devtools/eigen
1
0
Fork 0
mirror of https://gitlab.com/libeigen/eigen.git synced 2026-04-10 11:34:33 +08:00
Code Issues Packages Projects Releases Wiki Activity

Commit Graph

  • db9b154193 Add missing non-const reverse method in VectorwiseOp. Gael Guennebaud 2016-07-16 15:19:28 +02:00
  • 461cd819c2 Workaround VS2015 bug Gael Guennebaud 2016-07-13 18:46:01 +02:00
  • 5ea0864c81 Fix regression in a previous commit: some diagonal entry might not be treated by the 2x2 real preconditioner. Gael Guennebaud 2016-07-13 18:37:54 +02:00
  • 20f7ef2f89 An evalTo expression is only aligned iff both the lhs and the rhs are aligned. Benoit Steiner 2016-07-12 10:56:42 -07:00
  • b4343aa67e Avoid division by very small entries when extracting singularvalues, and explicitly handle the 1x1 complex case. Gael Guennebaud 2016-07-12 17:22:03 +02:00
  • e2aa58b631 Consider denormals as zero in makeJacobi and 2x2 SVD. This also fix serious issues with x387 for which values can be much smaller than the smallest denormal! Gael Guennebaud 2016-07-12 17:21:03 +02:00
  • 263993a7b6 Fix test for nearly null input Gael Guennebaud 2016-07-12 17:19:26 +02:00
  • 9ab35d8ba4 Fix compilation of doc Gael Guennebaud 2016-07-12 16:47:39 +02:00
  • 19614497ae Add some doxygen's images to support both old and recent doxygen versions (with some vague definitions of old and recent ;) ) Gael Guennebaud 2016-07-12 16:45:43 +02:00
  • c98bac2966 Manually add -stdd=c++11 to nvcc for old cmake versions Gael Guennebaud 2016-07-12 09:29:18 +02:00
  • 013a904237 Pulled latest updates from trunk Benoit Steiner 2016-07-11 14:29:05 -07:00
  • 40eb97516c reverted unintended change. Benoit Steiner 2016-07-11 14:28:03 -07:00
  • 8b3fc31b55 compile fix (SFINAE variant apparently didn't work for all compilers) for the following compiler/platform: gcc (GCC) 4.1.2 20080704 (Red Hat 4.1.2-46) Copyright (C) 2006 Free Software Foundation, Inc. klimpel 2016-07-11 17:42:22 +02:00
  • 03b71c273e Made the packetmath test compile again. A better fix would be to move the special function tests to the unsupported directory where the code now resides. Benoit Steiner 2016-07-11 13:50:24 -07:00
  • 3a2dd352ae Improved the contraction mapper to properly support tensor products Benoit Steiner 2016-07-11 13:43:41 -07:00
  • 0bc020be9d Improved the detection of packet size in the tensor scan evaluator. Benoit Steiner 2016-07-11 12:14:56 -07:00
  • a96a7ce3f7 Move CUDA's special functions to SpecialFunctions module. Gael Guennebaud 2016-07-11 18:39:11 +02:00
  • bec35f4c55 Clarify that SpecialFunctions is unsupported Gael Guennebaud 2016-07-11 18:38:40 +02:00
  • fd60966310 merge Gael Guennebaud 2016-07-11 18:11:47 +02:00
  • 7d636349dc Fix configuration of CUDA: - preserve user defined CUDA_NVCC_FLAGS - remove the -ansi flag that conflicts with -std=c++11 - do not add -std=c++11 if already there Gael Guennebaud 2016-07-11 18:09:04 +02:00
  • 3e348fdcf9 Workaround MSVC bug Gael Guennebaud 2016-07-11 15:24:52 +02:00
  • 131ee4bb8e Split test_slice_in_expr which seems to be huge for visual Gael Guennebaud 2016-07-11 11:46:55 +02:00
