Commit Graph

Select branches

Hide Pull Requests

2.0

3.0

3.1

3.2

3.3

3.4

5.0

gpu-cg-interop

gpu-dense-solvers

gpu-library-dispatch

gpu-modernize-minimum-versions

gpu-sparse-fft-spmv

master

revert-b1d2ce4c

selfadjoint-eigensolver-audit

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

3.4.0-rc1

3.4.1

5.0.0

5.0.1

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

• 4fb3459a23 Fix double-promotion warnings Christoph Hertzberg 2021-02-27 16:16:10 +01:00
• 4bfcee47b9 Idrs iterative linear solver Jens Wehner 2021-02-27 12:09:33 +00:00
• 29ebd84cb7 Fix NEON sqrt for 32-bit, add prsqrt. Antonio Sanchez 2021-02-26 13:59:46 -08:00
• fe19714f80 Merge branch 'rmlarsen1/eigen-nan_prop' Rasmus Munk Larsen 2021-02-26 09:21:24 -08:00
• e67672024d Merge branch 'nan_prop' of https://gitlab.com/rmlarsen1/eigen into nan_prop Rasmus Munk Larsen 2021-02-26 09:12:44 -08:00
• 5e7d4c33d6 Add TODO. Rasmus Munk Larsen 2021-02-26 09:08:45 -08:00
• fb5b59641a Defer default for minCoeff/maxCoeff to templated variant. Rasmus Munk Larsen 2021-02-26 09:07:00 -08:00
• e19829c3b0 Fix floor/ceil for NEON fp16. Antonio Sanchez 2021-02-25 20:39:56 -08:00
• 5529db7524 Fix SSE/NEON pfloor/pceil for saturated values. Antonio Sanchez 2021-02-25 14:29:49 -08:00
• 51eba8c3e2 Fix indentation. Rasmus Munk Larsen 2021-02-24 17:59:36 -08:00
• 5297b7162a Make it possible to specify NaN propagation strategy for maxCoeff/minCoeff reductions. Rasmus Munk Larsen 2021-02-24 17:49:20 -08:00
• ecb7b19dfa Disable new/delete test for HIP Antonio Sanchez 2021-02-25 08:04:05 -08:00
• 6eebe97bab Fix clang compile when no MMA flags are set. Simplify MMA compiler detection. Chip-Kerchner 2021-02-24 20:43:23 -06:00
• f284c8592b Don't crash when attempting to slice an empty tensor. Rasmus Munk Larsen 2021-02-24 18:12:51 -08:00
• 4cb0592af7 Fix indentation. Rasmus Munk Larsen 2021-02-24 17:59:36 -08:00
• 6b34568c74 Merge branch 'nan_prop' of https://gitlab.com/rmlarsen1/eigen into nan_prop Rasmus Munk Larsen 2021-02-24 17:54:58 -08:00
• 0065f9d322 Make it possible to specify NaN propagation strategy for maxCoeff/minCoeff reductions. Rasmus Munk Larsen 2021-02-24 17:49:20 -08:00
• 841c8986f8 Make it possible to specify NaN propagation strategy for maxCoeff/minCoeff reductions. Rasmus Munk Larsen 2021-02-24 17:49:20 -08:00
• 113e61f364 Remove unused function scalar_cmp_with_cast. Rasmus Munk Larsen 2021-02-24 15:58:23 -08:00
• 98ca58b02c Cast anonymous enums to int when used in expressions. Rasmus Munk Larsen 2021-02-24 14:12:27 -08:00
• c31ead8a15 Having forward template function declarations in a P10 file causes bad code in certain situations. Chip-Kerchner 2021-02-24 17:40:34 -06:00
• f44197fabd Some improvements for kissfft from Martin Reinecke(pocketfft author): 1.Only computing about half of the factors and use complex conjugate symmetry for the rest instead of all to save time. 2.All twiddles are calculated in double because that gives the maximum achievable precision when doing float transforms. 3.Reducing all angles to the range 0<angle<pi/4 which gives even more precision. Guoqiang QI 2021-02-24 21:36:47 +00:00
