idstatetitleauthorsource
299 OPEN Avoid use of build-directory Eigen3Config.cmake Christopher Dembia chrisdembia/eigen
488 OPEN Ensure that traits::Scalar and Xpr::Scalar are never const. g.gael@free.fr ggael/eigen3
125 OPEN #789: Memory allocation only when necessary jorisv31 jorisv31/eigen
613 OPEN Add additional FMA operations pmsub and pnmadd. Christoph Hertzberg chtz/eigen-add_more_fma
592 OPEN Bug 1655: Replace EIGEN_TEST_CXX11 by EIGEN_TEST_CXX_STANDARD option. Christoph Hertzberg chtz/eigen-test-cxx_version
556 OPEN Adding NEON specific Quaternion multiplication. Ritesh Gupta gupta_ritesh/eigen
600 OPEN Extend Quaternion with general purpose operators g.gael@free.fr ggael/eigen-quat2
340 OPEN New fast sparse QR solvers for special structure matrices Jan Svoboda jasvob/eigen_sparse_qr
37 OPEN Polynomials module derivative David J. Luitz dluitz/eigen
320 OPEN Add macro to define BLAS index type. Péricles Lopes Machado ar2tech/eigen
43 OPEN OpenGLSupport tweaks for OpenGL 3.0 Andrea Tagliasacchi drtaglia/eigen-opengl
73 OPEN Text-based DenseBase input Chen-Pang He jdh8/eigen
112 OPEN Added support for Unified Virtual Memory to the tensor library Benoit Steiner benoitsteiner/eigen
388 OPEN QRKit: Sparse, Composable QR Decompositions Jan Svoboda jasvob/eigen_sparse_qr_solvers
473 OPEN Add support of Intel (R) MKL Inspector-Executor SparseBLAS (current stage: SparseMatrix*DenseVector) Maria Zhukova mzhukova/mklspblassupport
758 OPEN Improved NEON support Joel Holdsworth realjhol/eigen
372 OPEN Enabled different compatible custom scalar types for the decomposition and the rhs/dest memory of HouseholderQR (Changed product_triangular_matrix_matrix to have LhsScalar and RhsScalar template parameters, and more) Patrick Peltzer patrick_peltzer/eigen
296 OPEN Swizzle operators for Eigen Julian Amann Vertexwahn/eigen
594 OPEN Add storage option KeepMemory jmirabel jmirabel/eigen
686 OPEN Tensor module - added enstein notation godeffroy valet godeffroy/eigen
769 MERGED Capture TensorMap by value inside tensor expression AST Eugene Zhulenev ezhulenev/eigen-01
754 MERGED Update ConfigureVectorization.h to not optimize fp16 routines when compiling with cuda. Anshul Jaiswal anshuljl/eigen-2
650 OPEN Added AlignedBox::transformed(AffineTransform). Martin Pecka peci1/eigen
767 MERGED Fix shadow warnings in AlignedBox and SparseBlock Eugene Zhulenev ezhulenev/eigen-02
768 MERGED Fix for the HIP build+test errors. Deven Desai deven-amd/eigen-hip-fix-191201
763 DECLINED Fix for the HIP build+test errors. Deven Desai deven-amd/eigen-hip-fix-191120
673 DECLINED [SYCL] Adding SYCL kernels for unsupported modules Mehdi Goli codeplaysoftware/eigen
691 MERGED SYCL Backend Mehdi Goli codeplaysoftware/eigen
760 MERGED Fix warnings Joel Holdsworth realjhol/eigen
766 MERGED Add async evaluation support to TensorReverse Eugene Zhulenev ezhulenev/eigen-02
765 MERGED Add async evaluation support to TensorPadding/TensorImagePatch/TensorShuffling Eugene Zhulenev ezhulenev/eigen-02
762 DECLINED Add async evaluation support to TensorImagePatch and TensorPadding Eugene Zhulenev ezhulenev/eigen-01
764 OPEN Fixed printing of char and unsigned char matrices Joel Holdsworth realjhol/eigen
761 DECLINED IO: Fixed printing of char and unsigned char matrices Joel Holdsworth realjhol/eigen
482 OPEN cmake: Support multi-configuration generator Jean-Christophe Fillion-Robin jcfr/eigen
759 DECLINED Use modern cmake paradigms for install directory Hans Johnson hjmjohnson/eigen_cmake_subdirectory_modernize
749 DECLINED REMOVE ME. This is mercurial mess that can not be fixed. Other PR's will be sent from clean forks. Hans Johnson hjmjohnson/eigen
757 DECLINED Fix prsqrt to return 0 for +inf arguments. Rasmus Larsen rmlarsen/eigen
751 DECLINED "Fixed compilation issue when compiling using MSVC with /arch:AVX512 flag" Sakshi Goynar sakshigoynar/eigen_avx512_msvc
748 DECLINED BUG: Xcode 11.0 with default -macosx-version-min=10.15 seg faults Hans Johnson hjmjohnson/eigen
756 MERGED Remove legacy block evaluation support Eugene Zhulenev ezhulenev/eigen-01
755 MERGED Break loop dependence in TensorGenerator block access Eugene Zhulenev ezhulenev/eigen-01
752 DECLINED Add EIGEN_HAS_INTRINSIC_INT128 macro Sam Hasinoff hasinoff/eigen
734 DECLINED Update ConfigureVectorization.h to not optimize fp16 routines when compiling with cuda. Anshul Jaiswal anshuljl/eigen-1
746 MERGED Remove internal::smart_copy and replace with std::copy Eugene Zhulenev ezhulenev/eigen-01
750 DECLINED Fix for the HIP build+test errors. Deven Desai null
753 MERGED Rollback of PR #746 and partial rollback of https://bitbucket.org/eigen/eigen/commits/aa92a050d7e0802454870b8662b117361da7b9ab. Rasmus Larsen rmlarsen/eigen
744 DECLINED BUG: Xcode 11.0 with default -macosx-version-min=10.15 seg faults Hans Johnson null
740 DECLINED Fix for the HIP build+test errors. Deven Desai null
747 MERGED Cleanup includes in Tensor module after switch to C++11 and above Eugene Zhulenev ezhulenev/eigen-02
745 DECLINED Remove internal::smart_copy and replace with std::copy Eugene Zhulenev ezhulenev/eigen-01
743 MERGED [CXX11Meta] Use std::array front to get to a first element pointer. Eugene Zhulenev ezhulenev/eigen-01
742 MERGED Prevent potential ODR in TensorExecutor Eugene Zhulenev ezhulenev/eigen-01
741 DECLINED Fix template specialization compilation error with GCC Eugene Zhulenev ezhulenev/eigen-01
739 MERGED This PR fixes array specialization and implicit call to std::array constructor for older GCC version Mehdi Goli codeplaysoftware/eigen-upstream-pure
720 DECLINED Add async evaluation support to reduction, forced evaluation, evalto, and reshape. This makes the cross entropy op in TensorFlow 18% faster on CPU. Rasmus Larsen null
738 MERGED Fix for the HIP build+test errors. Deven Desai null
736 DECLINED Fix for the HIP build+test errors. Deven Desai null
737 DECLINED Add block evaluation V2 to TensorAsyncExecutor. Rasmus Larsen rmlarsen/eigen2
735 MERGED Drop support for c++03 in Eigen tensor. Rasmus Larsen rmlarsen/eigen2
733 MERGED Propagate block evaluation preference through rvalue tensor expressions Eugene Zhulenev ezhulenev/eigen-01
732 MERGED Cleanup Tensor block destination and materialized block storage allocation Eugene Zhulenev ezhulenev/eigen-01
731 MERGED TensorBroadcasting support for random/uniform blocks Eugene Zhulenev ezhulenev/eigen-02
730 MERGED Block evaluation for TensorGenerator/TensorReverse/TensorShuffling Eugene Zhulenev ezhulenev/eigen-01
702 DECLINED MatrixMarketIterator.h pass for . and .. in Getnextvalidmatrix Vincent HUBER null
707 DECLINED Bug 1746: Add move-constructor and move-assignment-operator for PermutationMatrix and Transpositions Frank Zingsheim zingsheim/eigen
729 MERGED Block evaluation for TensorGenerator + TensorReverse + fixed bug in tensor reverse op Eugene Zhulenev ezhulenev/eigen-01
728 DECLINED Block evaluation for TensorGenerator + TensorReverse + fixed bug in tensor reverse op Eugene Zhulenev ezhulenev/eigen-01
727 DECLINED Block evaluation for TensorGenerator + TensorReverse + fixed bug in tensor reverse op Eugene Zhulenev ezhulenev/eigen-01
726 MERGED Block evaluation for TensorChipping + fixed bugs in TensorPadding and TensorSlicing Eugene Zhulenev ezhulenev/eigen-01
719 DECLINED Use std::real and std::imag for real and imaginary parts Sebastian Wolf sebwolf-TUM/eigen
725 MERGED Add block evaluation to TensorEvalTo and fix few small bugs Eugene Zhulenev ezhulenev/eigen-01
724 MERGED Fixing incorrect size in Tensor documentation. Brian Zhao bmzhao/eigen
723 MERGED Add block evaluation to TensorReshaping/TensorCasting/TensorPadding/TensorSelect Eugene Zhulenev ezhulenev/eigen-01
722 MERGED Add beta to TensorContractionKernel and make memset optional Eugene Zhulenev ezhulenev/eigen-01
721 DECLINED Prevent infinite loop in the nvcc compiler while unrolling the recurrent templates for Chebyshev polynomial evaluation. Rasmus Larsen null
706 DECLINED Add constructor for PermutationMatrix that accepts a variable number of arguments David Tellenbach tellenbach/eigen
718 DECLINED Move Implementation of vectorized erf() (error function) and fix various builds that were broken by https://bitbucket.org/eigen/eigen/commits/03b9312b2716bef62645fe85469be71edd0bfcd8 Rasmus Larsen null
717 MERGED Fix cxx11_tensor_block_io test Eugene Zhulenev ezhulenev/eigen-02
716 MERGED Fix erf in c++03 Eugene Zhulenev ezhulenev/eigen-02
715 MERGED Fix compilation warnings and errors with clang in TensorBlockV2 Eugene Zhulenev ezhulenev/eigen-02
709 MERGED Tensor block evaluation V2 support for unary/binary/broadcasting Eugene Zhulenev ezhulenev/eigen-02
704 MERGED Add generic PacketMath implementation of the Error Function (erf). Rasmus Larsen null
714 MERGED Fix for the HIP build+test errors. Deven Desai null
713 MERGED Fix a bug in a packed block type in TensorContractionThreadPool Eugene Zhulenev ezhulenev/eigen-02
712 MERGED Choose TensorBlock StridedLinearCopy type statically Eugene Zhulenev ezhulenev/eigen-02
711 DECLINED Choose TensorBlock StridedLinearCopy type statically Eugene Zhulenev ezhulenev/eigen-02
710 MERGED Add new TensorBlock api implementation + tests Eugene Zhulenev ezhulenev/eigen-02
