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

  • 306daa24a3 Fixed a compilation warning Benoit Steiner 2016-10-28 03:50:31 +00:00
  • 8471cf1996 Fixed compilation warning Benoit Steiner 2016-10-28 03:46:08 +00:00
  • b0c5bfdf78 Added missing template parameters Benoit Steiner 2016-10-28 03:43:41 +00:00
  • 2ebb314fa7 Use threadsafe versions of lgamma and lgammaf if possible. Rasmus Munk Larsen 2016-10-27 16:17:12 -07:00
  • 530f20c21a Workaround MSVC issue. Gael Guennebaud 2016-10-27 21:51:37 +02:00
  • c3ce4f9ac0 Merged in enricodetoma/eigen (pull request PR-241) Gael Guennebaud 2016-10-27 19:21:28 +00:00
  • 7d64e6752c Pulled latest updates from trunk Benoit Steiner 2016-10-26 18:48:06 -07:00
  • 0a4c4d40b4 Removed a template parameter for fixed sized tensors Benoit Steiner 2016-10-26 18:47:37 -07:00
  • 3ecb343dc3 Fix regression in X = (X*X.transpose())/s with X rectangular by deferring resizing of the destination after the creation of the evaluator of the source expression. Gael Guennebaud 2016-10-26 22:50:41 +02:00
  • 6ed571744b Always enable /bigobj for tests to avoid a compile error in MSVC 2015 enrico.detoma 2016-10-26 22:48:46 +02:00
  • 97feea9d39 add a generic EIGEN_HAS_CXX11 Gael Guennebaud 2016-10-26 15:53:13 +02:00
  • ca6a2a5248 Fix warning with ICC Gael Guennebaud 2016-10-26 14:13:05 +02:00
  • 5f2dd503ff Replaced tabs with spaces Benoit Steiner 2016-10-25 20:40:58 -07:00
  • 1644bafe29 Code cleanup Benoit Steiner 2016-10-25 20:36:14 -07:00
  • b15a5dc3f4 Fix ICC warnings Gael Guennebaud 2016-10-25 22:20:24 +02:00
  • aad72f3c6d Add missing inline keywords Gael Guennebaud 2016-10-25 20:20:09 +02:00
  • 3e194a6a73 Fixed a typo Benoit Steiner 2016-10-25 08:42:15 -07:00
  • 58146be99b bug #1004: one more rewrite of LinSpaced for floating point numbers to guarantee both interpolation and monotonicity. This version simply does low+i*step plus a branch to return high if i==size-1. Vectorization is accomplished with a branch and the help of pinsertlast. Some quick benchmark revealed that the overhead is really marginal, even when filling small vectors. Gael Guennebaud 2016-10-25 16:53:09 +02:00
  • 13fc18d3a2 Add a pinsertlast function replacing the last entry of a packet by a scalar. (useful to vectorize LinSpaced) Gael Guennebaud 2016-10-25 16:48:49 +02:00
  • 2634f9386c bug #1333: fix bad usage of const_cast_derived. Better use .data() for that purpose. Gael Guennebaud 2016-10-24 22:22:35 +02:00
  • 9e8f07d7b5 Cleanup ArrayWrapper and MatrixWrapper by removing redundant accessors. Gael Guennebaud 2016-10-24 22:16:48 +02:00
  • b027d7a8cf bug #1004: remove the inaccurate "sequential" path for LinSpaced, mark respective function as deprecated, and enforce strict interpolation of the higher range using a correction term. Now, even with floating point precision, both the 'low' and 'high' bounds are exactly reproduced at i=0 and i=size-1 respectively. Gael Guennebaud 2016-10-24 20:27:21 +02:00
  • b11aab5fcc Merged in benoitsteiner/opencl (pull request PR-238) Benoit Steiner 2016-10-24 15:30:45 +00:00
  • 53c77061f0 bug #698: rewrite LinSpaced for integer scalar types to avoid overflow and guarantee an even spacing when possible. Otherwise, the "high" bound is implicitly lowered to the largest value allowing for an even distribution. This changeset also disable vectorization for this integer path. Gael Guennebaud 2016-10-24 15:50:27 +02:00
  • e8e56c7642 Add unit test for overflow in LinSpaced Gael Guennebaud 2016-10-24 15:43:51 +02:00
  • 40f62974b7 bug #1328: workaround a compilation issue with gcc 4.2 Gael Guennebaud 2016-10-20 19:19:37 +02:00
  • cf20b30d65 Merge latest updates from trunk Benoit Steiner 2016-10-20 09:42:05 -07:00
  • 03b63e182c Added SYCL include in Tensor. Luke Iwanski 2016-10-20 15:32:44 +01:00
  • d3943cd50c Fixed a few typos in the ternary tensor expressions types Benoit Steiner 2016-10-19 12:56:12 -07:00
  • 15eca2432a Euler tests: Tighter precision when no roll exists and clean code. Tal Hadad 2016-10-18 23:24:57 +03:00
