Skip to content


Subversion checkout URL

You can clone with
Download ZIP

Comparing changes

Choose two branches to see what’s changed or to start a new pull request. If you need to, you can also compare across forks.

Open a pull request

Create a new pull request by comparing changes across two branches. If you need to, you can also compare across forks.
Commits on Nov 04, 2013
@prbrune prbrune Simplification of graph creation in PCGAMG_Classical 550383e
@prbrune prbrune Cleanup of PCGAMGClassical Prolongator creation. 167fb78
@prbrune prbrune PCGAMG_Classical Separate thresholding for graph coarsening and prolo…
@prbrune prbrune Merge branch 'madams/gamg-destroy' into prbrune/pcgamg-classicalinter…
Commits on Nov 06, 2013
@prbrune prbrune PCGAMG_Classical: Begin standard interpolation; match direct.
This matches the direct interpolation previously implemented as
a bug check.  The next step is to add the indirect entries.
Commits on Nov 07, 2013
@prbrune prbrune PCGAMGClassical: Standard interpolation now works for scalar problems. ed4e82e
Commits on Nov 19, 2013
@prbrune prbrune PCGAMG_Classical: Add thresholding for the prolongator entries 586a838
@prbrune prbrune PCGAMG_Classical: Added PCGAMGClassicalSetType() and registration 8eab0cc
@prbrune prbrune PCGAMG_Classical: Fixes for complex and 64-bit index builds in the ne…
…w interp.
@prbrune prbrune PCGAMG_Classical: Added Jacobi smoothing for the OptProl part b4dc3eb
Commits on Dec 17, 2013
@prbrune prbrune Merge branch 'madams/gamg-destroy' into prbrune/pcgamg-classicalinter…
@prbrune prbrune PCGAMG_Classical: PetscOptionsList() -> PetscOptionsFList() b54f5b7
@prbrune prbrune PCGAMG_Classical: Leave diagonal of the graph intact.
This was causing VecAssemblyBegin() to hang in the MIS coarsening code.
@prbrune prbrune PCGAMG_Classical: update test output 62eb343
Commits on Dec 19, 2013
@prbrune prbrune PCGAMG_Classical: Fix for Windows complex build errors
C:\cygwin\home\sbalay\PETSC~2.CLO\src\ksp\pc\impls\gamg\classical.c(639) : error C2784: '_Ty __cdecl std::abs(const class std::complex<_Ty> &)' : could not deduce template argument for 'const class std::complex<_Ty> &' from 'double'
C:\cygwin\home\sbalay\PETSC~2.CLO\src\ksp\pc\impls\gamg\classical.c(640) : error C2784: '_Ty __cdecl std::abs(const class std::complex<_Ty> &)' : could not deduce template argument for 'const class std::complex<_Ty> &' from 'double'
C:\cygwin\home\sbalay\PETSC~2.CLO\src\ksp\pc\impls\gamg\classical.c(1127) : error C2440: '=' : cannot convert from 'int (__cdecl *)(struct _p_PC *,struct _p_Mat *,struct _p_Mat ** )' to 'int (__cdecl *)(struct _p_PC *,struct _p_Mat *const ,struct _p_Mat ** )'
        This conversion requires a reinterpret_cast, a C-style cast or function-style cast
gmakefile:135: recipe for target 'arch-mswin-cxx-cmplx/obj/src/ksp/pc/impls/gamg/classical.o' failed

Reported-by: Satish Balay <>
Commits on Jan 01, 2014
@BarrySmith BarrySmith fix %d that should be %D 7904a33
@BarrySmith BarrySmith update docs to emphasis building PETSc with C compiler, not C++ 32ec92c
@BarrySmith BarrySmith fix for removal of fortran checks in 8ab2591
@hzhang hzhang add journal paper mszm2013 to petscapp.bib a8864bb
@hzhang hzhang Merge branch 'master' of d6a1543
Commits on Jan 02, 2014
@jedbrown jedbrown petscapp.bib: add D'Amore 2011
Thanks-to: Luisa Carracciuolo <>
@prbrune prbrune Merge branch 'prbrune/pcgamg-classicalinterpolationstrategies' 3c879ca
Commits on Jan 04, 2014
@BarrySmith BarrySmith Barry is a complete idiot, PETSC_DEFAULT_REAL_PRECISION? _PRECISION? …
…Fixed to PETSC_DEFAULT_REAL and added to changes file
@BarrySmith BarrySmith renamed to updated to latest PE…
…TSc etc
@BarrySmith BarrySmith enhancements to FASTMathInstaller 6065892
@BarrySmith BarrySmith slightly more details on the FASTMath installer, still pretty worthless ebe93c3
Commits on Jan 05, 2014
@BarrySmith BarrySmith Merge branch 'barry/fix-percent-d' 4f15227
@BarrySmith BarrySmith a couple more non-ISO 90 syntax things allowed in examples e89d78d
@BarrySmith BarrySmith go back to the use of MPICH that I changed wrongly in 7904a33 06283fd
Commits on Jan 06, 2014
@balay balay configure: add finishing timestamp for the configure completion.
'Starting' timestamp is set inconfig/BuildSystem/config/ - but
its not clear if finishing time stamp can also be set there [even when
errors occur]. So setting the timestamp in config/ instead.
@balay balay make.log add in 'Finishing' timestamp 44d0de3