Gitiles
Code Review Sign In
gerrit-public.fairphone.software / platform / external / XNNPACK / 0e8013760d7f1862fad194998d6f32339b7fe49f
  1. 0e80137 Refactor parameters in F32 VRND microkernels by Marat Dukhan · 2 years, 10 months ago
  2. d57a5ad Add const keyword for AVX load masks by Marat Dukhan · 2 years, 10 months ago
  3. bbfc27d Refactor NEON/NEONFMA VSIGMOID microkernels by Marat Dukhan · 2 years, 10 months ago
  4. 6853f23 Update amalgamated microkernels by Marat Dukhan · 2 years, 10 months ago
  5. ce834ad Refactor parameters in F32 VSIGMOID microkernels by Marat Dukhan · 2 years, 11 months ago
  6. 05b6cb1 Transpose microkernel tester uses iota instead of rng so that it's easier to debug tests by Alan Kelly · 2 years, 11 months ago
  7. 4a79ff2 Refactor parameters in F32 VELU microkernels by Marat Dukhan · 2 years, 11 months ago
  8. e5efb16 Refactor VUNARY microkernel parameters by Marat Dukhan · 2 years, 11 months ago
  9. a8b3994 Remove mask_table literal from F32 DWCONV AVX microkernels by Marat Dukhan · 2 years, 11 months ago
  10. 9084fc8 Quantized Sigmoid and ELU benchmarks by Marat Dukhan · 2 years, 11 months ago
  11. 3ddc20c Benchmarks for Abs, Negate, and Square operators by Marat Dukhan · 2 years, 11 months ago
  12. 5c7fd89 Benchmark for Leaky ReLU operator by Marat Dukhan · 2 years, 11 months ago
  13. a0129e9 Refactor benchmarks for elementwise operators by Marat Dukhan · 2 years, 11 months ago
  14. e72b282 Refactor parameters in F32 VSQRT microkernels by Marat Dukhan · 2 years, 11 months ago
  15. 98c5215 Move mask_table into VBINARY[C] AVX microkernel parameters by Marat Dukhan · 2 years, 11 months ago
  16. 0f28193 Minor optimization in F32 GEMM/IGEMM AVX512F microkernels by Marat Dukhan · 2 years, 11 months ago
  17. d57186a Refactor F32 VMULCADDC parameters by Marat Dukhan · 2 years, 11 months ago
  18. f600497 Refactor parameter initialization in Vector Binary Elementwise microkernels by Marat Dukhan · 2 years, 11 months ago
  19. b40ee63 Add comment about architecture for each set of inits for ARM. by Frank Barchard · 2 years, 11 months ago
  20. c83ef3b Refactor F32 MINMAX parameters for WAsm SIMD by Marat Dukhan · 2 years, 11 months ago
  21. 2894e99 Refactor F32 VLRELU microkernels by Marat Dukhan · 2 years, 11 months ago
  22. b7c1b71 Refactor F32->F16 VCVT microkernels by Marat Dukhan · 2 years, 11 months ago
  23. 7ddef84 Move mask_table into VCLAMP AVX microkernel parameters by Marat Dukhan · 2 years, 11 months ago
  24. e14e791 Move mask_table into VHSWISH AVX microkernel parameters by Marat Dukhan · 2 years, 11 months ago
  25. 134f984 Refactor F16->F32 VCVT microkernels by Marat Dukhan · 2 years, 11 months ago
  26. 3220551 Use --features wasm_simd rather than --copts -msimd128 for WAsm SIMD builds by Marat Dukhan · 2 years, 11 months ago
  27. 4d738ae Include vcvtnq_s32_f32 polyfill for older gcc versions by Marat Dukhan · 2 years, 11 months ago
  28. dc54e12 Replace vshll_n_u32(v, 0) with vmovl_u32 in C4/C4S2 GEMM/IGEMM microkernels by Marat Dukhan · 2 years, 11 months ago
  29. e1228e3 Fix .clang-format path by Marat Dukhan · 2 years, 11 months ago
  30. 2700809 Specify -mfp16-format=ieee for AArch32 GCC builds by Marat Dukhan · 2 years, 11 months ago
  31. 70137e4 Enable AArch32 QC8 neon dot product by Frank Barchard · 2 years, 11 months ago
  32. 87fe410 QC8 quantization for all aarch32 GEMM/IGEMM microkernels by Frank Barchard · 2 years, 11 months ago
