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

  • e6c183f8fd Fix doc issues regarding ndtri Gael Guennebaud 2019-09-04 23:00:21 +02:00
  • 5702a57926 Fix possible warning regarding strict equality comparisons Gael Guennebaud 2019-09-04 22:57:04 +02:00
  • e38dd48a27 PR 681: Add ndtri function, the inverse of the normal distribution function. Srinivas Vasudevan 2019-08-12 19:26:29 -04:00
  • f59bed7a13 Change typedefs from private to protected to fix MSVC compilation Eugene Zhulenev 2019-09-03 19:11:36 -07:00
  • 47fefa235f Allow move-only done callback in TensorAsyncDevice Eugene Zhulenev 2019-09-03 17:20:56 -07:00
  • a8d264fa9c Add test for const TensorMap underlying data mutation Eugene Zhulenev 2019-09-03 11:38:39 -07:00
  • f68f2bba09 TensorMap constness should not change underlying storage constness Eugene Zhulenev 2019-09-03 11:08:09 -07:00
  • 8e7e3d9bc8 Makes Scalar/RealScalar typedefs public in Pardiso's wrappers (see PR 688) Gael Guennebaud 2019-09-03 13:09:03 +02:00
  • c694be1214 Fixed Tensor documentation formatting. Alberto Luaces 2019-07-23 09:24:06 +00:00
  • 15f3d9d272 More colamd cleanup: - Move colamd implementation in its own namespace to avoid polluting the internal namespace with Ok, Status, etc. - Fix signed/unsigned warning - move some ugly free functions as member functions Gael Guennebaud 2019-09-03 00:50:51 +02:00
  • a4d1a6cd7d Eigen_Colamd.h updated to replace constexpr with consts and enums. Anshul Jaiswal 2019-08-17 05:29:23 +00:00
  • 283558face Ordering.h edited to fix dependencies on Eigen_Colamd.h Anshul Jaiswal 2019-08-15 20:21:56 +00:00
  • 39f30923c2 Eigen_Colamd.h edited replacing macros with constexprs and functions. Anshul Jaiswal 2019-08-15 20:15:19 +00:00
  • 0a6b553ecf Eigen_Colamd.h edited online with Bitbucket replacing constant #defines with const definitions Anshul Jaiswal 2019-07-21 04:53:31 +00:00
  • fab51d133e Updated Eigen_Colamd.h, namespacing macros ALIVE & DEAD as COLAMD_ALIVE & COLAMD_DEAD to prevent conflicts with other libraries / code. Anshul Jaiswal 2019-06-08 21:09:06 +00:00
  • 79c402e40e Fix shadow warnings in TensorContractionThreadPool Eugene Zhulenev 2019-08-30 15:38:31 -07:00
  • edf2ec28d8 Fix block mapper type name in TensorExecutor Eugene Zhulenev 2019-08-30 15:29:25 -07:00
  • f0b36fb9a4 evalSubExprsIfNeededAsync + async TensorContractionThreadPool Eugene Zhulenev 2019-08-30 15:13:38 -07:00
  • 619cea9491 Revert accidentally removed <memory> header from ThreadPool Eugene Zhulenev 2019-08-30 14:51:17 -07:00
  • 66665e7e76 Asynchronous expression evaluation with TensorAsyncDevice Eugene Zhulenev 2019-08-30 14:49:40 -07:00
  • f6c51d9209 Fix missing header inclusion and colliding definitions for half type casting, which broke build with -march=native on Haswell/Skylake. Rasmus Munk Larsen 2019-08-30 14:03:29 -07:00
  • bc40d4522c Const correctness in TensorMap<const Tensor<T, ...>> expressions Eugene Zhulenev 2019-08-28 17:46:05 -07:00
  • 1187bb65ad Add more tests for corner cases of log1p and expm1. Add handling of infinite arguments to log1p such that log1p(inf) = inf. Rasmus Munk Larsen 2019-08-28 12:20:21 -07:00
  • 6e77f9bef3 Remove shadow warnings in TensorDeviceThreadPool Eugene Zhulenev 2019-08-28 10:32:19 -07:00
  • 9aba527405 Revert changes to std_falback::log1p that broke handling of arguments less than -1. Fix packet op accordingly. Rasmus Munk Larsen 2019-08-27 15:35:29 -07:00
  • b021cdea6d Clean up float16 a.k.a. Eigen::half support in Eigen. Move the definition of half to Core/arch/Default and move arch-specific packet ops to their respective sub-directories. Rasmus Munk Larsen 2019-08-27 11:30:31 -07:00
