Commit Graph

Select branches

Hide Pull Requests

2.0

3.0

3.1

3.2

3.3

3.4

EIGEN_ARM_ALIGN_HINT

RVV_Chip

avx512fp16

ci

master

options

revert-3f3144f5

revert-7f06bcae

2.0-beta1

2.0-beta2

2.0-beta3

2.0-beta4

2.0-beta5

2.0-beta6

2.0-rc1

2.0.0

2.0.1

2.0.10

2.0.11

2.0.12

2.0.13

2.0.14

2.0.15

2.0.16

2.0.17

2.0.2

2.0.3

2.0.4

2.0.5

2.0.6

2.0.7

2.0.8

2.0.9

3.0-beta1

3.0-beta2

3.0-beta3

3.0-beta4

3.0-rc1

3.0.0

3.0.1

3.0.2

3.0.3

3.0.4

3.0.5

3.0.6

3.0.7

3.1.0

3.1.0-alpha1

3.1.0-alpha2

3.1.0-beta1

3.1.0-rc1

3.1.0-rc2

3.1.1

3.1.2

3.1.3

3.1.4

3.2-beta1

3.2-rc1

3.2-rc2

3.2.0

3.2.1

3.2.10

3.2.2

3.2.3

3.2.4

3.2.5

3.2.6

3.2.7

3.2.8

3.2.9

3.3-alpha1

3.3-beta1

3.3-beta2

3.3-rc1

3.3-rc2

3.3.0

3.3.1

3.3.2

3.3.3

3.3.4

3.3.5

3.3.6

3.3.7

3.3.8

3.3.8-rc1

3.3.9

3.4-rc1

3.4.0

actual-start-from-scratch

after-hg-migration

before-3.4

before-evaluators

before-git-migration

before-hg-migration

nightly

starting_new_generickernels

starting_new_packmapcalculator

Mono Color

• bf48d46338 Explicitly state that indices must be sorted. Antonio Sánchez 2022-10-19 18:15:29 +00:00
• 3bb6a48d8c Fix bug atan2 Rasmus Munk Larsen 2022-10-12 23:49:32 +00:00
• 14c847dc0e Refactor special values test for pow, and add a similar test for atan2 Rasmus Munk Larsen 2022-10-12 20:12:08 +00:00
• 462758e8a3 Don't use generic sign function for sign(complex) unless it is vectorizable Rasmus Munk Larsen 2022-10-12 16:03:29 +00:00
• c0d6a72611 Use pnegate(pzero(x)) as a generic way to generate -0.0. Some compiler do not handle the literal -0.0 properly in fastmath mode. Rasmus Munk Larsen 2022-10-12 01:57:05 +00:00
• 7846c7387c Eigen/Sparse: fix warnings -Wunused-but-set-variable Laurent Rineau 2022-10-11 17:37:04 +00:00
• 3167544873 Handle NaN inputs to atan2. Rasmus Munk Larsen 2022-10-10 19:36:36 -07:00
• 72db3f0fa5 Remove references to M_PI_2 and M_PI_4. Rasmus Munk Larsen 2022-10-11 00:27:16 +00:00
• d6bc062591 Remove reference to EIGEN_HAS_CXX11_MATH. Rasmus Munk Larsen 2022-10-10 23:38:28 +00:00
• 5ceed0d57f Guard GCC-specific pragmas with "#ifdef EIGEN_COMP_GNUC" Rasmus Munk Larsen 2022-10-10 20:38:53 +00:00
• 528b68674c [clang-format] Add a few macros to AttributeMacros Alexander Richardson 2022-10-10 16:44:47 +00:00
• e95c4a837f Simpler range reduction strategy for atan<float>(). Rasmus Munk Larsen 2022-10-04 18:11:00 +00:00
• 80efbfdeda Unconditionally enable CXX11 math. Antonio Sánchez 2022-10-04 17:37:47 +00:00
• e5794873cb Replace assert with eigen_assert. Antonio Sánchez 2022-10-04 17:11:23 +00:00
• 7d6a9925cc Fix 4x4 inverse when compiling with -Ofast. Antonio Sánchez 2022-10-04 16:05:49 +00:00
• 1414a76fa9 Only vectorize atan<double> for Altivec if VSX is available. Rasmus Munk Larsen 2022-10-03 22:06:58 +00:00
• c475228b28 Vectorize atan() for double. Rasmus Munk Larsen 2022-10-01 01:49:30 +00:00
