From 8f43d745b170aefca269a087fc045d8af3813c33 Mon Sep 17 00:00:00 2001 From: Pablo Tello Date: Wed, 27 Mar 2019 09:28:32 +0000 Subject: COMPMID-2063: New Winograd implementation Refactoring of winograd code reducing the size of the binaries about 8X. Change-Id: If8845bda324573e1a5cf436f354ac8603e88a92e Signed-off-by: Pablo Tello Reviewed-on: https://review.mlplatform.org/c/959 Comments-Addressed: Arm Jenkins Tested-by: Anthony Barbier Reviewed-by: Georgios Pinitas --- SConscript | 2 +- .../kernels/NEWinogradConvolutionLayerKernel.h | 183 +- .../NEON/kernels/convolution/common/padding.hpp | 17 + .../convolution/winograd/batched_blocked_gemm.hpp | 69 - .../NEON/kernels/convolution/winograd/gemm.hpp | 127 -- .../convolution/winograd/gemm/a64_sgemm.hpp | 355 ---- .../convolution/winograd/gemm/a64_sgemm_4x16.hpp | 1446 --------------- .../convolution/winograd/transforms/input.hpp | 349 ---- .../convolution/winograd/transforms/kernel.hpp | 77 - .../convolution/winograd/transforms/output.hpp | 278 --- .../NEON/kernels/convolution/winograd/winograd.hpp | 610 +++++++ .../kernels/convolution/winograd/winograd_gemm.hpp | 226 --- .../winograd/winograd_input_transform.hpp | 271 --- .../convolution/winograd/winograd_layer.hpp | 211 +++ .../winograd/winograd_output_transform.hpp | 232 --- .../NEON/functions/NEWinogradConvolutionLayer.h | 4 +- .../kernels/NEWinogradConvolutionLayerKernel.cpp | 142 +- .../convolution/winograd/batched_blocked_gemm.cpp | 82 - .../NEON/kernels/convolution/winograd/padding.cpp | 166 ++ .../winograd/transforms/input_1x8_fp32.cpp | 261 --- .../winograd/transforms/input_2x2_3x3_fp32.cpp | 311 ---- .../winograd/transforms/input_6x6_fp32.cpp | 376 ---- .../winograd/transforms/output_2_7_fp32.cpp | 163 -- .../winograd/transforms/output_2x2_3x3_fp32.cpp | 375 ---- .../winograd/transforms/output_2x2_5x5_fp32.cpp | 369 ---- .../winograd/transforms/output_4_5_fp32.cpp | 171 -- .../winograd/transforms/output_4x4_3x3_fp32.cpp | 428 ----- .../winograd/transforms/output_6_3_fp32.cpp | 179 -- .../winograd/transforms/weights_2_7_fp32.cpp | 124 -- .../winograd/transforms/weights_2x2_3x3_fp32.cpp | 228 --- .../winograd/transforms/weights_2x2_5x5_fp32.cpp | 408 ----- .../winograd/transforms/weights_4_5_fp32.cpp | 124 -- .../winograd/transforms/weights_4x4_3x3_fp32.cpp | 266 --- .../winograd/transforms/weights_6_3_fp32.cpp | 125 -- .../NEON/kernels/convolution/winograd/winograd.cpp | 236 +++ .../kernels/convolution/winograd/winograd_gemm.cpp | 240 --- .../winograd/winograd_transforms/input.hpp | 265 +++ .../input_1x8_fp32_fp32_integers.cpp | 158 ++ .../input_4x4_fp32_fp32_integers.cpp | 255 +++ .../input_6x6_fp32_fp32_integers.cpp | 1308 ++++++++++++++ .../winograd/winograd_transforms/kernel.hpp | 78 + .../winograd/winograd_transforms/output.hpp | 249 +++ .../output_2_7_fp32_fp32_integers.cpp | 137 ++ .../output_2x2_3x3_fp32_fp32_integers.cpp | 222 +++ .../output_2x2_5x5_fp32_fp32_integers.cpp | 216 +++ .../output_4_5_fp32_fp32_integers.cpp | 143 ++ .../output_4x4_3x3_fp32_fp32_integers.cpp | 1855 ++++++++++++++++++++ .../output_6_3_fp32_fp32_integers.cpp | 149 ++ .../weights_2_7_fp32_fp32_integers.cpp | 90 + .../weights_2x2_3x3_fp32_fp32_integers.cpp | 220 +++ .../weights_2x2_5x5_fp32_fp32_integers.cpp | 401 +++++ .../weights_4_5_fp32_fp32_integers.cpp | 90 + .../weights_4x4_3x3_fp32_fp32_integers.cpp | 257 +++ .../weights_6_3_fp32_fp32_integers.cpp | 90 + .../NEON/functions/NEWinogradConvolutionLayer.cpp | 41 +- 55 files changed, 7668 insertions(+), 7787 deletions(-) delete mode 100644 arm_compute/core/NEON/kernels/convolution/winograd/batched_blocked_gemm.hpp delete mode 100644 arm_compute/core/NEON/kernels/convolution/winograd/gemm.hpp delete mode 100644 arm_compute/core/NEON/kernels/convolution/winograd/gemm/a64_sgemm.hpp delete mode 100644 arm_compute/core/NEON/kernels/convolution/winograd/gemm/a64_sgemm_4x16.hpp delete mode 100644 arm_compute/core/NEON/kernels/convolution/winograd/transforms/input.hpp delete mode 100644 arm_compute/core/NEON/kernels/convolution/winograd/transforms/kernel.hpp delete mode 100644 arm_compute/core/NEON/kernels/convolution/winograd/transforms/output.hpp create mode 100644 arm_compute/core/NEON/kernels/convolution/winograd/winograd.hpp delete mode 100644 arm_compute/core/NEON/kernels/convolution/winograd/winograd_gemm.hpp delete mode 100644 arm_compute/core/NEON/kernels/convolution/winograd/winograd_input_transform.hpp create mode 100644 arm_compute/core/NEON/kernels/convolution/winograd/winograd_layer.hpp delete mode 100644 arm_compute/core/NEON/kernels/convolution/winograd/winograd_output_transform.hpp delete mode 100644 src/core/NEON/kernels/convolution/winograd/batched_blocked_gemm.cpp create mode 100644 src/core/NEON/kernels/convolution/winograd/padding.cpp delete mode 100644 src/core/NEON/kernels/convolution/winograd/transforms/input_1x8_fp32.cpp delete mode 100644 src/core/NEON/kernels/convolution/winograd/transforms/input_2x2_3x3_fp32.cpp delete mode 100644 src/core/NEON/kernels/convolution/winograd/transforms/input_6x6_fp32.cpp delete mode 100644 src/core/NEON/kernels/convolution/winograd/transforms/output_2_7_fp32.cpp delete mode 100644 src/core/NEON/kernels/convolution/winograd/transforms/output_2x2_3x3_fp32.cpp delete mode 100644 src/core/NEON/kernels/convolution/winograd/transforms/output_2x2_5x5_fp32.cpp delete mode 100644 src/core/NEON/kernels/convolution/winograd/transforms/output_4_5_fp32.cpp delete mode 100644 src/core/NEON/kernels/convolution/winograd/transforms/output_4x4_3x3_fp32.cpp delete mode 100644 src/core/NEON/kernels/convolution/winograd/transforms/output_6_3_fp32.cpp delete mode 100644 src/core/NEON/kernels/convolution/winograd/transforms/weights_2_7_fp32.cpp delete mode 100644 src/core/NEON/kernels/convolution/winograd/transforms/weights_2x2_3x3_fp32.cpp delete mode 100644 src/core/NEON/kernels/convolution/winograd/transforms/weights_2x2_5x5_fp32.cpp delete mode 100644 src/core/NEON/kernels/convolution/winograd/transforms/weights_4_5_fp32.cpp delete mode 100644 src/core/NEON/kernels/convolution/winograd/transforms/weights_4x4_3x3_fp32.cpp delete mode 100644 src/core/NEON/kernels/convolution/winograd/transforms/weights_6_3_fp32.cpp create mode 100644 src/core/NEON/kernels/convolution/winograd/winograd.cpp delete mode 100644 src/core/NEON/kernels/convolution/winograd/winograd_gemm.cpp create mode 100644 src/core/NEON/kernels/convolution/winograd/winograd_transforms/input.hpp create mode 100644 src/core/NEON/kernels/convolution/winograd/winograd_transforms/input_1x8_fp32_fp32_integers.cpp create mode 100644 src/core/NEON/kernels/convolution/winograd/winograd_transforms/input_4x4_fp32_fp32_integers.cpp create mode 100644 src/core/NEON/kernels/convolution/winograd/winograd_transforms/input_6x6_fp32_fp32_integers.cpp create mode 100644 src/core/NEON/kernels/convolution/winograd/winograd_transforms/kernel.hpp create mode 100644 src/core/NEON/kernels/convolution/winograd/winograd_transforms/output.hpp create mode 100644 src/core/NEON/kernels/convolution/winograd/winograd_transforms/output_2_7_fp32_fp32_integers.cpp create mode 100644 src/core/NEON/kernels/convolution/winograd/winograd_transforms/output_2x2_3x3_fp32_fp32_integers.cpp create mode 100644 src/core/NEON/kernels/convolution/winograd/winograd_transforms/output_2x2_5x5_fp32_fp32_integers.cpp create mode 100644 src/core/NEON/kernels/convolution/winograd/winograd_transforms/output_4_5_fp32_fp32_integers.cpp create mode 100644 src/core/NEON/kernels/convolution/winograd/winograd_transforms/output_4x4_3x3_fp32_fp32_integers.cpp create mode 100644 src/core/NEON/kernels/convolution/winograd/winograd_transforms/output_6_3_fp32_fp32_integers.cpp create mode 100644 src/core/NEON/kernels/convolution/winograd/winograd_transforms/weights_2_7_fp32_fp32_integers.cpp create mode 100644 src/core/NEON/kernels/convolution/winograd/winograd_transforms/weights_2x2_3x3_fp32_fp32_integers.cpp create mode 100644 src/core/NEON/kernels/convolution/winograd/winograd_transforms/weights_2x2_5x5_fp32_fp32_integers.cpp create mode 100644 src/core/NEON/kernels/convolution/winograd/winograd_transforms/weights_4_5_fp32_fp32_integers.cpp create mode 100644 src/core/NEON/kernels/convolution/winograd/winograd_transforms/weights_4x4_3x3_fp32_fp32_integers.cpp create mode 100644 src/core/NEON/kernels/convolution/winograd/winograd_transforms/weights_6_3_fp32_fp32_integers.cpp diff --git a/SConscript b/SConscript index 03976ea830..62b6073bf3 100644 --- a/SConscript +++ b/SConscript @@ -207,7 +207,7 @@ if env['neon']: # build winograd sources for either v7a / v8a core_files += Glob('src/core/NEON/kernels/convolution/*/*.cpp') core_files += Glob('src/core/NEON/kernels/convolution/winograd/*/*.cpp') - arm_compute_env.Append(CPPPATH = ["arm_compute/core/NEON/kernels/winograd/", "arm_compute/core/NEON/kernels/assembly/"]) + arm_compute_env.Append(CPPPATH = ["arm_compute/core/NEON/kernels/convolution/winograd/","arm_compute/core/NEON/kernels/convolution/common/" , "arm_compute/core/NEON/kernels/assembly/"]) graph_files += Glob('src/graph/backends/NEON/*.cpp') diff --git a/arm_compute/core/NEON/kernels/NEWinogradConvolutionLayerKernel.h b/arm_compute/core/NEON/kernels/NEWinogradConvolutionLayerKernel.h index 96580053dd..f6b189cb1c 100644 --- a/arm_compute/core/NEON/kernels/NEWinogradConvolutionLayerKernel.h +++ b/arm_compute/core/NEON/kernels/NEWinogradConvolutionLayerKernel.h @@ -1,5 +1,5 @@ /* - * Copyright (c) 2017-2018 ARM Limited. + * Copyright (c) 2017-2019 ARM Limited. * * SPDX-License-Identifier: MIT * @@ -27,8 +27,7 @@ #include "arm_compute/core/NEON/INEKernel.h" #include "arm_compute/core/NEON/kernels/convolution/common/convolution.hpp" #include "arm_compute/core/NEON/kernels/convolution/common/tensor.hpp" -#include "arm_compute/core/NEON/kernels/convolution/winograd/batched_blocked_gemm.hpp" -#include "arm_compute/core/NEON/kernels/convolution/winograd/winograd_gemm.hpp" +#include "arm_compute/core/NEON/kernels/convolution/winograd/winograd_layer.hpp" namespace arm_compute { @@ -39,6 +38,17 @@ template class INEWinogradLayerTransformInputKernel : public INEKernel { public: + /** Get the working space required to perform the transformation. + * + * Note, the working space is only required when performing the + * transformation - hence it can be reused whenever the transformation is + * not running. + * + * @param num_threads The greatest number of threads that will be used to execute the transform. + * @return Size of working space required in bytes. + */ + virtual unsigned int get_working_space_size(unsigned int num_threads) const = 0; + /** Determine how much memory (in units of TIn) to allocate for the * transformed input. * @@ -72,9 +82,10 @@ public: * @param[in] padding Padding type. * @param[out] output Base of output matrices. * @param[in] matrix_stride Stride between output matrices. + * @param[in] workspace Tensor to be used as the working space during the computation. */ virtual void configure(const ITensor *input_nhwc, const int num_batches, const int num_rows, const int num_cols, const int num_channels, - const PaddingType padding, ITensor *output, const int matrix_stride) = 0; + const PaddingType padding, ITensor *output, const int matrix_stride, ITensor *workspace) = 0; /** Destructor */ virtual ~INEWinogradLayerTransformInputKernel() @@ -116,6 +127,18 @@ public: int num_cols, bool same_padding) const override; + /** Get the working space required to perform the transformation. + * + * Note, the working space is only required when performing the + * transformation - hence it can be reused whenever the transformation is + * not running. + * + * @param[in] num_threads The greatest number of threads that will be used to execute the transform. + * + * @return Size of working space required in bytes. + */ + unsigned int get_working_space_size(unsigned int num_threads) const override; + /** Gets the stride between matrices in the input worspace * * @param[in] kernel_shape The shape of the weights tensor. @@ -144,6 +167,7 @@ public: * @param[in] padding Padding type. * @param[out] output Base of output matrices. * @param[in] matrix_stride Stride between output matrices. + * @param[in] workspace Tensor to be used as the working space during the computation. */ void configure( const ITensor *input_nhwc, @@ -153,13 +177,14 @@ public: const int num_channels, const PaddingType padding, ITensor *output, - const int matrix_stride) override; + const int matrix_stride, + ITensor *workspace) override; // Inherited methods overridden: void run(const Window &window, const ThreadInfo &info) override; /** Winograd base kernel */ - using WinogradBase = winograd::WinogradGEMM; + using WinogradBase = winograd::WinogradGEMM; /** Winograd convolution kernel */ using WinogradConv = typename WinogradBase::template Convolution; @@ -174,15 +199,22 @@ public: static Status validate(const ITensorInfo *input, const ITensorInfo *output, const WinogradInfo &winograd_info); private: - using InputTransform = typename WinogradBase::template InputTransform; - const ITensor *_input_nhwc; - int _num_batches; /**< Number of batches in input tensor. */ - int _num_rows; /**< Number of rows in input tensor. */ - int _num_cols; /**< Number of columns in input tensor. */ - int _num_channels; /**< Number of channels in input tensor. */ - PaddingType _padding; /**< Padding type. */ - ITensor *_output; /**< Base of output matrices. */ - int _matrix_stride; /**< Stride between output matrices. */ + using InputTransform = typename WinogradBase::template InputTransform; + + std::unique_ptr _transform{ nullptr }; + const ITensor *_input_nhwc; + int _num_batches; /**< Number of batches in input tensor. */ + int _num_rows; /**< Number of rows in input tensor. */ + int _num_cols; /**< Number of columns in input tensor. */ + int _num_channels; /**< Number of channels in input tensor. */ + PaddingType _padding; /**< Padding type. */ + ITensor *_output; /**< Base of output matrices. */ + int _matrix_stride; /**< Stride between output matrices. */ + int _padding_top; /**< Padding to apply to the top of the image. */ + int _padding_left; /**< Padding to apply to the left of the image. */ + int _padding_right; /**< Padding to apply to the right of the image. */ + int _padding_bottom; /**< Padding to apply to the bottom of the image. */ + ITensor *_workspace; }; /** Interface for the NEON kernel to perform Winograd output transform. */ @@ -190,6 +222,18 @@ template class INEWinogradLayerTransformOutputKernel : public INEKernel { public: + /** Get the working space required to perform the transformation. + * + * Note, the working space is only required when performing the + * transformation - hence it can be reused whenever the transformation is + * not running. + * + * @param[in] num_threads The greatest number of threads that will be used to execute the transform. + * + * @return Size of working space required in bytes. + */ + virtual unsigned int get_working_space_size(unsigned int num_threads) const = 0; + /** Determine how much memory (in units of TOut) to allocate for the * (Winograd domain) output. * @@ -225,24 +269,26 @@ public: /** Configure the output transform kernel. * - * @param[in] biases Pointer to the biases tensor. - * @param[in] output_workingspace Pointer to working space for the output tensor in the Winograd domain. - * @param[in] matrix_stride Output matrix stride, can be computed with winograd::WinogradGEMM<2, 2, 3, 3>::Convolution::get_output_matrix_stride() - * @param[out] output_nhwc Pointer to a tensor in NHWC data layout ordered output tensor, in the spatial domain. - * @param[in] num_batches Number of batches in the input tensor. - * @param[in] num_rows Number of rows in output tensor. - * @param[in] num_cols Number of columns in output tensor. - * @param[in] num_channels Number of feature maps in the output tensor. + * @param[in] biases Pointer to the biases tensor. + * @param[in] transformed_output Pointer to working space for the output tensor in the Winograd domain. + * @param[in] matrix_stride Output matrix stride, can be computed with winograd::WinogradGEMM<2, 2, 3, 3>::Convolution::get_output_matrix_stride() + * @param[out] output_nhwc Pointer to a tensor in NHWC data layout ordered output tensor, in the spatial domain. + * @param[in] num_batches Number of batches in the input tensor. + * @param[in] num_rows Number of rows in output tensor. + * @param[in] num_cols Number of columns in output tensor. + * @param[in] num_channels Number of feature maps in the output tensor. + * @param[in] workspace Tensor to be used as the working space during the computation. */ virtual void configure( const ITensor *biases, - const ITensor *output_workingspace, + const ITensor *transformed_output, const int matrix_stride, ITensor *output_nhwc, const int num_batches, const int num_rows, const int num_cols, - const int num_channels) = 0; + const int num_channels, + ITensor *workspace) = 0; virtual ~INEWinogradLayerTransformOutputKernel() { @@ -305,54 +351,70 @@ public: */ Tensor4DShape get_output_shape(const KernelShape &kernel_shape, const Tensor4DShape &in_shape, const PaddingType padding) const override; + /** Get the working space required to perform the transformation. + * + * Note, the working space is only required when performing the + * transformation - hence it can be reused whenever the transformation is + * not running. + * + * @param[in] num_threads The greatest number of threads that will be used to execute the transform. + * + * @return Size of working space required in bytes. + */ + unsigned int get_working_space_size(unsigned int num_threads) const override; + /** Configure the output transform kernel. * - * @param[in] biases Pointer to the biases tensor. - * @param[in] output_workingspace Pointer to working space for the output tensor in the Winograd domain. - * @param[in] matrix_stride Output matrix stride, can be computed with winograd::WinogradGEMM<2, 2, 3, 3>::Convolution::get_output_matrix_stride() - * @param[out] output_nhwc Pointer to a tensor with NHWC data layout, in the spatial domain. - * @param[in] num_batches Number of batches in the input tensor. - * @param[in] num_rows Number of rows in output tensor. - * @param[in] num_cols Number of columns in output tensor. - * @param[in] num_channels Number of feature maps in the output tensor. + * @param[in] biases Pointer to the biases tensor. + * @param[in] transformed_output Pointer to working space for the output tensor in the Winograd domain. + * @param[in] matrix_stride Output matrix stride, can be computed with winograd::WinogradGEMM<2, 2, 3, 3>::Convolution::get_output_matrix_stride() + * @param[out] output_nhwc Pointer to a tensor with NHWC data layout, in the spatial domain. + * @param[in] num_batches Number of batches in the input tensor. + * @param[in] num_rows Number of rows in output tensor. + * @param[in] num_cols Number of columns in output tensor. + * @param[in] num_channels Number of feature maps in the output tensor. + * @param[in] workspace Tensor to be used as the working space during the computation. */ void configure( const ITensor *biases, - const ITensor *output_workingspace, + const ITensor *transformed_output, const int matrix_stride, ITensor *output_nhwc, const int num_batches, const int num_rows, const int num_cols, - const int num_channels) override; + const int num_channels, + ITensor *workspace) override; void run(const Window &window, const ThreadInfo &info) override; /** Static function to check if given info will lead to a valid configuration of @ref NEWinogradLayerTransformOutputKernel * - * @param[in] input Source tensor with shape [C, N, 16, batches] or [C, N, 36, batches]. Data types supported: F32. - * @param[in] bias Biases tensor. Shared biases supported. Biases are 1D tensor with dimensions [OFM]. It can be a nullptr. Data type supported: as @p input - * @param[out] output Destination tensor with shape [output_convolved_dims.width, output_convolved_dims.height, C, batches]. Data type supported: same as @p input - * @param[in] winograd_info Contains Winograd's information described in @ref WinogradInfo + * @param[in] input Source tensor info with shape [C, N, 16, batches] or [C, N, 36, batches]. Data types supported: F32. + * @param[in] bias Biases tensor info. Shared biases supported. Biases are 1D tensor with dimensions [OFM]. It can be a nullptr. Data type supported: as @p input + * @param[in] output Destination tensor info with shape [output_convolved_dims.width, output_convolved_dims.height, C, batches]. Data type supported: same as @p input + * @param[in] winograd_info Contains Winograd's information described in @ref WinogradInfo * * @return a status */ static Status validate(const ITensorInfo *input, const ITensorInfo *bias, const ITensorInfo *output, const WinogradInfo &winograd_info); private: - using WinogradBase = winograd::WinogradGEMM; + using WinogradBase = winograd::WinogradGEMM; using WinogradConv = typename WinogradBase::template Convolution; - using OutputTransform = typename WinogradBase::template OutputTransform; - - const ITensor *_biases; - const ITensor *_output_workspace; - int _matrix_stride; - int _matrix_row_stride; - ITensor *_output_nhwc; - int _num_batches; - int _num_rows; - int _num_cols; - int _num_channels; + using OutputTransform = typename WinogradBase::template OutputTransform; + + std::unique_ptr _transform{ nullptr }; + const ITensor *_biases; + const ITensor *_transformed_output; + ITensor *_workspace; + int _matrix_stride; + int _matrix_row_stride; + ITensor *_output_nhwc; + int _num_batches; + int _num_rows; + int _num_cols; + int _num_channels; }; /** Interface for the NEON kernel to perform Winograd weights transform. */ @@ -482,15 +544,16 @@ public: bool is_parallelisable() const override; private: - using WinogradBase = winograd::WinogradGEMM; + using WinogradBase = winograd::WinogradGEMM; using WinogradConv = typename WinogradBase::template Convolution; - using WeightsTransform = typename WinogradBase::template WeightsTransform; - - const ITensor *_weights_hwio; - ITensor *_output; - int _matrix_stride; - int _num_output_channels; - int _num_input_channels; + using WeightsTransform = typename WinogradBase::template WeightsTransform; + + std::unique_ptr _transform{ nullptr }; + const ITensor *_weights_hwio; + ITensor *_output; + int _matrix_stride; + int _num_output_channels; + int _num_input_channels; }; /** NEON kernel to perform Winograd. */ @@ -499,7 +562,7 @@ class NEWinogradLayerConfiguration { public: /** Winograd base kernel */ - using WinogradBase = winograd::WinogradGEMM; + using WinogradBase = winograd::WinogradGEMM; /** Winograd convolution kernel */ using WinogradConv = typename WinogradBase::template Convolution; diff --git a/arm_compute/core/NEON/kernels/convolution/common/padding.hpp b/arm_compute/core/NEON/kernels/convolution/common/padding.hpp index 33f77d7ee9..97b21e0ff5 100644 --- a/arm_compute/core/NEON/kernels/convolution/common/padding.hpp +++ b/arm_compute/core/NEON/kernels/convolution/common/padding.hpp @@ -71,4 +71,21 @@ class CopyCropped ); }; +template +void crop_and_copy_tile( + unsigned int tile_rows, + unsigned int tile_cols, + unsigned int n_channels, + const T *inptr, + unsigned int in_row_stride, + unsigned int in_col_stride, + T *outptr, + unsigned int out_row_stride, + unsigned int out_col_stride, + unsigned int crop_top, + unsigned int crop_left, + unsigned int crop_bottom, + unsigned int crop_right +); + } diff --git a/arm_compute/core/NEON/kernels/convolution/winograd/batched_blocked_gemm.hpp b/arm_compute/core/NEON/kernels/convolution/winograd/batched_blocked_gemm.hpp deleted file mode 100644 index 663b3c414f..0000000000 --- a/arm_compute/core/NEON/kernels/convolution/winograd/batched_blocked_gemm.hpp +++ /dev/null @@ -1,69 +0,0 @@ -/* - * Copyright (c) 2017 ARM Limited. - * - * SPDX-License-Identifier: MIT - * - * Permission is hereby granted, free of charge, to any person obtaining a copy - * of this software and associated documentation files (the "Software"), to - * deal in the Software without restriction, including without limitation the - * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or - * sell copies of the Software, and to permit persons to whom the Software is - * furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in all - * copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE - * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER - * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, - * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. - */ - -#pragma once - -namespace winograd -{ - -template -class BatchedBlockedGemm -{ - public: - /** Create a new batched blocked GEMM operator. */ - BatchedBlockedGemm( - const unsigned int n_gemms, - const int M, const int K, const int N, - const int a_matrix_stride, - const int a_row_stride, - const int b_matrix_stride, - const int b_row_stride, - const int c_matrix_stride, - const int c_row_stride, - const TIn* const a_ptr, - const TIn* const b_ptr, - TOut* const c_ptr - ); - - BatchedBlockedGemm(const BatchedBlockedGemm&) = delete; - BatchedBlockedGemm operator=(const BatchedBlockedGemm&) = delete; - - /** Get a window of work performed by the operator. */ - unsigned int get_window() const; - - /** Perform a portion of the work of the operator. */ - void run(const unsigned int start, const unsigned int stop); - - private: - const unsigned int n_gemms; - const int M, N, K; - const int a_matrix_stride, a_row_stride; - const int b_matrix_stride, b_row_stride; - const int c_matrix_stride, c_row_stride; - const TIn* const a_ptr; - const TIn* const b_ptr; - TOut* const c_ptr; -}; - -} // namespace winograd diff --git a/arm_compute/core/NEON/kernels/convolution/winograd/gemm.hpp b/arm_compute/core/NEON/kernels/convolution/winograd/gemm.hpp deleted file mode 100644 index 6e06db324c..0000000000 --- a/arm_compute/core/NEON/kernels/convolution/winograd/gemm.hpp +++ /dev/null @@ -1,127 +0,0 @@ -/* - * Copyright (c) 2017 ARM Limited. - * - * SPDX-License-Identifier: MIT - * - * Permission is hereby granted, free of charge, to any person obtaining a copy - * of this software