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

  • b5df8cabd7 fix hip-clang compilation due to new HIP scalar accessor Jeff Daily 2020-01-20 21:08:52 +00:00
  • 6d284bb1b7 Fix for HIP breakage - 200115. Adding a missing EIGEN_DEVICE_FUNC attr Deven Desai 2020-01-16 00:51:43 +00:00
  • f6c6de5d63 Ensure Igamma does not NaN or Inf for large values. Srinivas Vasudevan 2020-01-14 21:32:48 +00:00
  • 6601abce86 Remove rogue include in TypeCasting.h. Meta.h is already included by the top-level header in Eigen/Core. Rasmus Munk Larsen 2020-01-14 21:03:53 +00:00
  • b9362fb8f7 Convert StridedLinearBufferCopy::Kind to enum class Eugene Zhulenev 2020-01-13 11:43:24 -08:00
  • 5a8b97b401 Switching unpacket_traits<Packet4i> to vectorizable=true. Everton Constantino 2020-01-13 12:07:58 -03:00
  • 42838c28b8 Adding correct cache sizes for PPC architecture. Everton Constantino 2020-01-13 16:58:14 +00:00
  • 1d0c45122a Removing executable bit from file mode Christoph Hertzberg 2020-01-11 15:02:29 +01:00
  • 35219cea68 Bug #1790: Make areApprox check numext::isnan instead of bitwise equality (NaNs don't have to be bitwise equal). Christoph Hertzberg 2020-01-11 14:57:22 +01:00
  • 2e099e8d8f Added special_packetmath test and tweaked bounds on tests. Refactor shared packetmath code to header file. (Squashed from PR !38) Srinivas Vasudevan 2020-01-11 10:31:21 +00:00
  • e1ecfc162d call Explicitly ::rint and ::rintf for targets without c++11. Without this, the Windows build breaks when trying to compile numext::rint<double>. Rasmus Munk Larsen 2020-01-10 21:14:08 +00:00
  • da5a7afed0 Improvements to the tidiness and completeness of the NEON implementation Joel Holdsworth 2020-01-10 18:31:15 +00:00
  • 452371cead Fix for gcc build error when using Eigen headers with AVX512 Anuj Rawat 2020-01-09 11:50:50 -08:00
  • 601f89dfd0 Adding RInt vector support for SYCL. mehdi-goli 2020-01-10 17:55:15 +00:00
  • 2ea5a715cf Properly initialize b vector in SplineFitting Matthew Powelson 2020-01-07 16:20:43 -06:00
  • 9254974115 Don't add EIGEN_DEVICE_FUNC to random() since ::rand is not available in Cuda. Rasmus Munk Larsen 2020-01-09 21:23:09 +00:00
  • a3ec89b5bd Add missing EIGEN_DEVICE_FUNC annotations in MathFunctions.h. Rasmus Munk Larsen 2020-01-09 21:06:34 +00:00
  • 8333e03590 Use data.data() instead of &data (since it is not obvious that Array is trivially copyable) Christoph Hertzberg 2020-01-09 11:38:19 +01:00
  • e6fcee995b Don't use the rational approximation to the logistic function on GPUs as it appears to be slightly slower. Rasmus Munk Larsen 2020-01-09 00:04:26 +00:00
  • 4217a9f090 The upper limits for where to use the rational approximation to the logistic function were not set carefully enough in the original commit, and some arguments would cause the function to return values greater than 1. This change set the versions found by scanning all floating point numbers (using std::nextafterf()). Rasmus Munk Larsen 2020-01-08 22:21:37 +00:00
  • 9623c0c4b9 Fix formatting Christoph Hertzberg 2020-01-08 13:58:18 +01:00
  • 19876ced76 Bug #1785: Introduce numext::rint. Ilya Tokar 2019-12-16 16:00:35 -05:00
  • d0ae052da4 [SYCL Backend] * Adding Missing operations for vector comparison in SYCL. This caused compiler error for vector comparison when compiling SYCL * Fixing the compiler error for placement new in TensorForcedEval.h This caused compiler error when compiling SYCL backend * Reducing the SYCL warning by removing the abort function inside the kernel * Adding Strong inline to functions inside SYCL interop. mehdi-goli 2020-01-07 15:13:37 +00:00
