Skip to content

HTTPS clone URL

Subversion checkout URL

You can clone with HTTPS or Subversion.

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.
This comparison is big! We're only showing the most recent 250 commits
Commits on Jan 30, 2014
@knepley knepley Sys: Fixed PetscFree to account for some NULL pointers e38f91f
@knepley knepley Merge branch 'knepley/malloc-0size-null' into next
* knepley/malloc-0size-null:
  Sys: Fixed PetscFree to account for some NULL pointers
0da7f9f
@knepley knepley KSP: It looks like someone forgot this declaration b2a7426
@knepley knepley KSP: Fix for name change in another branch 79bd946
@knepley knepley Merge branch 'knepley/feature-schurm-explicit-op' into next
* knepley/feature-schurm-explicit-op:
  KSP: Fix for name change in another branch
  KSP: It looks like someone forgot this declaration
da31d23
@knepley knepley Sys: Fuck Macros and their Unholy Enabler, The Preprocessor 06c0a5e
@knepley knepley Merge branch 'knepley/malloc-0size-null' into next
* knepley/malloc-0size-null:
  Sys: Fuck Macros and their Unholy Enabler, The Preprocessor
9d45432
@jedbrown jedbrown PetscFreeAlign: avoid unshifting NULL
When malloc has poor alignment and memalign is not available,
PetscMallocAlign shifts to retain alignment, tucking the actual offset
into a hidden variable.  It must be safe to pass NULL.

Reported-by: Matthew G. Knepley <knepley@gmail.com>
103af89
@jedbrown jedbrown Merge branch 'jed/malloc-zero' into next
* jed/malloc-zero:
  PetscFreeAlign: avoid unshifting NULL
dbdaaaa
@BarrySmith BarrySmith remove taolapack.h it duplicated prototypes already in petscblaslapack.h 04a443e
@knepley knepley Sys: More from the grimy underbelly of text substitution 3589d31
@knepley knepley Merge branch 'knepley/malloc-0size-null' into next
* knepley/malloc-0size-null:
  Sys: More from the grimy underbelly of text substitution
985969d
@BarrySmith BarrySmith merged TAO_INFINITY and SNES_VI_INF into PETSC_INFINITY e270355
Commits on Jan 31, 2014
@jedbrown jedbrown VecScatterCreateToZero: fix man page
Reported-by: Dave May <dave.mayhem23@gmail.com>
7e441cd
@knepley knepley PetscFE: Correct sizes for f0, f1 arrays
- Note that the sizes for OpenCL will still use dim
c5071df
@knepley knepley PetscFE: Fix loop order for Jacobian calculation 0f61caa
@knepley knepley SNES ex62: Fix syntax for test output c1e9349
@knepley knepley Merge branch 'knepley/fix-dtfe-jacobian-loops' into next
* knepley/fix-dtfe-jacobian-loops:
  SNES ex62: Fix syntax for test output
  PetscFE: Fix loop order for Jacobian calculation
  PetscFE: Correct sizes for f0, f1 arrays - Note that the sizes for OpenCL will still use dim
633fb24
@knepley knepley DMPlex: Fix hybrid 2d simplex refinement
- If hybrid end face is inverted, flip the new hybrid cells created
- Fix supports for this flip
ea00e70
@knepley knepley Merge branch 'knepley/fix-hybrid-tet-refinement' into next
* knepley/fix-hybrid-tet-refinement:
  DMPlex: Fix hybrid 2d simplex refinement - If hybrid end face is inverted, flip the new hybrid cells created - Fix supports for this flip
7933446
@jedbrown jedbrown PCMG: fix Galerkin when restriction not set
Reported-by: Torquil Macdonald Sørensen <torquil@gmail.com>
9f3d9e6
@jedbrown jedbrown Merge branch 'jed/pcmg-galerkin-rap' into next
* jed/pcmg-galerkin-rap:
  PCMG: fix Galerkin when restriction not set
a0efd8b
@BarrySmith BarrySmith move tao vector operations over to Vec directory, fix a couple names …
…and calling sequences
4473680
@BarrySmith BarrySmith stop using taosolver.h includes f273b95
@BarrySmith BarrySmith added petsc to front of tao public includes ba92ff5
@BarrySmith BarrySmith completed renaming of tao include files 21ec2d5
@BarrySmith BarrySmith removed taoinitialize and taofinalize b008026
@BarrySmith BarrySmith changed TaoSolver usage to Tao 441846f
Commits on Feb 01, 2014
@jedbrown jedbrown DMPlex: fix %d for PetscInt 796f034
@jedbrown jedbrown DMPlexStratify: use Function() so man pages have links 150b719
@jedbrown jedbrown Merge branch 'irving/doc-cleanup' into next
* irving/doc-cleanup:
  DMPlexStratify: use Function() so man pages have links
  DMPlex: fix %d for PetscInt
  Mention cell number in another DMPlexComputeCellGeometry error
  DMPlex: Don't imply we can't handle triangles with 3D coordinates
  snes ex62: Fix incompressibility equation typo
  DMPlexStratify: Clarify documentation
  plex.c: Fix typo: nubmer -> number
  manual: Fix typo (liek -> like)
  vsectionis.c: Fix a comment typo
  manual: Fix a grammatical error

Conflicts:
	src/dm/impls/plex/plex.c
	src/dm/impls/plex/plexgeometry.c
ce88b05
@jedbrown jedbrown Merge branch 'juhaj/hdf5-chunking'
HDF5 viewer: improved chunking, IO performance, and scalability with
respect to HDF5 limitations.

