Commit Graph

  • 014f12f11a GPU: Add BLAS-1 ops, DeviceScalar, device-resident SpMV, and CG interop (5/5) gpu-cg-interop Rasmus Munk Larsen 2026-04-09 19:54:13 -07:00
  • 43a95b62bb GPU: Add sparse solvers, FFT, and SpMV (cuDSS, cuFFT, cuSPARSE) gpu-sparse-fft-spmv Rasmus Munk Larsen 2026-04-09 19:11:49 -07:00
  • 8593c7f5a1 GPU: Add dense cuSOLVER solvers (QR, SVD, EigenSolver) gpu-dense-solvers Rasmus Munk Larsen 2026-04-09 19:11:25 -07:00
  • 58c44ef36d GPU: Add library dispatch module (DeviceMatrix, cuBLAS, cuSOLVER) gpu-library-dispatch Rasmus Munk Larsen 2026-04-09 16:15:39 -07:00
  • 6a9405bf7a GPU: Raise CUDA/HIP minimum and remove legacy guards gpu-modernize-minimum-versions Rasmus Munk Larsen 2026-04-09 13:38:04 -07:00
  • e055e4e415 Add plog_core_double with fallback for AVX without AVX2 master Rasmus Munk Larsen 2026-04-08 19:41:07 -07:00
  • 111c4d23a9 Revert "Revert "Speed up plog_double ~1.7x with fast integer range reduction"" revert-b1d2ce4c Rasmus Munk Larsen 2026-04-08 13:10:27 -07:00
  • b1d2ce4c85 Revert "Speed up plog_double ~1.7x with fast integer range reduction" Rasmus Munk Larsen 2026-04-08 13:03:48 -07:00
  • ab70739c9c Speed up plog_double ~1.7x with fast integer range reduction Rasmus Munk Larsen 2026-04-07 21:48:25 -07:00
  • e778b5d22b Switch ASAN/UBSAN smoketest pipelines to large runners Rasmus Munk Larsen 2026-04-07 21:37:58 -07:00
  • def45c5e1e Improve psincos_double: faster polynomials + accurate range reduction Rasmus Munk Larsen 2026-04-07 21:24:24 -07:00
  • 110530a4d8 Fix bugs and improve robustness of SelfAdjointEigenSolver, improve test coverage Rasmus Munk Larsen 2026-04-07 21:08:29 -07:00
  • bde3a68bae Improve dense linear solver docs with practical guidance Rasmus Munk Larsen 2026-04-05 21:40:42 -07:00
  • 8eabfb5342 Vectorize BLAS level 1/2 routines with Eigen expressions Rasmus Munk Larsen 2026-04-05 18:53:11 -07:00
  • 4ad90a60f1 Replace blas/f2c with clean C++ implementations Rasmus Munk Larsen 2026-04-05 16:04:41 -07:00
  • 8346cc3410 Fix three bugs in SelfAdjointEigenSolver and improve test coverage selfadjoint-eigensolver-audit Rasmus Munk Larsen 2026-04-04 18:02:37 -07:00
  • 93e9970964 Run clang-format on bench_small_matrix.cpp Rasmus Munk Larsen 2026-04-04 15:28:20 -07:00
  • 3eed3b0ab9 Fix Gram-Schmidt bug in SelfAdjointEigenSolver::computeDirect and add small matrix benchmarks Rasmus Munk Larsen 2026-04-04 15:24:42 -07:00
  • 8ddbe44799 Add small fixed-size matrix benchmarks for robotics/CV workloads Rasmus Munk Larsen 2026-04-04 15:06:32 -07:00
  • fe6ada10be Prevent nightly CI pipelines from being auto-cancelled Rasmus Munk Larsen 2026-04-04 11:52:15 -07:00
  • 8179474225 CI: Add AVX512-FP16 build tests with GCC 13 Alexander Grund 2026-04-04 18:32:31 +00:00
  • b57d860f3e Fix GCC maybe-uninitialized warning in InnerProduct Florian Maurin 2026-04-04 02:41:09 +00:00
  • a3074053a6 Speed up pexp_double by ~15-17% Rasmus Munk Larsen 2026-04-03 17:09:11 -07:00