  • eedb7eeacf Protecting integer_types's long long test with a check to see if we have CXX11 support. Everton Constantino 2020-01-07 14:35:35 +00:00
  • bcbaad6d87 Bug #1800: Guard against misleading indentation Christoph Hertzberg 2020-01-03 13:47:43 +01:00
  • 00de570793 Fix -Werror -Wfloat-conversion warning. Janek Kozicki 2019-12-23 23:52:44 +01:00
  • 636e2bb3fa Fix for HIP breakage - 191220 Deven Desai 2019-12-20 21:28:00 +00:00
  • 89449a0821 Remove .hgignore and copy .gitignore from master branch Christoph Hertzberg 2019-12-20 18:25:47 +01:00
  • e1e35a2246 Bug #1796: Make matrix squareroot usable for Map and Ref types Christoph Hertzberg 2019-12-20 18:15:35 +01:00
  • 1e9664b147 Bug #1796: Make matrix squareroot usable for Map and Ref types Christoph Hertzberg 2019-12-20 18:10:22 +01:00
  • d86544d654 Reduce code duplication and avoid confusing Doxygen Christoph Hertzberg 2019-12-19 19:48:39 +01:00
  • dde279f57d Hide recursive meta templates from Doxygen Christoph Hertzberg 2019-12-19 19:47:23 +01:00
  • c21771ac04 Use double-braces initialization (as everywhere else in the test-suite). Christoph Hertzberg 2019-12-19 19:20:48 +01:00
  • a3273aeff8 Fix trivial shadow warning Christoph Hertzberg 2019-12-19 19:13:11 +01:00
  • 870e53c0f2 Bug #1788: Fix rule-of-three violations inside the stable modules. This fixes deprecated-copy warnings when compiling with GCC>=9 Also protect some additional Base-constructors from getting called by user code code (#1587) Christoph Hertzberg 2019-12-19 17:30:11 +01:00
  • 6965f6de7f Fix unit-test which I broke in previous fix Christoph Hertzberg 2019-12-19 13:42:14 +01:00
  • 7a65219a2e Fix TensorPadding bug in squeezed reads from inner dimension Eugene Zhulenev 2019-12-19 05:43:57 +00:00
  • 73e55525e5 Return const data pointer from TensorRef evaluator.data() Eugene Zhulenev 2019-12-18 23:19:36 +00:00
  • ae07801dd8 Tensor block evaluation cost model Eugene Zhulenev 2019-12-18 20:07:00 +00:00
  • 72166d0e6e Fix some maybe-unitialized warnings Christoph Hertzberg 2019-12-18 18:26:20 +01:00
  • 5a3eaf88ac Workaround class-memaccess warnings on newer GCC versions Christoph Hertzberg 2019-12-18 16:37:26 +01:00
  • de07c4d1c2 fix compilation due to new HIP scalar accessor Jeff Daily 2019-12-11 22:00:55 +00:00
  • 788bef6ab5 Reduce block evaluation overhead for small tensor expressions Eugene Zhulenev 2019-12-17 19:06:14 +00:00
  • 7252163335 Add default definition for EIGEN_PREDICT_* Rasmus Munk Larsen 2019-12-16 22:31:59 +00:00
  • a566074480 Improve accuracy of fast approximate tanh and the logistic functions in Eigen, such that they preserve relative accuracy to within a few ULPs where their function values tend to zero (around x=0 for tanh, and for large negative x for the logistic function). Rasmus Munk Larsen 2019-12-16 21:33:42 +00:00
  • 8e5da71466 Resolve double-promotion warnings when compiling with clang. sin was calling sin(double) instead of std::sin(float) Christoph Hertzberg 2019-12-13 22:46:40 +01:00
  • 9b7a2b43c2 Renamed .hgignore to .gitignore (removing hg-specific "syntax" line) Christoph Hertzberg 2019-12-13 19:40:57 +01:00
  • 06e99aaf40 Bug 1785: fix pround on x86 to use the same rounding mode as std::round. Ilya Tokar 2019-12-12 14:04:56 -05:00
  • 73a8d572f5 Clamp tanh approximation outside [-c, c] where c is the smallest value where the approximation is exactly +/-1. Without FMA, c = 7.90531110763549805, with FMA c = 7.99881172180175781. Rasmus Munk Larsen 2019-12-12 19:34:25 +00:00
  • 88062b7fed Fix implementation of complex expm1. Add tests that fail with previous implementation, but pass with the current one. Srinivas Vasudevan 2019-12-12 01:56:54 +00:00
