Christoph Hertzberg
096af59799
Fix bug #784 : Assert if assigning a product to a triangularView does not match the size.
2014-04-04 17:48:37 +02:00
Benoit Steiner
8044b00a7f
bug #782 : Workaround for gcc <= 4.4 compilation error on the NEON PacketMath code.
2014-04-03 23:41:47 +02:00
Gael Guennebaud
fb03b56647
Fix warning
2014-03-27 11:38:35 +01:00
Gael Guennebaud
f0a4c9d5ab
Update gebp kernel to process a panle of 4 columns at once for the remaining ones.
2014-03-26 23:22:36 +01:00
Gael Guennebaud
8be011e776
Remove remaining bits of the dead working buffer
2014-03-26 23:14:44 +01:00
Gael Guennebaud
bc401eb6fa
Implement new 1 packet x 8 gebp kernel
2014-03-26 18:53:00 +01:00
Gael Guennebaud
b286a1e75c
add pbroadcast2/4 generic intrinsics
2014-03-26 16:46:36 +01:00
Benoit Steiner
c8c81c1e74
Improved the efficiency if the block-panel matrix multiplication code: the change reduces the pressure on the L1 cache by removing the calls to gebp_traits::unpackRhs(). Instead the packetization of the rhs blocks is done on the fly in gebp_traits::loadRhs(). This adds numerous calls to pset1<ResPacket> (since we're packetizing on the fly in the inner loop) but this is more than compensated by the fact that we're decreasing the memory transfers by a factor RhsPacketSize.
2014-01-02 16:18:32 -08:00
Gael Guennebaud
01fd880424
Revert previous change and introduce a new workaround regarding gcc generating a shufps instruction instead of the more efficient pshufd instruction.
...
The trick consists in introducing a new pload1 function to be used in low level product kernels for which bug #203 does not apply.
Indeed, it turned out that using inline assembly prevents gcc of doing a good job at instructtion reordering.
2014-03-20 16:03:46 +01:00
Gael Guennebaud
c39a3fa7a1
Makes gcc to generate a pshufd instruction for pset1
2014-03-20 10:14:26 +01:00
Gael Guennebaud
2a564695f0
Simpler and hopefully more future-proof fix for bug #503 (aligned_allocator with c++11)
2014-03-19 13:28:50 +01:00
Christoph Hertzberg
35a2c9cde7
clang does not accept this without template keyword
2014-03-14 16:48:29 +01:00
Gael Guennebaud
bb4b67cf39
Relax Ref such that Ref<MatrixXf> accepts a RowVectorXf which can be seen as a degenerate MatrixXf(1,N)
2014-03-13 18:04:19 +01:00
Christoph Hertzberg
2379ccffcb
bug #755 : CommaInitializer produced wrong assertions in absence of ReturnValueOptimization.
2014-03-12 13:48:09 +01:00
Benoit Steiner
ce99b502ce
Use vectorization when packing row-major rhs matrices. (bug #717 )
2013-12-17 10:49:43 -08:00
Gael Guennebaud
ba2f79e680
Fix selfadjoint_matrix_vector_product for complex with packet size > 2 (e.g., AVX)
2014-03-07 23:18:20 +01:00
Christoph Hertzberg
d5cc083782
Fixed bug #754 . Only inserted (!defined(_WIN32_WCE)) analog to alloc and free implementation (not tested, but should be correct).
2014-03-05 14:50:00 +01:00
Gael Guennebaud
7313f32efa
Help MSVC to inline some trivial functions
2014-03-04 17:24:00 +01:00
Christoph Hertzberg
04e1e38eed
bug #289 : Removed useless static keywords
2014-03-04 15:10:29 +01:00
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
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
0508af4287
Merged in martinhofernandes/eigen (pull request PR-40)
...
Better fix for bug #503
2014-02-14 15:31:39 +01:00
Gael Guennebaud
97965dde9b
alloca is not necessarily alligned on windows
2014-02-14 00:04:38 +01:00
Gael Guennebaud
3291580630
Fix bug #740 : overflow issue in stableNorm
2014-02-13 15:44:01 +01:00
Jitse Niesen
7ea6ef8969
Fix documentation of MatrixBase::applyOnTheLeft (bug #739 )
...