and associated documentation files (the "Software"), to - * deal in the Software without restriction, including without limitation the - * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or - * sell copies of the Software, and to permit persons to whom the Software is - * furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in all - * copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE - * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER - * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, - * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. - */ - -#pragma once -#include "arm_compute/core/NEON/kernels/convolution/common/utils.hpp" - -template -inline void Gemm(const TIn* const a, const TIn* const b, TOut *c, - const int M, const int K, const int N, - const int a_row_stride, - const int b_row_stride, - const int c_row_stride, - const bool a_transposed=false, - const bool b_transposed=false) { - // Array access methods - const auto A = [a, a_transposed, M, K, a_row_stride] (const int i, const int j) -> TIn { - return a[(!a_transposed) ? i*a_row_stride + j : i + j*M]; - }; - - const auto B = [b, b_transposed, K, N, b_row_stride] (const int i, const int j) -> TIn { - return b[(!b_transposed) ? i*b_row_stride + j : i + j*N]; - }; - - const auto C = [c, c_row_stride] (const int i, const int j) -> TOut& { - return c[i*c_row_stride + j]; - }; - - // Perform the matrix multiplication - for (int i = 0; i < M; i++) { - for (int j = 0; j < N; j++) { - for (int k = 0; k < K; k++) { - C(i, j) += A(i, k) * B(k, j); - } - } - } -} - -template -inline void BlockedGemm( - const TIn* const a, const TIn* const b, TOut *c, - const int M, const int K, const int N, - const int a_row_stride, - const int b_row_stride, - const int c_row_stride -) { - // Array access methods - const auto A = [a, a_row_stride] (const int i, const int j) -> TIn { - return a[i*a_row_stride + j]; - }; - - const auto B = [b, b_row_stride] (const int i, const int j) -> TIn { - return b[i*b_row_stride + j]; - }; - - const auto C = [c, c_row_stride] (const int i, const int j) -> TOut& { - return c[i*c_row_stride + j]; - }; - - const int M_BLOCKS = iceildiv(M, M_BLOCK); - const int N_BLOCKS = iceildiv(N, N_BLOCK); - - // For each block of output rows - for (int mblock = 0; mblock < M_BLOCKS; mblock++) { - // For each block of output columns - for (int nblock = 0; nblock < N_BLOCKS; nblock++) { - // Create an appropriately sized block of accumulators - TOut accum[M_BLOCK][N_BLOCK]; - for (int i = 0; i < M_BLOCK; i++) { - for (int j = 0; j < N_BLOCK; j++) { - accum[i][j] = static_cast(0); - } - } - - // Perform this portion of the matrix multiply - for (int k = 0; k < K; k++) { - // Load elements of A - TIn elems_a[M_BLOCK]; - for (int i = 0; i < M_BLOCK; i++) { - elems_a[i] = A(mblock*M_BLOCK + i, k); - } - - // Load elements of B - TIn elems_b[N_BLOCK]; - for (int j = 0; j < N_BLOCK; j++) { - elems_b[j] = B(k, nblock*N_BLOCK + j); - } - - // Perform the partial matrix multiply - for (int i = 0; i < M_BLOCK; i++) { - for (int j = 0; j < N_BLOCK; j++) { - accum[i][j] += elems_a[i] * elems_b[j]; - } - } - } - - // Store the partial product - for (int i = 0; i < M_BLOCK; i++) { - for (int j = 0; j < N_BLOCK; j++) { - C(mblock*M_BLOCK + i, nblock*N_BLOCK + j) = accum[i][j]; - } - } - } - } -} - -#include "gemm/a64_sgemm.hpp" diff --git a/arm_compute/core/NEON/kernels/convolution/winograd/gemm/a64_sgemm.hpp b/arm_compute/core/NEON/kernels/convolution/winograd/gemm/a64_sgemm.hpp deleted file mode 100644 index 8073cb1896..0000000000 --- a/arm_compute/core/NEON/kernels/convolution/winograd/gemm/a64_sgemm.hpp +++ /dev/null @@ -1,355 +0,0 @@ -/* - * Copyright (c) 2017 ARM Limited. - * - * SPDX-License-Identifier: MIT - * - * Permission is hereby granted, free of charge, to any person obtaining a copy - * of this software and associated documentation files (the "Software"), to - * deal in the Software without restriction, including without limitation the - * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or - * sell copies of the Software, and to permit persons to whom the Software is - * furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in all - * copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE - * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER - * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, - * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. - */ - -#pragma once -#include -#include "arm_compute/core/NEON/kernels/convolution/common/utils.hpp" - -#ifdef __aarch64__ - -template <> -inline void BlockedGemm<8, 12, float, float>( - const float* const a, const float* const b, float *c, - const int M, const int K, const int N, - const int a_row_stride, - const int b_row_stride, - const int c_row_stride -) { - const int M_BLOCK = 8; - const int N_BLOCK = 12; - - const int m_blocks = iceildiv(M, M_BLOCK); - const int n_blocks = iceildiv(N, N_BLOCK); - - // For each block of output rows - for (int mblock = 0; mblock < m_blocks; mblock++) { - // For each block of output columns - for (int nblock = 0; nblock < n_blocks; nblock++) { - const float *aptr = a + mblock*M_BLOCK*a_row_stride; - const float *bptr = b + nblock*N_BLOCK; - float *cptr = c + mblock*M_BLOCK*c_row_stride + nblock*N_BLOCK; - int k = K; - - asm volatile ( - // Create an 8x12 block of accumulators - " A_1 .req v27\n" - "sA_1 .req s27\n" - " A_2 .req v28\n" - "sA_2 .req s28\n" - " A_3 .req v29\n" - "sA_3 .req s29\n" - " A_4 .req v30\n" - "sA_4 .req s30\n" - - " B_1 .req v24\n" " B_2 .req v25\n" " B_3 .req v26\n" - "qB_1 .req q24\n" "qB_2 .req q25\n" "qB_3 .req q26\n" - - " C_11 .req v0\n" " C_12 .req v1\n" " C_13 .req v2\n" - " C_21 .req v3\n" " C_22 .req v4\n" " C_23 .req v5\n" - " C_31 .req v6\n" " C_32 .req v7\n" " C_33 .req v8\n" - " C_41 .req v9\n" " C_42 .req v10\n" " C_43 .req v11\n" - " C_51 .req v12\n" " C_52 .req v13\n" " C_53 .req v14\n" - " C_61 .req v15\n" " C_62 .req v16\n" " C_63 .req v17\n" - " C_71 .req v18\n" " C_72 .req v19\n" " C_73 .req v20\n" - " C_81 .req v21\n" " C_82 .req v22\n" " C_83 .req v23\n" - - "qC_11 .req q0\n" "qC_12 .req q1\n" "qC_13 .req q2\n" - "qC_21 .req q3\n" "qC_22 .req q4\n" "qC_23 .req q5\n" - "qC_31 .req q6\n" "qC_32 .req q7\n" "qC_33 .req q8\n" - "qC_41 .req q9\n" "qC_42 .req q10\n" "qC_43 .req q11\n" - "qC_51 .req q12\n" "qC_52 .req q13\n" "qC_53 .req q14\n" - "qC_61 .req q15\n" "qC_62 .req q16\n" "qC_63 .req q17\n" - "qC_71 .req q18\n" "qC_72 .req q19\n" "qC_73 .req q20\n" - "qC_81 .req q21\n" "qC_82 .req q22\n" "qC_83 .req q23\n" - - "aptr1 .req x17\n" - "aptr2 .req x18\n" - "aptr3 .req x19\n" - "aptr4 .req x20\n" - "aptr5 .req x21\n" - "aptr6 .req x22\n" - "aptr7 .req x23\n" - - // Initialise accumulators with 0 - // Initialise pointers - "movi C_11.4s, #0\n" - "add aptr1, %x[aptr], %x[a_row_stride]\n" - "movi C_12.4s, #0\n" - "add aptr2, aptr1, %x[a_row_stride]\n" - "movi C_13.4s, #0\n" - "add aptr3, aptr2, %x[a_row_stride]\n" - "movi C_21.4s, #0\n" - "add aptr4, aptr3, %x[a_row_stride]\n" - "movi C_22.4s, #0\n" - "add aptr5, aptr4, %x[a_row_stride]\n" - "movi C_23.4s, #0\n" - "add aptr6, aptr5, %x[a_row_stride]\n" - "movi C_31.4s, #0\n" - "add aptr7, aptr6, %x[a_row_stride]\n" - "movi C_32.4s, #0\n" - "ldr qB_1, [%x[bptr]]\n" - "movi C_33.4s, #0\n" - "ldr qB_2, [%x[bptr], #0x10]\n" - "movi C_41.4s, #0\n" - "prfm pldl1keep, [%x[bptr], #0x00]\n" - "movi C_42.4s, #0\n" - "prfm pldl1keep, [%x[bptr], #0x10]\n" - "movi C_43.4s, #0\n" - "prfm pldl1keep, [%x[bptr], #0x20]\n" - "movi C_51.4s, #0\n" - "prfm pldl1keep, [%x[aptr], #0x00]\n" - "movi C_52.4s, #0\n" - "prfm pldl1keep, [ aptr1, #0x00]\n" - "movi C_53.4s, #0\n" - "prfm pldl1keep, [ aptr2, #0x00]\n" - "movi C_61.4s, #0\n" - "prfm pldl1keep, [ aptr3, #0x00]\n" - "movi C_62.4s, #0\n" - "prfm pldl1keep, [ aptr4, #0x00]\n" - "movi C_63.4s, #0\n" - "prfm pldl1keep, [ aptr5, #0x00]\n" - "movi C_71.4s, #0\n" - "prfm pldl1keep, [ aptr6, #0x00]\n" - "movi C_72.4s, #0\n" - "prfm pldl1keep, [ aptr7, #0x00]\n" - "movi C_73.4s, #0\n" - "ldr sA_1, [%x[aptr]], #0x4\n" - "movi C_81.4s, #0\n" - "ldr sA_2, [ aptr1], #0x4\n" - "movi C_82.4s, #0\n" - "ldr sA_3, [ aptr2], #0x4\n" - "movi C_83.4s, #0\n" - "subs %x[k], %x[k], #1\n" - "beq 2f\n" - - "1:" - "fmla C_11.4s, B_1.4s, A_1.s[0]\n" - "ldr qB_3, [%x[bptr], #0x20]\n" - "fmla C_12.4s, B_2.4s, A_1.s[0]\n" - "ldr sA_4, [ aptr3], #0x4\n" - "fmla C_13.4s, B_3.4s, A_1.s[0]\n" - "ldr sA_1, [ aptr4], #0x04\n" - - "fmla C_21.4s, B_1.4s, A_2.s[0]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride]\n" - "fmla C_22.4s, B_2.4s, A_2.s[0]\n" - "prfm pldl1keep, [ aptr3, #0x10]\n" - "fmla C_23.4s, B_3.4s, A_2.s[0]\n" - "ldr sA_2, [ aptr5], #0x04\n" - - "fmla C_31.4s, B_1.4s, A_3.s[0]\n" - "prfm pldl1keep, [%x[bptr], #0x00]\n" - "fmla C_32.4s, B_2.4s, A_3.s[0]\n" - "prfm pldl1keep, [%x[bptr], #0x10]\n" - "fmla C_33.4s, B_3.4s, A_3.s[0]\n" - "ldr sA_3, [ aptr6], #0x04\n" - - "fmla C_41.4s, B_1.4s, A_4.s[0]\n" - "prfm pldl1keep, [%x[bptr], #0x20]\n" - "fmla C_42.4s, B_2.4s, A_4.s[0]\n" - "prfm pldl1keep, [ aptr4, #0x10]\n" - "fmla C_43.4s, B_3.4s, A_4.s[0]\n" - "ldr sA_4, [ aptr7], #0x04\n" - - "fmla C_51.4s, B_1.4s, A_1.s[0]\n" - "prfm pldl1keep, [ aptr5, #0x10]\n" - "fmla C_52.4s, B_2.4s, A_1.s[0]\n" - "prfm pldl1keep, [ aptr6, #0x10]\n" - "fmla C_53.4s, B_3.4s, A_1.s[0]\n" - "ldr sA_1, [%x[aptr]], #0x04\n" - - "fmla C_61.4s, B_1.4s, A_2.s[0]\n" - "prfm pldl1keep, [ aptr7, #0x10]\n" - "fmla C_62.4s, B_2.4s, A_2.s[0]\n" - "subs %x[k], %x[k], #1\n" - "fmla C_63.4s, B_3.4s, A_2.s[0]\n" - "ldr sA_2, [ aptr1], #0x04\n" - - "fmla C_71.4s, B_1.4s, A_3.s[0]\n" - "prfm pldl1keep, [%x[aptr], #0x10]\n" - "fmla C_72.4s, B_2.4s, A_3.s[0]\n" - "prfm pldl1keep, [ aptr1, #0x10]\n" - "fmla C_73.4s, B_3.4s, A_3.s[0]\n" - "ldr sA_3, [ aptr2], #0x04\n" - - "fmla C_81.4s, B_1.4s, A_4.s[0]\n" - "prfm pldl1keep, [ aptr2, #0x10]\n" - "fmla C_82.4s, B_2.4s, A_4.s[0]\n" - "ldp qB_1, qB_2, [%x[bptr]]\n" - "fmla C_83.4s, B_3.4s, A_4.s[0]\n" - "bne 1b\n" - - "2:" - "fmla C_11.4s, B_1.4s, A_1.s[0]\n" - "ldr qB_3, [%x[bptr], #0x20]\n" - "fmla C_12.4s, B_2.4s, A_1.s[0]\n" - "stp qC_11, qC_12, [%x[cptr]]\n" - "fmla C_13.4s, B_3.4s, A_1.s[0]\n" - "str qC_13, [%x[cptr], #0x20]\n" - "add %x[cptr], %x[cptr], %x[c_row_stride]\n" - "ldr sA_1, [ aptr4], #0x04\n" - - "fmla C_21.4s, B_1.4s, A_2.s[0]\n" - "ldr sA_4, [ aptr3], #0x4\n" - "fmla C_22.4s, B_2.4s, A_2.s[0]\n" - "stp qC_21, qC_22, [%x[cptr]]\n" - "fmla C_23.4s, B_3.4s, A_2.s[0]\n" - "str qC_23, [%x[cptr], #0x20]\n" - "add %x[cptr], %x[cptr], %x[c_row_stride]\n" - "ldr sA_2, [ aptr5], #0x04\n" - - "fmla C_31.4s, B_1.4s, A_3.s[0]\n" - "fmla C_32.4s, B_2.4s, A_3.s[0]\n" - "stp qC_31, qC_32, [%x[cptr]]\n" - "fmla C_33.4s, B_3.4s, A_3.s[0]\n" - "str qC_33, [%x[cptr], #0x20]\n" - "add %x[cptr], %x[cptr], %x[c_row_stride]\n" - "ldr sA_3, [ aptr6], #0x04\n" - - "fmla C_41.4s, B_1.4s, A_4.s[0]\n" - "fmla C_42.4s, B_2.4s, A_4.s[0]\n" - "stp qC_41, qC_42, [%x[cptr]]\n" - "fmla C_43.4s, B_3.4s, A_4.s[0]\n" - "str qC_43, [%x[cptr], #0x20]\n" - "add %x[cptr], %x[cptr], %x[c_row_stride]\n" - "ldr sA_4, [ aptr7], #0x04\n" - - "fmla C_51.4s, B_1.4s, A_1.s[0]\n" - "fmla C_52.4s, B_2.4s, A_1.s[0]\n" - "stp qC_51, qC_52, [%x[cptr]]\n" - "fmla C_53.4s, B_3.4s, A_1.s[0]\n" - "str qC_53, [%x[cptr], #0x20]\n" - "add %x[cptr], %x[cptr], %x[c_row_stride]\n" - - "fmla C_61.4s, B_1.4s, A_2.s[0]\n" - "fmla C_62.4s, B_2.4s, A_2.s[0]\n" - "stp qC_61, qC_62, [%x[cptr]]\n" - "fmla C_63.4s, B_3.4s, A_2.s[0]\n" - "str qC_63, [%x[cptr], #0x20]\n" - "add %x[cptr], %x[cptr], %x[c_row_stride]\n" - - "fmla C_71.4s, B_1.4s, A_3.s[0]\n" - "fmla C_72.4s, B_2.4s, A_3.s[0]\n" - "stp qC_71, qC_72, [%x[cptr]]\n" - "fmla C_73.4s, B_3.4s, A_3.s[0]\n" - "str qC_73, [%x[cptr], #0x20]\n" - "add %x[cptr], %x[cptr], %x[c_row_stride]\n" - - "fmla C_81.4s, B_1.4s, A_4.s[0]\n" - "fmla C_82.4s, B_2.4s, A_4.s[0]\n" - "stp qC_81, qC_82, [%x[cptr]]\n" - "fmla C_83.4s, B_3.4s, A_4.s[0]\n" - "str qC_83, [%x[cptr], #0x20]\n" - "add %x[cptr], %x[cptr], %x[c_row_stride]\n" - - // Clear aliases - ".unreq aptr1\n" - ".unreq aptr2\n" - ".unreq aptr3\n" - ".unreq aptr4\n" - ".unreq aptr5\n" - ".unreq aptr6\n" - ".unreq aptr7\n" - - ".unreq A_1\n" ".unreq A_2\n" ".unreq A_3\n" ".unreq A_4\n" - ".unreq sA_1\n" ".unreq sA_2\n" ".unreq sA_3\n" ".unreq sA_4\n" - - ".unreq B_1\n" ".unreq B_2\n" ".unreq B_3\n" - ".unreq qB_1\n" ".unreq qB_2\n" ".unreq qB_3\n" - - ".unreq C_11\n" ".unreq C_12\n" ".unreq C_13\n" - ".unreq C_21\n" ".unreq C_22\n" ".unreq C_23\n" - ".unreq C_31\n" ".unreq C_32\n" ".unreq C_33\n" - ".unreq C_41\n" ".unreq C_42\n" ".unreq C_43\n" - ".unreq C_51\n" ".unreq C_52\n" ".unreq C_53\n" - ".unreq C_61\n" ".unreq C_62\n" ".unreq C_63\n" - ".unreq C_71\n" ".unreq C_72\n" ".unreq C_73\n" - ".unreq C_81\n" ".unreq C_82\n" ".unreq C_83\n" - - ".unreq qC_11\n" ".unreq qC_12\n" ".unreq qC_13\n" - ".unreq qC_21\n" ".unreq qC_22\n" ".unreq qC_23\n" - ".unreq qC_31\n" ".unreq qC_32\n" ".unreq qC_33\n" - ".unreq qC_41\n" ".unreq qC_42\n" ".unreq qC_43\n" - ".unreq qC_51\n" ".unreq qC_52\n" ".unreq qC_53\n" - ".unreq qC_61\n" ".unreq qC_62\n" ".unreq qC_63\n" - ".unreq qC_71\n" ".unreq qC_72\n" ".unreq qC_73\n" - ".unreq qC_81\n" ".unreq qC_82\n" ".unreq qC_83\n" - : [aptr] "+r" (aptr), - [bptr] "+r" (bptr), - [cptr] "+r" (cptr), - [k] "+r" (k) - : [a_row_stride] "r" (a_row_stride * sizeof(float)), - [b_row_stride] "r" (b_row_stride * sizeof(float)), - [c_row_stride] "r" (c_row_stride * sizeof(float)) - : "cc", "memory", - "v0", "v1", "v2", "v3", "v4", "v5", "v6", "v7", "v8", "v9", "v10", - "v11", "v12", "v13", "v14", "v15", "v16", "v17", "v18", "v19", - "v20", "v21", "v22", "v23", "v24", "v25", "v26", "v27", "v28", - "v29", "v30", "x17", "x18", "x19", "x20", "x21", "x22", "x23" - ); - } - } -} - -/*****************************************************************************/ -/* 4x16 blocked GEMM with specialised tails - */ -#include "a64_sgemm_4x16.hpp" - -template <> -inline void BlockedGemm<4, 16, float, float>( - const float* const a, const float* const b, float *c, - const int M, const int K, const int N, - const int a_row_stride, - const int b_row_stride, - const int c_row_stride -) { - // Despatch based on tail of K - switch (K % 4) { - case 3: - sgemm_4x16_impl<3>( - a, b, c, M, K, N, a_row_stride, b_row_stride, c_row_stride - ); - break; - case 2: - sgemm_4x16_impl<2>( - a, b, c, M, K, N, a_row_stride, b_row_stride, c_row_stride - ); - break; - case 1: - sgemm_4x16_impl<1>( - a, b, c, M, K, N, a_row_stride, b_row_stride, c_row_stride - ); - break; - case 0: - sgemm_4x16_impl<0>( - a, b, c, M, K, N, a_row_stride, b_row_stride, c_row_stride - ); - break; - default: - assert(false); - } -} - -#endif // __aarch64__ diff --git a/arm_compute/core/NEON/kernels/convolution/winograd/gemm/a64_sgemm_4x16.hpp b/arm_compute/core/NEON/kernels/convolution/winograd/gemm/a64_sgemm_4x16.hpp deleted file mode 100644 index 5cd37de7a0..0000000000 --- a/arm_compute/core/NEON/kernels/convolution/winograd/gemm/a64_sgemm_4x16.hpp +++ /dev/null @@ -1,1446 +0,0 @@ -/* - * Copyright (c) 2017 ARM Limited. - * - * SPDX-License-Identifier: MIT - * - * Permission is hereby granted, free of charge, to any person obtaining a copy - * of this software and associated documentation files (the "Software"), to - * deal in the Software without restriction, including without limitation the - * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or - * sell copies of the Software, and to permit persons to whom the Software is - * furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in all - * copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE - * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER - * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, - * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. - */ - -template -inline void sgemm_4x16_impl( - const float* const a, const float* const b, float *c, - const int M, const int K, const int N, - const int a_row_stride, - const int b_row_stride, - const int c_row_stride -); - -template <> -inline void sgemm_4x16_impl<0>( - const float* const a, const float* const b, float *c, - const int M, const int K, const int N, - const int a_row_stride, - const int b_row_stride, - const int c_row_stride -) { - const int TAIL_SIZE = 0; - const int M_BLOCK = 4; - const int N_BLOCK = 16; - - const int m_blocks = iceildiv(M, M_BLOCK); - const int n_blocks = iceildiv(N, N_BLOCK); - - // For each block of output rows - for (int mblock = 0; mblock < m_blocks; mblock++) { - // For each block of output columns - for (int nblock = 0; nblock < n_blocks; nblock++) { - const float *aptr = a + mblock*M_BLOCK*a_row_stride; - const float *bptr = b + nblock*N_BLOCK; - float *cptr = c + mblock*M_BLOCK*c_row_stride + nblock*N_BLOCK; - int k = (K - TAIL_SIZE) / 4; - - asm volatile( - "aptr2 .req X20\n" - "aptr3 .req X21\n" - "aptr4 .req X22\n" - "vC11 .req v0\n" "vC12 .req v1\n" "vC13 .req v2\n" "vC14 .req v3\n" - "qC11 .req q0\n" "qC12 .req q1\n" "qC13 .req q2\n" "qC14 .req q3\n" - "vC21 .req v4\n" "vC22 .req v5\n" "vC23 .req v6\n" "vC24 .req v7\n" - "qC21 .req q4\n" "qC22 .req q5\n" "qC23 .req q6\n" "qC24 .req q7\n" - "vC31 .req v8\n" "vC32 .req v9\n" "vC33 .req v10\n" "vC34 .req v11\n" - "qC31 .req q8\n" "qC32 .req q9\n" "qC33 .req q10\n" "qC34 .req q11\n" - "vC41 .req v12\n" "vC42 .req v13\n" "vC43 .req v14\n" "vC44 .req v15\n" - "qC41 .req q12\n" "qC42 .req q13\n" "qC43 .req q14\n" "qC44 .req q15\n" - "vA1 .req v16\n" "qA1 .req q16\n" "dA1 .req d16\n" "sA1 .req s16\n" - "vA2 .req v17\n" "qA2 .req q17\n" "dA2 .req d17\n" "sA2 .req s17\n" - "vA3 .req v18\n" "qA3 .req q18\n" "dA3 .req d18\n" "sA3 .req s18\n" - "vA4 .req v19\n" "qA4 .req q19\n" "dA4 .req d19\n" "sA4 .req s19\n" - "vB1 .req v20\n" "qB1 .req q20\n" - "vB2 .req v21\n" "qB2 .req q21\n" - "vB3 .req v22\n" "qB3 .req q22\n" - "vB4 .req v23\n" "qB4 .req q23\n" - - // Clear accumulators, initialise pointers - "movi vC11.4s, #0\n" - "add aptr2, %x[aptr], %x[a_row_stride_bytes]\n" - "movi vC12.4s, #0\n" - "add aptr3, aptr2, %x[a_row_stride_bytes]\n" - "movi vC13.4s, #0\n" - "add aptr4, aptr3, %x[a_row_stride_bytes]\n" - "movi vC14.4s, #0\n" - "ldr qA1, [%x[aptr]], #0x10\n" - "movi vC21.4s, #0\n" - "ldr qA2, [ aptr2], #0x10\n" - "movi vC22.4s, #0\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "movi vC23.4s, #0\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "movi vC24.4s, #0\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "movi vC31.4s, #0\n" - "movi vC32.4s, #0\n" - "movi vC33.4s, #0\n" - "movi vC34.4s, #0\n" - "movi vC41.4s, #0\n" - "movi vC42.4s, #0\n" - "movi vC43.4s, #0\n" - "movi vC44.4s, #0\n" - "subs %x[k], %x[k], #1\n" - "beq 2f\n" - - "1:" // Loop proper - "fmla vC11.4s, vB1.4s, vA1.s[0]\n" - "ldr qA3, [ aptr3], #0x10\n" - "fmla vC21.4s, vB1.4s, vA2.s[0]\n" - "ldr qA4, [ aptr4], #0x10\n" - "fmla vC31.4s, vB1.4s, vA3.s[0]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[0]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[0]\n" - "fmla vC22.4s, vB2.4s, vA2.s[0]\n" - "fmla vC32.4s, vB2.4s, vA3.s[0]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[0]\n" - "fmla vC13.4s, vB3.4s, vA1.s[0]\n" - "fmla vC23.4s, vB3.4s, vA2.s[0]\n" - "fmla vC33.4s, vB3.4s, vA3.s[0]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[0]\n" - "fmla vC14.4s, vB4.4s, vA1.s[0]\n" - "fmla vC24.4s, vB4.4s, vA2.s[0]\n" - "fmla vC34.4s, vB4.4s, vA3.s[0]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[0]\n" - - "fmla vC11.4s, vB1.4s, vA1.s[1]\n" - "fmla vC21.4s, vB1.4s, vA2.s[1]\n" - "fmla vC31.4s, vB1.4s, vA3.s[1]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[1]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[1]\n" - "fmla vC22.4s, vB2.4s, vA2.s[1]\n" - "fmla vC32.4s, vB2.4s, vA3.s[1]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[1]\n" - "fmla vC13.4s, vB3.4s, vA1.s[1]\n" - "fmla vC23.4s, vB3.4s, vA2.s[1]\n" - "fmla vC33.4s, vB3.4s, vA3.s[1]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[1]\n" - "fmla vC14.4s, vB4.4s, vA1.s[1]\n" - "fmla vC24.4s, vB4.4s, vA2.s[1]\n" - "fmla vC34.4s, vB4.4s, vA3.s[1]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[1]\n" - - "fmla vC11.4s, vB1.4s, vA1.s[2]\n" - "fmla vC21.4s, vB1.4s, vA2.s[2]\n" - "fmla vC31.4s, vB1.4s, vA3.s[2]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[2]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[2]\n" - "fmla vC22.4s, vB2.4s, vA2.s[2]\n" - "fmla vC32.4s, vB2.4s, vA3.s[2]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[2]\n" - "fmla vC13.4s, vB3.4s, vA1.s[2]\n" - "fmla vC23.4s, vB3.4s, vA2.s[2]\n" - "fmla vC33.4s, vB3.4s, vA3.s[2]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[2]\n" - "fmla vC14.4s, vB4.4s, vA1.s[2]\n" - "fmla vC24.4s, vB4.4s, vA2.s[2]\n" - "fmla vC34.4s, vB4.4s, vA3.s[2]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[2]\n" - - "fmla vC11.4s, vB1.4s, vA1.s[3]\n" - "fmla vC21.4s, vB1.4s, vA2.s[3]\n" - "fmla vC31.4s, vB1.4s, vA3.s[3]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[3]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[3]\n" - "fmla vC22.4s, vB2.4s, vA2.s[3]\n" - "fmla vC32.4s, vB2.4s, vA3.s[3]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[3]\n" - "fmla vC13.4s, vB3.4s, vA1.s[3]\n" - "fmla vC23.4s, vB3.4s, vA2.s[3]\n" - "fmla vC33.4s, vB3.4s, vA3.s[3]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[3]\n" - "subs %x[k], %x[k], #1\n" - "fmla vC14.4s, vB4.4s, vA1.s[3]\n" - "ldr qA1, [%x[aptr]], #0x10\n" - "fmla vC24.4s, vB4.4s, vA2.s[3]\n" - "ldr qA2, [ aptr2], #0x10\n" - "fmla vC34.4s, vB4.4s, vA3.s[3]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[3]\n" - "bne 1b\n" - - "2:" // Tail - "fmla vC11.4s, vB1.4s, vA1.s[0]\n" - "ldr qA3, [ aptr3], #0x10\n" - "fmla vC21.4s, vB1.4s, vA2.s[0]\n" - "ldr qA4, [ aptr4], #0x10\n" - "fmla vC31.4s, vB1.4s, vA3.s[0]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[0]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[0]\n" - "fmla vC22.4s, vB2.4s, vA2.s[0]\n" - "fmla vC32.4s, vB2.4s, vA3.s[0]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[0]\n" - "fmla vC13.4s, vB3.4s, vA1.s[0]\n" - "fmla vC23.4s, vB3.4s, vA2.s[0]\n" - "fmla vC33.4s, vB3.4s, vA3.s[0]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[0]\n" - "fmla vC14.4s, vB4.4s, vA1.s[0]\n" - "fmla vC24.4s, vB4.4s, vA2.s[0]\n" - "fmla vC34.4s, vB4.4s, vA3.s[0]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[0]\n" - - "fmla vC11.4s, vB1.4s, vA1.s[1]\n" - "fmla vC21.4s, vB1.4s, vA2.s[1]\n" - "fmla vC31.4s, vB1.4s, vA3.s[1]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[1]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[1]\n" - "fmla vC22.4s, vB2.4s, vA2.s[1]\n" - "fmla vC32.4s, vB2.4s, vA3.s[1]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[1]\n" - "fmla vC13.4s, vB3.4s, vA1.s[1]\n" - "fmla vC23.4s, vB3.4s, vA2.s[1]\n" - "fmla vC33.4s, vB3.4s, vA3.s[1]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[1]\n" - "fmla vC14.4s, vB4.4s, vA1.s[1]\n" - "fmla vC24.4s, vB4.4s, vA2.s[1]\n" - "fmla vC34.4s, vB4.4s, vA3.s[1]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[1]\n" - - "fmla vC11.4s, vB1.4s, vA1.s[2]\n" - "fmla vC21.4s, vB1.4s, vA2.s[2]\n" - "fmla vC31.4s, vB1.4s, vA3.s[2]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[2]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[2]\n" - "fmla vC22.4s, vB2.4s, vA2.s[2]\n" - "fmla vC32.4s, vB2.4s, vA3.s[2]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[2]\n" - "fmla vC13.4s, vB3.4s, vA1.s[2]\n" - "fmla vC23.4s, vB3.4s, vA2.s[2]\n" - "fmla vC33.4s, vB3.4s, vA3.s[2]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[2]\n" - "fmla vC14.4s, vB4.4s, vA1.s[2]\n" - "fmla vC24.4s, vB4.4s, vA2.s[2]\n" - "fmla vC34.4s, vB4.4s, vA3.s[2]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[2]\n" - - "fmla vC11.4s, vB1.4s, vA1.s[3]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC12.4s, vB2.4s, vA1.s[3]\n" - "stp qC11, qC12, [%x[cptr], #0x00]\n" - "fmla vC13.4s, vB3.4s, vA1.s[3]\n" - "fmla vC14.4s, vB4.4s, vA1.s[3]\n" - "stp qC13, qC14, [%x[cptr], #0x20]\n" - "fmla vC21.4s, vB1.4s, vA2.s[3]\n" - "add %x[cptr], %x[cptr], %x[c_row_stride_bytes]\n" - "fmla vC22.4s, vB2.4s, vA2.s[3]\n" - "stp qC21, qC22, [%x[cptr], #0x00]\n" - "fmla vC23.4s, vB3.4s, vA2.s[3]\n" - "fmla vC24.4s, vB4.4s, vA2.s[3]\n" - "stp qC23, qC24, [%x[cptr], #0x20]\n" - "fmla vC31.4s, vB1.4s, vA3.s[3]\n" - "add %x[cptr], %x[cptr], %x[c_row_stride_bytes]\n" - "fmla vC32.4s, vB2.4s, vA3.s[3]\n" - "stp qC31, qC32, [%x[cptr], #0x00]\n" - "fmla vC33.4s, vB3.4s, vA3.s[3]\n" - "fmla vC34.4s, vB4.4s, vA3.s[3]\n" - "stp qC33, qC34, [%x[cptr], #0x20]\n" - "fmla vC41.4s, vB1.4s, vA4.s[3]\n" - "add %x[cptr], %x[cptr], %x[c_row_stride_bytes]\n" - "fmla vC42.4s, vB2.4s, vA4.s[3]\n" - "stp qC41, qC42, [%x[cptr], #0x00]\n" - "fmla vC43.4s, vB3.4s, vA4.s[3]\n" - "fmla vC44.4s, vB4.4s, vA4.s[3]\n" - "stp qC43, qC44, [%x[cptr], #0x20]\n" - "add %x[cptr], %x[cptr], %x[c_row_stride_bytes]\n" - - ".unreq vB4\n" ".unreq qB4\n" - ".unreq vB3\n" ".unreq qB3\n" - ".unreq vB2\n" ".unreq qB2\n" - ".unreq vB1\n" ".unreq qB1\n" - ".unreq vA4\n" ".unreq qA4\n" ".unreq dA4\n" ".unreq sA4\n" - ".unreq vA3\n" ".unreq qA3\n" ".unreq dA3\n" ".unreq sA3\n" - ".unreq vA2\n" ".unreq qA2\n" ".unreq dA2\n" ".unreq sA2\n" - ".unreq vA1\n" ".unreq qA1\n" ".unreq dA1\n" ".unreq sA1\n" - ".unreq qC41\n" ".unreq qC42\n" ".unreq qC43\n" ".unreq qC44\n" - ".unreq vC41\n" ".unreq vC42\n" ".unreq vC43\n" ".unreq vC44\n" - ".unreq qC31\n" ".unreq qC32\n" ".unreq qC33\n" ".unreq qC34\n" - ".unreq vC31\n" ".unreq vC32\n" ".unreq vC33\n" ".unreq vC34\n" - ".unreq qC21\n" ".unreq qC22\n" ".unreq qC23\n" ".unreq qC24\n" - ".unreq vC21\n" ".unreq vC22\n" ".unreq vC23\n" ".unreq vC24\n" - ".unreq qC11\n" ".unreq qC12\n" ".unreq qC13\n" ".unreq qC14\n" - ".unreq vC11\n" ".unreq vC12\n" ".unreq vC13\n" ".unreq vC14\n" - ".unreq aptr2\n" - ".unreq aptr3\n" - ".unreq aptr4\n" - - : [aptr] "+r" (aptr), - [bptr] "+r" (bptr), - [cptr] "+r" (cptr), - [k] "+r" (k) - : [a_row_stride_bytes] "r" (a_row_stride * sizeof(float)), - [b_row_stride_bytes] "r" (b_row_stride * sizeof(float)), - [c_row_stride_bytes] "r" (c_row_stride * sizeof(float)) - : "cc", "memory", "x20", "x21", "x22", - "v0", "v1", "v2", "v3", "v4", "v5", "v6", "v7", "v8", "v9", "v10", - "v11", "v12", "v13", "v14", "v15", "v16", "v17", "v18", "v19", "v20", - "v21", "v22", "v23" - ); - } - } -} - -template <> -inline void sgemm_4x16_impl<1>( - const float* const a, const float* const b, float *c, - const int M, const int K, const int N, - const int a_row_stride, - const int b_row_stride, - const int c_row_stride -) { - const int TAIL_SIZE = 1; - const int M_BLOCK = 4; - const int N_BLOCK = 16; - - const int m_blocks = iceildiv(M, M_BLOCK); - const int n_blocks = iceildiv(N, N_BLOCK); - - // For each block of output rows - for (int mblock = 0; mblock < m_blocks; mblock++) { - // For each block of output columns - for (int nblock = 0; nblock < n_blocks; nblock++) { - const float *aptr = a + mblock*M_BLOCK*a_row_stride; - const float *bptr = b + nblock*N_BLOCK; - float *cptr = c + mblock*M_BLOCK*c_row_stride + nblock*N_BLOCK; - int k = (K - TAIL_SIZE) / 4; - - asm volatile( - "aptr2 .req X20\n" - "aptr3 .req X21\n" - "aptr4 .req X22\n" - "vC11 .req v0\n" "vC12 .req v1\n" "vC13 .req v2\n" "vC14 .req v3\n" - "qC11 .req q0\n" "qC12 .req q1\n" "qC13 .req q2\n" "qC14 .req q3\n" - "vC21 .req v4\n" "vC22 .req v5\n" "vC23 .req v6\n" "vC24 .req v7\n" - "qC21 .req q4\n" "qC22 .req q5\n" "qC23 .req q6\n" "qC24 .req q7\n" - "vC31 .req v8\n" "vC32 .req v9\n" "vC33 .req v10\n" "vC34 .req v11\n" - "qC31 .req q8\n" "qC32 .req q9\n" "qC33 .req q10\n" "qC34 .req q11\n" - "vC41 .req v12\n" "vC42 .req v13\n" "vC43 .req v14\n" "vC44 .req v15\n" - "qC41 .req q12\n" "qC42 .req q13\n" "qC43 .req q14\n" "qC44 .req q15\n" - "vA1 .req v16\n" "qA1 .req q16\n" "dA1 .req d16\n" "sA1 .req s16\n" - "vA2 .req v17\n" "qA2 .req q17\n" "dA2 .req d17\n" "sA2 .req s17\n" - "vA3 .req v18\n" "qA3 .req q18\n" "dA3 .req d18\n" "sA3 .req s18\n" - "vA4 .req v19\n" "qA4 .req q19\n" "dA4 .req d19\n" "sA4 .req s19\n" - "vB1 .req v20\n" "qB1 .req q20\n" - "vB2 .req v21\n" "qB2 .req q21\n" - "vB3 .req v22\n" "qB3 .req q22\n" - "vB4 .req v23\n" "qB4 .req q23\n" - - // Clear accumulators, initialise pointers - "movi vC11.4s, #0\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "movi vC12.4s, #0\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "movi vC13.4s, #0\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "movi vC14.4s, #0\n" - "add aptr2, %x[aptr], %x[a_row_stride_bytes]\n" - "movi vC21.4s, #0\n" - "add aptr3, aptr2, %x[a_row_stride_bytes]\n" - "movi vC22.4s, #0\n" - "add aptr4, aptr3, %x[a_row_stride_bytes]\n" - "movi vC23.4s, #0\n" - "cbnz %x[k], 3f\n" - - // Prepare for tail in K - "movi vC24.4s, #0\n" - "ldr sA1, [%x[aptr]], #0x04\n" - "movi vC31.4s, #0\n" - "ldr sA2, [ aptr2], #0x04\n" - "movi vC32.4s, #0\n" - "movi vC33.4s, #0\n" - "movi vC34.4s, #0\n" - "movi vC41.4s, #0\n" - "movi vC42.4s, #0\n" - "movi vC43.4s, #0\n" - "movi vC44.4s, #0\n" - "b 2f\n" // Jump to tail - - "3:" // Prepare for loop over K - "movi vC24.4s, #0\n" - "ldr qA1, [%x[aptr]], #0x10\n" - "movi vC31.4s, #0\n" - "ldr qA2, [ aptr2], #0x10\n" - "movi vC32.4s, #0\n" - "movi vC33.4s, #0\n" - "movi vC34.4s, #0\n" - "movi vC41.4s, #0\n" - "movi vC42.4s, #0\n" - "movi vC43.4s, #0\n" - "movi vC44.4s, #0\n" - "subs %x[k], %x[k], #1\n" - "beq 4f\n" - - "1:" // Loop proper - "fmla vC11.4s, vB1.4s, vA1.s[0]\n" - "ldr qA3, [ aptr3], #0x10\n" - "fmla vC21.4s, vB1.4s, vA2.s[0]\n" - "ldr qA4, [ aptr4], #0x10\n" - "fmla vC31.4s, vB1.4s, vA3.s[0]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[0]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[0]\n" - "fmla vC22.4s, vB2.4s, vA2.s[0]\n" - "fmla vC32.4s, vB2.4s, vA3.s[0]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[0]\n" - "fmla vC13.4s, vB3.4s, vA1.s[0]\n" - "fmla vC23.4s, vB3.4s, vA2.s[0]\n" - "fmla vC33.4s, vB3.4s, vA3.s[0]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[0]\n" - "fmla vC14.4s, vB4.4s, vA1.s[0]\n" - "fmla vC24.4s, vB4.4s, vA2.s[0]\n" - "fmla vC34.4s, vB4.4s, vA3.s[0]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[0]\n" - - "fmla vC11.4s, vB1.4s, vA1.s[1]\n" - "fmla vC21.4s, vB1.4s, vA2.s[1]\n" - "fmla vC31.4s, vB1.4s, vA3.s[1]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[1]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[1]\n" - "fmla vC22.4s, vB2.4s, vA2.s[1]\n" - "fmla vC32.4s, vB2.4s, vA3.s[1]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[1]\n" - "fmla vC13.4s, vB3.4s, vA1.s[1]\n" - "fmla vC23.4s, vB3.4s, vA2.s[1]\n" - "fmla vC33.4s, vB3.4s, vA3.s[1]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[1]\n" - "fmla vC14.4s, vB4.4s, vA1.s[1]\n" - "fmla vC24.4s, vB4.4s, vA2.s[1]\n" - "fmla vC34.4s, vB4.4s, vA3.s[1]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[1]\n" - - "fmla vC11.4s, vB1.4s, vA1.s[2]\n" - "fmla vC21.4s, vB1.4s, vA2.s[2]\n" - "fmla vC31.4s, vB1.4s, vA3.s[2]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[2]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[2]\n" - "fmla vC22.4s, vB2.4s, vA2.s[2]\n" - "fmla vC32.4s, vB2.4s, vA3.s[2]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[2]\n" - "fmla vC13.4s, vB3.4s, vA1.s[2]\n" - "fmla vC23.4s, vB3.4s, vA2.s[2]\n" - "fmla vC33.4s, vB3.4s, vA3.s[2]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[2]\n" - "fmla vC14.4s, vB4.4s, vA1.s[2]\n" - "fmla vC24.4s, vB4.4s, vA2.s[2]\n" - "fmla vC34.4s, vB4.4s, vA3.s[2]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[2]\n" - - "fmla vC11.4s, vB1.4s, vA1.s[3]\n" - "fmla vC21.4s, vB1.4s, vA2.s[3]\n" - "fmla vC31.4s, vB1.4s, vA3.s[3]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[3]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[3]\n" - "fmla vC22.4s, vB2.4s, vA2.s[3]\n" - "fmla vC32.4s, vB2.4s, vA3.s[3]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[3]\n" - "fmla vC13.4s, vB3.4s, vA1.s[3]\n" - "fmla vC23.4s, vB3.4s, vA2.s[3]\n" - "fmla vC33.4s, vB3.4s, vA3.s[3]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[3]\n" - "subs %x[k], %x[k], #1\n" - "fmla vC14.4s, vB4.4s, vA1.s[3]\n" - "ldr qA1, [%x[aptr]], #0x10\n" - "fmla vC24.4s, vB4.4s, vA2.s[3]\n" - "ldr qA2, [ aptr2], #0x10\n" - "fmla vC34.4s, vB4.4s, vA3.s[3]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[3]\n" - "bne 1b\n" - - "4:" // Tail iteration - "fmla vC11.4s, vB1.4s, vA1.s[0]\n" - "ldr qA3, [ aptr3], #0x10\n" - "fmla vC21.4s, vB1.4s, vA2.s[0]\n" - "ldr qA4, [ aptr4], #0x10\n" - "fmla vC31.4s, vB1.4s, vA3.s[0]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[0]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[0]\n" - "fmla vC22.4s, vB2.4s, vA2.s[0]\n" - "fmla vC32.4s, vB2.4s, vA3.s[0]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[0]\n" - "fmla vC13.4s, vB3.4s, vA1.s[0]\n" - "fmla vC23.4s, vB3.4s, vA2.s[0]\n" - "fmla vC33.4s, vB3.4s, vA3.s[0]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[0]\n" - "fmla vC14.4s, vB4.4s, vA1.s[0]\n" - "fmla vC24.4s, vB4.4s, vA2.s[0]\n" - "fmla vC34.4s, vB4.4s, vA3.s[0]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[0]\n" - - "fmla vC11.4s, vB1.4s, vA1.s[1]\n" - "fmla vC21.4s, vB1.4s, vA2.s[1]\n" - "fmla vC31.4s, vB1.4s, vA3.s[1]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[1]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[1]\n" - "fmla vC22.4s, vB2.4s, vA2.s[1]\n" - "fmla vC32.4s, vB2.4s, vA3.s[1]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[1]\n" - "fmla vC13.4s, vB3.4s, vA1.s[1]\n" - "fmla vC23.4s, vB3.4s, vA2.s[1]\n" - "fmla vC33.4s, vB3.4s, vA3.s[1]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[1]\n" - "fmla vC14.4s, vB4.4s, vA1.s[1]\n" - "fmla vC24.4s, vB4.4s, vA2.s[1]\n" - "fmla vC34.4s, vB4.4s, vA3.s[1]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[1]\n" - - "fmla vC11.4s, vB1.4s, vA1.s[2]\n" - "fmla vC21.4s, vB1.4s, vA2.s[2]\n" - "fmla vC31.4s, vB1.4s, vA3.s[2]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[2]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[2]\n" - "fmla vC22.4s, vB2.4s, vA2.s[2]\n" - "fmla vC32.4s, vB2.4s, vA3.s[2]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[2]\n" - "fmla vC13.4s, vB3.4s, vA1.s[2]\n" - "fmla vC23.4s, vB3.4s, vA2.s[2]\n" - "fmla vC33.4s, vB3.4s, vA3.s[2]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[2]\n" - "fmla vC14.4s, vB4.4s, vA1.s[2]\n" - "fmla vC24.4s, vB4.4s, vA2.s[2]\n" - "fmla vC34.4s, vB4.4s, vA3.s[2]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[2]\n" - - "fmla vC11.4s, vB1.4s, vA1.s[3]\n" - "fmla vC21.4s, vB1.4s, vA2.s[3]\n" - "fmla vC31.4s, vB1.4s, vA3.s[3]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[3]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[3]\n" - "fmla vC22.4s, vB2.4s, vA2.s[3]\n" - "fmla vC32.4s, vB2.4s, vA3.s[3]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[3]\n" - "fmla vC13.4s, vB3.4s, vA1.s[3]\n" - "fmla vC23.4s, vB3.4s, vA2.s[3]\n" - "fmla vC33.4s, vB3.4s, vA3.s[3]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[3]\n" - "fmla vC14.4s, vB4.4s, vA1.s[3]\n" - "ldr sA1, [%x[aptr]], #0x04\n" - "fmla vC24.4s, vB4.4s, vA2.s[3]\n" - "ldr sA2, [ aptr2], #0x04\n" - "fmla vC34.4s, vB4.4s, vA3.s[3]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[3]\n" - - "2:" // Common tail - "fmla vC11.4s, vB1.4s, vA1.s[0]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC12.4s, vB2.4s, vA1.s[0]\n" - "stp qC11, qC12, [%x[cptr], #0x00]\n" - "fmla vC13.4s, vB3.4s, vA1.s[0]\n" - "ldr sA3, [ aptr3], #0x04\n" - "fmla vC14.4s, vB4.4s, vA1.s[0]\n" - "stp qC13, qC14, [%x[cptr], #0x20]\n" - "fmla vC21.4s, vB1.4s, vA2.s[0]\n" - "add %x[cptr], %x[cptr], %x[c_row_stride_bytes]\n" - "fmla vC22.4s, vB2.4s, vA2.s[0]\n" - "stp qC21, qC22, [%x[cptr], #0x00]\n" - "fmla vC23.4s, vB3.4s, vA2.s[0]\n" - "ldr sA4, [ aptr4], #0x04\n" - "fmla vC24.4s, vB4.4s, vA2.s[0]\n" - "stp qC23, qC24, [%x[cptr], #0x20]\n" - "fmla vC31.4s, vB1.4s, vA3.s[0]\n" - "add %x[cptr], %x[cptr], %x[c_row_stride_bytes]\n" - "fmla vC32.4s, vB2.4s, vA3.s[0]\n" - "stp qC31, qC32, [%x[cptr], #0x00]\n" - "fmla vC33.4s, vB3.4s, vA3.s[0]\n" - "fmla vC34.4s, vB4.4s, vA3.s[0]\n" - "stp qC33, qC34, [%x[cptr], #0x20]\n" - "fmla vC41.4s, vB1.4s, vA4.s[0]\n" - "add %x[cptr], %x[cptr], %x[c_row_stride_bytes]\n" - "fmla vC42.4s, vB2.4s, vA4.s[0]\n" - "stp qC41, qC42, [%x[cptr], #0x00]\n" - "fmla vC43.4s, vB3.4s, vA4.s[0]\n" - "fmla vC44.4s, vB4.4s, vA4.s[0]\n" - "stp qC43, qC44, [%x[cptr], #0x20]\n" - "add %x[cptr], %x[cptr], %x[c_row_stride_bytes]\n" - - ".unreq vB4\n" ".unreq qB4\n" - ".unreq vB3\n" ".unreq qB3\n" - ".unreq vB2\n" ".unreq qB2\n" - ".unreq vB1\n" ".unreq qB1\n" - ".unreq vA4\n" ".unreq qA4\n" ".unreq dA4\n" ".unreq sA4\n" - ".unreq vA3\n" ".unreq qA3\n" ".unreq dA3\n" ".unreq sA3\n" - ".unreq vA2\n" ".unreq qA2\n" ".unreq dA2\n" ".unreq sA2\n" - ".unreq vA1\n" ".unreq qA1\n" ".unreq dA1\n" ".unreq sA1\n" - ".unreq qC41\n" ".unreq qC42\n" ".unreq qC43\n" ".unreq qC44\n" - ".unreq vC41\n" ".unreq vC42\n" ".unreq vC43\n" ".unreq vC44\n" - ".unreq qC31\n" ".unreq qC32\n" ".unreq qC33\n" ".unreq qC34\n" - ".unreq vC31\n" ".unreq vC32\n" ".unreq vC33\n" ".unreq vC34\n" - ".unreq qC21\n" ".unreq qC22\n" ".unreq qC23\n" ".unreq qC24\n" - ".unreq vC21\n" ".unreq vC22\n" ".unreq vC23\n" ".unreq vC24\n" - ".unreq qC11\n" ".unreq qC12\n" ".unreq qC13\n" ".unreq qC14\n" - ".unreq vC11\n" ".unreq vC12\n" ".unreq vC13\n" ".unreq vC14\n" - ".unreq aptr2\n" - ".unreq aptr3\n" - ".unreq aptr4\n" - - : [aptr] "+r" (aptr), - [bptr] "+r" (bptr), - [cptr] "+r" (cptr), - [k] "+r" (k) - : [a_row_stride_bytes] "r" (a_row_stride * sizeof(float)), - [b_row_stride_bytes] "r" (b_row_stride * sizeof(float)), - [c_row_stride_bytes] "r" (c_row_stride * sizeof(float)) - : "cc", "memory", "x20", "x21", "x22", - "v0", "v1", "v2", "v3", "v4", "v5", "v6", "v7", "v8", "v9", "v10", - "v11", "v12", "v13", "v14", "v15", "v16", "v17", "v18", "v19", "v20", - "v21", "v22", "v23" - ); - } - } -} - -template <> -inline void sgemm_4x16_impl<2>( - const float* const a, const float* const b, float *c, - const int M, const int K, const int N, - const int a_row_stride, - const int b_row_stride, - const int c_row_stride -) { - const int TAIL_SIZE = 2; - const int M_BLOCK = 4; - const int N_BLOCK = 16; - - const int m_blocks = iceildiv(M, M_BLOCK); - const int n_blocks = iceildiv(N, N_BLOCK); - - // For each block of output rows - for (int mblock = 0; mblock < m_blocks; mblock++) { - // For each block of output columns - for (int nblock = 0; nblock < n_blocks; nblock++) { - const float *aptr = a + mblock*M_BLOCK*a_row_stride; - const float *bptr = b + nblock*N_BLOCK; - float *cptr = c + mblock*M_BLOCK*c_row_stride + nblock*N_BLOCK; - int k = (K - TAIL_SIZE) / 4; - - asm volatile( - "aptr2 .req X20\n" - "aptr3 .req X21\n" - "aptr4 .req X22\n" - "vC11 .req v0\n" "vC12 .req v1\n" "vC13 .req v2\n" "vC14 .req v3\n" - "qC11 .req q0\n" "qC12 .req q1\n" "qC13 .req q2\n" "qC14 .req q3\n" - "vC21 .req v4\n" "vC22 .req v5\n" "vC23 .req v6\n" "vC24 .req v7\n" - "qC21 .req q4\n" "qC22 .req q5\n" "qC23 .req q6\n" "qC24 .req q7\n" - "vC31 .req v8\n" "vC32 .req v9\n" "vC33 .req v10\n" "vC34 .req v11\n" - "qC31 .req q8\n" "qC32 .req q9\n" "qC33 .req q10\n" "qC34 .req q11\n" - "vC41 .req v12\n" "vC42 .req v13\n" "vC43 .req v14\n" "vC44 .req v15\n" - "qC41 .req q12\n" "qC42 .req q13\n" "qC43 .req q14\n" "qC44 .req q15\n" - "vA1 .req v16\n" "qA1 .req q16\n" "dA1 .req d16\n" "sA1 .req s16\n" - "vA2 .req v17\n" "qA2 .req q17\n" "dA2 .req d17\n" "sA2 .req s17\n" - "vA3 .req v18\n" "qA3 .req q18\n" "dA3 .req d18\n" "sA3 .req s18\n" - "vA4 .req v19\n" "qA4 .req q19\n" "dA4 .req d19\n" "sA4 .req s19\n" - "vB1 .req v20\n" "qB1 .req q20\n" - "vB2 .req v21\n" "qB2 .req q21\n" - "vB3 .req v22\n" "qB3 .req q22\n" - "vB4 .req v23\n" "qB4 .req q23\n" - - // Clear accumulators, initialise pointers - "movi vC11.4s, #0\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "movi vC12.4s, #0\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "movi vC13.4s, #0\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "movi vC14.4s, #0\n" - "add aptr2, %x[aptr], %x[a_row_stride_bytes]\n" - "movi vC21.4s, #0\n" - "add aptr3, aptr2, %x[a_row_stride_bytes]\n" - "movi vC22.4s, #0\n" - "add aptr4, aptr3, %x[a_row_stride_bytes]\n" - "movi vC23.4s, #0\n" - "cbnz %x[k], 3f\n" - - // Prepare for tail in K - "movi vC24.4s, #0\n" - "ldr dA1, [%x[aptr]], #0x08\n" - "movi vC31.4s, #0\n" - "ldr dA2, [ aptr2], #0x08\n" - "movi vC32.4s, #0\n" - "movi vC33.4s, #0\n" - "movi vC34.4s, #0\n" - "movi vC41.4s, #0\n" - "movi vC42.4s, #0\n" - "movi vC43.4s, #0\n" - "movi vC44.4s, #0\n" - "b 2f\n" // Jump to tail - - "3:" // Prepare for loop over K - "movi vC24.4s, #0\n" - "ldr qA1, [%x[aptr]], #0x10\n" - "movi vC31.4s, #0\n" - "ldr qA2, [ aptr2], #0x10\n" - "movi vC32.4s, #0\n" - "movi vC33.4s, #0\n" - "movi vC34.4s, #0\n" - "movi vC41.4s, #0\n" - "movi vC42.4s, #0\n" - "movi vC43.4s, #0\n" - "movi vC44.4s, #0\n" - "subs %x[k], %x[k], #1\n" - "beq 4f\n" - - "1:" // Loop proper - "fmla vC11.4s, vB1.4s, vA1.s[0]\n" - "ldr qA3, [ aptr3], #0x10\n" - "fmla vC21.4s, vB1.4s, vA2.s[0]\n" - "ldr qA4, [ aptr4], #0x10\n" - "fmla vC31.4s, vB1.4s, vA3.s[0]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[0]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[0]\n" - "fmla vC22.4s, vB2.4s, vA2.s[0]\n" - "fmla vC32.4s, vB2.4s, vA3.s[0]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[0]\n" - "fmla vC13.4s, vB3.4s, vA1.s[0]\n" - "fmla vC23.4s, vB3.4s, vA2.s[0]\n" - "fmla vC33.4s, vB3.4s, vA3.s[0]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[0]\n" - "fmla vC14.4s, vB4.4s, vA1.s[0]\n" - "fmla vC24.4s, vB4.4s, vA2.s[0]\n" - "fmla vC34.4s, vB4.4s, vA3.s[0]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[0]\n" - - "fmla vC11.4s, vB1.4s, vA1.s[1]\n" - "fmla vC21.4s, vB1.4s, vA2.s[1]\n" - "fmla vC31.4s, vB1.4s, vA3.s[1]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[1]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[1]\n" - "fmla vC22.4s, vB2.4s, vA2.s[1]\n" - "fmla vC32.4s, vB2.4s, vA3.s[1]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[1]\n" - "fmla vC13.4s, vB3.4s, vA1.s[1]\n" - "fmla vC23.4s, vB3.4s, vA2.s[1]\n" - "fmla vC33.4s, vB3.4s, vA3.s[1]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[1]\n" - "fmla vC14.4s, vB4.4s, vA1.s[1]\n" - "fmla vC24.4s, vB4.4s, vA2.s[1]\n" - "fmla vC34.4s, vB4.4s, vA3.s[1]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[1]\n" - - "fmla vC11.4s, vB1.4s, vA1.s[2]\n" - "fmla vC21.4s, vB1.4s, vA2.s[2]\n" - "fmla vC31.4s, vB1.4s, vA3.s[2]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[2]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[2]\n" - "fmla vC22.4s, vB2.4s, vA2.s[2]\n" - "fmla vC32.4s, vB2.4s, vA3.s[2]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[2]\n" - "fmla vC13.4s, vB3.4s, vA1.s[2]\n" - "fmla vC23.4s, vB3.4s, vA2.s[2]\n" - "fmla vC33.4s, vB3.4s, vA3.s[2]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[2]\n" - "fmla vC14.4s, vB4.4s, vA1.s[2]\n" - "fmla vC24.4s, vB4.4s, vA2.s[2]\n" - "fmla vC34.4s, vB4.4s, vA3.s[2]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[2]\n" - - "fmla vC11.4s, vB1.4s, vA1.s[3]\n" - "fmla vC21.4s, vB1.4s, vA2.s[3]\n" - "fmla vC31.4s, vB1.4s, vA3.s[3]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[3]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[3]\n" - "fmla vC22.4s, vB2.4s, vA2.s[3]\n" - "fmla vC32.4s, vB2.4s, vA3.s[3]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[3]\n" - "fmla vC13.4s, vB3.4s, vA1.s[3]\n" - "fmla vC23.4s, vB3.4s, vA2.s[3]\n" - "fmla vC33.4s, vB3.4s, vA3.s[3]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[3]\n" - "subs %x[k], %x[k], #1\n" - "fmla vC14.4s, vB4.4s, vA1.s[3]\n" - "ldr qA1, [%x[aptr]], #0x10\n" - "fmla vC24.4s, vB4.4s, vA2.s[3]\n" - "ldr qA2, [ aptr2], #0x10\n" - "fmla vC34.4s, vB4.4s, vA3.s[3]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[3]\n" - "bne 1b\n" - - "4:" // Tail iteration - "fmla vC11.4s, vB1.4s, vA1.s[0]\n" - "ldr qA3, [ aptr3], #0x10\n" - "fmla vC21.4s, vB1.4s, vA2.s[0]\n" - "ldr qA4, [ aptr4], #0x10\n" - "fmla vC31.4s, vB1.4s, vA3.s[0]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[0]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[0]\n" - "fmla vC22.4s, vB2.4s, vA2.s[0]\n" - "fmla vC32.4s, vB2.4s, vA3.s[0]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[0]\n" - "fmla vC13.4s, vB3.4s, vA1.s[0]\n" - "fmla vC23.4s, vB3.4s, vA2.s[0]\n" - "fmla vC33.4s, vB3.4s, vA3.s[0]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[0]\n" - "fmla vC14.4s, vB4.4s, vA1.s[0]\n" - "fmla vC24.4s, vB4.4s, vA2.s[0]\n" - "fmla vC34.4s, vB4.4s, vA3.s[0]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[0]\n" - - "fmla vC11.4s, vB1.4s, vA1.s[1]\n" - "fmla vC21.4s, vB1.4s, vA2.s[1]\n" - "fmla vC31.4s, vB1.4s, vA3.s[1]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[1]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[1]\n" - "fmla vC22.4s, vB2.4s, vA2.s[1]\n" - "fmla vC32.4s, vB2.4s, vA3.s[1]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[1]\n" - "fmla vC13.4s, vB3.4s, vA1.s[1]\n" - "fmla vC23.4s, vB3.4s, vA2.s[1]\n" - "fmla vC33.4s, vB3.4s, vA3.s[1]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[1]\n" - "fmla vC14.4s, vB4.4s, vA1.s[1]\n" - "fmla vC24.4s, vB4.4s, vA2.s[1]\n" - "fmla vC34.4s, vB4.4s, vA3.s[1]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[1]\n" - - "fmla vC11.4s, vB1.4s, vA1.s[2]\n" - "fmla vC21.4s, vB1.4s, vA2.s[2]\n" - "fmla vC31.4s, vB1.4s, vA3.s[2]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[2]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[2]\n" - "fmla vC22.4s, vB2.4s, vA2.s[2]\n" - "fmla vC32.4s, vB2.4s, vA3.s[2]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[2]\n" - "fmla vC13.4s, vB3.4s, vA1.s[2]\n" - "fmla vC23.4s, vB3.4s, vA2.s[2]\n" - "fmla vC33.4s, vB3.4s, vA3.s[2]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[2]\n" - "fmla vC14.4s, vB4.4s, vA1.s[2]\n" - "fmla vC24.4s, vB4.4s, vA2.s[2]\n" - "fmla vC34.4s, vB4.4s, vA3.s[2]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[2]\n" - - "fmla vC11.4s, vB1.4s, vA1.s[3]\n" - "fmla vC21.4s, vB1.4s, vA2.s[3]\n" - "fmla vC31.4s, vB1.4s, vA3.s[3]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[3]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[3]\n" - "fmla vC22.4s, vB2.4s, vA2.s[3]\n" - "fmla vC32.4s, vB2.4s, vA3.s[3]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[3]\n" - "fmla vC13.4s, vB3.4s, vA1.s[3]\n" - "fmla vC23.4s, vB3.4s, vA2.s[3]\n" - "fmla vC33.4s, vB3.4s, vA3.s[3]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[3]\n" - "fmla vC14.4s, vB4.4s, vA1.s[3]\n" - "ldr dA1, [%x[aptr]], #0x08\n" - "fmla vC24.4s, vB4.4s, vA2.s[3]\n" - "ldr dA2, [ aptr2], #0x08\n" - "fmla vC34.4s, vB4.4s, vA3.s[3]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[3]\n" - - "2:" // Common tail - "fmla vC11.4s, vB1.4s, vA1.s[0]\n" - "ldr dA3, [ aptr3], #0x08\n" - "fmla vC21.4s, vB1.4s, vA2.s[0]\n" - "ldr dA4, [ aptr4], #0x08\n" - "fmla vC31.4s, vB1.4s, vA3.s[0]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[0]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[0]\n" - "fmla vC22.4s, vB2.4s, vA2.s[0]\n" - "fmla vC32.4s, vB2.4s, vA3.s[0]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[0]\n" - "fmla vC13.4s, vB3.4s, vA1.s[0]\n" - "fmla vC23.4s, vB3.4s, vA2.s[0]\n" - "fmla vC33.4s, vB3.4s, vA3.s[0]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[0]\n" - "fmla vC14.4s, vB4.4s, vA1.s[0]\n" - "fmla vC24.4s, vB4.4s, vA2.s[0]\n" - "fmla vC34.4s, vB4.4s, vA3.s[0]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[0]\n" - - "fmla vC11.4s, vB1.4s, vA1.s[1]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC12.4s, vB2.4s, vA1.s[1]\n" - "stp qC11, qC12, [%x[cptr], #0x00]\n" - "fmla vC13.4s, vB3.4s, vA1.s[1]\n" - "fmla vC14.4s, vB4.4s, vA1.s[1]\n" - "stp qC13, qC14, [%x[cptr], #0x20]\n" - "fmla vC21.4s, vB1.4s, vA2.s[1]\n" - "add %x[cptr], %x[cptr], %x[c_row_stride_bytes]\n" - "fmla vC22.4s, vB2.4s, vA2.s[1]\n" - "stp qC21, qC22, [%x[cptr], #0x00]\n" - "fmla vC23.4s, vB3.4s, vA2.s[1]\n" - "fmla vC24.4s, vB4.4s, vA2.s[1]\n" - "stp qC23, qC24, [%x[cptr], #0x20]\n" - "fmla vC31.4s, vB1.4s, vA3.s[1]\n" - "add %x[cptr], %x[cptr], %x[c_row_stride_bytes]\n" - "fmla vC32.4s, vB2.4s, vA3.s[1]\n" - "stp qC31, qC32, [%x[cptr], #0x00]\n" - "fmla vC33.4s, vB3.4s, vA3.s[1]\n" - "fmla vC34.4s, vB4.4s, vA3.s[1]\n" - "stp qC33, qC34, [%x[cptr], #0x20]\n" - "fmla vC41.4s, vB1.4s, vA4.s[1]\n" - "add %x[cptr], %x[cptr], %x[c_row_stride_bytes]\n" - "fmla vC42.4s, vB2.4s, vA4.s[1]\n" - "stp qC41, qC42, [%x[cptr], #0x00]\n" - "fmla vC43.4s, vB3.4s, vA4.s[1]\n" - "fmla vC44.4s, vB4.4s, vA4.s[1]\n" - "stp qC43, qC44, [%x[cptr], #0x20]\n" - "add %x[cptr], %x[cptr], %x[c_row_stride_bytes]\n" - - ".unreq vB4\n" ".unreq qB4\n" - ".unreq vB3\n" ".unreq qB3\n" - ".unreq vB2\n" ".unreq qB2\n" - ".unreq vB1\n" ".unreq qB1\n" - ".unreq vA4\n" ".unreq qA4\n" ".unreq dA4\n" ".unreq sA4\n" - ".unreq vA3\n" ".unreq qA3\n" ".unreq dA3\n" ".unreq sA3\n" - ".unreq vA2\n" ".unreq qA2\n" ".unreq dA2\n" ".unreq sA2\n" - ".unreq vA1\n" ".unreq qA1\n" ".unreq dA1\n" ".unreq sA1\n" - ".unreq qC41\n" ".unreq qC42\n" ".unreq qC43\n" ".unreq qC44\n" - ".unreq vC41\n" ".unreq vC42\n" ".unreq vC43\n" ".unreq vC44\n" - ".unreq qC31\n" ".unreq qC32\n" ".unreq qC33\n" ".unreq qC34\n" - ".unreq vC31\n" ".unreq vC32\n" ".unreq vC33\n" ".unreq vC34\n" - ".unreq qC21\n" ".unreq qC22\n" ".unreq qC23\n" ".unreq qC24\n" - ".unreq vC21\n" ".unreq vC22\n" ".unreq vC23\n" ".unreq vC24\n" - ".unreq qC11\n" ".unreq qC12\n" ".unreq qC13\n" ".unreq qC14\n" - ".unreq vC11\n" ".unreq vC12\n" ".unreq vC13\n" ".unreq vC14\n" - ".unreq aptr2\n" - ".unreq aptr3\n" - ".unreq aptr4\n" - - : [aptr] "+r" (aptr), - [bptr] "+r" (bptr), - [cptr] "+r" (cptr), - [k] "+r" (k) - : [a_row_stride_bytes] "r" (a_row_stride * sizeof(float)), - [b_row_stride_bytes] "r" (b_row_stride * sizeof(float)), - [c_row_stride_bytes] "r" (c_row_stride * sizeof(float)) - : "cc", "memory", "x20", "x21", "x22", - "v0", "v1", "v2", "v3", "v4", "v5", "v6", "v7", "v8", "v9", "v10", - "v11", "v12", "v13", "v14", "v15", "v16", "v17", "v18", "v19", "v20", - "v21", "v22", "v23" - ); - } - } -} - -template <> -inline void sgemm_4x16_impl<3>( - const float* const a, const float* const b, float *c, - const int M, const int K, const int N, - const int a_row_stride, - const int b_row_stride, - const int c_row_stride -) { - const int TAIL_SIZE = 3; - const int M_BLOCK = 4; - const int N_BLOCK = 16; - - const int m_blocks = iceildiv(M, M_BLOCK); - const int n_blocks = iceildiv(N, N_BLOCK); - - // For each block of output rows - for (int mblock = 0; mblock < m_blocks; mblock++) { - // For each block of output columns - for (int nblock = 0; nblock < n_blocks; nblock++) { - const float *aptr = a + mblock*M_BLOCK*a_row_stride; - const float *bptr = b + nblock*N_BLOCK; - float *cptr = c + mblock*M_BLOCK*c_row_stride + nblock*N_BLOCK; - int k = (K - TAIL_SIZE) / 4; - - asm volatile( - "aptr2 .req X20\n" - "aptr3 .req X21\n" - "aptr4 .req X22\n" - "vC11 .req v0\n" "vC12 .req v1\n" "vC13 .req v2\n" "vC14 .req v3\n" - "qC11 .req q0\n" "qC12 .req q1\n" "qC13 .req q2\n" "qC14 .req q3\n" - "vC21 .req v4\n" "vC22 .req v5\n" "vC23 .req v6\n" "vC24 .req v7\n" - "qC21 .req q4\n" "qC22 .req q5\n" "qC23 .req q6\n" "qC24 .req q7\n" - "vC31 .req v8\n" "vC32 .req v9\n" "vC33 .req v10\n" "vC34 .req v11\n" - "qC31 .req q8\n" "qC32 .req q9\n" "qC33 .req q10\n" "qC34 .req q11\n" - "vC41 .req v12\n" "vC42 .req v13\n" "vC43 .req v14\n" "vC44 .req v15\n" - "qC41 .req q12\n" "qC42 .req q13\n" "qC43 .req q14\n" "qC44 .req q15\n" - "vA1 .req v16\n" "qA1 .req q16\n" "dA1 .req d16\n" "sA1 .req s16\n" - "vA2 .req v17\n" "qA2 .req q17\n" "dA2 .req d17\n" "sA2 .req s17\n" - "vA3 .req v18\n" "qA3 .req q18\n" "dA3 .req d18\n" "sA3 .req s18\n" - "vA4 .req v19\n" "qA4 .req q19\n" "dA4 .req d19\n" "sA4 .req s19\n" - "vB1 .req v20\n" "qB1 .req q20\n" - "vB2 .req v21\n" "qB2 .req q21\n" - "vB3 .req v22\n" "qB3 .req q22\n" - "vB4 .req v23\n" "qB4 .req q23\n" - - // Clear accumulators, initialise pointers - "movi vC11.4s, #0\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "movi vC12.4s, #0\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "movi