| 334eac09 | 19-May-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
removed use of PetscGlobalXXX() from source where it should not have been used
removed use of MPI_IN_PLACE where it should not have been used
added text to developers.tex about not using > 1.1 MPI wi
removed use of PetscGlobalXXX() from source where it should not have been used
removed use of MPI_IN_PLACE where it should not have been used
added text to developers.tex about not using > 1.1 MPI without checks and not using PetscGlobalXXX()
Hg-commit: caac1bef6b4be19d1c0d5e507683ea2000c7a62a
show more ...
|
| c0e490dd | 18-May-2010 |
Shri Abhyankar <abhyshr@mcs.anl.gov> |
Small fix
Hg-commit: 969a0439c57c59d0faa74159cc595ede51671bbb |
| b1550197 | 18-May-2010 |
Shri Abhyankar <abhyshr@mcs.anl.gov> |
Started working on the new VecLoad interface
Fixed missing value in MatLUFactorNumeric_SeqAIJ_Inode
Hg-commit: 3244cd11c6aa30c0fda11e54a655412157e3a23f |
| 60154eb2 | 17-May-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
some fixes for dynamic libraries load including adding characteristic, fixes for multiple libraries
removed Usage Statistics from webpages since had no useful information
Hg-commit: 60bc2b7630581287
some fixes for dynamic libraries load including adding characteristic, fixes for multiple libraries
removed Usage Statistics from webpages since had no useful information
Hg-commit: 60bc2b76305812878b4dd93b664f25326281ac25
show more ...
|
| d736bfeb | 11-May-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
fixed enum argument to PetscError() in many places that arise only in C++
changed CHKERRXX() to be more rational, see petsc-dev email
Hg-commit: bf64cb01b8b40e01034bd22ad18ed6a259752ed7 |
| c1235816 | 10-May-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
removed unneeded { around SETERRQX() calls}
Hg-commit: d55148dbbe8a10f500a7e181b91ba166aec6f665 |
| 17186662 | 10-May-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
finished formating of SETERRQ() calls and changed a few PETSC_COMM_SELF to appropriate comm
Hg-commit: 099173676d5157f4b09175cd1f5104bf5e549784 |
| e32f2f54 | 07-May-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
added MPI_Comm as first argument to PetscError() and SETERRQX()
Hg-commit: fe5a2ff050abc00ca2c2979c0af22b117b62874e |
| f23f966c | 06-May-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
commit after merge of bug-fix to petsc-3.1 adding calls to MPI_Waitall() for ready receiver posted messages after cancel.
Hg-commit: c1b45858b86e4eb019634112f9645ee58d756dfa |
| 293bbe37 | 06-May-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
In ready receiver mode for VecScatter the recvieve requests need to be MPI_Waited on after being canceled. Most systems worked without this wait but IBM POwersystems seem to require it. Otherwise the
In ready receiver mode for VecScatter the recvieve requests need to be MPI_Waited on after being canceled. Most systems worked without this wait but IBM POwersystems seem to require it. Otherwise they generate Can't free an active receive request
From: Mathias PUETZ <mpuetz@de.ibm.com>
Subject: Re: [petsc-maint #45786] problems in PETSC VecScatter code which cause problems in IBM PE with suggested fixes
Date: May 6, 2010 4:18:38 AM CDT
Hg-commit: 09b519fe2c89bbd57182a06d8a708469fa25b00b
show more ...
|
| 503cfb0c | 22-Apr-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
general cleanup of petscsys.h
removed PetscStrfree() since PetscFree() does the same thing
Hg-commit: 383ece197d31b92016b12ff3a62a2e744f8d981d |
| e2e64c6b | 22-Apr-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
systematically changed use of phrase of ./config/configure.py to ./configure to have one consistent (and standardish) way of refering to it. Having two ways of refering to the same thing is confusing
systematically changed use of phrase of ./config/configure.py to ./configure to have one consistent (and standardish) way of refering to it. Having two ways of refering to the same thing is confusing and unneccesary
Hg-commit: fcd327f3f4fcb82c63cd1afe8da55ad84e8250f3
show more ...
|
| f4ea02b3 | 12-Apr-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
basic DM in PCMG works,
Hg-commit: b2fcfd86c7b1fa80b7b0bbaaac001e0340916b7a |
| 0e1b073a | 10-Apr-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
changed VecSqrt() to VecSqrtAbs()
Hg-commit: 271746aefb509c82730a03fa4fbf53a1092237ba |
| fcd96beb | 10-Apr-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
commit after pull of manual pages patches for 3.1
Hg-commit: fdda79b9b0cfa271aff815250a8f6761d22b5ee8 |
| eca87e8d | 09-Apr-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
many more manual page fixes, mostly for badly formatted manual pages for macros
Hg-commit: 97f5f25f18c2f9879fa942b30e59e9d3b8604142 |
| 1890ba74 | 09-Apr-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
minor fixes/updates to manual pages
Hg-commit: 98609ad2d9f7c6c11108358322be1092152924d7 |
| 834dbeb0 | 09-Apr-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
removed PetscObjectExists() very dangerous function, little used
minor manual page fixes
Hg-commit: cb01c277f10bfdc4f569c4aaf2838cbe32ff5320 |
| 6aa30d07 | 09-Apr-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
PetscLayoutInitialize() was still listed in many manual pages when it no longer exists and manaul pages said you work with pointer to object instead of object which is now wrong (use to be that way).
PetscLayoutInitialize() was still listed in many manual pages when it no longer exists and manaul pages said you work with pointer to object instead of object which is now wrong (use to be that way).
Hg-commit: 7017bfc4c6f127702183568c7bcd88cf6b97c833
show more ...
|
| 01cb0274 | 09-Apr-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
PetscInitializeFortran() manual page was not generated. Docs improved and test example improved.
Hg-commit: 576bbe2394af0a89ae4d5dcc0b113fb3e9a189f5 |
| 8710a43c | 08-Apr-2010 |
Satish Balay <balay@mcs.anl.gov> |
merge from 31-p1
Hg-commit: 862efa09182c710aba624593b4f9b5c375d445be |
| f6680f47 | 08-Apr-2010 |
Satish Balay <balay@mcs.anl.gov> |
doc fixes [still p1]
Hg-commit: a72848c873413904e3af55111ad18683dbdc9296 |
| 0700a824 | 08-Apr-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
changed cookie to classid cause that is what it is.
Hg-commit: ffa25d1edea7afc409fcc13e7f2b63587b1494c1 |
| 18c4d256 | 02-Apr-2010 |
Jed Brown <jed@59A2.org> |
merge release
Hg-commit: 1128870f63f1ed1a640bab8e365c4ed33a81e66e |
| 5514b3c3 | 02-Apr-2010 |
Jed Brown <jed@59A2.org> |
Free winstarts (closes memory leak with -vecscatter_window)
Hg-commit: 692f2379c82dc8307129cc08d98f26aca26ed92b |