  • 194daa3048 Fix assertion (it did not make sense for static_val types) Gael Guennebaud 2016-07-11 11:39:27 +02:00
  • 18c35747ce Emulate _BitScanReverse64 for 32 bits builds Gael Guennebaud 2016-07-11 11:38:04 +02:00
  • ef05463fcf Merged kmargar/eigen/tip into default, Altivec/VSX port should be working ok now. Konstantinos Margaritis 2016-07-10 16:11:46 +03:00
  • 9f7caa7e7d minor fixes for big endian altivec/vsx Konstantinos Margaritis 2016-07-10 07:05:10 -03:00
  • 3c795c6923 bug #1119: Adjust call to ?gssvx for SuperLU 5 Also improved corresponding cmake module to detect versions 5.x Christoph Hertzberg 2016-07-10 02:29:57 +02:00
  • 57113e00f9 Relax strict equality Gael Guennebaud 2016-07-09 23:37:11 +02:00
  • 599f8ba617 Change runtime to compile-time conditional. Gael Guennebaud 2016-07-08 11:39:43 +02:00
  • 544935101a Fix warnings Gael Guennebaud 2016-07-08 11:38:52 +02:00
  • 59bf2774a3 Fix warnings Gael Guennebaud 2016-07-08 11:38:11 +02:00
  • 2f7e2614e7 bug #1232: refactor special functions as a new SpecialFunctions module, currently in unsupported/. Gael Guennebaud 2016-07-08 11:13:55 +02:00
  • 8b7431d8fd fix compilation with c++11 Gael Guennebaud 2016-07-07 15:18:23 +02:00
  • 69378eed0b Split huge unit test Gael Guennebaud 2016-07-07 15:18:04 +02:00
  • c684e37d32 Prevent division by zero. Gael Guennebaud 2016-07-07 11:03:01 +02:00
  • 179ebb88f9 Fix warning Gael Guennebaud 2016-07-07 09:16:40 +02:00
  • 5d2dada197 Fix warnings Gael Guennebaud 2016-07-07 09:05:15 +02:00
  • f5e780fb05 split huge unit test Gael Guennebaud 2016-07-07 08:59:59 +02:00
  • 66917299a9 Add debug output Gael Guennebaud 2016-07-06 22:27:15 +02:00
  • 5ca2457fa5 Fix unit test. Gael Guennebaud 2016-07-06 22:25:24 +02:00
  • 9b68ed4537 Relax is_equal to is_approx because scaling might modify last bit. Gael Guennebaud 2016-07-06 15:02:49 +02:00
  • c3b23d7dbf Fix support of Intel's VML Gael Guennebaud 2016-07-06 14:07:32 +02:00
  • 8ec4d6480d Fix compilation with recent updates of icc 2016 Gael Guennebaud 2016-07-06 14:07:14 +02:00
  • 5b3a6f51d3 Improve numerical robustness of RealSchur: add scaling and compare sub-diag entries to largest diagonal entry instead of the 2 neighbors. Gael Guennebaud 2016-07-06 13:45:30 +02:00
  • d2b5a19e0f Fix warning. Gael Guennebaud 2016-07-06 11:05:30 +02:00
  • 367ef66af3 Re-enable some specializations for Assignment<.,Product<>> Gael Guennebaud 2016-07-05 22:58:14 +02:00
  • 155d8d8603 Fix compilation with msvc Gael Guennebaud 2016-07-05 14:43:42 +02:00
  • 43696ede8f Revert unwanted changes. Gael Guennebaud 2016-07-04 22:40:36 +02:00
  • b39fd8217f Fix nesting of SolveWithGuess, and add unit test. Gael Guennebaud 2016-07-04 17:47:47 +02:00
  • ec02af1047 Fix template resolution. Gael Guennebaud 2016-07-04 17:37:33 +02:00
  • fbcfc2f862 Add unit test for solveWithGuess, and fix template resolution. Gael Guennebaud 2016-07-04 17:19:38 +02:00
  • 7f7839c12f Add documentation and exemples for inplace decomposition. Gael Guennebaud 2016-07-04 17:18:26 +02:00
  • 32a41ee659 bug #707: add inplace decomposition through Ref<> for Cholesky, LU and QR decompositions. Gael Guennebaud 2016-07-04 15:13:35 +02:00