  33. a2f1891 Add _prfm to names on Neon microkernels in a consistent way. by Frank Barchard · 2 years, 11 months ago
  34. ef0f09c Add cpu clockrate to x16/x32_transpose benchmarks. by Frank Barchard · 2 years, 11 months ago
  35. 447aa7b #include allocator.h header to gemm tests. by Frank Barchard · 2 years, 11 months ago
  36. 1945f0b SSE transpose x16 microkernel (4x8) by Alan Kelly · 2 years, 11 months ago
  37. 57719a2 clang-format for microkernels by Alan Kelly · 2 years, 11 months ago
  38. 0d10cc7 Split VHSWISH parameter initialization functions per ISA by Marat Dukhan · 2 years, 11 months ago
  39. b43b47a Add a script to convert existing assembly microkernels to JIT codegen. by Zhi An Ng · 2 years, 11 months ago
  40. 561d068 Refactor parameter initialization for VHSWISH microkernels by Marat Dukhan · 2 years, 11 months ago
  41. e4d3f76 Mark aarch64 microkernels as assembly for tests by Frank Barchard · 2 years, 11 months ago
  42. 7a03a0f Merge pull request #2191 from xbwee1024:bugfix by XNNPACK Team · 2 years, 11 months ago
  43. 1c852c9 Enable PRFM version of QS8 4x8 lane AArch32 microkernels by Frank Barchard · 2 years, 11 months ago
  44. 0db2e4c Support - (minus) operator for creating S/D register lists, this looks closer to native assembly. by Zhi An Ng · 2 years, 11 months ago
  45. 2493de9 WAsmSIMD transpose microkernel by Alan Kelly · 2 years, 11 months ago
  46. 77b694c Fixes style issues with SSE microkernel by Alan Kelly · 2 years, 11 months ago
  47. 691ec40 Use proper intrinsics header in SSE F32 VHSWISH microkernels by Marat Dukhan · 2 years, 11 months ago
  48. c025831 Refactor declarations of parameter initialization functions by Marat Dukhan · 2 years, 11 months ago
  49. 51c6134 Amalgamate SSE and AVX512 microkernels for TFLite build by Marat Dukhan · 2 years, 11 months ago
  50. e0f15ad Split scalar production microkernels into portable, AArch32, and Wasm by Marat Dukhan · 2 years, 11 months ago
  51. c80ffb0 Fix generation of gemm tests for ADJBLOCK and rerun scripts. by Zhi An Ng · 2 years, 11 months ago
  52. f98f58d Lowering to c++11 as c++14 literals was converted to c++11 in #2192 by xbwee · 2 years, 11 months ago
  53. 0fd983b Adds -Wcast-qual flag to detect cast dropping const. by Alan Kelly · 2 years, 11 months ago
  54. f527d56 Avoid using C++14 features in AArch32 assembler test by Marat Dukhan · 2 years, 11 months ago
  55. 562112e Fix build error with cmake for src/jit. by xbwee · 2 years, 11 months ago
  56. 19bfefe Support Relaxed SIMD in xnnpack_cc_library and xnnpack_aggregate_library by Marat Dukhan · 2 years, 11 months ago
  57. 9519816 Enable QS8 4x8 LD64 Neon on AArch32 by Frank Barchard · 2 years, 11 months ago
  58. e31f29e Declare assembly for QS8 microkernels by Frank Barchard · 2 years, 11 months ago
  59. 4c61779 Minimally support WebAssembly Relaxed SIMD builds by Marat Dukhan · 2 years, 11 months ago
  60. 8c7355a Enable QS8 4x8 LD64 dot product on AArch32 by Frank Barchard · 2 years, 11 months ago
  61. 1e9c5ac Fix CMake build by Marat Dukhan · 2 years, 11 months ago
  62. c3c6632 Improve compatibility with GCC in AVX512-SKX microkernels by Marat Dukhan · 2 years, 11 months ago
  63. 50b0bd9 Fix encoding and supported immediate values for vldr and vstr. by Zhi An Ng · 2 years, 11 months ago
  64. 1aac8e8 Implement vmrs (FPSCR) by Zhi An Ng · 2 years, 11 months ago
  65. 0a1b7b6 Implement ldrd (immediate) by Zhi An Ng · 2 years, 11 months ago
