Gitiles
Code Review Sign In
gerrit-public.fairphone.software / platform / external / XNNPACK / 490febefc20fb13d66e98bdbd25615aaba276236
  1. 490febe Cortex A7 microkernel based on LD64 with PLD added. 3.2% faster in end to end mobilenet v2 by Frank Barchard · 4 years, 2 months ago
  2. 1483c53 WAsm SIMD version of F32 PAVGPOOL microkernels by Marat Dukhan · 4 years, 2 months ago
  3. 3b7432d WAsm SIMD versions of F32 AVGPOOL microkernels by Marat Dukhan · 4 years, 2 months ago
  4. efaac27 Prefix generated HSWISH microkernels with hswish- by Marat Dukhan · 4 years, 2 months ago
  5. f4935a2 Enable WAsm SIMD microkernels for Leaky ReLU by Marat Dukhan · 4 years, 2 months ago
  6. 9306ae0 WAsm SIMD version of X32 PAD microkernel by Marat Dukhan · 4 years, 2 months ago
  7. 52238f0 WAsm SIMD versions of RADDSTOREEXPMINUSMAX microkernels by Marat Dukhan · 4 years, 2 months ago
  8. 8ee3701 WAsm SIMD version of X32 FILL microkernel by Marat Dukhan · 4 years, 2 months ago
  9. 66d99e9 WAsm SIMD version of PACKX microkernel by Marat Dukhan · 4 years, 2 months ago
  10. b3635ed Port SIGMOID microkernels to WAsm SIMD by Marat Dukhan · 4 years, 2 months ago
  11. 19dd91d Prefix generated VLRELU microkernels with vlrelu- by Marat Dukhan · 4 years, 2 months ago
  12. 1827af1 Fix incompatibilities with MSVC in unit tests by Marat Dukhan · 4 years, 2 months ago
  13. b82b2cd WAsm SIMD conversion-based variants of VRND microkernels by Marat Dukhan · 4 years, 2 months ago
  14. 88da62c WAsm SIMD conversion-based variants of rounding evaluation stubs by Marat Dukhan · 4 years, 2 months ago
  15. 7829928 Reoptimize WAsm SIMD PReLU microkernels by Marat Dukhan · 4 years, 2 months ago
  16. 320cb46 Update CLAMP microkernels by Frank Barchard · 4 years, 2 months ago
  17. e6502da WAsm SIMD versions of Leaky ReLU microkernels by Marat Dukhan · 4 years, 2 months ago
  18. 04d6cf1 assert non-null pointer in CLAMP microkernels by Frank Barchard · 4 years, 2 months ago
  19. fb158e2 RELU microkernel to clamp values to 0 for a specialized clamp operator by Frank Barchard · 4 years, 2 months ago
  20. d816f62 WAsm SIMD versions of VMULCADDC microkernels by Marat Dukhan · 4 years, 2 months ago
  21. 08b7a97 Rename Q8 microkernels and operators to QU8 by Marat Dukhan · 4 years, 2 months ago
  22. 688f6d8 Unify x86 and ARM flavors of WAsm SIMD GEMM/IGEMM/DWCONV with RELU by Marat Dukhan · 4 years, 2 months ago
  23. 55dde5b NEON F32 HSWISH microkernel unrolled by 16 by Marat Dukhan · 4 years, 2 months ago
  24. 201ea0e Reoptimize F16 HSWISH microkernels by Marat Dukhan · 4 years, 2 months ago
  25. 9df9dc6 Reoptimize HSWISH microkernels by Marat Dukhan · 4 years, 2 months ago
  26. faa2e5c Fix CHW depthwise convolution in WAsm SIMD by Marat Dukhan · 4 years, 2 months ago
  27. d27202d Support Reshape Node in Subgraph API by Marat Dukhan · 4 years, 3 months ago
  28. 51a01c6 Support Square Root node in Subgraph API by Marat Dukhan · 4 years, 3 months ago
  29. 56772e5 Generalize Max Pooling operator creation across data types by Marat Dukhan · 4 years, 3 months ago
  30. 00d1d6e WAsm SIMD variants of F32 IBILINEAR microkernels by Marat Dukhan · 4 years, 3 months ago
  31. e39e646 WAsm SIMD versions of [I]GEMM microkernels with NR=2 by Marat Dukhan · 4 years, 3 months ago
  32. f6e2480 WAsm SIMD variants of F32 MAXPOOL microkernels by Marat Dukhan · 4 years, 3 months ago
  33. 3fa52c8 WAsm SIMD versions of F32 CLAMP microkernel by Marat Dukhan · 4 years, 3 months ago