* juhaj/hdf5-chunking:
  VecView HDF5: PETSc style, lispy closing parens
  VecView HDF5: fix -Wsign-compare (comparing signed vs unsigned)
  VecView HDF5: static function, cast, error handling
  Some minor fixes in comments, changed some float divisions to integer divisions and added error checking.
  First attempt at sanitising chunking.
1f49d5c
@jedbrown jedbrown Merge branch 'jose/mat-vec-fixes' into maint
* jose/mat-vec-fixes:
  Correctly check input type of Mat argument in MatMPIAIJGetLocalMat
  Fix Matlab viewer for parallel complex vectors
7a2f779
@BarrySmith BarrySmith fixes to tao to compile in single precision ba4b436
@BarrySmith BarrySmith set tao tolerances for single precision
BTW: there are all kinds of crazy tolerances all over the place; needs to be rationalized
6f4723b
@BarrySmith BarrySmith fixed more Tao print format statements 335036c
@BarrySmith BarrySmith need to cast PetscReal to double in PetscInfo calls fe386f6
@BarrySmith BarrySmith added PetscStackCallBLAS() wrappers in tao code
removed all CHKMEMQ macros in tao code that have PetscStack calls that also check the memory
0cbffdb
Commits on Feb 02, 2014
@BarrySmith BarrySmith update to tao users manual for integration into PETSc repository b4bb1fb
@BarrySmith BarrySmith Merge branch 'master' into balay/tao-to-petsc b2bbe45
@BarrySmith BarrySmith Merge branch 'balay/tao-to-petsc' into next
Conflicts:
	src/dm/examples/tests/ex42.c
	src/dm/examples/tutorials/ex65dm.c
	src/ksp/ksp/examples/tutorials/ex42.c
	src/ksp/ksp/impls/cg/stcg/stcg.c
	src/snes/examples/tutorials/cahn-hilliard-schemes/ex1.c
	src/snes/examples/tutorials/ex54.c
	src/snes/examples/tutorials/ex55.c
	src/snes/examples/tutorials/ex60.c
	src/snes/examples/tutorials/ex61.c
	src/snes/examples/tutorials/ex63.c
	src/snes/examples/tutorials/ex633d_db.c
	src/snes/examples/tutorials/ex64.c
	src/snes/examples/tutorials/ex65.c
	src/snes/examples/tutorials/ex653d.c
	src/snes/impls/fas/fasfunc.c
	src/snes/impls/fas/fasgalerkin.c
	src/ts/examples/tutorials/ex23.c
e8624b4
@BarrySmith BarrySmith git sucks 27ea85c
@BarrySmith BarrySmith remove wierd handling of options and options prefixes in Tao 72b7fd4
@BarrySmith BarrySmith Merge branch 'balay/tao-to-petsc' into next 9e05ea1
@balay balay docs: update header_TaoSolver -> header_Tao 1e655f3
@balay balay docs: fix 'Concepts' keyword in tao examples 01b3cee
@knepley knepley DMPlex: Fix for 3d hybrid simplicial refinement
- Fixed the order in which hybrid faces are attached to the cell
3b61eb6
@knepley knepley DMPlex ex4: Fixed test output for hybrid tet refinement faae8ed
@knepley knepley Merge branch 'knepley/fix-hybrid-tet-refinement' into next
* knepley/fix-hybrid-tet-refinement:
  DMPlex ex4: Fixed test output for hybrid tet refinement
  DMPlex: Fix for 3d hybrid simplicial refinement - Fixed the order in which hybrid faces are attached to the cell
d05d7b7
@knepley knepley TS: Remove use of DMADDA in the F90 module
- One could complain that irresponsible individuals are checking broken shit into next, but that is passe
ab8eaee
@knepley knepley Merge branch 'barry/rm-pcasa' into next
* barry/rm-pcasa:
  TS: Remove use of DMADDA in the F90 module - One could complain that irresponsible individuals are checking broken shit into next, but that is passe
03754d9
@BarrySmith BarrySmith starting to add optimized MatSeqAIJSetValuesLocalFast() bd04181
Commits on Feb 03, 2014
@jedbrown jedbrown TAO: s/PETSC_NULL/NULL/ in manual
Touches a few non-TAO files.
7afc1a8
@jedbrown jedbrown Fix <> includes of local files [bug in parent commit] 1c7d246
@jedbrown jedbrown Merge commit 'Fix <> includes of local files [bug in parent commit]' …
…into balay/tao-to-petsc
54ef5d7
@jedbrown jedbrown Merge branch 'balay/tao-to-petsc' into next
* balay/tao-to-petsc:
  Fix <> includes of local files [bug in parent commit]
  TAO: s/PETSC_NULL/NULL/ in manual
  docs: fix 'Concepts' keyword in tao examples
  docs: update header_TaoSolver -> header_Tao

Conflicts:
	src/snes/examples/tutorials/cahn-hilliard-schemes/ex1.c
1a95a80
Jason Sarich tao: add max_it to least-square examples to prevent runaways c320ee0
@knepley knepley DMPlex: Documentation and declaration for creation functions 26492d9
@knepley knepley Merge branch 'knepley/malloc-0size-null'
* knepley/malloc-0size-null:
  Sys: More from the grimy underbelly of text substitution
  Sys: Fuck Macros and their Unholy Enabler, The Preprocessor
  Sys: Fixed PetscFree to account for some NULL pointers
  Sys: Always return NULL for a 0 size allocation
7db01ba
@knepley knepley Merge branch 'master' into next
* master:
  DMPlex: Documentation and declaration for creation functions

Conflicts:
	src/dm/impls/plex/plexcreate.c
1232f7a
@knepley knepley Fix include format 3d13b8f
@knepley knepley DMPlex: Fix compiler warning 3e86960
@knepley knepley Merge branch 'knepley/fix-include-paths' into next
* knepley/fix-include-paths:
  DMPlex: Fix compiler warning
  Fix include format

Conflicts:
	src/mat/impls/aij/mpi/mpicusparse/mpiaijcusparse.cu
	src/mat/impls/aij/seq/seqcusparse/aijcusparse.cu
6b3b755
@knepley knepley Merge branch 'knepley/fix-dtfe-jacobian-loops'
* knepley/fix-dtfe-jacobian-loops:
  SNES ex62: Fix syntax for test output
  PetscFE: Fix loop order for Jacobian calculation
  PetscFE: Correct sizes for f0, f1 arrays - Note that the sizes for OpenCL will still use dim
e90cffd
@knepley knepley Merge branch 'knepley/feature-schurm-explicit-op'
* knepley/feature-schurm-explicit-op:
  KSP: Fix for name change in another branch
  KSP: It looks like someone forgot this declaration
  Mat: Improve conversion of MATDENSE to MATAIJ - Previously, we were creating a fully dense AIJ matrix
  Section: Small doc fix
  PC: Allow FieldSplit to explicitly compute S for preconditioning - Yes I know it is generally slow
  KSP: Added MatSchurComplementComputeExplicitOperator - This could be optimized to compute a sparse S for preconditioning (Saad has some papers here)
296d98f
@knepley knepley Bib: New reference 801dc27
@knepley knepley Merge branch 'knepley/feature-periodic-mesh'
* knepley/feature-periodic-mesh:
  Matt moved DMDA_BOUNDARY.. to DM_BOUNDARY... but did not fix ex2f.F includes
  DM Fortran examples: fix petscdm.h -> finclude/petscdm.h
  DMPlex: Fixed markers on periodic boundary
  DM examples: include petscdm.h in examples that use DM_BOUNDARY_*
  DMBoundary: fix line length for Fortran
  DMPlex: Now DMPlexCreateHexBox() can have a Mobius topology using DM_BOUNDARY_TWIST
  DM: Moved DMDABoundaryTypes to DMBoundaryTypes - Changed the enum values as well - Changed docs and updated Changes - Updated examples
  DMPlex: DMPlexCreateHexBoxMesh() now allows periodicity - Changed declaration and uses - DMPlexCreateSquareMesh() now allows periodicity - Works with refinement

