Gael Guennebaud
ec39a39cb6
fix LLT for zero matrix
2010-07-15 20:00:34 +02:00
Gael Guennebaud
ccc6731f86
fix use of rank in QR
2010-07-15 19:59:21 +02:00
Gael Guennebaud
b09bb50aeb
fix QR solving with m>n
2010-07-15 19:52:11 +02:00
Gael Guennebaud
c7b8de77c0
fix aligned_delete for null pointers
2010-07-15 09:28:29 +02:00
Gael Guennebaud
c44bbabdcc
fix LU and QR solve when rank==0, fix LLT when the matrix is purely 0
2010-07-12 16:42:38 +02:00
Gael Guennebaud
468863f3a0
fix bad Map in row-vector by matrix products
2010-07-09 20:27:05 +02:00
Gael Guennebaud
81f36b0e21
disable MSVC optimization when the underlying compiler is ICC
2010-07-09 19:36:43 +02:00
Benoit Jacob
c196a49f67
Added tag 2.0.14 for changeset e18e51d8913fb3acc6aa2f934c8de7c96651ebb9
2010-06-22 22:23:39 -04:00
Benoit Jacob
e18e51d891
bump
2010-06-22 22:22:11 -04:00
Stuart Glaser
58fc972ed3
LU on limited-size matrices no longer allocates for temporaries.
2010-06-21 21:29:45 -07:00
Benoit Jacob
e7b6a4bcba
fix bug introduced yesterday preventing vectorization of vectors when the storage order is not "the right one".
...
expand a little the vectorization_logic test and backport EIGEN_DEBUG_ASSIGN.
2010-06-22 09:24:07 -04:00
Benoit Jacob
eaa81c135a
fix brain dead computation of the aligned bit.
...
When using a max-size that is fixed and not a multiple of 16 bit, we're not aligned.
2010-06-21 21:07:53 -04:00
Benoit Jacob
ad8b6c2342
fix #127 : remove static keywords that had no effect anyway since the forward-declaration wasn't static,
...
and that would have been bad if they had taken effect.
2010-06-16 08:28:34 -04:00
Benoit Jacob
37fe67372b
Added tag 2.0.13 for changeset ee499a855c066609123c5c5091f6e052e774e8d0
2010-06-10 08:06:38 -04:00
Benoit Jacob
ee499a855c
bump version
2010-06-10 08:06:06 -04:00
Benoit Jacob
bc0ce37395
mention that that issue has been solved in GCC 4.5
2010-06-10 08:02:20 -04:00
Benoit Jacob
65c0b2a04d
install the Eigen and Dense public headers !!!!!
2010-06-10 08:02:02 -04:00
Benoit Jacob
93d8d0e1b5
add unit test for bug #132
2010-06-09 20:50:18 -04:00
Benoit Jacob
501063d9e9
Fix bug #132
...
In the matrix-vector products, we were calling coeffRef on the vector xpr without checking it has DirectAccess.
Will add unit test (since it's in 2.0, just import the test case provided in the bug report).
Confirming that this can't happen in the devel branch, and that if we tried to call coeffRef on an xpr
without DirectAccess, that would not compile (since the DenseCoeffsBase class was introduced).
2010-06-09 19:39:05 -04:00
Gael Guennebaud
c76c8d6917
fix issue #125
2010-05-31 22:53:02 +02:00
Thomas Capricelli
0bb8688d70
sync .hgignore with tip
2010-05-26 02:37:52 +02:00
Thomas Capricelli
ea87337647
fix readcost for complex types (backport from tip)
2010-05-26 02:34:55 +02:00
Piotr Trojanek
12557fb2a2
more std:: namespace fixups for 2.0 branch
2010-05-01 09:25:36 +09:00
Benoit Jacob
8a95876825
with QCC, don't try passing --version
2010-04-29 07:58:47 -04:00
Benoit Jacob
82d7c4e1d0
forgot hg add
2010-04-26 07:54:19 -04:00
Benoit Jacob
8d4b0aae04
Fix bug #93 : add platform check for how to link to the standard math library.
...
This allows to support QNX.
