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

  • c5e8546306 Adding asynchandler to sycl queue as lack of it can cause undefined behaviour. Mehdi Goli 2016-12-15 16:59:57 +00:00
  • 4247d35d4b Fixed bug which (extremely rarely) could end in an infinite loop Christoph Hertzberg 2016-12-15 17:22:12 +01:00
  • 642dddcce2 Fix nonnull-compare warning Christoph Hertzberg 2016-12-15 17:16:56 +01:00
  • 1324ffef2f Reenabled the use of constexpr on OpenCL devices Benoit Steiner 2016-12-15 06:49:38 -08:00
  • 5d00fdf0e8 bug #1363: fix mingw's ABI issue Gael Guennebaud 2016-12-15 11:58:31 +01:00
  • 2c2e218471 Avoid using #define since they can conflict with user code Benoit Steiner 2016-12-14 19:49:15 -08:00
  • 3beb180ee5 Don't call EnvThread::OnCancel by default since it doesn't do anything. Benoit Steiner 2016-12-14 18:33:39 -08:00
  • 9ff5d0f821 Merged eigen/eigen into default Benoit Steiner 2016-12-14 17:32:16 -08:00
  • 730eb9fe1c Adding asynchronous execution as it improves the performance. Mehdi Goli 2016-12-14 17:38:53 +00:00
  • 11b492e993 bug #1358: fix compilation for sparse += sparse.selfadjointView(); Gael Guennebaud 2016-12-14 17:53:47 +01:00
  • e67397bfa7 bug #1359: fix compilation of col_major_sparse.row() *= scalar (used to work in 3.2.9 though the expression is not really writable) Gael Guennebaud 2016-12-14 17:05:26 +01:00
  • 98d7458275 bug #1359: fix sparse /=scalar and *=scalar implementation. InnerIterators must be obtained from an evaluator. Gael Guennebaud 2016-12-14 17:03:13 +01:00
  • 2d4a091beb Adding tensor contraction operation backend for Sycl; adding test for contractionOp sycl backend; adding temporary solution to prevent memory leak in buffer; cleaning up cxx11_tensor_buildins_sycl.h Mehdi Goli 2016-12-14 15:30:37 +00:00
  • c817ce3ba3 bug #1361: fix compilation issue in mat=perm.inverse() Gael Guennebaud 2016-12-13 23:10:27 +01:00
  • a432fc102d Moved the choice of ThreadPool to unsupported/Eigen/CXX11/ThreadPool Benoit Steiner 2016-12-12 15:24:16 -08:00
  • 8ae68924ed Made ThreadPoolInterface::Cancel() an optional functionality Benoit Steiner 2016-12-12 11:58:38 -08:00
  • 57acb05eef Update and extend doc on alignment issues. Gael Guennebaud 2016-12-11 22:45:32 +01:00
  • 76fca22134 Use a more accurate timer to sleep on Linux systems. Benoit Steiner 2016-12-09 15:12:24 -08:00
  • 4deafd35b7 Introduce a portable EIGEN_SLEEP macro. Benoit Steiner 2016-12-09 14:52:15 -08:00
  • aafa97f4d2 Fixed build error with MSVC Benoit Steiner 2016-12-09 14:42:32 -08:00
  • 2f5b7a199b Reworked the threadpool cancellation mechanism to not depend on pthread_cancel since it turns out that pthread_cancel doesn't work properly on numerous platforms. Benoit Steiner 2016-12-09 13:05:14 -08:00
  • 3d59a47720 Added a message to ease the detection of platforms on which thread cancellation isn't supported. Benoit Steiner 2016-12-08 14:51:46 -08:00
  • 28ee8f42b2 Added a Flush method to the RunQueue Benoit Steiner 2016-12-08 14:07:56 -08:00
  • 69ef267a77 Added the new threadpool cancel method to the threadpool interface based class. Benoit Steiner 2016-12-08 14:03:25 -08:00
  • 7bfff85355 Added support for thread cancellation on Linux Benoit Steiner 2016-12-08 08:12:49 -08:00