708 DECLINED Tensor block evaluation V2 initial commit Eugene Zhulenev ezhulenev/eigen-01
124 DECLINED Einstein notation for Tensor module godeffroy valet godeffroy/eigen_
705 MERGED Add support for asynchronous evaluation of tensor casting expressions. Rasmus Larsen rmlarsen/eigen_cuda
703 MERGED Fix for the HIP build+test errors. Deven Desai null
701 MERGED Add additional bessel functions Srinivas Vasudevan srvasude/eigen
700 MERGED Use ThreadLocal container in TensorContractionThreadPool Eugene Zhulenev ezhulenev/eigen-01
699 MERGED Add packetized versions of i0e and i1e special functions Srinivas Vasudevan srvasude/eigen
697 DECLINED Fix for the HIP build+test errors introduced by the ndtri support. Deven Desai null
698 MERGED ThreadLocal container that does not rely on thread local storage Eugene Zhulenev ezhulenev/eigen-01
621 DECLINED Fix documentation of EIGEN_COMP_EMSCRIPTEN David Tellenbach tellenbach/eigen
681 DECLINED Add ndtri function, the inverse of the normal distribution function. Srinivas Vasudevan srvasude/eigen
696 MERGED Change typedefs from private to protected to fix MSVC compilation Eugene Zhulenev ezhulenev/eigen-01
695 MERGED Allow move-only done callback in TensorAsyncDevice Eugene Zhulenev ezhulenev/eigen-01
694 MERGED TensorMap constness should not change underlying storage constness Eugene Zhulenev ezhulenev/eigen-01
657 DECLINED Updated Eigen_Colamd.h, namespacing macros ALIVE & DEAD as COLAMD_ALIVE & COLAMD_DEAD Anshul Jaiswal anshuljl/eigen
688 DECLINED public typedefs for PardisoSupport to match other SparseSolvers typdefs Vincent HUBER null
676 DECLINED Fixed Tensor documentation formatting. Alberto Luaces aluaces/eigen
675 DECLINED Please add space for class DenseCoeffsBase at line 35 Jiacheng_Wu Jiacheng_Wu/eigen
693 MERGED evalSubExprsIfNeededAsync + async TensorContractionThreadPool Eugene Zhulenev ezhulenev/eigen-01
692 MERGED Asynchronous expression evaluation with TensorAsyncDevice Eugene Zhulenev ezhulenev/eigen-01
690 MERGED Const correctness in TensorMap> expressions Eugene Zhulenev ezhulenev/eigen-01
689 MERGED Remove shadow warnings in TensorDeviceThreadPool Eugene Zhulenev ezhulenev/eigen-01
683 MERGED Asynchronous parallelFor in Eigen ThreadPoolDevice Eugene Zhulenev ezhulenev/eigen-01
684 DECLINED Clean up float16 a.k.a. Eigen::half support in Eigen. Rasmus Larsen null
685 MERGED Fix get_random_seed on Native Client Marat Dukhan GooglerMaratDukhan/eigen
687 DECLINED public typedefs for PardisoSupport to match other SparseSolvers typdefs Vincent HUBER null
679 MERGED Fixes for Altivec/VSX and compilation with clang on PowerPC João Paulo Labegalini de Carvalho jaopaulolc/eigen
680 MERGED Implement vectorized versions of log1p and expm1 in Eigen using Kahan's formulas, and change the scalar implementations to properly handle infinite arguments. Rasmus Larsen null
682 MERGED Remove XSMM support from Tensor module Eugene Zhulenev ezhulenev/eigen-01
658 DECLINED Make pmax/pmin behave like std::fmax/std::fmin rather than std::max/std::min Rasmus Larsen null
678 DECLINED Couple of fixes for Altivec/VSX and the compilation with clang on PowerPC João Paulo Labegalini de Carvalho null
677 DECLINED Reinstate vectorized evaluation of coefficient-wise functions over tensor blocks. Rasmus Larsen null
662 MERGED Modifying tesnor module to support SYCL bakcend. Mehdi Goli codeplaysoftware/eigen-upstream-pure
442 SUPERSEDED Enabling Optimised Eigen SYCL backend for the unsupported Tensor module Mehdi Goli codeplaysoftware/eigen
541 OPEN Add structured bindings support Bruno Manganelli bmanga95/eigen
674 DECLINED Added leading asterisk for Doxygen to consume as it was removing asterisk intended to be part of the code. Kyle Vedder kylevedder/eigen
672 DECLINED Fix typo in Umeyama method documentation Michael Grupp MichaelGrupp/eigen
671 MERGED Fix expression evaluation heuristic for TensorSliceOp Eugene Zhulenev ezhulenev/eigen-01
670 MERGED Add outer/inner chipping optimization for chipping dimension specified at runtime Eugene Zhulenev ezhulenev/eigen-01
669 MERGED adding the EIGEN_DEVICE_FUNC attribute to the constCast routine. Deven Desai null
667 MERGED [SYCL] : Mehdi Goli codeplaysoftware/eigen
668 MERGED Allocate non-const scalar buffer for block evaluation with DefaultDevice Eugene Zhulenev ezhulenev/eigen-01
655 DECLINED Fix missing Eigen namespace in Macros Justin Carpentier JustinCarpentier/eigen
666 DECLINED [SYCL] Adding the SYCL memory model. Mehdi Goli codeplaysoftware/eigen
665 MERGED Fix TensorReverse on GPU with m_stride[i]==0 Eugene Zhulenev ezhulenev/eigen-01
660 MERGED fix for ROCm/HIP specific compile error introduced by a recent commit. Deven Desai null
664 MERGED Add block access to TensorReverseOp and make sure that TensorForcedEval uses block access when preferred Eugene Zhulenev ezhulenev/eigen-01
661 DECLINED Required modification in Eigen core to make sycl-backend working. Mehdi Goli codeplaysoftware/eigen-upstream-pure
663 MERGED Optimize evaluation strategy for TensorSlicingOp and TensorChippingOp Eugene Zhulenev ezhulenev/eigen-01
659 MERGED Fix CUDA build on Mac. Rasmus Larsen rmlarsen/eigen_cuda
635 MERGED Speed up GEMV on AVX-512 builds, just as done for GEBP previously. Gustavo Lima Chaves glchaves/eigen
656 DECLINED Add workaround for choosing the right include files with FP16C support with clang. Rasmus Larsen null
654 MERGED Minor build improvements Artem Belevich Artem-B/eigen
652 DECLINED Clean up CUDA/NVCC version macros and their use in Eigen. Rasmus Larsen null
653 DECLINED Minor build improvements. Artem Belevich Artem-B/eigen
651 DECLINED Clean up CUDA/NVCC version macros and their use in Eigen. Rasmus Larsen null
647 DECLINED Revert change of EIGEN_CUDACC_VER definition. It should map to the NVCC version (without requiring that cuda_runtime.h is included first) and map to zero for cuda clang. Christian Sigg csigg/eigen
648 MERGED GEMV: remove double declaration of constant. Gustavo Lima Chaves glchaves/eigen
649 MERGED fix for HIP build errors that were introduced by a commit earlier this week Deven Desai null
643 MERGED Make Eigen build with cuda 10 and clang. Rasmus Larsen null
378 MERGED Add interface to umfpack_*l_* functions Vincent HUBER null
646 MERGED Eigen: Fix MSVC C++17 language standard detection logic Scott Ramsby scramsby/eigen
645 MERGED Prevent potential division by zero in TensorExecutor Eugene Zhulenev ezhulenev/eigen-01
644 MERGED Always evaluate Tensor expressions with broadcasting via tiled evaluation code path Eugene Zhulenev ezhulenev/eigen-01
640 MERGED Fix deadlocks in thread pool. Rasmus Larsen null
641 DECLINED SparseLU.h - corrected example, it didn't compile Dario Figueira dariofigueira/eigen
642 MERGED Removing an unused masked load API introduced in commit-id 1f84827 Anuj Rawat anuj-rawat/eigen
639 MERGED Fix AVX512 & GCC 6.3 compilation Eugene Zhulenev ezhulenev/eigen-01
628 MERGED Adding lowlevel APIs for optimized RHS packet load in TensorFlow SpatialConvolution Anuj Rawat null
638 OPEN Added test for relativity of CMAKEPACKAGE_INSTALL_DIR so that Eigen is relocatable. Wojtek Mamrak wmamrak/eigen
637 MERGED Add masked pstoreu to AVX and AVX512 PacketMath Eugene Zhulenev ezhulenev/eigen-01
636 DECLINED Avoid checking PGI version when not PGI Andy May may/eigen
633 MERGED Check if gpu_assert was overridden in TensorGpuHipCudaDefines Eugene Zhulenev ezhulenev/eigen-01
634 MERGED Fix compilation with PGI version 19 Andy May may/eigen
559 DECLINED gebp: work on having fraction of packet vectorization on the multiple rows attack (before going row-by-row) Gustavo Lima Chaves glchaves/eigen
632 MERGED Fix doxygen warnings Eugene Zhulenev ezhulenev/eigen-01
631 MERGED Get rid of SequentialLinSpacedReturnType deprecation warnings in DenseBase.h Eugene Zhulenev ezhulenev/eigen-01
630 MERGED Add missing EIGEN_DEPRECATED annotations to deprecated functions and fix few other doxygen warnings Eugene Zhulenev ezhulenev/eigen-01
629 MERGED Use packet ops instead of AVX2 intrinsics Eugene Zhulenev ezhulenev/eigen-01
627 DECLINED Split the implementation of i?amax into two. Sameer Agarwal null
626 OPEN Add natvis to VS projects when importing Eigen target using CMake João Maia jpmmaia/eigen
625 MERGED Update TheadPoolDevice example. jonathon_k jonathon_k/eigen
624 MERGED adding EIGEN_DEVICE_FUNC to the recently added TensorContractionKernel constructor. Deven Desai null
623 MERGED Add missing semicolon Eugene Zhulenev ezhulenev/eigen-01
622 MERGED Add support for custom packed Lhs/Rhs blocks in tensor contractions Eugene Zhulenev ezhulenev/eigen-01
619 DECLINED Add support for pcmp_eq in AltiVec/Complex.h William Irons wdirons/eigen
620 DECLINED Disable -fno-unsafe-math-optimizations for ICC David Tellenbach tellenbach/eigen
546 MERGED ROCm/HIP specfic fixes + updates Deven Desai null
618 MERGED Make clipping outside [-18:18] consistent for vectorized and non-vectorized paths of scalar_logistic_op. Rasmus Larsen null
617 MERGED Fix include guard comments for Skyline module David Tellenbach tellenbach/eigen
616 DECLINED Remove undefined std::complex David Tellenbach tellenbach/eigen
593 DECLINED Initializer list + variadic ctor for DiagonalMatrix David Tellenbach tellenbach/eigen
615 MERGED Clean up PacketMathHalf.h and add a few missing logical packet ops. Rasmus Larsen null