  34. 8c41796 WAsm SIMD versions of F32 RMAX microkernel by Marat Dukhan · 4 years, 3 months ago
  35. efc1014 ld64 aarch32 GEMM 4x8 microkernel do all loads before MLA by Frank Barchard · 4 years, 3 months ago
  36. c8caea6 Generalize Convolution operator creation across data types by Marat Dukhan · 4 years, 3 months ago
  37. 5a698bb Generalize Deconvolution operator creation across data types by Marat Dukhan · 4 years, 3 months ago
  38. bc08f31 Generalize Fully Connected operator creation across data types by Marat Dukhan · 4 years, 3 months ago
  39. 8e229db Add scalar binary ops with linear activation by Frank Barchard · 4 years, 3 months ago
  40. b42f866 Unify interface of weights packing functions by Marat Dukhan · 4 years, 3 months ago
  41. 6543872 Improve error logging in NCHW Convolution operator by Marat Dukhan · 4 years, 3 months ago
  42. a6879bd Re-arrange packing functions by Marat Dukhan · 4 years, 3 months ago
  43. ab58238 Make packing functions non-inline by Marat Dukhan · 4 years, 3 months ago
  44. c67dd7f Initialize linear vs minmax binary operator microkernels for web assembly. by Frank Barchard · 4 years, 3 months ago
  45. 1d6d403 Exclude WAsm SIMD microkernels from CMake build by Marat Dukhan · 4 years, 3 months ago
  46. 0822dde vbinary and vunary microkernels assert pointers are not NULL by Frank Barchard · 4 years, 3 months ago
  47. d5b9f1c Add WASMSIMD binary ops with linear activation by Frank Barchard · 4 years, 3 months ago
  48. 609ac84 Update avgpool tests with ifdef for WASMSIMD by Frank Barchard · 4 years, 3 months ago
  49. 6804bbd Square Root operator by Marat Dukhan · 4 years, 3 months ago
  50. 676d582 Fix CMake build by Marat Dukhan · 4 years, 3 months ago
  51. f4df5fe Cortex-A7 use prefetch version of GEMM microkernel. by Frank Barchard · 4 years, 3 months ago
  52. f4db2f3 Vector SQRT microkernels by Marat Dukhan · 4 years, 3 months ago
  53. 8400076 Square root evaluation stubs by Marat Dukhan · 4 years, 3 months ago
  54. 37c8351 Port unary elementwise microkernels to WAsm SIMD by Marat Dukhan · 4 years, 3 months ago
  55. 47387d6 Fix parameter initializion and WAsm microkernel tests by Marat Dukhan · 4 years, 3 months ago
  56. 606889f Remove -munimplemented-simd128 from WAsm SIMD build by Marat Dukhan · 4 years, 3 months ago
  57. 72b399a Port RND microkernels to WAsm SIMD intrinsics by Marat Dukhan · 4 years, 3 months ago
  58. 2285547 Fix inefficiencies in WAsm SIMD MIN/MAX/MINC/MAXC microkernels by Marat Dukhan · 4 years, 3 months ago
  59. f2ebd89 Remove VRSQRDIFFC microkernels by Marat Dukhan · 4 years, 3 months ago
  60. cdc5655 Enable WebAssembly SIMD kernels for binary elementwise operators by Marat Dukhan · 4 years, 3 months ago
  61. 0e97d6f Vector Elementwise Binary with Scalar microkernels in WAsm SIMD by Marat Dukhan · 4 years, 3 months ago
  62. 6b73c4f FP16 use 6x16 aarch64 microkernel by Frank Barchard · 4 years, 3 months ago
  63. 270a2c4 FP16 End-to-End benchmarks by Marat Dukhan · 4 years, 3 months ago
  64. 49b4dcc FP16 Convolution NHWC operator by Frank Barchard · 4 years, 3 months ago
  65. fb5b20a Enable WAsm MVP tests in WAsm SIMD builds by Marat Dukhan · 4 years, 3 months ago
  66. 93d1ba1 WAsm SIMD versions of Vector Elementwise Binary microkernels by Marat Dukhan · 4 years, 3 months ago
  67. c303fe6 Optimize selection of HSWISH microkernels in WAsm SIMD by Marat Dukhan · 4 years, 3 months ago
  68. 1d89da8 Enable WAsm HSWISH microbenchmarks in WAsm SIMD build by Marat Dukhan · 4 years, 3 months ago
