| #
a0036e3b
|
| 28-Aug-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
damn
Hg-commit: e3a7c92ab17287bab1a86b3f92d92972bb76aea5
|
| #
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 ...
|
| #
551c0e33
|
| 27-Aug-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
commit after merge
Hg-commit: e8da3dbee7adf5e5a3bff8c938bdf7c8d59b9f19
|
| #
50d8bf02
|
| 27-Aug-2010 |
Jed Brown <jed@59A2.org> |
Prefetch updates
Use __builtin_prefetch when it's available. The control is at least as
precise as the _mm_prefetch intrinsics.
Add PETSC_PREFETCH_HINT_{NTA,T0,T1,T2} macros to handle implementati
Prefetch updates
Use __builtin_prefetch when it's available. The control is at least as
precise as the _mm_prefetch intrinsics.
Add PETSC_PREFETCH_HINT_{NTA,T0,T1,T2} macros to handle implementations
disagreeing about _mm_prefetch prototypes.
Changed prefetch call in relax.h to from T0 to NTA, it seems very
unlikely that this function really intended to prefetch only to
highest-level cache.
Hg-commit: 1bd3600f61c29cec4aa844c152ce9cf8b01912e8
show more ...
|
| #
d5bac5b1
|
| 09-Jul-2010 |
Victor Minden <victorminden@gmail.com> |
commit after merge
Hg-commit: dbe17f4cd57aa220aaa7f3a4929b4662f1fd3cc2
|
| #
d2ffe925
|
| 09-Jul-2010 |
Matthew G Knepley <knepley@gmail.com> |
Merge
Hg-commit: fb3fb77fa7d8bd71eb94b0f289024414b41b2354
|
| #
3f9fe445
|
| 08-Jul-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
finished labelling manual pages as Logically or Neighbor-wise Collective
Hg-commit: 8a714c954b7624fb9d7cacbaa621300eebb99517
|
| #
5148efb2
|
| 16-Jun-2010 |
Victor Minden <victorminden@gmail.com> |
commit after merge
Hg-commit: 8f08ad9f64d125fb53316ec5474fbfc0e8ec6eb9
|
| #
fa1c05ea
|
| 16-Jun-2010 |
Matthew G Knepley <knepley@gmail.com> |
Merge
Hg-commit: 41c58157a6a224c1869477ba893b728b564d6719
|
| #
a8b09249
|
| 16-Jun-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
three variables with bad const issues
Hg-commit: 167b35b9f2cf6d95f1d3c227aeaa3ca038b67448
|
| #
3649974f
|
| 16-Jun-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
changed many VecGetArray() to VecGetArrayRead()
Hg-commit: d7f64a5632750c1ca905eea86db48747b7367a8a
|
| #
05772624
|
| 07-Jun-2010 |
Victor Minden <victorminden@gmail.com> |
commit after merge
Hg-commit: 8eb1b0a369f9ab4b13dd8ccc92d9c82da0822d87
|
| #
ad8e3b70
|
| 07-Jun-2010 |
Matthew G Knepley <knepley@gmail.com> |
Merge
Hg-commit: 534539781f71eb6ba71ef89f4c82d23acf24e8e4
|
| #
ded6e084
|
| 06-Jun-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
make AIJ matrices be able to turn off use of inodes when inodes no longer exist. Previously once an AIJ matrix used Inodes it always used inodes.
Hg-commit: 3dfe5fed95605badef5e0b0e882202c04fe873e0
|
| #
4b9149d0
|
| 04-Jun-2010 |
Victor Minden <victorminden@gmail.com> |
Added functionality to do VecScale_Seq with CUDA
Hg-commit: 03fc0a1743fa5058e74e5c64d61114027987800a
|
| #
864a5514
|
| 04-Jun-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
commit after merge
Hg-commit: 46185e7fd336cffd34e37d6fb3000ca9cd647fbe
|
| #
936c6a71
|
| 04-Jun-2010 |
Matthew G Knepley <knepley@gmail.com> |
Merge
Hg-commit: f22d1eddb75a19febb7e6bbaead2dc4196ee1cfa
|
| #
17454e89
|
| 04-Jun-2010 |
Shri Abhyankar <abhyshr@mcs.anl.gov> |
Conversion of a few macros to static inline functions
Fixed fill ratio info for SeqAIJ LU symbolic factorization
Hg-commit: 70f9e9faca93960c98b60d63404804672fe0204b
|
| #
c2b86aee
|
| 24-May-2010 |
Hong Zhang <hzhang@mcs.anl.gov> |
bugfix of shift_pd for inode implementation
Hg-commit: b89092042bcc53148820ab79013a23a5c36160a9
|
| #
93cf5eb5
|
| 18-May-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
commit after merge
Hg-commit: aca5ed378a5658ef7df14846128d98fb472f8dd7
|
| #
b23f91a4
|
| 18-May-2010 |
Matthew G Knepley <knepley@gmail.com> |
Merge
Hg-commit: 3d6e115ed50e0a91aeb25dbd4b36559e2e6d6008
|
| #
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
|
| #
048b5e81
|
| 18-May-2010 |
Shri Abhyankar <abhyshr@mcs.anl.gov> |
Added factorization routines for BAIJ block size 1 (new data structure).
Inode with block size 4 was not being used in MatLUFactorNumeric_SeqAIJ_Inode.Fixed this.
Hg-commit: 5e4d1d7488fdaab08536c98c
Added factorization routines for BAIJ block size 1 (new data structure).
Inode with block size 4 was not being used in MatLUFactorNumeric_SeqAIJ_Inode.Fixed this.
Hg-commit: 5e4d1d7488fdaab08536c98c8a456ca9fc016047
show more ...
|
| #
d2d7d3aa
|
| 17-May-2010 |
Barry Smith <bsmith@mcs.anl.gov> |
commit after merge
Hg-commit: 072ee9112db7234a610102046f587237a18bdd45
|
| #
07b50cab
|
| 17-May-2010 |
Hong Zhang <hzhang@mcs.anl.gov> |
replace sctx.useshift with sctx.newshift
Hg-commit: 2f3e02680ad1d957577174697f9d3c9a99ba67d4
|