MatMultAdd_SeqAIJCUSP(): Fixed incorrect calculation when using the compressed row format.

Instead of z = y + Ax, the implementation computed z = Ax,
which caused MPI-based calculations with more th

MatMultAdd_SeqAIJCUSP(): Fixed incorrect calculation when using the compressed row format.

Instead of z = y + Ax, the implementation computed z = Ax,
which caused MPI-based calculations with more than one rank to fail.

show more ...

Merge branch 'master' into barry/wirth-fusion-materials
bad manual merge needs to be fixed
this is what happens when branches hang around too long

Conflicts:
src/ts/examples/tutorials/advection-dif

Merge branch 'master' into barry/wirth-fusion-materials
bad manual merge needs to be fixed
this is what happens when branches hang around too long

Conflicts:
src/ts/examples/tutorials/advection-diffusion-reaction/ex10.c

show more ...

aijcusp: Freeing up temporary buffers before reallocation.

A CUSP matrix uses internal index buffers and temporary vectors.
The old implementation reallocated these buffers with each copy to the dev

aijcusp: Freeing up temporary buffers before reallocation.

A CUSP matrix uses internal index buffers and temporary vectors.
The old implementation reallocated these buffers with each copy to the device,
thus leaking memory. This commit fixes these leaks.

show more ...

Merge branch 'jose/mat-vec-fixes'

* jose/mat-vec-fixes:
Correctly check input type of Mat argument in MatMPIAIJGetLocalMat
Fix Matlab viewer for parallel complex vectors

Conflicts:
src/vec/vec

Merge branch 'jose/mat-vec-fixes'

* jose/mat-vec-fixes:
Correctly check input type of Mat argument in MatMPIAIJGetLocalMat
Fix Matlab viewer for parallel complex vectors

Conflicts:
src/vec/vec/impls/mpi/pdvec.c

show more ...

Mat: Fix MatrixMarket format
- It is 1-based, and the PETSc header does not belong

Merge branch 'barry/compressed-coloring-only-for-square-matrices'

Correctly check input type of Mat argument in MatMPIAIJGetLocalMat

MatMPIAIJGetLocalMat requires MPIAIJ matrix as input, but it used
a explicit cast in the declaration of variables, before checking

Correctly check input type of Mat argument in MatMPIAIJGetLocalMat

MatMPIAIJGetLocalMat requires MPIAIJ matrix as input, but it used
a explicit cast in the declaration of variables, before checking that
the input Mat is actually of this type.

show more ...

Mat: Fixed bug in MatZeroRowsColumns()
- In MPIAIJ, the scatter of rhs values was using ADD instead of INSERT

Merge branch 'barry/fix-matlab'

fix a couple compiler warnings/errors (depending on compiler) for Matlab engine code

cannot use inode compressed colorings on non-square matrices

seqaij complex matrices printed in ascii full form only print imaginary part if some are nonzero
two examples had integer complex operations with c++ complex does not like

Merge branch 'barry/fix-mpiint-petscoptionsgetint'

Conflicts:
src/ksp/pc/impls/redistribute/redistribute.c
src/mat/impls/aij/mpi/mpiaij.c
src/mat/impls/dense/mpi/mpidense.c
src/mat/utils/matstas

Merge branch 'barry/fix-mpiint-petscoptionsgetint'

Conflicts:
src/ksp/pc/impls/redistribute/redistribute.c
src/mat/impls/aij/mpi/mpiaij.c
src/mat/impls/dense/mpi/mpidense.c
src/mat/utils/matstash.c
src/snes/examples/tutorials/ex28.c
src/sys/objects/pinit.c
src/vec/is/ao/impls/memscalable/aomemscalable.c

show more ...

Sys: drop explicit type arguments from PetscNew() and PetscNewLog()

git grep -l 'PetscNew\(Log\)\?(' -- $1 | xargs perl -pi -e '
s@PetscNew\([^,;()]+ *, *@PetscNew(@;
s@PetscNewLog\(([^,;()]+) *,[^,

Sys: drop explicit type arguments from PetscNew() and PetscNewLog()

git grep -l 'PetscNew\(Log\)\?(' -- $1 | xargs perl -pi -e '
s@PetscNew\([^,;()]+ *, *@PetscNew(@;
s@PetscNewLog\(([^,;()]+) *,[^,;()]+, *@PetscNewLog($1,@'

show more ...

Sys: add PetscCalloc[1-7]

Replace lots of places where PetscMemzero is called immediately after
allocation.