vC13.4s, #0\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "movi vC14.4s, #0\n" - "add aptr2, %x[aptr], %x[a_row_stride_bytes]\n" - "movi vC21.4s, #0\n" - "add aptr3, aptr2, %x[a_row_stride_bytes]\n" - "movi vC22.4s, #0\n" - "add aptr4, aptr3, %x[a_row_stride_bytes]\n" - "movi vC23.4s, #0\n" - "cbnz %x[k], 3f\n" - - // Prepare for tail in K - "movi vC24.4s, #0\n" - "ldr dA1, [%x[aptr]], #0x08\n" - "movi vC31.4s, #0\n" - "ldr dA2, [ aptr2], #0x08\n" - "movi vC32.4s, #0\n" - "movi vC33.4s, #0\n" - "movi vC34.4s, #0\n" - "movi vC41.4s, #0\n" - "movi vC42.4s, #0\n" - "movi vC43.4s, #0\n" - "movi vC44.4s, #0\n" - "b 2f\n" // Jump to tail - - "3:" // Prepare for loop over K - "movi vC24.4s, #0\n" - "ldr qA1, [%x[aptr]], #0x10\n" - "movi vC31.4s, #0\n" - "ldr qA2, [ aptr2], #0x10\n" - "movi vC32.4s, #0\n" - "movi vC33.4s, #0\n" - "movi vC34.4s, #0\n" - "movi vC41.4s, #0\n" - "movi vC42.4s, #0\n" - "movi vC43.4s, #0\n" - "movi vC44.4s, #0\n" - "subs %x[k], %x[k], #1\n" - "beq 4f\n" - - "1:" // Loop proper - "fmla vC11.4s, vB1.4s, vA1.s[0]\n" - "ldr qA3, [ aptr3], #0x10\n" - "fmla vC21.4s, vB1.4s, vA2.s[0]\n" - "ldr qA4, [ aptr4], #0x10\n" - "fmla vC31.4s, vB1.4s, vA3.s[0]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[0]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[0]\n" - "fmla vC22.4s, vB2.4s, vA2.s[0]\n" - "fmla vC32.4s, vB2.4s, vA3.s[0]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[0]\n" - "fmla vC13.4s, vB3.4s, vA1.s[0]\n" - "fmla vC23.4s, vB3.4s, vA2.s[0]\n" - "fmla vC33.4s, vB3.4s, vA3.s[0]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[0]\n" - "fmla vC14.4s, vB4.4s, vA1.s[0]\n" - "fmla vC24.4s, vB4.4s, vA2.s[0]\n" - "fmla vC34.4s, vB4.4s, vA3.s[0]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[0]\n" - - "fmla vC11.4s, vB1.4s, vA1.s[1]\n" - "fmla vC21.4s, vB1.4s, vA2.s[1]\n" - "fmla vC31.4s, vB1.4s, vA3.s[1]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[1]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[1]\n" - "fmla vC22.4s, vB2.4s, vA2.s[1]\n" - "fmla vC32.4s, vB2.4s, vA3.s[1]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[1]\n" - "fmla vC13.4s, vB3.4s, vA1.s[1]\n" - "fmla vC23.4s, vB3.4s, vA2.s[1]\n" - "fmla vC33.4s, vB3.4s, vA3.s[1]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[1]\n" - "fmla vC14.4s, vB4.4s, vA1.s[1]\n" - "fmla vC24.4s, vB4.4s, vA2.s[1]\n" - "fmla vC34.4s, vB4.4s, vA3.s[1]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[1]\n" - - "fmla vC11.4s, vB1.4s, vA1.s[2]\n" - "fmla vC21.4s, vB1.4s, vA2.s[2]\n" - "fmla vC31.4s, vB1.4s, vA3.s[2]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[2]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[2]\n" - "fmla vC22.4s, vB2.4s, vA2.s[2]\n" - "fmla vC32.4s, vB2.4s, vA3.s[2]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[2]\n" - "fmla vC13.4s, vB3.4s, vA1.s[2]\n" - "fmla vC23.4s, vB3.4s, vA2.s[2]\n" - "fmla vC33.4s, vB3.4s, vA3.s[2]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[2]\n" - "fmla vC14.4s, vB4.4s, vA1.s[2]\n" - "fmla vC24.4s, vB4.4s, vA2.s[2]\n" - "fmla vC34.4s, vB4.4s, vA3.s[2]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[2]\n" - - "fmla vC11.4s, vB1.4s, vA1.s[3]\n" - "fmla vC21.4s, vB1.4s, vA2.s[3]\n" - "fmla vC31.4s, vB1.4s, vA3.s[3]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[3]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[3]\n" - "fmla vC22.4s, vB2.4s, vA2.s[3]\n" - "fmla vC32.4s, vB2.4s, vA3.s[3]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[3]\n" - "fmla vC13.4s, vB3.4s, vA1.s[3]\n" - "fmla vC23.4s, vB3.4s, vA2.s[3]\n" - "fmla vC33.4s, vB3.4s, vA3.s[3]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[3]\n" - "subs %x[k], %x[k], #1\n" - "fmla vC14.4s, vB4.4s, vA1.s[3]\n" - "ldr qA1, [%x[aptr]], #0x10\n" - "fmla vC24.4s, vB4.4s, vA2.s[3]\n" - "ldr qA2, [ aptr2], #0x10\n" - "fmla vC34.4s, vB4.4s, vA3.s[3]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[3]\n" - "bne 1b\n" - - "4:" // Tail iteration - "fmla vC11.4s, vB1.4s, vA1.s[0]\n" - "ldr qA3, [ aptr3], #0x10\n" - "fmla vC21.4s, vB1.4s, vA2.s[0]\n" - "ldr qA4, [ aptr4], #0x10\n" - "fmla vC31.4s, vB1.4s, vA3.s[0]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[0]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[0]\n" - "fmla vC22.4s, vB2.4s, vA2.s[0]\n" - "fmla vC32.4s, vB2.4s, vA3.s[0]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[0]\n" - "fmla vC13.4s, vB3.4s, vA1.s[0]\n" - "fmla vC23.4s, vB3.4s, vA2.s[0]\n" - "fmla vC33.4s, vB3.4s, vA3.s[0]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[0]\n" - "fmla vC14.4s, vB4.4s, vA1.s[0]\n" - "fmla vC24.4s, vB4.4s, vA2.s[0]\n" - "fmla vC34.4s, vB4.4s, vA3.s[0]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[0]\n" - - "fmla vC11.4s, vB1.4s, vA1.s[1]\n" - "fmla vC21.4s, vB1.4s, vA2.s[1]\n" - "fmla vC31.4s, vB1.4s, vA3.s[1]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[1]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[1]\n" - "fmla vC22.4s, vB2.4s, vA2.s[1]\n" - "fmla vC32.4s, vB2.4s, vA3.s[1]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[1]\n" - "fmla vC13.4s, vB3.4s, vA1.s[1]\n" - "fmla vC23.4s, vB3.4s, vA2.s[1]\n" - "fmla vC33.4s, vB3.4s, vA3.s[1]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[1]\n" - "fmla vC14.4s, vB4.4s, vA1.s[1]\n" - "fmla vC24.4s, vB4.4s, vA2.s[1]\n" - "fmla vC34.4s, vB4.4s, vA3.s[1]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[1]\n" - - "fmla vC11.4s, vB1.4s, vA1.s[2]\n" - "fmla vC21.4s, vB1.4s, vA2.s[2]\n" - "fmla vC31.4s, vB1.4s, vA3.s[2]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[2]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[2]\n" - "fmla vC22.4s, vB2.4s, vA2.s[2]\n" - "fmla vC32.4s, vB2.4s, vA3.s[2]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[2]\n" - "fmla vC13.4s, vB3.4s, vA1.s[2]\n" - "fmla vC23.4s, vB3.4s, vA2.s[2]\n" - "fmla vC33.4s, vB3.4s, vA3.s[2]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[2]\n" - "fmla vC14.4s, vB4.4s, vA1.s[2]\n" - "fmla vC24.4s, vB4.4s, vA2.s[2]\n" - "fmla vC34.4s, vB4.4s, vA3.s[2]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[2]\n" - - "fmla vC11.4s, vB1.4s, vA1.s[3]\n" - "fmla vC21.4s, vB1.4s, vA2.s[3]\n" - "fmla vC31.4s, vB1.4s, vA3.s[3]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[3]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[3]\n" - "fmla vC22.4s, vB2.4s, vA2.s[3]\n" - "fmla vC32.4s, vB2.4s, vA3.s[3]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[3]\n" - "fmla vC13.4s, vB3.4s, vA1.s[3]\n" - "fmla vC23.4s, vB3.4s, vA2.s[3]\n" - "fmla vC33.4s, vB3.4s, vA3.s[3]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[3]\n" - "fmla vC14.4s, vB4.4s, vA1.s[3]\n" - "ldr dA1, [%x[aptr]], #0x08\n" - "fmla vC24.4s, vB4.4s, vA2.s[3]\n" - "ldr dA2, [ aptr2], #0x08\n" - "fmla vC34.4s, vB4.4s, vA3.s[3]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[3]\n" - - "2:" // Common tail - "fmla vC11.4s, vB1.4s, vA1.s[0]\n" - "ldr dA3, [ aptr3], #0x08\n" - "fmla vC21.4s, vB1.4s, vA2.s[0]\n" - "ldr dA4, [ aptr4], #0x08\n" - "fmla vC31.4s, vB1.4s, vA3.s[0]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[0]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[0]\n" - "fmla vC22.4s, vB2.4s, vA2.s[0]\n" - "fmla vC32.4s, vB2.4s, vA3.s[0]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[0]\n" - "fmla vC13.4s, vB3.4s, vA1.s[0]\n" - "fmla vC23.4s, vB3.4s, vA2.s[0]\n" - "fmla vC33.4s, vB3.4s, vA3.s[0]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[0]\n" - "fmla vC14.4s, vB4.4s, vA1.s[0]\n" - "fmla vC24.4s, vB4.4s, vA2.s[0]\n" - "fmla vC34.4s, vB4.4s, vA3.s[0]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[0]\n" - - "fmla vC11.4s, vB1.4s, vA1.s[1]\n" - "fmla vC21.4s, vB1.4s, vA2.s[1]\n" - "fmla vC31.4s, vB1.4s, vA3.s[1]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC41.4s, vB1.4s, vA4.s[1]\n" - "add %x[bptr], %x[bptr], %x[b_row_stride_bytes]\n" - "fmla vC12.4s, vB2.4s, vA1.s[1]\n" - "fmla vC22.4s, vB2.4s, vA2.s[1]\n" - "fmla vC32.4s, vB2.4s, vA3.s[1]\n" - "ldr qB1, [%x[bptr], #0x00]\n" - "fmla vC42.4s, vB2.4s, vA4.s[1]\n" - "fmla vC13.4s, vB3.4s, vA1.s[1]\n" - "fmla vC23.4s, vB3.4s, vA2.s[1]\n" - "fmla vC33.4s, vB3.4s, vA3.s[1]\n" - "ldr qB2, [%x[bptr], #0x10]\n" - "fmla vC43.4s, vB3.4s, vA4.s[1]\n" - "fmla vC14.4s, vB4.4s, vA1.s[1]\n" - "ldr sA1, [%x[aptr]], #0x04\n" - "fmla vC24.4s, vB4.4s, vA2.s[1]\n" - "ldr sA2, [ aptr2], #0x04\n" - "fmla vC34.4s, vB4.4s, vA3.s[1]\n" - "ldr qB3, [%x[bptr], #0x20]\n" - "fmla vC44.4s, vB4.4s, vA4.s[1]\n" - - "fmla vC11.4s, vB1.4s, vA1.s[0]\n" - "ldr qB4, [%x[bptr], #0x30]\n" - "fmla vC12.4s, vB2.4s, vA1.s[0]\n" - "stp qC11, qC12, [%x[cptr], #0x00]\n" - "fmla vC13.4s, vB3.4s, vA1.s[0]\n" - "ldr sA3, [ aptr3], #0x04\n" - "fmla vC14.4s, vB4.4s, vA1.s[0]\n" - "stp qC13, qC14, [%x[cptr], #0x20]\n" - "fmla vC21.4s, vB1.4s, vA2.s[0]\n" - "add %x[cptr], %x[cptr], %x[c_row_stride_bytes]\n" - "fmla vC22.4s, vB2.4s, vA2.s[0]\n" - "stp qC21, qC22, [%x[cptr], #0x00]\n" - "fmla vC23.4s, vB3.4s, vA2.s[0]\n" - "ldr sA4, [ aptr4], #0x04\n" - "fmla vC24.4s, vB4.4s, vA2.s[0]\n" - "stp qC23, qC24, [%x[cptr], #0x20]\n" - "fmla vC31.4s, vB1.4s, vA3.s[0]\n" - "add %x[cptr], %x[cptr], %x[c_row_stride_bytes]\n" - "fmla vC32.4s, vB2.4s, vA3.s[0]\n" - "stp qC31, qC32, [%x[cptr], #0x00]\n" - "fmla vC33.4s, vB3.4s, vA3.s[0]\n" - "fmla vC34.4s, vB4.4s, vA3.s[0]\n" - "stp qC33, qC34, [%x[cptr], #0x20]\n" - "fmla vC41.4s, vB1.4s, vA4.s[0]\n" - "add %x[cptr], %x[cptr], %x[c_row_stride_bytes]\n" - "fmla vC42.4s, vB2.4s, vA4.s[0]\n" - "stp qC41, qC42, [%x[cptr], #0x00]\n" - "fmla vC43.4s, vB3.4s, vA4.s[0]\n" - "fmla vC44.4s, vB4.4s, vA4.s[0]\n" - "stp qC43, qC44, [%x[cptr], #0x20]\n" - "add %x[cptr], %x[cptr], %x[c_row_stride_bytes]\n" - - ".unreq vB4\n" ".unreq qB4\n" - ".unreq vB3\n" ".unreq qB3\n" - ".unreq vB2\n" ".unreq qB2\n" - ".unreq vB1\n" ".unreq qB1\n" - ".unreq vA4\n" ".unreq qA4\n" ".unreq dA4\n" ".unreq sA4\n" - ".unreq vA3\n" ".unreq qA3\n" ".unreq dA3\n" ".unreq sA3\n" - ".unreq vA2\n" ".unreq qA2\n" ".unreq dA2\n" ".unreq sA2\n" - ".unreq vA1\n" ".unreq qA1\n" ".unreq dA1\n" ".unreq sA1\n" - ".unreq qC41\n" ".unreq qC42\n" ".unreq qC43\n" ".unreq qC44\n" - ".unreq vC41\n" ".unreq vC42\n" ".unreq vC43\n" ".unreq vC44\n" - ".unreq qC31\n" ".unreq qC32\n" ".unreq qC33\n" ".unreq qC34\n" - ".unreq vC31\n" ".unreq vC32\n" ".unreq vC33\n" ".unreq vC34\n" - ".unreq qC21\n" ".unreq qC22\n" ".unreq qC23\n" ".unreq qC24\n" - ".unreq vC21\n" ".unreq vC22\n" ".unreq vC23\n" ".unreq vC24\n" - ".unreq qC11\n" ".unreq qC12\n" ".unreq qC13\n" ".unreq qC14\n" - ".unreq vC11\n" ".unreq vC12\n" ".unreq vC13\n" ".unreq vC14\n" - ".unreq aptr2\n" - ".unreq aptr3\n" - ".unreq aptr4\n" - - : [aptr] "+r" (aptr), - [bptr] "+r" (bptr), - [cptr] "+r" (cptr), - [k] "+r" (k) - : [a_row_stride_bytes] "r" (a_row_stride * sizeof(float)), - [b_row_stride_bytes] "r" (b_row_stride * sizeof(float)), - [c_row_stride_bytes] "r" (c_row_stride * sizeof(float)) - : "cc", "memory", "x20", "x21", "x22", - "v0", "v1", "v2", "v3", "v4", "v5", "v6", "v7", "v8", "v9", "v10", - "v11", "v12", "v13", "v14", "v15", "v16", "v17", "v18", "v19", "v20", - "v21", "v22", "v23" - ); - } - } -} diff --git a/arm_compute/core/NEON/kernels/convolution/winograd/transforms/input.hpp b/arm_compute/core/NEON/kernels/convolution/winograd/transforms/input.hpp deleted file mode 100644 index b813bbb25c..0000000000 --- a/arm_compute/core/NEON/kernels/convolution/winograd/transforms/input.hpp +++ /dev/null @@ -1,349 +0,0 @@ -/* - * Copyright (c) 2017 ARM Limited. - * - * SPDX-License-Identifier: MIT - * - * Permission is hereby granted, free of charge, to any person obtaining a copy - * of this software and associated documentation files (the "Software"), to - * deal in the Software without restriction, including without limitation the - * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or - * sell copies of the Software, and to permit persons to whom the Software is - * furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in all - * copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE - * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER - * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, - * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. - */ - -#pragma once -#include "../winograd_gemm.hpp" - -namespace winograd -{ - /***************************************************************************/ - /* Instance-less API */ - template - void InputTransformImpl::execute( - const T* const input, /** Input tensor data */ - const int n_batches, /** Number of batches in input tensor. */ - const int in_batch_stride, /** Stride between batches of the input. */ - const int n_rows, /** Number of rows in input tensor. */ - const int in_row_stride, /** Stride between rows of the input. */ - const int n_cols, /** Number of columns in input tensor. */ - const int in_col_stride, /** Stride between columns of the input. */ - const int n_channels, /** Number of channels in input tensor. */ - const PaddingType padding, /** Padding type. */ - const int tile_M, - const int tile_N, - T* const output, /** Base of output matrices. */ - const int matrix_stride, /** Stride between output matrices. */ - const int matrix_batch_stride, /** Stride between batches within the matrix. */ - const int matrix_row_stride /** Stride within matrices. */ - ) - { - // Compute the padding required on each edge of the image - const int pad_top = (padding == PADDING_SAME) ? (KernelRows - 1) / 2 : 0; - const int pad_left = (padding == PADDING_SAME) ? (KernelCols - 1) / 2 : 0; - - // Compute striding values (assuming NHWC ordered data) - const int output_col_stride = matrix_row_stride; - const int output_row_stride = tile_N * output_col_stride; - - // Loop over batches - for (int batch = 0; batch < n_batches; batch++) - { - // Pointer to the batch - const T* const input_base_batch = input + batch * in_batch_stride; - T* const outptr_base_batch = output + batch * matrix_batch_stride; - - // Loop over rows of tiles - for (int tile_i = 0; tile_i < tile_M; tile_i++) - { - // Padding (top + bottom) for the row - const int row_top = tile_i*(InnerTileRows - overlap_rows) - pad_top; - const int row_bottom = row_top + InnerTileRows; - const int row_pad_top = std::max(0, pad_top - tile_i*(InnerTileRows - overlap_rows)); - const int row_pad_bottom = (row_bottom <= n_rows) ? 0 : row_bottom - n_rows; - - // Pointer to the row - const int row_offset = std::min(0, row_pad_top - pad_top); - const T* const input_base_row = ( - input_base_batch + ((InnerTileRows - overlap_rows)*tile_i + row_offset)*in_row_stride - ); - T* const outptr_base_row = outptr_base_batch + tile_i*output_row_stride; - - // Process the row - process_tile_row( - tile_N, n_channels, - input_base_row, in_row_stride, in_col_stride, - outptr_base_row, matrix_stride, matrix_row_stride, - row_pad_top, pad_left, row_pad_bottom, n_cols - ); - } - } - } - - - template - void InputTransformImpl::execute( - const T* const input, /** Input tensor data */ - const int n_batches, /** Number of batches in input tensor. */ - const int in_batch_stride, /** Stride between batches of the input. */ - const int n_rows, /** Number of rows in input tensor. */ - const int in_row_stride, /** Stride between rows of the input. */ - const int n_cols, /** Number of columns in input tensor. */ - const int in_col_stride, /** Stride between columns of the input. */ - const int n_channels, /** Number of channels in input tensor. */ - const PaddingType padding, /** Padding type. */ - const int tile_M, - const int tile_N, - T* const output, /** Base of output matrices. */ - const int matrix_stride, /** Stride between output matrices. */ - const int matrix_batch_stride, /** Stride between batches within the matrix. */ - const int matrix_row_stride /** Stride within matrices. */ - ) - { - // If an Nx1 kernel then transpose and redirect to the 1xN implementation - InputTransformImpl<1, KernelRows, 1, InnerTileRows, T>::execute( - input, - n_batches, in_batch_stride, - n_cols, in_col_stride, - n_rows, in_row_stride, - n_channels, padding, - tile_N, tile_M, - output, matrix_stride, matrix_batch_stride, matrix_row_stride - ); - } - - template - void InputTransformImpl::process_tile_row( - const int tile_N, - int n_channels, - const T* const input_base, - const int input_row_stride, - const int input_col_stride, - T* const matrix_base, - const int matrix_stride, - const int matrix_row_stride, - const int pad_top, - const int row_pad_left, - const int pad_bottom, - const int n_cols - ) - { - // Loop over columns of tiles - for (int tile_j = 0; tile_j < tile_N; tile_j++) - { - // Padding (left + right) for the tile - const int t_start = tile_j*(InnerTileCols - overlap_cols) - row_pad_left; - const int t_end = t_start + InnerTileCols; - const int t_pad_left = std::max(0, row_pad_left - tile_j*(InnerTileCols - overlap_cols)); - const int t_pad_right = (t_end <= n_cols) ? 0 : t_end - n_cols; - - // Get pointers into the inputs and outputs - const int col_offset = std::min(0, t_pad_left - row_pad_left); - const T* const input_base_col = ( - input_base + ((InnerTileCols - overlap_cols)*tile_j + col_offset)*input_col_stride - ); - T* const outptr = matrix_base + tile_j*matrix_row_stride; - - // Apply the specific tile processing function - const typename Tiles::TileFn tilefn = Tiles::get_tile_specialization( - pad_top, t_pad_left, pad_bottom, t_pad_right - ); - - tilefn( - n_channels, - input_base_col, input_row_stride, input_col_stride, - outptr, matrix_stride, - pad_top, t_pad_left, pad_bottom, t_pad_right - ); - } - } - - /***************************************************************************/ - template - InputTransform::InputTransform( - const T* const input, /** Input tensor data */ - const int n_batches, /** Number of batches in input tensor. */ - const int n_rows, /** Number of rows in input tensor. */ - const int n_cols, /** Number of columns in input tensor. */ - const int n_channels, /** Number of channels in input tensor. */ - const PaddingType padding, /** Padding type. */ - T* const output, /** Base of output matrices. */ - const int matrix_stride, /** Stride between output matrices. */ - const int matrix_row_stride, /** Stride within matrices. */ - const int in_batch_stride, /** Stride between input batches. */ - const int in_row_stride, /** Stride between input rows. */ - const int in_col_stride /** Stride between input columns. */ - ) : _inptr(input), _outptr(output), - _n_batches(n_batches), _n_rows(n_rows), _n_cols(n_cols), _n_channels(n_channels), - _matrix_stride(matrix_stride), _matrix_row_stride(matrix_row_stride), - _tiles_M(iceildiv((padding == PADDING_SAME) ? n_rows : n_rows - KernelRows + 1, - InnerTileRows - KernelRows + 1)), - _tiles_N(iceildiv((padding == PADDING_SAME) ? n_cols : n_cols - KernelCols + 1, - InnerTileCols - KernelCols + 1)), - _in_col_stride(in_col_stride ? in_col_stride : n_channels), - _in_row_stride(in_row_stride ? in_row_stride : n_cols * _in_col_stride), - _in_batch_stride(in_batch_stride ? in_batch_stride : n_rows * _in_row_stride), - _padding_type(padding) - { - } - - template - unsigned int InputTransform::get_window() const - { - // The final window includes the tail, all other windows will be a multiple - // of the window block in size. - return iceildiv(_n_channels, WINDOW_BLOCK); - } - - template - void InputTransform::run( - const unsigned int start, const unsigned int stop - ) - { - if (start >= get_window()) - { - return; - } - - // Determine the window of work to perform - const unsigned int start_channel = start * WINDOW_BLOCK; - const unsigned int stop_channel = std::min( - stop * WINDOW_BLOCK, _n_channels - ); - const unsigned int n_channels = stop_channel - start_channel; - - // Perform the work - execute( - _inptr + start_channel, - _n_batches, _in_batch_stride, - _n_rows, _in_row_stride, - _n_cols, _in_col_stride, - n_channels, - _padding_type, - _tiles_M, - _tiles_N, - _outptr + start_channel, - _matrix_stride, - _matrix_row_stride * _tiles_M * _tiles_N, - _matrix_row_stride - ); - } - - template - void InputTransform::execute( - const T* const input, /** Input tensor data */ - const int n_batches, /** Number of batches in input tensor. */ - const int in_batch_stride, /** Stride between batches of the input. */ - const int n_rows, /** Number of rows in input tensor. */ - const int in_row_stride, /** Stride between rows of the input. */ - const int n_cols, /** Number of columns in input tensor. */ - const int in_col_stride, /** Stride between columns of the input. */ - const int n_channels, /** Number of channels in input tensor. */ - const PaddingType padding, /** Padding type. */ - const int tile_M, - const int tile_N, - T* const output, /** Base of output matrices. */ - const int matrix_stride, /** Stride between output matrices. */ - const int matrix_batch_stride, /** Stride between batches within the matrix. */ - const int matrix_row_stride /** Stride within matrices. */ - ) - { - Transform::execute( - input, n_batches, in_batch_stride, n_rows, in_row_stride, n_cols, - in_col_stride, n_channels, padding, tile_M, tile_N, output, - matrix_stride, matrix_batch_stride, matrix_row_stride - ); - } - - template - typename InputTransformImplTiles::TileFn - InputTransformImplTiles:: - get_tile_specialization( - const int pad_top, - const int pad_left, - const int pad_bottom, - const int pad_right - ) - { - if (!(pad_top || pad_left || pad_bottom || pad_right)) - { - // No padding, return unpadded specialisation - return tilefn_unpadded; - } - else if (pad_top && !(pad_left || pad_bottom || pad_right)) - { - // Top padding only - const int index = (pad_top - min_pad_top) / (InnerTileRows - overlap_rows); - return tilefn_top_padded[index]; - } - else if (!(pad_top) && pad_left && !(pad_bottom || pad_right)) - { - // Left padding only - const int index = (pad_left - min_pad_left) / (InnerTileCols - overlap_cols); - return tilefn_left_padded[index]; - } - else if (!(pad_top || pad_left) && pad_bottom && !(pad_right)) - { - // Bottom padding only - return tilefn_bottom_padded[pad_bottom - 1]; - } - else if (!(pad_top || pad_left || pad_bottom) && pad_right) - { - // Right padding only - return tilefn_right_padded[pad_right - 1]; - } - else - { - // Combination of paddings, return an unspecialised method - return tilefn_generic; - } - } - - template - typename InputTransformImplTiles<1, KernelCols, 1, InnerTileCols, T>::TileFn - InputTransformImplTiles<1, KernelCols, 1, InnerTileCols, T>:: - get_tile_specialization( - const int pad_top, - const int pad_left, - const int pad_bottom, - const int pad_right - ) - { - (void) pad_top; - (void) pad_bottom; - - if (!(pad_left || pad_right)) - { - // No padding, return unpadded specialisation - return tilefn_unpadded; - } - else if (pad_left && !pad_right) - { - // Left padding only - const int index = (pad_left - min_pad_left) / (InnerTileCols - overlap_cols); - return tilefn_left_padded[index]; - } - else if (!pad_left && pad_right) - { - // Right padding only - return tilefn_right_padded[pad_right - 1]; - } - else - { - // Combination of paddings, return an unspecialised method - return tilefn_generic; - } - } -} - - diff --git a/arm_compute/core/NEON/kernels/convolution/winograd/transforms/kernel.hpp b/arm_compute/core/NEON/kernels/convolution/winograd/transforms/kernel.hpp deleted file mode 100644 index bad3ef2249..0000000000 --- a/arm_compute/core/NEON/kernels/convolution/winograd/transforms/kernel.hpp +++ /dev/null @@ -1,77 +0,0 @@ -/* - * Copyright (c) 2017 ARM Limited. - * - * SPDX-License-Identifier: MIT - * - * Permission is hereby granted, free of charge, to any person obtaining a copy - * of this software and associated documentation files (the "Software"), to - * deal in the Software without restriction, including without limitation the - * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or - * sell copies of the Software, and to permit persons to whom the Software is - * furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in all - * copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE - * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER - * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, - * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. - */ - -#include "arm_compute/core/NEON/kernels/convolution/winograd/winograd_gemm.hpp" -using namespace winograd; - - -template -template -WinogradGEMM::WeightsTransform::WeightsTransform( - const T* const input, - T* const output, - const int matrix_stride, /** Stride across matrices in the output. */ - const int matrix_row_stride, /** Stride across rows of the matrix. */ - const int n_output_channels, - const int n_input_channels -) : inptr(input), outptr(output), - matrix_stride(matrix_stride), matrix_row_stride(matrix_row_stride), - n_output_channels(n_output_channels), n_input_channels(n_input_channels) -{ -} - - -template -template -unsigned int WinogradGEMM::WeightsTransform::get_window() const -{ - // TODO When the weights transform supports multithreading, return the number - // of output channels. For now we return 1 to indicate that the weights must - // be transformed as a single block. - // return n_output_channels; - return 1; -} - - -template -template -void WinogradGEMM::WeightsTransform::run( - const unsigned int start, const unsigned int stop -) -{ - // TODO When the weights transform supports multithreading call execute for a - // portion of the output channels. - (void) start; - (void) stop; - - // For now, just do all of the work. - execute( - n_output_channels, - n_input_channels, - inptr, - outptr, - matrix_stride, - matrix_row_stride - ); -} diff --git a/arm_compute/core/NEON/kernels/convolution/winograd/transforms/output.hpp b/arm_compute/core/NEON/kernels/convolution/winograd/transforms/output.hpp deleted file mode 100644 index 77cd9de513..0000000000 --- a/arm_compute/core/NEON/kernels/convolution/winograd/transforms/output.hpp +++ /dev/null @@ -1,278 +0,0 @@ -/* - * Copyright (c) 2017 ARM Limited. - * - * SPDX-License-Identifier: MIT - * - * Permission is hereby granted, free of charge, to any person obtaining a copy - * of this software and associated documentation files (the "Software"), to - * deal in the Software without restriction, including without limitation the - * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or - * sell copies of the Software, and to permit persons to whom the Software is - * furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in all - * copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE - * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER - * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, - * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. - */ - -#pragma once -#include "../winograd_gemm.hpp" - -namespace winograd -{ -/***************************************************************************/ - /* Instance-less API */ - template - void OutputTransformImpl::execute( - const int n_batches, - const int output_batch_stride, - const int n_rows, - const int output_row_stride, - const int n_cols, - const int output_col_stride, - const int n_channels, - const T* const matrix_base, - const int matrix_stride, - const int matrix_row_stride, - const T* const biases, - T* const output - ) - { - // Compute the number of tiles and hence the padding required on the bottom - // and right of the image. - const int tile_M = iceildiv(n_rows, OutputTileRows); - const int tile_N = iceildiv(n_cols, OutputTileCols); - const int pad_bottom = OutputTileRows*tile_M - n_rows; - const int pad_right = OutputTileCols*tile_N - n_cols; - - const int matrix_tile_row_stride = tile_N * matrix_row_stride; - const int matrix_batch_stride = tile_M * matrix_tile_row_stride; - - // Perform the output transformation for each batch - for (int batch = 0; batch < n_batches; batch++) - { - // Get batch offset for input and outputs. - const T* const matrix_batch = matrix_base + batch*matrix_batch_stride; - T* const outptr_batch = output + batch*output_batch_stride; - - // Perform the output transformation for each row of the output tensor. - for (int tile_i = 0; tile_i < tile_M; tile_i++) - { - // Compute properties of this row of output tiles - const int row_pad_bottom = (tile_i < tile_M - 1) ? 