add MatConvert_SeqAIJ_Elemental()

Merge branch 'maint'

MATMPIDENSE used in documentation in several places that should have been MATSEQDENSE

Reported-by: Dave May <dave.mayhem23@gmail.com>

Merge branch 'jed/matzerorows-nocomm'

* jed/matzerorows-nocomm:
MatZeroRows_MPIAIJ: cast to PetscInt* because clang attributes don't understand const
MatZeroRowsColumns_MPI(B)AIJ: avoid nrows*co

Merge branch 'jed/matzerorows-nocomm'

* jed/matzerorows-nocomm:
MatZeroRows_MPIAIJ: cast to PetscInt* because clang attributes don't understand const
MatZeroRowsColumns_MPI(B)AIJ: avoid nrows*comm_size complexity
MatZeroRows_MPI(B)AIJ: avoid communication when MAT_NO_OFF_PROC_ZERO_ROWS
MatZeroRows_MPI(B)AIJ: avoid nrows*comm_size complexity

show more ...

Merge branch 'barry/always-include-complex'

Merge branch 'barry/optimize-matdiagonalset'

MatZeroRows_MPIAIJ: cast to PetscInt* because clang attributes don't understand const

Clang provides the attribute pointer_with_type_tag, but even though
the argument is const in our function, the a

MatZeroRows_MPIAIJ: cast to PetscInt* because clang attributes don't understand const

Clang provides the attribute pointer_with_type_tag, but even though
the argument is const in our function, the attribute does not
recognize this and thus warns about a mismatch.

Reported-by: Barry Smith <bsmith@mcs.anl.gov>

show more ...

MatZeroRowsColumns_MPI(B)AIJ: avoid nrows*comm_size complexity

Switch from bad "trick" to short-circuited binary search.

MatZeroRows_MPI(B)AIJ: avoid communication when MAT_NO_OFF_PROC_ZERO_ROWS

This functionality existed prior to Matt's refactor to use PetscSF and
was considered useful to PFLOTRAN.

Reported-by: Glen

MatZeroRows_MPI(B)AIJ: avoid communication when MAT_NO_OFF_PROC_ZERO_ROWS

This functionality existed prior to Matt's refactor to use PetscSF and
was considered useful to PFLOTRAN.

Reported-by: Glenn Hammond <gehammo@sandia.gov>

show more ...

MatZeroRows_MPI(B)AIJ: avoid nrows*comm_size complexity

If the rows are in reverse order, the old implementation would linear
search the ownership ranges for each row. Use binary search instead.

CUDA: Added PETSC_SKIP_COMPLEX to remaining .cu files

Merge branch 'maint'

Merge branch 'hzhang/mataxpy-bugfix' into maint

reduce code duplication for MatAXPYGetPreallocation_MPIXXX()

Merge branch 'maint'

reduce code duplication for MatAXPYGetPreallocation_SeqXXX()

pastix: fix compile errors [linux-cxx-cmplx-pkgs-64idx]

cannot use DCOMPLEX from pastix.h - so use PetscScalar directly

bugfix for MatAXPY_ for mpibaij and mpisbaij

bugfix for MatAXPY_ for seqbaij and seqsbaij

added MatDiagonalSet_MPIAIJ()
removes searches for square matrices that already have diagonal entries

Merge branch 'maint'

with with -> with the the -> the

Delete commented out code in cuda related files.

Note that there are still a fairly large number of commented out chunks
of code in the rest of the code base:

[dmeiser@ivy petsc]$ grep -R '#if\s0'

Delete commented out code in cuda related files.

Note that there are still a fairly large number of commented out chunks
of code in the rest of the code base:

[dmeiser@ivy petsc]$ grep -R '#if\s0' * |grep '^src\|include'| wc -l
87

show more ...

Adding type checks for Vector types in MatSolve_SeqAIJCUSPARSE.

Short circuit for zero matrices in MatMul_SeqAIJCUSP and MatMulAdd_SeqAIJCUSP.

As of cusp 0.4.0 the case of zero matrices is not handled well: cusp
produces segfaults. Adding checks to guard agains

Short circuit for zero matrices in MatMul_SeqAIJCUSP and MatMulAdd_SeqAIJCUSP.

As of cusp 0.4.0 the case of zero matrices is not handled well: cusp
produces segfaults. Adding checks to guard against this case.

show more ...