  • 6f4f12d1ed Add isApprox() and cast() functions. Tal Hadad 2016-10-17 22:23:47 +03:00
  • 7402cfd4cc Add safty for near pole cases and test them better. Tal Hadad 2016-10-17 20:42:08 +03:00
  • 8fb162fc85 Fixing the typo regarding missing #if needed for proper handling of exceptions in Eigen/Core. Mehdi Goli 2016-10-16 12:52:34 +01:00
  • 58f5d7d058 Fix calc bug, docs and better testing. Tal Hadad 2016-10-16 14:39:26 +03:00
  • e36cb91c99 Fixing the code indentation in the TensorReduction.h file. Mehdi Goli 2016-10-14 18:03:00 +01:00
  • 2e188dd4d4 Merged ComputeCpp to default. Luke Iwanski 2016-10-14 16:47:40 +01:00
  • 15380f9a87 Applyiing Benoit's comment to return the missing line back in Eigen/Core Mehdi Goli 2016-10-14 16:39:41 +01:00
  • 692b30ca95 Fix previous merge. Gael Guennebaud 2016-10-14 17:16:28 +02:00
  • 050c681bdd Merged in rmlarsen/eigen2 (pull request PR-232) Gael Guennebaud 2016-10-14 14:51:09 +00:00
  • 078a202621 Merge Hongkai Dai correct range calculation, and remove ranges from API. Docs updated. Tal Hadad 2016-10-14 16:03:28 +03:00
  • e742da8b28 Merged ComputeCpp into default. Luke Iwanski 2016-10-14 13:36:51 +01:00
  • 524fa4c46f Reducing the code by generalising sycl backend functions/structs. Mehdi Goli 2016-10-14 12:09:55 +01:00
  • 014d9f1d9b implement euler angles with the right ranges Hongkai Dai 2016-10-13 14:45:51 -07:00
  • 737e4152c3 Merged in lukier/eigen (pull request PR-234) Benoit Steiner 2016-10-13 18:09:28 +00:00
  • d0ee2267d6 Relaxed the resizing checks so that they don't fail with gcc >= 5.3 Benoit Steiner 2016-10-13 10:59:46 -07:00
  • a94791b69a Fixes for min and abs after Benoit's comments, switched to numext. Robert Lukierski 2016-10-13 15:00:22 +01:00
  • ac63d6891c Patch to allow VS2015 & CUDA 8.0 to compile with Eigen included. I'm not sure whether to limit the check to this compiler combination ( || (EIGEN_COMP_MSVC == 1900 && __CUDACC_VER__) ) or to leave it as it is. I also don't know if this will have any affect on including Eigen in device code (I'm not in my current project). Avi Ginsburg 2016-10-13 08:47:32 +00:00
  • 7e4a6754b2 Merged eigen/eigen into default Benoit Steiner 2016-10-12 22:42:33 -07:00
  • 38b6048e14 Deleted redundant implementation of predux Benoit Steiner 2016-10-12 14:37:56 -07:00
  • e74612b9a0 Remove double ;; Gael Guennebaud 2016-10-12 22:49:47 +02:00
  • 78d2926508 Merged eigen/eigen into default Benoit Steiner 2016-10-12 13:46:29 -07:00
  • 2e2f48e30e Take advantage of AVX512 instructions whenever possible to speedup the processing of 16 bit floats. Benoit Steiner 2016-10-12 13:45:39 -07:00
  • 03fd417f66 Fix SPQR for rectangular matrices (grafted from f939c351cb ) Gael Guennebaud 2016-10-12 22:39:33 +02:00
  • f939c351cb Fix SPQR for rectangular matrices Gael Guennebaud 2016-10-12 22:39:33 +02:00
  • 091d373ee9 Fix outer-stride. Gael Guennebaud 2016-10-12 21:47:52 +02:00
  • 471075f7ad Fixes min() warnings. Robert Lukierski 2016-10-12 18:59:05 +01:00
  • 5c366fe1d7 Merged in rmlarsen/eigen (pull request PR-230) Gael Guennebaud 2016-10-12 16:30:51 +00:00
  • 86711497c4 Adding EIGEN_DEVICE_FUNC in the Geometry module. Additional CUDA necessary fixes in the Core (mostly usage of EIGEN_USING_STD_MATH). Robert Lukierski 2016-10-12 16:35:17 +01:00
  • 47150af1c8 Fix copy-paste error: Must use _mm256_cmp_ps for AVX. Rasmus Munk Larsen 2016-10-12 08:34:39 -07:00
  • 89e315152c bug #1325: fix compilation on NEON with clang Gael Guennebaud 2016-10-12 16:55:47 +02:00
  • 7f0599b6eb Manually define int16_t and uint16_t when compiling with Visual Studio Benoit Steiner 2016-10-08 22:56:32 -07:00
  • 5727e4d89c Reenabled the use of variadic templates on tegra x1 provides that the latest version (i.e. JetPack 2.3) is used. Benoit Steiner 2016-10-08 22:19:03 +00:00
  • 5266ff8966 Cleaned up a regression test Benoit Steiner 2016-10-08 19:12:44 +00:00