0: pad_bottom; - const T* const matrix_tile_row = matrix_batch + tile_i * matrix_tile_row_stride; - T* const outptr_row = outptr_batch + OutputTileRows*tile_i*output_row_stride; - - // Process the row - process_tile_row( - tile_N, n_channels, matrix_tile_row, matrix_stride, - matrix_row_stride, biases, - outptr_row, output_row_stride, output_col_stride, row_pad_bottom, - pad_right - ); - } - } - } - -template - void OutputTransformImpl::execute( - const int n_batches, - const int output_batch_stride, - const int n_rows, - const int output_row_stride, - const int n_cols, - const int output_col_stride, - const int n_channels, - const T* const matrix_base, - const int matrix_stride, - const int matrix_row_stride, - const T* const biases, - T* const output - ) - { - // If an Nx1 kernel then transpose and redirect to the 1xN implementation. - OutputTransformImpl<1, KernelRows, 1, InnerTileRows, T>::execute( - n_batches, - output_batch_stride, - n_cols, output_col_stride, - n_rows, output_row_stride, - n_channels, - matrix_base, matrix_stride, matrix_row_stride, - biases, output - ); - } - - template - void OutputTransformImpl::process_tile_row( - const int tile_N, - const int n_channels, - const T* const matrix_base, - const int matrix_stride, - const int matrix_row_stride, - const T* const biases, - T* const output, - const int output_row_stride, - const int output_col_stride, - const int row_pad_bottom, - const int row_pad_right - ) - { - // Loop over columns of tiles - for (int tile_j = 0; tile_j < tile_N; tile_j++) - { - // Properties of this tile - const int tile_pad_right = (tile_j < tile_N - 1) ? 0 : row_pad_right; - const T* const matrix_row = matrix_base + tile_j * matrix_row_stride; - T* const outptr = output + OutputTileCols *tile_j*output_col_stride; - - // Perform the output transformation - const typename Tiles::TileFn tilefn = Tiles::get_tile_specialization(row_pad_bottom, tile_pad_right); - tilefn( - n_channels, matrix_row, matrix_stride, biases, - outptr, output_row_stride, output_col_stride, - row_pad_bottom, tile_pad_right - ); - } - } - -/***************************************************************************/ - template - OutputTransform::OutputTransform( - const T* const matrix_base, - const int matrix_stride, - const int matrix_row_stride, - const T* const biases, - T* const output, - const int n_batches, - const int n_rows, - const int n_cols, - const int n_channels, - const int out_batch_stride, - const int out_row_stride, - const int out_col_stride - ) : _matrix_base(matrix_base), _biases(biases), - _matrix_stride(matrix_stride), _matrix_row_stride(matrix_row_stride), - _outptr(output), _n_batches(n_batches), _n_rows(n_rows), _n_cols(n_cols), - _n_channels(n_channels), _tile_M(iceildiv(n_rows, OutputTileRows)), - _tile_N(iceildiv(n_cols, OutputTileCols)), - _out_col_stride(out_col_stride ? out_col_stride : n_channels), - _out_row_stride(out_row_stride ? out_row_stride : n_cols * _out_col_stride), - _out_batch_stride(out_batch_stride ? out_batch_stride : n_rows * _out_row_stride) - { - } - - template - unsigned int OutputTransform::get_window() const - { - // The final window includes the tail, all other windows will be a multiple - // of the window block in size. - return iceildiv(_n_channels, WINDOW_BLOCK); - } - -template - void OutputTransform::run( - const unsigned int start, const unsigned int stop - ) - { - if (start >= get_window()) - { - return; - } - - // Determine the window of work to perform - const unsigned int start_channel = start * WINDOW_BLOCK; - const unsigned int stop_channel = std::min( - stop * WINDOW_BLOCK, _n_channels - ); - const unsigned int n_channels = stop_channel - start_channel; - - execute( - _n_batches, - _out_batch_stride, - _n_rows, - _out_row_stride, - _n_cols, - _out_col_stride, - n_channels, - _matrix_base + start_channel, - _matrix_stride, - _matrix_row_stride, - (_biases != nullptr) ? _biases + start_channel : nullptr, - _outptr + start_channel - ); - } - - template - void OutputTransform::execute( - const int n_batches, - const int out_batch_stride, - const int n_rows, - const int out_row_stride, - const int n_cols, - const int out_col_stride, - const int n_channels, - const T* const matrix_base, - const int matrix_stride, - const int matrix_row_stride, - const T* const biases, - T* const output - ) - { - Transform::execute( - n_batches, out_batch_stride, - n_rows, out_row_stride, - n_cols, out_col_stride, n_channels, - matrix_base, matrix_stride, matrix_row_stride, - biases, output - ); - } - - template - typename OutputTransformImplTiles<1, KernelCols, 1, InnerTileCols, T>::TileFn - OutputTransformImplTiles<1, KernelCols, 1, InnerTileCols, T>:: - get_tile_specialization(const int pad_bottom, const int pad_right) - { - (void) pad_bottom; - - if (!pad_right) - { - // No padding, return unpadded specialisation - return tilefn_unpadded; - } - else - { - return tilefn_right_padded[pad_right - 1]; - } - } - - template - typename OutputTransformImplTiles::TileFn - OutputTransformImplTiles:: - get_tile_specialization(const int pad_bottom, const int pad_right) - { - if (!(pad_bottom || pad_right)) - { - // No padding, return unpadded specialisation - return tilefn_unpadded; - } - else if (pad_bottom && !pad_right) - { - return tilefn_bottom_padded[pad_bottom - 1]; - } - else if (!pad_bottom && pad_right) - { - return tilefn_right_padded[pad_right - 1]; - } - else - { - return tilefn_generic; - } - } -} // namespace winograd diff --git a/arm_compute/core/NEON/kernels/convolution/winograd/winograd.hpp b/arm_compute/core/NEON/kernels/convolution/winograd/winograd.hpp new file mode 100644 index 0000000000..183c9c1061 --- /dev/null +++ b/arm_compute/core/NEON/kernels/convolution/winograd/winograd.hpp @@ -0,0 +1,610 @@ +/* + * Copyright (c) 2019 ARM Limited. + * + * SPDX-License-Identifier: MIT + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#pragma once + +#include "convolution.hpp" +#include "tensor.hpp" +#include "utils.hpp" + +namespace winograd +{ + +class ITransform +{ + public: + virtual ~ITransform() = default; + + /** + * Get the working space required to perform the transformation. + * + * Note, the working space is only required when performing the + * transformation - hence it can be reused whenever the transformation is + * not running. + * + * @param nthreads The greatest number of threads that will be used to execute the transform. + * @return Size of working space required in bytes. + */ + virtual size_t get_working_space_size(unsigned int nthreads=1) const = 0; + + /** + * Set the working space to be used by the transformation. + * + * Note, the working space is only required when performing the + * transformation - hence it can be reused whenever the transformation is + * not running. + * + * @param Pointer to the working space. + */ + virtual void set_working_space(void *buffer) = 0; + + /** + * Get the window of work a given operator can perform. + */ + virtual unsigned int get_window() const = 0; + + /** + * Perform work upon a window of the transform. + */ + virtual void run(unsigned int start, unsigned int stop, unsigned int threadid=0) = 0; +}; + +class IInputTransform : public ITransform +{ + public: + virtual ~IInputTransform() = default; + + /** + * Set the pointer to the (NHWC-ordered) tensor to be transformed. + */ + virtual void set_input_tensor(const void *input) = 0; + + /** + * Set the pointer to the (NHWC-ordered) tensor to be transformed. + * @param col_stride Stride between columns of the tensor, measured in elements (not bytes). + */ + virtual void set_input_tensor(const void *input, int col_stride) = 0; + + /** + * Set the pointer to the (NHWC-ordered) tensor to be transformed. + * @param row_stride Stride between rows of the tensor, measured in elements (not bytes). + * @param col_stride Stride between columns of the tensor, measured in elements (not bytes). + */ + virtual void set_input_tensor(const void *input, int row_stride, int col_stride) = 0; + + /** + * Set the pointer to the (NHWC-ordered) tensor to be transformed. + * @param batch_stride Stride between batches of the tensor, measured in elements (not bytes). + * @param row_stride Stride between rows of the tensor, measured in elements (not bytes). + * @param col_stride Stride between columns of the tensor, measured in elements (not bytes). + */ + virtual void set_input_tensor(const void *input, int batch_stride, int row_stride, int col_stride) = 0; + + /** + * Set pointers to the matrices written by the transform. + * @param matrices Pointer to the start of the first matrix representing the transformed input. + * @param inter_matrix_stride Stride (in elements) between matrices. + * @param matrix_row_stride Stride (in elements) between the rows within a single matrix. + */ + virtual void set_output_matrices(void *matrices, int inter_matrix_stride, int matrix_row_stride) = 0; +}; + +class IOutputTransform : public ITransform +{ + public: + virtual ~IOutputTransform() = default; + + /** + * Set pointers to the matrices written by the transform. + * @param matrices Pointer to the start of the first matrix representing the input to the transform. + * @param inter_matrix_stride Stride (in elements) between matrices. + * @param matrix_row_stride Stride (in elements) between the rows within a single matrix. + */ + virtual void set_input_matrices(const void *matrices, int inter_matrix_stride, int matrix_row_stride) = 0; + + /** + * Set pointer to the bias tensor (can be ignored or called with nullptr for no bias. + */ + virtual void set_bias(const void *bias=nullptr) = 0; + + /** + * Set pointer to the output tensor produced by the transform. + */ + virtual void set_output_tensor(void *output) = 0; + + /** + * Set pointer to the output tensor produced by the transform. + * @param col_stride Stride between columns of the tensor, measured in elements (not bytes). + */ + virtual void set_output_tensor(void *output, int col_stride) = 0; + + /** + * Set pointer to the output tensor produced by the transform. + * @param row_stride Stride between rows of the tensor, measured in elements (not bytes). + * @param col_stride Stride between columns of the tensor, measured in elements (not bytes). + */ + virtual void set_output_tensor(void *output, int row_stride, int col_stride) = 0; + + /** + * Set pointer to the output tensor produced by the transform. + * @param batch_stride Stride between batches of the tensor, measured in elements (not bytes). + * @param row_stride Stride between rows of the tensor, measured in elements (not bytes). + * @param col_stride Stride between columns of the tensor, measured in elements (not bytes). + */ + virtual void set_output_tensor(void *output, int batch_stride, int row_stride, int col_stride) = 0; +}; + +class IWeightTransform : public ITransform +{ + public: + virtual ~IWeightTransform() = default; + + /** Set pointer to the weight tensor read by the transform. */ + virtual void set_weight_tensor(const void *weights) = 0; + + /** + * Set pointers to the matrices written by the transform. + * @param matrices Pointer to the start of the first matrix representing the transformed input. + * @param inter_matrix_stride Stride (in elements) between matrices. + * @param matrix_row_stride Stride (in elements) between the rows within a single matrix. + */ + virtual void set_output_matrices(void *matrices, int inter_matrix_stride, int matrix_row_stride) = 0; +}; + +enum class WinogradRoots +{ + Integers, +}; + +template +class InputTransform : public IInputTransform +{ + public: + /** Create an InputTransform operator fixed on a given problem and set of + * pointers. + */ + InputTransform( + int kernel_rows, /**< Number of rows in the kernel */ + int kernel_cols, /**< Number of columns in the kernel */ + int n_batches, /**< Number of batches in input tensor. */ + int n_rows, /**< Number of rows in input tensor. */ + int n_cols, /**< Number of columns in input tensor. */ + int n_channels, /**< Number of channels in input tensor. */ + int padding_top, /**< Padding to apply to the top of the image. */ + int padding_left, /**< Padding to apply to the left of the image. */ + int padding_bottom, /**< Padding to apply to the bottom of the image. */ + int padding_right /**< Padding to apply to the right of the image. */ + ); + + InputTransform(InputTransform&) = delete; + InputTransform operator=(InputTransform&) = delete; + + /** Set pointers to the input tensor read by the transform. */ + void set_input_tensor(const void *input) override; + void set_input_tensor(const void *input, int col_stride) override; + void set_input_tensor(const void *input, int row_stride, int col_stride) override; + void set_input_tensor(const void *input, int batch_stride, int row_stride, int col_stride) override; + + /** Set pointers to the matrices written by the transform. */ + void set_output_matrices(void *matrices, int iter_matrix_stride, int matrix_row_stride) override; + + /** Get the working space required to perform the transformation. */ + size_t get_working_space_size(unsigned int nthreads=1) const override; + void set_working_space(void *buffer) override; + + /** Get the window of work a given operator can perform. */ + unsigned int get_window() const override; + static constexpr unsigned int WINDOW_BLOCK = 16; // Base size of window + + /** Perform work upon a window of the input. */ + void run(unsigned int start, unsigned int stop, unsigned int threadid=0) override; + + protected: + const int _n_batches, _n_rows, _n_cols, _n_channels; + + private: + void transform_unpadded_tile( + unsigned int threadid, + int n_channels, + TOut *outptr, + const TIn *inptr + ); + + void transform_padded_tile( + unsigned int threadid, + int n_channels, + TOut *outptr, + const TIn *inptr, + int padding_top, + int padding_left, + int padding_bottom, + int padding_right + ); + + /* Tile implementation */ + static void transform_tile( + int n_channels, /** @param[in] Number of channels in the tensor. */ + const TIn* inptr_base, /** @param[in] Pointer to the base of the input tile. */ + int input_row_stride, /** @param[in] Stride between rows of the input tensor. */ + int input_col_stride, /** @param[in] Stride between columns of the input tensor. */ + TOut* mptr_base, /** @param[out] Base pointer to transformed input matrices. */ + int matrix_stride /** @param[in] Stride between matrices in the input space. */ + ); + + /** Get the working space for a thread. */ + void * get_working_space(unsigned int threadid) const; + + const TIn* _inptr; + TOut* _outptr; + + const int _overlap_rows, _overlap_cols; + const int _padding_top, _padding_left, _padding_bottom, _padding_right; + const int _tiles_M, _tiles_N; + int _matrix_stride, _matrix_row_stride, _matrix_batch_stride; + int _in_col_stride, _in_row_stride, _in_batch_stride; + + const int _working_space_col_stride, _working_space_row_stride; + TIn *_working_space; +}; + +template +class InputTransform : + public InputTransform<1, InnerTileRows, TIn, TOut, Roots> +{ + using Base = InputTransform<1, InnerTileRows, TIn, TOut, Roots>; + + public: + InputTransform( + int kernel_rows, /**< Number of rows in the kernel. */ + int kernel_cols, /**< Number of columns in the kernel. */ + int n_batches, /**< Number of batches in input tensor. */ + int n_rows, /**< Number of rows in input tensor. */ + int n_cols, /**< Number of columns in input tensor. */ + int n_channels, /**< Number of channels in input tensor. */ + int padding_top, /**< Padding to apply to the top of the image. */ + int padding_left, /**< Padding to apply to the left of the image. */ + int padding_bottom, /**< Padding to apply to the bottom of the image. */ + int padding_right /**< Padding to apply to the right of the image. */ + ); + + /** Set pointers to the input tensor read by the transform. */ + void set_input_tensor(const void *input) override; + void set_input_tensor(const void *input, int col_stride) override; + void set_input_tensor(const void *input, int row_stride, int col_stride) override; + void set_input_tensor(const void *input, int batch_stride, int row_stride, int col_stride) override; +}; + +template < + int KernelRows, int KernelCols, + int InnerTileRows, int InnerTileCols, + typename TIn, typename TOut, + WinogradRoots Roots +> +class OutputTransform : public IOutputTransform +{ + public: + OutputTransform( + int n_batches, /**< Number of batches in output tensor. */ + int n_rows, /**< Number of rows in output tensor. */ + int n_cols, /**< Number of columns in output tensor. */ + int n_channels /**< Number of channels in output tensor. */ + ); + + OutputTransform(OutputTransform&) = delete; + OutputTransform operator=(OutputTransform&) = delete; + + /** Set pointers to the matrices read by the transform. */ + void set_input_matrices(const void *matrices, int iter_matrix_stride, int matrix_row_stride) override; + + /** Set pointer to the bias tensor (can be ignored or called with nullptr for no bias */ + void set_bias(const void *bias=nullptr) override; + + /** Set pointers to the output tensor written by the transform. */ + void set_output_tensor(void *output) override; + void set_output_tensor(void *output, int col_stride) override; + void set_output_tensor(void *output, int row_stride, int col_stride) override; + void set_output_tensor(void *output, int batch_stride, int row_stride, int col_stride) override; + + /** Get the working space required to perform the transformation. */ + size_t get_working_space_size(unsigned int nthreads=1) const override; + void set_working_space(void *buffer) override; + + /** Get the window of work a given operator can perform. */ + unsigned int get_window() const override; + static constexpr unsigned int WINDOW_BLOCK = 16; // Base size of window + + /** Perform work upon a window of the input. */ + void run(unsigned int start, unsigned int stop, unsigned int threadid=0) override; + + protected: + static constexpr int inner_tile_rows = InnerTileRows; + static constexpr int inner_tile_cols = InnerTileCols; + static constexpr int output_tile_rows = InnerTileRows - KernelRows + 1; + static constexpr int output_tile_cols = InnerTileCols - KernelCols + 1; + + const int _n_batches, _n_rows, _n_cols, _n_channels; + + private: + void transform_uncropped_tile( + unsigned int threadid, + int n_channels, + TOut *outptr, + const TIn *inptr, + const TOut *biases + ); + + void transform_cropped_tile( + unsigned int threadid, + int n_channels, + TOut *outptr, + const TIn *inptr, + const TOut *biases, + int pad_bottom, + int pad_right + ); + + /** Implementation of the tile transformation method. */ + static void transform_tile( + int n_channels, + const TIn* matrix_base, + int matrix_stride, + const TOut* biases, + TOut* output, + int output_row_stride, + int output_col_stride + ); + + /** Get the working space for a thread. */ + void * get_working_space(unsigned int threadid) const; + + const TIn* _matrix_base; + const TOut* _biases; + int _matrix_stride, _matrix_row_stride, _matrix_batch_stride; + TOut* _outptr; + const int _tiles_M, _tiles_N; + int _out_col_stride, _out_row_stride, _out_batch_stride; + + const int _working_space_col_stride, _working_space_row_stride; + TOut *_working_space; +}; + +template < + int KernelRows, + int InnerTileRows, + typename TIn, typename TOut, + WinogradRoots Roots +> +class OutputTransform : + public OutputTransform<1, KernelRows, 1, InnerTileRows, TIn, TOut, Roots> +{ + using Base = OutputTransform<1, KernelRows, 1, InnerTileRows, TIn, TOut, Roots>; + + public: + OutputTransform( + int n_batches, /**< Number of batches in output tensor. */ + int n_rows, /**< Number of rows in output tensor. */ + int n_cols, /**< Number of columns in output tensor. */ + int n_channels /**< Number of channels in output tensor. */ + ); + + /** Set pointers to the output tensor written by the transform. */ + void set_output_tensor(void *output) override; + void set_output_tensor(void *output, int col_stride) override; + void set_output_tensor(void *output, int row_stride, int col_stride) override; + void set_output_tensor(void *output, int batch_stride, int row_stride, int col_stride) override; +}; + +template < + int KernelRows, int KernelCols, + int InnerTileRows, int InnerTileCols, + typename TIn, typename TOut, + WinogradRoots Roots +> +class WeightTransform : public IWeightTransform +{ + public: + WeightTransform( + int n_output_channels, /**< Number of output channels in the kernel. */ + int n_input_channels /**< Number of input channels in the kernel. */ + ); + + WeightTransform(WeightTransform&) = delete; + WeightTransform operator=(WeightTransform&) = delete; + + /** Set pointer to the weight tensor read by the transform. */ + void set_weight_tensor(const void *weights) override; + + /** Set pointer to the matrices written by the transform. */ + void set_output_matrices(void *matrices, int inter_matrix_stride, int matrix_row_stride) override; + + /** Get the working space required to perform the transformation. */ + size_t get_working_space_size(unsigned int nthreads=1) const override; + void set_working_space(void *buffer) override; + + /** Get the window of work a given operator can perform. */ + unsigned int get_window() const override; + static constexpr unsigned int WINDOW_BLOCK = 16; // Base size of window + + /** Perform work upon a window of the input. */ + void run(unsigned int start, unsigned int stop, unsigned int threadid=0) override; + + protected: + static const int kernel_rows = KernelRows; + static const int kernel_cols = KernelCols; + static const int inner_tile_rows = InnerTileRows; + static const int inner_tile_cols = InnerTileCols; + + private: + /** Apply the transform to a tensor. */ + static void execute( + int n_output_channels, + int n_input_channels, + const TIn* input, + TOut* output, + int matrix_stride, + int matrix_row_stride + ); + + const int _n_output_channels, _n_input_channels; + TOut *_matrices; + int _matrix_stride, _matrix_row_stride; + const TIn *_weights; +}; + +template +class WeightTransform : + public WeightTransform<1, KernelRows, 1, InnerTileRows, TIn, TOut, Roots> +{ + public: + using WeightTransform<1, KernelRows, 1, InnerTileRows, TIn, TOut, Roots>::WeightTransform; +}; + +template +class WinogradGEMM +{ + public: + // Information about the specific Winograd instance + static constexpr int output_tile_rows = OutputTileRows; + static constexpr int output_tile_cols = OutputTileCols; + static constexpr int kernel_rows = KernelRows; + static constexpr int kernel_cols = KernelCols; + static constexpr int inner_tile_rows = output_tile_rows + kernel_rows - 1; + static constexpr int inner_tile_cols = output_tile_cols + kernel_cols - 1; + static constexpr int N_GEMMS = inner_tile_rows * inner_tile_cols; + + /** Transform weights from the spatial to the Winograd domain. */ + template + using WeightsTransform = WeightTransform< + KernelRows, KernelCols, inner_tile_rows, inner_tile_cols, + TIn, TOut, Roots + >; + + /** Transform input feature maps from the spatial to the Winograd domain. + */ + template + using InputTransform = InputTransform< + inner_tile_rows, inner_tile_cols, TIn, TOut, Roots + >; + + /** Transform output feature maps from the Winograd to the spatial domain. + */ + template + using OutputTransform = OutputTransform< + KernelRows, KernelCols, inner_tile_rows, inner_tile_cols, + TIn, TOut, Roots + >; + + /** Perform a convolution. + */ + template + class Convolution + { + public: + // Information about the typed Winograd instance + typedef TOut OutputType; + typedef TOutGEMM GemmOutputType; + typedef TInGEMM GemmInputType; + typedef TIn InputType; + + /** Get the output shape of a convolution. */ + static Tensor4DShape get_output_shape( + const KernelShape &kernel_shape, + const Tensor4DShape &in_shape, + const PaddingType padding + ); + + /* Get the memory required to transform the kernel. + */ + static size_t get_kernel_transform_working_size(const KernelShape &shape); + + /** Get the memory required to store the kernel transformed into the + * Winograd domain. + */ + static size_t get_kernel_storage_size(const