  69. 1c6e389 Include WAsm MVP (scalar) sources in WAsm SIMD builds by Marat Dukhan · 4 years, 3 months ago
  70. a11ca34 Microbenchmarks for HSWISH microkernels by Marat Dukhan · 4 years, 3 months ago
  71. 18a7fd0 Add assertions for MR field in Global Average Pooling by Frank Barchard · 4 years, 3 months ago
  72. ad35260 HardSwish operator benchmark by Marat Dukhan · 4 years, 3 months ago
  73. 9baec80 WAsm SIMD versions of HSWISH microkernels by Marat Dukhan · 4 years, 3 months ago
  74. 43cb20e Fix initialization of inputs & weights in end-to-end models by Marat Dukhan · 4 years, 3 months ago
  75. 0d3f467 SSE2 and SSE4.1 versions of Leaky ReLU microkernels by Marat Dukhan · 4 years, 3 months ago
  76. 5fd21b0 Reoptimize SSE PReLU microkernels by Marat Dukhan · 4 years, 3 months ago
  77. 7c1f808 WAsm implementation of PReLU microkernels by Marat Dukhan · 4 years, 3 months ago
  78. 195f8eb WAsm SIMD implementation of PReLU microkernels by Marat Dukhan · 4 years, 3 months ago
  79. 8772714 Benchmarks for rounding operators by Marat Dukhan · 4 years, 3 months ago
  80. 39b5e94 SSE versions of PReLU microkernels by Marat Dukhan · 4 years, 3 months ago
  81. d3f3d87 ROUND evaluation stubs in WAsm SIMD intrinsics by Marat Dukhan · 4 years, 3 months ago
  82. 812a7b8 xnn_create_global_average_pooling_nwc_f16 check FP16 output_min < output_max by Frank Barchard · 4 years, 3 months ago
  83. 01898c0 FP16 binary add operator by Frank Barchard · 4 years, 3 months ago
  84. 5fd403b Fix incompatibility with CMake 3.5 by Marat Dukhan · 4 years, 3 months ago
  85. b8171da Binary operator create test flags for hardware support by Frank Barchard · 4 years, 3 months ago
  86. c08701d Generalize binary operator setup to be type independent by Frank Barchard · 4 years, 3 months ago
  87. fa75714 Remove XNN_NO_Q8_OPERATORS and XNN_NO_F16_OPERATORS guards from tests by Frank Barchard · 4 years, 3 months ago
  88. d6ca9d8 4x8-minmax-aarch32-neon-pld-cortex-a75 Fix prefetch offset to not skip a cache line by Frank Barchard · 4 years, 3 months ago
  89. 0ccccf1 Group FP16 operators together in a section guarded by #ifndef XNN_NO_F16_OPERATORS by Frank Barchard · 4 years, 3 months ago
  90. 3913370 FP16 global average pool NHWC operator test round accumulated min and max by Frank Barchard · 4 years, 3 months ago
  91. 7ae00ad Remove FP16 dependency from Global Average Pooling NWC test by Frank Barchard · 4 years, 3 months ago
  92. ef61d02 Guard FP16 Global Average Pooling by initialization flags by Marat Dukhan · 4 years, 3 months ago
  93. 8d5d259 Check NEON FP16 arithmetics support by Marat Dukhan · 4 years, 3 months ago
  94. 854fb6b Replace xnn_params.initialized with fine-grained xnn_params.init_flags by Marat Dukhan · 4 years, 3 months ago
  95. af4dad4 Add ISA check for FP16 to GLOBAL_AVERAGE_POOLING_NWC_F16 tests by Frank Barchard · 4 years, 3 months ago
  96. 8b283aa Disable sparse inference by default by Marat Dukhan · 4 years, 3 months ago
  97. 9de90e0 Graph rewriting for sparse inference in NCHW layout by Marat Dukhan · 4 years, 3 months ago
  98. 3303271 Support Depthwise kernel layout in NCHW Convolution by Marat Dukhan · 4 years, 3 months ago
  99. c3d52cf Parametrise Convolution operators by strides along the channel dimension by Marat Dukhan · 4 years, 3 months ago
  100. 69a6a76 Polyfill _cvtu32_mask16 only for Clang 8 from Android NDK r19 by Marat Dukhan · 4 years, 3 months ago