• a31effc3bc Add invoke_result and eliminate result_of warnings for C++17+. Antonio Sanchez 2021-02-17 11:59:33 -08:00
• 8523d447a1 Fixes to support old and new versions of the compilers for built-ins. Cast to non-const when using vector_pair with certain built-ins. Chip-Kerchner 2021-02-24 14:46:04 -06:00
• 5908aeeaba Fix CUDA device new and delete, and add test. Antonio Sanchez 2021-02-17 15:09:37 -08:00
• 119763cf38 Eliminate CMake FindPackageHandleStandardArgs warnings. Antonio Sanchez 2021-02-23 20:16:52 -08:00
• 6cf0ab5e99 Disable fast psqrt for NEON. Antonio Sanchez 2021-02-23 19:48:46 -08:00
• aba3998278 Fix check if GPU compile phase for std::hash Antonio Sanchez 2021-02-23 19:52:08 -08:00
• db5691ff2b Fix some CUDA warnings. Antonio Sanchez 2021-02-19 08:52:31 -08:00
• 88d4c6d4c8 Accurate pow, part 2. This change adds specializations of log2 and exp2 for double that make pow<double> accurate the 1 ULP. Speed for AVX-512 is within 0.5% of the currect implementation. Rasmus Munk Larsen 2021-02-22 16:06:00 -08:00
• 53a7864c48 Fixed sparse conservativeResize() when both num cols and rows decreased. Adam Shapiro 2021-02-23 21:32:39 +00:00
• 2ac0b78739 Fixed sparse conservativeResize() when both num cols and rows decreased. Adam Shapiro 2021-02-23 21:32:39 +00:00
• 10c77b0ff4 Fix compilation errors with later versions of GCC and use of MMA. Chip-Kerchner 2021-02-22 15:01:47 -06:00
• 73922b0174 Fixes Bug #1925. Packets should be passed by const reference, even to inline functions. Christoph Hertzberg 2021-02-20 18:56:42 +01:00
• 5f9cfb2529 Add missing adolc isinf/isnan. Antonio Sanchez 2021-02-19 14:17:10 -08:00
• ce4af0b38f Missing change regarding #1910 Christoph Hertzberg 2021-02-19 20:51:35 +01:00
• 9fc3d9f3ca Fix some implicit literal to Scalar conversions in SparseCore Gael Guennebaud 2019-09-10 22:03:07 +00:00
• a7749c09bc Bug #1910: Make SparseCholesky work for RowMajor matrices Christoph Hertzberg 2021-02-19 19:36:18 +01:00
• 128eebf05e Revert "add EIGEN_DEVICE_FUNC to EIGEN_MAKE_ALIGNED_OPERATOR_NEW_IF macros (only if not HIPCC)." Antonio Sánchez 2021-02-19 17:09:16 +00:00
• 33e0af0130 Return nan at poles of polygamma, digamma, and zeta if limit is not defined frgossen 2021-02-19 16:35:11 +00:00
• 7f09d3487d Use the Cephes double subtraction trick in pexp<float> even when FMA is available. Otherwise the accuracy drops from 1 ulp to 3 ulp. Rasmus Munk Larsen 2021-02-18 20:49:18 +00:00
• 12fd3dd655 add EIGEN_DEVICE_FUNC to EIGEN_MAKE_ALIGNED_OPERATOR_NEW_IF macros (only if not HIPCC). Masaki Murooka 2020-03-25 17:26:22 +09:00
• aa8b22e776 Bump to 3.4.99 David Tellenbach 2021-02-17 23:23:17 +01:00
• 831133cc76 Bump to 3.4.0 3.4.0-rc1 David Tellenbach 2021-02-17 23:19:19 +01:00
• 5336ad8591 Define internal::make_unsigned for [unsigned]long long on macOS. David Tellenbach 2021-02-17 23:03:10 +01:00