• 1e1848fdb1 Add a vectorized implementation of atan2 to Eigen. Rasmus Munk Larsen 2022-09-28 20:46:49 +00:00
• b3bf8d6a13 Try to reduce size of GEBP kernel for non-ARM targets. Rasmus Munk Larsen 2022-09-28 02:37:18 +00:00
• 13b69fc1b0 Try to reduce compilation time/memory for GEBP kernel using EIGEN_IF_CONSTEXPR Rasmus Munk Larsen 2022-09-23 20:09:42 +00:00
• 3c4637640b Remove unused typedef. Rasmus Munk Larsen 2022-09-23 19:11:31 +00:00
• ed8cda3ce4 Move EIGEN_NEON_GEBP_NR macro to the right place in GeneralBlockPanelKernel.h Rasmus Munk Larsen 2022-09-23 02:24:27 +00:00
• e2ea866515 Add a macro to set the nr trait in the BEBP kernel for NEON. Rasmus Munk Larsen 2022-09-22 23:56:34 +00:00
• 23299632c2 Use 3px8/2px8/1px8/1x8 gebp_kernel on arm64-neon Lianhuang Li 2022-09-21 16:36:40 +00:00
• 7b2901e2aa Add vectorized integer division for int32 with AVX512, AVX or SSE. Rasmus Munk Larsen 2022-09-21 00:27:23 +00:00
• 5ffe7b92e0 [ROCm] fixed gpuGetDevice unused message Chao Chen 2022-09-20 21:38:20 +00:00
• f913a40678 Revert "Add AVX int32_t pdiv" Rasmus Munk Larsen 2022-09-16 22:48:08 +00:00
• 273e0c884e Revert "Add constexpr, test for C++14 constexpr." Rasmus Munk Larsen 2022-09-16 21:14:29 +00:00
• ea84e7ad63 Add AVX int32_t pdiv Charles Schlosser 2022-09-16 17:06:29 +00:00
• dceb779ecd Fix test for pow with mixed integer types. We do not convert the exponent if it is an integer type. Rasmus Munk Larsen 2022-09-12 15:51:27 -07:00
• afc014f1b5 Allow mixed types for pow(), as long as the exponent is exactly representable in the base type. Rasmus Munk Larsen 2022-09-12 21:55:30 +00:00
• b2c82a9347 Remove bad skew_symmetric_matrix3 test. Antonio Sánchez 2022-09-10 07:08:37 +00:00
• e8a2aa24a2 Fix a couple of issues with unary pow(): Rasmus Munk Larsen 2022-09-09 17:21:11 +00:00
• 07d0759951 [ROCm] Fix for sparse matrix related breakage on ROCm. Rohit Santhanam 2022-09-09 14:41:00 +00:00
• fb212c745d Fix g++-6 constexpr and c++20 constexpr build errors. Antonio Sánchez 2022-09-09 03:41:45 +00:00
• ec9c7163a3 Feature/skew symmetric matrix3 Thomas Gloor 2022-09-08 20:44:40 +00:00
• 311ba66f7c Fix realloc for non-trivial types. Antonio Sánchez 2022-09-08 19:39:36 +00:00
• 3c37dd2a1d Tweak bound for pow to account for floating-point types. Antonio Sánchez 2022-09-08 17:40:45 +00:00
• f9dfda28ab Add missing comparison operators for GPU packets. Rasmus Munk Larsen 2022-09-07 14:10:02 -07:00
• 242325eca7 Remove unused variable. Rasmus Munk Larsen 2022-09-07 20:46:44 +00:00
• 133498c329 Add constexpr, test for C++14 constexpr. Tobias Schlüter 2022-03-05 19:02:12 +09:00
• 69f50e3a67 Adjust overflow threshold bound for pow tests. Antonio Sánchez 2022-09-06 19:53:29 +00:00
• 3e44f960ed Reduce compiler warnings for tests. Antonio Sanchez 2022-09-06 11:19:38 -07:00
• b7e21d4e38 Call check_that_malloc_is_allowed() in aligned_realloc() Florian Richer 2022-09-05 09:11:08 +02:00
