Gitiles
Code Review Sign In
review.mlplatform.org / ml / ComputeLibrary / c8cc024603cb1db084227196a52e562bf251d339 / src / cpu / operators / internal / CpuGemmAssemblyDispatch.cpp
  1. c8cc024 Adding documentation section explaining how BF16 is used by Ramy Elgammal · 1 year, 9 months ago
  2. 93581a5 [ONCPUML-970] Fast math mode for fixed format kernels by Pablo Marquez Tello · 2 years ago
  3. 13b623e [ONCPUML-968] Fixed format kernel support in additional APIs by Milos Puzovic · 1 year, 11 months ago
  4. aa52b7d Fix compilation error rasied in Nightly_NEW by Ramy Elgammal · 1 year, 11 months ago
  5. 9178002 Fix for inclusion of "arm_gemm" from src into "Types.h" from core by Ramy Elgammal · 2 years ago
  6. d208f4f Enable march=armv8.6-a in non multi-isa builds by Pablo Marquez Tello · 2 years ago
  7. 553f695 [ONCPUML-951] Variable weight support for Convolution. by Francesco Petrogalli · 2 years ago
  8. 5fcf22d [arm_gemm] Import fixed-format kernels from gemm_linux. by Francesco.Petrogalli@arm.com · 2 years, 3 months ago
  9. 168d6a8 Use svcreate instead of list initializations. by Michalis Spyrou · 2 years, 2 months ago
  10. 5d606cc Fix CpuGemmAssemblyDispatch::has_opt_impl. by Francesco.Petrogalli@arm.com · 2 years, 3 months ago
  11. e33c556 [arm_gemm] Use static validate to find arm_gemm kernels. by Francesco.Petrogalli@arm.com · 2 years, 3 months ago
  12. 193cad3 Remove deprecated interface from arm_compute. by Francesco.Petrogalli@arm.com · 2 years, 4 months ago
  13. 63e0beb Add support for non-constant weights and biases in CpuFullyConnected by Giorgio Arena · 2 years, 9 months ago
  14. 9ac7b99 Revert "Add support for non-constant weights and biases in CpuFullyConnected" by Pablo Marquez Tello · 2 years, 10 months ago
  15. aed63ee Add support for non-constant weights and biases in CpuFullyConnected by Michele Di Giorgio · 3 years ago
  16. 7891a73 Move CPU/GPU files from Core/Runtime to the respective backend folders by Georgios Pinitas · 2 years, 11 months ago[Renamed (98%) from src/runtime/cpu/operators/internal/CpuGemmAssemblyDispatch.cpp]
  17. fa1db17 Ensure that correct transformed matrices are used in CpuFullyConnected by Georgios Pinitas · 2 years, 11 months ago
  18. 4ee8b15 Update GEMM assembly kernels by Georgios Pinitas · 3 years ago
  19. d7316eb Port NEGEMMConv2d to memory injecting interface by Michele Di Giorgio · 3 years, 1 month ago
  20. d02d5ed Integrate improved CPU depthwise convolution kernels by Michele Di Giorgio · 3 years, 5 months ago
  21. 8ae3cda Revert "Implement memory injection in CpuDirectGemmConv2d" by Michele Di Giorgio · 3 years, 1 month ago
  22. b3be457 Implement memory injection in CpuDirectGemmConv2d by Sang-Hoon Park · 3 years, 1 month ago
  23. d89e2fa Create CpuGemmDirectConv2d by Sang-Hoon Park · 3 years, 2 months ago
  24. 4f7693d Rename NEGEMMAssembly to CpuGemmAssembly by Sang-Hoon Park · 3 years, 2 months ago[Renamed (95%) from src/runtime/NEON/functions/NEGEMMAssemblyDispatch.cpp]
  25. 78e7691 Remove TODOs by Sheri Zhang · 3 years, 2 months ago
  26. 33f41fa Fix trademarks throughout the codebase by Michele Di Giorgio · 3 years, 4 months ago
  27. 5f91b5d Remove alignment FIXME by Michele Di Giorgio · 3 years, 3 months ago
  28. ac6499a Comply with Trademark rules for use of Neon, Arm and Mali by Sheri Zhang · 3 years, 5 months ago
  29. 8d5337e Add explicit cast to fix gcc10 build error by Sang-Hoon Park · 3 years, 6 months ago
  30. ec2256b Move NEGEMMAssemblyDispatch as an internal function by Georgios Pinitas · 3 years, 7 months ago
  31. 4c634e0 Initialize batches variable in NEGEMMAssemblyDispatch by Georgios Pinitas · 3 years, 7 months ago
  32. 40f51a6 Update default C++ standard to C++14 by Georgios Pinitas · 3 years, 7 months ago
  33. 71696ba Initialize unitialized M parameter in NEGEMMAssemblyDispatch by Georgios Pinitas · 3 years, 7 months ago
  34. c0b6f76 COMPMID-3776: Indirect GEMM by Georgios Pinitas · 3 years, 8 months ago
  35. 68dd25f COMPMID-3637: Move utility headers from arm_compute to src by Sang-Hoon Park · 3 years, 9 months ago
  36. 9e698d5 COMPMID-3757: (u)int8: Don't select the 16-bit route on A53 for cases with very few rows by David Mansell · 3 years, 10 months ago
