| 26283091 | 29-Oct-2009 |
Barry Smith <bsmith@mcs.anl.gov> |
changed PetscMap to PetscLayout
Hg-commit: b8f4632bd00d1563dcf8108fd4199169f8fd320b |
| 7a2fc3fe | 29-Oct-2009 |
Barry Smith <bsmith@mcs.anl.gov> |
changed PetscMap to be pointer to _p_PetscMap
changed PetscMapInitialize() to PetscMapCreate() and also have it malloc the space
Hg-commit: 2d1edc265bc1b52b0bfe730b4711ac37a11744fd |
| dc9360f3 | 28-Oct-2009 |
Barry Smith <bsmith@mcs.anl.gov> |
add support for PCREDISTRIBUTE for setupcalled already called
Hg-commit: 2a5f400ff6d98c17f76077e2cacdb59ed90e2771 |
| ca320bd4 | 27-Oct-2009 |
Barry Smith <bsmith@mcs.anl.gov> |
basic PCRedistribute works
Hg-commit: e251c191299557d57332808ebd0734370cceb141 |
| 8e395302 | 05-Nov-2009 |
Jed Brown <jed@59A2.org> |
Fix unclosed structured comments
Hg-commit: 508975e438d67dc06d3677c6ba78999b11ca8636 |
| 427efc14 | 23-Oct-2009 |
Matthew Knepley <knepley@mcs.anl.gov> |
Fixed ASA problems
Hg-commit: f949e59dac4578d5ead383433a9df881be31a886 |
| 46bc77b6 | 08-Oct-2009 |
Barry Smith <bsmith@mcs.anl.gov> |
cleanup of many makefiles removed outdated and old style crud
Hg-commit: eaa8d3fd8168d422b0b437697ce75e0f89c760b6 |
| e2dee56b | 07-Oct-2009 |
Barry Smith <bsmith@mcs.anl.gov> |
commit after merge of patch to 3.0.0 in tfs
Hg-commit: 3782a3cc1a1f038eb250ec0984baa32350fa3da4 |
| 9182e22c | 07-Oct-2009 |
Barry Smith <bsmith@mcs.anl.gov> |
From: s.kramer@imperial.ac.uk
Subject: bug in tfs preconditioner
think I found a nasty bug in the tfs preconditioner/solver. The first symptons were MPI complaints about corrupted messages in MPI_W
From: s.kramer@imperial.ac.uk
Subject: bug in tfs preconditioner
think I found a nasty bug in the tfs preconditioner/solver. The first symptons were MPI complaints about corrupted messages in MPI_WAIT on line 1552 of ksp/pc/impls/tfs/gs.c (the file is the same in petsc-dev and 3.0.0). Valgrind suggested a buffer overrun (in reading) in MPI_Isend of line 1533:
ierr = MPI_Isend(dptr3, *msg_size++, MPIU_SCALAR, *list++, MSGTAG1+my_id, gs->gs_comm, msg_ids_out++);CHKERRQ(ierr);
Stepping through with a debugger however it looked like everything going into the MPI_Isends and MPI_Irecvs was perfectly fine. Until I realised that they were both replaced by a macro from petsclog.h:
#define MPI_Isend(buf,count,datatype,dest,tag,comm,request) \
((isend_ct++,0) || TypeSize(&isend_len,count,datatype) || MPI_Isend(buf,count,datatype,dest,tag,comm,request))
Because count is used twice in that expression, the argument *msg_size++ is evaluated twice, and only gives the right integer value in calling TypeSize, and will be wrong in the actual MPI_Isend call. The same thing is going on on line 1327 of gs.c btw. If someone has time to look into this, it would be much appreciated. Has the tfs solver been used/applied much, as far as people know?
Hg-commit: 434a17925af32ecc820090666284b512ae1455e8
show more ...
|
| 911f9fe8 | 07-Oct-2009 |
Barry Smith <bsmith@mcs.anl.gov> |
minor additions to PCSetUp_Redistribute()
Hg-commit: bf5980cc405a5a920d4c400fc91e2bddaa037481 |
| 5604da14 | 29-Sep-2009 |
Matthew Knepley <knepley@mcs.anl.gov> |
Merge
Hg-commit: b70f4367eb5351c6e8d5d2a868decc5b50bd9108 |
| 864b637d | 29-Sep-2009 |
Matthew Knepley <knepley@mcs.anl.gov> |
Corrected behavior of ML for different matrix types
Hg-commit: 7c4f52b3fb2dd3b0c0c86e2b4616420332ae07a2 |
| a1db5e05 | 28-Sep-2009 |
Barry Smith <bsmith@mcs.anl.gov> |
changed bit_mask.c to bitmask.c source file names in PETSc are not suppose to have _ in them.
