Merge branch 'tisaac/dmforest' into mlange/fix-plex-multi-level-overlap

Conflicts:
src/dm/impls/plex/examples/tests/output/ex12_5.out
src/dm/label/dmlabel.c

Merge branch 'pr413/scools/siegfriedcools/pipecgrr/master'

Merge branch 'pr403/pascgeopc/psanan/ksp-pipe-flex/master'

Added pipecgrr linear solver

Merge branch 'master' into barry/add-pcfieldsplit-logevent

KSPPIPEFCG, KSPPIPEGCR, KSPFCG: remove macro

This removes the macro to compute the number of previous search directions to use. This is code duplication, and perhaps an inline function added to a he

KSPPIPEFCG, KSPPIPEGCR, KSPFCG: remove macro

This removes the macro to compute the number of previous search directions to use. This is code duplication, and perhaps an inline function added to a header (perhaps include/petsc/private/kspimpl.h) might be an alternative.

show more ...

KSPFCG, KSPPIPEFCG, KSPPIPEGCR: fixes to respect changes in naming conventions for truncation strategies, and removal of counter of search space resets

Merge branch 'tisaac/dmforest' of bitbucket.org:petsc/petsc into tisaac/dmforest

KSP: Add KSPPIPEFGMRES

KSPPIPEFGMRES is modification of the pipelined GMRES algorithm described in

Ghysels, P., Ashby, T. J., Meerbergen, K., & Vanroose, W. (2013). Hiding Global Communication Late

KSP: Add KSPPIPEFGMRES

KSPPIPEFGMRES is modification of the pipelined GMRES algorithm described in

Ghysels, P., Ashby, T. J., Meerbergen, K., & Vanroose, W. (2013). Hiding Global Communication Latency in the GMRES Algorithm on Massively Parallel Machines. SIAM J. Sci. Comput., 35(1), 48–71

to include variable preconditioning, using the same approach as employed in extending GMRES to FGMRES.

Note that it includes a ``shift'' parameter, in contrast to the ``explicitly normalized'' algorithm implemented in KSPPGMRES.

For more, see
"Pipelined, Flexible Krylov Subspace Methods"
P. Sanan, S. M. Schnepp, D. A. May

show more ...

KSP: Add KSPPIPEGCR

Implements the PIPEGCR method, as described in
"Pipelined, Flexible Krylov Subspace Methods"
P. Sanan, S. M. Schnepp, D. A. May

KSP: Add KSPPIPEFCG

Implements the PIPEFCG algorithm as described in
"Pipelined, Flexible Krylov Subspace Methods"
P. Sanan, S. M. Schnepp, D. A. May

KSPFCG - Define truncation strategies to be used with related solvers

DM: create DMProjectField() from DMPlexProjectField()

Merge branch 'master' into barry/add-pcreset-hypre-for-master

Needed to move PCDestroy_HYPRE() added to master to PCReset_HYPRE() since that was introduced
as a bug fix for maint

Plex: Fixed DMPlexProjectField()

Merge branch 'master' into tisaac/dmforest

* master: (177 commits)
DT ex3: The test should check the number of digits requested
DT ex3: One test fails in single precision
Plex: Do not automati

Merge branch 'master' into tisaac/dmforest

* master: (177 commits)
DT ex3: The test should check the number of digits requested
DT ex3: One test fails in single precision
Plex: Do not automatically create a default section if no fields are defined
Plex: DMPlexCopyCoordinates() needs to create the coordinate section if it is missing
superlu: no longer requires xerbla slamch dlamch ?
superlu: update to v5.1
skip MatSolve() if matrix factorization fails
replacing SETERRQ with error output info and continue execution
SNES ex12: Added parallel tests - These test parallel, unstructured interpolation/restriction using FAS
Plex+SNES: Now in DMPlexComputeJacobian_Internal(), we also form the Jacobian preconditioner matrix
FE: Now we can form either the Jacobian or the Jacobian preconditioned matrix using a flag
DS: Now have point wise function slots for the Jacobian preconditioned matrix - This will allow us to form the "right" operator for preconditioning, like M in the (2,2) block for Stokes
Plex: When swapping Plexes, also swap point SF
Plex: Fix parallel error in Injector
Mat+Plex: Added MatPreallocator which can be called with MatSetValues() and preallocates another Mat - This could have been another "mode" of Mat, but I thought this was cleaner - Use MatPreallocator to preallocator Plex interpolators
DT ex3: Change tolerance and eliminate tests which need MPFR from regular runs
superlu: fix complex build
superlu: update to using 5.0 tarball with the fix for -Werror=format-security
superlu: update to use tarball with -Werror=format-security fix
DM: Clone the coordinate DM instead of reusing in Clone()
...

Conflicts:
config/builder.py
src/dm/impls/plex/plexcreate.c

show more ...

Merge branch 'master' into hongzh/checkpointing

Manual merges for a couple of PetscOptions changes in master

Merge branch 'pr391/zianekhodja/lilia/tsirm/master'

Merge branch 'master' into pr391/zianekhodja/lilia/tsirm/master

Merge branch 'balay/add-KSPComputeRitz-patch'

Computation of Ritz pairs within GMRES

Signed-off-by: Sylvain Mercier <sylvainmercier85@gmail.com>

rm KSPConveredSubReason

add enum KSPConvergedSubReason

Merge branch 'tisaac/dm-label-promote' into tisaac/dmforest

* tisaac/dm-label-promote: (319 commits)
snes tutorial ex69: give mesh a name for regression purposes
DM: make DMLabel labels a part o

Merge branch 'tisaac/dm-label-promote' into tisaac/dmforest

* tisaac/dm-label-promote: (319 commits)
snes tutorial ex69: give mesh a name for regression purposes
DM: make DMLabel labels a part of all DMs
builder: update regression tests
builder: update some missing package requirements
DM: set coordinate vector name
ts tutorials ex11_sa: make it compile
mark variable as unused so compiler doesn't warn about unused value
configure: use 'git cat-file -e' to check for existance of a commit in localrepo. ['git rev-parse' does not give error for a hash thats not in repo]
petsc4py: revert to manually syncing petsc4py with petsc master changes
configure: revert gitcommit default value to Null. If a pacakge using giturl does not change this value - raise an error
fix: revert change in a44bcdf that breaks with pgf90
update test output for fixed to hypre interface that now actually pass the interp vectors to BoomerAMG
Error message should say use negative threshold value to prevent filtering out zeros
Calls to HYPRE_BoomerAMGSetInterpVectors() did not actually copy the vector data pointer from the PETSc vector to the hypre vector thus resulting in the hypre interp vectors being identically zero and hence not affecting the quality of the AMG hierarchy produced/
DMPlex: Silence overzealous compiler warnings
add row/col orientation to MatConvert_Elemental_Dense
fix formatting error in previous commit [broke nightlybuilds]
fix formating of TSTHETA manual page and add missing command line option
run hwloc with MPIEXEC and binding so it runs on the correct machine
fix: memory leaks [in valgrind test]
...

show more ...

Merge branch 'master' into barry-allow-vecset-nan-3.6