  37. fa269bb MLCE-229: Fixed requantization per channel in asm kernel by morgolock · 3 years, 10 months ago
  38. 0bc80da MLCE-229: Support for negative shifts in asm kernels by morgolock · 3 years, 11 months ago
  39. f932d2c COMPMID-3386: Support memory injection in CLConcatenate functions/kernels by Michele Di Giorgio · 4 years ago
  40. a084b46 [ONCPUML-97]: Implement "int8" support for 2D decomposition at high core counts by Aleksandr Nikolaev · 4 years ago
  41. 534fdea [ONCPUML-96] FP16 support for 2D decomposition at high core counts. by cfRod · 4 years ago
  42. d9eaf61 COMPMID-3324: Adjusting capitalization of Arm copyright claim to reflect Arm preferred presentation by Michele Di Giorgio · 4 years ago
  43. 0cc50ed COMPMID-3324: Remove pretransposed support from NEON backend by Georgios Pinitas · 4 years ago
  44. 5aa1a0b COMPID-3324: Clean GEMM kernels by Georgios Pinitas · 4 years ago
  45. 0f954eb COMPMID-3324: Remove pretranspose_B support from NEGEMM by Georgios Pinitas · 4 years ago
  46. 6ad60af COMPMID-3520: Move ndrange.hpp header from arm_gemm to assembly by Michele Di Giorgio · 4 years, 1 month ago
  47. 6f8b17d [ONCPUML-7] arm_compute support for ND parallelism by Joseph Dobson · 4 years, 5 months ago
  48. c7b183a COMPMID-3160: Add Bfloat16 support in NEGEMMConvolutionLayer by Georgios Pinitas · 4 years, 4 months ago
  49. 47fd61f COMPMID-2819: Remove use of std::bind1st as is deprecated since C++11 by Georgios Pinitas · 4 years, 5 months ago
  50. 71ac903 COMPMID-2923 Integrate arm_gemm per channel quantization by Michalis Spyrou · 4 years, 8 months ago
  51. 5255672 COMPMID-2998: NEON/GEMMConvolutionLayer/Quantized/QASYMM8_SIGNED failing on 32-bit by Michele Di Giorgio · 4 years, 6 months ago
  52. 5cb49dc COMPMID-2962 Fix compiler warnings for g++8 by Michalis Spyrou · 4 years, 7 months ago
  53. 6011f24 COMPMID-2853: Fix VGG16 for fp16. by Georgios Pinitas · 4 years, 8 months ago
  54. 77d4252 COMPMID-2853: VGG16 regression for fp32 by Georgios Pinitas · 4 years, 8 months ago
  55. dbdea0d COMPMID-2308: NEConvolutionLayer: support QUANT8_SYMM_PER_CHANNEL filters by Georgios Pinitas · 4 years, 9 months ago
  56. 48b3ef8 COMPMID-2577: Fuse bias addition and activation in gemm assembly kernels by Georgios Pinitas · 4 years, 9 months ago
  57. 1a569a3 COMPMID-2161 [NEON] Create IWeightManager class by Michalis Spyrou · 4 years, 10 months ago
  58. cfa2bba COMPMID-2178: Update GEMM assembly code. by Georgios Pinitas · 5 years ago
  59. 37d080f COMPMID-2378: Sanitize GEMM configuration for NEON by Georgios Pinitas · 5 years ago
  60. bcfd09a COMPMID-2177 Fix clang warnings by Michalis Spyrou · 5 years ago
  61. da953f2 COMPMID-2098: Scope handling of memory group resources. by Georgios Pinitas · 5 years ago
  62. 3dbfd23 COMPMID-1710: Introduce GEMM strategy name in GEMMAssemblyWrapper. by Georgios Pinitas · 5 years ago
  63. 7cd26d4 COMPMID-1867: Add NEON/SVE GEMM Hybrid kernels. by Georgios Pinitas · 5 years ago
  64. ca1250d COMPMID-1451: Fix MemoryManager usage. by Georgios Pinitas · 6 years ago
  65. 1e92896 COMPMID-1451: Dispatch fp16 gemms through the function interface. by Georgios Pinitas · 6 years ago
  66. 20394d5 COMPMID-1248 Enabled memory manager in NEWinogradConvolutionLayer by Anthony Barbier · 6 years ago
  67. 3d677cc COMPMID-1406: Refactor gemm_interleaved to use our own types and scheduler by Anthony Barbier · 6 years ago
  68. 9036749 COMPMID-1445: Fixed NEON GEMM assembly dispatch for QASYMM8 by Anthony Barbier · 6 years ago
  69. 40ed6d8 COMPMID-1440: Access original B in gemm assembly when not pretransposed. by Georgios Pinitas · 6 years ago
  70. eb84d6b COMPMID-1434: Fix NEWinograd for NHWC and sub-tensors by Georgios Pinitas · 6 years ago
  71. eaefd00 COMPMID-1419: Make NEGEMMAssemblyDispatch dynamically typed instead of templated by Anthony Barbier · 6 years ago
  72. c8e84b5 COMPMID-1405: Create our own gemm_native kernel / function. by Anthony Barbier · 6 years ago
  73. 67fd4e8 COMPMID-1381: Fix nightly build failure on armv7 by Anthony Barbier · 6 years ago
  74. 71d9b57 COMPMID-1381: Cleaned up the AssemblyHelper interface by Anthony Barbier · 6 years ago