  • 6811e6cf49 Merged in srvasude/eigen/fix_cuda_exp (pull request PR-268) Benoit Steiner 2016-12-08 05:14:11 -08:00
  • 747202d338 typo Gael Guennebaud 2016-12-08 12:48:15 +01:00
  • bb297abb9e make sure we use the right eigen version Gael Guennebaud 2016-12-08 12:00:11 +01:00
  • 8b4b00d277 fix usage of custom compiler Gael Guennebaud 2016-12-08 11:59:39 +01:00
  • 7105596899 Add missing include and use -O3 Gael Guennebaud 2016-12-07 16:56:08 +01:00
  • 780f3c1adf Fix call to convert on linux Gael Guennebaud 2016-12-07 16:30:11 +01:00
  • 3855ab472f Cleanup file structure Gael Guennebaud 2016-12-07 14:23:49 +01:00
  • 59a59fa8e7 Update perf monitoring scripts to generate html/svg outputs Gael Guennebaud 2016-12-07 13:36:56 +01:00
  • f2c506b03d Add a script example to run and upload performance tests Gael Guennebaud 2016-12-06 16:46:52 +01:00
  • 1b4e085a7f generate png file for web upload Gael Guennebaud 2016-12-06 16:46:22 +01:00
  • f725f1cebc Mention the CMAKE_PREFIX_PATH variable. Gael Guennebaud 2016-12-06 15:23:45 +01:00
  • f90c4aebc5 Update monitored changeset lists Gael Guennebaud 2016-12-06 15:07:46 +01:00
  • eb621413c1 Revert vec/y to vec*(1/y) in row-major TRSM: - div is extremely costly - this is consistent with the column-major case - this is consistent with all other BLAS implementations Gael Guennebaud 2016-12-06 15:04:50 +01:00
  • 8365c2c941 Fix BLAS backend for symmetric rank K updates. Gael Guennebaud 2016-12-06 14:47:09 +01:00
  • 0c4d05b009 Explain how to choose your favorite Eigen version Gael Guennebaud 2016-12-06 11:34:06 +01:00
  • e049a2a72a Added relocatable cmake support also for CMake before 3.0 and after 2.8.8 Silvio Traversaro 2016-12-06 10:37:34 +01:00
  • e6c8b5500c Change comparisons to use Scalar instead of RealScalar. Srinivas Vasudevan 2016-12-05 14:01:45 -08:00
  • f7d7c33a28 Fix expm1 CUDA implementation (do not shadow exp CUDA implementation). Srinivas Vasudevan 2016-12-05 12:19:01 -08:00
  • 18481b518f Make CMake config file relocatable Silvio Traversaro 2016-12-05 10:39:52 +01:00
  • c68c8631e7 fix compilation of BTL's blaze interface Gael Guennebaud 2016-12-05 23:02:16 +01:00
  • 1ff1d4a124 Add performance monitoring for LLT Gael Guennebaud 2016-12-05 23:01:52 +01:00
  • 18de92329e use numext::abs (grafted from 0a08d4c60b ) Angelos Mantzaflaris 2016-12-02 11:48:06 +01:00
  • e8a6aa518e 1. Add explicit template to abs2 (resolves deduction for some arithmetic types) 2. Avoid signed-unsigned conversion in comparison (warning in case Scalar is unsigned) (grafted from 4086187e49 ) Angelos Mantzaflaris 2016-12-02 11:39:18 +01:00
  • a6b971e291 Fix memory leak in Ref<Sparse> Gael Guennebaud 2016-12-05 16:59:30 +01:00
  • 8640ffac65 Optimize SparseLU::solve for rhs vectors Gael Guennebaud 2016-12-05 15:41:14 +01:00
  • 62acd67903 remove temporary in SparseLU::solve Gael Guennebaud 2016-12-05 15:11:57 +01:00
  • 0db6d5b3f4 bug #1356: fix calls to evaluator::coeffRef(0,0) to get the address of the destination by adding a dstDataPtr() member to the kernel. This fixes undefined behavior if dst is empty (nullptr). Gael Guennebaud 2016-12-05 15:08:09 +01:00
  • 91003f3b86 typo Gael Guennebaud 2016-12-05 13:51:07 +01:00
  • 445c015751 extend monitoring benchmarks with transpose matrix-vector and triangular matrix-vectors. Gael Guennebaud 2016-12-05 13:36:26 +01:00
