Commit Graph

  • 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
  • ba4d7304e2 Document DiagonalBase Arthur 2022-06-08 17:46:32 +00:00
  • 95463b59bc Mark index_remap as EIGEN_DEVICE_FUNC in src/Core/Reshaped.h (Fixes #2493) Binhao Qin 2022-06-07 20:10:47 +00:00
  • 28812d2ebb AVX512 TRSM Kernels respect EIGEN_NO_MALLOC Shi, Brian 2022-06-06 17:03:10 -07:00
  • 9960a30422 Fix row vs column vector typo in Matrix class tutorial sfalmo 2022-06-07 17:28:19 +00:00
  • 8c2e0e3cb8 Fix ambiguous comparisons for c++20 (again again) Antonio Sánchez 2022-06-07 17:06:17 +00:00
  • 14aae29470 Provide DiagonalMatrix Product and Initializers Arthur 2022-06-06 21:43:22 +00:00
  • 76cf6204f3 Revert "Fix c++20 ambiguity of comparisons." Antonio Sánchez 2022-06-04 02:32:10 +00:00
  • 8fbb76a043 Fix build issues with MSVC for AVX512 aaraujom 2022-06-03 14:55:40 +00:00
  • 4f6354128f Fix c++20 ambiguity of comparisons. Antonio Sánchez 2022-06-03 05:11:07 +00:00
  • f542b0a71f Adding an MKL adapter in FFT module. Oleg Shirokobrod 2022-06-02 18:10:43 +00:00
  • d49ede4dc4 Add AVX512 s/dgemm optimizations for compute kernel (2nd try) aaraujom 2022-05-28 02:00:21 +00:00
  • 510f6b9f15 Fix integer shortening warnings in visitor tests. Rasmus Munk Larsen 2022-05-27 18:51:37 +00:00
  • 705ae70646 Add R-Bidiagonalization step to BDCSVD Arthur 2022-05-27 02:00:24 +00:00
  • e99163e732 fix: issue 2481: LDLT produce wrong results with AutoDiffScalar Mario Rincon-Nigro 2022-05-25 15:26:10 +00:00
  • 477eb7f630 Revert "Avoid ambiguous Tensor comparison operators for C++20 compatibility" Antonio Sánchez 2022-05-24 16:09:59 +00:00
  • c5a5ac680c [SYCL] SYCL-2020 range does not have default constructor. Mehdi Goli 2022-05-24 03:11:46 +00:00
  • 5c2179b6c3 Avoid ambiguous Tensor comparison operators for C++20 compatibility Benjamin Kramer 2022-05-23 17:36:03 +00:00
  • aa8b7e2c37 Add subMappers to Power GEMM packing - simplifies the address calculations (10% faster) Chip Kerchner 2022-05-23 15:18:29 +00:00
  • 32348091ba Avoid signed integer overflow in adjoint test. Antonio Sánchez 2022-05-23 14:46:16 +00:00
  • cbe03f3531 [SYCL] Extending SYCL queue interface extension. Mehdi Goli 2022-05-23 14:45:27 +00:00
  • 32a3f9ac33 Improve plogical_shift_* implementations and fix typo in SVE/PacketMath.h Guoqiang QI 2022-05-23 09:32:32 +00:00
  • ac5c83a3f5
    unset executable flag Eisuke Kawashima 2022-05-22 22:46:53 +09:00
  • 481a4a8c31 Fix BDCSVD condition for failing with numerical issue. Antonio Sanchez 2022-05-20 08:17:07 -07:00
  • a9868bd5be Add arg() to tensor Tobias Wood 2022-05-17 12:01:39 +01:00
  • e7248b26a1 Prevent BDCSVD crash caused by index out of bounds. Antonio Sánchez 2022-05-19 22:29:48 +00:00
  • 028ab12586 Prevent BDCSVD crash caused by index out of bounds. Antonio Sánchez 2022-05-19 22:29:48 +00:00
  • 798fc1c577 Fix 'Incorrect reference code in STL_interface.hh for ata_product' eigen/isses/2425 Rohan Ghige 2022-05-18 14:42:57 +00:00
  • 9b9496ad98 Revert "Add AVX512 optimizations for matrix multiply" Antonio Sánchez 2022-05-13 18:50:33 +00:00
  • 25db0b4a82 Add AVX512 optimizations for matrix multiply aaraujom 2022-05-12 23:41:19 +00:00
  • 00b75375e7 Adding PocketFFT support in FFT module since kissfft has some flaw in accuracy and performance Guoqiang QI 2022-05-11 17:44:22 +00:00
  • 73d65dbc43 Update README.md. Remove obsolete comment about RowMajor not being fully supported. Rasmus Munk Larsen 2022-05-06 18:19:35 +00:00
