aboutsummaryrefslogtreecommitdiffhomepage
path: root/Eigen/src/Core/AssignEvaluator.h
Commit message (Collapse)AuthorAge
* Fix c++20 warnings about using enums in arithmetic expressions.Gravatar Rasmus Munk Larsen2021-06-10
|
* Revert "Revert "Adds EIGEN_CONSTEXPR and EIGEN_NOEXCEPT to rows(), cols(), ↵Gravatar Steve Bronder2021-03-24
| | | | | | innerStride(), outerStride(), and size()"" This reverts commit 5f0b4a4010af4cbf6161a0d1a03a747addc44a5d.
* Revert "Adds EIGEN_CONSTEXPR and EIGEN_NOEXCEPT to rows(), cols(), ↵Gravatar David Tellenbach2021-03-05
| | | | | | | innerStride(), outerStride(), and size()" This reverts commit 6cbb3038ac48cb5fe17eba4dfbf26e3e798041f1 because it breaks clang-10 builds on x86 and aarch64 when C++11 is enabled.
* Adds EIGEN_CONSTEXPR and EIGEN_NOEXCEPT to rows(), cols(), innerStride(), ↵Gravatar Steve Bronder2021-03-04
| | | | outerStride(), and size()
* Cast anonymous enums to int when used in expressions.Gravatar Rasmus Munk Larsen2021-02-24
|
* Fix unused warning on new `dense_assignment_loop` impl.Gravatar Antonio Sanchez2020-12-07
|
* Add specialization for compile-time zero-sized dense assignment.Gravatar Antonio Sanchez2020-12-07
| | | | | | | | | | | | | | | In the current `dense_assignment_loop` implementations, if the destination's inner or outer size is zero at compile time and if the kernel involves a product, we currently get a compile error (#2080). This is triggered by attempting to multiply a non-existent row by a column (or vice-versa). To address this, we add a specialization for zero-sized assignments (`AllAtOnceTraversal`) which evaluates to a no-op. We also add a static check to ensure the size is in-fact zero. This now seems to be the only existing use of `AllAtOnceTraversal`. Fixes #2080.
* bug #1680: improve MSVC inlining by declaring many triavial constructors and ↵Gravatar Gael Guennebaud2019-02-15
| | | | accessors as STRONG_INLINE.
* Make MaxPacketSize a true upper bound, even for fixed-size inputsGravatar Gael Guennebaud2018-11-16
|
* PR 544: Set requestedAlignment correctly for SliceVectorizedTraversalsGravatar Mark D Ryan2018-11-13
| | | | | | | | | | | | | | | | | | | | | | | Commit aa110e681b8b2237757a652ba47da49e1fbd2cd6 optimised the multiplication of small dyanmically sized matrices by restricting the packet size to a maximum of 4, increasing the chances that SIMD instructions are used in the computation. However, it introduced a mismatch between the packet size and the requestedAlignment. This mismatch can lead to crashes when the destination is not aligned. This patch fixes the issue by ensuring that the AssignmentTraits are correctly computed when using a restricted packet size. * * * Bind LinearPacketType to MaxPacketSize This commit applies any packet size limit specified when instantiating copy_using_evaluator_traits to the LinearPacketType, providing that the size of the destination is not known at compile time. * * * Add unit test for restricted packet assignment A new unit test is added to check that multiplication of small dynamically sized matrices works correctly when the packet size is restricted to 4 and the destination is unaligned.