612 DECLINED Add misc. missing packet ops for half float types in GPU. Rasmus Larsen null
614 MERGED Fix multiple bugs in unsupported/Tensor module Eugene Zhulenev ezhulenev/eigen-01
611 MERGED Remove redundant steal loop Eugene Zhulenev ezhulenev/eigen-01
606 MERGED Remove EIGEN_MPL2_ONLY guards around code re-licensed from LGPL to MPL2 in https://bitbucket.org/eigen/eigen/commits/db27a223a0d2ddf857e1232a5365292db5b55ef3 Rasmus Larsen null
610 MERGED Block evaluation for TensorGeneratorOp Eugene Zhulenev ezhulenev/eigen-01
609 MERGED Tune tensor contraction threadpool heuristics Eugene Zhulenev ezhulenev/eigen-01
608 MERGED Add an extra check for the RunQueue size estimate Eugene Zhulenev ezhulenev/eigen-01
607 MERGED Do not create Tensor in cxx11_tensor_forced_eval test Eugene Zhulenev ezhulenev/eigen-01
604 MERGED Add tiled evaluation for TensorForcedEvalOp Eugene Zhulenev ezhulenev/eigen-01
605 MERGED Do not initialize invalid fast_strides in TensorGeneratorOp Eugene Zhulenev ezhulenev/eigen-01
603 MERGED Use fast divisors in TensorGeneratorOp Eugene Zhulenev ezhulenev/eigen-01
598 DECLINED Do not keep latex logs Bernhard M. Wiedemann bmwiedemann/eigen
602 DECLINED Fully qualify Eigen::internal::aligned_free Sam Hasinoff hasinoff/eigen
601 DECLINED [Pardiso] fix compilation error as reported in bug 1629 Jason Juang jasjuang/eigen
595 DECLINED Add a general QuaternionNumber g.gael@free.fr ggael/eigen-quat
599 MERGED Consistently use EIGEN_BLAS_FUNC in BLAS. Sameer Agarwal null
596 MERGED Improve EventCount used by the non-blocking threadpool. Rasmus Larsen null
597 MERGED Change licensing of OrderingMethods/Amd.h and SparseCholesky/SimplicialCholesky_impl.h from LGPL to MPL2. Rasmus Larsen null
587 DECLINED PartialPivLU.h Fixed-size specializations of computing the decomposition for 2x2-, 3x3- and 4x4-matrices Antti Ylitepsa DogeMajor/eigen
180 MERGED alias template for matrix and array classes Heiko Bauke ra_bauke/eigen
576 DECLINED Fix a typo that contraction Op dimension should be leftRank + rightRank - 2*ContractionDim 张浩 hzhangxyz/eigen
591 MERGED Fix signed-unsigned comparison warning in RunQueue Eugene Zhulenev ezhulenev/eigen-01
590 MERGED Do not generate no-op cast() and conjugate() expressions Eugene Zhulenev ezhulenev/eigen-01
589 MERGED Speedup Tensor ThreadPool RunQueu::Empty() Eugene Zhulenev ezhulenev/eigen-01
588 OPEN lu.cpp New tests to make sure that the (faster) fixed-sized-specializations for 2x2-, 3x3- and 4x4-matrices work Antti Ylitepsa DogeMajor/eigen-1
586 MERGED Add PacketConv implementation for non-vectorizable src expressions Eugene Zhulenev ezhulenev/eigen-01
585 MERGED Optimize TensorConversion evaluator: do not convert same type Eugene Zhulenev ezhulenev/eigen-01
584 DECLINED Optimize TensorConversion evaluator: do not convert same type Eugene Zhulenev ezhulenev/eigen-01
363 DECLINED Adding fallback for isfinite, isinf, isnan; fix for conversion from 0 to scalar in RealSchur Dmitriy Korchemkin dkorchemkin/eigen
583 DECLINED Spline.h: fix spelling "spang" -> "span" Steve Peters null
582 MERGED Don't do parallel_pack if we can use thread_local memory in tensor contractions Eugene Zhulenev ezhulenev/eigen-01
578 MERGED Speed up Eigen matrix*vector and vector*matrix multiplication. Rasmus Larsen null
581 MERGED Parallelize tensor contraction only by sharding dimension and use 'thread-local' memory for packing Eugene Zhulenev ezhulenev/eigen-01
579 MERGED Speed up row-major matrix-vector product on ARM Sameer Agarwal null
580 MERGED Fix GeneralBlockPanelKernel Android compilation Eugene Zhulenev ezhulenev/eigen-01
577 DECLINED Speed up Eigen matrix*vector and vector*matrix multiplication Rasmus Larsen null
574 DECLINED Replacement of Matrix/Array/PlainObjectBase Ctor that takes initializer_list with variadic template ctor David Tellenbach tellenbach/eigen
575 OPEN Add cmake option to turn off parallelization Brad Kotsopoulos bkotzz/eigen
572 DECLINED Initializer list construction David Tellenbach tellenbach/eigen
573 DECLINED Replace host_define.h with cuda_runtime_api.h Nathan Luehr ncluehr/eigen
397 DECLINED Hyperlink DOIs against preferred resolver kalei null
567 DECLINED Restructuring of all DenseSolvers Patrick Peltzer patrick_peltzer/eigen
570 MERGED Add support for inverse hyperbolic functions. Fix cost of division. Rasmus Larsen null
566 DECLINED Remove reinterpret_cast from AVX512 complex implementation Mark Ryan markdryan/eigen
571 DECLINED Faster sin/cos for large inputs. g.gael@free.fr ggael/eigen3
569 DECLINED Fix warnings in ptrue for complex and half types. Rasmus Larsen null
563 MERGED Add support for pcmp_eq and pnot, including for complex types. Rasmus Larsen null
568 MERGED Fix shorten-64-to-32 warning in TensorContractionThreadPool Eugene Zhulenev ezhulenev/eigen-01
565 MERGED Fix shorten-64-to-32 warning in TensorContractionThreadPool Eugene Zhulenev ezhulenev/eigen-01
564 MERGED Optimize evalShardedByInnerDim Eugene Zhulenev ezhulenev/eigen-01
558 DECLINED Add support and tests for pcmp_eq and pnot. Rasmus Larsen null
562 MERGED Explicitly set fill character when printing aligned data to ostream Eugene Zhulenev ezhulenev/eigen-01
561 DECLINED Explicitly set fill character when printing aligned data to ostream Eugene Zhulenev ezhulenev/eigen-01
560 DECLINED Fix the AVX512f only builds Mark Ryan markdryan/eigen
557 DECLINED Add support and tests for pcmp_eq and pnot. Rasmus Larsen null
555 DECLINED Adding NEON specific Quaternion multiplication. Here are the benchmarks. This is on Pixel 3. Overall we see around 15% improvement for floats, but only 5% improvements for doubles. Ritesh Gupta gupta_ritesh/eigen
543 MERGED Add parallel memcpy to TensorThreadPoolDevice in Eigen, but limit the number of threads to 4, beyond which we just seem to be wasting CPU cycles as the threads contend for memory bandwidth. Rasmus Larsen null
553 MERGED Do not disable alignment with EIGEN_GPUCC Eugene Zhulenev ezhulenev/eigen-01
554 MERGED Fix tensor contraction on AVX512 builds Mark Ryan markdryan/eigen
443 DECLINED Fix tensor contraction for AVX512 machines Mark Ryan markdryan/eigen
552 OPEN Add FindEigen3 support for policy CMP0074 Frederic Simonis frederic_simonis/eigen-1
549 MERGED fix the build on 64-bit ARM when NEON is disabled Benoit Jacob bjacob/eigen
551 MERGED Use explicit packet type in SSE/PacketMath pldexp Eugene Zhulenev ezhulenev/eigen-01
550 DECLINED Neon fixes. Note: 2 separate changesets, fixing 2 separate things. Benoit Jacob bjacob/eigen
538 DECLINED Vectorize row-by-row gebp loop iterations on 16 packets as well Gustavo Lima Chaves glchaves/eigen
548 DECLINED Fix typos Patrik Huber patrikhuber/eigen-1
547 DECLINED Fix small typo in TutorialSlicingIndexing.dox Patrik Huber patrikhuber/eigen-1
354 DECLINED fix aligment issue with EIGEN_DEFAULT_ALIGN_BYTES value greater than 16 Justin Carpentier carpent/eigen
544 DECLINED Set requestedAlignment correctly for SliceVectorizedTraversals Mark Ryan markdryan/eigen
545 DECLINED Fix typo in comment on EIGEN_MAX_STATIC_ALIGN_BYTES Nikolaus Demmel NikolausDemmel/eigen
542 MERGED Add optimized version of logistic function for float. Rasmus Larsen null
364 DECLINED QRKit: Sparse, Composable QR Decompositions for Efficient and Stable Solutions to Problems in Computer Vision Jan Svoboda jasvob/eigen_sparse_qr
536 OPEN Allow C++17 structured bindings on vectors Bruno Manganelli bmanga95/eigen
540 MERGED A few small fixes to a) prevent throwing in ctors and dtors of the threading code, and b) supporting matrix exponential on platforms with 113 bits of mantissa for long doubles. Rasmus Larsen null
539 MERGED Vectorize row-by-row gebp loop iterations on 16 packets as well Gustavo Lima Chaves glchaves/eigen
527 DECLINED run{all}.sh: be able to skip plotting data Gustavo Lima Chaves glchaves/eigen
526 DECLINED Speed up multiplication of small, dynamically sized matrices Mark Ryan markdryan/eigen
484 OPEN cmake: Support configuring Eigen without the blas/lapack convenience targets Jean-Christophe Fillion-Robin jcfr/eigen
537 MERGED Workaround nvcc+msvc compiler bug Eugene Zhulenev ezhulenev/eigen-01
535 DECLINED Don't store the reference to a temporary value David Wagner easymile/eigen
281 DECLINED Improve FindEigen3.cmake Silvio Traversaro traversaro/eigen
282 DECLINED Add cmake config file for source dir Silvio Traversaro traversaro/eigen
534 MERGED Fix cxx11_tensor_{block_access, reduction} tests Eugene Zhulenev ezhulenev/eigen-02
533 DECLINED Tutorial typo fix Halie Murray-Davis hmurraydavis/eigen
325 DECLINED Support compiling with STLport without IO streams Christian von Schultz vonschultz/eigen
532 MERGED Only set EIGEN_CONSTEXPR_ARE_DEVICE_FUNC for clang++ if cxx_relaxed_constexpr is available. Rasmus Larsen null
502 OPEN Improve CUDA matrix inversion Andrea Bocci fwyzard/eigen
531 DECLINED Only set EIGEN_CONSTEXPR_ARE_DEVICE_FUNC for clang++ if cxx_relaxed_constexpr is available. Rasmus Larsen null
530 MERGED Move from rvalue arguments in ThreadPool enqueue* methods Eugene Zhulenev ezhulenev/eigen-02
529 MERGED Reduce thread scheduling overhead in parallelFor Eugene Zhulenev ezhulenev/eigen-02