  • 84fefdf321 Merged in ezhulenev/eigen-01 (pull request PR-683) Rasmus Larsen 2019-08-26 21:49:17 +00:00
  • 8b5ab0e4dd Fix get_random_seed on Native Client maratek 2019-08-23 15:25:56 -07:00
  • 6901788013 Asynchronous parallelFor in Eigen ThreadPoolDevice Eugene Zhulenev 2019-08-22 10:50:51 -07:00
  • 2fb24384c9 Merged in jaopaulolc/eigen (pull request PR-679) Christoph Hertzberg 2019-08-22 15:57:33 +00:00
  • 57f6b62597 Merged in rmlarsen/eigen (pull request PR-680) Rasmus Larsen 2019-08-22 00:25:29 +00:00
  • 071311821e Remove XSMM support from Tensor module Eugene Zhulenev 2019-08-19 11:44:25 -07:00
  • 5ac7984ffa Fix debug macros in p{load,store}u João P. L. de Carvalho 2019-08-14 11:59:12 -06:00
  • db9147ae40 Add missing pcmp_XX methods for double/Packet2d João P. L. de Carvalho 2019-08-14 10:37:39 -06:00
  • a3298b22ec Implement vectorized versions of log1p and expm1 in Eigen using Kahan's formulas, and change the scalar implementations to properly handle infinite arguments. Rasmus Munk Larsen 2019-08-12 13:53:28 -07:00
  • 787f6ef025 Fix packed load/store for PowerPC's VSX João P. L. de Carvalho 2019-08-09 16:02:55 -06:00
  • 4d29aa0294 Fix offset argument of ploadu/pstoreu for Altivec João P. L. de Carvalho 2019-08-09 15:59:26 -06:00
  • 66d073c38e bug #1718: Add cast to successfully compile with clang on PowerPC João P. L. de Carvalho 2019-08-09 15:56:26 -06:00
  • d55d392e7b Fix bugs in log1p and expm1 where repeated using statements would clobber each other. Add specializations for complex types since std::log1p and std::exp1m do not support complex. Rasmus Munk Larsen 2019-08-08 16:27:32 -07:00
  • 85928e5f47 Guard against repeated definition of EIGEN_MPL2_ONLY Rasmus Munk Larsen 2019-08-07 14:19:00 -07:00
  • facc4e4536 Disable tests for contraction with output kernels when using libxsmm, which does not support this. Rasmus Munk Larsen 2019-08-07 14:11:15 -07:00
  • eab7e52db2 [Eigen] Vectorize evaluation of coefficient-wise functions over tensor blocks if the strides are known to be 1. Provides up to 20-25% speedup of the TF cross entropy op with AVX. Rasmus Munk Larsen 2019-08-07 12:57:42 -07:00
  • 0987126165 Clean up unnecessary namespace specifiers in TensorBlock.h. Rasmus Munk Larsen 2019-08-07 12:12:52 -07:00
  • 0050644b23 Fix doc regarding alignment and c++17 Gael Guennebaud 2019-08-04 01:09:41 +02:00
  • e2999d4c38 Fix performance regressions due to https://bitbucket.org/eigen/eigen/pull-requests/662. Rasmus Munk Larsen 2019-08-02 11:18:13 -07:00
  • f22b7283a3 Added leading asterisk for Doxygen to consume as it was removing asterisk intended to be part of the code. Kyle Vedder 2019-07-18 18:12:14 +00:00
  • 6e17491f45 Fix typo in Umeyama method documentation Michael Grupp 2019-07-17 11:20:41 +00:00
  • e0f5a2a456 Remove {} accidentally added in previous commit Christoph Hertzberg 2019-07-18 20:22:17 +02:00
  • ea6d7eb32f Move variadic constructors outside #ifndef EIGEN_PARSED_BY_DOXYGEN block, to make it actually appear in the generated documentation. Christoph Hertzberg 2019-07-12 19:46:37 +02:00
  • 9237883ff1 Escape \# inside doxygen docu Christoph Hertzberg 2019-07-12 19:45:13 +02:00
  • c2671e5315 Build deprecated snippets with -DEIGEN_NO_DEPRECATED_WARNING Also, document LinSpaced only where it is implemented Christoph Hertzberg 2019-07-12 19:43:32 +02:00
  • 3cd148f983 Fix expression evaluation heuristic for TensorSliceOp Eugene Zhulenev 2019-07-09 12:10:26 -07:00
  • 23b958818e Fix compiler for unsigned integers. Rasmus Munk Larsen 2019-07-09 11:18:25 -07:00
  • 6083014594 Add outer/inner chipping optimization for chipping dimension specified at runtime Eugene Zhulenev 2019-07-03 11:35:25 -07:00
