Google Git
Sign in
eigen / mirror / HEAD
  1. 5610a13 Simplify and speed up pow() by 5-6% by Rasmus Munk Larsen · 20 hours ago master
  2. 6c6ce9d Enable vectorized erf<double>(x) for SSE and AVX, which was accidentally removed in merge request 1750. by Rasmus Munk Larsen · 35 hours ago
  3. e7c799b Prevent premature overflow to infinity in exp(x). The changes also provide a 3-4% speedup. by Rasmus Munk Larsen · 36 hours ago
  4. 00af471 Revert https://gitlab.com/libeigen/eigen/-/commit/040180078db70b8673932d7e5615920d64ceeaf5 by Rasmus Munk Larsen · 2 days ago
  5. 8ee6f84 Speed up exp(x). by Rasmus Munk Larsen · 2 days ago
  6. 93ec545 disable fill_n optimization for msvc by Charles Schlosser · 2 days ago
  7. 0af6ab4 Remove unnecessary check for HasBlend trait. by Rasmus Munk Larsen · 2 days ago
  8. d5eec78 Get rid of redundant computation for large arguments to erf(x). by Rasmus Munk Larsen · 3 days ago
  9. 2fc6380 Fix C++20 constexpr test compilation failures by Tyler Veness · 3 days ago
  10. 5133c83 Vectorize erf(x) for double. by Rasmus Munk Larsen · 5 days ago
  11. d6e3b52 Update Assign_MKL.h to cast disparate enum type to int, so it can be compared... by Conrad Poelman · 6 days ago
  12. 0401800 Ensure that destructor's needed by lldb make it into binary in non-inlined fashion by breathe1 · 6 days ago
  13. 0fb2ed1 Make element accessors constexpr by Tyler Veness · 7 days ago
  14. 8b4efc8 check_size_for_overflow: use numeric limits instead of c99 macro by Charles Schlosser · 8 days ago
  15. 489dbbc make fixed_size matrices conform to std::is_standard_layout by Charles Schlosser · 8 days ago
  16. 283d871 Add missing EIGEN_DEVICE_FUNCTION decorations. by Rasmus Munk Larsen · 12 days ago
  17. 0d366f6 Vectorize erfc(x) for double and improve erfc(x) for float. by Rasmus Munk Larsen · 13 days ago
  18. 8adf436 more avx predux_any by Charles Schlosser · 14 days ago
  19. bc424f6 add missing avx predux_any functions by Charles Schlosser · 14 days ago avx_predux
  20. e52ac76 use EIGEN_CPLUSPLUS instead of checking cpp version by Charles Schlosser · 2 weeks ago
  21. 122be16 Revert "make fixed-size objects trivially move assignable" by Rasmus Munk Larsen · 2 weeks ago
  22. d490212 Tensor Roll / Circular Shift / Rotate by Tobias Wood · 2 weeks ago
  23. bb73be8 make fixed-size objects trivially move assignable by Charles Schlosser · 2 weeks ago
  24. 7fd305e Fix GPU builds. by Antonio Sánchez · 3 weeks ago
  25. c826765 Don't use __builtin_alloca_with_align with nvc++ by Morris Hafner · 3 weeks ago
  26. 84c446d Fix macro redefinition warning in FFTW test by Tyler Veness · 3 weeks ago
  27. a9584d8 Fix clang6 failures. by Antonio Sánchez · 3 weeks ago
  28. dd4c280 Fix clang6 failures. by Antonio Sánchez · 3 weeks ago
  29. 9e962d9 Fix OOB access in triangular matrix multiplication. by Antonio Sánchez · 3 weeks ago
  30. 695e49d Fix NVCC builds for CUDA 10+. by Antonio Sánchez · 3 weeks ago
  31. dae0977 Don't pass matrices by value. by Antonio Sánchez · 3 weeks ago
  32. c23ec34 Add tests for sizeof() with one dynamic dimension. by Rasmus Munk Larsen · 3 weeks ago