  • 75e80792cc Update relevent list of changesets. Gael Guennebaud 2016-07-04 14:32:34 +02:00
  • dacc544b84 asm escape was not strong enough to prevent too aggressive compiler optimization let's fallback to no-inline. Gael Guennebaud 2016-07-04 14:32:15 +02:00
  • b74e45906c Few fixes in perf-monitoring. Gael Guennebaud 2016-07-04 14:30:50 +02:00
  • ce9fc0ce14 fix clang compilation Gael Guennebaud 2016-07-04 12:59:02 +02:00
  • 440020474c Workaround compilation issue with msvc Gael Guennebaud 2016-07-04 12:49:19 +02:00
  • e61cee7a50 Fix compilation of some unit tests with msvc Gael Guennebaud 2016-07-04 11:49:03 +02:00
  • 91b3039013 Change the semantic of the last template parameter of Assignment from "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. Gael Guennebaud 2016-07-04 11:02:00 +02:00
  • 0fa9e4a15c Fix performance regression in dgemm introduced by changeset 5d51a7f12c Gael Guennebaud 2016-07-02 17:35:08 +02:00
  • 672076db5d Fix performance regression introduced in changeset e56aabf205 . Register blocking sizes are better handled by the cache size heuristics. The current code introduced very small blocks, for instance for 9x9 matrix, thus killing performance. Gael Guennebaud 2016-07-02 15:40:56 +02:00
  • eeb0d880ee Enable efficient Tensor reduction for doubles Igor Babuschkin 2016-07-01 19:08:26 +01:00
  • 78f37ca03c Expose real and imag methods on Tensors Igor Babuschkin 2016-07-01 17:34:31 +01:00
  • d161b8f03a Merged in carpent/eigen (pull request PR-204) Gael Guennebaud 2016-07-01 09:56:44 +02:00
  • cb2d8b8fa6 Made it possible to compile reductions for an old cuda architecture and run them on a recent gpu. Benoit Steiner 2016-06-29 15:42:01 -07:00
  • b2a47641ce Made the code compile when using CUDA architecture < 300 Benoit Steiner 2016-06-29 15:32:47 -07:00
  • b047ca765f Merged in ibab/eigen/fix-tensor-scan-gpu (pull request PR-205) Benoit Steiner 2016-06-29 14:52:19 -07:00
  • 85699850d9 Add missing CUDA kernel to tensor scan op Igor Babuschkin 2016-06-29 11:54:35 +01:00
  • 6126886a67 Use complete nested namespace Eigen::internal Justin Carpentier 2016-06-28 20:09:25 +02:00
  • 328c5d876a Undo changes in AltiVec --- I don't have any way to test there. Benoit Jacob 2016-06-28 11:15:25 -04:00
  • 38fb606052 Avoid global variables with static constructors in NEON/Complex.h Benoit Jacob 2016-06-28 11:12:49 -04:00
  • 1a9f92e781 Added a test to validate the tensor scan evaluation on GPU. The test is currently disabled since the code segfaults. Benoit Steiner 2016-06-27 16:02:52 -07:00
  • 75c333f94c Don't store the scan axis in the evaluator of the tensor scan operation since it's only used in the constructor. Also avoid taking references to values that may becomes stale after a copy construction. Benoit Steiner 2016-06-27 10:32:38 -07:00
  • d476cadbb8 bug #1247: fix regression in compilation of pow(integer,integer), and add respective unit tests. Gael Guennebaud 2016-06-25 10:12:06 +02:00
  • cfff370549 Fix hyperbolic functions for autodiff. Gael Guennebaud 2016-06-24 23:21:35 +02:00
  • c50c73cae2 Fix missing specialization. Gael Guennebaud 2016-06-24 23:10:39 +02:00