Add examples; move methods from EigenBase.h to MatrixBase.h
2014-02-12 14:03:39 +00:00
Gael Guennebaud
a7621809fe
Remove useless register keyword, and optimize predux_min/max for SSE4
2014-01-25 16:54:13 +01:00
Gael Guennebaud
deab937d45
NVCC: fix closed-form eigenvalue decomposition, workaround gcc4.7/nvcc5.5 issue
2014-01-24 12:50:29 +01:00
Martinho Fernandes
4c08385b74
Merged eigen/eigen into default
2014-01-10 11:22:24 +01:00
Martinho Fernandes
4ccff2d028
Placement new must use void* to avoid user-specific overloads.
2014-01-10 11:20:40 +01:00
Martinho Fernandes
3a4616d6e3
Add C++11 allocator overloads to avoid implicit conversions.
2014-01-10 11:02:11 +01:00
Gael Guennebaud
92190a1caf
Add an example showing how to use C++11 random distributions
2014-01-07 20:23:35 +01:00
Gael Guennebaud
ac409f51f1
Document the fact that Random and setRandom are not reentrant (so not thread-safe)
2014-01-07 20:17:59 +01:00
Gael Guennebaud
49034d1570
Fix bug #708 : add placement new/delete for array
2013-11-27 09:46:59 +01:00
Gael Guennebaud
46dd1bb1be
Workaround fixing aliasing issue in x = SparseLU::solve(x)
2013-11-15 11:19:19 +01:00
Christoph Hertzberg
e59b38abef
Implement boolean reductions for zero-sized objects
2013-11-13 16:47:02 +01:00
Leszek Swirski
b93520b1a5
Install functor folder with cmake
2013-11-08 14:07:11 +00:00
Gael Guennebaud
cb8da751a0
fix broken commit
2013-11-07 22:44:37 +01:00
Gael Guennebaud
76c230a84d
Add an option to test evaluators globally
2013-11-07 16:38:14 +01:00
Gael Guennebaud
57327cc2d5
Drop evaluators for SwapWrapper and SelfCwiseBinaryOp
2013-11-07 14:07:27 +01:00
Gael Guennebaud
5887e82729
Clean evaluator_impl_base. It will probably be removed in the future
2013-11-07 14:02:47 +01:00
Gael Guennebaud
af9851d1d7
bug #99 : move the creation of the evaluator to a central place, and make generic_dense_assignment_kernel hold the destination and source evaluators
2013-11-07 12:03:12 +01:00
Gael Guennebaud
8fe609311d
Move internal::swap to numext to fix ambiguous call with std::swap
2013-11-07 09:01:26 +01:00
Gael Guennebaud
8edc964734
bug #99 : refactor assignment and compound assignment mechanism through "assignment functors" and "assignement kernels".
...
The former is very low level and generic. The later abstarct the former for dense expressions. This refactoring permits
to get rid of the very ugly SwapWrapper and SelfCwiseBinaryOp classes.
In the future, this will also permit to simplify all these evaluation loops and perhaps to reuse them for reduxions.
That will also permit to specialize for operations like expr1 += expr2 outside Eigen, and so for any kind
of expressions (dense, sparse, tensor, etc.)
2013-11-06 18:17:59 +01:00
Gael Guennebaud
a37bdfc955
Fix static/inline order
2013-11-06 11:13:31 +01:00
Gael Guennebaud
03de5c2410
Split the huge Functors.h file
2013-11-06 10:36:10 +01:00
Gael Guennebaud
87aee5fda1
Allow calling attributes of dynamic size objects from device
2013-11-05 15:40:58 +01:00
Gael Guennebaud
1bb1a57ef7
merge with default branch
2013-11-05 10:31:59 +01:00
Gael Guennebaud
19521c83b8
bug #677 : fix usage of pld instrinsics for ccomplexes
2013-11-02 12:10:48 +01:00
Gael Guennebaud
6dc0e59b1e
Fix bug #677 : compilation issue on arm64 which does not have the PLD instruction
2013-10-31 13:52:43 +01:00