• 0845df7f77 Fix uninitialized warning on AVX. Antonio Sanchez 2021-02-17 13:13:39 -08:00
• 9b51dc7972 Fixed performance issues for VSX and P10 MMA in general_matrix_matrix_product Chip Kerchner 2021-02-17 17:49:23 +00:00
• be0574e215 New accurate algorithm for pow(x,y). This version is accurate to 1.4 ulps for float, while still being 10x faster than std::pow for AVX512. A future change will introduce a specialization for double. Rasmus Munk Larsen 2021-02-17 02:50:32 +00:00
• 7ff0b7a980 Updated pfrexp implementation. Antonio Sanchez 2021-02-12 11:32:29 -08:00
• 9ad4096ccb Document possible inconsistencies when using Matrix<bool, ...> David Tellenbach 2021-02-17 00:50:26 +01:00
• f702792a7c missing method in packetmath.h void ptranspose(PacketBlock<Packet16uc, 4>& kernel) Ashutosh Sharma 2021-02-16 08:25:33 -08:00
• db61b8d478 Avoid -Wunused warnings in NDEBUG builds. Jan van Dijk 2021-02-12 18:35:35 +00:00
• 622c598944 Don't allow all test jobs to fail but only the currently failing ones. David Tellenbach 2021-02-12 14:01:17 +01:00
• 90ee821c56 Use vrsqrts for rsqrt Newton iterations. Antonio Sanchez 2021-02-11 11:33:51 -08:00
• 9fde9cce5d Adjust bounds for pexp_float/double Antonio Sanchez 2021-02-09 12:01:09 -08:00
• 4cb563a01e Fix ldexp implementations. Antonio Sanchez 2020-10-12 12:24:08 +01:00
• 7eb07da538 loop less ptranspose Ashutosh Sharma 2021-02-10 10:21:37 -08:00
• 36200b7855 Remove vim specific comments to recognoize correct file-type. David Tellenbach 2021-02-09 09:13:09 +01:00
• 54589635ad Replace nullptr by NULL in SparseLU.h to be C++03 compliant. David Tellenbach 2021-02-09 09:08:06 +01:00
• 984d010b7b add specialization of check_sparse_solving() for SuperLU solver, in order to test adjoint and transpose solves Ralf Hannemann-Tamas 2021-02-08 22:00:31 +00:00
• b578930657 Fix documentation typos in LDLT.h Nikolaus Demmel 2021-01-08 13:09:54 +00:00
• 66841ea070 Enable bdcsvd on host. Antonio Sanchez 2021-02-08 12:51:29 -08:00
• 6e3b795f81 Add more tests for pow and fix a corner case for huge exponent where the result is always zero or infinite unless x is one. Rasmus Munk Larsen 2021-02-05 16:58:49 -08:00
• abcde69a79 Disable vectorized pow for half/bfloat16. Antonio Sanchez 2021-02-05 12:17:34 -08:00
• f85038b7f3 Fix excessive GEBP register spilling for 32-bit NEON. Antonio Sanchez 2021-02-03 08:18:28 -08:00
• 56c8b14d87 Eliminate implicit conversions from float to double. Antonio Sanchez 2021-02-01 15:31:01 -08:00
• fb4548e27b Implement bit_* for device. Antonio Sanchez 2021-02-01 13:23:53 -08:00
• 84911f9c05 Include <cstdint> in one place, remove custom typedefs Antonio Sanchez 2021-01-26 13:39:34 -08:00
• 1615a27993 Fix altivec packetmath. Antonio Sanchez 2021-01-26 10:23:23 -08:00
• 1414e2212c Fix clang compilation for AltiVec from previous check-in Chip Kerchner 2021-01-28 18:36:40 +00:00
• 170a504c2f Add the following functions David Tellenbach 2021-01-28 15:13:07 +01:00