KernelShape &shape); + + /** Get the memory required to store the input tensor transformed into + * the Winograd domain. + */ + static size_t get_input_storage_size( + const KernelShape &kernel_shape, + const Tensor4DShape &input_shape, + const PaddingType padding_type + ); + + /** Get the memory required to store the output tensor in the Winograd + * domain. + */ + static size_t get_output_storage_size( + const KernelShape &kernel_shape, + const Tensor4DShape &input_shape, + const PaddingType padding_type + ); + + /** Get the memory required to apply a Winograd operator to some input. + */ + static size_t get_working_space_size( + const KernelShape &kernel_shape, + const Tensor4DShape &input_shape, + const PaddingType padding_type + ); + + /* Get the memory required by a single "input" matrix. + */ + static size_t get_input_matrix_size( + const KernelShape &kernel_shape, + const Tensor4DShape &input_shape, + const PaddingType padding_type + ); + + static int get_input_matrix_stride( + const KernelShape &kernel_shape, + const Tensor4DShape &input_shape, + const PaddingType padding_type + ); + + /* Get the memory required by a single "output" matrix. + */ + static size_t get_output_matrix_size( + const KernelShape &kernel_shape, + const Tensor4DShape &input_shape, + const PaddingType padding_type + ); + + static int get_output_matrix_stride( + const KernelShape &kernel_shape, + const Tensor4DShape &input_shape, + const PaddingType padding_type + ); + + /* Get the memory required by a single "kernel" matrix. + */ + static size_t get_kernel_matrix_size(const KernelShape &shape); + static int get_kernel_matrix_stride(const KernelShape &shape); + + static constexpr int M_BLOCK = 4; /** Size of block used by GEMM. */ + static constexpr int N_BLOCK = 16; /** Size of block used by GEMM. */ + }; +}; + +} // namespace winograd diff --git a/arm_compute/core/NEON/kernels/convolution/winograd/winograd_gemm.hpp b/arm_compute/core/NEON/kernels/convolution/winograd/winograd_gemm.hpp deleted file mode 100644 index 71b5fd516f..0000000000 --- a/arm_compute/core/NEON/kernels/convolution/winograd/winograd_gemm.hpp +++ /dev/null @@ -1,226 +0,0 @@ -/* - * Copyright (c) 2017 ARM Limited. - * - * SPDX-License-Identifier: MIT - * - * Permission is hereby granted, free of charge, to any person obtaining a copy - * of this software and associated documentation files (the "Software"), to - * deal in the Software without restriction, including without limitation the - * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or - * sell copies of the Software, and to permit persons to whom the Software is - * furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in all - * copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE - * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER - * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, - * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. - */ - -#pragma once - -#include "arm_compute/core/NEON/kernels/convolution/common/alloc.hpp" -#include "arm_compute/core/NEON/kernels/convolution/common/convolution.hpp" -#include "gemm.hpp" -#include "arm_compute/core/NEON/kernels/convolution/common/shims.hpp" -#include "arm_compute/core/NEON/kernels/convolution/common/tensor.hpp" -#include "arm_compute/core/NEON/kernels/convolution/common/utils.hpp" -#include "winograd_input_transform.hpp" -#include "winograd_output_transform.hpp" - -#include -#include -#include - -// Generic Winograd implementation using GEMM -namespace winograd -{ - -template -class WinogradGEMM -{ - public: - // Information about the specific Winograd instance - static constexpr int output_tile_rows = OutputTileRows; - static constexpr int output_tile_cols = OutputTileCols; - static constexpr int kernel_rows = KernelRows; - static constexpr int kernel_cols = KernelCols; - static constexpr int inner_tile_rows = output_tile_rows + kernel_rows - 1; - static constexpr int inner_tile_cols = output_tile_cols + kernel_cols - 1; - static constexpr int N_GEMMS = inner_tile_rows * inner_tile_cols; - - /** Transform weights from the spatial to the Winograd domain. */ - template - struct WeightsTransform - { - /** Get the bytes read during the transform. */ - static inline size_t bytes_read(const KernelShape &shape) - { - return shape.size() * sizeof(T); - } - - /** Get the bytes written during the transform. */ - static inline size_t bytes_written(const KernelShape &shape) - { - const int inner_tile_size = inner_tile_rows * inner_tile_cols; - return (inner_tile_size * shape.n_input_channels * - shape.n_output_channels * sizeof(T)); - } - - /** Get the count of operations performed by the transform. */ - static int ops_performed(const KernelShape &shape); - - /** Apply the transform to a tensor. */ - static void execute( - const int n_output_channels, - const int n_input_channels, - const T* const input, - T* const output, - const int matrix_stride, - const int matrix_row_stride - ); - - /** Create a WeightsTransform operator fixed on a given problem and set - * of pointers. - */ - WeightsTransform( - const T* const input, - T* const output, - const int matrix_stride, /** Stride across matrices in the output. */ - const int matrix_row_stride, /** Stride across rows of the matrix. */ - const int n_output_channels, /** Number of filters. */ - const int n_input_channels /** Number of channels in each filter. */ - ); - - /** Get the window of work a given operator can perform. */ - unsigned int get_window() const; - - /** Perform work upon a window of the input. */ - void run(const unsigned int start, const unsigned int stop); - - private: - const T* const inptr; /** Fixed pointer to input data. */ - T* const outptr; /** Fixed pointer to output memory. */ - const int matrix_stride; /** Stride between output matrices. */ - const int matrix_row_stride; /** Stride within output matrices. */ - const int n_output_channels; /** Number of filters. */ - const int n_input_channels; /** Number of channels in each filter. */ - }; - - /** Transform input feature maps from the spatial to the Winograd domain. - */ - template - using InputTransform = InputTransform< - KernelRows, KernelCols, - (OutputTileRows + KernelRows - 1), - (OutputTileCols + KernelCols - 1), - T - >; - - /** Transform output feature maps from the Winograd to the spatial domain. - */ - template - using OutputTransform = OutputTransform< - KernelRows, KernelCols, - (OutputTileRows + KernelRows - 1), - (OutputTileCols + KernelCols - 1), - T - >; - - - /** Perform a convolution. - */ - template - class Convolution - { - public: - // Information about the typed Winograd instance - typedef TOut OutputType; - typedef TIn InputType; - - /** Get the output shape of a convolution. */ - static Tensor4DShape get_output_shape( - const KernelShape &kernel_shape, - const Tensor4DShape &in_shape, - const PaddingType padding - ); - - /* Get the memory required to transform the kernel. - */ - static size_t get_kernel_transform_working_size(const KernelShape &shape); - - /** Get the memory required to store the kernel transformed into the - * Winograd domain. - */ - static size_t get_kernel_storage_size(const KernelShape &shape); - - /** Get the memory required to store the input tensor transformed into - * the Winograd domain. - */ - static size_t get_input_storage_size( - const KernelShape &kernel_shape, - const Tensor4DShape &input_shape, - const PaddingType padding_type - ); - - /** Get the memory required to store the output tensor in the Winograd - * domain. - */ - static size_t get_output_storage_size( - const KernelShape &kernel_shape, - const Tensor4DShape &input_shape, - const PaddingType padding_type - ); - - /** Get the memory required to apply a Winograd operator to some input. - */ - static size_t get_working_space_size( - const KernelShape &kernel_shape, - const Tensor4DShape &input_shape, - const PaddingType padding_type - ); - - /* Get the memory required by a single "input" matrix. - */ - static size_t get_input_matrix_size( - const KernelShape &kernel_shape, - const Tensor4DShape &input_shape, - const PaddingType padding_type - ); - - static int get_input_matrix_stride( - const KernelShape &kernel_shape, - const Tensor4DShape &input_shape, - const PaddingType padding_type - ); - - /* Get the memory required by a single "output" matrix. - */ - static size_t get_output_matrix_size( - const KernelShape &kernel_shape, - const Tensor4DShape &input_shape, - const PaddingType padding_type - ); - - static int get_output_matrix_stride( - const KernelShape &kernel_shape, - const Tensor4DShape &input_shape, - const PaddingType padding_type - ); - - /* Get the memory required by a single "kernel" matrix. - */ - static size_t get_kernel_matrix_size(const KernelShape &shape); - static int get_kernel_matrix_stride(const KernelShape &shape); - - static constexpr int M_BLOCK = 4; /** Size of block used by GEMM. */ - static constexpr int N_BLOCK = 16; /** Size of block used by GEMM. */ - }; -}; - -} // namespace winograd diff --git a/arm_compute/core/NEON/kernels/convolution/winograd/winograd_input_transform.hpp b/arm_compute/core/NEON/kernels/convolution/winograd/winograd_input_transform.hpp deleted file mode 100644 index 995554d7f2..0000000000 --- a/arm_compute/core/NEON/kernels/convolution/winograd/winograd_input_transform.hpp +++ /dev/null @@ -1,271 +0,0 @@ -/* - * Copyright (c) 2018 ARM Limited. - * - * SPDX-License-Identifier: MIT - * - * Permission is hereby granted, free of charge, to any person obtaining a copy - * of this software and associated documentation files (the "Software"), to - * deal in the Software without restriction, including without limitation the - * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or - * sell copies of the Software, and to permit persons to whom the Software is - * furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in all - * copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE - * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER - * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, - * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. - */ - -#pragma once - -namespace winograd -{ - -namespace -{ - -template -class InputTransformImplTiles -{ - public: - /** Method to transform a tile of the input tensor into the Winograd domain. */ - typedef void (*TileFn)( - const int n_channels, /** @param[in] Number of channels in the tensor. */ - const T* const inptr_base, /** @param[in] Pointer to the base of the input tile. */ - const int input_row_stride, /** @param[in] Stride between rows of the input tensor. */ - const int input_col_stride, /** @param[in] Stride between columns of the input tensor. */ - T* const mptr_base, /** @param[out] Base pointer to transformed input matrices. */ - const int matrix_stride, /** @param[in] Stride between matrices in the input space. */ - const int _pad_top, /** @param[in] Top padding for unspecialised tiles. */ - const int _pad_left, /** @param[in] Left padding for unspecialised tiles. */ - const int _pad_bottom, /** @param[in] Bottom padding for unspecialised tiles. */ - const int _pad_right /** @param[in] Right padding for unspecialised tiles. */ - ); - - static TileFn get_tile_specialization( - const int pad_top, - const int pad_left, - const int pad_bottom, - const int pad_right - ); - - // Tile overlaps - static constexpr int overlap_rows = KernelRows - 1; - static constexpr int overlap_cols = KernelCols - 1; - - private: - - // Maximum padding and number of distinct paddings - static constexpr int max_pad_top = KernelRows / 2; - static constexpr int min_pad_top = KernelRows % (InnerTileRows - overlap_rows); - static constexpr int n_pad_top = iceildiv(max_pad_top, InnerTileRows - overlap_rows); - - static constexpr int max_pad_left = KernelCols / 2; - static constexpr int min_pad_left = KernelCols % (InnerTileCols - overlap_cols); - static constexpr int n_pad_left = iceildiv(max_pad_left, InnerTileCols - overlap_cols); - - static constexpr int n_pad_bottom = InnerTileRows; - static constexpr int n_pad_right = InnerTileCols; - - // Pointers to methods implementing a generically padded tile and a totally unpadded tile. - static const TileFn tilefn_generic; /** Generic tile processing function. */ - static const TileFn tilefn_unpadded; /** Tile processor for unpadded tiles. */ - - // Arrays of methods covering tiles which are padded only on a single side. - static const TileFn tilefn_top_padded[n_pad_top]; - static const TileFn tilefn_left_padded[n_pad_left]; - static const TileFn tilefn_bottom_padded[n_pad_bottom]; - static const TileFn tilefn_right_padded[n_pad_right]; -}; - - -template < int KernelCols, int InnerTileCols, typename T> -class InputTransformImplTiles<1, KernelCols, 1, InnerTileCols, T> -{ - public: - /** Method to transform a tile of the input tensor into the Winograd domain. */ - typedef void (*TileFn)( - const int n_channels, /** @param[in] Number of channels in the tensor. */ - const T* const inptr_base, /** @param[in] Pointer to the base of the input tile. */ - const int input_row_stride, /** @param[in] Stride between rows of the input tensor. */ - const int input_col_stride, /** @param[in] Stride between columns of the input tensor. */ - T* const mptr_base, /** @param[out] Base pointer to transformed input matrices. */ - const int matrix_stride, /** @param[in] Stride between matrices in the input space. */ - const int _pad_top, /** @param[in] Top padding for unspecialised tiles. */ - const int _pad_left, /** @param[in] Left padding for unspecialised tiles. */ - const int _pad_bottom, /** @param[in] Bottom padding for unspecialised tiles. */ - const int _pad_right /** @param[in] Right padding for unspecialised tiles. */ - ); - - static TileFn get_tile_specialization( - const int pad_top, - const int pad_left, - const int pad_bottom, - const int pad_right - ); - - // Tile overlaps - static constexpr int overlap_rows = 0; - static constexpr int overlap_cols = KernelCols - 1; - - private: - // Maximum padding and number of distinct paddings - static constexpr int max_pad_left = KernelCols / 2; - static constexpr int min_pad_left = KernelCols % (InnerTileCols - overlap_cols); - static constexpr int n_pad_left = iceildiv(max_pad_left, InnerTileCols - overlap_cols); - - static constexpr int n_pad_right = InnerTileCols; - - // Pointers to methods implementing a generically padded tile and a totally unpadded tile. - static const TileFn tilefn_generic; /** Generic tile processing function. */ - static const TileFn tilefn_unpadded; /** Tile processor for unpadded tiles. */ - - // Arrays of methods covering tiles which are padded only on a single side. - static const TileFn tilefn_left_padded[n_pad_left]; - static const TileFn tilefn_right_padded[n_pad_right]; -}; - - - -template -class InputTransformImpl -{ - public: - /** Apply the transform to a tensor. */ - static void execute( - const T* const input, /** Input tensor data */ - const int n_batches, /** Number of batches in input tensor. */ - const int in_batch_stride, /** Stride between batches of the input. */ - const int n_rows, /** Number of rows in input tensor. */ - const int in_row_stride, /** Stride between rows of the input. */ - const int n_cols, /** Number of columns in input tensor. */ - const int in_col_stride, /** Stride between columns of the input. */ - const int n_channels, /** Number of channels in input tensor. */ - const PaddingType padding, /** Padding type. */ - const int tile_M, - const int tile_N, - T* const output, /** Base of output matrices. */ - const int matrix_stride, /** Stride between output matrices. */ - const int matrix_batch_stride, /** Stride between batches within the matrix. */ - const int matrix_row_stride /** Stride within matrices. */ - ); - - private: - static void process_tile_row( - const int tile_N, - int n_channels, - const T* const input_base, - const int input_row_stride, - const int input_col_stride, - T* const matrix_base, - const int matrix_stride, - const int matrix_row_stride, - const int row_pad_top, - const int row_pad_left, - const int row_pad_bottom, - const int n_cols - ); - - using Tiles = InputTransformImplTiles; - - static constexpr int overlap_rows = Tiles::overlap_rows; - static constexpr int overlap_cols = Tiles::overlap_cols; - - - }; - - -template -class InputTransformImpl -{ - public: - /** Apply the transform to a tensor. */ - static void execute( - const T* const input, /** Input tensor data */ - const int n_batches, /** Number of batches in input tensor. */ - const int in_batch_stride, /** Stride between batches of the input. */ - const int n_rows, /** Number of rows in input tensor. */ - const int in_row_stride, /** Stride between rows of the input. */ - const int n_cols, /** Number of columns in input tensor. */ - const int in_col_stride, /** Stride between columns of the input. */ - const int n_channels, /** Number of channels in input tensor. */ - const PaddingType padding, /** Padding type. */ - const int tile_M, - const int tile_N, - T* const output, /** Base of output matrices. */ - const int matrix_stride, /** Stride between output matrices. */ - const int matrix_batch_stride, /** Stride between batches within the matrix. */ - const int matrix_row_stride /** Stride within matrices. */ - ); -}; - -} // namespace (anonymous) - -template -class InputTransform -{ - public: - /***********************************************************************/ - /** Create an InputTransform operator fixed on a given problem and set of - * pointers. - */ - InputTransform( - const T* const input, /** Input tensor data */ - const int n_batches, /** Number of batches in input tensor. */ - const int n_rows, /** Number of rows in input tensor. */ - const int n_cols, /** Number of columns in input tensor. */ - const int n_channels, /** Number of channels in input tensor. */ - const PaddingType padding, /** Padding type. */ - T* const output, /** Base of output matrices. */ - const int matrix_stride, /** Stride between output matrices. */ - const int matrix_row_stride, /** Stride within matrices. */ - const int in_batch_stride=0, /** Stride between input batches. */ - const int in_row_stride=0, /** Stride between input rows. */ - const int in_col_stride=0 /** Stride between input columns. */ - ); - - /** Get the window of work a given operator can perform. */ - unsigned int get_window() const; - static constexpr unsigned int WINDOW_BLOCK = 16; // Base size of window - - /** Perform work upon a window of the input. */ - void run(const unsigned int start, const unsigned int stop); - - /** Apply the transform to a tensor. */ - static void execute( - const T* const input, /** Input tensor data */ - const int n_batches, /** Number of batches in input tensor. */ - const int in_batch_stride, /** Stride between batches of the input. */ - const int n_rows, /** Number of rows in input tensor. */ - const int in_row_stride, /** Stride between rows of the input. */ - const int n_cols, /** Number of columns in input tensor. */ - const int in_col_stride, /** Stride between columns of the input. */ - const int n_channels, /** Number of channels in input tensor. */ - const PaddingType padding, /** Padding type. */ - const int tile_M, - const int tile_N, - T* const output, /** Base of output matrices. */ - const int matrix_stride, /** Stride between output matrices. */ - const int matrix_batch_stride, /** Stride between batches within the matrix. */ - const int matrix_row_stride /** Stride within matrices. */ - ); - - protected: - using Transform = InputTransformImpl; - - /* Member values for instance-based API. */ - const T* const _inptr; - T* const _outptr; - const int _n_batches, _n_rows, _n_cols, _n_channels, _matrix_stride, - _matrix_row_stride, _tiles_M, _tiles_N; - const int _in_col_stride, _in_row_stride, _in_batch_stride; - const PaddingType _padding_type; -}; - -} // namespace winograd diff --git a/arm_compute/core/NEON/kernels/convolution/winograd/winograd_layer.hpp b/arm_compute/core/NEON/kernels/convolution/winograd/winograd_layer.hpp new file mode 100644 index 0000000000..9d418bebb4 --- /dev/null +++ b/arm_compute/core/NEON/kernels/convolution/winograd/winograd_layer.hpp @@ -0,0 +1,211 @@ +/* + * Copyright (c) 2019 ARM Limited. + * + * SPDX-License-Identifier: MIT + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#pragma once + +#include + +#include "arm_gemm_local.hpp" +#include "arm_gemm.hpp" +#include "winograd.hpp" + +namespace winograd +{ + + +class IWinogradConvolutionLayer +{ + public: + virtual ~IWinogradConvolutionLayer() = default; + + virtual unsigned int weight_transform_get_window(void) const = 0; + virtual void weight_transform_run(unsigned int start, unsigned int stop) = 0; + + virtual ITransform& input_transform(void) = 0; // Expose the input transform + virtual ITransform& output_transform(void) = 0; // Expose the output transform + virtual arm_gemm::IGemmCommon *gemm(void) = 0; // Expose the underlying GEMM +}; + +/** Example of how to construct an ACL-like interface. + * + * Use `get_weight_storage_size`, `get_input_storage_size` and + * `get_output_storage_size` to allocate memory for the convolution engine. + * Then create a `WinogradConvolutionLayer`. + * + * Initialise the weights using `weights_transform.run(...)`. + * + * For each inference: + * 1. Transform the inputs to the Winograd domain using `input_transform.run(...)` + * 2. Perform a number of GEMMs using `gemms.run(...)` + * 3. Transform the output to the spatial domain using `output_transform.run(...)` + */ +template +class WinogradConvolutionLayer : public IWinogradConvolutionLayer +{ + private: + static constexpr int InnerTileRows = OutputTileRows + KernelRows - 1; + static constexpr int InnerTileCols = OutputTileCols + KernelCols - 1; + static constexpr int N_GEMMS = InnerTileRows * InnerTileCols; + + const KernelShape _kernel_shape; + const Tensor4DShape _input_shape; + const PaddingType _padding; + const Tensor4DShape _output_shape; + const int _n_output_rows, _n_output_cols; + const int _kernel_matrix_stride, _kernel_matrix_row_stride; + const int _input_matrix_stride, _input_matrix_row_stride; + const int _output_matrix_stride, _output_matrix_row_stride; + const int _tile_rows, _tile_cols; + const int _m, _k, _n; + + public: + using WinogradBase = winograd::WinogradGEMM; + using WeightsTransform = typename WinogradBase::template WeightsTransform; + using InputTransform = typename WinogradBase::template InputTransform; + using WinogradConv = typename WinogradBase::template Convolution; + using OutputTransform = typename WinogradBase::template OutputTransform; + + /* Public member variables. */ + WeightsTransform weights_transform; /** Operator to transform weights to Winograd domain. */ + InputTransform _input_transform; /** Operator to transform input to Winograd domain. */ + arm_gemm::UniqueGemmCommon gemms; /** Operator to perform multiple GEMMs. */ + OutputTransform _output_transform; /** Operator to transform output from Winograd domain. */ + + /** Determine how much memory (in units of TIn) to allocate for the + * transformed weights. + */ + static unsigned int get_weight_storage_size( + const int n_output_channels, /** Number of output feature maps. */ + const int n_input_channels /** Number of input feature maps. */ + ); + + static unsigned int get_weight_stride( + const int n_output_channels, /** Number of output feature maps. */ + const int n_input_channels /** Number of input feature maps. */ + ); + + static unsigned int get_weight_multi_stride( + const int n_output_channels, /** Number of output feature maps. */ + const int n_input_channels /** Number of input feature maps. */ + ); + + /** Determine how much memory (in units of TIn) to allocate for the + * transformed input. + */ + static unsigned int get_input_storage_size( + const int n_batches, /** Number of batches in the input tensor. */ + const int n_channels, /** Number of feature maps in the input tensor. */ + const int n_rows, /** Number of rows in each feature map. */ + const int n_cols, /** Number of columns in each feature map. */ + const bool same_padding /** Use "SAME" padding, otherwise use "VALID". */ + ); + + /** Get the row stride for the A matrix in the Winograd domain. */ + static unsigned int get_input_stride( + const int n_batches, /** Number of batches in the input tensor. */ + const int n_channels, /** Number of feature maps in the input tensor. */ + const int n_rows, /** Number of rows in each feature map. */ + const int n_cols, /** Number of columns in each feature map. */ + const bool same_padding /** Use "SAME" padding, otherwise use "VALID". */ + ); + + /** Get the stride between A matrices in the Winograd domain. */ + static unsigned int get_input_multi_stride( + const int n_batches, /** Number of batches in the input tensor. */ + const int n_channels, /** Number of feature maps in the input tensor. */ + const int n_rows, /** Number of rows in each feature map. */ + const int n_cols, /** Number of columns in each feature map. */ + const bool same_padding /** Use "SAME" padding, otherwise use "VALID". */ + ); + + /** Determine how much memory (in units of TOut) to allocate for the + * (Winograd domain) output. + */ + static unsigned int get_output_storage_size( + const int n_batches, /** Number of batches in the output tensor. */ + const int n_rows, /** Number of rows in each feature map of the input tensor. */ + const int n_cols, /** Number of columns in each feature map of the input tensor. */ + const int n_output_channels, /** Number of feature maps in the output tensor. */ + const bool same_padding /** Use "SAME" padding, otherwise use "VALID". */ + ); + + static unsigned int get_output_stride( + const int n_batches, /** Number of batches in the output tensor. */ + const int n_rows, /** Number of rows in each feature map of the input tensor. */ + const int n_cols, /** Number of columns in each feature map of the input tensor. */ + const int n_output_channels, /** Number of feature maps in the output tensor. */ + const bool same_padding /** Use "SAME" padding, otherwise use "VALID". */ + ); + + static unsigned int get_output_multi_stride( + const int