  • 7eb2e0a95b adding the EIGEN_DEVICE_FUNC attribute to the constCast routine. Deven Desai 2019-07-02 20:02:46 +00:00
  • ef8aca6a89 Merged in codeplaysoftware/eigen (pull request PR-667) Gael Guennebaud 2019-07-02 12:45:23 +00:00
  • 4ac93f8edc Allocate non-const scalar buffer for block evaluation with DefaultDevice Eugene Zhulenev 2019-07-01 10:55:19 -07:00
  • 9ea490c82c [SYCL] : * Modifying TensorDeviceSYCL to use EIGEN_THROW_X. * Modifying TensorMacro to use EIGEN_TRY/CATCH(X) macro. * Modifying TensorReverse.h to use EIGEN_DEVICE_REF instead of &. * Fixing the SYCL device macro in SpecialFunctionsImpl.h. Mehdi Goli 2019-07-01 16:27:28 +01:00
  • ffaf658ecd PR 655: Fix missing Eigen namespace in Macros Justin Carpentier 2019-06-05 09:51:59 +02:00
  • 0b24e1cb5c [SYCL] Adding the SYCL memory model. The SYCL memory model provides : * an interface for SYCL buffers to behave as a non-dereferenceable pointer * an interface for placeholder accessor to behave like a pointer on both host and device Mehdi Goli 2019-07-01 16:02:30 +01:00
  • 81a03bec75 Fix TensorReverse on GPU with m_stride[i]==0 Eugene Zhulenev 2019-06-28 15:50:39 -07:00
  • 8053eeb51e Fix CUDA compilation error for pselect<half>. Rasmus Munk Larsen 2019-06-28 12:07:29 -07:00
  • 74a9dd1102 Fix preprocessor condition to only generate a warning when calling eigen::GpuDevice::synchronize() from device code, but not when calling from a non-GPU compilation unit. Rasmus Munk Larsen 2019-06-28 11:56:21 -07:00
  • 70d4020ad9 Remove comma causing warning in c++03 mode. Rasmus Munk Larsen 2019-06-28 11:39:45 -07:00
  • 6e7c76481a Merge with Eigen head Eugene Zhulenev 2019-06-28 11:22:46 -07:00
  • 878845cb25 Add block access to TensorReverseOp and make sure that TensorForcedEval uses block access when preferred Eugene Zhulenev 2019-06-28 11:13:44 -07:00
  • 1f61aee5ca [SYCL] This PR adds the minimum modifications to the Eigen unsupported module required to run it on devices supporting SYCL. * Abstracting the pointer type so that both SYCL memory and pointer can be captured. * Converting SYCL virtual pointer to SYCL device memory in Eigen evaluator class. * Binding SYCL placeholder accessor to command group handler by using bind method in Eigen evaluator node. * Adding SYCL macro for controlling loop unrolling. * Modifying the TensorDeviceSycl.h and SYCL executor method to adopt the above changes. Rasmus Munk Larsen 2019-06-28 10:11:56 -07:00
  • 7d08fa805a [SYCL] This PR adds the minimum modifications to the Eigen unsupported module required to run it on devices supporting SYCL. * Abstracting the pointer type so that both SYCL memory and pointer can be captured. * Converting SYCL virtual pointer to SYCL device memory in Eigen evaluator class. * Binding SYCL placeholder accessor to command group handler by using bind method in Eigen evaluator node. * Adding SYCL macro for controlling loop unrolling. * Modifying the TensorDeviceSycl.h and SYCL executor method to adopt the above changes. Mehdi Goli 2019-06-28 10:08:23 +01:00
  • 16a56b2ddd [SYCL] This PR adds the minimum modifications to Eigen core required to run Eigen unsupported modules on devices supporting SYCL. * Adding SYCL memory model * Enabling/Disabling SYCL backend in Core * Supporting Vectorization Mehdi Goli 2019-06-27 12:25:09 +01:00
  • adec097c61 Remove extra comma (causes warnings in C++03) Christoph Hertzberg 2019-06-26 16:14:28 +02:00
  • 229db81572 Optimize evaluation strategy for TensorSlicingOp and TensorChippingOp Eugene Zhulenev 2019-06-25 15:41:37 -07:00
  • ba506d5bd2 fix for a ROCm/HIP specificcompile errror introduced by a recent commit. Deven Desai 2019-06-22 00:06:05 +00:00