  • 68e03ab240 Add uninstall target only if not already defined. Francesco Romano 2022-05-05 17:43:08 +00:00
  • 2c055f8633 make diagonal matrix cols() and rows() methods constexpr Alex_M 2022-05-03 06:32:48 +00:00
  • c2f15edc43 Add load vector_pairs for RHS of GEMM MMA. Improved predux GEMV. Chip Kerchner 2022-04-25 16:23:01 +00:00
  • 9e026e5e28 Removed need to supply the Symmetric flag to UpLo argument for Accelerate LLT and LDLT John Mather 2022-04-21 20:02:10 +00:00
  • 44ba7a0da3 Fix compiler bugs for GCC 10 & 11 for Power GEMM Chip Kerchner 2022-04-20 15:59:00 +00:00
  • b02c384ef4 Add fused multiply functions for PowerPC - pmsub, pnmadd and pnmsub Chip Kerchner 2022-04-18 16:16:32 +00:00
  • 3de96caeaa Fix HouseholderSequence.h Rohit Santhanam 2022-04-17 02:46:56 +00:00
  • a1e1612c28 Fix cwise NaN propagation for scalar input. Antonio Sanchez 2022-04-15 21:45:53 -07:00
  • f845a8bb1a Fix cwise NaN propagation for scalar input. Antonio Sánchez 2022-04-16 05:07:44 +00:00
  • a4bb513b99 Update HouseholderSequence.h Charles Schlosser 2022-04-15 16:56:17 +00:00
  • fc1d888415 Remove AVX512VL dependency in trsm Shi, Brian 2022-04-14 11:35:26 -07:00
  • 07db964bde Restrict new AVX512 trsm to AVX512VL, rename files for consistency. Antonio Sánchez 2022-04-14 16:58:32 +00:00
  • 67eeba6e72 Avoidable heap allocation in applyHouseholderToTheLeft Charles Schlosser 2022-04-13 18:45:36 +00:00
  • 3342fc7e4d Allow all tests to pass with EIGEN_TEST_NO_EXPLICIT_VECTORIZATION Antonio Sánchez 2022-04-12 14:48:22 +00:00
  • efb08e0bb5 Revert "Fix ambiguous DiagonalMatrix constructors." Antonio Sánchez 2022-04-12 03:54:31 +00:00
  • 53eec53d2a Fix Power GEMV order of operations in predux for MMA. Chip Kerchner 2022-04-11 21:29:05 +00:00
  • a81bba962a Fix ambiguous DiagonalMatrix constructors. Antonio Sánchez 2022-04-11 19:13:25 +00:00
  • f3aaba8705 Revert "Replace call to FixedDimensions() with a singleton instance of" Antonio Sánchez 2022-04-10 15:30:33 +00:00
  • f7b31f864c Revert "Replace call to FixedDimensions() with a singleton instance of" Antonio Sánchez 2022-04-10 15:30:33 +00:00
  • f3ba220c5d Remove EIGEN_EMPTY_STRUCT_CTOR Tobias Schlüter 2022-04-07 09:30:51 +09:00
  • 5ed7a86ae9 Fix MSVC+CUDA issues. Antonio Sánchez 2022-04-08 18:05:32 +00:00
  • 734ed1efa6 Fix ODR issues in lapacke_helpers. Antonio Sánchez 2022-04-08 15:31:30 +00:00
  • 2c45a3846e Fix some max size expressions. Antonio Sánchez 2022-04-06 22:19:57 +00:00
  • edc822666d Fix navbar scroll with toc. Antonio Sánchez 2022-04-05 20:14:22 +00:00
  • df87d40e34 constexpr reshape helper Erik Schultheis 2022-04-05 17:32:17 +00:00
  • 403fa33409 Performance improvements in GEMM for Power Chip Kerchner 2022-04-05 12:18:53 +00:00
  • e1df3636b2 More constexpr helpers Erik Schultheis 2022-04-04 18:38:34 +00:00
  • 64909b82bd static const class members turned into constexpr Erik Schultheis 2022-04-04 17:33:33 +00:00
  • 2c0ef43b48 Added Scaling function overload for vector rvalue reference William Talbot 2022-04-04 16:50:09 +00:00
  • ba2cb835aa Add back std::remove* aliases - third-party libraries rely on these. Antonio Sanchez 2022-03-31 15:08:01 -07:00
  • 0c859cf35d Consider inf/nan in scalar test_isApprox. Antonio Sánchez 2022-04-01 17:00:24 +00:00
  • 1ddd3e29cb fixed order of arguments in blas syrk Erik Schultheis 2022-03-26 14:39:31 +02:00
  • 2c56442805 Don't include .cpp in lapack. Antonio Sánchez 2022-03-30 21:41:56 +00:00