n_batches, /** Number of batches in the output tensor. */ + const int n_rows, /** Number of rows in each feature map of the input tensor. */ + const int n_cols, /** Number of columns in each feature map of the input tensor. */ + const int n_output_channels, /** Number of feature maps in the output tensor. */ + const bool same_padding /** Use "SAME" padding, otherwise use "VALID". */ + ); + + /** Get the shape (rows, cols) of a feature map of the output tensor. */ + static std::pair get_output_feature_map_shape( + const int n_input_rows, /** Number of rows in the input feature map. */ + const int n_input_cols, /** Number of columns in the input feature map. */ + const bool same_padding /** Use "SAME" padding, otherwise use "VALID". */ + ); + + /** Create a new Winograd convolution layer. + */ + WinogradConvolutionLayer( + const arm_gemm::CPUInfo &cpuinfo, /** Describes CPU properties. */ + const int n_threads, /** Maximum number of threads used to execute the convolution. */ + const int n_batches, /** Number of batches in the input and output tensors. */ + const int n_input_channels, /** Number of feature maps in a batch of the input tensor. */ + const int n_input_rows, /** Number of rows in a feature map of the input tensor. */ + const int n_input_cols, /** Number of columns in a feature map of the input tensor. */ + const int n_output_channels, /** Number of feature maps in the output tensor. */ + const bool same_padding, /** Use "SAME" padding, otherwise use "VALID". */ + const TIn* const weights, /** Pointer to weight tensor in spatial domain. Must be ordered as "Height x Rows x Input Feature Maps x Output Feature Maps. */ + TInGEMM* const weights_storage, /** Pointer to storage for weight tensor in the Winograd domain. Must be at least the size returned by `get_weight_storage_size`. */ + const TIn* const input, /** Pointer to NHWC ordered input tensor, in the spatial domain. */ + TInGEMM* const winograd_input, /** Pointer to working space for the input tensor in the Winograd domain. Must be at least the size returned by `get_input_storage_size`. */ + const TOut* const biases, /** Pointer to biases vector. Pass nullptr if no bias is provided. */ + TOut* const output, /** Pointer to NHWC ordered output tensor, in the spatial domain. */ + TOutGEMM* const winograd_output, /** Pointer to working space for the output tensor in the Winograd domain. Must be at least the size returned by `get_output_storage_size`. */ + const bool pretranspose_B=true, /** Hint that the B matrix can be pretransposed. */ + arm_gemm::GemmConfig *gemm_cfg=nullptr /** Pointer to GEMM configuration. */ + ); + + /* Utility methods for interacting with the layer. */ + unsigned int weight_transform_get_window(void) const; + void weight_transform_run(const unsigned int start, const unsigned int stop); + + ITransform& input_transform(void); + ITransform& output_transform(void); + + /* Get a pointer to the GEMM underlying the Winograd transform. */ + arm_gemm::IGemmCommon *gemm(void); +}; + +} diff --git a/arm_compute/core/NEON/kernels/convolution/winograd/winograd_output_transform.hpp b/arm_compute/core/NEON/kernels/convolution/winograd/winograd_output_transform.hpp deleted file mode 100644 index 07a0b8666a..0000000000 --- a/arm_compute/core/NEON/kernels/convolution/winograd/winograd_output_transform.hpp +++ /dev/null @@ -1,232 +0,0 @@ -/* - * Copyright (c) 2018 ARM Limited. - * - * SPDX-License-Identifier: MIT - * - * Permission is hereby granted, free of charge, to any person obtaining a copy - * of this software and associated documentation files (the "Software"), to - * deal in the Software without restriction, including without limitation the - * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or - * sell copies of the Software, and to permit persons to whom the Software is - * furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in all - * copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE - * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER - * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, - * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. - */ - -#pragma once - -namespace winograd -{ - - -namespace -{ - -template -class OutputTransformImplTiles -{ - public: - typedef void (*TileFn)( - const int n_channels, /** @param[in] Number of channels in output tensor */ - const T* const matrix_base, /** @param[in] Base pointer to Winograd output matrices. */ - const int matrix_stride, /** @param[in] Stride between matrices in the output space. */ - const T* const biases, /** @param[in] Pointer to bias vector (may be nullptr). */ - T* const output, /** @param[out] Pointer to output tensor. */ - const int output_row_stride, /** @param[in] Stride across rows of the output tensor. */ - const int output_col_stride, /** @param[in] Stride between columns of the output tensor. */ - const int _pad_bottom, /** @param[in] Bottom padding for unspecialised tiles. */ - const int _pad_right /** @param[in] Right padding for unspecialised tiles. */ - ); - - static TileFn get_tile_specialization( - const int pad_bottom, - const int pad_right - ); - - static constexpr unsigned int OutputTileRows = InnerTileRows - KernelRows + 1; - static constexpr unsigned int OutputTileCols = InnerTileCols - KernelCols + 1; - - private: - static constexpr unsigned int n_pad_bottom = OutputTileRows - 1; - static constexpr unsigned int n_pad_right = OutputTileCols - 1; - - static const TileFn tilefn_generic; /** Generic tile processing function. */ - static const TileFn tilefn_unpadded; /** Tile processor for unpadded tiles. */ - static const TileFn tilefn_bottom_padded[n_pad_bottom]; /** Bottom padding only. */ - static const TileFn tilefn_right_padded[n_pad_right]; /** Right padding only. */ -}; - -template -class OutputTransformImplTiles<1, KernelCols, 1, InnerTileCols, T> -{ - public: - typedef void (*TileFn)( - const int n_channels, /** @param[in] Number of channels in output tensor */ - const T* const matrix_base, /** @param[in] Base pointer to Winograd output matrices. */ - const int matrix_stride, /** @param[in] Stride between matrices in the output space. */ - const T* const biases, /** @param[in] Pointer to bias vector (may be nullptr). */ - T* const output, /** @param[out] Pointer to output tensor. */ - const int output_row_stride, /** @param[in] Stride across rows of the output tensor. */ - const int output_col_stride, /** @param[in] Stride between columns of the output tensor. */ - const int _pad_bottom, /** @param[in] Bottom padding for unspecialised tiles. */ - const int _pad_right /** @param[in] Right padding for unspecialised tiles. */ - ); - - static TileFn get_tile_specialization( - const int pad_bottom, - const int pad_right - ); - - static constexpr unsigned int OutputTileRows = 1; - static constexpr unsigned int OutputTileCols = InnerTileCols - KernelCols + 1; - - private: - static constexpr unsigned int n_pad_right = OutputTileCols - 1; - - static const TileFn tilefn_unpadded; /** Tile processor for unpadded tiles. */ - static const TileFn tilefn_right_padded[n_pad_right]; /** Right padding only. */ -}; - -template -class OutputTransformImpl -{ - private: - static void process_tile_row( - const int tile_N, - const int n_channels, - const T* const matrix_base, - const int matrix_stride, - const int matrix_row_stride, - const T* const biases, - T* const output, - const int output_row_stride, - const int output_col_stride, - const int row_pad_bottom, - const int row_pad_right - ); - - using Tiles = OutputTransformImplTiles< - KernelRows, KernelCols, InnerTileRows, InnerTileCols, T - >; - - public: - /** Apply the output transform to a tensor. */ - static void execute( - const int n_batches, - const int out_batch_stride, - const int n_rows, - const int out_row_stride, - const int n_cols, - const int out_col_stride, - const int n_channels, - const T* const matrix_base, - const int matrix_stride, - const int matrix_row_stride, - const T* const biases, - T* const output - ); - - static constexpr unsigned int OutputTileRows = Tiles::OutputTileRows; - static constexpr unsigned int OutputTileCols = Tiles::OutputTileCols; -}; - -template -class OutputTransformImpl -{ - public: - /** Apply the output transform to a tensor. */ - static void execute( - const int n_batches, - const int out_batch_stride, - const int n_rows, - const int out_row_stride, - const int n_cols, - const int out_col_stride, - const int n_channels, - const T* const matrix_base, - const int matrix_stride, - const int matrix_row_stride, - const T* const biases, - T* const output - ); - - static constexpr unsigned int OutputTileRows = InnerTileRows - KernelRows + 1; - static constexpr unsigned int OutputTileCols = 1; -}; - -} // namespace (anonymous) - -template -class OutputTransform -{ - public: - /***********************************************************************/ - /** Create an OutputTransform operator fixed on a given problem and set - * of pointers. - */ - OutputTransform( - const T* const matrix_base, /** Pointer to base of matrices. */ - const int matrix_stride, /** Stride between matrices. */ - const int matrix_row_stride, /** Stride within a matrix. */ - const T* const biases, /** Pointer to biases vector. */ - T* const output, /** Pointer to output tensor. */ - const int n_batches, /** Number of batches in output tensor. */ - const int n_rows, /** Number of rows in output tensor. */ - const int n_cols, /** Number of columns in output tensor. */ - const int n_channels, /** Number of channels in output tensor. */ - const int out_batch_stride=0, /** Output batch stride. */ - const int out_row_stride=0, /** Output row stride. */ - const int out_col_stride=0 /** Output column stride. */ - ); - - /** Get the window of work a given operator can perform. */ - unsigned int get_window() const; - static constexpr unsigned int WINDOW_BLOCK = 16; // Base size of window - - /** Perform work upon a window of the input. */ - void run(const unsigned int start, const unsigned int stop); - - /** Apply the transform to create a tensor. */ - static void execute( - const int n_batches, - const int out_batch_stride, - const int n_rows, - const int out_row_stride, - const int n_cols, - const int out_col_stride, - const int n_channels, - const T* const matrix_base, - const int matrix_stride, - const int matrix_row_stride, - const T* const biases, - T* const output - ); - - private: - using Transform = OutputTransformImpl< - KernelRows, KernelCols, InnerTileRows, InnerTileCols, T - >; - - static constexpr unsigned int OutputTileRows = Transform::OutputTileRows; - static constexpr unsigned int OutputTileCols = Transform::OutputTileCols; - - /** Member constants for instances of the transform. */ - const T* const _matrix_base; - const T* const _biases; - const int _matrix_stride, _matrix_row_stride; - T* const _outptr; - const int _n_batches, _n_rows, _n_cols, _n_channels, _tile_M, _tile_N; - const int _out_col_stride, _out_row_stride, _out_batch_stride; -}; - -} // namespace winograd - diff --git a/arm_compute/runtime/NEON/functions/NEWinogradConvolutionLayer.h b/arm_compute/runtime/NEON/functions/NEWinogradConvolutionLayer.h index 292c70b87c..ad37ba51ab 100644 --- a/arm_compute/runtime/NEON/functions/NEWinogradConvolutionLayer.h +++ b/arm_compute/runtime/NEON/functions/NEWinogradConvolutionLayer.h @@ -1,5 +1,5 @@ /* - * Copyright (c) 2017-2018 ARM Limited. + * Copyright (c) 2017-2019 ARM Limited. * * SPDX-License-Identifier: MIT * @@ -113,6 +113,8 @@ private: CPPPermute _permute_input; CPPPermute _permute_weights; CPPPermute _permute_output; + Tensor _input_transformed; + Tensor _output_transformed; Tensor _input_workspace; Tensor _output_workspace; Tensor _kernel_storage; diff --git a/src/core/NEON/kernels/NEWinogradConvolutionLayerKernel.cpp b/src/core/NEON/kernels/NEWinogradConvolutionLayerKernel.cpp index 3e76a080fd..263ded0b84 100644 --- a/src/core/NEON/kernels/NEWinogradConvolutionLayerKernel.cpp +++ b/src/core/NEON/kernels/NEWinogradConvolutionLayerKernel.cpp @@ -1,5 +1,5 @@ /* - * Copyright (c) 2017-2018 ARM Limited. + * Copyright (c) 2017-2019 ARM Limited. * * SPDX-License-Identifier: MIT * @@ -238,8 +238,7 @@ unsigned int NEWinogradLayerTransformWeightsKernel NEWinogradLayerTransformWeightsKernel::NEWinogradLayerTransformWeightsKernel() - : _weights_hwio(nullptr), _output(nullptr), _matrix_stride(0), _num_output_channels(0), _num_input_channels(0) - + : _transform(nullptr), _weights_hwio(nullptr), _output(nullptr), _matrix_stride(0), _num_output_channels(0), _num_input_channels(0) { } @@ -263,11 +262,10 @@ void NEWinogradLayerTransformWeightsKernel(num_output_channels, num_input_channels); - const int matrix_row_stride = roundup(num_output_channels, WinogradConv::N_BLOCK); - WeightsTransform transform(nullptr, nullptr, matrix_stride, matrix_row_stride, num_output_channels, num_input_channels); - Window win; - auto win_last = transform.get_window(); + Window win; + auto win_last = _transform->get_window(); win.set(Window::DimX, Window::Dimension(0, win_last, 1)); INEKernel::configure(win); } @@ -278,12 +276,14 @@ void NEWinogradLayerTransformWeightsKernelset_weight_tensor(_weights_hwio->buffer()); + const int matrix_row_stride = roundup(_num_output_channels, WinogradConv::N_BLOCK); + _transform->set_output_matrices(_output->buffer(), _matrix_stride, matrix_row_stride); + _transform->set_working_space(_output->buffer()); - const int matrix_row_stride = roundup(_num_output_channels, WinogradConv::N_BLOCK); - WeightsTransform transform(reinterpret_cast(_weights_hwio->buffer()), reinterpret_cast(_output->buffer()), _matrix_stride, matrix_row_stride, _num_output_channels, _num_input_channels); - const size_t fst = window.x().start(); - const size_t lst = window.x().end(); - transform.run(fst, lst); + _transform->run(fst, lst); } template @@ -330,6 +330,12 @@ unsigned int NEWinogradLayerTransformInputKernel(WinogradConv::get_input_storage_size(kern_shape, input_shape, padding) / sizeof(T)); } +template +unsigned int NEWinogradLayerTransformInputKernel::get_working_space_size(unsigned int num_threads) const +{ + return _transform->get_working_space_size(num_threads) / sizeof(T); +} + template int NEWinogradLayerTransformInputKernel::get_matrix_stride( const KernelShape &kernel_shape, const Tensor4DShape &input_shape, const PaddingType padding_type) const @@ -339,7 +345,8 @@ int NEWinogradLayerTransformInputKernel NEWinogradLayerTransformInputKernel::NEWinogradLayerTransformInputKernel() - : _input_nhwc(), _num_batches(0), _num_rows(0), _num_cols(0), _num_channels(0), _padding(), _output(nullptr), _matrix_stride(0) + : _transform(nullptr), _input_nhwc(nullptr), _num_batches(0), _num_rows(0), _num_cols(0), _num_channels(0), _padding(), _output(nullptr), _matrix_stride(0), _padding_top(), _padding_left(), + _padding_right(), _padding_bottom(), _workspace(nullptr) { } @@ -352,7 +359,8 @@ void NEWinogradLayerTransformInputKernel( + KernelRows, + KernelCols, + num_batches, + num_rows, + num_cols, + num_channels, + _padding_top, /**< Padding to apply to the top of the image. */ + _padding_left, /**< Padding to apply to the left of the image. */ + _padding_bottom, /**< Padding to apply to the bottom of the image. */ + _padding_right /**< Padding to apply to the right of the image. */ + ); + + Window win; + auto win_last = _transform->get_window(); win.set(Window::DimX, Window::Dimension(0, win_last, 1)); INEKernel::configure(win); } @@ -374,22 +401,25 @@ void NEWinogradLayerTransformInputKernelinfo()->element_size(); - const int input_col_stride = _input_nhwc->info()->strides_in_bytes().y() / element_size_in_bytes; - const int input_row_stride = _input_nhwc->info()->strides_in_bytes().z() / element_size_in_bytes; - const int input_batch_stride = _input_nhwc->info()->strides_in_bytes()[3] / element_size_in_bytes; - const auto input_nhwc_ptr = reinterpret_cast(_input_nhwc->buffer() + _input_nhwc->info()->offset_first_element_in_bytes()); - auto output_ptr = reinterpret_cast(_output->buffer() + _output->info()->offset_first_element_in_bytes()); - InputTransform input_transform(input_nhwc_ptr, - _num_batches, _num_rows, _num_cols, _num_channels, _padding, - output_ptr, - _matrix_stride, _num_channels, input_batch_stride, input_row_stride, input_col_stride); + const int element_size_in_bytes = _input_nhwc->info()->element_size(); + const int input_col_stride = _input_nhwc->info()->strides_in_bytes().y() / element_size_in_bytes; + const int input_row_stride = _input_nhwc->info()->strides_in_bytes().z() / element_size_in_bytes; + const int input_batch_stride = _input_nhwc->info()->strides_in_bytes()[3] / element_size_in_bytes; + const auto input_nhwc_ptr = reinterpret_cast(_input_nhwc->buffer() + _input_nhwc->info()->offset_first_element_in_bytes()); + auto output_ptr = reinterpret_cast(_output->buffer() + _output->info()->offset_first_element_in_bytes()); + ARM_COMPUTE_ERROR_ON_NULLPTR(output_ptr); + + _transform->set_input_tensor(input_nhwc_ptr, input_batch_stride, input_row_stride, input_col_stride); + _transform->set_output_matrices(output_ptr, _matrix_stride, _num_channels); + + _transform->set_working_space(_workspace->buffer()); // The code below cannot be moved to configure because biases hasn't been allocated at that point const size_t fst = window.x().start(); const size_t lst = window.x().end(); - input_transform.run(fst, lst); + _transform->run(fst, lst, info.thread_id); } template @@ -435,10 +465,17 @@ unsigned int NEWinogradLayerTransformOutputKernel NEWinogradLayerTransformOutputKernel::NEWinogradLayerTransformOutputKernel() - : _biases(nullptr), _output_workspace(nullptr), _matrix_stride(0), _matrix_row_stride(0), _output_nhwc(nullptr), _num_batches(0), _num_rows(0), _num_cols(0), _num_channels(0) + : _transform(nullptr), _biases(nullptr), _transformed_output(nullptr), _workspace(nullptr), _matrix_stride(0), _matrix_row_stride(0), _output_nhwc(nullptr), _num_batches(0), _num_rows(0), + _num_cols(0), _num_channels(0) { } +template +unsigned int NEWinogradLayerTransformOutputKernel::get_working_space_size(unsigned int num_threads) const +{ + return _transform->get_working_space_size(num_threads) / sizeof(T); +} + template int NEWinogradLayerTransformOutputKernel::get_matrix_stride( const KernelShape &kernel_shape, const Tensor4DShape &input_shape, const PaddingType padding_type) const @@ -455,28 +492,29 @@ Tensor4DShape NEWinogradLayerTransformOutputKernel void NEWinogradLayerTransformOutputKernel::configure( const ITensor *biases, - const ITensor *output_workingspace, + const ITensor *transformed_output, const int matrix_stride, ITensor *output_nhwc, const int num_batches, const int num_rows, const int num_cols, - const int num_channels) -{ - _biases = biases; - _output_workspace = output_workingspace; - _matrix_stride = matrix_stride; - _matrix_row_stride = roundup(num_channels, WinogradConv::N_BLOCK); - _output_nhwc = output_nhwc; - _num_batches = num_batches; - _num_rows = num_rows; - _num_cols = num_cols; - _num_channels = num_channels; + const int num_channels, + ITensor *workspace) +{ + _biases = biases; + _workspace = workspace; + _transformed_output = transformed_output; + _matrix_stride = matrix_stride; + _matrix_row_stride = roundup(num_channels, WinogradConv::N_BLOCK); + _output_nhwc = output_nhwc; + _num_batches = num_batches; + _num_rows = num_rows; + _num_cols = num_cols; + _num_channels = num_channels; // We don't have the biases buffer at this stage as it hasn't been allocated, we pass in nullptr OutputTransform is only used here to compute the window - OutputTransform output_transform(nullptr, _matrix_stride, _matrix_row_stride, nullptr, nullptr, _num_batches, _num_rows, _num_cols, _num_channels); - + _transform = arm_compute::support::cpp14::make_unique(num_batches, num_rows, num_cols, num_channels); Window win; - auto win_last = output_transform.get_window(); + auto win_last = _transform->get_window(); win.set(Window::DimX, Window::Dimension(0, win_last, 1)); _output_nhwc->info()->set_valid_region(ValidRegion(Coordinates(), _output_nhwc->info()->tensor_shape())); @@ -488,22 +526,22 @@ void NEWinogradLayerTransformOutputKernelinfo()->strides_in_bytes()[3] / sizeof(T); const int out_row_stride = _output_nhwc->info()->strides_in_bytes()[2] / sizeof(T); const int out_col_stride = _output_nhwc->info()->strides_in_bytes()[1] / sizeof(T); - OutputTransform output_transform(reinterpret_cast(_output_workspace->buffer()), _matrix_stride, _matrix_row_stride, - (_biases ? reinterpret_cast(_biases->buffer() + _biases->info()->offset_first_element_in_bytes()) : nullptr), - reinterpret_cast(_output_nhwc->buffer() + _output_nhwc->info()->offset_first_element_in_bytes()), - _num_batches, _num_rows, _num_cols, _num_channels, out_batch_stride, out_row_stride, out_col_stride); - + _transform->set_input_matrices(_transformed_output->buffer(), _matrix_stride, _matrix_row_stride); + _transform->set_bias((_biases ? reinterpret_cast(_biases->buffer() + _biases->info()->offset_first_element_in_bytes()) : nullptr)); + _transform->set_output_tensor(_output_nhwc->buffer() + _output_nhwc->info()->offset_first_element_in_bytes(), out_batch_stride, out_row_stride, out_col_stride); + _transform->set_working_space(_workspace->buffer()); // The code below cannot be moved to configure because biases hasn't been allocated at that point const size_t fst = window.x().start(); const size_t lst = window.x().end(); - output_transform.run(fst, lst); + _transform->run(fst, lst, info.thread_id); } template diff --git a/src/core/NEON/kernels/convolution/winograd/batched_blocked_gemm.cpp b/src/core/NEON/kernels/convolution/winograd/batched_blocked_gemm.cpp deleted file mode 100644 index ac83bf9dd2..0000000000 --- a/src/core/NEON/kernels/convolution/winograd/batched_blocked_gemm.cpp +++ /dev/null @@ -1,82 +0,0 @@ -/* - * Copyright (c) 2017 ARM Limited. - * - * SPDX-License-Identifier: MIT - * - * Permission is hereby granted, free of charge, to any person obtaining a copy - * of this software and associated documentation files (the "Software"), to - * deal in the Software without restriction, including without limitation the - * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or - * sell copies of the Software, and to permit persons to whom the Software is - * furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in all - * copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE - * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER - * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, - * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. - */ - -#include "arm_compute/core/NEON/kernels/convolution/winograd/batched_blocked_gemm.hpp" -#include "arm_compute/core/NEON/kernels/convolution/winograd/gemm.hpp" - -using namespace winograd; - -template -BatchedBlockedGemm::BatchedBlockedGemm( - const unsigned int n_gemms, - const int M, const int K, const int N, - const int a_matrix_stride, - const int a_row_stride, - const int b_matrix_stride, - const int b_row_stride, - const int c_matrix_stride, - const int c_row_stride, - const TIn* const a_ptr, - const TIn* const b_ptr, - TOut* const c_ptr -) : n_gemms(n_gemms), M(M), N(N), K(K), - a_matrix_stride(a_matrix_stride), - a_row_stride(a_row_stride), - b_matrix_stride(b_matrix_stride), - b_row_stride(b_row_stride), - c_matrix_stride(c_matrix_stride), - c_row_stride(c_row_stride), - a_ptr(a_ptr), b_ptr(b_ptr), c_ptr(c_ptr) -{ -} - -template -unsigned int BatchedBlockedGemm::get_window() const -{ - return n_gemms; -} - -template -void BatchedBlockedGemm::run( - const unsigned int start, const unsigned int stop -) -{ - // Perform the specified GEMMs - for (unsigned int i = start; i < stop; i++) - { - // Get pointers to the relevant matrices - const TIn* const mtr_a = a_ptr + i*a_matrix_stride; - const TIn* const mtr_b = b_ptr + i*b_matrix_stride; - TOut* const mtr_c = c_ptr + i*c_matrix_stride; - - // Perform the GEMM - BlockedGemm( - mtr_a, mtr_b, mtr_c, M, K, N, - a_row_stride, b_row_stride, c_row_stride - ); - } -} - -template class winograd::BatchedBlockedGemm<4, 16, float, float>; - diff --git a/src/core/NEON/kernels/convolution/winograd/padding.cpp b/src/core/NEON/kernels/convolution/winograd/padding.cpp new file mode 100644 index 0000000000..46fe57c7c9 --- /dev/null +++ b/src/core/NEON/kernels/convolution/winograd/padding.cpp @@ -0,0 +1,166 @@ +/* + * Copyright (c) 2019 ARM Limited. + * + * SPDX-License-Identifier: MIT + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#include +#include + +#include "padding.hpp" + +namespace padding +{ + +template +void copy_and_pad_tile( + const unsigned int tile_rows, + const unsigned int tile_cols, + const unsigned int n_channels, + const T* const inptr, + const unsigned int in_row_stride, + const unsigned int in_col_stride, + T* const outptr, + const unsigned int out_row_stride, + const unsigned int out_col_stride, + const unsigned int pad_top, + const unsigned int pad_left, + const unsigned int pad_bottom, + const unsigned int pad_right, + const T pad_value +) +{ + for (unsigned int out_i = 0; out_i < tile_rows; out_i++) + { + for (unsigned int out_j = 0; out_j < tile_cols; out_j++) + { + T* const output = outptr + out_i*out_row_stride + out_j*out_col_stride; + + if (out_i < pad_top || tile_rows - pad_bottom <= out_i || + out_j < pad_left || tile_cols - pad_right <= out_j) + { + for (unsigned int n = 0; n < n_channels; n++) + { + output[n] = pad_value; + } + } + else + { + const auto in_i = out_i - pad_top, in_j = out_j - pad_left; + const T* const input = inptr + in_i*in_row_stride + in_j*in_col_stride; + std::memcpy(output, input, n_channels * sizeof(T)); + } + } + } +} + +template void copy_and_pad_tile( + unsigned int, unsigned int, unsigned int, + const uint8_t *, unsigned int, unsigned int, + uint8_t *, unsigned int, unsigned int, + unsigned int, unsigned int, unsigned int, unsigned int, uint8_t +); + +template void copy_and_pad_tile( + unsigned int, unsigned int, unsigned int, + const float *, unsigned int, unsigned int, + float *, unsigned int, unsigned int, + unsigned int, unsigned int, unsigned int, unsigned int, float +); + +template +void CopyCropped::execute( + const