  • e3f613cbd4 Improve performance of row-major-dense-matrix * vector products for recent CPUs. This revised version does not bother about aligned loads/stores, and rather processes 8 rows at ones for better instruction pipelining. Gael Guennebaud 2016-12-05 13:02:01 +01:00
  • 3abc827354 Clean debugging code Gael Guennebaud 2016-12-05 12:59:32 +01:00
  • 462c28e77a Merged in srvasude/eigen (pull request PR-265) Benoit Steiner 2016-12-05 02:31:11 +00:00
  • 4465d20403 Add missing generic load methods. Gael Guennebaud 2016-12-03 21:25:04 +01:00
  • 6a5fe86098 Complete rewrite of column-major-matrix * vector product to deliver higher performance of modern CPU. The previous code has been optimized for Intel core2 for which unaligned loads/stores were prohibitively expensive. This new version exhibits much higher instruction independence (better pipelining) and explicitly leverage FMA. According to my benchmark, on Haswell this new kernel is always faster than the previous one, and sometimes even twice as fast. Even higher performance could be achieved with a better blocking size heuristic and, perhaps, with explicit prefetching. We should also check triangular product/solve to optimally exploit this new kernel (working on vertical panel of 4 columns is probably not optimal anymore). Gael Guennebaud 2016-12-03 21:14:14 +01:00
  • 2bfece5cd1 Merged eigen/eigen into default Benoit Steiner 2016-12-02 16:30:14 -08:00
  • 09ee7f0c80 Fix small nit where I changed name of plog1p to pexpm1. Srinivas Vasudevan 2016-12-02 15:30:12 -08:00
  • a0d3ac760f Sync from Head. Srinivas Vasudevan 2016-12-02 14:14:45 -08:00
  • 218764ee1f Added support for expm1 in Eigen. Srinivas Vasudevan 2016-12-02 14:13:01 -08:00
  • 66f65ccc36 Ease compiler job to generate clean and efficient code in mat*vec. Gael Guennebaud 2016-12-02 22:41:26 +01:00
  • fe696022ec Operators += and -= do not resize! Gael Guennebaud 2016-12-02 22:40:25 +01:00
  • 592acc5bfa Makingt default numeric_list works with sycl. Mehdi Goli 2016-12-02 17:58:30 +00:00
  • 8dfb3e00b8 merge Gael Guennebaud 2016-12-02 11:34:21 +01:00
  • 4c0d5f3c01 Add perf monitoring for gemv Gael Guennebaud 2016-12-02 11:34:12 +01:00
  • d2718d662c Re-enable A^T*A action in BTL Gael Guennebaud 2016-12-02 11:32:03 +01:00
  • 22f7d398e2 bug #1355: Fixed wrong line-endings on two files Christoph Hertzberg 2016-12-02 11:22:05 +01:00
  • 8c24723a09 typo UIntPtr (grafted from b6f04a2dd4 ) Angelos Mantzaflaris 2016-12-01 21:25:58 +01:00
  • aeba0d8655 fix two warnings(unused typedef, unused variable) and a typo (grafted from a9aa3bcf50 ) Angelos Mantzaflaris 2016-12-01 21:23:43 +01:00
  • 27873008d4 Clean up SparseCore module regarding ReverseInnerIterator Gael Guennebaud 2016-12-01 21:55:10 +01:00
  • 181138a1cb fix member order Gael Guennebaud 2016-12-01 17:06:20 +01:00
  • 9f297d57ae Merged in rmlarsen/eigen (pull request PR-256) Gael Guennebaud 2016-12-01 15:27:33 +00:00
  • f95e3b84a5 merge Gael Guennebaud 2016-12-01 16:18:57 +01:00
  • 7ff26ddcbb Merged eigen/eigen into default Benoit Steiner 2016-12-01 07:13:17 -08:00