  • 3852351793 merge pull request 198 Gael Guennebaud 2016-06-24 11:48:17 +02:00
  • 6dd9077070 Fix some unused typedef warnings. Gael Guennebaud 2016-06-24 11:34:21 +02:00
  • ce90647fa5 Fix NumTraits<AutoDiff> Gael Guennebaud 2016-06-24 11:34:02 +02:00
  • fa39f81b48 Fix instantiation of ScalarBinaryOpTraits for AutoDiff. Gael Guennebaud 2016-06-24 11:33:30 +02:00
  • cd577a275c Relax promote_scalar_arg logic to enable promotion to Expr::Scalar if conversion to Expr::Literal fails. This is useful to cancel expression template at the scalar level, e.g. with AutoDiff<AutoDiff<>>. This patch also defers calls to NumTraits in cases for which types are not directly compatible. Gael Guennebaud 2016-06-24 11:28:54 +02:00
  • deb45ad4bc bug #1245: fix compilation with msvc Gael Guennebaud 2016-06-24 09:52:25 +02:00
  • a9c1e4d7b7 Return -1 from CurrentThreadId when called by thread outside the pool. Rasmus Munk Larsen 2016-06-23 16:40:07 -07:00
  • d39df320d2 Resolve merge. Rasmus Munk Larsen 2016-06-23 15:08:03 -07:00
  • 361dbd246d Add unit test for printing empty tensors Gael Guennebaud 2016-06-23 18:54:30 +02:00
  • 360a743a10 bug #1241: does not emmit anything for empty tensors Gael Guennebaud 2016-06-23 18:47:31 +02:00
  • 55fc04e8b5 Fix operator priority Gael Guennebaud 2016-06-23 15:36:42 +02:00
  • bf2d5edecc Fix warning. Gael Guennebaud 2016-06-23 15:35:17 +02:00
  • 7c6561485a merge PR 194 Gael Guennebaud 2016-06-23 15:29:57 +02:00
  • be107e387b fix compilation with clang 3.9, fix performance with pset1, use vector operators instead of intrinsics in some cases Konstantinos Margaritis 2016-06-23 10:19:05 -03:00
  • 76faf4a965 Introduce a NumTraits<T>::Literal type to be used for literals, and improve mixing type support in operations between arrays and scalars: - 2 * ArrayXcf is now optimized in the sense that the integer 2 is properly promoted to a float instead of a complex<float> (fix a regression) - 2.1 * ArrayXi is now forbiden (previously, 2.1 was converted to 2) - This mechanism should be applicable to any custom scalar type, assuming NumTraits<T>::Literal is properly defined (it defaults to T) Gael Guennebaud 2016-06-23 14:27:20 +02:00
  • a3f7edf7e7 Biug 1242: fix comma init with empty matrices. Gael Guennebaud 2016-06-23 10:25:04 +02:00
  • a29a2cb4ff Silenced a couple of compilation warnings generated by xcode Benoit Steiner 2016-06-22 16:43:02 -07:00
  • f8fcd6b32d Turned the constructor of the PerThread struct into what is effectively a constant expression to make the code compatible with a wider range of compilers Benoit Steiner 2016-06-22 16:03:11 -07:00
  • c58df31747 Handle empty tensors in the print functions Benoit Steiner 2016-06-21 09:22:43 -07:00
  • de32f8d656 Fixed the printing of rank-0 tensors Benoit Steiner 2016-06-20 10:46:45 -07:00
  • 8c34b5a0e3 mostly cleanups and modernizing code Konstantinos Margaritis 2016-06-19 16:13:17 -03:00
  • b410d46482 mostly cleanups and modernizing code Konstantinos Margaritis 2016-06-19 16:12:52 -03:00
  • b80379bda0 fixed pexp<Packet2d>, was failing tests Konstantinos Margaritis 2016-06-19 16:11:58 -03:00