* typoGravatar Gael Guennebaud2018-11-14
|
* PR 526: Speed up multiplication of small, dynamically sized matricesGravatar Mark D Ryan2018-10-12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The Packet16f, Packet8f and Packet8d types are too large to use with dynamically sized matrices typically processed by the SliceVectorizedTraversal specialization of the dense_assignment_loop. Using these types is likely to lead to little or no vectorization. Significant slowdown in the multiplication of these small matrices can be observed when building with AVX and AVX512 enabled. This patch introduces a new dense_assignment_kernel that is used when computing small products whose operands have dynamic dimensions. It ensures that the PacketSize used is no larger than 4, thereby increasing the chance that vectorized instructions will be used when computing the product. I tested all 969 possible combinations of M, K, and N that are handled by the dense_assignment_loop on x86 builds. Although a few combinations are slowed down by this patch they are far outnumbered by the cases that are sped up, as the following results demonstrate. Disabling Packed8d on AVX512 builds: Total Cases: 969 Better: 511 Worse: 85 Same: 373 Max Improvement: 169.00% (4 8 6) Max Degradation: 36.50% (8 5 3) Median Improvement: 35.46% Median Degradation: 17.41% Total FLOPs Improvement: 19.42% Disabling Packet16f and Packed8f on AVX512 builds: Total Cases: 969 Better: 658 Worse: 5 Same: 306 Max Improvement: 214.05% (8 6 5) Max Degradation: 22.26% (16 2 1) Median Improvement: 60.05% Median Degradation: 13.32% Total FLOPs Improvement: 59.58% Disabling Packed8f on AVX builds: Total Cases: 969 Better: 663 Worse: 96 Same: 210 Max Improvement: 155.29% (4 10 5) Max Degradation: 35.12% (8 3 2) Median Improvement: 34.28% Median Degradation: 15.05% Total FLOPs Improvement: 26.02%
* Add more debug outputGravatar Gael Guennebaud2018-09-21
|
* MIsc. source and comment typosGravatar luz.paz2018-03-11
| | | | Found using `codespell` and `grep` from downstream FreeCAD
* silent MSVC warningGravatar Gael Guennebaud2017-11-27
|
* Fix a gcc7 warning: Wint-in-bool-contextGravatar Gael Guennebaud2017-06-26
|
* fixed inlining issue with clang-cl on visual studioGravatar Alexander Neumann2017-02-08
| | | | | (grafted from 7962ac1a5855e8b7a60d5d90e61365b71f5501a5 )
* Fix warningGravatar Gael Guennebaud2017-01-27
|
* bug #1376: add missing assertion on size mismatch with compound assignment ↵Gravatar Gael Guennebaud2017-01-23
| | | | operators (e.g., mat += mat.col(j))
* bug #1356: fix calls to evaluator::coeffRef(0,0) to get the address of the ↵Gravatar Gael Guennebaud2016-12-05
| | | | | | destination by adding a dstDataPtr() member to the kernel. This fixes undefined behavior if dst is empty (nullptr).
* Fix compilation issue in mat = permutation (regression introduced in ↵Gravatar Gael Guennebaud2016-11-20
| | | | | | 8193ffb3d38b56c9295f204dc57dc6bac74f58aa )
* bug #1343: fix compilation regression in mat+=selfadjoint_view.Gravatar Gael Guennebaud2016-11-18
| | | | Generic EigenBase2EigenBase assignment was incomplete.
* Fix regression in X = (X*X.transpose())/s with X rectangular by deferring ↵Gravatar Gael Guennebaud2016-10-26
| | | | resizing of the destination after the creation of the evaluator of the source expression.
* Fixes for min and abs after Benoit's comments, switched to numext.Gravatar Robert Lukierski2016-10-13
|
* Fixes min() warnings.Gravatar Robert Lukierski2016-10-12
|
* Adding EIGEN_DEVICE_FUNC in the Geometry module.Gravatar Robert Lukierski2016-10-12
| | | | | Additional CUDA necessary fixes in the Core (mostly usage of EIGEN_USING_STD_MATH).
* Enable slice-vectorization+inner-unrolling when unaligned vectorization is ↵Gravatar Gael Guennebaud2016-07-28
| | | | allowed. For instance, this permits to vectorize 5x5 matrices (including product)
* Add debug outputGravatar Gael Guennebaud2016-07-06
|
* Re-enable some specializations for Assignment<.,Product<>>Gravatar Gael Guennebaud2016-07-05
|
* Change the semantic of the last template parameter of Assignment from ↵Gravatar Gael Guennebaud2016-07-04
| | | | | | | "Scalar" to "SFINAE" only. The previous "Scalar" semantic was obsolete since we allow for different scalar types in the source and destination expressions. On can still specialize on scalar types through SFINAE and/or assignment functor.