528 MERGED [TensorBlockIO] Check if it's allowed to squeeze inner dimensions Eugene Zhulenev ezhulenev/eigen-02
520 DECLINED Speed up multiplication of small matrices on x86-64 Mark Ryan markdryan/eigen
413 DECLINED [Bug 1558] Fix MINRES edge case David Hyde DABH/eigen
525 MERGED Fix bug in partial reduction of expressions requiring evaluation Eugene Zhulenev ezhulenev/eigen-02
524 MERGED Use void type if stl-style iterators are not supported Eugene Zhulenev ezhulenev/eigen-01
523 MERGED Compile time detection for unimplemented stl-style iterators Eugene Zhulenev ezhulenev/eigen-01
522 DECLINED Fix bug in TensorArgMax for return_dim == -1. Rasmus Larsen null
521 DECLINED Fix lazy_gemm_settings.txt Mark Ryan markdryan/eigen
519 MERGED Bug 231: implementation of STL iterators for dense expressions g.gael@free.fr ggael/eigen-iterators
245 DECLINED Expose publicly VectorwiseOp subvector access Federico Allocati fallocati/eigen
518 MERGED PR with HIP specific fixes (for the eigen nightly regression failures in HIP mode) Deven Desai null
517 MERGED Get rid of unused variable warning. Rasmus Larsen null
508 DECLINED Provide EIGEN_OVERRIDE and EIGEN_FINAL macros to mark virtual function overrides Christoph Hertzberg chtz/eigen-override
516 MERGED Fix bug in copy optimization in Tensor slicing. Eugene Zhulenev ezhulenev/eigen-01
515 MERGED Fix a few warnings and rename a variable to not shadow "last". Rasmus Larsen null
514 MERGED Add tests for evalShardedByInnerDim contraction + fix bugs Eugene Zhulenev ezhulenev/eigen-01
512 DECLINED Fix test/packetmath.cpp for AVX512 builds. MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Gustavo Lima Chaves glchaves/eigen
509 MERGED Support multiple contraction kernel types in TensorContraction Eugene Zhulenev ezhulenev/eigen-01
513 MERGED Optimize TensorBlockCopyOp Eugene Zhulenev ezhulenev/eigen-01
510 DECLINED Fix test/packetmath.cpp for AVX512 builds. Gustavo Lima Chaves glchaves/eigen
511 MERGED Parallelize tensor contraction over the inner dimension. Rasmus Larsen null
301 MERGED Add DenseBase::reshaped() and operator()(all) g.gael@free.fr ggael/eigen-reshape
41 MERGED reshape yoco xiao yoco/eigen
506 MERGED 8096f5ec0ff3 g.gael@free.fr ggael/eigen3
474 DECLINED Add support for partitioning a single Eigen::ThreadPool into multiple virtual pool(s). Ravi Kiran seeravikiran/eigen_partitions
507 DECLINED Split cxx11_tensor_executor test Eugene Zhulenev ezhulenev/eigen-02
496 DECLINED Cleanup in placeholders g.gael@free.fr ggael/eigen3
505 MERGED Initialize BlockIteratorState in a C++03 compatible way. Rasmus Larsen null
504 MERGED Silence compiler warning. Rasmus Larsen null
503 MERGED Silence compiler warnings. Rasmus Larsen null
500 DECLINED Casting IndexList to DSize explicitly, as the DSizes constructor is required to be explicit. Mehdi Goli codeplaysoftware/eigen-upstream-pure
501 MERGED Enable DSizes type promotion with c++03 Eugene Zhulenev ezhulenev/eigen-02
498 MERGED Add DSizes index type promotion Eugene Zhulenev ezhulenev/eigen-02
497 MERGED Fix warnings in IndexList array_prod Eugene Zhulenev ezhulenev/eigen-01
444 MERGED Tiled evaluation for Tensor ops Eugene Zhulenev ezhulenev/eigen
490 MERGED Let MaxSizeVector respect alignment of objects Christoph Hertzberg chtz/eigen-maxsizevector
499 DECLINED fix typo for Tensor README lijiansong lijiansong/eigen
495 DECLINED Fix IndexList array_prod Eugene Zhulenev ezhulenev/eigen
494 MERGED Fix DSizes IndexList constructor Eugene Zhulenev ezhulenev/eigen
493 MERGED Const cast scalar pointer in TensorSlicingOp evaluator Eugene Zhulenev ezhulenev/eigen
492 MERGED Explicitly construct tensor block dimensions from evaluator dimensions Eugene Zhulenev ezhulenev/eigen
491 MERGED Support reshaping with static shapes and dimensions conversion in tensor broadcasting Eugene Zhulenev ezhulenev/eigen
480 MERGED Avoid compilation error in C++11 test when EIGEN_AVOID_STL_ARRAY is set. Rasmus Larsen null
486 MERGED Use numerically stable tree reduction in TensorReduction. Rasmus Larsen null
489 DECLINED MSVC specific C++11 fixes Jiandong Ruan jd_ruan/eigen
477 DECLINED Updates to fix HIP-clang specific compile errors. Deven Desai null
478 DECLINED Use padding instead of alignment attribute, which MaxSizeVector does not respect. Rasmus Larsen null
487 DECLINED Remove const from ScalarBinaryOpTraits::ResultType Jiandong Ruan jd_ruan/eigen
485 DECLINED tensorflow build break caused by the new block feature Jiandong Ruan null
483 OPEN cmake: Support disabling test when including eigen as sub-project Jean-Christophe Fillion-Robin jcfr/eigen
290 OPEN Add CMake option 'EIGEN_DISABLE_CXX11_MATH' Ruslan Baratov ruslo/eigen
481 MERGED Bug 1526 - CUDA compilation fails on CUDA 9.x SDK when arch is set to compute_60 and/or above Jiandong Ruan null
479 DECLINED Make param name and docs constistent for JacobiRotation::makeGivens Christopher Suter cgsuter/eigen
357 DECLINED Documentation for thrust::complex Henri Menke null
468 MERGED Add support for emulating thread local. Rasmus Larsen null
476 DECLINED adding EIGEN_DEVICE_FUNC attribute to fix some GPU unit tests that are broken in HIP mode Deven Desai null
475 DECLINED Separating SYCL math function. Mehdi Goli codeplaysoftware/eigen-upstream-pure
467 OPEN HermitianMatrix David Tellenbach tellenbach/eigen
465 DECLINED Fix issue in RowMajor assignment in plain_matrix_type_row_major::type Justin Carpentier carpent/eigen
461 MERGED Creating a pointer type in TensorCustomOp.h Mehdi Goli codeplaysoftware/eigen-upstream-pure
471 DECLINED replaced alias templates as alias template are not allowed on typedef. Mehdi Goli codeplaysoftware/eigen-upstream-pure
472 MERGED Removed an used variable (PacketSize) from TensorExecutor Sameer Agarwal null
470 MERGED Fix a few compiler warnings in CXX11 tests. Rasmus Larsen null
469 MERGED Fix cholmod build when SuiteSparse_long != long Rasmus Larsen null
466 MERGED Move sigmoid functor to core and rename it to 'logistic'. Rasmus Larsen null
456 DECLINED Move sigmoid functor to core. Rasmus Larsen null
464 MERGED Cast to diagonalSize to RealScalar instead Scalar. Rasmus Larsen null
463 MERGED Cast diagonalSize() to Scalar before multiplication. Rasmus Larsen null
462 MERGED Fixing compiler warning in TensorBlock.h as it was creating a lot of noise at compilation. Mehdi Goli codeplaysoftware/eigen-upstream-pure
459 MERGED Disabling assert inside SYCL kernel. Mehdi Goli codeplaysoftware/eigen-upstream-pure
458 MERGED Fix init order. Rasmus Larsen null
460 MERGED Adding EIGEN_UNROLL_LOOP macro. Mehdi Goli codeplaysoftware/eigen-upstream-pure
457 MERGED Forward declare NoOpOutputKernel as struct rather than class to be consistent with implementation. Rasmus Larsen null
450 DECLINED Fix cuda clang build. Rasmus Larsen null
455 MERGED Fix initialization order. Rasmus Larsen null
454 MERGED SYCL required changes Mehdi Goli codeplaysoftware/eigen-upstream-pure
453 MERGED Fix 256 bit packet size assumptions in unit tests. Gustavo Lima Chaves glchaves/eigen
451 MERGED Fixing ambigous constructor error for Clang compiler. Mehdi Goli codeplaysoftware/eigen-upstream-pure
452 MERGED Fixing compilation error for cxx11_tensor_trace.cpp on Microsoft Visual Studio. Mehdi Goli codeplaysoftware/eigen-upstream-pure
431 MERGED Optional ThreadPoolDevice allocator Paul Tucker paultucker/eigen
447 MERGED Adding variadic version of assert which can take a parameter pack as its input. Mehdi Goli codeplaysoftware/eigen-upstream-pure
446 MERGED Distinguishing between internal memory allocation/deallocation from explicit user memory allocation/deallocation. Mehdi Goli codeplaysoftware/eigen-upstream-pure
448 MERGED Adding new arch/SYCL headers, used for SYCL vectorization. Mehdi Goli codeplaysoftware/eigen-upstream-pure
449 MERGED Enabling per device specialisation of packetSize. Mehdi Goli codeplaysoftware/eigen-upstream-pure
445 MERGED Replacing ad-hoc inline keyword with EIGEN_STRONG_INLINE MACRO. Mehdi Goli codeplaysoftware/eigen-upstream-pure
370 MERGED Use device's allocate function instead of internal::aligned_malloc. yuefengz yuefengz/eigen
437 DECLINED Force unaligned load in copy mode when Broadcasting. Rasmus Larsen null
438 MERGED Tiled tensor executor Eugene Zhulenev ezhulenev/eigen
440 DECLINED Fix two small typos in the documentation Patrik Huber patrikhuber/eigen
441 MERGED Reduce the number of template specializations of classes related to tensor contraction to reduce binary size. Rasmus Larsen null
412 DECLINED Fix AVX512 implementations of psqrt Mark Ryan markdryan/eigen
439 DECLINED Add pcast packet op for NEON. Rasmus Larsen null
435 DECLINED TensorBlockIO: reading and writing tensor blocks Eugene Zhulenev ezhulenev/eigen
376 DECLINED Allow is_convertible_impl to apply to objects that are not default constructible jasonmercer jasonmercer/eigen
436 DECLINED Account for missing change on commit "Remove SimpleThreadPool and..." Gustavo Lima Chaves glchaves/eigen
434 DECLINED Iterate over Tensor memory using TensorBlocks. Eugene Zhulenev ezhulenev/eigen
433 MERGED Move cxx11_tensor_uint128 test under an EIGEN_TEST_CXX11 guarded block Gustavo Lima Chaves glchaves/eigen
423 DECLINED Reduce number of allocations in TensorContractionThreadPool. Eugene Zhulenev ezhulenev/eigen
