in PCView output
-ksp_type preonly does not support -ksp_monitor_short so remove from test line to prevent extranious warning output
missing CHKERRQ; and ugly double declaration of ierr
Plex: Reorganized PlexPoint interface - Added DMPlexGetPointLocalField() and DMPlexGetPointGlobalField() - Now the DMPlexGetLocalOffset*_Private() functions return start and end - Uniformly use private functions
Plex: Put in explicit cast
MatSetValuesMPISBAIJ will now print correct global indices when error generated due to new nonzero location
...

# Please enter a commit message to explain why this merge is necessary,
# especially if it merges an updated upstream into a topic branch.
#
# Lines starting with '#' will be ignored, and an empty message aborts
# the commit.

show more ...

Merge branch 'master' into tisaac/dmforest

If we're going to start developing on this again, the recent changes to the
location of configuration files will make it annoying to switch to and from
thi

Merge branch 'master' into tisaac/dmforest

If we're going to start developing on this again, the recent changes to the
location of configuration files will make it annoying to switch to and from
this branch. Matt's and my fixes in DMPlex will also be relevant.

* master: (1557 commits)
MatLoad_MPIDense() always allocated array space even if it was previously provided
Clarified lumping options for PCFieldSplitSchurPreType SELFP.
fixed up PetscInfo in GAMG
TSErrorWeightedNorm[2|Infinity]: API changes
MatLoad_SeqDense() always allocated new space for the data even if the user already provided it
Fixed DMCompositeGetAccessArray() to support readonly vectors
Fix obviously-wrong aspects of documentation on SNESVINEWTONRSLS and SNESVINEWTONSSLS.
fixed example to compile - demonstrates a bug
var name 'len' was F90 key word
For user provided Jacobian functions they should always assembly into the second matrix, the one from which the preconditioner is built because the first may be used as matrix-free with -snes_mf_operator
Suggest --with-64-bit-indices if users mistakenly use --with-64bit-indices
Fix configure logging of integer size.
missing underscore in #define of vecgetarrayread translation of fortran stub for VECGETARRAYREAD
updated test output files to reflect differences introduced previously in PCView output
-ksp_type preonly does not support -ksp_monitor_short so remove from test line to prevent extranious warning output
missing CHKERRQ; and ugly double declaration of ierr
Plex ex3: For want of a semicolon, the kingdom was lost
MatSetValuesMPISBAIJ will now print correct global indices when error generated due to new nonzero location
completed basic use of MatSetValuesBlocked_SeqBAIJ_Inlined() by MatSetValuesBlocked_MPIBAIJ() to allow proper global indices being printed when error generated by new nonzero locations.
Added inline version of MatSetValuesBlocked_SeqBAIJ() to be used by MatSetValuesBlocked_MPIBAIJ() to print error message for nonzero block with global indices
...

show more ...

Merge branch 'karpeev/ksp-pcfieldsplit-selfp-doc-fix'

Clarified lumping options for PCFieldSplitSchurPreType SELFP.