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

  • 8719b9c5bc Disable test for 32-bit systems (e.g. ARM, i386) Antonio Sánchez 2020-05-28 17:40:15 +00:00
  • 8e1df5b082 Fix incorrect usage of if defined(EIGEN_ARCH_PPC) => if EIGEN_ARCH_PPC Yong Tang 2020-05-28 05:50:56 -07:00
  • 4e7046063b Fix #1874: it works on both MSVC 2017 and other platforms. Kan Chen 2020-05-21 18:42:56 +08:00
  • 2d67af2d2b Add pscatter for Packet16{u}c (int8) Pedro Caldeira 2020-05-20 09:57:03 -05:00
  • 5328cd62b3 Guard usage of decltype since it's a C++11 feature David Tellenbach 2020-05-20 16:01:41 +02:00
  • cc86a31e20 Add guard around specialization for bool, which is only currently implemented for SSE. Rasmus Munk Larsen 2020-05-19 16:21:56 -07:00
  • 8a7f360ec3 - Vectorizing MMA packing. - Optimizing MMA kernel. - Adding PacketBlock store to blas_data_mapper. Everton Constantino 2020-05-19 19:24:11 +00:00
  • a145e4adf5 Add newline at the end of StlIterators.h. Rasmus Munk Larsen 2020-05-15 20:36:00 +00:00
  • 8ce9630ddb Fix #1874: workaround MSVC 2017 compilation issue. Gael Guennebaud 2020-05-15 20:47:32 +02:00
  • 9b411757ab Add missing packet ops for bool, and make it pass the same packet op unit tests as other arithmetic types. Rasmus Munk Larsen 2020-05-11 13:23:31 -07:00
  • d640276d31 Added support for reverse iterators for Vectorwise operations. Felipe Attanasio 2020-05-14 22:38:20 +00:00
  • fa8fd4b4d5 Indexed view should have RowMajorBit when there is staticly a single row Christopher Moore 2020-05-14 22:11:19 +00:00
  • a187ffea28 Resolve "IndexedView of a vector should allow linear access" Christopher Moore 2020-05-13 19:24:42 +00:00
  • ba9d18b938 Add KLU support to spbenchsolver Mark Eberlein 2020-01-24 16:28:09 -08:00
  • 5fdc179241 Altivec template functions to better code reusability Pedro Caldeira 2020-05-11 16:38:56 -03:00
  • d3e81db6c5 Eigen moved the scanLauncehr function inside the internal namespace. This commit applies the following changes: - Moving the scamLauncher specialization inside internal namespace to fix compiler crash on TensorScan for SYCL backend. - Replacing SYCL/sycl.hpp to CL/sycl.hpp in order to follow SYCL 1.2.1 standard. - minor fixes: commenting out an unused variable to avoid compiler warnings. mehdi-goli 2020-05-11 16:10:33 +01:00
  • c1d944dd91 Remove packet ops pinsertfirst and pinsertlast that are only used in a single place, and can be replaced by other ops when constructing the first/final packet in linspaced_op_impl::packetOp. Rasmus Munk Larsen 2020-05-08 15:41:50 -07:00
  • 5c4e19fbe7 Possibility to specify user-defined default cache sizes for GEBP kernel David Tellenbach 2020-05-07 17:25:42 +02:00
  • 225ab040e0 Remove unused packet op "palign". Clean up a compiler warning in c++03 mode in AVX512/Complex.h. Rasmus Munk Larsen 2020-05-07 17:14:26 -07:00
  • 74ec8e6618 Make size odd for transposeInPlace test to make sure we hit the scalar path. Rasmus Munk Larsen 2020-05-07 17:29:56 +00:00
  • 49f1aeb60d Remove traits declaring NEON vectorized casts that do not actually have packet op implementations. Rasmus Munk Larsen 2020-05-07 09:49:22 -07:00
  • 2fd8a5a08f Add parallelization of TensorScanOp for types without packet ops. Rasmus Munk Larsen 2020-05-06 14:48:37 -07:00