  • 79aa2b784e Adding sycl backend for TensorPadding.h; disbaling __unit128 for sycl in TensorIntDiv.h; disabling cashsize for sycl in tensorDeviceDefault.h; adding sycl backend for StrideSliceOP ; removing sycl compiler warning for creating an array of size 0 in CXX11Meta.h; cleaning up the sycl backend code. Mehdi Goli 2016-12-01 13:02:27 +00:00
  • 5e3c5c42f6 cmake: remove architecture dependency from Eigen3ConfigVersion.cmake Sergiu Deitsch 2016-11-30 15:46:46 +01:00
  • 3440b46e2f doc: mention the NO_MODULE option and target availability (grafted from 65f09be8d2 ) Sergiu Deitsch 2016-11-30 15:41:38 +01:00
  • a0329f64fb Add a default constructor for the "fake" __half class when not using the __half class provided by CUDA. Rasmus Munk Larsen 2016-11-29 13:18:09 -08:00
  • e777674a87 Extend polynomial solver unit tests to complexes (grafted from f12b368417 ) Gael Guennebaud 2016-11-23 16:05:45 +01:00
  • 222ce4b49d Automatically switch between EigenSolver and ComplexEigenSolver, and fix a few Real versus Scalar issues. (grafted from 56e5ec07c6 ) Gael Guennebaud 2016-11-23 16:05:10 +01:00
  • 4415d4e2d4 Patch from Oleg Shirokobrod to extend polynomial solver to complexes (grafted from 9246587122 ) Gael Guennebaud 2016-11-23 15:42:26 +01:00
  • 07c2081aac fix unit test in c++03: c++03 does not allow passing local or anonymous enum as template param (grafted from 48898a988a ) Gael Guennebaud 2019-03-18 11:38:36 +01:00
  • dd93c41618 Complete previous backport with is_valid_index_type Gael Guennebaud 2019-03-17 22:02:18 +01:00
  • 441b3511de bug #1692: enable enum as sizes of Matrix and Array (grafted from cf7e2e277f ) Gael Guennebaud 2019-03-17 21:59:30 +01:00
  • cf0bf73edc use proper id Thomas Capricelli 2019-03-12 13:54:24 +01:00
  • f719b23ef7 update tracking code for 3.3 branch Thomas Capricelli 2019-03-12 13:49:07 +01:00
  • 5110d803e4 Change license from LGPL to MPL2 with agreement from David Harmon. (grafted from 2df4f00246 ) Gael Guennebaud 2019-03-07 18:17:10 +01:00
  • 6b59aa705a bug #1689 fix used-but-marked-unused warning (grafted from bfbf7da047 ) Gael Guennebaud 2019-03-05 23:46:24 +01:00
  • 7b93328baf Enable construction of Ref<VectorType> from a runtime vector. (grafted from b0d406d91c ) Gael Guennebaud 2019-03-03 15:25:25 +01:00
  • 6adb70d3c4 Enable documentation of Array's typedefs (grafted from 475295b5ff ) Gael Guennebaud 2019-02-20 15:18:07 +01:00
  • c354f59fb6 bug #1679: avoid possible division by 0 in complex-schur (grafted from f2970819a2 ) Gael Guennebaud 2019-02-15 09:39:25 +01:00
  • 637302a4c2 Spline.h: fix spelling "spang" -> "span" (grafted from 953ca5ba2f ) Steven Peters 2019-02-08 06:23:24 +00:00
  • 2aa9eb3ce8 bug #1676: workaround GCC's bug in c++17 mode. (grafted from b3c4344a68 ) Gael Guennebaud 2019-02-07 15:21:35 +01:00
  • f1c12d8ff0 Workaround gcc's alloc-size-larger-than= warning (grafted from bcb7c66b53 ) Gael Guennebaud 2018-10-07 21:55:59 +02:00
  • 6c4d57dc9e Fix a gcc7 warning about bool * bool in abs2 default implementation. (grafted from 561f777075 ) Gael Guennebaud 2017-06-27 12:05:17 +02:00
  • 6870a39feb Hide some annoying unused variable warnings in g++8.1 (grafted from a7779a9b42 ) Christoph Hertzberg 2019-01-29 16:48:21 +01:00
  • bb9981e24b Fix gcc 8.1 warning: "maybe use uninitialized" (grafted from 16b2001ece ) Gael Guennebaud 2018-10-07 21:54:49 +02:00