Google Git Google Git
Sign in
eigen/mirror/HEAD
  1. 7177d99 Doc: Warn that product results depend on the detected cache sizes by Rasmus Munk Larsen · 12 hours ago master
  2. cb842af Jacobi: Keep complex Givens rotations unitary for subnormal inputs by DrWisdome · 15 hours ago
  3. 584581e NEON: Add Packet8bf as the full bfloat16 packet by Charles Schlosser · 15 hours ago
  4. c848e90 Tensor: Handle oversized GPU convolution kernels by Florian Maurin · 16 hours ago
  5. 0a746d8 Tests: Derive the subnormal betainc tolerance from the exponent's terms by Rasmus Munk Larsen · 16 hours ago
  6. 3081d7a CI: Point relocated Windows test jobs at their own build tree by Rasmus Munk Larsen · 16 hours ago
  7. 68cb54e CI: Pin the nvhpc builds to nvc++'s generic x86-64 target by Rasmus Munk Larsen · 16 hours ago
  8. d194702 Riccati: Add a discrete-time algebraic Riccati equation solver by DrWisdome · 18 hours ago
  9. 674e4a4 Eigenvalues: Start ComplexQZ sweeps below two small subdiagonals by Rasmus Munk Larsen · 20 hours ago
  10. e69e1e7 Core: Keep GEBP accumulators in registers with clang at AVX2 by Rasmus Munk Larsen · 21 hours ago
  11. 656945e Core: Let GEMM overwrite the destination for C = A*B instead of zeroing it first by Kasper Nielsen · 32 hours ago
  12. d8220da Core: Clean up moved elements when reallocation throws by Florian Maurin · 2 days ago
  13. 4ae021a Core: Iterate one-packet GEBP row panels inside the column loop by Rasmus Munk Larsen · 2 days ago
  14. f9dcf76 Core: Use portable math overloads for half and bfloat16 by Florian Maurin · 3 days ago
  15. 84f7421 Cholesky: Pivot LDLT on the Schur complement's diagonal, and block it by Rasmus Munk Larsen · 3 days ago
  16. f3982a9 Test: Check AutoDiff scalar derivatives analytically by Florian Maurin · 3 days ago
  17. 451bd23 Core: Work around a GCC 15.1/15.2 miscompile of mixed real/complex TRMV by Rasmus Munk Larsen · 3 days ago
  18. 2915a85 Splines: Avoid integer overflow in basis derivative scaling by Florian Maurin · 3 days ago
  19. aa93ded Tests: Bound the conditioning of svd_min_norm's generated systems by Rasmus Munk Larsen · 3 days ago
  20. 07965f9 Core: Keep complex GEBP accumulators in registers with 16 vector registers by Michael Tesch · 3 days ago
  21. e2ca51a Geometry: Reset last row when building Affine/Isometry Transform from matrix by Amirishetty Sai Vignesh · 4 days ago
  22. 0934a0b Core: Avoid spurious alignment warnings in GEMM packing storage by Florian Maurin · 4 days ago
  23. 394fb09 Scripts: Run the clang-tidy Claude Code hook in the background on C++ files by Rasmus Munk Larsen · 4 days ago
  24. eaf8a79 Build: Remove the demos and Qt4 support by Rasmus Munk Larsen · 4 days ago
  25. 8cd908a GPU: Use cudaDataType_t with cuDSS 1.0 by Rasmus Munk Larsen · 4 days ago
  26. ac17087 CMake: Standardize compiler and architecture detection for build names by Florian Maurin · 4 days ago
  27. e7ac941 NEON: Vectorize bfloat16 sqrt and rsqrt by Charles Schlosser · 4 days ago
  28. b17f29a Tests: Fix empty annotation stringification with MSVC by Florian Maurin · 4 days ago
  29. 01a0896 Core: Avoid copies when conservatively resizing vectors through two-argument overloads by Florian Maurin · 4 days ago
  30. e2fa2e8 Core: Fix first_aligned and unaligned complex vectorization by eigen-review-bot · 4 days ago
  31. fe23fa5 Core: Fix SYCL/CUDA include-order compatibility by Florian Maurin · 5 days ago
  32. d9b48d3 Core: Fix unused BLAS strides for singleton maps by Florian Maurin · 5 days ago
  33. dfcc848 Core: Work around GCC PR124138 in unrolled self-adjoint copies by Rasmus Munk Larsen · 5 days ago