Conflicts:
	src/dm/impls/plex/plexcreate.c
	src/ts/examples/tutorials/advection-diffusion-reaction/ex10.c
fbf5b48
@knepley knepley DMPlex: Fix missing declaration 1a77d57
@knepley knepley Merge branch 'master' into next
* master:
  DMPlex: Fix missing declaration
  Bib: New reference

Conflicts:
	src/ts/examples/tutorials/advection-diffusion-reaction/ex10.c
7724349
@knepley knepley Merge remote-tracking branch 'origin/karpeev/ksp-matschurcomplement'
* origin/karpeev/ksp-matschurcomplement:
  Fixed memory leak in MatSchurComplement.
  MatSchurComplementXXXSubMatrices renamed for consistency.
  PCFieldSplit: enable extraction of MatSchurComplement to configure it.
  MatSchurComplement can use different approximations to inv(A00) when forming Sp.
  PCFieldSplit: Schur can use lumping with MatSchurComplement.
  MatSchurComplement: Allow lumping of A00 when forming Sp.
  PCFieldSplit implements SELFP preconditioning for Schur using MatSchurComplementGetPmat to assemble Sp.
  MatCreateSchurComplementPmat() now public.
  MatSchurComplement now uses MatDiagonalScale to form Sp.  Factored out Sp constructor for consistency and reuse.
  More consistent matrix naming in arguments and docs to MatSchurComplement-related routines.
  MatSchurComplement can now assemble Sp on demand.

Conflicts:
	include/petscksp.h
	include/petscpc.h
	src/ksp/pc/impls/fieldsplit/fieldsplit.c
2e71c61
@prbrune prbrune Merge branch 'prbrune/matcoloring-fixminpacklf' 2151b98
Commits on Feb 04, 2014
@knepley knepley SNES ex12: Added full tests, and a p-Laplacian mode 843f6ba
@knepley knepley DM: Barry missed a %G e3556bc
@knepley knepley Merge branch 'knepley/feature-plex-fas' into next
* knepley/feature-plex-fas:
  DM: Barry missed a %G
  SNES ex12: Added full tests, and a p-Laplacian mode

Conflicts:
	config/builder.py
	include/petscksp.h
	src/ksp/pc/impls/fieldsplit/fieldsplit.c
ba42239
@knepley knepley DMShell: Added support for DMCreateSubDM()
Reported-by: Luc Berger-Vergiat <lb2653@columbia.edu>
6e44b4c
@knepley knepley Merge branch 'knepley/feature-dmshell-subdm' into next
* knepley/feature-dmshell-subdm:
  DMShell: Added support for DMCreateSubDM() Reported-by: Luc Berger-Vergiat <lb2653@columbia.edu>
2930850
@BarrySmith BarrySmith make MAT_NEW_NONZERO_LOCATION_ERR docs clearer 169f685
@BarrySmith BarrySmith PetscViewerGetSingleton() does not work in parallel for binary viewers c203d5c
@knepley knepley Sys: Fix memory leak in PetscStrreplace() 31936d5
@knepley knepley Merge branch 'knepley/fix-str-leak' into next
* knepley/fix-str-leak:
  Sys: Fix memory leak in PetscStrreplace()
  PetscViewerGetSingleton() does not work in parallel for binary viewers
ed7a3de
@knepley knepley Merge branch 'knepley/fix-hybrid-tet-refinement'
* knepley/fix-hybrid-tet-refinement:
  DMPlex ex4: Fixed test output for hybrid tet refinement
  DMPlex: Fix for 3d hybrid simplicial refinement - Fixed the order in which hybrid faces are attached to the cell
  DMPlex: Fix hybrid 2d simplex refinement - If hybrid end face is inverted, flip the new hybrid cells created - Fix supports for this flip
  DMPlex ex4: Added test for hybrid quads
  DMPlex: Added in hybrid quad refinement - Strange merge in the middle for some reason
  DMPlex: Cleanup and fix for SF construction in refinement - Had wrong numbering for vertices in quads
  DMPlex: Bug fix for hybrid tet refinement - The new edges for hybrid faces do not depend on the cap face orientation - This happens because we are not attaching hybrid faces in an order dependent on the cap face orientation. Maybe we should.
  DMPlex: Bug fix for hybrid tet refinement - The orientation that we get for hybrid faces depends on both the edge and face orientation - This happens because we are not attaching hybrid faces in an order dependent on the cap face orientation. Maybe we should.
81906eb
Commits on Feb 05, 2014
@BarrySmith BarrySmith more (double) case in print statements
some badly formatted code
001a771
Jason Sarich tao: fix pounders indexing bug babe810
@balay balay f2cblaslapack: fix quad version of ila*l[rc] i*max1 functions
Fixed-by: Eloy Romero Alcalde <elroal@upvnet.upv.es>

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

      I stumbled across bugs in f2cblaslapack-3.4.2.q/  for quad/__float128 precision that need fixing.

      In the Lapack directory ila*lc.c  ila*lr.c are single/double real/complex versions but the toclapack.sh
(or whatever generates the quad precision files) does not generate the needed ilaqlc.c  ilaqlr.c and ilawlc.c
ilawlr.c files. In addition, calls to iladlc_ and iladlr_ (same for z) in the q/w source files are not
correctly mapped to the q (same for w) versions instead they are left as is (in for example qlarf.c)

    extern integer iladlc_(integer *, integer *, quadreal *, integer *),
            iladlr_(integer *, integer *, quadreal *, integer *);

/*     Scan for the last non-zero column in C(1:lastv,:). */
            lastc = iladlc_(&lastv, n, &c__[c_offset], ldc);
        } else {
/*     Scan for the last non-zero row in C(:,1:lastv). */
            lastc = iladlr_(m, &lastv, &c__[c_offset], ldc);
        }

Thus when compiling qlarf.c the C compiler thinks iladlc_ takes a quadreal when in fact it takes a double and
errors occur at runtime in the call.

