Benoit Steiner
131027ee0a
Merged eigen/eigen into default
2014-02-24 13:54:07 -08:00
Benoit Steiner
db7d49efbb
Added support for FMA instructions
2014-02-24 13:45:32 -08:00
Gael Guennebaud
9fdc6258cf
Implement bug #317 : use a template function call to suppress unused variable warnings. This also fix the issue of the previous changeset in a much nicer way.
2014-02-24 18:13:49 +01:00
Gael Guennebaud
21fecd5252
Workaround clang ABI change with unsed arguments (ugly fix)
2014-02-24 17:12:17 +01:00
Jitse Niesen
6fecb6f1b6
Fix bug #748 - array_5 test fails for seed 1392781168.
2014-02-24 14:10:17 +00:00
Christoph Hertzberg
3e439889e0
Specify what non-resizeable objects are in transposeInPlace and adjointInPlace (cf bug #749 )
2014-02-24 13:12:10 +01:00
Gael Guennebaud
cbc572caf7
Split LU/Inverse.h to Core/Inverse.h for the generic Inverse expression, and LU/InverseImpl.h for the dense implementation of dense.inverse()
2014-02-24 11:49:30 +01:00
Gael Guennebaud
1e0c2f6ddb
Hide some deprecated classes.
2014-02-24 11:41:19 +01:00
Gael Guennebaud
c98881e130
By-pass ProductBase for triangular and selfadjoint products and get rid of ProductBase
2014-02-23 22:51:13 +01:00
Gael Guennebaud
d67548f345
Get rid of GeneralProduct<> for GemvProduct
2014-02-21 17:13:28 +01:00
Gael Guennebaud
6c7ab50811
Get rid of GeneralProduct<> for GemmProduct
2014-02-21 16:43:03 +01:00
Gael Guennebaud
728c3d2cb9
Get rid of GeneralProduct for outer-products, and get rid of ScaledProduct
2014-02-21 16:27:24 +01:00
Gael Guennebaud
af31b6c37a
Generalize evaluator<Inverse<>> such that there is no need to specialize it
2014-02-21 15:22:08 +01:00
Gael Guennebaud
93125e372d
Port LU module to evaluators (except image() and kernel())
2014-02-20 15:26:15 +01:00
Gael Guennebaud
b2e1453e1e
Some bit flags and internal structures are deprecated
2014-02-20 15:25:06 +01:00
Gael Guennebaud
9621333545
Fix dimension of Solve expression
2014-02-20 15:24:21 +01:00
Gael Guennebaud
5f6ec95291
Propagate LvalueBit flag to TriangularView
2014-02-20 15:24:00 +01:00
Gael Guennebaud
ecd2c8f37b
Add general Inverse<> expression with evaluator
2014-02-20 14:18:24 +01:00
Gael Guennebaud
5960befc20
More int versus Index fixes
2014-02-19 21:42:29 +01:00
Gael Guennebaud
2eee6eaf3c
Fix mixing scalar types with evaluators
2014-02-19 16:30:17 +01:00
Gael Guennebaud
8af02d19b2
ExprType::Nested has a new meaning now...
2014-02-19 15:16:11 +01:00
Gael Guennebaud
95b0a6707b
evaluator<Replicate> must evaluate its argument to avoid redundant evaluations
2014-02-19 14:51:46 +01:00
Gael Guennebaud
b1ab6a8e0b
Add missing assertion in swap()
2014-02-19 14:06:35 +01:00
Gael Guennebaud
61cff28618
Disable Flagged and ForceAlignedAccess
2014-02-19 14:05:56 +01:00
Gael Guennebaud
68e8ddaf94
Fix vectorization logic wrt assignment functors
2014-02-19 13:26:07 +01:00
Gael Guennebaud
3a735a6cf1
Fix lazy evaluation in Ref
2014-02-19 13:17:41 +01:00
Gael Guennebaud
ccc41128fb
Add a Solve expression for uniform treatment of solve() methods.
2014-02-19 11:33:29 +01:00
Gael Guennebaud
b3a07eecc5
Fix CoeffReadCost of products to handle Dynamic costs
2014-02-19 11:32:04 +01:00
Gael Guennebaud
c16b80746a
isApprox must honors nested_eval
2014-02-19 11:30:58 +01:00
Benoit Steiner
cbd7e98174
Merged the latest version of the code from eigen/eigen
2014-02-18 18:51:24 -08:00
Benoit Steiner
7ed9441ea4
Reverted the definition of the EIGEN_ALIGN to its former meaning (i.e. a boolean)
...
Created a new EIGEN_ALIGN_BYTES define to encode how the data should be aligned
Fixed a few remaining alignment issues exposed when the Eigen code is compiled with avx enabled.
Created a new EIGEN_ALIGN_DEFAULT define, which is set to the minimum alignment value required for the chosen instruction set. Use this value instead of EIGEN_ALIGN32 to preserve the existing alignment on SSE/Altivec/Neon.
2014-02-18 18:06:44 -08:00
Gael Guennebaud
5b78780def
Add evaluator shortcut for triangular ?= product
2014-02-18 17:43:16 +01:00
Gael Guennebaud
8169c6ac59
Simplify implementation of coeff-based products to fully exploit our reduxion mechanisms.
...
If this results in performance regressions, then we should optimize reduxion rather than
somehow duplicate the code.
2014-02-18 16:57:25 +01:00
Gael Guennebaud
463554c254
Merge with default branch
2014-02-18 15:45:39 +01:00
Gael Guennebaud
82c066b3c4
Cleaning
2014-02-18 15:44:32 +01:00
Gael Guennebaud
0543cb51b5
Product::coeff method are also OK for lazy products (including diagonal products)
2014-02-18 14:51:41 +01:00
Gael Guennebaud
99e27916cf
Fix all()/any() for evaluators
2014-02-18 14:26:25 +01:00
Gael Guennebaud
37a1d736bf
_MatrixTypeNested must be public in sparse Block
2014-02-18 13:35:24 +01:00
Gael Guennebaud
06545058bb
Temporary workaround for permutations
2014-02-18 13:33:04 +01:00
Gael Guennebaud
7002aa858f
Support Product::coeff(0,0) even for dynamic matrices
2014-02-18 13:32:30 +01:00
Gael Guennebaud
8cfb138e73
Finally, the simplest remains to deffer resizing at the latest
2014-02-18 13:31:44 +01:00
Gael Guennebaud
1b5de5a37b
Add evaluator for Ref
2014-02-18 13:30:16 +01:00
Gael Guennebaud
a08cba6b5f
Move is_diagonal to XprHelper, forward declare Ref
2014-02-18 11:03:59 +01:00
Gael Guennebaud
573c587e3d
New design for handling automatic transposition
2014-02-18 10:53:14 +01:00
Gael Guennebaud
551bf5c66a
Get rid of DiagonalProduct
2014-02-18 10:52:26 +01:00
Gael Guennebaud
2d136d3d7f
Get rid of SeflCwiseBinaryOp
2014-02-18 10:52:00 +01:00
Gael Guennebaud
873401032b
Fix scalar * product optimization when 'product' includes a selfadjoint matrix
2014-02-17 19:00:45 +01:00
Gael Guennebaud
d595fd31f5
Deal with automatic transposition in call_assignment, fix a few shortcomings
2014-02-17 16:11:55 +01:00
Gael Guennebaud
bffa15142c
Add evaluator support for diagonal products
2014-02-17 16:10:55 +01:00
Christoph Hertzberg
b14a4628af
Relaxed umeyama test. Problem was ill-posed if linear part was scaled with very small number. This should fix bug #744 .
2014-02-17 13:48:00 +01:00