• 598e1b6e54 Add the following functions: David Tellenbach 2021-01-28 01:10:36 +01:00
• 0668c68b03 Allow for negative strides. Gael Guennebaud 2021-01-27 23:29:33 +01:00
• 288d456c29 Replace language_support module with builtin CheckLanguage Samir Benmendil 2020-10-12 12:24:08 +01:00
• 3f4684f87d Include <cstdint> in one place, remove custom typedefs Antonio Sanchez 2021-01-26 13:39:34 -08:00
• 0784d9f87b Fix sqrt, ldexp and frexp compilation errors. Chip Kerchner 2021-01-25 15:22:19 -06:00
• a4edb1079c fix test of ExtractVolumePatchesOp Gmc2 2021-01-25 03:23:46 +00:00
• 4c42d5ee41 Eliminate implicit conversion warning in test/array_cwise.cpp Antonio Sanchez 2021-01-23 11:54:00 -08:00
• e0d13ead90 Replace std::isnan with numext::isnan for c++03 Antonio Sanchez 2021-01-23 11:02:35 -08:00
• c35965b381 Remove unused variable in SparseLU.h Florian Maurin 2021-01-22 22:24:11 +00:00
• f0e46ed5d4 Fix pow and other cwise ops for half/bfloat16. Antonio Sanchez 2021-01-22 11:10:54 -08:00
• f19bcffee6 Specialize std::complex operators for use on GPU device. Antonio Sanchez 2021-01-06 09:41:15 -08:00
• 65e2169c45 Add support for Arm SVE David Tellenbach 2021-01-21 21:11:57 +00:00
• b2126fd6b5 Fix pfrexp/pldexp for half. Antonio Sanchez 2021-01-20 19:00:09 -08:00
• 25d8498f8b Fix stable_norm_1 test. Antonio Sanchez 2021-01-21 09:44:42 -08:00
• 660c6b857c Remove std::cerr in iterative solver since we don't have iostream. David Tellenbach 2021-01-21 11:40:05 +01:00
• 77dc6dbb44 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
• d5b7981119 Fix signed-unsigned comparison. Antonio Sanchez 2021-01-20 08:34:00 -08:00
• e409795d6b Proper CPUID Ivan Popivanov 2021-01-05 12:51:10 -08:00
• cdd8fdc32e Vectorize pow(x, y). This closes https://gitlab.com/libeigen/eigen/-/issues/2085, which also contains a description of the algorithm. Rasmus Munk Larsen 2021-01-18 13:25:16 +00:00
• bde6741641 Improved std::complex sqrt and rsqrt. Antonio Sanchez 2021-01-16 10:22:07 -08:00
• 21a8a2487c fix paddings of TensorVolumePatchOp Maozhou, Ge 2021-01-15 11:51:49 +08:00
• 38ae5353ab 1)provide a better generic paddsub op implementation 2)make paddsub op support the Packet2cf/Packet4f/Packet2f in NEON 3)make paddsub op support the Packet2cf/Packet4f in SSE Guoqiang QI 2021-01-13 22:54:03 +00:00
• 352f1422d3 Remove inf local variable. Antonio Sanchez 2021-01-12 10:33:15 -08:00
• 2044084979 Remove TODO from Transform::computeScaleRotation() Antonio Sanchez 2021-01-11 11:30:01 -08:00
• 3daf92c7a5 Transform::computeScalingRotation flush determinant to +/- 1. Antonio Sanchez 2021-01-11 09:27:25 -08:00
• 587fd6ab70 Only specialize complex sqrt_impl for CUDA if not MSVC. Antonio Sanchez 2021-01-11 09:15:45 -08:00
• 2a6addb4f9 Fix for breakage in ROCm support - 210108 Deven Desai 2021-01-08 18:04:40 +00:00
• f149e0ebc3 Fix MSVC complex sqrt and packetmath test. Antonio Sanchez 2021-01-07 09:39:05 -08:00
• 8d9cfba799 Fix rand test for MSVC. Antonio Sanchez 2021-01-07 12:45:20 -08:00