| 8cbc2665 | 17-Sep-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
someone removed the /usr from the getSearchDirectories() for pthread.py; it stopped working for me because it could not find libpthread.a so I put it back, please email me if you have concerns. BTW:
someone removed the /usr from the getSearchDirectories() for pthread.py; it stopped working for me because it could not find libpthread.a so I put it back, please email me if you have concerns. BTW: it doesn't appear to look in /usr if getSearchDirectories() returns [] so I don't understand how it could work without the /usr?
Hg-commit: 02c800600f9e243c12ad73e0678f45f9a2f01af1
show more ...
|
| d9f04c2d | 17-Sep-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
commit after merge
Hg-commit: 40959d786871b3f854447e1ae958a7d8951865b1 |
| cc9833d2 | 17-Sep-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
fixed pthread to support single precision 64 bit int windows etc
Hg-commit: 72a65071199249112327e74df3d050626ea6fb7c |
| 17224f6f | 17-Sep-2010 |
Matthew G Knepley <knepley@gmail.com> |
Spelling error
Hg-commit: 5bea1b5b37a06f66533f131e9d02a455e79a3770 |
| 14dc4a0f | 17-Sep-2010 |
Satish Balay <balay@mcs.anl.gov> |
do not use '/usr' as default search path - as it can be /usr/lib or /usr/lib64
Hg-commit: b1fd812e6ffee115d4cecf2a7b905d6d04d6e97c |
| e29b2329 | 16-Sep-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
commit after merge
Hg-commit: c1237763191796e794819bb5e56688ca83406d39 |
| ffc871a5 | 16-Sep-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
basic thread based webserver is working.
Hg-commit: 1f1782a7f75a7afa334923daa4504805899ecfce |
| 63d6bff0 | 16-Sep-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
added --with-server for configure
added PETSC_USE_PTHREAD that should turn off any non-thread safe code, like building the stack
Hg-commit: 48bd7885c44ef999002edd180c6945605b41bebb |
| 36a48117 | 16-Sep-2010 |
Jed Brown <jed@59A2.org> |
Update petsc4py and mpi4py versions
Hg-commit: 2624bead07b8276e91c42a58d98f4c766b05eb7e |
| 277aa27a | 15-Sep-2010 |
Victor Minden <victorminden@gmail.com> |
added check that void* is the same size in CUDA as in C
Hg-commit: d659ce87dbf3af92efe6aec6046b379f5ee20043 |
| 545236c8 | 15-Sep-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
fixes to update to the latest petsc4py and mpi4py
Hg-commit: 502ee218465a19d2f1d514ee636c6275f2abe648 |
| f563f603 | 09-Sep-2010 |
Satish Balay <balay@mcs.anl.gov> |
set 'CCASPEC_VARS' in configure instead of makefile
[otherwise 'make clean' gives error '--var: Command not found']
Hg-commit: 332275c09cb2ad4d7444798a67867f9dd5e5927e |
| 439d5d18 | 08-Sep-2010 |
Satish Balay <balay@mcs.anl.gov> |
update to use ParMetis-3.1.1
Hg-commit: bdaffee2e98fe4b3f8476ff5e319566949cc98d0 |
| 6708bc5d | 01-Sep-2010 |
Satish Balay <balay@mcs.anl.gov> |
go back to the correct prometheus tarball.
Backed out changeset 70f516a96002
Hg-commit: 4a947be7a3db1c2da2be8deff9306be239082d0a |
| f2648109 | 01-Sep-2010 |
Satish Balay <balay@mcs.anl.gov> |
use temp prometheus tarball [with PetscStackPop fixes]
Hg-commit: 70f516a960025b619e4daf2c3d854f180b1740ed |
| 89667278 | 01-Sep-2010 |
Matthew G Knepley <knepley@gmail.com> |
Started support for deprecation of options
Hg-commit: 6f55f3d89ae91dcc06938a61ec2d345c15b31d5c |
| e68ebbec | 30-Aug-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
added PETSc setup.py for building PETSc with easy_install
Hg-commit: e676044ffb5c237267fff3cfbf3fdea328c33b16 |
| a20a2d1d | 29-Aug-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
stuff in __init__.py was garbage so removed it to prevent confusion
Hg-commit: 81580c2f0bb84134e89ff7612cb50d3ca49904d8 |
| 494c0aa3 | 29-Aug-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
petsc4py.py and mpi4py.py set ARCHFLAGS properly on Apple before setting up the two packages
Hg-commit: c851ab8ee5f62e731d8d634379ad7a5d7ca54730 |
| c612facf | 28-Aug-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
add ARCHFLAGS so that Mac doesn't try to build petsc4py for ppc
Hg-commit: 1c65342adba8ee2b6553a041fb76dad2f4ead3fe |
| a0036e3b | 28-Aug-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
damn
Hg-commit: e3a7c92ab17287bab1a86b3f92d92972bb76aea5 |
| 61fc94c7 | 28-Aug-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
commit after merge
Hg-commit: 4fa83538c42e070b73ebf5c2d2226bf76d74b79d |
| 53c77d0a | 27-Aug-2010 |
Jed Brown <jed@59A2.org> |
Consistency checking for __FUNCT__ when __func__ or __FUNCTION__ are available.
Uses the compiler definitions in CHKERRQ/SETERRQ when available to get
better traces through user code that doesn't ma
Consistency checking for __FUNCT__ when __func__ or __FUNCTION__ are available.
Uses the compiler definitions in CHKERRQ/SETERRQ when available to get
better traces through user code that doesn't manually defined __FUNCT__.
Fix lots of places that __FUNCT__ was used inconsistently.
Hg-commit: fdbdc93647ff934a12906fd51bc39302de07ae77
show more ...
|
| 10699583 | 27-Aug-2010 |
Jed Brown <jed@59A2.org> |
Go back to using _mm_prefetch by default over __builtin_prefetch.
The Clang headers are buggy [1] and _mm_prefetch has more precise semantics.
[1] http://llvm.org/bugs/show_bug.cgi?id=8011
Hg-comm
Go back to using _mm_prefetch by default over __builtin_prefetch.
The Clang headers are buggy [1] and _mm_prefetch has more precise semantics.
[1] http://llvm.org/bugs/show_bug.cgi?id=8011
Hg-commit: 7c0cae9c100ef15ddd8944a88001f778ee2c471a
show more ...
|
| 63cf1ef0 | 27-Aug-2010 |
Jed Brown <jed@59A2.org> |
Correct prefetch hint definitions to agree with Intel manual and GCC docs
Hg-commit: 51fe6ba42cf669ceabf76163530668e9dbe3d2d7 |