  33. 58b252e Fix typo in PacketMath.h by Rasmus Munk Larsen · 3 weeks ago
  34. 6c04d0c Add missing exp2 definition for Altivec. by Rasmus Munk Larsen · 3 weeks ago
  35. b15ebb1 add nextafter for bfloat16 by Peter Gavin · 4 weeks ago
  36. 53b83cd Include <type_traits> in main.h for std::is_trivial* by Rasmus Munk Larsen · 4 weeks ago
  37. 3756385 Fix stack allocation assert by Charles Schlosser · 4 weeks ago
  38. 3f067c4 Add exp2() as a packet op and array method. by Rasmus Munk Larsen · 4 weeks ago
  39. 4e5136d make fixed size matrices and arrays trivially_default_constructible by Charles Schlosser · 4 weeks ago
  40. b396a6f Add free-function swap. by Antonio Sánchez · 5 weeks ago
  41. 820e8a4 add compile time info to reverse in place by Charles Schlosser · 6 weeks ago
  42. b55dab7 Fix DenseBase::tail for Dynamic template argument by Charles Schlosser · 6 weeks ago
  43. e0cbc55 Update README.md by Charles Schlosser · 6 weeks ago
  44. 7eea0a9 Vectorize erfc() for float by Rasmus Munk Larsen · 6 weeks ago
  45. 78f3c65 Don't use constexpr with half. by Rasmus Munk Larsen · 6 weeks ago
  46. 6d7af23 Adjust array_cwise for 32-bit arm. by Antonio Sánchez · 6 weeks ago
  47. 74dcfbb Use ppolevl for polynomial evaluation in more places. by Rasmus Munk Larsen · 6 weeks ago
  48. a097f72 Avoid producing erf(x) = NaN for large |x|. by Rasmus Munk Larsen · 7 weeks ago
  49. 44b16f4 Improve speed and accuracy or erf() by Rasmus Munk Larsen · 7 weeks ago
  50. 12068cb Fix inverse evaluator for running on CUDA device. by Antonio Sánchez · 7 weeks ago
  51. 4e8e5e7 Add max_digits10 in NumTraits for mpreal types. by Rasmus Munk Larsen · 7 weeks ago
  52. 8e8c319 Add missing EIGEN_DEVICE_FUNC annotations. by Rasmus Munk Larsen · 7 weeks ago
  53. 7ad7c1d fix implicit conversion warning (again) by Charles Schlosser · 8 weeks ago
  54. d052b7f add extra debugging info to float_pow_test_impl, clean up array_cwise tests by Charles Schlosser · 8 weeks ago
  55. ba5183f fix warning in EigenSolver::pseudoEigenvalueMatrix() by Charles Schlosser · 8 weeks ago
  56. 3ffb4e5 fix implicit conversion in TensorChipping by Charles Schlosser · 8 weeks ago
  57. b6b8b54 Fixed issue #2858: removed unneeded call to _mm_setzero_si128 by Sean McBride · 8 weeks ago
  58. 2a34651 Refactor code to use constexpr for data() functions. by Frédéric BRIOL · 8 weeks ago
  59. 2d4c9b4 make fixed size matrices and arrays trivially_copy_constructible and trivially_move_constructible by Charles Schlosser · 9 weeks ago
  60. 132f281 Fix generic ceil for SSE2. by Antonio Sánchez · 10 weeks ago
  61. 84282c4 optimize new dot product by Charles Schlosser · 2 months ago
  62. fb477b8 Better dot products by Charles Schlosser · 2 months ago
  63. 134b526 Update NonBlockingThreadPool.h plain asserts to use eigen_plain_assert by Sophie Chang · 2 months ago
  64. 072ec9d Fix a bug for pcmp_lt_or_nan and Add sqrt support for SVE by qile lin · 3 months ago
  65. 9315389 Fix bug in bug fix for atanh. by Rasmus Munk Larsen · 3 months ago
  66. f33af05 Fix bug for atanh(-1). by Rasmus Munk Larsen · 3 months ago
  67. 66927f7 Fix out-of-range arguments to _mm_permute_pd. by Rasmus Munk Larsen · 3 months ago