The fix would be to have the script generate the q and w versions of ila*lc.c  ila*lr.c  and when generating
the quad q*.c and w*.c files, replace the calls to iladlc_  iladlr_ (and z version) with ilqdlc_  ilqdlr_  (and
z versions)
3d838d8
@balay balay Merge branch 'balay/fix-qaud-f2cblas' into next 34baa37
Commits on Feb 06, 2014
@BarrySmith BarrySmith support MPIU_REAL,SCALAR and SUM in Fortran for quad precision 5167406
@knepley knepley DMPlex: Replaced a PetscMalloc3() with PetscMalloc1()s
- Fieldsplit was freeing these individually
03dc339
@knepley knepley Sys: Have to check for all zero sizes in PetscMalloc*()
- Produced a leak in optimized mode
ceba6b7
@knepley knepley Merge branch 'knepley/fix-str-leak' into next
* knepley/fix-str-leak:
  Sys: Have to check for all zero sizes in PetscMalloc*() - Produced a leak in optimized mode
  DMPlex: Replaced a PetscMalloc3() with PetscMalloc1()s - Fieldsplit was freeing these individually
9e85a16
@knepley knepley KSP: Fixed Fortran enum definition
- I cannot see how this did not come up in 'next'
ec475d1
@knepley knepley Web: Moved tutorial to previous, and added slides a911845
Commits on Feb 07, 2014
@BarrySmith BarrySmith support for MPIU operations for quad precision in Fortran now works 122e117
@BarrySmith BarrySmith MPIUni bindings for Fortran did not work for quad precision
  Please enter the commit message for your changes. Lines starting
1572cfd
@BarrySmith BarrySmith fortran definitions of MPI_INTEGER etc were wrong for MPIUNI thus giv…
…ing wrong results for tao fortran examples that used them in MPI_Allreduce()
bd592de
@BarrySmith BarrySmith Merge branch 'master' into balay/tao-to-petsc
Conflicts:
	src/dm/examples/tests/ex42.c
	src/dm/examples/tutorials/ex65dm.c
	src/ksp/ksp/examples/tutorials/ex42.c
	src/snes/examples/tutorials/cahn-hilliard-schemes/ex1.c
	src/snes/examples/tutorials/ex54.c
	src/snes/examples/tutorials/ex55.c
	src/snes/examples/tutorials/ex60.c
	src/snes/examples/tutorials/ex61.c
	src/snes/examples/tutorials/ex63.c
	src/snes/examples/tutorials/ex633d_db.c
	src/snes/examples/tutorials/ex64.c
	src/snes/examples/tutorials/ex65.c
	src/snes/examples/tutorials/ex653d.c
	src/ts/examples/tutorials/ex23.c
bab9b9e
@BarrySmith BarrySmith fixes after merge with master due to DMDA_BOUNDARY -> DM_BOUNDARY e8871b4
@BarrySmith BarrySmith added citation for Tao users manual when Tao solver is used e2379f4
@BarrySmith BarrySmith Merge branch 'balay/tao-to-petsc' into next
Conflicts:
	src/dm/dt/interface/dt.c
	src/dm/examples/tests/ex42.c
	src/snes/examples/tutorials/cahn-hilliard-schemes/ex1.c
6eb5e12
@BarrySmith BarrySmith new version of fast matrix setter since old version was fundamentally…
… broken
189e400
@BarrySmith BarrySmith additional optimizations (trivial) in fast aij matsetvalues f38dd0b
@BarrySmith BarrySmith completed simple ever so slightly faster specialized MatSetValues 1f763a6
Commits on Feb 08, 2014
@BarrySmith BarrySmith added citations for TS and SNES papers fffbeea
@BarrySmith BarrySmith type caste of function missing void 0a9920b
@BarrySmith BarrySmith ignore another silly Fortran compiler warning message 653fb27
@BarrySmith BarrySmith Merge branch 'barry/rm-petscsnesfas_h' b603e5f
@BarrySmith BarrySmith Merge branch 'barry/rm-pcasa'
Conflicts:
	src/ksp/ksp/examples/tutorials/ex38.c
	src/ksp/ksp/examples/tutorials/ex39.c
1f289cd
@BarrySmith BarrySmith Merge branch 'balay/tao-to-petsc'
Conflicts:
	src/snes/impls/fas/fasfunc.c
	src/snes/impls/fas/fasgalerkin.c
a7b5fb5
@d-meiser d-meiser Fix include in mpiaijcusp.cu. fbf4e39
@BarrySmith BarrySmith Merged in dmeiser/petsc/fix-mpicusparse-compilation (pull request #141)
This should fix the cuda compilation in next
c6121aa
@knepley knepley DMPlex: Fixed SF created during refinement
- Remote edge had wrong starting point
- Forgot an entry for hybrid edge
7d5cd7d
@knepley knepley SF: Added additional checking fc1ede2
@knepley knepley Merge branch 'knepley/fix-plex-refine-sf' into next
* knepley/fix-plex-refine-sf:
  SF: Added additional checking
  DMPlex: Fixed SF created during refinement - Remote edge had wrong starting point - Forgot an entry for hybrid edge
  ignore another silly Fortran compiler warning message
  added citations for TS and SNES papers

Conflicts:
	src/ts/interface/ts.c
b6ce54f
@knepley knepley Merge branch 'knepley/fix-str-leak'
* knepley/fix-str-leak:
  Sys: Have to check for all zero sizes in PetscMalloc*() - Produced a leak in optimized mode
  DMPlex: Replaced a PetscMalloc3() with PetscMalloc1()s - Fieldsplit was freeing these individually
  Sys: Fix memory leak in PetscStrreplace()
47e1e29
@knepley knepley PetscBT: Added PetscBTNegate() f47f4e3
@knepley knepley DMPlex: Added parallel stage to DMPlexOrient()
- The coarse solver is not implemented, so only works for p=2
fedde07
@knepley knepley Merge branch 'knepley/fix-plex-orient-parallel' into next
* knepley/fix-plex-orient-parallel:
  DMPlex: Added parallel stage to DMPlexOrient() - The coarse solver is not implemented, so only works for p=2
  PetscBT: Added PetscBTNegate()
40cb36e
@BarrySmith BarrySmith fix for change to include files with <> c5483a7
@BarrySmith BarrySmith Merge branch 'master' of bitbucket.org:petsc/petsc 462104a
@BarrySmith BarrySmith change PetscStrncpy() to ALWAYS null terminate result ff32304
Shri Abhyankar update-SuiteSparse: Single configure script for SuiteSparse
PETSc currently has interfaces for three packages (UMFPACK, CHOLMOD, and KLU)
from SuiteSparse and each have their own configure file.
This single SuiteSparse.py file downloads the entire SuiteSparse package and
installs the three packages (and their associated
dependencies). This also updates CHOLMOD and UMFPACK to use the latest versions
available via SuiteSparse version 4.2.1.
350ae3a
Shri Abhyankar update-SuiteSparse: Rempved UMFPACK.py and CHOLMOD.py, these packages
get installed now by SuiteSparse.py
e74fa32
@balay balay Merge branch 'balay/fix-qaud-f2cblas' e15cd23
@BarrySmith BarrySmith opstring was never initialized ecf335a
@BarrySmith BarrySmith Merge branch 'master' of bitbucket.org:petsc/petsc 7cb340a
Shri Abhyankar update-SuiteSparse: Need to also remove CHOLMOD and UMFPACK from list…
… of package names in config/BuildSystem/config/packages/__init__.py
7a617fa
@BarrySmith BarrySmith convert to PetscMalloc1()
better handling of string viewer when running out of space
89d949e
Commits on Feb 09, 2014
@BarrySmith BarrySmith decrease indefinite tolerance for minres and update example to get co…
…nvergence
6ac5689
@BarrySmith BarrySmith don't use pow() directly in source code 97bf3f2
@BarrySmith BarrySmith removed outdated contrib a127210
@BarrySmith BarrySmith add quad support for acos,asin,tgamma 0646a65
@BarrySmith BarrySmith match Jed's capitalization for Acos and Asin 255453a
@BarrySmith BarrySmith Merge branch 'master' into next
Conflicts:
	include/petscmath.h
	src/snes/impls/fas/fasfunc.c
	src/snes/linesearch/interface/linesearch.c
	src/ts/examples/tutorials/power_grid/ex2.c
	src/vec/is/sf/examples/tutorials/ex1.c
f481161
@BarrySmith BarrySmith numerical constants such as PETSC_PI, PETSC_REAL_MAX are automaticall…
…y placed in Fortran common block