  34. d020da3 NEON: Round float to bfloat16 with RSHRN by Charles Schlosser · 5 days ago
  35. e5820bb Core: Optimize single-column GEMV updates by Florian Maurin · 5 days ago
  36. 57e4f04 Eigenvalues: Protect GCC 10/11 Schur splitting from postreload GCSE by Rasmus Munk Larsen · 6 days ago
  37. 4b9786e Core: Keep AArch64 GCC GEBP accumulators in registers by Rasmus Munk Larsen · 6 days ago
  38. f9b7781 Deduplicate async TensorContractionEvaluatorBase template instantiations... by Ce Zheng · 6 days ago
  39. 2d273ed SME: Pack narrow panels into whole vectors by Rasmus Munk Larsen · 6 days ago
  40. ab73a04 Licensing: Label netlib LAPACK files as BSD-3-Clause-Open-MPI by Rasmus Munk Larsen · 6 days ago
  41. afeafad Tests: Keep ridders' tiny slopes normal where NEON flushes subnormals by Rasmus Munk Larsen · 6 days ago
  42. d0b1b7c Core: Fix fast-math trig reduction and optimization barrier spills by Rasmus Munk Larsen · 6 days ago
  43. ff931b4 Core: Coalesce heap packing buffers in GEMM and SYRK by Rasmus Munk Larsen · 6 days ago
  44. 2502f4c Tests: Scale the sparse determinant tolerance with conditioning by Rasmus Munk Larsen · 6 days ago
  45. 7a03903 Core: Use pmsub for the error term of the non-FMA twoprod by Charles Schlosser · 6 days ago
  46. 6be0f5c Core: Add in-place self-adjoint permutations and share Cholesky pivots by Rasmus Munk Larsen · 6 days ago
  47. 4d45fc8 Added asserts and templated .dimension() by Tobias Wood · 6 days ago
  48. 55ecf0c Test: Add verifyProductUpdate and bound product_sme's update checks absolutely by Kasper Nielsen · 6 days ago
  49. 1d1dec5 NumericalDiff: Implement Ridders' method by onalante-ebay · 7 days ago
  50. 73255cf Tests: Preserve subnormal beta reference values on ARMv7 by Rasmus Munk Larsen · 7 days ago
  51. 7525813 Core: Unroll cheap unary packet assignments with GCC by Rasmus Munk Larsen · 7 days ago
  52. 10658a6 Core: Resize fixed-capacity storage without copying scalars by Rasmus Munk Larsen · 7 days ago
  53. c62f1b7 Core: Respect scalar type alignment in scratch buffers by Florian Maurin · 7 days ago
  54. 3b4bb6d Prune unreachable rhs_inner_dim_reordered=true contraction instantiations by Ce Zheng · 7 days ago
  55. 312a349 Core: Bring phypot_complex within half an ulp and scale plog_complex only when needed by Charles Schlosser · 8 days ago
  56. 1da510b Core: Avoid speculative division by zero in complex_divide_smith by eigen-review-bot · 8 days ago
  57. 53e7993 Tensor: Use TensorCostModel overhead constants in numThreadsInnerDim by eigen-review-bot · 8 days ago
  58. 70b785e Core: Match scalar ptanh clamp threshold to pmadd FMA fusion by Antonio Sánchez · 8 days ago
  59. 3c3bf45 SME: Run small and thin products on the NEON kernel from the GEMM driver by Rasmus Munk Larsen · 9 days ago
  60. 454dd6f Core: Rename Complex alias in complex_exp2_impl to avoid X11 macro collision by eigen-review-bot · 9 days ago
  61. f3ec3aa Core: Test predicate segments shorter than a packet with one masked load, and OR integer packets in any() by Charles Schlosser · 9 days ago
  62. f28e5bf Core: Fix compiler warnings from the CI build logs by Charles Schlosser · 10 days ago
  63. 82d27c5 Core: Fix Clang packet rounding and complex division exceptions by Rasmus Munk Larsen · 10 days ago
  64. c0e5aab SparseCore: Fix random insertion when StorageIndex is wider than Index by Rasmus Munk Larsen · 10 days ago
  65. c44c4c8 Tests: Fix division-by-zero warning in packetmath_gather by Antonio Sánchez · 10 days ago
  66. 3502bd4 Tests: Return 0 from positive-control failtest binaries by Antonio Sánchez · 10 days ago
  67. 172426f GPU: Fix clang-cuda math overload ambiguity and unused variable warnings in tests by Antonio Sánchez · 10 days ago