432 DECLINED Disable type traits for stdlibc++ <= 4.9.3 Eugene Zhulenev ezhulenev/eigen
430 DECLINED Convert packet count to the reducer type in MeanReducer Eugene Zhulenev ezhulenev/eigen
428 DECLINED Assert that no output kernel is defined for GPU contraction Eugene Zhulenev ezhulenev/eigen
429 DECLINED Disable type traits for GCC < 5.1.0 Eugene Zhulenev ezhulenev/eigen
427 DECLINED Specify default output kernel for TensorContractionOp Eugene Zhulenev ezhulenev/eigen
417 DECLINED Fuse computations into the Tensor contractions using output kernel Eugene Zhulenev ezhulenev/eigen
426 DECLINED Fuse computations into the Tensor contractions using output kernel Eugene Zhulenev ezhulenev/eigen
425 MERGED applying EIGEN_DECLARE_TEST to *gpu unit tests Deven Desai null
424 DECLINED Add optional Allocator argument to ThreadPoolDevice constructor. Paul Tucker paultucker/eigen
422 MERGED Optimize the case where broadcasting is a no-op. Rasmus Larsen null
421 MERGED Clang produces incorrect Thumb2 assembler when using alloca. Rasmus Larsen null
391 DECLINED Implement fast approximate exponential using the "classic" algorithm in https://goo.gl/aUVTK3. Rasmus Larsen null
383 DECLINED specialize scalar_opposite_op functor for CUDA on clang Nicolò Valigi null
414 DECLINED Optimize sparse Q generation and multiplication jetrull jetrull/eigen-sparseq-speedup
402 MERGED Adding support for using Eigen in HIP kernels. Deven Desai null
306 DECLINED Add bitwise and, or, and xor binary functors. ringwalt ringwalt/eigen
420 DECLINED Fix cxx11_tensor_fft not building on Windows. Thales Sabino tluisrs/eigen
398 DECLINED Some documentation fixes in unsupported/NonLinearOptimization Max Kontak null
366 DECLINED Add missing consts for rows and cols functions in SparseLU Julian Kent jkflying/eigen-splu-const
419 DECLINED FindEigen3: Set Eigen3_FOUND variable Daniele E. Domenichelli drdanz/eigen
418 DECLINED Remove double semicolon in PacketMath.h Tobias Klauser tklauser/eigen
416 DECLINED Fuse computations into the Tensor contractions using output kernel Eugene Zhulenev ezhulenev/eigen
415 DECLINED Fuse computations into the Tensor contractions using output kernel Eugene Zhulenev ezhulenev/eigen
407 DECLINED Fix the Packet16h version of ptranspose Mark Ryan markdryan/eigen
408 DECLINED Improve CUDA compatibility in eigen Andrea Bocci null
406 OPEN Removing (most) memory allocations in preallocated decompositions Escande NA Escande/eigen
411 MERGED Use std::complex constructor instead of assignment from scalar Jonathan Liu net147/eigen
410 MERGED Fix typo in pblend for AltiVec. Rasmus Larsen null
409 MERGED Fix oversharding bug in parallelFor. Rasmus Larsen null
389 DECLINED Add a fast path for the trivial case in tensor broadcasting. Rasmus Larsen null
405 DECLINED Improve support for using EIgen inside CUDA kernels Andrea Bocci null
403 MERGED Derivative of the incomplete Gamma function and the sample of a Gamma random variable Michael Figurnov mfigurnov/eigen
374 DECLINED Adding EIGEN_DEVICE_FUNC to Products, especially Dense2Dense Assignment Robert Lukierski lukier/eigen
371 DECLINED Add EIGEN_DEVICE_FUNC to a bunch of functions. Ilya Biryukov ilya-biryukov/eigen
404 MERGED Fix compilation of special functions without C99 math. Michael Figurnov mfigurnov/eigen
396 DECLINED Set EIGEN_IDEAL_MAX_ALIGN_BYTES correctly for AVX512 builds Mark Ryan markdryan/eigen
400 MERGED Exponentially scaled modified Bessel functions of order zero and one. Michael Figurnov mfigurnov/eigen
401 MERGED Add a ThreadPoolInterface* getter for ThreadPoolDevice. Penporn Koanantakool penpornk/eigen
390 DECLINED cmake: only run hg if we are in hg repository Christoph Junghans junghans/eigen
399 DECLINED Exponentially scaled modified Bessel functions of order zero and one. Michael Figurnov mfigurnov/eigen
395 OPEN Cholmod error handling René Wagner renewagner/eigen
394 OPEN Cholmodsupport self contained header René Wagner renewagner/eigen
367 DECLINED Make sparse QR result sizes consistent with dense QR jetrull jetrull/eigen
386 DECLINED Bugfix/sparseq complex jetrull jetrull/eigen
393 MERGED Rename scalar_clip_op to scalar_clamp_op to prevent collision with existing functor in TensorFlow. Rasmus Larsen null
360 MERGED Fix bugs and typos in the contraction example of the tensor README Didier Jansen didierjansen/eigen
392 MERGED Add vectorized clip functor for Eigen Tensors Rasmus Larsen null
387 DECLINED Default test please ignore Jan Svoboda jasvob/eigen_sparse_qr_extra
385 MERGED fix AVX512 plog Jayaram Bobba jbobba/eigen
368 DECLINED Fixed some spaces in Core header files Tanmay Rustagi rtanmay/eigen
313 DECLINED add helper functions for NEON Koan-Sin Tan koansin/eigen
362 DECLINED Fix warnings Ben Boeckel mathstuf/eigen
384 DECLINED Cast zeros to Scalar in RealSchur Dmitriy Korchemkin dkorchemkin/eigen_patches
314 DECLINED Remove warnings from TensorConvolution.h compile Hugh Perkins hughperkins/eigen
382 DECLINED Add a specialization of Eigen::numext::conj for std::complex to be used when compiling a cuda kernel. This fixes the compilation of TensorFlow 1.4 with clang 6.0 used as CUDA compiler with libc++. Nicolò Valigi null
381 OPEN Specialize is_arithmetic and is_integral for long long Nicolò Valigi null
322 DECLINED Locally supress some GCC warnings Diazonium Diazonium/eigen
326 DECLINED Suppressing a Division by Zero in TriangularSolverVector.h Mmanu Chaturvedi m-chaturvedi/eigen
365 DECLINED Bug1493: Fix complex HouseholderSequence representation Christoph Hertzberg null
380 DECLINED removed multiple definitions issue Geoffrey Viola geoffviola/eigen
379 MERGED Fix code sample output in block(int, int, int, int) doxygen Guillaume Jacob easymile_gjacob/eigen
377 DECLINED scripts: add MKL install tool via APT, YUM, or Conda repositories Maria Kraynyuk mkraynyu/eigen-mkl
361 DECLINED Fix for AVX512 Md Faijul Amin mdfaijul/eigen
345 MERGED Adds missing EIGEN_STRONG_INLINE to support MSVC properly inlining small vector calculations Basil Fierz bfierz/eigen
359 MERGED ENH: exp supports complex type for cuda Yan Facai (颜发才) facaiy/eigen
346 MERGED a small mistake QuickReference.dox edited online with Bitbucket zhouzhaoping zzp11/eigen
369 MERGED Move up the specialization of std::numeric_limits Daniel Trebbien null
375 MERGED Improve robustness of igamma and igammac to bad inputs. Rasmus Larsen null
373 DECLINED Added eigen_assert() to MatrixBase::diagonal(Index) MAYANK AGARWAL mayank1495/eigen
355 DECLINED Fix dense Q application in the complex case with judicious use of adjoint()/conjugate() jetrull jetrull/eigen
352 MERGED Update the padding computation for PADDING_SAME to be consistent with TensorFlow. Yangzihao Wang yangzihao/eigen
358 DECLINED Fix incorrect ldvt in LAPACKE call from JacobiSVD Eugene Chereshnev echeresh/eigen
356 MERGED Disable use of recurrence for computing twiddle factors. RJ Ryan rryan/eigen
353 DECLINED For cuda 9.1 replace math_functions.hpp with cuda_runtime.h Nathan Luehr ncluehr/eigen
350 DECLINED Remove forceinline option for EIGEN_STRONG_INLINE. Gunhan Gulsoy gunan/eigen
351 DECLINED Win nvcc Gunhan Gulsoy gunan/eigen
349 MERGED Replace __float2half_rn with __float2half Nathan Luehr ncluehr/eigen
348 MERGED Fix incorrect integer cast in half2 predux. Nathan Luehr ncluehr/eigen
347 DECLINED Disable gcc-specific workaround for Clang to allow build with AVX512 Zvi Rackover zvirack/eigen
317 DECLINED Make buildtests.in more robust Robin Richtsfeld Androbin42/eigen-1
316 DECLINED Make eigen_monitor_perf.sh more robust Robin Richtsfeld Androbin42/eigen
324 MERGED In the CMake find module, define the Eigen imported target as when installing with CMake Alain Vaucher alainvaucher/eigen
337 MERGED Add support for SuiteSparse's KLU routines Kyle Macfarlan kylemacfarlan/eigen
305 MERGED Issue with mpreal and std::numeric_limits::digits Simon Praetorius spraetor/eigen
302 MERGED Make scaling work with non-square matrices Julian Kent null
342 MERGED Use col method for column-major matrix Justin Carpentier carpent/eigen
329 MERGED Added an example for a contraction to a scalar value to README.md Jonas Harsch JonasMu/eigen
334 MERGED - Add support for NEON plog PacketMath function Guillaume Michel guillaume_michel/eigen
335 MERGED Typo fix alignmeent ->alignment LaFeuille LaFeuille/eigen-1
344 MERGED Branch 3.3: Fixing missing inlines on device functions for newer CUDA cards Henry Schreiner henryiii/eigen
343 MERGED Fixing missing inlines on device functions for newer CUDA cards Henry Schreiner henryiii/eigen
341 MERGED Default Benoit Steiner benoitsteiner/opencl
330 DECLINED SparseQR improved efficiency of evaluation of product of Househoulder vectors. Jan Svoboda jasvob/eigen
336 DECLINED signature_of_eigen3_matrix_library edited online with Bitbucket Creative Learner shahzad123/eigen
339 MERGED Specialize ThreadPoolDevice::enqueueNotification for the case with no args. Rasmus Larsen null
338 MERGED Full Support for s390x z14 arch, in particular proper vector float 32-bit Konstantinos Margaritis kmargar/eigen
321 OPEN Changed the install include directory exported. The current implementation (with full path) leads to a machine specific path being printed in the Target.cmake file. The new implementation let cmake compute the path to be relocatable. Pierre Venet XB32ZPV/eigen