  • c9394d7a0e Remove extra "one" in comment. Rasmus Munk Larsen 2019-06-20 16:23:19 -07:00
  • b8f8dac4eb Update comment as suggested by tra@google.com. Rasmus Munk Larsen 2019-06-20 16:18:37 -07:00
  • e5e63c2cad Fix grammar. Rasmus Munk Larsen 2019-06-20 16:03:59 -07:00
  • 302a404b7e Added comment explaining the surprising EIGEN_COMP_CLANG && !EIGEN_COMP_NVCC clause. Rasmus Munk Larsen 2019-06-20 15:59:08 -07:00
  • b5237f53b1 Fix CUDA build on Mac. Rasmus Munk Larsen 2019-06-20 15:44:14 -07:00
  • 988f24b730 Various fixes for packet ops. 1. Fix buggy pcmp_eq and unit test for half types. 2. Add unit test for pselect and add specializations for SSE 4.1, AVX512, and half types. 3. Get rid of FIXME: Implement faster pnegate for half by XOR'ing with a sign bit mask. Rasmus Munk Larsen 2019-06-20 11:47:49 -07:00
  • e0be7f30e1 bug #1724: Mask buggy warnings with g++-7 (grafted from 427f2f66d6 ) Christoph Hertzberg 2019-06-14 14:57:46 +02:00
  • 6d432eae5d Make is_valid_index_type return false for float and double when EIGEN_HAS_TYPE_TRAITS is off. Rasmus Munk Larsen 2019-06-05 16:42:27 -07:00
  • f715f6e816 Add workaround for choosing the right include files with FP16C support with clang. Rasmus Munk Larsen 2019-06-05 13:36:37 -07:00
  • c1b0aea653 Merged in Artem-B/eigen (pull request PR-654) Rasmus Larsen 2019-05-31 22:27:04 +00:00
  • b08527b0c1 Clean up CUDA/NVCC version macros and their use in Eigen, and a few other CUDA build failures. Rasmus Munk Larsen 2019-05-31 15:26:06 -07:00
  • b4c49bf00e Minor build improvements tra 2019-05-31 14:08:34 -07:00
  • 5614400581 digits10() needs to return an integer Problem reported on https://stackoverflow.com/questions/56395899 Christoph Hertzberg 2019-05-31 15:45:41 +02:00
  • 36e0a2b93f Merged in deven-amd/eigen-hip-fix-190524 (pull request PR-649) Rasmus Larsen 2019-05-24 16:05:31 +00:00
  • 2c38930161 fix for HIP build errors that were introduced by a commit earlier this week Deven Desai 2019-05-24 14:25:32 +00:00
  • c5019f722b Use pade for matrix exponential also for complex values. Michael Tesch 2019-05-08 17:04:55 +02:00
  • 56bc4974fb GEMV: remove double declaration of constant. Gustavo Lima Chaves 2019-05-23 14:50:29 -07:00
  • ac21a08c13 Cast Index to RealScalar This fixes compilation issues with RealScalar types that are not implicitly castable from Index (e.g. ceres Jet types). Reported by Peter Anderson-Sprecher via eMail Christoph Hertzberg 2019-05-23 15:31:12 +02:00
  • 3eb5ad0ed0 Enable support for F16C with Clang. The required intrinsics were added here: https://reviews.llvm.org/D16177 and are part of LLVM 3.8.0. Rasmus Munk Larsen 2019-05-20 17:19:20 -07:00
  • e92486b8c3 Merged in rmlarsen/eigen (pull request PR-643) Rasmus Larsen 2019-05-20 17:02:39 +00:00
  • fd595d42a7 Merge Rasmus Munk Larsen 2019-05-20 09:39:11 -07:00
  • cc7ecbb124 Merged in scramsby/eigen (pull request PR-646) Gael Guennebaud 2019-05-20 07:19:10 +00:00
  • 01654d97fa Prevent potential division by zero in TensorExecutor Eugene Zhulenev 2019-05-17 14:02:25 -07:00
  • 78d3015722 Merged in ezhulenev/eigen-01 (pull request PR-644) Rasmus Larsen 2019-05-17 19:44:25 +00:00
  • bf9cbed8d0 Merged in glchaves/eigen (pull request PR-635) Rasmus Larsen 2019-05-17 19:40:50 +00:00
  • 96a276803c Always evaluate Tensor expressions with broadcasting via tiled evaluation code path Eugene Zhulenev 2019-05-16 16:15:45 -07:00
  • ab0a30e429 Make Eigen build with cuda 10 and clang. Rasmus Munk Larsen 2019-05-15 13:32:15 -07:00
  • 734a50dc60 Make Eigen build with cuda 10 and clang. Rasmus Munk Larsen 2019-05-15 13:32:15 -07:00