  • a91913e961 Speed up plog_float by 1.6x with improved accuracy Rasmus Munk Larsen 2026-04-03 13:45:01 -07:00
  • ebae0c7c10 ulp_accuracy: use dynamic work queue for thread load balancing Rasmus Munk Larsen 2026-04-02 22:40:03 -07:00
  • 5977635d64 fix singed integer overflow UB in integer_types and other trivial compiler warnings Charles Schlosser 2026-04-03 03:36:28 +00:00
  • 60df12437e Fix ulp_accuracy crashes in Release builds Rasmus Munk Larsen 2026-04-02 20:12:13 -07:00
  • e315a8cdd0 Inline IndexedViewMethods.inc into DenseBase.h Pavel Guzenfeld 2026-04-02 22:26:56 +00:00
  • 8ec68856a6 Fix basicstuff_8 casting test failure on loongarch64 Rasmus Munk Larsen 2026-04-02 14:14:54 -07:00
  • 61a8662876 Improve log1p accuracy and speed with direct range reduction Rasmus Munk Larsen 2026-04-02 11:29:25 -07:00
  • d31a73437f Vectorize asinh and acosh for float and double Rasmus Munk Larsen 2026-04-01 21:46:36 -07:00
  • 9513d3878e Vectorize sinh, cosh, and log10 Rasmus Munk Larsen 2026-04-01 20:41:18 -07:00
  • 30e669cfe1 Tensor module: const-correctness and constexpr improvements Rasmus Munk Larsen 2026-04-01 17:49:56 -07:00
  • 64885cc6a3 Fix remaining MSVC warnings in Windows CI (C4804, C4244, C4146, C4305) Rasmus Munk Larsen 2026-04-01 17:20:31 -07:00
  • 6a07970d7d CI: split NVHPC build and make fallback parallelism configurable Rasmus Munk Larsen 2026-04-01 16:43:33 -07:00
  • 4be66f2830 CI: fail test jobs when no tests are found (--no-tests=error) Rasmus Munk Larsen 2026-04-01 12:50:54 -07:00
  • 1df89cbc21 Right-size CI runners to reduce waste and shuffle build order to avoid OOM Rasmus Munk Larsen 2026-03-31 19:10:34 -07:00
  • b54640df19 Fix NVHPC warnings in Visitor.h and Memory.h Rasmus Munk Larsen 2026-03-31 15:09:37 -07:00
  • 7fcbed7acb Fill packet math coverage gaps across multiple architectures Rasmus Munk Larsen 2026-03-31 14:52:32 -07:00
  • 80ab2898e2 CI: install libclang-rt-14-dev for sanitizer smoketest Rasmus Munk Larsen 2026-03-31 00:16:18 -07:00
  • 798d7f2bec CI: drop Clang-6, bump base image to Ubuntu 24.04 and Clang 12 to 14 Rasmus Munk Larsen 2026-03-30 22:00:17 -07:00
  • 1ade3636b9 Fix BDCSVD bidiagonal hard-case failures on ARM with GCC Rasmus Munk Larsen 2026-03-30 20:17:37 -07:00
  • 801a9ee690 Fix ~1,460 MSVC warnings from generic code instantiated with bool Rasmus Munk Larsen 2026-03-29 21:05:49 -07:00
  • 806c7b6590 CI: fix Windows build cache key containing invalid path characters Rasmus Munk Larsen 2026-03-29 19:57:45 -07:00
  • 2776ba55eb Update slicing tutorial docs to reflect Eigen::placeholders namespace Rasmus Munk Larsen 2026-03-29 18:51:02 -07:00
  • 09581fda38 Modernize tensor contraction code: bug fixes, dead code removal, and cleanup Rasmus Munk Larsen 2026-03-29 18:03:06 -07:00
  • 732ebc8cc2 Modernize evaluator files Rasmus Munk Larsen 2026-03-29 17:40:39 -07:00
  • 255f522e2e Fix bugs, docs, and structure in unsupported/ public headers Rasmus Munk Larsen 2026-03-29 17:06:40 -07:00