Hg-commit: bdc3fbaac6a05b53d5845da1b2abe1f9c39e12f2 |
| 098c0d8d | 24-Sep-2009 |
Matthew Knepley <knepley@mcs.anl.gov> |
Merge
Hg-commit: a7710b011ce3f2c4ac169ab040d429403a541f7f |
| de8939e3 | 24-Sep-2009 |
Barry Smith <bsmith@mcs.anl.gov> |
needed (MatSORType) caste
Hg-commit: dd8198117db07a544e45b476c75f7b971d904edc |
| 7319c654 | 24-Sep-2009 |
Barry Smith <bsmith@mcs.anl.gov> |
PCApplyRichardson() now takes a flag indicating if initial guess is zero, this saves work on application of SOR preconditioner in first iteration for example
KSPSolve() checks if the norm of x is kno
PCApplyRichardson() now takes a flag indicating if initial guess is zero, this saves work on application of SOR preconditioner in first iteration for example
KSPSolve() checks if the norm of x is known to be zero and if so temporarily sets the guess_zero flag in KSP (it is only temporarily cause in the next solve it may not be zero.
Hg-commit: 7c18a45dad7e4c7bb780a7cf7ffbc18bb03598c8
show more ...
|
| 361c1e09 | 23-Sep-2009 |
Matthew Knepley <knepley@mcs.anl.gov> |
Fixed tag in PC
Hg-commit: 227204f7c76b85d4d0f3a8fb994e01ef05ec3b4b |
| e8dd6687 | 17-Sep-2009 |
Hong Zhang <hzhang@mcs.anl.gov> |
rm unused variables
Hg-commit: 9237ba2c5a8040a722bd5cbbf72e2bcb6f147a1a |
| e032938d | 16-Sep-2009 |
Barry Smith <bsmith@mcs.anl.gov> |
removed dead pcdmmg; added comments for gcov rules in rules makefile
Hg-commit: 2505ec1c893ed70485ad3887a2246a1aea7a7e39 |
| df826632 | 08-Sep-2009 |
Barry Smith <bsmith@mcs.anl.gov> |
minor improvements/code modifications to MatRelax_SeqSBAIJ()
Hg-commit: 9de31ba8922331ba61bee731740c3ae28ff76b36 |
| 1c6d4a29 | 01-Sep-2009 |
Barry Smith <bsmith@mcs.anl.gov> |
added sor.tex discussion of SOR and Eisentat to the repository.
Note it is not yet linked to by anything else in the repository
Hg-commit: a0734403d7dae2dc059c3f6902082a79505fc729 |
| 5c99c7da | 26-Aug-2009 |
Barry Smith <bsmith@mcs.anl.gov> |
added "" around compiler string passed to hypre.py CC
That worked, but now I am running into errors when comfiguring hypre. I believe that it is a matter of not having quotes around the TAU C compil
added "" around compiler string passed to hypre.py CC
That worked, but now I am running into errors when comfiguring hypre. I believe that it is a matter of not having quotes around the TAU C compiler specification in the hypre configure arguments, i.e.
CC=/autofs/na2_proj/geo002/petsc-dev/bin/taucc.py -cc=cc -pdt_parse=/ccs/proj/geo002/xt5/pdtoolkit-3.14.1/x86_64/bin/cxxparse -tau_lib_dir=/ccs/proj/geo002/xt5/tau-2.18.2p4/x86_64/lib
I have quotes around this argument in the petsc configure arguments. Do I need to add brackets like you do with comma delimited lists? Other options?
Hg-commit: 32644898bea493c42c01be0928b3a26c980ff151
show more ...
|
| eaf59616 | 11-Aug-2009 |
Matthew Knepley <knepley@mcs.anl.gov> |
Fix for ASM
- There were some scatters that needed to be created on all procs
Hg-commit: e61a35b66f55ab39a731ae7214d4b3cf84b5fe10 |
| 3b46a515 | 27-Jul-2009 |
Glenn Hammond <unknown> |
Added command line arguments for maximum # levels for hypre PFMG solver
Hg-commit: d703bf0cde9be5121b2ca2f68d36363e0559466f |
| 6ed231c7 | 21-Jul-2009 |
Matthew Knepley <knepley@mcs.anl.gov> |
Allow unsorted subdomain indices in ASM
Hg-commit: fc2dccd6e2317e488ce5d2d36b578a430de1302a |