  68. 3b70ae5 Core bug fixes: MSA, AltiVec, RVV, LSX and SVE backends by Rasmus Munk Larsen · 11 days ago
  69. ba3ca95 Core, SparseCore, Tensor: Fix some CI compiler warnings by Florian Maurin · 11 days ago
  70. f124c5f Enable constant folding for half(int) via float_to_half_rtne<true> by eigen-review-bot · 11 days ago
  71. 1039c3e SME: Scale by alpha before adding C in the NEON complex store by Michael Tesch · 11 days ago
  72. dbb99fe MSVC: Fix the MSVC 14.29 build of nine tests by Rasmus Munk Larsen · 11 days ago
  73. 1888fad Core bug fixes: reductions, norms and condition estimates by Rasmus Munk Larsen · 11 days ago
  74. 8bcd1d4 HVX: Improve reduction accuracy by Cheng Wang · 12 days ago
  75. e776f34 Fix Tensor thread-pool cost model, shuffling tiling, and contraction bandwidth by eigen-review-bot · 12 days ago
  76. 0a9871b Vectorize TensorSelectOp condition mask loads via Packet16b pcast by eigen-review-bot · 12 days ago
  77. de8b810 AVX-512 Pack32To16 / pstore cleanup and triangular stack allocation limit guard by eigen-review-bot · 12 days ago
  78. e993b2d SME: Store tail-panel lines whole by Kasper Nielsen · 12 days ago
  79. 2dd6575 StructuredMatrices: Worked examples from PDEs, signal processing, approximation, statistics and physics by Rasmus Munk Larsen · 12 days ago
  80. 12bc807 Core: Centralize shape checks and elide proven runtime assertions by Florian Maurin · 12 days ago
  81. 5f13c7a Updated NumericalDiff API to expose abs_tol and make setting rel_tol simpler by Carl Leake · 12 days ago
  82. 505f9cf Core: Fix C++14 linkage of metadata constants by Florian Maurin · 12 days ago
  83. e5e17be Core: Share empty-input dispatch for reductions by Florian Maurin · 12 days ago
  84. 43f0cac Core bug fixes: SSE, AVX and AVX-512 backends by Rasmus Munk Larsen · 12 days ago
  85. 78da6e8 Tests: Stage packetmath FP-exception operands through volatile objects by Rasmus Munk Larsen · 13 days ago
  86. fc7f87c CI: Sandbox Windows MSVC temp directory to prevent disk exhaustion by eigen-review-bot · 13 days ago
  87. ba45c44 CI: Fix the levenberg_marquardt, QEMU-timeout and nvhpc failures on master by Rasmus Munk Larsen · 13 days ago
  88. 791af28 Core: Define binary_floating_point_traits constants out of class for C++14 by Rasmus Munk Larsen · 13 days ago
  89. 29ed23b SparseCore: Compare sparse matrices without squaring their norms by Rasmus Munk Larsen · 13 days ago
  90. ae6c893 Fix unaligned packet loads in diagonal products by Antonio Sánchez · 14 days ago
  91. 81ccc4e Tests: Check always-defined platform macros with #if, not #ifdef by Rasmus Munk Larsen · 14 days ago
  92. 15bcfa7 Core bug fixes: scalar math, NumTraits and half/bfloat16 by Rasmus Munk Larsen · 14 days ago
  93. 7d254a4 Core bug fixes: expressions and views by Rasmus Munk Larsen · 14 days ago
  94. 8435869 Core bug fixes: storage, allocation and special matrices by Rasmus Munk Larsen · 14 days ago
  95. b20119f Core bug fixes: products, triangular and self-adjoint views by Rasmus Munk Larsen · 14 days ago
  96. 80778f5 Core bug fixes: vectorized math functions by Rasmus Munk Larsen · 14 days ago
  97. bc7719e SME: Single-vector stores in the full-panel transposer; four depth chunks per fill for 64-bit elements by Kasper Nielsen · 14 days ago
  98. 0089094 Core: Search all(), any(), allFinite() and isZero() with a blocked predicate search by Charles Schlosser · 2 weeks ago
  99. bc5d897 GPU bug fixes: device-side results and device-callable helpers by Rasmus Munk Larsen · 2 weeks ago
  100. 496efcc Core: Count bits with the CUDA intrinsics in device code by Rasmus Munk Larsen · 2 weeks ago