  • 5c68051cd7 Merge the content of the ComputeCpp branch into the default branch Benoit Steiner 2016-10-07 11:04:16 -07:00
  • 4860727ac2 Remove static qualifier of free-functions (inline is enough and this helps ICC to find the right overload) Gael Guennebaud 2016-10-07 09:21:12 +02:00
  • 507b661106 Renamed predux_half into predux_downto4 Benoit Steiner 2016-10-06 17:57:04 -07:00
  • a498ff7df6 Fixed incorrect comment Benoit Steiner 2016-10-06 15:27:27 -07:00
  • 8ba3c41fcf Revergted unecessary change Benoit Steiner 2016-10-06 15:12:15 -07:00
  • a7473d6d5a Fixed compilation error with gcc >= 5.3 Benoit Steiner 2016-10-06 14:33:22 -07:00
  • 5e64cea896 Silenced a compilation warning Benoit Steiner 2016-10-06 14:24:17 -07:00
  • 33fba3f08d Merged in rryan/eigen/tensorfunctors (pull request PR-233) Benoit Steiner 2016-10-06 12:29:19 -07:00
  • bfc264abe8 Add a test that GPU complex product reductions match CPU reductions. RJ Ryan 2016-10-06 11:10:14 -07:00
  • e2e9cdd169 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 2016-10-06 10:49:48 -07:00
  • d485d12c51 Added missing AVX intrinsics for fp16: in particular, implemented predux which is required by the matrix-vector code. Benoit Steiner 2016-10-06 10:41:03 -07:00
  • 48c635e223 Add a simple cost model to prevent Eigen's parallel GEMM from using too many threads when the inner dimension is small. Rasmus Munk Larsen 2016-10-06 10:33:10 -07:00
  • 9f3276981c Enabling AVX512 should also enable AVX2. Benoit Steiner 2016-10-06 10:29:48 -07:00
  • 80b5133789 Fix compilation of qr.inverse() for column and full pivoting variants. Gael Guennebaud 2016-10-06 09:55:50 +02:00
  • 4131074818 Deleted unecessary CMakeLists.txt file Benoit Steiner 2016-10-05 18:54:35 -07:00
  • cb5cd69872 Silenced a compilation warning. Benoit Steiner 2016-10-05 18:50:53 -07:00
  • 78b569f685 Merged latest updates from trunk Benoit Steiner 2016-10-05 18:48:55 -07:00
  • 9c2b6c049b Silenced a few compilation warnings Benoit Steiner 2016-10-05 18:37:31 -07:00
  • 6f3cd529af Pulled latest updates from trunk Benoit Steiner 2016-10-05 18:31:43 -07:00
  • d7f9679a34 Fixed a couple of compilation warnings Benoit Steiner 2016-10-05 15:00:32 -07:00
  • ae1385c7e4 Pull the latest updates from trunk Benoit Steiner 2016-10-05 14:54:36 -07:00
  • 73b0012945 Fixed compilation warnings Benoit Steiner 2016-10-05 14:24:24 -07:00
  • c84084c0c0 Fixed compilation warning Benoit Steiner 2016-10-05 14:15:41 -07:00
  • 4387433acf Increased the robustness of the reduction tests on fp16 Benoit Steiner 2016-10-05 10:42:41 -07:00
  • aad20d700d Increase the tolerance to numerical noise. Benoit Steiner 2016-10-05 10:39:24 -07:00
  • 8b69d5d730 ::rand() returns a signed integer on win32 Benoit Steiner 2016-10-05 08:55:02 -07:00
  • ed7a220b04 Fixed a typo that impacts windows builds Benoit Steiner 2016-10-05 08:51:31 -07:00
  • ceee1c008b Silenced compilation warning Benoit Steiner 2016-10-04 18:47:53 -07:00
  • 698ff69450 Properly characterize the CUDA packet primitives for fp16 as device only Benoit Steiner 2016-10-04 16:53:30 -07:00
  • 7f67e6dfdb Update comment for fast sqrt. Rasmus Munk Larsen 2016-10-04 15:09:11 -07:00
  • 765615609d Update comment for fast sqrt. Rasmus Munk Larsen 2016-10-04 15:08:41 -07:00
  • 3ed67cb0bb Fix a bug in the implementation of Carmack's fast sqrt algorithm in Eigen (enabled by EIGEN_FAST_MATH), which causes the vectorized parts of the computation to return -0.0 instead of NaN for negative arguments. Rasmus Munk Larsen 2016-10-04 14:22:56 -07:00
  • 6af5ac7e27 Cleanup the cuda executor code. Benoit Steiner 2016-10-04 08:52:13 -07:00
  • 2f6d1607c8 Cleaned up the random number generation code. Benoit Steiner 2016-10-04 08:38:23 -07:00
  • 881b90e984 Use explicit type casting to generate packets of zeros. Benoit Steiner 2016-10-04 08:23:38 -07:00
  • 91207cbae3 Added tag 3.2.10 for changeset 1c9aa054c7 Gael Guennebaud 2016-10-04 09:21:27 +02:00
  • 1c9aa054c7 bump to 3.2.10 3.2.10 Gael Guennebaud 2016-10-04 09:21:16 +02:00