333 OPEN CBLAS API Ryan Richard ryan_richard/eigen
332 MERGED Avoid undefined behavior in Eigen::TensorCostModel::numThreads. Rasmus Larsen null
331 MERGED Fixed compilation with cuda_clang. Ilya Biryukov ilya-biryukov/eigen
328 MERGED Bug 1464 : Fixes construction of EulerAngles from 3D vector expression. Abhijit Kundu infinitei/eigen
327 MERGED CUDA 9.0 support Nathan Luehr ncluehr/eigen
312 MERGED Work around a compilation error seen with nvcc V8.0.61 Daniel Trebbien null
288 DECLINED LLT: Solve in place fixes Jim Radford jimradford/eigen
323 MERGED Improved support for OpenCL Benoit Steiner benoitsteiner/opencl
315 MERGED Add labels to #ifdef, in TensorReductionCuda.h Hugh Perkins hughperkins/eigen
319 MERGED Tensor Trace op Gagan Goel tntnatbry/eigen
285 DECLINED Align allocations made in TensorDeviceSycl.h Ted Ying yingted/eigen
318 MERGED Improved support for OpenCL Benoit Steiner benoitsteiner/opencl
240 OPEN OpenCL 1.2 on Mac Hugh Perkins hughperkins/eigen
308 DECLINED Auto diff scalar bug Mmanu Chaturvedi m-chaturvedi/eigen
311 MERGED Eigen compilation with nvcc Abhijit Kundu infinitei/eigen
310 MERGED Use scalar_sum_op and scalar_quotient_op instead of operator+ and operator/ in MeanReducer. RJ Ryan rryan/eigen
309 MERGED OpenCL improvements Benoit Steiner benoitsteiner/opencl
307 MERGED fix typos in the Tensor readme Gagan Goel tntnatbry/eigen
304 MERGED Fixed compilation with cuda-clang Ilya Biryukov ilya-biryukov/eigen
303 MERGED Make the non-blocking threadpool more flexible and less wasteful of CPU cycles for high-latency use-cases. Rasmus Larsen null
300 OPEN DualQuaternion initial implementation Andrew Hundt ahundt/eigen
298 MERGED Improved support for OpenCL Benoit Steiner benoitsteiner/opencl
297 DECLINED Fix vector indexing with int64_t and uint64_t indexes Jonathan Hseu jhseu/eigen
295 MERGED Revert PR #292. Rasmus Larsen null
294 MERGED generalized operator() for indexed access and slicing g.gael@free.fr ggael/eigen-flexidexing
292 MERGED Adds a fast memcpy function to Eigen. Rasmus Larsen null
293 MERGED Reverse arguments for pmin in AVX. Rasmus Larsen null
291 MERGED Make NaN propagation consistent between the pmax/pmin and std::max/std::min Rasmus Larsen null
289 MERGED Fix a typo LaFeuille LaFeuille/eigen
287 DECLINED add cmake-option to enable/disable creation of tests Nero Burner neroburner/eigen
284 DECLINED Proposed patch for bug #1340 vroussellet vroussellet/eigen
286 DECLINED doc: Fix trivial typos in AsciiQuickReference.txt Marco Falke null
283 DECLINED Fix not being able to have binary operations among dense and sparse matrix expressions. Ng Kelvin Kelvin-Ng/eigen
257 MERGED CMake imported target (take #2) Sergiu Deitsch sergiu/eigen
280 DECLINED Move ENV var from HINTS to PATHS in find_path Silvio Traversaro traversaro/eigen-3
279 MERGED Fix for auto appearing in functor template argument. Benoit Steiner benoitsteiner/opencl
276 MERGED Minor improvements to Umfpack support João Rui Leal joaoruileal/eigen
278 MERGED Default Benoit Steiner benoitsteiner/opencl
269 MERGED Remove superfluous const's (can cause warnings on some Intel compilers) Angelos Mantzaflaris angelos_m/eigen
277 DECLINED Add support for long indexes and for (real-valued) row-major matrices to CholmodSupport module Christoph Hertzberg null
274 DECLINED Transformation methods added to ParametrizedLine class. kamre kamre/eigen
273 DECLINED NumTraits.h: Rafael Guglielmetti rgugliel/eigen
275 MERGED Improved support for OpenCL Benoit Steiner benoitsteiner/opencl
270 DECLINED Add gdb prettyprinter for sparse matrices Jeff Trull jefftrull/eigen
272 MERGED Adding asynchandler to sycl queue as lack of it can cause undefined behaviour. Benoit Steiner benoitsteiner/opencl
271 MERGED Improved support for OpenCL Benoit Steiner benoitsteiner/opencl
268 MERGED Fix expm1 CUDA implementation (do not shadow exp CUDA implementation). Srinivas Vasudevan srvasude/eigen
267 DECLINED Make CMake config file relocatable Silvio Traversaro traversaro/eigen
264 MERGED add explicit template to numext::abs2 and fix signed/unsigned warning Angelos Mantzaflaris angelos_m/eigen
265 MERGED Add Expm1 support to Eigen. Srinivas Vasudevan srvasude/eigen
266 MERGED Makingt default numeric_list works with sycl. Benoit Steiner benoitsteiner/opencl
263 MERGED fix two warnings(unused typedef, unused variable) and a typo Angelos Mantzaflaris angelos_m/eigen
255 DECLINED cmake: added Eigen3::Eigen imported target Sergiu Deitsch sergiu/eigen
256 MERGED Add a default constructor for the "fake" __half class when not using the __half class provided by CUDA. Rasmus Larsen null
262 MERGED Adding sycl backend for TensorPadding.h and misc opencl fixes and leanups. Benoit Steiner benoitsteiner/opencl
259 DECLINED Fix warning: this ‘if’ clause does not guard... [-Wmisleading-indentation] Angelos Mantzaflaris null
261 MERGED Improved support for OpenCL Benoit Steiner benoitsteiner/opencl
260 MERGED Improved support for OpenCL Benoit Steiner benoitsteiner/opencl
258 DECLINED Fix warning: this ‘if’ clause does not guard... [-Wmisleading-indentation] Angelos Mantzaflaris null
237 MERGED Euler angles Tal Hadad tal500/eigen-eulerangles
254 MERGED Default Benoit Steiner benoitsteiner/opencl
253 MERGED OpenCL improvements Benoit Steiner benoitsteiner/opencl
252 MERGED Workaround for error in VS2012 with /clr chun.wang ChunW/eigen
251 MERGED ZVector float/std:complex for ZVector Konstantinos Margaritis kmargar/eigen
250 MERGED Default Benoit Steiner benoitsteiner/opencl
249 MERGED Reduce dispatch overhead in parallelFor. Rasmus Larsen null
248 MERGED Make sure not to call numext::maxi on expression templates Niels Ole Salscheider olesalscheider/eigen
247 MERGED Default Benoit Steiner benoitsteiner/opencl
246 MERGED Improved support for OpenCL Benoit Steiner benoitsteiner/opencl
235 MERGED Added support for AVX512 to Eigen Benoit Steiner benoitsteiner/eigen-avx512
244 MERGED Disable vectorization on device only when compiling for sycl Benoit Steiner benoitsteiner/opencl
243 MERGED Fixed the ambiguity in callig make_tuple for sycl backend. Benoit Steiner benoitsteiner/opencl
242 MERGED Use reentrant versions of lgamma and lgammaf if available. Rasmus Larsen null
241 MERGED Always enable /bigobj for tests to avoid a compile error in MSVC 2015 enricodetoma enricodetoma/eigen
239 DECLINED Initial changes to compile ReduceFunctor with clang-3.8 Hugh Perkins hughperkins/eigen
238 MERGED Added support for OpenCL to the Tensor Module Benoit Steiner benoitsteiner/opencl
232 MERGED Improve performance of parallelized matrix multiply for rectangular matrices Rasmus Larsen null
234 MERGED Enabling CUDA in Geometry Robert Lukierski lukier/eigen
236 MERGED Patch for Bug 1324 - Compilation error(s) when compiling 3.3rc1 with CUDA 8 on Windows Avi Ginsburg aviginsburg/eigen-fork
230 MERGED Fix a bug in psqrt for SSE and AVX when EIGEN_FAST_MATH=1 Rasmus Larsen null
233 MERGED Fully support complex types in SumReducer and MeanReducer when building for CUDA by using scalar_sum_op and scalar_product_op instead of operator+ and operator*. RJ Ryan rryan/eigen
227 DECLINED Update CUDA-specialized scalar_sum_op, scalar_difference_op, scalar_product_op and scalar_quotient_op to include changes made in b8b8c42 and 03556a1. RJ Ryan rryan/eigen
231 OPEN FindEigen3.cmake: Do not try to read the Macros.h file if it doesn't exist Daniele E. Domenichelli drdanz/eigen
229 MERGED Disabled MSVC level 4 warning C4714 Sergiu Deitsch sergiu/eigen
75 DECLINED Various warnings and compilation errors fixes Sergiu Deitsch sergiu/eigen
228 MERGED fix documentation of LinSpaced Felix Gruber felGru/eigen
61 DECLINED Fixed linking to boost from the blaze benchmark under MSVC FlorianGeorge FlorianGeorge/eigen_blaze_fork_2
220 DECLINED Add negation operator and hemispherization method to QuaternionBase Jonathan Brandmeyer jbrandmeyer/eigen
226 DECLINED Add default values for HasSelect and HasEq (== 0) for the base generic NumTraits. Eugene Brevdo ebrevdo/eigen
225 MERGED Add CUDA-specific std::complex specializations for scalar_sum_op, scalar_difference_op, scalar_product_op, and scalar_quotient_op. RJ Ryan rryan/eigen
224 MERGED remove ternary operator in euler angles hongkai dai hongkai-dai/eigen
223 DECLINED Made AutoDiffJacobian more intuitive to use and updated for C++11 Emil Fresk korken/eigen
222 MERGED Fix CUDA build broken by changes to min and max reduction. Rasmus Larsen null
221 MERGED Fix bugs to make min- and max reducers work with correctly with IEEE infinities. Rasmus Larsen null
201 OPEN Matrix sign function & refactor matrix power Chen-Pang He jdh8/eigen
215 MERGED Fix_tanh_inconsistent_for_tensorflow Ziming Dong suiyuan2009/eigen
219 MERGED Fix compilation on CUDA 8 by removing call to h2log1p (second attempt) Igor Babuschkin ibab/eigen
213 MERGED Modify FindEigen3.cmake to find Eigen3Config.cmake Silvio Traversaro traversaro/eigen-1
203 DECLINED Disable pkgconfig only for native windows builds xantares xantares/eigen-3
216 MERGED Enable efficient Tensor reduction for doubles on the GPU (continued) Igor Babuschkin ibab/eigen
218 MERGED Fix compilation on CUDA 8 due to missing h2log1p function Igor Babuschkin ibab/eigen