now numerical constants are always equal in both languages
639f5a5
@BarrySmith BarrySmith Merge branch 'barry/add-fortran-num-consts' into next cc647ed
Commits on Feb 11, 2014
@jedbrown jedbrown Math: fix typo 3b7821d
@jedbrown jedbrown Merge branch 'prbrune/removeunwrappedmathfunctions' into next
Merge fixes conflicted definition of PetscCosComplex().

* prbrune/removeunwrappedmathfunctions:
  Math: fix typo
071be99
@prbrune prbrune MatColoring: Fix for MPI types used in PetscSF reductions
Reported-by: Jed Brown <jedbrown@mcs.anl.gov>
0c4b1c2
@prbrune prbrune Merge branch 'prbrune/matcoloring-fixmpitypes' into next 4f5edb6
@prbrune prbrune Mat: Added MatGetLayouts() which allows for access to row and column …
…layouts.
a93ff8c
@prbrune prbrune Merge branch 'prbrune/mat-matgetlayouts' into next fc5acc9
@jedbrown jedbrown make shared: use configure to determine shared linking
Now supports OS-X -dynamiclib instead of less portable -shared (which
apparently does not work with Intel 14.0).

Reported-by: Blaise A Bourdin <bourdin@lsu.edu>
accebfa
@jedbrown jedbrown make shared: default SL_LINKER_FUNCTION (revert before release)
This default allows old petscvariables to work while we transition to
using this function for linking.  This commit should be reverted before
v3.5.
1dabf71
@jedbrown jedbrown Merge branch 'jed/gnumake' into next
* jed/gnumake:
  make shared: default SL_LINKER_FUNCTION (revert before release)
  make shared: use configure to determine shared linking

Conflicts:
	gmakefile
0228242
Commits on Feb 12, 2014
@slepc slepc vec/ex1.c: Remove PetscFunctionBegin from main() ebf869b
@slepc slepc vec/ex18.c: Correctly fill the last entry of vector xend 6de8f9d
@slepc slepc ksp/ex20.c: Add MatSetUp() and fix typo f7024bc
@jedbrown jedbrown Merge branch 'jed/pcmg-galerkin-rap'
* jed/pcmg-galerkin-rap:
  PCMG: fix Galerkin when restriction not set
  PCMG Galerkin: use A_coarse = R*A*P when R != P^T
90a6321
@jedbrown jedbrown Merge remote-tracking branch 'bb/knepley/fix-include-paths'
* bb/knepley/fix-include-paths:
  DMPlex: Fix compiler warning
  Fix include format

Conflicts:
	src/mat/impls/aij/mpi/mpicusparse/mpiaijcusparse.cu
	src/mat/impls/aij/seq/seqcusparse/aijcusparse.cu
6551ba1
Shri Abhyankar update-SuiteSparse: Changed a bunch of flags.
Use PETSC_HAVE_SUITESPARSE instead of PETSC_HAVE_UMFPACK and PETSC_HAVE_CHOLMOD.
Use SUITESPARSE_INCLUDE instead of UMFPACK_INCLUDE and CHOLMOD_INCLUDE.
913c410
Commits on Feb 13, 2014
@balay balay suitesparse: configure will suggest --download-suitesparse if --downl…
…oad-umfpack is used
1699bcc
@balay balay nightlybuilds: change --download-umfpack to --download-suitesparse d882683
@BarrySmith BarrySmith remove my silly naming of monitor function passed to SNESMonitorSet()…
… as SNESMonitorFunction
6e4dcb1
@BarrySmith BarrySmith added calling sequence of user provided function to manual page for M…
…atMFFDSetFunction()
14f633e
@BarrySmith BarrySmith cleanup of SNESApplyPC docs and code 37703b4
@prbrune prbrune SNESLineSearch: Fix documentation levels to highlight standard use case
Reported-by: Barry Smith <bsmith@mcs.anl.gov>
162e0bf
@BarrySmith BarrySmith fix MATSNESMF so that MatMFFDSetFunction() works correctly with it 5eb111a
@prbrune prbrune Merge branch 'prbrune/sneslinesearch-doclevelfix' into next be94adb
@prbrune prbrune SNES: Removed now-extraneous fnorm variable from SNESApplyPC()
Since c1c7507 the fnorm argument
has been unused.  Fix documentation also.

Reported-by: Barry Smith <bsmith@mcs.anl.gov>
6a63999
@prbrune prbrune Merge branch 'prbrune/snes-snesapplypcfix' into next 4da8008
@jedbrown jedbrown Web: update link to PyClaw website
Reported-by: David Ketcheson <david.ketcheson@kaust.edu.sa>
3c45446
@BarrySmith BarrySmith added faq item for snes matrix-free usage 722329f
@BarrySmith BarrySmith Merge branch 'barry/fix-setfunction-for-snesmf' into next
Conflicts:
	src/snes/interface/snespc.c
0064441
@BarrySmith BarrySmith typo in nrnorm, should have been found earlier
example with outdated %F and use of %d that should be %D
2da8f66
Shri Abhyankar update-SuiteSparse: change name of the test example targets for UMFPACK. a277f1d
Shri Abhyankar update-SuiteSparse: update UMFPACK and CHOLMOD version info in the in…
…terface source file.
d515b9b
Shri Abhyankar Merge branch 'shri/update-SuiteSparse' into shri/external-package-KLU
Conflicts:
	conf/test
	config/examples/arch-linux-pkgs-dbg-ftn-interfaces.py
	config/examples/arch-linux-pkgs-opt.py
6d808a5
@BarrySmith BarrySmith make sure example does not send uninitialized values through MPI to g…
…et caught by valgrind
4ff6614
Shri Abhyankar MATSOLVERKLU: Update KLU interface to use common SuiteSparse config. 7087e0d
Shri Abhyankar Merge branch 'shri/external-package-KLU' into next
Conflicts:
	config/BuildSystem/config/packages/__init__.py
5100f83
Commits on Feb 14, 2014
@jedbrown jedbrown make shared: fix library versioning, sonames, and symbolic linking
Uses $(eval ...) to eliminate some redundancy between packages.