  • 381f8f3139 Initialize non-trivially constructible types when allocating a temp buffer. Eugene Zhulenev 2019-12-12 01:31:30 +00:00
  • 64272c7f40 Squeeze reads from two inner dimensions in TensorPadding Eugene Zhulenev 2019-12-11 16:54:51 -08:00
  • 963ba1015b Add back accidentally deleted default constructor to TensorExecutorTilingContext. Eugene Zhulenev 2019-12-11 18:47:55 +00:00
  • 1b6e0395e6 Added io test Joel Holdsworth 2019-11-25 15:23:55 +00:00
  • 3c0ef9f394 IO: Fixed printing of char and unsigned char matrices Joel Holdsworth 2019-11-05 12:17:58 +00:00
  • e87af0ed37 Added Eigen::numext typedefs for uint8_t, int8_t, uint16_t and int16_t Joel Holdsworth 2019-11-25 13:59:25 +00:00
  • 15b3bcfca0 Bug 1786: fix compilation with MSVC Gael Guennebaud 2019-12-11 16:16:38 +01:00
  • c9220c035f Remove block memory allocation required by removed block evaluation API Eugene Zhulenev 2019-12-10 17:15:55 -08:00
  • 1c879eb010 Remove V2 suffix from TensorBlock Eugene Zhulenev 2019-12-10 15:40:23 -08:00
  • dbca11e880 Remove TensorBlock.h and old TensorBlock/BlockMapper Eugene Zhulenev 2019-12-10 11:58:30 -08:00
  • c49f0d851a Fix for HIP breakage detected on 191210 Deven Desai 2019-12-10 22:14:05 +00:00
  • 2918f85ba9 Do not use std::vector in getResourceRequirements Eugene Zhulenev 2019-12-09 16:19:38 -08:00
  • 8056a05b54 Undo the block size change. Artem Belevich 2019-12-09 11:10:29 -08:00
  • dbb703d44e Add async evaluation support to TensorSelectOp Eugene Zhulenev 2019-12-09 18:36:13 +00:00
  • 11d6465326 fix AlignedVector3 inconsisent interface with other Vector classes, default constructor and operator- were missing. Janek Kozicki 2019-12-05 17:18:05 +01:00
  • bb7ccac3af Add recursive work splitting to EvalShardedByInnerDimContext Eugene Zhulenev 2019-12-05 14:50:19 -08:00
  • 25230d1862 Improve performance of contraction kernels Artem Belevich 2019-12-05 12:48:34 -08:00
  • 08eeb648ea update hg to git hashes Gael Guennebaud 2019-12-05 16:33:24 +01:00
  • 366cf005b0 Add missing initialization in cxx11_tensor_trace.cpp. Rasmus Munk Larsen 2019-12-04 23:56:37 +00:00
  • c488b8b32f Replace calls to "hg" by calls to "git" Gael Guennebaud 2019-12-04 11:24:06 +01:00
  • 8fbe0e4699 Update old links to bitbucket to point to gitlab.com Gael Guennebaud 2019-12-04 10:57:07 +01:00
  • 114a15c66a Added tag before-git-migration for changeset a7c7d329d8 Gael Guennebaud 2019-12-04 10:06:00 +01:00
  • a7c7d329d8 Merged in ezhulenev/eigen-01 (pull request PR-769) before-git-migration Rasmus Larsen 2019-12-04 00:49:10 +00:00
  • cacf433975 Merged in anshuljl/eigen-2/Anshul-Jaiswal/update-configurevectorizationh-to-not-op-1573079916090 (pull request PR-754) Rasmus Larsen 2019-12-04 00:45:42 +00:00
  • 8f4536e852 Capture TensorMap by value inside tensor expression AST Eugene Zhulenev 2019-12-03 16:39:05 -08:00
  • 4e696901f8 Remove __host__ annotation for device-only function. Rasmus Munk Larsen 2019-12-03 14:33:19 -08:00
  • ead81559c8 Use EIGEN_DEVICE_FUNC macro instead of __device__. Rasmus Munk Larsen 2019-12-03 12:08:22 -08:00
  • 5f1082d0b1 Fix QuaternionBase::cast for quaternion map and wrapper. (grafted from 6358599ecb ) Gael Guennebaud 2019-12-03 14:51:14 +01:00
  • 6358599ecb Fix QuaternionBase::cast for quaternion map and wrapper. Gael Guennebaud 2019-12-03 14:51:14 +01:00
  • 7745f69013 bug #1776: fix vector-wise STL iterator's operator-> using a proxy as pointer type. This changeset fixes also the value_type definition. Gael Guennebaud 2019-12-03 14:40:15 +01:00
  • 66f07efeae Revert the specialization for scalar_logistic_op<float> introduced in: Rasmus Munk Larsen 2019-12-02 17:00:58 -08:00