  • 0e59f786e1 Fix accidental copy of loop variable. Rasmus Munk Larsen 2020-05-05 21:35:38 +00:00
  • 7b76c85daf Vectorize and parallelize TensorScanOp. Rasmus Munk Larsen 2020-05-05 00:19:43 +00:00
  • a74a278abd Fix confusing template param name for Stride fwd decl. Xiaoxiang Cao 2020-04-30 01:43:05 +00:00
  • 923ee9aba3 Fix the embarrassingly incomplete fix to the embarrassing bug in blocked transpose. Rasmus Munk Larsen 2020-04-29 17:27:36 +00:00
  • a32923a439 Fix (embarrassing) bug in blocked transpose. Rasmus Munk Larsen 2020-04-29 17:02:27 +00:00
  • 1e41406c36 Add missing transpose in cleanup loop. Without it, we trip an assertion in debug mode. Rasmus Munk Larsen 2020-04-29 01:30:51 +00:00
  • fbe7916c55 Fix compilation error with Clang on Android: _mm_extract_epi64 fails to compile. Rasmus Munk Larsen 2020-04-29 00:58:41 +00:00
  • 82f54ad144 Fix perf monitoring merge function Clément Grégoire 2020-04-23 13:58:01 +00:00
  • ab773c7e91 Extend support for Packet16b: Rasmus Munk Larsen 2020-04-24 17:29:25 -07:00
  • b47c777993 Block transposeInPlace() when the matrix is real and square. This yields a large speedup because we transpose in registers (or L1 if we spill), instead of one packet at a time, which in the worst case makes the code write to the same cache line PacketSize times instead of once. Rasmus Munk Larsen 2020-04-27 18:55:15 -07:00
  • 29f0917a43 Add support to vector instructions to Packet16uc and Packet16c Pedro Caldeira 2020-03-18 10:25:26 -05:00
  • e80ec24357 Remove unused packet op "preduxp". Rasmus Munk Larsen 2020-04-23 18:17:14 +00:00
  • 0aebe19aca BooleanRedux.h: Add more EIGEN_DEVICE_FUNC qualifiers. René Wagner 2020-04-23 17:25:08 +02:00
  • 3c02fefec5 Add async evaluation support to TensorSlicingOp. Eugene Zhulenev 2020-04-22 19:55:01 +00:00
  • 0c67b855d2 Add Packet8s and Packet8us to support signed/unsigned int16/short Altivec vector operations Pedro Caldeira 2020-03-09 13:33:00 -05:00
  • e8f40e4670 Fix bug in ptrue for Packet16b. Rasmus Munk Larsen 2020-04-20 21:45:10 +00:00
  • 2f6ddaa25c Add partial vectorization for matrices and tensors of bool. This speeds up boolean operations on Tensors by up to 25x. Rasmus Munk Larsen 2020-04-20 20:16:28 +00:00
  • 00f6340153 Update PreprocessorDirectives.dox - Added line for the new VectorwiseOp plugin directive (and re-alphabatized the plugin section) dlazenby 2020-04-17 21:43:37 +00:00
  • 5ab87d8aba Move eigen_packet_wrapper to GenericPacketMath.h and use it for SSE/AVX/AVX512 as it is already used for NEON. This will allow us to define multiple packet types backed by the same vector type, e.g., __m128i. Use this machanism to define packets for half and clean up the packet op implementations. Rasmus Munk Larsen 2020-04-15 18:17:19 +00:00
  • 4aae8ac693 Fix typo in TypeCasting.h Rasmus Munk Larsen 2020-04-14 02:55:51 +00:00
  • 1d674003b2 Fix big in vectorized casting of Rasmus Munk Larsen 2020-04-14 02:11:06 +00:00
  • b1aa07a8d3 Fix a bug in TensorIndexList.h Changming Sun 2020-04-10 20:48:54 +00:00
  • d46d726e9d CommaInitializer wrongfully asserted for 0-sized blocks commainitialier unit-test never actually called test_block_recursion, which also was not correctly implemented and would have caused too deep template recursion. Christoph Hertzberg 2020-04-13 16:41:20 +02:00