* Include the cost of stores in unrolling (also fix infinite unrolling with ↵Gravatar Gael Guennebaud2016-06-15
| | | | expression costing 0 like Constant)
* Relax mixing-type constraints for binary coefficient-wise operators:Gravatar Gael Guennebaud2016-06-06
| | | | | | | | | | - Replace internal::scalar_product_traits<A,B> by Eigen::ScalarBinaryOpTraits<A,B,OP> - Remove the "functor_is_product_like" helper (was pretty ugly) - Currently, OP is not used, but it is available to the user for fine grained tuning - Currently, only the following operators have been generalized: *,/,+,-,=,*=,/=,+=,-= - TODO: generalize all other binray operators (comparisons,pow,etc.) - TODO: handle "scalar op array" operators (currently only * is handled) - TODO: move the handling of the "void" scalar type to ScalarBinaryOpTraits
* Remove debuging code.Gravatar Gael Guennebaud2016-05-26
|
* Introduce internal's UIntPtr and IntPtr types for pointer to integer ↵Gravatar Gael Guennebaud2016-05-26
| | | | | | | | conversions. This fixes "conversion from pointer to same-sized integral type" warnings by ICC. Ideally, we would use the std::[u]intptr_t types all the time, but since they are C99/C++11 only, let's be safe.
* bug #256: enable vectorization with unaligned loads/stores.Gravatar Gael Guennebaud2016-05-24
| | | | | This concerns all architectures and all sizes. This new behavior can be disabled by defining EIGEN_UNALIGNED_VECTORIZE=0
* Clean propagation of Dest/Src alignments.Gravatar Gael Guennebaud2016-05-24
|
* bug #1207: Add and fix logical-op warningsGravatar Christoph Hertzberg2016-05-11
|
* Fix performance regression: with AVX, unaligned stores were emitted instead ↵Gravatar Gael Guennebaud2016-05-01
| | | | of aligned ones for fixed size assignement.
* Improve half-packet vectorization logic to distinguish linear versus inner ↵Gravatar Gael Guennebaud2016-04-13
| | | | traversal modes.
* Fix static/inline keywords order.Gravatar Gael Guennebaud2016-04-11
|
* Removed executable bit from header filesGravatar Benoit Steiner2016-03-23
|
* Improve inliningGravatar Gael Guennebaud2016-02-08
|
* bug #667: declare several critical functions as FORECE_INLINE to make ICC ↵Gravatar Gael Guennebaud2016-01-31
| | | | | | | happier. <g.gael@free.fr> HG: branch 'default' HG: changed Eigen/src/Core/ArrayBase.h HG: changed Eigen/src/Core/AssignEvaluator.h HG: changed Eigen/src/Core/CoreEvaluators.h HG: changed Eigen/src/Core/CwiseUnaryOp.h HG: changed Eigen/src/Core/DenseBase.h HG: changed Eigen/src/Core/MatrixBase.h
* bug #1144: fix regression in x=y+A*x (aliasing), and move ↵Gravatar Gael Guennebaud2016-01-09
| | | | evaluator_traits::AssumeAliasing to evaluator_assume_aliasing.
* Remove useless "explicit", and fix inline/static order.Gravatar Gael Guennebaud2015-12-11
|
* Simplify cost computations based on HugeCost being smaller that unrolling limitGravatar Gael Guennebaud2015-10-28
|
* Refactoring of the cost model:Gravatar Gael Guennebaud2015-10-28
| | | | | | | | | | | - Dynamic is now an invalid value - introduce a HugeCost constant to be used for runtime-cost values or arbitrarily huge cost - add sanity checks for cost values: must be >=0 and not too large This change provides several benefits: - it fixes shortcoming is some cost computation where the Dynamic case was not properly handled. - it simplifies cost computation logic, and should avoid future similar shortcomings. - it allows to distinguish between different level of dynamic/huge/infinite cost - it should enable further simplifications in the computation of costs (save compilation time)
* Fix assign vectorization logic with respect to fixed outer-strideGravatar Gael Guennebaud2015-10-27
|
* Improve readibility of EIGEN_DEBUG_ASSIGN mode.Gravatar Gael Guennebaud2015-10-27
|
* Some cleaning in evaluatorsGravatar Gael Guennebaud2015-10-08
|
generated by cgit v1.2.3 (git 2.39.1) at 2024-05-18 01:08:39 +0000