  • 73b2c13bf2 Disable f16c scalar conversions for MSVC. Antonio Sánchez 2022-03-30 18:35:32 +00:00
  • 9bc9992dd3 Eliminate trace unused warning. Antonio Sanchez 2022-03-29 14:20:13 -07:00
  • e22d58e816 Add is_constant_evaluated, update alignment checks Tobias Schlüter 2022-03-25 04:00:58 +00:00
  • f0a91838aa Enable Aarch64 CI Everton Constantino 2022-03-24 16:46:42 -03:00
  • b9d2900e8f added a missing typename and fixed a unused typedef warning Erik Schultheis 2022-03-24 12:07:18 +02:00
  • 0611f7fff0 Add missing explicit reinterprets b-shi 2022-03-23 21:10:26 +00:00
  • cd3c81c3bc Add a NNLS solver to unsupported - issue #655 Essex Edwards 2022-03-23 20:20:44 +00:00
  • 0699fa06fe Split general_matrix_vector_product interface for Power into two macros - one ColMajor and RowMajor. Chip Kerchner 2022-03-23 18:09:33 +00:00
  • 19a6a827c4 Optimize visitor traversal in case of RowMajor. Antonio Sánchez 2022-03-23 15:27:57 +00:00
  • 34e5f34b39 Update warning suppression to latest. Antonio Sánchez 2022-03-21 15:56:03 +00:00
  • f2a3e03e9b Fix usages of wrong namespace Romain Biessy 2022-03-21 15:07:53 +00:00
  • 4451823fb4 Fix ODR violation in trsm. Antonio Sánchez 2022-03-20 15:56:53 +00:00
  • 9a14d91a99 Fix AVX512 builds with MSVC. Antonio Sánchez 2022-03-18 16:04:53 +00:00
  • 7b10795e39 Change EIGEN_ALTIVEC_ENABLE_MMA_DYNAMIC_DISPATCH and EIGEN_ALTIVEC_DISABLE_MMA flags to be like TensorFlow's... Chip Kerchner 2022-03-17 22:35:27 +00:00
  • 3ca1228d45 Work around MSVC compiler bug dropping const. Antonio Sánchez 2022-03-17 20:50:26 +00:00
  • 40eb34bc5d Fix RowMajorBit <-> RowMajor mixup. Tobias Schlüter 2022-03-17 15:28:12 +00:00
  • c062983464 Completed a missing parenthesis in tutorial. Øystein Sørensen 2022-03-17 14:52:07 +00:00
  • 9deaa19121 Work around g++-10 docker issue for geo_orthomethods_4. Antonio Sánchez 2022-03-16 21:46:04 +00:00
  • e34db1239d Fix missing pound Antonio Sanchez 2022-03-16 12:26:12 -07:00
  • 591906477b Fix up PowerPC MMA flags so it builds by default. Antonio Sánchez 2022-03-16 19:16:28 +00:00
  • 518fc321cb AVX512 Optimizations for Triangular Solve b-shi 2022-03-16 18:04:50 +00:00
  • 01b5bc48cc Disable schur non-convergence test. Antonio Sánchez 2022-03-16 17:33:53 +00:00
  • 421cbf0866 Replace Eigen type metaprogramming with corresponding std types and make use of alias templates Erik Schultheis 2022-03-16 16:43:40 +00:00
  • 514f90c9ff Remove workarounds for bad GCC-4 warnings Arthur 2022-03-16 00:08:16 +00:00
  • 9ad5661482 Revert "Fix up PowerPC MMA flags so it builds by default." Rasmus Munk Larsen 2022-03-15 20:51:03 +00:00
  • 65eeedf964 Fix up PowerPC MMA flags so it builds by default. Antonio Sánchez 2022-03-15 20:22:23 +00:00
  • cb1e8228e9 Convert bit calculation to constexpr, avoid casts. Tobias Schlüter 2022-03-13 22:27:06 +09:00
  • baf9a985ec Fix swap test for size 1 inputs. Antonio Sánchez 2022-03-10 15:05:58 +00:00
  • 7882408856 Temporarily disable aarch64 CI. Everton Constantino 2022-03-10 09:34:19 -03:00
  • 2a6be5492f Fix construct_at compilation breakage on ROCm. Rohit Santhanam 2022-03-09 16:47:53 +00:00
  • a3b64625e3 Remove ComputeCpp-specific code from SYCL Vptr Duncan McBain 2022-03-08 22:44:18 +00:00
  • 9296bb4b93 Fix edge-case in zeta for large inputs. Antonio Sánchez 2022-03-08 21:21:20 +00:00
  • cd2ba9d03e Add construct_at, destroy_at wrappers. Use throughout. Tobias Schlüter 2022-03-08 20:43:22 +00:00