2010-04-19 11:30:46 -04:00
Gael Guennebaud
4785e27d6a
fix btl compilation
2010-04-01 12:34:55 +02:00
Benoit Jacob
20b544b444
disable all alignment with QCC/QNX in eigen 2.0
2010-03-06 00:31:55 -05:00
Piotr Trojanek
135013c608
EIGEN_ALIGN std:: fixup
2010-03-04 16:38:37 +01:00
Piotr Trojanek
1625a5e3f8
fixups for clean QCC compilation (add std:: in front of size_t, memcpy, etc.)
2010-03-05 11:21:11 +01:00
Piotr Trojanek
47a61bbd80
posix_memalign check for QNX
2010-03-04 17:12:30 +01:00
Benoit Jacob
12bcfae0c5
clarify EIGEN_DONT_ALIGN docs wrt versions
2010-02-28 15:56:50 -05:00
Benoit Jacob
5f42104e0a
really fix the LDLt, at the expense of letting isPositiveDefinite() always return true (it was fundamentally broken anyway, especially as in 2.0 we don't even pivot at all).
...
also, fix compilation
2010-02-23 18:10:31 -05:00
Benoit Jacob
6b3f81b414
precision improvement. Wtf were we using sqrt(precision) for the cutoff? Replaced by precision*biggest.
2010-02-23 16:26:39 -05:00
Gael Guennebaud
1f4b8e6a36
compilation fix in ldlt() for non matrix types
...
(transplanted from 608959aa6fce375abb92872350267074a1d90283
)
2010-02-21 10:29:19 +01:00
Benoit Jacob
e1f61b40c8
oops, this had to be done here, not at the end.
2010-02-12 09:02:25 -05:00
Benoit Jacob
f369dc873e
Piotr's patch was missing many occurences of size_t. So,
...
using std::size_t;
This is the only way that we can ensure QCC support in the long term without having to think about it everytime.
2010-02-12 08:57:04 -05:00
Benoit Jacob
c03bca21c4
Added tag 2.0.12 for changeset ed6eb5a625756b0e7123c0b9a40e5c4dbe3c4547
2010-02-11 21:39:46 -05:00
Benoit Jacob
ed6eb5a625
bump
2010-02-11 21:39:41 -05:00
Benoit Jacob
9488a12125
work around brain dead ICC
2010-02-11 19:32:56 -05:00
Piotr Trojanek
7b44957c4b
std:: namespace fixup for more restricive compilers such as QNX's QCC
2010-02-10 22:27:35 +01:00
Hauke Heibel
743ad75595
BenchTimer backport (clock_gettime & QueryPerformanceCounter).
2010-02-03 21:55:01 +01:00
Benoit Jacob
a9eabed421
Patch by 'Wolf' from the issue tracker:
...
Fix bug #90 , missing type cast in LU, allow to use LU with MPFR.
2010-02-02 07:06:15 -05:00
Benoit Jacob
cd34a1d351
backport bug fix by Jitse.
2010-01-28 14:00:09 -05:00
Benoit Jacob
3e963ee69d
EIGEN_ENUM_MIN ---> EIGEN_SIZE_MIN
2010-01-26 20:37:57 -05:00
Benoit Jacob
6cc9dc17f2
In LU / Inverse, decouple the output type from the input type.
...
This has long been done in the default branch
2010-01-26 18:45:23 -05:00
Gael Guennebaud
7852a48a2f
fix matrix product with EIGEN_DEFAULT_TO_ROW_MAJOR
2010-01-25 21:56:01 +01:00
Benoit Jacob
d209120180
* Introduce EIGEN_DEFAULT_TO_ROW_MAJOR tests option
...
---> Now only product_large fails with EIGEN_DEFAULT_TO_ROW_MAJOR.
* Fix EIGEN_NO_ASSERTION_CHECKING tests option
* Fix a crash in Tridiagonalization on row-major matrices + SSE
* Fix inverse test (numeric stability noise)
* Extend map test (see previous fixes in MapBase)
* Fix vectorization_logic test for row-major
* Disable sparse tests with EIGEN_DEFAULT_TO_ROW_MAJOR
2010-01-25 14:00:02 -05:00
Thomas Capricelli
55c0707b1d
fix the script again (definitely?) + cleaning
2010-01-22 19:28:33 +01:00
Benoit Jacob
72044ca925
fix a super nasty bug: on row-major expressions that are NOT vectors but that
...
do have LinearAccess, the MapBase::coeff(int) and MapBase::coeffRef(int)
methods were broken.
2010-01-21 23:33:20 -05:00
