| #
9a07f4df
|
| 20-Nov-2012 |
Jed Brown <jed@59A2.org> |
Some real math functions
Hg-commit: 0c7e3dc215326861e3fdde67be0229a7248b021b
|
| #
7a3afff6
|
| 15-Oct-2012 |
Barry Smith <bsmith@mcs.anl.gov> |
commit after merge
Hg-commit: 89503a96f70dd4d497719d2597342d8944de9c22
|
| #
9f20b660
|
| 15-Oct-2012 |
Satish Balay <balay@mcs.anl.gov> |
petscmath.h: <complex> should not get exported via public includes for c++/non-complex build.
So use PETSC_DESIRE_COMPLEX for c++ aswell [similar to current usage for C].
Its one more fix for d00825f
petscmath.h: <complex> should not get exported via public includes for c++/non-complex build.
So use PETSC_DESIRE_COMPLEX for c++ aswell [similar to current usage for C].
Its one more fix for d00825fc28cd. This broke c++ build on OSX [with isinf() in mathinf.c]
Hg-commit: c3957891ee3ad3e67521ccfddb5d1f2234762909
show more ...
|
| #
3351b394
|
| 15-Oct-2012 |
Satish Balay <balay@mcs.anl.gov> |
petscmath.h: fix Precision: __float128 - bug introduced in d00825fc28cd
Hg-commit: 94f31d2b6fc27496da3a48a1674778ff4462c95b
|
| #
50f81f78
|
| 14-Oct-2012 |
Jed Brown <jed@59A2.org> |
Refactor complex support to allow portable use of complex in otherwise real builds
Defines PETSC_HAVE_COMPLEX when _any_ complex type is available
Hg-commit: d00825fc28cd13b46bc0f1016da5b9d91a442bc4
|
| #
4dd37ed8
|
| 29-Aug-2012 |
Barry Smith <bsmith@mcs.anl.gov> |
commit after merge
Hg-commit: f6a88fee9ba1b8041ee9b459d0ad54470439160a
|
| #
8c764dc5
|
| 23-Aug-2012 |
Jose Roman <jroman@dsic.upv.es> |
Enable __float128 precision with complex scalars
Hg-commit: 260360ea8a4fc88aed535bc381fba60e70cd1504
|
| #
48fab831
|
| 09-Jul-2012 |
Brad Aagaard <baagaard@usgs.gov> |
Merge
Hg-commit: 8f74ca6da3dda2e92c9063777436a7e360bcc637
|
| #
0f55b88d
|
| 03-Jul-2012 |
Satish Balay <balay@mcs.anl.gov> |
backout all chmod stuff in bf971ba58125
[but keep diffs to src/mat/impls/aij/mpi/mpimatmatmult.c ]
Hg-commit: 4a812fc1f35496ce2717c7da9ef12ffabbd13258
|
| #
c0174eb7
|
| 03-Jul-2012 |
Mark F. Adams <mark.adams@columbia.edu> |
removed a spurious call to MatSetBlockSize.
Hg-commit: bf971ba58125bd8cf7ceb2a9d0f5cdf27c27cc6a
|
| #
519e2a1f
|
| 27-Jun-2012 |
Paul Mullowney <paulm@txcorp.com> |
Slight fix.
Hg-commit: 66ca8db0d5f8c65f46466a1889320bb308d2c39b
|
| #
debe9ee2
|
| 26-Jun-2012 |
Paul Mullowney <paulm@txcorp.com> |
Some local changes.
Hg-commit: 554d9907a87259cc15d90b3fa0540f3e10379a06
|
| #
f1301baf
|
| 25-Jun-2012 |
Matthew G Knepley <knepley@gmail.com> |
Merge
Hg-commit: 5fd7344ebb01d47fc51348189c58908ebeb94702
|
| #
10e9db80
|
| 30-May-2012 |
Paul Mullowney <paulm@txcorp.com> |
Pulled in changes from petsc-dev. make tests passing. Everything looks good.
Hg-commit: 94eb5d50210da3db6d654576e30f2a31ec2adf25
|
| #
574fde7b
|
| 25-May-2012 |
Satish Balay <balay@mcs.anl.gov> |
looks like quadmath.h requires EXTERN_C_BEGIN/END [for c++ build to work].
However these macros are not available in petscmath.h - so expand them manually.
Hg-commit: 4776db33cd9c4538948c076ffbff71f
looks like quadmath.h requires EXTERN_C_BEGIN/END [for c++ build to work].
However these macros are not available in petscmath.h - so expand them manually.
Hg-commit: 4776db33cd9c4538948c076ffbff71f0508d9a4d
show more ...
|
| #
014dd563
|
| 16-May-2012 |
Jed Brown <jed@59A2.org> |
Add --with-visibility, all public symbols are marked with PETSC_EXTERN
Hg-commit: 91d6ccc58aec44b109582d7decb1c094fd081024
|
| #
9ae82921
|
| 11-May-2012 |
Paul Mullowney <paulm@txcorp.com> |
Lots of changes. Also addign new files for cusparse matrix type.
Hg-commit: d2f118b395b24fdc836f857e3f221f6c0c9eca58
|
| #
a5057860
|
| 30-Apr-2012 |
Barry Smith <bsmith@mcs.anl.gov> |
cleanup include files slightly
removed warning generating code
Hg-commit: e1c9e44476917d298f5bb1d12e7f3b51de0af5ca
|
| #
faa6e9b0
|
| 13-Mar-2012 |
Matthew G Knepley <knepley@gmail.com> |
Increased the precision of PETSC_PI
Hg-commit: f2a228c1861f1900a8aaaf2f6582bb98ce131282
|
| #
51322221
|
| 28-Feb-2012 |
Jungho Lee <julee@mcs.anl.gov> |
commit after merge
Hg-commit: e8d19a834d936a4816c3abf3af0b68733c00be11
|
| #
1996d6cf
|
| 02-Oct-2011 |
Peter Brune <brune@mcs.anl.gov> |
commit after merge
Hg-commit: e8b7dd1aa8e3cc4c5c8c0c0901a58568ca5faaab
|
| #
d9a4bb16
|
| 01-Oct-2011 |
Jed Brown <jed@59A2.org> |
Add PetscClipInterval() because I'm tired of getting Max and Min mixed up when I write it myself
Hg-commit: d7cbb52ce93e90711bcb7d7b248ca3e225e1a286
|
| #
8c9b41c2
|
| 23-Sep-2011 |
Jed Brown <jed@59A2.org> |
merge Rosenbrock work
Hg-commit: c912d4ef2e7db4f1f10a2f0924022e28387de729
|
| #
ab824b78
|
| 20-Sep-2011 |
Barry Smith <bsmith@mcs.anl.gov> |
removed definition of specific machine constants in tfs.h
put more "correct" machine constants into petscmath.h
Hg-commit: 5473102dc88e9b9b88dbc2eead8916daf2611023
|
| #
82a7e548
|
| 20-Sep-2011 |
Barry Smith <bsmith@mcs.anl.gov> |
fixes to allow runs without ctable
changed default values for machine epsilon and its sqrt
Hg-commit: c636d6b71ea127a1acb84fe0383b490d3b14f6a2
|