  68. bbdabeb Vectorize atanh<double>. Make atanh(x) standard compliant for |x| >= 1. by Rasmus Munk Larsen · 3 months ago
  69. 26e2c4f Add nvc++ support by Morris Hafner · 3 months ago
  70. c59332d Detect "effectively inner/outer" chipping in TensorChipping by Eugene Zhulenev · 3 months ago
  71. 648bce6 SSE/AVX Complex FMA by Charles Schlosser · 3 months ago
  72. c21a80b BDCSVD: Suppress Wmaybe-uninitialized by Charles Schlosser · 3 months ago
  73. 9d3d37c Complex Numtraits::HasSign and nmsub test by Charles Schlosser · 3 months ago
  74. c5189ac Fix GeneralizedEigenSolver::eigenvectors() not appearing in documentation by Valentin Sarthou · 3 months ago
  75. 3b5a1b4 sve instrinsics with "_x" suffix will be faster than "_z" suffix by qile lin · 3 months ago
  76. 98f1ac5 Fix breakage in GPU build. by Rasmus Munk Larsen · 3 months ago
  77. 231308f TensorVolumePatchOp: Suppress Wmaybe-uninitialized caused by unreachable code by Charles Schlosser · 3 months ago
  78. 2bf8fe1 NEON Complex Intrinsics by Tobias Wood · 3 months ago
  79. f91f8e9 Consolidate float and double implementations of patan(). by Rasmus Munk Larsen · 3 months ago
  80. 87239e0 vectorize squaredNorm() for complex types by Charles Schlosser · 3 months ago
  81. 32d95bb Add vectorized implementation of tanh<double> by Rasmus Munk Larsen · 3 months ago
  82. cc240ee Speed up and improve accuracy of tanh. by Rasmus Munk Larsen · 3 months ago
  83. 92e373e Speed up StableNorm for non-trivial sizes and improve consistency between aligned and unaligned inputs. by Rasmus Munk Larsen · 3 months ago
  84. 1dbc758 Include <thread> for std::this_thread::yield(). by Rasmus Munk Larsen · 3 months ago
  85. ab31094 Add a yield instruction in the two spinloops of the threaded matmul implementation. by Rasmus Munk Larsen · 3 months ago
  86. 99ffad1 A few cleanups to threaded product code and test. by Rasmus Munk Larsen · 3 months ago
  87. 59498c9 SSE/AVX use fmaddsub for complex products by Charles Schlosser · 4 months ago
  88. 1dcae7c Revert "BDCSVD fix -Wmaybe-uninitialized" by Rasmus Munk Larsen · 4 months ago
  89. d14b0a4 Remove C++23 check around has_denorm deprecation suppression by Tyler Veness · 4 months ago
  90. 24db460 hlog symbol lookup should not restricted to global namespace by Jatin Chaudhary · 4 months ago
  91. 767e60e Fix Woverflow warnings in PacketMathFP16 by Alexander Grund · 4 months ago
  92. 8025683 Fix conversion of `Eigen::half` to `_Float16` in AVX512 code by Alexander Grund · 4 months ago
  93. ec18dd0 fix pi in kissfft by Alexey Korepanov · 4 months ago
  94. 2b7b7aa Speed up complex * complex matrix multiplication. by Rasmus Munk Larsen · 4 months ago
  95. b3e3b7b Remove implicit this capture in lambdas by Devon Loehr · 4 months ago
  96. e44db21 Optimize ThreadPool spinning by Eugene Zhulenev · 4 months ago
  97. c593e9e Fix typos by Mike Taves · 4 months ago
  98. fd98cc4 Avoid atomic false sharing in RunQueue by Eugene Zhulenev · 4 months ago
  99. 0b646f3 Update file .clang-format by Charles Schlosser · 4 months ago
  100. 1dcb07b Update file eigen_navtree_hacks.js by Charles Schlosser · 4 months ago