Sys: add PetscMalloc1 macro, array allocation without redundant types

The type is inferred from the pointer return type. This patch is
automated via the following script:

git grep -l 'PetscMalloc(

Sys: add PetscMalloc1 macro, array allocation without redundant types

The type is inferred from the pointer return type. This patch is
automated via the following script:

git grep -l 'PetscMalloc(.*sizeof' src | xargs perl -pi -e 's@PetscMalloc\(([^,;]*[^,; ]) *\* *sizeof\([^,;()]+\),@PetscMalloc1($1,@'

This commit contains an additional bug-fix in csrperm.c, fixing pointer
arity. The code was introduced in 2006, but the allocation could not
have been correct at any time. This probably means that
MatDuplicate_SeqAIJPERM has never been tested.

a54129beb540034ba105796c682d589e7e1111f2
Richard Tran Mills <rmills@ornl.gov>

Added MATSEQCSRPERM support for MatDuplicate() and conversion to/from
MATSEQAIJ. Note that these changes are not quite debugged.

show more ...

PetscMalloc[2-7]: remove type arguments, infer from pointer type

This change can be performed almost entirely by this script:

git grep -l PetscMalloc[2-7] $1 | xargs perl -pi \
-e '
s@PetscMalloc

PetscMalloc[2-7]: remove type arguments, infer from pointer type

This change can be performed almost entirely by this script:

git grep -l PetscMalloc[2-7] $1 | xargs perl -pi \
-e '
s@PetscMalloc2\(([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+)\)@PetscMalloc2($1,$3,$4,$6)@;
s@PetscMalloc3\(([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+)\)@PetscMalloc3($1,$3,$4,$6,$7,$9)@;
s@PetscMalloc4\(([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+)\)@PetscMalloc4($1,$3,$4,$6,$7,$9,$10,$12)@;
s@PetscMalloc5\(([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+)\)@PetscMalloc5($1,$3,$4,$6,$7,$9,$10,$12,$13,$15)@;
s@PetscMalloc6\(([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+)\)@PetscMalloc6($1,$3,$4,$6,$7,$9,$10,$12,$13,$15,$16,$18)@;
s@PetscMalloc7\(([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+),([^,;()]+)\)@PetscMalloc7($1,$3,$4,$6,$7,$9,$10,$12,$13,$15,$16,$18,$19,$21)@;
'

show more ...

fixed many examples that worked incorrectly for 64 bit indices

added FILE* argument to PetscSynchronizedFlush() so it knows where to flush the off process values

Previously if process 0 did not call PetscSynchronizedFPrintf() then it output the off process valu

added FILE* argument to PetscSynchronizedFlush() so it knows where to flush the off process values

Previously if process 0 did not call PetscSynchronizedFPrintf() then it output the off process values
to stdout even if the file was not stdout. Also affected PetscViewerASCIISynchronizeFlush()

Reported-by: Randall Mackie <rlmackie862@gmail.com>

show more ...

since external packages use a variety of different types for 64 bit indices
(though all have the same bit values) casts are needed for some compilers to
prevent warning messages when using 64 bit ind

since external packages use a variety of different types for 64 bit indices
(though all have the same bit values) casts are needed for some compilers to
prevent warning messages when using 64 bit indices

show more ...

Merge branch 'knepley/fix-mat-zero-rows'

* knepley/fix-mat-zero-rows:
Mat: Replaced communication in MatZeroRows_MPIBAIJ() with PetscSF - Tested with Mat ex12
Mat: Cast to remove warning
Mat:

Merge branch 'knepley/fix-mat-zero-rows'

* knepley/fix-mat-zero-rows:
Mat: Replaced communication in MatZeroRows_MPIBAIJ() with PetscSF - Tested with Mat ex12
Mat: Cast to remove warning
Mat: Fix MatZeroRows_MPIAIJ() - Must initialize local rows with -1 instead of 0
MatZeroRows_MPIAIJ: add missing PetscSFReduceEnd
Mat: Rewrote MPIAIJ MatZeroRows() to use PetscSF - Passes test ex12

show more ...

I have no freaking idea what git did or why this file is changed but Jed
is most definitely wrong about thinking having unmerged branches to master hanging around
being harmless. Not everyone remembe

I have no freaking idea what git did or why this file is changed but Jed
is most definitely wrong about thinking having unmerged branches to master hanging around
being harmless. Not everyone remembers everything they did two months ago or two hours ago.

show more ...

Merge remote-tracking branch 'origin/barry/remove-timelogging'

Merge remote-tracking branch 'origin/barry/fix-mat-checkinode-factorlu'

Conflicts:
src/mat/impls/aij/seq/aij.c

fix incorrect caste to superlu_dist int_t type when using 64 bit indices