170 MERGED Inline dot operator and eval* methods in the DiagonalMatrix Simon Rutishauser rutishauser/eigen
217 MERGED Add log1p support for CUDA and half floats Igor Babuschkin ibab/eigen
207 DECLINED Enable efficient Tensor reduction for doubles on the GPU Igor Babuschkin null
206 MERGED Expose real and imag methods on Tensors Igor Babuschkin null
214 DECLINED Fix Eigen3Config.cmake install logic in Windows Silvio Traversaro traversaro/eigen-2
135 DECLINED Fix Eigen3Config.cmake install logic in Windows Silvio Traversaro traversaro/eigen
212 MERGED Fixed a code comment error Yi Lin ITimer/eigen
210 MERGED Bug 1232: refactor special functions as a new SpecialFunctions module, currently in unsupported/. g.gael@free.fr null
211 OPEN Quaternion::slerp Chen-Pang He jdh8/eigen
209 DECLINED Optimize QuaternionBase::slerp by removing one branch Chen-Pang He jdh8/eigen
208 OPEN Improve InnerIterator Chen-Pang He jdh8/eigen
204 MERGED Use complete nested namespace Eigen::internal Justin Carpentier carpent/eigen
205 MERGED Add missing CUDA kernel to tensor scan op Igor Babuschkin null
134 DECLINED image_retval image: corrected an error in documentation null null
202 DECLINED Enable pkgconfig for MinGW xantares xantares/eigen-2
200 DECLINED Fix pow_impl for integer types and add missing unit test. Rasmus Larsen null
198 DECLINED Add autodiff coverage for standard library hyperbolic functions. Geoffrey Lalonde lalonde/eigen
199 MERGED Return -1 from CurrentThreadId when called by thread outside the pool. Rasmus Larsen null
193 MERGED Add CurrentThreadId and NumThreads methods to Eigen threadpools and TensorDeviceThreadPool. Rasmus Larsen null
194 MERGED Relax mixing-type constraints for binary coefficient-wise operators (bug #279) g.gael@free.fr null
197 MERGED Implement exclusive scan option for Tensor library Igor Babuschkin null
195 MERGED Add small fixes to TensorScanOp Igor Babuschkin null
196 MERGED Fixed type conversion from int in Transform * vec Abhijit Kundu null
190 DECLINED Add TernaryFunctors and the betainc SpecialFunction. Eugene Brevdo ebrevdo/eigen
155 DECLINED Remove executable bit from header files. Ville Kallioniemi ville-k/eigen
192 MERGED Add generic scan method Igor Babuschkin null
191 MERGED Add syntactic sugar to Eigen tensors to allow more natural syntax. Rasmus Larsen null
189 MERGED Add scan op to Tensor module Igor Babuschkin null
82 MERGED fix for MKL_BLAS not defined in MKL 11.2 null null
188 MERGED Minor cleanups: 1. Get rid of a few unused variables. 2. Get rid of last uses of EIGEN_USE_COST_MODEL. Rasmus Larsen null
187 DECLINED Reduce overhead for small tensors and cheap ops by short-circuiting the const computation and block size calculation in parallelFor. Rasmus Larsen null
186 MERGED Improvements to parallelFor. Rasmus Larsen null
185 DECLINED Add custom converter op. Rasmus Larsen null
184 MERGED Eliminate mutual recursion in igamma{,c}_impl::Run. Justin Lebar jlebar/eigen
183 MERGED Detect cxx_constexpr support when compiling with clang. Rasmus Larsen null
182 MERGED Use computeProductBlockingSizes to compute blocking for both ShardByCol and ShardByRow cases. Rasmus Larsen null
181 DECLINED Implement stricter argument checking for SYRK and SY2K on complex matrices to match behavior of reference BLAS Rasmus Larsen null
179 MERGED Prevent crash in CompleteOrthogonalDecomposition if object was default constructed. Rasmus Larsen null
178 MERGED Eigen Tensor cost model part 2: Thread scheduling for standard evaluators and reductions. Rasmus Larsen null
177 MERGED Eigen Tensor cost model part 1. Rasmus Larsen null
174 MERGED Add matrix condition number estimation module. Rasmus Larsen null
176 MERGED Fix missing domain errors for polygamma and zeta function Till Hoffmann tillahoffmann/eigen
175 MERGED Static casting scalar types so as to let chlesky module of eigen work with ceres Partha Ghosh parthaEth/eigen
173 MERGED Added zeta function of two arguments and polygamma function Till Hoffmann tillahoffmann/eigen
172 DECLINED Fix cross-compiling windows version detection xantares xantares/eigen-1
171 DECLINED Fix windows detection when cross-compiling xantares xantares/eigen
168 DECLINED Warning fixes in headers Vijay M vijaysm/eigen
169 MERGED Bugfixes to cuda tests, igamma & igammac implemented, & tests for digamma, igamma, igammac on CPU & GPU. Eugene Brevdo ebrevdo/eigen
167 MERGED Add infinity() support to numext::numeric_limits, use it in lgamma. Eugene Brevdo ebrevdo/eigen
166 MERGED Speed up tensor FFT by up ~25-50%. Rasmus Larsen null
154 DECLINED Fix clang warnings Jan Prach jendap/eigen
109 DECLINED Numerical integration module for Eigen Sreekumar Thaithara Balan tbs1980/eigen-numerical-integration-module
108 DECLINED Add stable versions of normalize() and normalized() Deanna Hood d_hood/eigen
165 MERGED Tiny bugfix in SpecialFunctions Eugene Brevdo ebrevdo/eigen
164 DECLINED Add functor which composes a unary functor and a binary functor into a new binary functor. This enables building fused primitives for cwise operations. Vincent Vanhoucke vanhoucke/eigen
48 MERGED fixed Eigen::Spline constructor dimension bug Bo Li null
163 MERGED Implement complete orthogonal decomposition in Eigen. Rasmus Larsen null
138 MERGED [doc] grammar fix: "linearly space" -> "linearly spaced" nnyby nnyby/eigen
162 MERGED Fix condition that made the unit test spam stdout with bogus error messages. Rasmus Larsen null
161 MERGED Change Eigen's ColPivHouseholderQR to use numerically stable norm downdate formula Rasmus Larsen null
158 MERGED Add constructor for long types. Ville Kallioniemi ville-k/eigen
160 MERGED Making ceil() functor generic w.r.t packet type Abhijit Kundu null
159 MERGED Modifications to the tensor benchmarks to allow compilation in a standalone fashion. Yangqing Jia jiayq/eigen
150 MERGED Add digamma for CPU + CUDA. Eugene Brevdo ebrevdo/eigen
157 DECLINED bug 1153: replace nullptr with 0 for old gcc versions travis_abbott travis_abbott/eigen
156 MERGED Documentation fixes larsmans null
153 MERGED Add ctor for long Ville Kallioniemi ville-k/eigen
152 MERGED Alternative way of forcing instantiation of device kernels without causing warnings or requiring device to device kernel invocations. Jeremy Barnes jeremy_barnes/eigen
151 DECLINED Fix compiler error: more than one constructors matched the argument list. Ville Kallioniemi ville-k/eigen
149 MERGED [doc] Remove extra ';' in Advanced Initialization sample Connor connor-k/eigen
148 MERGED Add special functions to eigen: lgamma, erf, erfc. Eugene Brevdo ebrevdo/eigen
145 DECLINED Further fixes for CMAKE_INSTALL_PREFIX correctness Taylor Braun-Jones nocnokneo/eigen
143 DECLINED Further fixes for CMAKE_INSTALL_PREFIX correctness Taylor Braun-Jones nocnokneo/eigen
144 DECLINED Further fixes for CMAKE_INSTALL_PREFIX correctness Taylor Braun-Jones nocnokneo/eigen
147 MERGED Fixes internal compiler error while compiling with VC2015 Update1 x64. Nikolay Fedorov null
146 MERGED Use a class constructor to initialize CPU cache sizes Chris Jones dr15jones/eigen
142 MERGED Add round, ceil and floor for SSE4.1/AVX (Bug #70) Alexandre Avenel aavenel/eigen
119 MERGED Use GCC flags in mingw/win32 Michael Abrahams michael_abrahams/eigen_mingw
141 DECLINED Fix typo in TutorialSparse: laplace equation contains gradient symbol (\nabla) instead of laplacian (\Delta). Ilya Popov null
140 MERGED Bug 1097 Added ArpackSupport to cmake install target Abhijit Kundu null
139 DECLINED Add a functor to setFromTriplets to handle duplicated entries Calixte Denizet calixte_denizet/eigen
137 MERGED Specified signedness of char type in test Doug Kwan doug_kwan/eigen
131 DECLINED Implement plog and pexp for AltiVec. Doug Kwan null
136 MERGED Removed documentation that did not match the member function DenseBase::outerSize() daalpa daalpa/eigen
132 MERGED Refactoring of alignment control & handling of multiple packet types per scalar type g.gael@free.fr ggael/eigen-align
113 DECLINED Allows user decide whether to use posix_memealign or not, since some OS such as Solaris 10 has no posix_memealign. Changjiang Yang yangcha/eigen
133 MERGED [Doc] Fix a spelling error in TopicMultithreading.dox Yi Lin ITimer/eigen
130 MERGED Allowed tensor contraction operation with an empty array of dimension pairs, which performs a tensor product. godeffroy valet null
129 DECLINED Fix tests which break under C++11 because they throw from destructors. Brian Silverman bsilver8192/eigen
127 DECLINED Fixed some compiler bugs in NVCC, now compiles with CUDA and MSVC Jonas Adler adler-j/eigen_cuda_msvc
128 DECLINED Allow AlignedVector3 conversion with operators. vroussellet vroussellet/eigen
126 MERGED Add const getters for LM parameters Nicolas Mellado nmellado/eigen
122 MERGED Protect against compilation errors with nvcc and numext/complex. Nicolas Mellado nmellado/eigen
123 MERGED Modify GEMM to handle m=0, n=0, and k=0 cases. Eric Martin emartin/eigen
121 MERGED typo correction in mathFunction Emilie Guy Emie/eigen
120 MERGED Fix double to Scalar unwanted promotions Nicolas Mellado nmellado/eigen
118 MERGED Fix two small undefined behaviors caught by static analysis. Vincent Vanhoucke vanhoucke/eigen_vanhoucke
105 MERGED Add more coefficient-wise unary array functors Deanna Hood d_hood/eigen