  • bd276fbb28 Map .inc files to C++ in Doxygen extension mapping Pavel Guzenfeld 2026-03-29 23:48:13 +00:00
  • c8633ceeea Clean up top-level Eigen headers Rasmus Munk Larsen 2026-03-29 16:28:09 -07:00
  • 409296d91d Add nightly benchmark regression detection pipeline Rasmus Munk Larsen 2026-03-29 16:03:56 -07:00
  • 753a6ac5b3 Fix private shadowing of protected base members in iterative solvers Pavel Guzenfeld 2026-03-29 22:40:48 +00:00
  • 9fe2f03fa4 Revert "Lower BDCSVD crossover threshold from 16 to 8" Rasmus Munk Larsen 2026-03-29 15:25:09 -07:00
  • 12fe90db8b Lower BDCSVD crossover threshold from 16 to 8 Rasmus Munk Larsen 2026-03-29 14:33:22 -07:00
  • b7f6aed1b9 Fix dangling reference in IndexedView with expression indices Pavel Guzenfeld 2026-03-29 16:39:13 +00:00
  • 624ab58e8d Add bidiagonal SVD API to BDCSVD and remove dead debug code Rasmus Munk Larsen 2026-03-28 20:38:31 -07:00
  • ba9871e46b fix and enable realview unit tests Charles Schlosser 2026-03-29 03:13:54 +00:00
  • b8dab89663 CI: remove broken NVHPC CUDA pipeline Rasmus Munk Larsen 2026-03-28 19:52:01 -07:00
  • 0fe8cdfa3b Extract RankRevealingBase CRTP mixin to eliminate decomposition code duplication Rasmus Munk Larsen 2026-03-28 19:12:23 -07:00
  • 5e521f3e45 Revert "add realview test" Rasmus Munk Larsen 2026-03-28 17:27:01 -07:00
  • 87ae1dbe7f add realview test Charles Schlosser 2026-03-28 23:26:51 +00:00
  • 49a137ca24 CI: limit NVHPC build parallelism to avoid OOM kills Rasmus Munk Larsen 2026-03-28 16:10:04 -07:00
  • f928a9f534 Fix static alignment for generic clang vector backend Rasmus Munk Larsen 2026-03-28 15:50:58 -07:00
  • 9706546a14 Add Householder blocked-right regression test Rasmus Munk Larsen 2026-03-27 20:49:43 -07:00
  • 90ca5bfd9a Strip lapacke.h to only the declarations used by Eigen Pavel Guzenfeld 2026-03-28 03:16:46 +00:00
  • cf508c096b Add block Householder right-side application for HouseholderSequence Rasmus Munk Larsen 2026-03-27 19:56:08 -07:00
  • 79d7d280a5 Fix bugs in evaluator files Rasmus Munk Larsen 2026-03-27 18:25:51 -07:00
  • b8baa2c49c Split eigensolver_selfadjoint test to fix NVHPC OOM Rasmus Munk Larsen 2026-03-27 18:09:26 -07:00
  • eb4b2eeffa UBSAN: use appropriate SSE intrinsics for loading 4 and 8 bytes Charles Schlosser 2026-03-27 19:54:10 +00:00
  • 9939a4c6e3 Fix SparseLU and SparseQR for custom scalar types Tyler Veness 2026-03-27 00:13:11 -07:00
  • 002229ce47 Fix RowMajor gemm_pack_lhs for backends without half/quarter packets Rasmus Munk Larsen 2026-03-23 23:33:42 -07:00
  • f574cb9b18 Fix vectorization_logic test for generic clang backend Rasmus Munk Larsen 2026-03-22 22:12:57 -07:00
  • 843ffcec8b Fix warnings reported by NVHPC 26.1 Rasmus Munk Larsen 2026-03-22 11:43:40 -07:00
  • 71ef987edb Fixes triangular solves on indexed/sliced dense expressions Florian Maurin 2026-03-22 18:12:21 +00:00