• 6e83e906c2 fix typo in doc/TutorialSparse.dox Gilles Aouizerate 2022-09-04 21:20:21 +00:00
• 525f066671 fixed msvc compilation error in GeneralizedEigenSolver.h Michael Palomas 2022-09-04 17:50:43 +00:00
• f241a2c18a Add asserts for index-out-of-bounds in IndexedView. Antonio Sánchez 2022-09-02 17:28:03 +00:00
• f5364331eb Fix some cmake issues. Antonio Sánchez 2022-09-02 16:43:14 +00:00
• d816044b6e Fix mixingtypes tests. Antonio Sánchez 2022-09-02 15:30:13 +00:00
• 94cc83faa1 2 typos fix in the 3rd table. Gilles Aouizerate 2022-08-31 19:54:42 +00:00
• 30c42222a6 Fix some test build errors in new unary pow. Antonio Sánchez 2022-08-30 17:24:14 +00:00
• bd393e15c3 Vectorize acos, asin, and atan for float. Rasmus Munk Larsen 2022-08-29 19:49:33 +00:00
• e5af9f87f2 Vectorize pow for integer base / exponent types Charles Schlosser 2022-08-29 19:23:54 +00:00
• 8acbf5c11c re-enable pow for complex types chuckyschluz 2022-08-26 17:29:02 -04:00
• 7064ed1345 Specialize psign<Packet8i> for AVX2, don't vectorize psign<bool>. Rasmus Munk Larsen 2022-08-26 17:02:37 +00:00
• 98e51c9e24 Avoid undefined behavior in array_cwise test due to signed integer overflow Rasmus Munk Larsen 2022-08-26 16:19:03 +00:00
• a7c1cac18b Fix GeneralizedEigenSolver::info() and Asserts Arthur 2022-08-25 22:05:04 +00:00
• 714678fc6c Add missing ptr in realloc call. Antonio Sanchez 2022-08-24 22:04:04 -07:00
• b2a13c9dd1 Sparse Core: Replace malloc/free with conditional_aligned Charles Schlosser 2022-08-23 21:44:22 +00:00
• 6aad0f821b Fix psign for unsigned integer types, such as bool. Rasmus Munk Larsen 2022-08-22 20:19:35 +00:00
• 1a09defce7 Protect new pblend implementation with EIGEN_VECTORIZE_AVX2 Rasmus Munk Larsen 2022-08-22 18:28:03 +00:00
• 7c67dc67ae Use proper double word division algorithm for pow<double>. Gives 11-15% speedup. Rasmus Munk Larsen 2022-08-17 18:36:23 +00:00
• 7a3b667c43 Add support for AVX512-FP16 for vectorizing half precision math Matthew Sterrett 2022-08-17 18:15:21 +00:00
• 76a669fb45 add fixed power unary operation Charles Schlosser 2022-08-16 21:32:36 +00:00
• 39fcc89798 Removed unnecessary checks for FP16C Matthew Sterrett 2022-08-16 18:14:41 +00:00
• 2f7cce2dd5 [SYCL] Fix some SYCL tests Romain Biessy 2022-08-16 17:37:54 +00:00
• 27367017bd Disable bad "deprecated warning" edge-case in BDCSVD Arthur 2022-08-11 18:43:31 +00:00
• b8e93bf589 Eliminate bool bitwise warnings. Antonio Sánchez 2022-08-09 22:42:30 +00:00
• 66ea0c09fd Don't double-define Half functions on aarch64 Lexi Bromfield 2022-08-09 20:00:34 +00:00
• 97e0784dc6 Vectorize the sign operator in Eigen. Rasmus Munk Larsen 2022-08-09 19:54:57 +00:00
• be20207d10 Fix vectorized Jacobi Rotation Arthur 2022-08-08 19:29:56 +00:00
• 7a87ed1b6a Fix code and unit test for a few corner cases in vectorized pow() Rasmus Munk Larsen 2022-08-08 18:48:36 +00:00
• 9e0afe0f02 Fix non-VSX PowerPC build Chip Kerchner 2022-08-08 18:18:17 +00:00