  • c854e189e6 Fixed commainitializer test. Antonio Sanchez 2020-04-10 13:53:26 -07:00
  • 39142904cc Resolve C4346 when building eigen on windows jangsoopark 2020-04-08 14:55:39 +09:00
  • f0577a2bfd Speed up matrix multiplication for small to medium size matrices by using half- or quarter-packet vectorized loads in gemm_pack_rhs if they have size 4, instead of dropping down the the scalar path. Rasmus Munk Larsen 2020-04-07 22:09:51 +00:00
  • 8e875719b3 Replace norm() with squaredNorm() to address integer overflows Antonio Sanchez 2020-04-07 09:30:16 -07:00
  • 9dda5eb7d2 Missing struct definition in NumTraits Antonio Sanchez 2020-04-07 09:01:11 -07:00
  • bcc0e9e15c Add numeric_limits min and max for bool Akshay Naresh Modi 2020-04-06 23:38:57 +00:00
  • 54a0a9c9dd Bugfix: conjugate_gradient did not compile with lazy-evaluated RealScalar Bernardo Bahia Monteiro 2020-03-29 18:53:18 -04:00
  • 4fd5d1477b Fix packetmath test build for AVX. Rasmus Munk Larsen 2020-03-27 17:05:39 +00:00
  • 393dbd8ee9 Fix bug in 52d54278be Rasmus Munk Larsen 2020-03-27 16:42:18 +00:00
  • 55c8fe8d0f Fix bug in 52d54278be Rasmus Munk Larsen 2020-03-27 16:41:15 +00:00
  • 6d2dbfc453 NEON: Fixed MSVC types definitions Joel Holdsworth 2020-03-19 11:25:56 +00:00
  • 52d54278be Additional NEON packet-math operations Joel Holdsworth 2020-03-26 20:18:19 +00:00
  • 0dd9643ad5 Fix Yade high precision Real compilation Janek Kozicki 2020-03-23 18:24:49 +00:00
  • deb93ed1bf Adhere to recommended load/store intrinsics for pp64le Everton Constantino 2020-03-18 12:27:39 -03:00
  • 5c22c7a7de Make file formatting comply with POSIX and Unix standards Aaron Franke 2020-03-21 21:30:06 -04:00
  • 5afdaa473a Fixing float32's pround halfway criteria to match STL's criteria. Everton Constantino 2020-03-16 18:17:03 -03:00
  • 96cd1ff718 Fixed: - access violation when initializing 0x0 matrices - exception can be thrown during stack unwind while comma-initializing a matrix if eigen_assert if configured to throw Alessio M 2020-03-21 05:11:21 +00:00
  • cc954777f2 Update VectorwiseOp.h to allow Plugins similar to MatrixBase.h or ArrayBase.h dlazenby 2020-03-19 16:43:38 +00:00
  • 55ecd58a3c Bug https://gitlab.com/libeigen/eigen/-/issues/1415: add missing EIGEN_DEVICE_FUNC to diagonal_product_evaluator_base. Masaki Murooka 2020-03-11 12:26:17 +09:00
  • 4da2c6b197 Remove reference to non-existent unary_op_base class. Rasmus Munk Larsen 2020-03-19 18:23:06 +00:00
  • eda90baf35 Add missing arguments to numext::absdiff(). Rasmus Munk Larsen 2020-03-19 18:16:55 +00:00
  • d5c665742b Add absolute_difference coefficient-wise binary Array function Joel Holdsworth 2020-03-19 17:45:20 +00:00
  • 6ff5a14091 Reenabling packetmath unsigned tests, adding dummy pabs for relevant unsigned types. Everton Constantino 2020-03-19 17:31:49 +00:00
  • 232f904082 Add shift_left<N> and shift_right<N> coefficient-wise unary Array functions Joel Holdsworth 2020-03-19 17:24:06 +00:00
  • 54aa8fa186 Implement integer square-root for NEON Joel Holdsworth 2020-03-19 17:05:13 +00:00
  • 37ccb86916 Update NullaryFunctors.h Allan Leal 2020-03-16 11:59:02 +00:00