Reported-by: Barry Smith <bsmith@mcs.anl.gov>
cbb9ddc
@jedbrown jedbrown Merge branch 'jed/gnumake' into next
* jed/gnumake:
  make shared: fix library versioning, sonames, and symbolic linking

Conflicts:
	gmakefile
e0887e2
@jedbrown jedbrown make shared: symlink without directory and install by recreating syml…
…inks
18e261d
@jedbrown jedbrown Merge branch 'jed/gnumake' into next
* jed/gnumake:
  make shared: symlink without directory and install by recreating symlinks
09da2c9
@BarrySmith BarrySmith check for ddot_() in chaco and reject it
print download packages before downloading
e685f91
@BarrySmith BarrySmith Merge branch 'barry/fix-chaco' into next f6430e7
@BarrySmith BarrySmith added how to find locations of tarballs to faq b6b28c8
@jedbrown jedbrown make shared: link directly to the actual library rather than via soname c21f8c7
@jedbrown jedbrown make: simply list objects as dependencies instead of via target-speci…
…fic var
9fad598
@jedbrown jedbrown make shared: pass path to SL_LINKER_FUNCTION
OS-X needs -install_name to contain the full path (similar to RPATH on
other unixy systems except it is set be the library rather than the
executable/dependent.)
30a89fe
@jedbrown jedbrown make shared: use leading 0 to indicate development version
These are distinct sonames on Linux.  Apple -compatibility_version is
unaffected because they convert to a 32-bit integer, so the leading 0
vanishes.
9769a5f
@jedbrown jedbrown make: hide library symlinking
Presumably the user is not interested in seeing this.  It should just
work.
b031592
@jedbrown jedbrown Merge branch 'jed/gnumake' into next
* jed/gnumake:
  make: hide library symlinking
  make shared: use leading 0 to indicate development version
  make shared: pass path to SL_LINKER_FUNCTION
  make: simply list objects as dependencies instead of via target-specific var
  make shared: link directly to the actual library rather than via soname
2fb305d
@prbrune prbrune DM: Fix for recursive DMDestroy() causing valgrind issues
runex26_2 in src/ts/examples/tutorials caused the following valgrind output:

> ==21447== Invalid read of size 8
> ==21447==    at 0x599093E: DMDestroy (dm.c:401)
> ==21447==    by 0x4F3F83F: PetscObjectDereference (inherit.c:612)
> ==21447==    by 0x4F30C6A: PetscObjectListDestroy (olist.c:160)
> ==21447==    by 0x4F3AA1F: PetscHeaderDestroy_Private (inherit.c:117)
> ==21447==    by 0x5184D16: VecDestroy (vector.c:499)
> ==21447==    by 0x5990EA5: DMDestroy (dm.c:425)
> ==21447==    by 0x5B2833D: PCDestroy (precon.c:122)
> ==21447==    by 0x5C0D6EF: KSPDestroy (itfunc.c:772)
> ==21447==    by 0x5C93387: SNESDestroy (snes.c:2790)
> ==21447==    by 0x5D35130: SNESReset_NASM (nasm.c:46)
> ==21447==    by 0x5C92705: SNESReset (snes.c:2732)
> ==21447==    by 0x5C92689: SNESReset (snes.c:2728)
> ==21447==    by 0x5DA00E8: TSReset (ts.c:1872)
> ==21447==    by 0x5DA09EB: TSDestroy (ts.c:1912)
> ==21447==    by 0x402671: main (ex26.c:172)
> ==21447==  Address 0x9b39778 is 24 bytes inside a block of size 32 free'd
> ==21447==    at 0x4C2A82E: free (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
> ==21447==    by 0x4F85AEB: PetscFreeAlign (mal.c:70)
> ==21447==    by 0x5990F29: DMDestroy (dm.c:426)
> ==21447==    by 0x5B2833D: PCDestroy (precon.c:122)
> ==21447==    by 0x5C0D6EF: KSPDestroy (itfunc.c:772)
> ==21447==    by 0x5C93387: SNESDestroy (snes.c:2790)
> ==21447==    by 0x5D35130: SNESReset_NASM (nasm.c:46)
> ==21447==    by 0x5C92705: SNESReset (snes.c:2732)
> ==21447==    by 0x5C92689: SNESReset (snes.c:2728)
> ==21447==    by 0x5DA00E8: TSReset (ts.c:1872)
> ==21447==    by 0x5DA09EB: TSDestroy (ts.c:1912)
> ==21447==    by 0x402671: main (ex26.c:172)

This was caused by recursive calls to DMDestroy().  When freeing the named local vector list, the linked list is freed
before being set to NULL.  The recursive call of DMDestroy() by VecDestroy() then tries to access the value in the
structure.  The fix to this is to zero out the DM member before the recursive call instead of after it.

Reported-by: Barry Smith <bsmith@mcs.anl.gov>
f490541
@BarrySmith BarrySmith Merge branch 'barry/add-fortran-num-consts' 1d449cb
@BarrySmith BarrySmith Merge branch 'barry/fix-chaco' 501c201
@BarrySmith BarrySmith Merge branch 'barry/fix-setfunction-for-snesmf' 987a76e
@jedbrown jedbrown MatPermute_MPIAIJ: fix memory error repurposing arrays
The maximum row length is not known in advance, yet rowlen could be
larger than number of rows m or the number of local columns.  This sums
in batches to avoid potential allocating per-row.