  • ac6aedc60a Fix flaky matrix_power test Rasmus Munk Larsen 2026-03-22 09:54:32 -07:00
  • 6490b17e6f Fix sanitizer regressions in sparse serializer and packet tests Rasmus Munk Larsen 2026-03-22 09:10:16 -07:00
  • 835e5615a9 Prefer SuiteSparse config-mode packages in Find modules Pavel Guzenfeld 2026-03-22 15:44:01 +00:00
  • f5774b014e Fix Doxygen build failure for comparison operator links Rasmus Munk Larsen 2026-03-21 18:48:07 -07:00
  • a0e30732a7 Remove trailing semicolon from EIGEN_UNUSED_VARIABLE macro Pavel Guzenfeld 2026-03-21 23:54:13 +00:00
  • e0b8498eef CI: Add nightly clang C++20 full test pipeline Rasmus Munk Larsen 2026-03-21 11:10:58 -07:00
  • 7e8a3040bb Fix Doxygen errors for ArrayBase comparison operators Rasmus Munk Larsen 2026-03-21 10:37:17 -07:00
  • 54b04fc6b1 Fix mixed-type GEMM packing for backends without half/quarter packets Rasmus Munk Larsen 2026-03-21 09:46:54 -07:00
  • 1d21d62fbc Fix computeInverseAndDetWithCheck for dynamic result matrices Pavel Guzenfeld 2026-03-21 15:38:27 +00:00
  • cc8c7cf0e6 Fix bugs and clean up SparseCore module Rasmus Munk Larsen 2026-03-21 06:31:12 -07:00
  • daecd28cd5 Add Array relational operator docs and FetchContent CMake guide Pavel Guzenfeld 2026-03-21 01:50:58 +00:00
  • 9d1e5f3915 Remove benchmark::internal::Benchmark* from all benchmarks Rasmus Munk Larsen 2026-03-20 17:42:07 -07:00
  • 8115b45e50 Fix integer sanitizer issues in shifts and test ranges Rasmus Munk Larsen 2026-03-20 17:27:02 -07:00
  • 89621d1024 CI: Remove GCC 6 pipeline Rasmus Munk Larsen 2026-03-20 17:09:19 -07:00
  • 6540bf4787 Harden unsupported tensor tests for sanitizers Rasmus Munk Larsen 2026-03-20 15:12:41 -07:00
  • 9d161e0c87 Fine-tune gebp_kernel for aarch64 Yu You 2026-03-20 14:29:03 -07:00
  • a0b16a7e1b Fix flaky product and eigensolver_selfadjoint tests Rasmus Munk Larsen 2026-03-20 13:44:03 -07:00
  • a72172e563 Add blocking and vectorization boundary tests for LU and Cholesky Pavel Guzenfeld 2026-03-20 20:27:49 +00:00
  • 30128de0e3 Guard eigen_fill_helper on trivially copyable scalars Pavel Guzenfeld 2026-03-20 19:03:13 +00:00
  • 8a47aa334b Replace empirical product test tolerances with principled Higham-Mary bounds Rasmus Munk Larsen 2026-03-20 11:03:58 -07:00
  • 821ab7d3e6 Fix TensorUInt128 division infinite loop on overflow Pavel Guzenfeld 2026-03-20 15:41:00 +00:00
  • 3578883bb3 CI: Split ASAN/UBSAN build into official/unsupported jobs Rasmus Munk Larsen 2026-03-19 17:56:43 -07:00
  • 3e5a2f9245 Fix vectorized erf returning NaN at ±inf instead of ±1 Pavel Guzenfeld 2026-03-19 21:12:15 +00:00
  • 36ca36d0de Guard redundant constexpr static member redeclarations for C++17+ Pavel Guzenfeld 2026-03-19 03:24:09 +00:00
  • 62e23f79dd Fix GCC 13 array-bounds warning in TensorContraction Pavel Guzenfeld 2026-03-19 03:08:21 +00:00
  • 05295a818b Fix undefined behavior in matrix_cwise test for signed integers Pavel Guzenfeld 2026-03-18 18:51:01 +00:00