  66. 26e55ed Implement vstr instruction by Zhi An Ng · 2 years, 11 months ago
  67. a787832 PUSH lr instead of r14 in AArch32 assembly microkernels by Frank Barchard · 2 years, 11 months ago
  68. e0ac223 QS8 IGEMM neon dot comment change float* to int8_t* by Frank Barchard · 2 years, 11 months ago
  69. 97f99fc Return error if fail to get page size by Zhi An Ng · 2 years, 11 months ago
  70. 932e823 Implement str (imm) by Zhi An Ng · 2 years, 11 months ago
  71. 4ebd680 Implement moveq, cmp (imm), sub (imm). by Zhi An Ng · 2 years, 11 months ago
  72. 2b74ddd Implement vld1_8 with offset register by Zhi An Ng · 2 years, 11 months ago
  73. fea422d Implement vld1_32 (single element to one lane). by Zhi An Ng · 2 years, 11 months ago
  74. e48b5c1 QS8 4x8 Neon Lane LD64 IGEMM AArch32 microkernel by Frank Barchard · 2 years, 11 months ago
  75. 1669dd0 aarch32 avoid the VPUSH/VPOP of unused registers by Frank Barchard · 2 years, 11 months ago
  76. 4841021 QS8 4x8 dot product LD64 IGEMM AArch32 microkernel by Frank Barchard · 2 years, 11 months ago
  77. 938ee9b Implement bic, vld1_8 and vld1_32 for QRegisterList, assert encodings don't error out in tests. by Zhi An Ng · 2 years, 11 months ago
  78. 9364bdc Implement vsdot_s8 instruction by Zhi An Ng · 2 years, 11 months ago
  79. a251f87 Implement vqmovn_s16, and_, adds. by Zhi An Ng · 2 years, 11 months ago
  80. 7c8090d Implement vcmpe_f32, vmovpl_f32, vmovmi_f32. by Zhi An Ng · 2 years, 11 months ago
  81. 2d8180c Implement 2-argument add, vmla_f32, vmov_f32, vmov_f64, vstm. by Zhi An Ng · 2 years, 11 months ago
  82. 70e8c99 Format source and BUILD file by Frank Barchard · 2 years, 11 months ago
  83. 9f3f420 QS8 4x8 LD64 dot product GEMM AArch32 microkernel by Frank Barchard · 2 years, 11 months ago
  84. b63e84c Implement b (unconditional branch) by Zhi An Ng · 2 years, 11 months ago
  85. be4e6a5 Add align for aligning instructions (similar to .align in assembly) by Zhi An Ng · 2 years, 11 months ago
  86. ec17e99 Add license to files by Zhi An Ng · 2 years, 11 months ago
  87. 98393ad AVX512 QS8->F32 and QU8->F32 VCVT microkernels by Marat Dukhan · 2 years, 11 months ago
  88. fda06cb SSE transpose microkernel by Alan Kelly · 2 years, 11 months ago
  89. 7b5f779 AVX2 QS8->F32 and QU8->F32 VCVT microkernels by Marat Dukhan · 2 years, 11 months ago
  90. cd4089f AVX QS8->F32 and QU8->F32 VCVT microkernels by Marat Dukhan · 2 years, 11 months ago
  91. 2edf863 AVX512 F32->QS8 and F32->QU8 VCVT microkernels by Marat Dukhan · 2 years, 11 months ago
  92. 0d399ca AVX2 F32->QS8 and F32->QU8 VCVT microkernels by Marat Dukhan · 2 years, 11 months ago
  93. 3bdbe9f Fix xnn_release_code_memory to unmap entire capacity of buffer by Zhi An Ng · 2 years, 11 months ago
  94. b91432c AVX F32->QS8 and F32->QU8 VCVT microkernels by Marat Dukhan · 2 years, 11 months ago
  95. 6fac719 Implement vqmovn_s32 and vext_8 by Zhi An Ng · 2 years, 11 months ago
  96. 4a58583 Implement vdup_8, vdup_16, vdup_32 by Zhi An Ng · 2 years, 11 months ago
  97. 2649014 Implement vmax_s8, vmin_s8, vqadd_s16, vqdmulh_s32, vqshl_s32, vrshl_s32 by Zhi An Ng · 2 years, 11 months ago
  98. 4ef8d51 Implement vst1_16, add some more test cases by Zhi An Ng · 2 years, 11 months ago
  99. 00a929f Implement vst1_8 and fix vst1_32 encoding by Zhi An Ng · 2 years, 11 months ago
  100. 9820234 Full set of benchmarks for Convert operator by Marat Dukhan · 2 years, 11 months ago