  • 7158ed4e0e Fixing HIP breakage caused by the recent commit that introduces Packet4h2 as the Eigen::Half packet type Deven Desai 2020-03-11 23:06:56 +00:00
  • d53ae40f7b NEON: Added int64_t and uint64_t packet math Joel Holdsworth 2020-01-13 14:04:48 +00:00
  • 4b9ecf2924 NEON: Added int8_t and uint8_t packet math Joel Holdsworth 2019-11-14 11:46:40 +00:00
  • ceaabd4e16 NEON: Added int16_t and uint16_t packet math Joel Holdsworth 2019-11-15 13:58:57 +00:00
  • d5d3cf9339 NEON: Added uint32_t packet math Joel Holdsworth 2019-12-17 18:41:16 +00:00
  • eacf97f727 NEON: Implemented half-size vectors Joel Holdsworth 2019-11-26 18:31:22 +00:00
  • 5f411b729e NEON: Set packet_traits<double> flags Joel Holdsworth 2020-01-13 18:53:49 +00:00
  • 88337acae2 test/packetmath: Add tests for all integer types Joel Holdsworth 2020-02-10 14:58:37 +00:00
  • 9e68977578 test/packetmath: Made negate non-mandatory Joel Holdsworth 2019-11-14 16:03:48 +00:00
  • b733b8b680 remove duplicate pset1 for half and add some comments about why we need expose pmul/add/div/min/max on host Sami Kama 2020-03-10 20:28:43 +00:00
  • a45d28256d Don't restrict CMAKE_BUILD_TYPE Ram-Z 2020-01-10 19:12:48 +00:00
  • 14db78c53b Fix some maybe-uninitialized warnings Christoph Hertzberg 2020-02-28 19:32:10 +01:00
  • 84364ad11d 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 2020-02-28 19:04:33 +01:00
  • 98bfc5aaa8 Update MarketIO.h Cédric Hubert 2020-02-14 19:35:30 +00:00
  • 52a2fbbb00 Revert "avoid selecting half-packets when unnecessary" Rasmus Munk Larsen 2020-02-25 01:07:43 +00:00
  • 235bcfe08d Revert "Pick full packet unconditionally when EIGEN_UNALIGNED_VECTORIZE" Rasmus Munk Larsen 2020-02-25 01:07:28 +00:00
  • d7a42eade6 Revert "do not pick full-packet if it'd result in more operations" Rasmus Munk Larsen 2020-02-25 01:07:15 +00:00
  • 6ac37768a9 Revert "add some static checks for packet-picking logic" Rasmus Munk Larsen 2020-02-25 01:07:04 +00:00
  • 87cfa4862f Revert "Disable test in test/vectorization_logic.cpp, which is currently failing with AVX." Rasmus Munk Larsen 2020-02-25 01:04:56 +00:00
  • b625adffd8 Disable test in test/vectorization_logic.cpp, which is currently failing with AVX. Rasmus Munk Larsen 2020-02-24 23:28:25 +00:00
  • f0ce88cff7 Include <sstream> explicitly, and don't rely on the implicit include via <complex>. Tobias Bosch 2020-02-24 23:09:36 +00:00
  • 160c0a3404 Change typedefs from private to protected to fix MSVC compilation Eugene Zhulenev 2019-09-03 19:11:36 -07:00
  • eb6cc29583 Avoid a division in NonBlockingThreadPool::Steal. Ilya Tokar 2020-02-14 16:02:57 -05:00
  • 7769600245 add some static checks for packet-picking logic Francesco Mazzoli 2020-01-24 11:06:36 +01:00
  • e9cc0cd353 do not pick full-packet if it'd result in more operations Francesco Mazzoli 2020-01-14 15:20:24 +01:00
  • 44df2109c8 Pick full packet unconditionally when EIGEN_UNALIGNED_VECTORIZE Francesco Mazzoli 2020-01-13 16:15:30 +01:00
  • 5ca10480b0 avoid selecting half-packets when unnecessary Francesco Mazzoli 2020-01-13 15:11:22 +01:00
  • f584bd9b30 Fail at compile time if default executor tries to use non-default device Eugene Zhulenev 2020-02-06 22:43:24 +00:00
  • 3fda850c46 Remove dead code from TensorReduction.h Eugene Zhulenev 2020-01-29 18:45:31 +00:00