60 MERGED Use trans(X) instead of X.transpose() in Blaze Benchmark FlorianGeorge FlorianGeorge/eigen_blaze_fork_2
117 MERGED Add check for older NEON macro petewarden petewarden/eigen
116 MERGED [Doc] Fix wording / typos in TemplateKeyword.dox Matt D MattPD/eigen
115 MERGED [[DOC]] Topic Multithreading dox compile error in example code resolved as in default branch Michiel Van Dyck mvdyck/eigen-3
103 MERGED Fix bug in pdiv which swaps 32-bit halves of a pair of Doug Kwan null
114 MERGED [Doc] Multi-threading fix Michiel Van Dyck mvdyck/eigen-3
106 DECLINED Adapted for biicode (C/C++ dependency manager) Francisco Ramirez franchuti688/eigen
111 DECLINED Add value_type typedef to Array and Matrix bases. nasailja nasailja/eigen
102 MERGED Enhancements to the Tensor codebase Benoit Steiner benoitsteiner/eigen
110 OPEN Bug 423 - Allow printing of rows of matrix using IOFormat of entire matrix Deanna Hood d_hood/eigen
107 MERGED [Doc] Fix missing image in sparse tutorial Deanna Hood d_hood/eigen
104 DECLINED Add more coefficient-wise unary array functors Deanna Hood null
100 MERGED Really use zero guess in ConjugateGradient::solve as documented Jan Blechta null
99 DECLINED Corrected the normalized() and normalize() functions in order not to return a NaN when the norm's value is null. Now, the functions will return its current value instead. Rafael Saback null
101 MERGED Added CMake support for Tensor module. CMake now installs CXX11 Tensor module like the rest of the unsupported modules Abhijit Kundu null
96 MERGED Add unsupported/CXX11/Tensor module g.gael@free.fr benoitsteiner/eigen
98 DECLINED Add _UpLo == Upper|Lower option for CG and MINRES. Jan Blechta null
92 MERGED Bug 877, Bug 572: Get rid of Index conversion warnings Christoph Hertzberg null
42 MERGED Added support for AVX and FMA to Eigen. Benoit Steiner benoitsteiner/eigen
68 DECLINED Tensor support Benoit Steiner null
97 DECLINED Add _UpLo == Whole option for CG and MINRES. Jan Blechta null
95 MERGED Using numext::pow instead of std::pow in poly_eval function. Filippo Basso bassofil/eigen
93 DECLINED fixed the outerproduct issue for un-evaluated variables Bo Li null
94 DECLINED Fix arm(32) prefetch usage broken in [e6b9611397611935cd89f8956a12e303fb1f3309]. Alexis Ballier aballier/eigen
36 MERGED Add cmake config files xantares null
91 MERGED Added cmake uninstall target Abhijit Kundu null
90 MERGED Adding missing OPENGL_LIBRARIES for openglsupport test. Abhijit Kundu null
89 DECLINED Bug 877: Get rid of most integer conversion warnings Christoph Hertzberg null
87 MERGED Extend NEON to add ARMv8 64-bit double support Konstantinos Margaritis kmargar/eigen
88 MERGED BDCSVD: fix CMake install (missing separator). Benjamin Chrétien bchretien/eigen
83 DECLINED Warnings on implicit conversion loses integer precision Georg Drenkhahn georg_drenkhahn/eigen
85 DECLINED Fixing some implicit integer conversion problems Georg Drenkhahn georg_drenkhahn/eigen
86 MERGED Make constructors explicit if they could lead to unintended implicit conversion Christoph Hertzberg null
84 MERGED Add VSX support Konstantinos Margaritis kmargar/eigen
79 DECLINED Added initial quadrature files. Jeff J null
81 MERGED Fix for warning on macro definitions of max() and min() in test.h Georg Drenkhahn georg_drenkhahn/eigen
80 MERGED Allow LevenbergMarquardt to work with non-standard types. Freddie Witherden FreddieWitherden/eigen
78 DECLINED CMake 3.0.0 adaptations to CMakeLists.txt files Georg Drenkhahn georg_drenkhahn/eigen
77 MERGED Fix uninitialized variable warning in SparseQR Vladimír Chalupecký null
76 MERGED FindEigen3.cmake: Add reading hints of Eigen directory location from environment variables EIGEN3_ROOT and EIGEN3_ROOT_DIR . Silvio Traversaro traversaro/eigen
74 MERGED Fix some typos in the documentation Benjamin Chrétien bchretien/eigen
69 MERGED Added Spline interpolation with derivatives. Jeff J null
72 MERGED Fix bug 839 Chen-Pang He jdh8/eigen
71 MERGED Find benchmark opponents more aggressively Chen-Pang He jdh8/eigen
70 MERGED Fix dox for namespaces Chen-Pang He jdh8/eigen
67 MERGED Change variable names in Eigen3Config.cmake to EIGEN3_* Vladimír Chalupecký null
66 DECLINED Tensor support Benoit Steiner benoitsteiner/eigen
65 DECLINED Bug 796: Fix variable names in UseEigen3.cmake to match those in Eigen3Config.cmake Vladimír Chalupecký null
63 DECLINED Sets the fill to ' ' when pretty printing matrices. This makes it easier to print Eigen matrices using logging systems that use a non-standard default fill. Benoit Steiner null
64 MERGED PolynomialSolver: fix bugs related to linear polynomials. Benjamin Chrétien bchretien/eigen
62 MERGED Made it possible to call the assignment operator on an Eigen::Block from a CUDA kernel. Benoit Steiner null
52 DECLINED Added support for Blaze to the benchmark suite FlorianGeorge FlorianGeorge/eigen_blaze_fork_1
59 MERGED Fix typo in SparseMatrix assert. Benjamin Chrétien bchretien/eigen
58 DECLINED Minor fix for Missing AVX subdirectory in CMake Abhijit Kundu null
57 MERGED Fix typo in Reductions tutorial. Benjamin Chrétien bchretien/eigen
56 MERGED Fixed a typo in CXX11Meta.h Benoit Steiner null
55 MERGED CUDA fixes Benoit Steiner null
47 MERGED Move the Base typedef's from private to public scope abachrac abachrac/eigen
54 MERGED Include LevenbergMarquardt in CMakeLists.txt. Freddie Witherden FreddieWitherden/eigen
53 MERGED Silenced a compilation warning produced by nvcc. Benoit Steiner null
51 MERGED Fixed typo: symmretric -> symmetric FlorianGeorge FlorianGeorge/eigen_blaze_fork_1
50 MERGED Fixed compilation error due to obsolete internal::abs and internal::sqrt function calls Abhijit Kundu null
49 MERGED fixed a template type conversion bug in AngleAxis found by Pei Luo Bo Li null
46 MERGED Fix typo: 'explicitely' -> 'explicitly' hdevalence hdevalence/eigen
45 MERGED Typo in the example for Eigen::SelfAdjointEigenSolver::eigenvectors, the first eigenvector should be col(0) not col(1) Olivier Saut osaut/eigen
40 MERGED Better fix for bug #503 Martinho Fernandes martinhofernandes/eigen
44 MERGED fix typo in evalSolverSugarFunction() Naumov Maks maksqwe/eigen
39 MERGED Tensor support for Eigen Christian Seiler chris-se/eigen
38 MERGED Fix functor installation Leszek Leszek/eigen
24 DECLINED Factorization of the JacobiSVD class into a SVDBase class to prepare the arrival of the BDCSVD (Bidiagonalisation Divide and Conquer algorithm) class GL27 gl27/eigen-bdcsvd
33 MERGED Fix for bug #503 Martinho Fernandes martinhofernandes/eigen
35 DECLINED Put exceptions under a #ifdef EIGEN_EXCEPTIONS guard Vincent Vanhoucke null
34 MERGED Silence unused variable warning. Vincent Vanhoucke null
32 MERGED Fixes for SparseMatrix to support non-POD scalar types Pavel Holoborodko null
31 MERGED Added support for custom scalars Pavel Holoborodko null
30 MERGED Added support for custom-scalars Pavel Holoborodko null
29 MERGED Quick fix in order to be custom-scalar friendly. Pavel Holoborodko advanpix/eigen-mp-3.2
28 MERGED Improve matrix power and (sparse) Kronecker product Chen-Pang He jdh8/eigen
27 MERGED Matrix power cleanup Chen-Pang He jdh8/eigen
26 MERGED new unsupported and not finished SVD, using a divide and conquert algorithm, with tests and benchmark GL27 gl27/eigen-bdcsvd-unsupported
25 MERGED Fix documentation typo Rhys Ulerich null
20 DECLINED Boost matrix square root. Implement triangular matrix power. Improve API of Kronecker product. Chen-Pang He jdh8/eigen
23 MERGED Benchmark class for the Geometry package jschwendner jschwendner/eigen
22 MERGED Fixed mpreal for IA64 architectures Pavel Holoborodko null
21 MERGED Updated mpreal support module Pavel Holoborodko null
19 MERGED the MINRES sparse solver for Eigen (for symmetric-indefinite matrices) giacomopo@gmail.com gpo/eigen-minres
18 MERGED Improve API for matrix power Chen-Pang He jdh8/eigen
17 MERGED Implement some functors for BLAS. Chen-Pang He jdh8/eigen
16 MERGED Write dox for assertions; move atanh2 to internal; improve BLAS Chen-Pang He jdh8/eigen
15 DECLINED Write dox for assertions; move atanh2 to internal; improve BLAS Chen-Pang He jdh8/eigen
14 DECLINED Add test and fix matrix power Chen-Pang He null
13 MERGED MatrixBase::pow() Chen-Pang He null
12 MERGED Fix typos in MatrixLogarithm - Branch 3.1 Chen-Pang He null
11 DECLINED Fixed clang++ warnings in debug mode. Janick Martinez Esturo ph03/eigen
10 MERGED Got rid of a warning message by doing an explicit cast Vlad-Andrei Lazar null
9 MERGED mainly enhance MatrixLogarithm's performance for RealScalar != double Chen-Pang He null
7 MERGED resolve pkgconfig destination - http://eigen.tuxfamily.org/bz/show_bug.cgi?id=338 Trevor Wennblom trevorw/eigen
8 MERGED enhance efficacy via avoiding exception handling Chen-Pang He null
6 DECLINED now MatrixExponential can take care of custom types (inclusive of mpreal) Chen-Pang He null
5 MERGED fix: is necessary for srand(time(NULL)) Chen-Pang He null
4 DECLINED now matrix exponential can take care of all types Chen-Pang He null
2 DECLINED add compatibility with long double Chen-Pang He null
3 MERGED add compatibility with long double Chen-Pang He null
1 MERGED Fix typos in Eigen documentation Zach Ploskey zploskey/eigen-doc