  • 3b15373bb3 Merged in ezhulenev/eigen-02 (pull request PR-767) Rasmus Larsen 2019-12-02 18:23:11 +00:00
  • 312c8e77ff Fix for the HIP build+test errors. Deven Desai 2019-12-02 17:41:32 +00:00
  • 956131d0e6 Merged in codeplaysoftware/eigen/SYCL-Backend (pull request PR-691) Rasmus Larsen 2019-11-28 16:19:25 +00:00
  • 00f32752f7 [SYCL] Rebasing the SYCL support branch on top of the Einge upstream master branch. * Unifying all loadLocalTile from lhs and rhs to an extract_block function. * Adding get_tensor operation which was missing in TensorContractionMapper. * Adding the -D method missing from cmake for Disable_Skinny Contraction operation. * Wrapping all the indices in TensorScanSycl into Scan parameter struct. * Fixing typo in Device SYCL * Unifying load to private register for tall/skinny no shared * Unifying load to vector tile for tensor-vector/vector-tensor operation * Removing all the LHS/RHS class for extracting data from global * Removing Outputfunction from TensorContractionSkinnyNoshared. * Combining the local memory version of tall/skinny and normal tensor contraction into one kernel. * Combining the no-local memory version of tall/skinny and normal tensor contraction into one kernel. * Combining General Tensor-Vector and VectorTensor contraction into one kernel. * Making double buffering optional for Tensor contraction when local memory is version is used. * Modifying benchmark to accept custom Reduction Sizes * Disabling AVX optimization for SYCL backend on the host to allow SSE optimization to the host * Adding Test for SYCL * Modifying SYCL CMake Mehdi Goli 2019-11-28 10:08:54 +00:00
  • 82a47338df Fix shadow warnings in AlignedBox and SparseBlock Eugene Zhulenev 2019-11-27 16:22:27 -08:00
  • ea51a9eace Add missing EIGEN_DEVICE_FUNC attribute to template specializations for pexp to fix GPU build. Rasmus Munk Larsen 2019-11-27 10:17:09 -08:00
  • 5a3ebda36b Fix warning due to missing cast for exponent arguments for std::frexp and std::lexp. Rasmus Munk Larsen 2019-11-26 16:18:29 -08:00
  • 2df57be856 Merged in realjhol/eigen/fix-warnings (pull request PR-760) Rasmus Larsen 2019-11-26 23:24:23 +00:00
  • 5496d0da0b Add async evaluation support to TensorReverse Eugene Zhulenev 2019-11-26 15:02:24 -08:00
  • bc66c88255 Add async evaluation support to TensorPadding/TensorImagePatch/TensorShuffling Eugene Zhulenev 2019-11-26 11:41:57 -08:00
  • 1039348f12 backport is_valid_index_type fix from default branch and c++03 Gael Guennebaud 2019-11-22 13:49:17 +01:00
  • a8d516b04e bug #1281: fix AutoDiffScalar's make_coherent for nested expression of constant ADs. (grafted from b9837ca9ae ) Gael Guennebaud 2019-11-14 14:58:08 +01:00
  • f1b1f13d3c silent cmake warnings Gael Guennebaud 2019-11-14 14:22:35 +01:00
  • a153dbae9b Fix case issue with Lapack unit tests (grafted from 0fb6e24408 ) Gael Guennebaud 2019-11-14 14:16:05 +01:00
  • 3d7e2a1f3d Fix possible conflict with an externally defined "real" type. This issue was detected by gcc5 only. Gael Guennebaud 2019-11-14 13:31:26 +01:00
  • e1e9b3f9f6 Swicth the 3.3 branch to my.cdash.org and add the "-v3.3" suffix to build name so that cdash can filter them to the proper groups. Gael Guennebaud 2019-11-14 10:46:02 +01:00
  • 86eb41f1cb SparseRef: Fixed alignment warning on ARM GCC Joel Holdsworth 2019-11-07 14:34:06 +00:00
  • c1a67cb5af Update ConfigureVectorization.h to not optimize fp16 routines when compiling with cuda. Anshul Jaiswal 2019-11-06 22:40:38 +00:00
  • 743c925286 test/packetmath: Silence alignment warnings Joel Holdsworth 2019-11-05 19:06:12 +00:00