Reported-by: Barry Smith <bsmith@mcs.anl.gov>
970468b
@jedbrown jedbrown Merge branch 'master' into next
* master:
  added how to find locations of tarballs to faq
  make sure example does not send uninitialized values through MPI to get caught by valgrind
  typo in nrnorm, should have been found earlier example with outdated %F and use of %d that should be %D
  Web: update link to PyClaw website
5fcd75d
@jedbrown jedbrown Merge branch 'jed/fix-mat-permute-mpiaij' into next
* jed/fix-mat-permute-mpiaij:
  MatPermute_MPIAIJ: fix memory error repurposing arrays
e9db2cb
@prbrune prbrune Merge branch 'prbrune/dm-destroylinkedlistfix' into next 4c7bbc2
@BarrySmith BarrySmith Merge branch 'master' into next d0fb336
@BarrySmith BarrySmith Merged in jose/fix-examples (pull request #143)
Fix some C examples in Vec and KSP
ad9a81e
@balay balay suitesparse: requires gnumake 559a4be
@balay balay nightlybuilds: update suitesparse builds on bsd as they don't build w…
…ith bsd-make
1e3b5c2
@balay balay Merge branch 'shri/update-SuiteSparse' into next 993c0c8
Commits on Feb 15, 2014
@jedbrown jedbrown PetscLayoutSetBlockSize: allow PETSC_DECIDE for use when block size h…
…as already been set

This slightly simplifies logic in some places, where otherwise we would
have to MatGetBlockSize() and pass it back in.
69bbac9
@jedbrown jedbrown Mat & Vec: make block sizes lazy
In the past, MatSetUp eagerly set the block sizes (to 1) if not already
set.  This prevents use of the "convenience" constructors because they
call MatSetUp before returning.  We could add bs as an explicit argument
to those constructors, but block size is an extraneous concept for most
users.  The default value of the block size is -1, so now we use
PetscAbs(map->bs) anywhere it is not known in advance that bs has been
explicitly set.  This allows block sizes to be lazy, and they can be set
arbitrarily late, so long as the new size is compatible with the
existing distribution.  For (S)BAIJ matrices, the block size is
structural so it gets set explicitly and can thus be referenced directly
rather than via MatGetBlockSize or wrapped in PetscAbs.
33d5767
@jedbrown jedbrown Merge branch 'master' into next
* master:
  ksp/ex20.c: Add MatSetUp() and fix typo
  vec/ex18.c: Correctly fill the last entry of vector xend
  vec/ex1.c: Remove PetscFunctionBegin from main()
4ee26ec
@jedbrown jedbrown PetscFE: declare viewing functions
They don't do anything yet, but they are certainly intended to be public.
fd41f35
@BarrySmith BarrySmith added PETSC_UNUSED in a couple places
fixed variables declared in code, as opposed to at the beginning
removed non-portable use of %lld
cf3d2d9
@BarrySmith BarrySmith Merge branch 'master' into barry/ts-rhsjacobian-only 1f1b5a6
@BarrySmith BarrySmith Merge branch 'barry/ts-rhsjacobian-only' into next 1c441a7
@balay balay fortran: fix typo for MatSeqAIJRestoreArray()
ex79f.obj : error LNK2001: unresolved external symbol _MATSEQAIJRESTOREARRAY@16
6778691
@balay balay nightlybuilds: switch freebsd-cxx-cmplx-pkgs-dbg build to use g++46
/usr/home/balay/petsc.clone-3/arch-freebsd-cxx-cmplx-pkgs-dbg/bin/mpicxx -o ex10.o -c -Wall -Wwrite-strings -Wno-strict-aliasing -Wno-unknown-pragmas -g -O0  -fPIC  -I/usr/home/balay/petsc.clone-3/include -I/usr/home/balay/petsc.clone-3/arch-freebsd-cxx-cmplx-pkgs-dbg/include -I/usr/local/include    `pwd`/ex10.c
/home/balay/petsc.clone-3/src/ts/examples/tutorials/ex10.c: In function 'PetscErrorCode RDIJacobian_FE(_p_TS*, PetscReal, _p_Vec*, _p_Vec*, PetscReal, _p_Mat**, _p_Mat**, MatStructure*, void*)':
/home/balay/petsc.clone-3/src/ts/examples/tutorials/ex10.c:715: internal compiler error: Segmentation fault: 11
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://gcc.gnu.org/bugs.html> for instructions.
*** [ex10.o] Error code 1
30f8f98
@knepley knepley DMPlex: Fixed some small mistakes
- Initialize var
- Declare at the top of the block
- Added cone size call
62ad4a6
@knepley knepley Merge branch 'knepley/fix-plex-orient-parallel' into next
* knepley/fix-plex-orient-parallel:
  DMPlex: Fixed some small mistakes - Initialize var - Declare at the top of the block - Added cone size call

# 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.
f3a2942
@knepley knepley Merge branch 'knepley/fix-plex-refine-sf'
* knepley/fix-plex-refine-sf:
  SF: Added additional checking
  DMPlex: Fixed SF created during refinement - Remote edge had wrong starting point - Forgot an entry for hybrid edge
93b9b95
@balay balay nightlybuilds: freebsd g++ requires -lexecinfo 2c441fb
@balay balay nightlybuilds: variable declared as PETSC_UNUSED - but its actually used 8869cef
@BarrySmith BarrySmith test example output file was missing and test not listed in makefile b0c3903
@BarrySmith BarrySmith Merge branch 'master' of bitbucket.org:petsc/petsc e630791
@knepley knepley Merge branch 'knepley/feature-dmshell-subdm'
* knepley/feature-dmshell-subdm:
  DMShell: Added DMCreateSubDM() support
d4c0edb
Commits on Feb 16, 2014
@BarrySmith BarrySmith in ex6.c warn against using low level PetscBinaryRead/Write() and rec…
…ommend VecView/VecLoad() instead
199802f
@jedbrown jedbrown DT: remove spurious printing c7322fd
@jedbrown jedbrown Merge branch 'jed/dm-fe-view-declarations' into next
* jed/dm-fe-view-declarations:
  PetscFE: declare viewing functions
98b9057
@balay balay freebsd: Add back PETSC_UNUSED - but reformat to prevent gcc warning
/home/balay/petsc.test/src/ts/examples/tutorials/ex10.c: In function 'PetscErrorCode RDIJacobian_FE(TS, PetscReal, Vec, Vec, PetscReal, _p_Mat**, _p_Mat**, MatStructure*, void*)':
/home/balay/petsc.test/src/ts/examples/tutorials/ex10.c:711:36: warning: ignoring attributes applied to 'std::complex<double>' after definition [-Wattributes]
52b409b
@jedbrown jedbrown Merge branch 'master' after racy merge
* master:
  freebsd: Add back PETSC_UNUSED - but reformat to prevent gcc warning
  in ex6.c warn against using low level PetscBinaryRead/Write() and recommend VecView/VecLoad() instead
  test example output file was missing and test not listed in makefile
  DMShell: Added support for DMCreateSubDM() Reported-by: Luc Berger-Vergiat <lb2653@columbia.edu>
8c7fd50
@jedbrown jedbrown Merge branch 'maint'
* maint:
  DT: remove spurious printing