• 84a9d6fac9 Fix use of Packet2d type for non-VSX. Chip Kerchner 2022-08-03 20:48:13 +00:00
• ce60a7be83 Partial Packet support for GEMM real-only (PowerPC). Also fix compilation warnings & errors for some conditions in new API. Chip Kerchner 2022-08-03 18:15:19 +00:00
• 5a1c7807e6 Fix inner iterator for sparse block. Antonio Sánchez 2022-08-03 17:26:12 +00:00
• 39d22ef46b Fix flaky packetmath_1 test. Antonio Sánchez 2022-08-02 17:42:45 +00:00
• 7896c7dc6b Use numext::sqrt in ConjugateGradient. Antonio Sánchez 2022-07-29 20:17:23 +00:00
• e618c4a5e9 Improve pblend AVX implementation Ilya Tokar 2022-07-29 18:45:33 +00:00
• ef4654bae7 Add true determinant to QR and it's variants sjusju 2022-07-29 18:24:14 +00:00
• b7668c0371 Avoid including <sstream> with EIGEN_NO_IO Alexander Richardson 2022-07-29 18:02:51 +00:00
• 7dd3dda3da Updated AccelerateSupport documentation after PR 966. John Mather 2022-07-29 17:42:31 +00:00
• 69714ff613 Add Sparse Subset of Matrix Inverse Julian Kent 2022-07-28 18:04:35 +00:00
• 34780d8bd1 Include immintrin.h header for enscripten. Antonio Sánchez 2022-07-22 02:27:42 +00:00
• 2cf4d18c9c Disable AVX512 GEMM kernels by default. Antonio Sánchez 2022-07-20 21:22:48 +00:00
• a678a3e052 Fix aligned_realloc to call check_that_malloc_is_allowed() if ptr == 0 Charles Schlosser 2022-07-19 20:59:07 +00:00
• 4a56359406 Add option to disable avx512 GEBP kernels b-shi 2022-07-18 17:59:09 +00:00
• 1092574b26 Fix wrong doxygen group usage Mathieu Westphal 2022-07-12 13:22:46 +02:00
• e1165dbf9a AutoDiff depends on Core, so include appropriate header. Antonio Sánchez 2022-07-09 23:57:09 +00:00
• bb51d9f4fa Fix ODR violations. Antonio Sánchez 2022-07-09 04:56:36 +00:00
• 06a458a13d Enable subtests which use device side malloc since this has been fixed in ROCm 5.2. Rohit Santhanam 2022-06-29 17:09:43 +00:00
• 84cf3ff18d Add pload_partial, pstore_partial (and unaligned versions), pgather_partial, pscatter_partial, loadPacketPartial and storePacketPartial. Chip Kerchner 2022-06-27 19:18:00 +00:00
• c603275dc9 Better performance for Power10 using more load and store vector pairs for GEMV Chip Kerchner 2022-06-27 18:11:55 +00:00
• 0e18714167 Fix clang-tidy warnings about function definitions in headers. Antonio Sanchez 2022-06-23 13:47:32 -07:00
• 8ed3b9dcd6 Skip f16/bf16 bessel specializations on AVX512 if unavailable. Antonio Sánchez 2022-06-24 15:10:36 +00:00
• bc2ab81634 Eliminate undef warnings when not compiling for AVX512. Antonio Sánchez 2022-06-24 15:10:10 +00:00
• 0e083b172e Use numext::sqrt in Householder.h. Antonio Sánchez 2022-06-21 16:29:59 +00:00
• 37673ca1bc AVX512 TRSM kernels use alloca if EIGEN_NO_MALLOC requested b-shi 2022-06-17 18:05:26 +00:00
• 4d1c16eab8 Fix tanh and erf to use vectorized version for EIGEN_FAST_MATH in VSX. Chip Kerchner 2022-06-15 16:06:43 +00:00
• 7ea823e824 [SYCL-Spec] According to [SYCL-2020 spec](... Mehdi Goli 2022-06-13 15:52:29 +00:00