Conflicts:
	src/dm/dt/interface/dt.c
9b01c52
@jedbrown jedbrown Merge branch 'master' into next
* master:
  freebsd: Add back PETSC_UNUSED - but reformat to prevent gcc warning
  DT: remove spurious printing
  in ex6.c warn against using low level PetscBinaryRead/Write() and recommend VecView/VecLoad() instead
  test example output file was missing and test not listed in makefile
  nightlybuilds: variable declared as PETSC_UNUSED - but its actually used
  nightlybuilds: freebsd g++ requires -lexecinfo
  nightlybuilds: switch freebsd-cxx-cmplx-pkgs-dbg build to use g++46
  fortran: fix typo for MatSeqAIJRestoreArray() ex79f.obj : error LNK2001: unresolved external symbol _MATSEQAIJRESTOREARRAY@16
1ab73cf
Commits on Feb 17, 2014
@BarrySmith BarrySmith updates to Tao per request of Lisandro for Petsc4py
* I would like to have TaoGetConstraintTolerances(), right now I'm
defining it in petsc4py as shown in the link below:
https://bitbucket.org/petsc/petsc4py/commits/ab01ebe336491190a2db85c179c48b1c64fdc673?at=master#chg-src/include/custom.h

* Look at all these definitions, they should be exposed in the C API
to be able to call e.g. TaoSetType(tao,TAOLMVM);
https://bitbucket.org/petsc/petsc4py/commits/ab01ebe336491190a2db85c179c48b1c64fdc673?at=master#Lsrc/PETSc/petsctao.pxiT4

Also removed -tao_method, removed tao_ from names of tao methods and fixed some tao source code formatting
58417fe
@BarrySmith BarrySmith Merge branch 'barry/tao-fixes-for-petsc4py' into next 157e62f
@jedbrown jedbrown Merge branch 'jed/lazy-blocksize' into next
* jed/lazy-blocksize:
  Mat & Vec: make block sizes lazy
  PetscLayoutSetBlockSize: allow PETSC_DECIDE for use when block size has already been set
5bc09d4
@BarrySmith BarrySmith changed TaoTerminationReason TaoConvergedReason
changed TaoLineSearchTerminationReason to TaoLineSearchConvergedReason

reason: consistency with the rest of PETSc
e4cb33b
@BarrySmith BarrySmith Merge branch 'barry/tao-fixes-for-petsc4py' into next c74a923
@jedbrown jedbrown Merge branch 'jed/elemental-int64'
* jed/elemental-int64:
  Elemental: disable Qt5 so we don't have to pass -I for those include paths
  nightlybuilds: disable elemental build when conflicting with hdf5
  elemental: update nightly builds. [add with-cxx-dialect=C++11 to linux - disable elsewere]
  Elemental: use self.framework.getCompilerFlags()
  Elemental: add support for 64-bit indices
  Elemental: upgrade to development version that requires C++11
  config: add with-cxx-dialect and check for C++11
  Elemental: fix setting of CMAKE_CXX_FLAGS
  config: write CXX_FLAGS for building C++ even when clanguage=C
38bdd0f
@BarrySmith BarrySmith fix some VecGetArray() to VecGetArrayRead() and const in Tao source code 5e08136
@BarrySmith BarrySmith fix Tao unconventional naming of TAOLINESEARCH_XXX_CTX TAOLINESEARCH_…
…MT etc
8caf6e8
@BarrySmith BarrySmith Merge branch 'barry/tao-fixes-for-petsc4py' into next c83dbde
@BarrySmith BarrySmith update some /*MC manual pages, correcting errors and occassionally ad…
…ding more information
1957e95
@BarrySmith BarrySmith Merge branch 'barry/update-some-manualpages' into next 88573df
@balay balay doc: fix errors from builddist.log 86da477
@balay balay cuda: fix mismatched include for cublasInit()
/home/balay/petsc.clone-3/src/sys/objects/ftn-custom/zstart.c: In function ‘petscinitialize_’:
/home/balay/petsc.clone-3/src/sys/objects/ftn-custom/zstart.c:446:3: warning: implicit declaration of function ‘cublasInit’ [-Wimplicit-function-declaration]
76c028a
@balay balay win32: fix fortran upcase for petscsetcommonblocknumeric_()
libpetsc.lib(zstartf.o) : error LNK2001: unresolved external symbol _petscsetcommonblocknumeric_@32
d5410ed
@balay balay Merge branch 'barry/add-fortran-num-consts' 5aedad7
Commits on Feb 18, 2014
@jedbrown jedbrown Merge branch 'jed/fix-mat-permute-mpiaij'
* jed/fix-mat-permute-mpiaij:
  MatPermute_MPIAIJ: fix memory error repurposing arrays
6f717fb
@jedbrown jedbrown Merge branch 'jed/malloc-zero'
* jed/malloc-zero:
  PetscFreeAlign: avoid unshifting NULL
  PetscMalloc: allow ptr=malloc(0) and free(ptr)
78ccc32
@jedbrown jedbrown PCBDDC: fix headers after batch change from <> to "" ab5c6b0
@jedbrown jedbrown Merge commit 'PCBDDC: fix headers after batch change from <> to ""' bbbf719
@jedbrown jedbrown DMPlex: quash -Wmaybe-uninitialized 2a5d012
@jedbrown jedbrown Merge branch 'master' into next
* master:
  PCBDDC: fix headers after batch change from <> to ""
  win32: fix fortran upcase for petscsetcommonblocknumeric_()
  cuda: fix mismatched include for cublasInit()
  doc: fix errors from builddist.log
d4f5a57
@jedbrown jedbrown Merge branch 'jed/plex-Wmaybe-uninitialized' into next
* jed/plex-Wmaybe-uninitialized:
  DMPlex: quash -Wmaybe-uninitialized
44c1122
@jedbrown jedbrown DMPlex: quash more -Wmaybe-uninitialized 38ea73c
@jedbrown jedbrown Merge branch 'jed/plex-Wmaybe-uninitialized' into next
* jed/plex-Wmaybe-uninitialized:
  DMPlex: quash more -Wmaybe-uninitialized
b77854e
@dalcinl dalcinl Fix missing PETSC_EXTERN in TaoMonitor() 6e6473c
@jedbrown jedbrown Mat: close memory leaks in MatGetOrdering_Spectral 8c6db3d
@jedbrown jedbrown DMClearGlobalVectors: fix reference-counting loop
Reported-by: Patrick Farrell <patrick.farrell@maths.ox.ac.uk>
62d839e
@jedbrown jedbrown Merge branch 'master' into next
* master:
  DMClearGlobalVectors: fix reference-counting loop
  Mat: close memory leaks in MatGetOrdering_Spectral
919f8e0
Jason Sarich tao: namespace TAO subvec and submat routines and make intern b98f30f
@BarrySmith BarrySmith fixed argument to TaoVecGetSubVec() 3a831ad
@BarrySmith BarrySmith Merge branch 'barry/tao-fixes-for-petsc4py' into next d62e7e9