size_t size, + const void * const inptr, + const size_t in_row_stride, + const size_t in_col_stride, + void * const outptr, + const size_t out_row_stride, + const size_t out_col_stride, + const unsigned int pad_top, + const unsigned int pad_left, + const unsigned int pad_bottom, + const unsigned int pad_right +) +{ + for (unsigned int out_i = 0, in_i = pad_top; in_i < TileRows - pad_bottom; out_i++, in_i++) + { + for (unsigned int out_j = 0, in_j = pad_left; in_j < TileCols - pad_right; out_j++, in_j++) + { + std::memcpy( + static_cast(outptr) + out_i*out_row_stride + out_j*out_col_stride, + static_cast(inptr) + in_i*in_row_stride + in_j*in_col_stride, + size + ); + } + } +} + +template class CopyCropped<2, 2>; +template class CopyCropped<3, 3>; +template class CopyCropped<4, 4>; + +template +void crop_and_copy_tile( + unsigned int tile_rows, + unsigned int tile_cols, + unsigned int n_channels, + const T *inptr, + unsigned int in_row_stride, + unsigned int in_col_stride, + T *outptr, + unsigned int out_row_stride, + unsigned int out_col_stride, + unsigned int crop_top, + unsigned int crop_left, + unsigned int crop_bottom, + unsigned int crop_right +) +{ + for (unsigned int out_i = 0, in_i = crop_top; in_i < tile_rows - crop_bottom; out_i++, in_i++) + { + for (unsigned int out_j = 0, in_j = crop_left; in_j < tile_cols - crop_right; out_j++, in_j++) + { + std::memcpy( + outptr + out_i*out_row_stride + out_j*out_col_stride, + inptr + in_i*in_row_stride + in_j*in_col_stride, + sizeof(T) * n_channels + ); + } + } +} + +template void crop_and_copy_tile( + unsigned int tile_rows, + unsigned int tile_cols, + unsigned int n_channels, + const float *inptr, + unsigned int in_row_stride, + unsigned int in_col_stride, + float *outptr, + unsigned int out_row_stride, + unsigned int out_col_stride, + unsigned int crop_top, + unsigned int crop_left, + unsigned int crop_bottom, + unsigned int crop_right +); + +} // namespace padding diff --git a/src/core/NEON/kernels/convolution/winograd/transforms/input_1x8_fp32.cpp b/src/core/NEON/kernels/convolution/winograd/transforms/input_1x8_fp32.cpp deleted file mode 100644 index e66300d39a..0000000000 --- a/src/core/NEON/kernels/convolution/winograd/transforms/input_1x8_fp32.cpp +++ /dev/null @@ -1,261 +0,0 @@ -/* - * Copyright (c) 2017 ARM Limited. - * - * SPDX-License-Identifier: MIT - * - * Permission is hereby granted, free of charge, to any person obtaining a copy - * of this software and associated documentation files (the "Software"), to - * deal in the Software without restriction, including without limitation the - * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or - * sell copies of the Software, and to permit persons to whom the Software is - * furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in all - * copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE - * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER - * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, - * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. - */ - -#include "arm_compute/core/NEON/kernels/convolution/winograd/transforms/input.hpp" -#include "arm_compute/core/NEON/kernels/convolution/winograd/winograd_gemm.hpp" -#include "arm_compute/core/NEON/kernels/convolution/common/arm.hpp" - -namespace -{ - -template -void winograd_input_transform_1x8_fp32_process_tile( - int n_channels, - const float* const input_base, - const int input_row_stride, - const int input_col_stride, - float* const matrix_base, - const int matrix_stride, - const int _pad_top, - const int _pad_left, - const int _pad_bottom, - const int _pad_right -) -{ - (void) input_row_stride; // No rows over which to stride - (void) _pad_top; // Never any top padding - (void) _pad_bottom; // Never any bottom padding - - // Extract padding arguments - const int pad_left = Specialized ? PadLeft : _pad_left; - const int pad_right = Specialized ? PadRight : _pad_right; - - constexpr int inner_tile_cols = 8; - const int cells_j = inner_tile_cols - pad_right; - - float *outptr = matrix_base; - - // Get pointers into the input tile - const float *x_ptrs[inner_tile_cols]; - for (int j = pad_left, xj = 0; j < cells_j; j++, xj++) - { - x_ptrs[j] = input_base + xj*input_col_stride; - } - - // Vectors used/computed in this kernel. - float x[inner_tile_cols]; - float U[inner_tile_cols]; - - for (int j = 0; j < inner_tile_cols; j++) - { - x[j] = 0.0f; - } - - // Perform the Winograd input transformation for each channel in the input - // tensor. - int channels_remaining = n_channels; -#ifdef __arm_any__ - for (; channels_remaining >= 4; channels_remaining -= 4) - { - float32x4_t x[inner_tile_cols], U[inner_tile_cols]; - for (int j = 0; j < inner_tile_cols; j++) - { - x[j] = vdupq_n_f32(0.0f); - } - - // Load x - for (int j = pad_left; j < cells_j; j++) - { - x[j] = vld1q_f32(x_ptrs[j]); - x_ptrs[j] += 4; - } - - // Compute U = x . X - U[0] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(x[6], 1), x[2], 49), x[4], -14), x[0], -36); - U[1] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(x[6], 1), x[2], 36), x[3], 13), x[4], -13), x[1], -36), x[5], -1); - U[2] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(x[6], 1), x[5], 1), x[2], 36), x[1], 36), x[4], -13), x[3], -13); - U[3] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(x[6], 1), x[3], 20), x[2], 9), x[5], -2), x[4], -10), x[1], -18); - U[4] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(x[6], 1), x[1], 18), x[2], 9), x[5], 2), x[4], -10), x[3], -20); - U[5] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(x[6], 1), x[3], 15), x[2], 4), x[5], -3), x[4], -5), x[1], -12); - U[6] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(x[6], 1), x[1], 12), x[2], 4), x[5], 3), x[4], -5), x[3], -15); - U[7] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(x[7], 1), x[3], 49), x[5], -14), x[1], -36); - - // Store the transformed vector - for (int j = 0; j < inner_tile_cols; j++) - { - vst1q_f32(outptr + j*matrix_stride, U[j]); - } - outptr += 4; - } - for (; channels_remaining >= 2; channels_remaining -= 2) - { - float32x2_t x[inner_tile_cols], U[inner_tile_cols]; - for (int j = 0; j < inner_tile_cols; j++) - { - x[j] = vdup_n_f32(0.0f); - } - - // Load x - for (int j = pad_left; j < cells_j; j++) - { - x[j] = vld1_f32(x_ptrs[j]); - x_ptrs[j] += 2; - } - - // Compute U = x . X - U[0] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(x[6], 1), x[2], 49), x[4], -14), x[0], -36); - U[1] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(x[6], 1), x[2], 36), x[3], 13), x[4], -13), x[1], -36), x[5], -1); - U[2] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(x[6], 1), x[5], 1), x[2], 36), x[1], 36), x[4], -13), x[3], -13); - U[3] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(x[6], 1), x[3], 20), x[2], 9), x[5], -2), x[4], -10), x[1], -18); - U[4] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(x[6], 1), x[1], 18), x[2], 9), x[5], 2), x[4], -10), x[3], -20); - U[5] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(x[6], 1), x[3], 15), x[2], 4), x[5], -3), x[4], -5), x[1], -12); - U[6] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(x[6], 1), x[1], 12), x[2], 4), x[5], 3), x[4], -5), x[3], -15); - U[7] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(x[7], 1), x[3], 49), x[5], -14), x[1], -36); - - // Store the transformed vector - for (int j = 0; j < inner_tile_cols; j++) - { - vst1_f32(outptr + j*matrix_stride, U[j]); - } - outptr += 2; - } -#endif // __arm_any__ - for (; channels_remaining; channels_remaining--) - { - // Load x - for (int j = pad_left; j < cells_j; j++) - { - x[j] = *(x_ptrs[j]++); - } - - // Compute U = x . X - U[0] = x[0]*-36 + x[4]*-14 + x[2]*49 + x[6]*1; - U[1] = x[5]*-1 + x[1]*-36 + x[4]*-13 + x[3]*13 + x[2]*36 + x[6]*1; - U[2] = x[3]*-13 + x[4]*-13 + x[1]*36 + x[2]*36 + x[5]*1 + x[6]*1; - U[3] = x[1]*-18 + x[4]*-10 + x[5]*-2 + x[2]*9 + x[3]*20 + x[6]*1; - U[4] = x[3]*-20 + x[4]*-10 + x[5]*2 + x[2]*9 + x[1]*18 + x[6]*1; - U[5] = x[1]*-12 + x[4]*-5 + x[5]*-3 + x[2]*4 + x[3]*15 + x[6]*1; - U[6] = x[3]*-15 + x[4]*-5 + x[5]*3 + x[2]*4 + x[1]*12 + x[6]*1; - U[7] = x[1]*-36 + x[5]*-14 + x[3]*49 + x[7]*1; - - // Store the transformed vector - for (int j = 0; j < inner_tile_cols; j++) - { - *(outptr + j*matrix_stride) = U[j]; - } - outptr++; - } -} - -} - -namespace winograd -{ -template -using Tiles = InputTransformImplTiles<1, x, 1, 8, float>; - -/*****************************************************************************/ -// 1x3 specialisations -template <> -const Tiles<3>::TileFn Tiles<3>::tilefn_generic = winograd_input_transform_1x8_fp32_process_tile; - -template <> -const Tiles<3>::TileFn Tiles<3>::tilefn_unpadded = winograd_input_transform_1x8_fp32_process_tile; - -template <> -const Tiles<3>::TileFn Tiles<3>::tilefn_left_padded[n_pad_left] = { - winograd_input_transform_1x8_fp32_process_tile, -}; - -template <> -const Tiles<3>::TileFn Tiles<3>::tilefn_right_padded[n_pad_right] = { - winograd_input_transform_1x8_fp32_process_tile, - winograd_input_transform_1x8_fp32_process_tile, - winograd_input_transform_1x8_fp32_process_tile, - winograd_input_transform_1x8_fp32_process_tile, - winograd_input_transform_1x8_fp32_process_tile, - winograd_input_transform_1x8_fp32_process_tile, - winograd_input_transform_1x8_fp32_process_tile, -}; -/*****************************************************************************/ - -/*****************************************************************************/ -// 1x5 specialisations -template <> -const Tiles<5>::TileFn Tiles<5>::tilefn_generic = winograd_input_transform_1x8_fp32_process_tile; - -template <> -const Tiles<5>::TileFn Tiles<5>::tilefn_unpadded = winograd_input_transform_1x8_fp32_process_tile; - -template <> -const Tiles<5>::TileFn Tiles<5>::tilefn_left_padded[n_pad_left] = { - winograd_input_transform_1x8_fp32_process_tile, -}; - -template <> -const Tiles<5>::TileFn Tiles<5>::tilefn_right_padded[n_pad_right] = { - winograd_input_transform_1x8_fp32_process_tile, - winograd_input_transform_1x8_fp32_process_tile, - winograd_input_transform_1x8_fp32_process_tile, - winograd_input_transform_1x8_fp32_process_tile, - winograd_input_transform_1x8_fp32_process_tile, - winograd_input_transform_1x8_fp32_process_tile, - winograd_input_transform_1x8_fp32_process_tile, -}; -/*****************************************************************************/ - -/*****************************************************************************/ -// 1x7 specialisations -template <> -const Tiles<7>::TileFn Tiles<7>::tilefn_generic = winograd_input_transform_1x8_fp32_process_tile; - -template <> -const Tiles<7>::TileFn Tiles<7>::tilefn_unpadded = winograd_input_transform_1x8_fp32_process_tile; - -template <> -const Tiles<7>::TileFn Tiles<7>::tilefn_left_padded[n_pad_left] = { - winograd_input_transform_1x8_fp32_process_tile, - winograd_input_transform_1x8_fp32_process_tile, -}; - -template <> -const Tiles<7>::TileFn Tiles<7>::tilefn_right_padded[n_pad_right] = { - winograd_input_transform_1x8_fp32_process_tile, - winograd_input_transform_1x8_fp32_process_tile, - winograd_input_transform_1x8_fp32_process_tile, - winograd_input_transform_1x8_fp32_process_tile, - winograd_input_transform_1x8_fp32_process_tile, - winograd_input_transform_1x8_fp32_process_tile, - winograd_input_transform_1x8_fp32_process_tile, -}; -/*****************************************************************************/ - - -template class InputTransform<1, 3, 1, 8, float>; -template class InputTransform<3, 1, 8, 1, float>; -template class InputTransform<1, 5, 1, 8, float>; -template class InputTransform<5, 1, 8, 1, float>; -template class InputTransform<1, 7, 1, 8, float>; -template class InputTransform<7, 1, 8, 1, float>; -} // namespace winograd diff --git a/src/core/NEON/kernels/convolution/winograd/transforms/input_2x2_3x3_fp32.cpp b/src/core/NEON/kernels/convolution/winograd/transforms/input_2x2_3x3_fp32.cpp deleted file mode 100644 index 4203945dd3..0000000000 --- a/src/core/NEON/kernels/convolution/winograd/transforms/input_2x2_3x3_fp32.cpp +++ /dev/null @@ -1,311 +0,0 @@ -/* - * Copyright (c) 2017 ARM Limited. - * - * SPDX-License-Identifier: MIT - * - * Permission is hereby granted, free of charge, to any person obtaining a copy - * of this software and associated documentation files (the "Software"), to - * deal in the Software without restriction, including without limitation the - * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or - * sell copies of the Software, and to permit persons to whom the Software is - * furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in all - * copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE - * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER - * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, - * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. - */ - -#include "arm_compute/core/NEON/kernels/convolution/winograd/transforms/input.hpp" -#include "arm_compute/core/NEON/kernels/convolution/winograd/winograd_gemm.hpp" -#include "arm_compute/core/NEON/kernels/convolution/common/arm.hpp" - -namespace winograd -{ - -using Tiles = InputTransformImplTiles<3, 3, 4, 4, float>; - -namespace -{ - - -template -void winograd_input_transform_4x4_fp32_process_tile( - int n_channels, - const float* const input_base, - const int input_row_stride, - const int input_col_stride, - float* const matrix_base, - const int matrix_stride, - const int _pad_top, - const int _pad_left, - const int _pad_bottom, - const int _pad_right - ) -{ -const int pad_top = Specialized ? PadTop : _pad_top; - const int pad_left = Specialized ? PadLeft : _pad_left; - const int pad_bottom = Specialized ? PadBottom : _pad_bottom; - const int pad_right = Specialized ? PadRight : _pad_right; - - constexpr int inner_tile_i = 4, inner_tile_j = 4; - const int cells_i = inner_tile_i - pad_bottom; - const int cells_j = inner_tile_i - pad_right; - - - - float *outptr = matrix_base; - - // Get pointers into the input tile - const float *x_ptrs[inner_tile_i][inner_tile_j]; - for (int i = pad_top, xi = 0; i < cells_i; i++, xi++) - { - // Get a pointer into the row - const float* const row_ptr = input_base + xi*input_row_stride; - - for (int j = pad_left, xj = 0; j < cells_j; j++, xj++) - { - x_ptrs[i][j] = row_ptr + xj*input_col_stride; - } - } - - // Matrices used/computed in this kernel. - float x[inner_tile_i][inner_tile_j]; - float XTx[inner_tile_i][inner_tile_j]; - float U[inner_tile_i][inner_tile_j]; - - for (int i = 0; i < inner_tile_i; i++) - { - for (int j = 0; j < inner_tile_j; j++) - { - x[i][j] = XTx[i][j] = 0.0f; - } - } - - // Perform the Winograd input transformation for each channel in the input - // tensor. - int channels_remaining = n_channels; -#ifdef __aarch64__ - for (; channels_remaining >= 4; channels_remaining -= 4) - { - // Matrices used/computed in this kernel. - float32x4_t x[inner_tile_i][inner_tile_j]; - float32x4_t XTx[inner_tile_i][inner_tile_j]; - float32x4_t U[inner_tile_i][inner_tile_j]; - - for (int i = 0; i < inner_tile_i; i++) - { - for (int j = 0; j < inner_tile_j; j++) - { - x[i][j] = vdupq_n_f32(0.0f); - XTx[i][j] = vdupq_n_f32(0.0f); - } - } - - // Load x - for (int i = pad_top; i < cells_i; i++) - { - for (int j = pad_left; j < cells_j; j++) - { - x[i][j] = vld1q_f32(x_ptrs[i][j]); - x_ptrs[i][j] += 4; - } - } - - // Compute XT . x - for (int j = pad_left; j < cells_j; j++) - { - // XTx[0][j] = x[0][j] - x[2][j]; - XTx[0][j] = vsubq_f32(x[0][j], x[2][j]); - - // XTx[1][j] = x[1][j] + x[2][j]; - XTx[1][j] = vaddq_f32(x[1][j], x[2][j]); - - // XTx[2][j] = x[2][j] - x[1][j]; - XTx[2][j] = vsubq_f32(x[2][j], x[1][j]); - - // XTx[3][j] = x[1][j] - x[3][j]; - XTx[3][j] = vsubq_f32(x[1][j], x[3][j]); - } - - // Compute U = XT . x . X - for (int i = 0; i < inner_tile_i; i++) - { - // U[i][0] = XTx[i][0] - XTx[i][2]; - U[i][0] = vsubq_f32(XTx[i][0], XTx[i][2]); - - // U[i][1] = XTx[i][1] + XTx[i][2]; - U[i][1] = vaddq_f32(XTx[i][1], XTx[i][2]); - - // U[i][2] = XTx[i][2] - XTx[i][1]; - U[i][2] = vsubq_f32(XTx[i][2], XTx[i][1]); - - // U[i][3] = XTx[i][1] - XTx[i][3]; - U[i][3] = vsubq_f32(XTx[i][1], XTx[i][3]); - } - - // Store the transformed matrix - for (int i = 0, m = 0; i < inner_tile_i; i++) - { - for (int j = 0; j < inner_tile_j; j++, m++) - { - vst1q_f32(outptr + m*matrix_stride, U[i][j]); - } - } - outptr += 4; - } -#endif // __aarch64__ -#ifdef __arm_any__ - for (; channels_remaining >= 2; channels_remaining -= 2) - { - // Matrices used/computed in this kernel. - float32x2_t x[inner_tile_i][inner_tile_j]; - float32x2_t XTx[inner_tile_i][inner_tile_j]; - float32x2_t U[inner_tile_i][inner_tile_j]; - - for (int i = 0; i < inner_tile_i; i++) - { - for (int j = 0; j < inner_tile_j; j++) - { - x[i][j] = vdup_n_f32(0.0f); - XTx[i][j] = vdup_n_f32(0.0f); - } - } - - // Load x - for (int i = pad_top; i < cells_i; i++) - { - for (int j = pad_left; j < cells_j; j++) - { - x[i][j] = vld1_f32(x_ptrs[i][j]); - x_ptrs[i][j] += 2; - } - } - - // Compute XT . x - for (int j = pad_left; j < cells_j; j++) - { - // XTx[0][j] = x[0][j] - x[2][j]; - XTx[0][j] = vsub_f32(x[0][j], x[2][j]); - - // XTx[1][j] = x[1][j] + x[2][j]; - XTx[1][j] = vadd_f32(x[1][j], x[2][j]); - - // XTx[2][j] = x[2][j] - x[1][j]; - XTx[2][j] = vsub_f32(x[2][j], x[1][j]); - - // XTx[3][j] = x[1][j] - x[3][j]; - XTx[3][j] = vsub_f32(x[1][j], x[3][j]); - } - - // Compute U = XT . x . X - for (int i = 0; i < inner_tile_i; i++) - { - // U[i][0] = XTx[i][0] - XTx[i][2]; - U[i][0] = vsub_f32(XTx[i][0], XTx[i][2]); - - // U[i][1] = XTx[i][1] + XTx[i][2]; - U[i][1] = vadd_f32(XTx[i][1], XTx[i][2]); - - // U[i][2] = XTx[i][2] - XTx[i][1]; - U[i][2] = vsub_f32(XTx[i][2], XTx[i][1]); - - // U[i][3] = XTx[i][1] - XTx[i][3]; - U[i][3] = vsub_f32(XTx[i][1], XTx[i][3]); - } - - // Store the transformed matrix - for (int i = 0, m = 0; i < inner_tile_i; i++) - { - for (int j = 0; j < inner_tile_j; j++, m++) - { - vst1_f32(outptr + m*matrix_stride, U[i][j]); - } - } - outptr += 2; - } -#endif // __arm_any__ - for (; channels_remaining; channels_remaining--) - { - // Load x - for (int i = pad_top; i < cells_i; i++) - { - for (int j = pad_left; j < cells_j; j++) - { - x[i][j] = *(x_ptrs[i][j]++); - } - } - - // Compute XT . x - for (int j = pad_left; j < cells_j; j++) - { - XTx[0][j] = x[0][j] - x[2][j]; - XTx[1][j] = x[1][j] + x[2][j]; - XTx[2][j] = x[2][j] - x[1][j]; - XTx[3][j] = x[1][j] - x[3][j]; - } - - // Compute U = XT . x . X - for (int i = 0; i < inner_tile_i; i++) - { - U[i][0] = XTx[i][0] - XTx[i][2]; - U[i][1] = XTx[i][1] + XTx[i][2]; - U[i][2] = XTx[i][2] - XTx[i][1]; - U[i][3] = XTx[i][1] - XTx[i][3]; - } - - // Store the transformed matrix - for (int i = 0, m = 0; i < inner_tile_i; i++) - { - for (int j = 0; j < inner_tile_j; j++, m++) - { - *(outptr + m*matrix_stride) = U[i][j]; - } - } - outptr++; - } -} - -} // namespace (anonymous) - -template <> -const Tiles::TileFn Tiles::tilefn_generic = winograd_input_transform_4x4_fp32_process_tile; - -template <> -const Tiles::TileFn Tiles::tilefn_unpadded = winograd_input_transform_4x4_fp32_process_tile; - - -template <> -const Tiles::TileFn Tiles::tilefn_top_padded[n_pad_top] = { - winograd_input_transform_4x4_fp32_process_tile, -}; - -template <> -const Tiles::TileFn Tiles::tilefn_left_padded[n_pad_left] = { - winograd_input_transform_4x4_fp32_process_tile, -}; - -template <> -const Tiles::TileFn Tiles::tilefn_bottom_padded[n_pad_bottom] = { - winograd_input_transform_4x4_fp32_process_tile, - winograd_input_transform_4x4_fp32_process_tile, - winograd_input_transform_4x4_fp32_process_tile, - winograd_input_transform_4x4_fp32_process_tile, -}; - -template <> -const Tiles::TileFn Tiles::tilefn_right_padded[n_pad_right] = { - winograd_input_transform_4x4_fp32_process_tile, - winograd_input_transform_4x4_fp32_process_tile, - winograd_input_transform_4x4_fp32_process_tile, - winograd_input_transform_4x4_fp32_process_tile, -}; - -template class InputTransform<3, 3, 4, 4, float>; -} // namespace winograd diff --git a/src/core/NEON/kernels/convolution/winograd/transforms/input_6x6_fp32.cpp b/src/core/NEON/kernels/convolution/winograd/transforms/input_6x6_fp32.cpp deleted file mode 100644 index 893122cc45..0000000000 --- a/src/core/NEON/kernels/convolution/winograd/transforms/input_6x6_fp32.cpp +++ /dev/null @@ -1,376 +0,0 @@ -/* - * Copyright (c) 2017 ARM Limited. - * - * SPDX-License-Identifier: MIT - * - * Permission is hereby granted, free of charge, to any person obtaining a copy - * of this software and associated documentation files (the "Software"), to - * deal in the Software without restriction, including without limitation the - * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or - * sell copies of the Software, and to permit persons to whom the Software is - * furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in all - * copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE - * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER - * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, - * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. - */ - -#include "arm_compute/core/NEON/kernels/convolution/winograd/transforms/input.hpp" -#include "arm_compute/core/NEON/kernels/convolution/winograd/winograd_gemm.hpp" -#include "arm_compute/core/NEON/kernels/convolution/common/arm.hpp" - -namespace -{ - -template -void winograd_input_transform_6x6_fp32_process_tile( - int n_channels, - const float* const input_base, - const int input_row_stride, - const int input_col_stride, - float* const matrix_base, -const int matrix_stride, - const int _pad_top, - const int _pad_left, - const int _pad_bottom, - const int _pad_right -) -{ - const int pad_top = Specialized ? PadTop : _pad_top; - const int pad_left = Specialized ? PadLeft : _pad_left; - const int pad_bottom = Specialized ? PadBottom : _pad_bottom; - const int pad_right = Specialized ? PadRight : _pad_right; - - constexpr int inner_tile_rows = 6; - constexpr int inner_tile_cols = 6; - - const int cells_i = inner_tile_rows - pad_bottom; - const int cells_j = inner_tile_cols - pad_right; - - float *outptr = matrix_base; - - // Get pointers into the input tile - const float *x_ptrs[inner_tile_rows][inner_tile_cols]; - for (int i = pad_top, xi = 0; i < cells_i; i++, xi++) - { - // Get a pointer into the row - const float* const row_ptr = input_base + xi*input_row_stride; - - for (int j = pad_left, xj = 0; j < cells_j; j++, xj++) - { - x_ptrs[i][j] = row_ptr + xj*input_col_stride; - } - } - - // Matrices used/computed in this kernel. - float x[inner_tile_rows][inner_tile_cols]; - float XTx[inner_tile_rows][inner_tile_cols]; - float U[inner_tile_rows][inner_tile_cols]; - for (int i = 0; i < inner_tile_rows; i++) - { - for (int j = 0; j < inner_tile_cols; j++) - { - x[i][j] = XTx[i][j] = 0.0f; - } - } - - // Perform the Winograd input transformation for each channel in the input - // tensor. - int channels_remaining = n_channels; -#ifdef __aarch64__ - for (; channels_remaining >= 4; channels_remaining -= 4) - { - // Matrices used/computed in this kernel - float32x4_t x[inner_tile_rows][inner_tile_cols]; - float32x4_t XTx[inner_tile_rows][inner_tile_cols]; - float32x4_t U[inner_tile_rows][inner_tile_cols]; - for (int i = 0; i < inner_tile_rows; i++) - { - for (int j = 0; j < inner_tile_cols; j++) - { - x[i][j] = vdupq_n_f32(0.0f); - XTx[i][j] = vdupq_n_f32(0.0f); - } - } - - // Read a 6x6 tile in the Winograd domain - for (int i = pad_top; i < cells_i; i++) - { - for (int j = pad_left; j < cells_j; j++) - { - x[i][j] = vld1q_f32(x_ptrs[i][j]); - x_ptrs[i][j] += 4; - } - } - - // Compute XT . x - for (int j = pad_left; j < cells_j; j++) - { - // XTx[0][j] = 4*x[0][j] + -5*x[2][j] + 1*x[4][j]; - XTx[0][j] = vmlsq_n_f32(vmlaq_n_f32(x[4][j], x[0][j], 4.0f), x[2][j], 5.0f); - - // XTx[1][j] = -4*x[1][j] + -4*x[2][j] + 1*x[3][j] + 1*x[4][j]; - XTx[1][j] = vmlsq_n_f32(vaddq_f32(x[3][j], x[4][j]), vaddq_f32(x[1][j], x[2][j]), 4.0f); - - // XTx[2][j] = 4*x[1][j] + -4*x[2][j] + -1*x[3][j] + 1*x[4][j]; - XTx[2][j] = vmlaq_n_f32(vsubq_f32(x[4][j], x[3][j]), vsubq_f32(x[1][j], x[2][j]), 4.0f); - - // XTx[3][j] = -2*x[1][j] + -1*x[2][j] + 2*x[3][j] + 1*x[4][j]; - XTx[3][j] = vmlaq_n_f32(vsubq_f32(x[4][j], x[2][j]), vsubq_f32(x[3][j], x[1][j]), 2.0f); - - // XTx[4][j] = 2*x[1][j] + -1*x[2][j] + -2*x[3][j] + 1*x[4][j]; - XTx[4][j] = vmlaq_n_f32(vsubq_f32(x[4][j], x[2][j]), vsubq_f32(x[1][j], x[3][j]), 2.0f); - - // XTx[5][j] = 4*x[1][j] + -5*x[3][j] + 1*x[5][j]; - XTx[5][j] = vmlsq_n_f32(vmlaq_n_f32(x[5][j], x[1][j], 4.0f), x[3][j], 5.0f); - } - - // Compute U = XT . x . X - for (int i = 0; i < inner_tile_rows; i++) - { - // U[i][0] = 4*XTx[i][0] + -5*XTx[i][2] + 1*XTx[i][4]; - U[i][0] = vmlsq_n_f32(vmlaq_n_f32(XTx[i][4], XTx[i][0], 4.0f), XTx[i][2], 5.0f); - - // U[i][1] = -4*XTx[i][1] + -4*XTx[i][2] + 1*XTx[i][3] + 1*XTx[i][4]; - U[i][1] = vmlsq_n_f32(vaddq_f32(XTx[i][3], XTx[i][4]), vaddq_f32(XTx[i][1], XTx[i][2]), 4.0f); - - // U[i][2] = 4*XTx[i][1] + -4*XTx[i][2] + -1*XTx[i][3] + 1*XTx[i][4]; - U[i][2] = vmlaq_n_f32(vsubq_f32(XTx[i][4], XTx[i][3]), vsubq_f32(XTx[i][1], XTx[i][2]), 4.0f); - - // U[i][3] = -2*XTx[i][1] + -1*XTx[i][2] + 2*XTx[i][3] + 1*XTx[i][4]; - U[i][3] = vmlaq_n_f32(vsubq_f32(XTx[i][4], XTx[i][2]), vsubq_f32(XTx[i][3], XTx[i][1]), 2.0f); - - // U[i][4] = 2*XTx[i][1] + -1*XTx[i][2] + -2*XTx[i][3] + 1*XTx[i][4]; - U[i][4] = vmlaq_n_f32(vsubq_f32(XTx[i][4], XTx[i][2]), vsubq_f32(XTx[i][1], XTx[i][3]), 2.0f); - - // U[i][5] = 4*XTx[i][1] + -5*XTx[i][3] + 1*XTx[i][5]; - U[i][5] = vmlsq_n_f32(vmlaq_n_f32(XTx[i][5], XTx[i][1], 4.0f), XTx[i][3], 5.0f); - } - - // Store the transformed matrix - for (int i = 0, m = 0; i < inner_tile_rows; i++) - { - for (int j = 0; j < inner_tile_cols; j++, m++) - { - vst1q_f32(outptr + m*matrix_stride, U[i][j]); - } - } - outptr += 4; - } -#endif // __aarch64__ -#ifdef __arm_any__ - for (; channels_remaining >= 2; channels_remaining -= 2) - { - // Matrices used/computed in this kernel - float32x2_t x[inner_tile_rows][inner_tile_cols]; - float32x2_t XTx[inner_tile_rows][inner_tile_cols]; - float32x2_t U[inner_tile_rows][inner_tile_cols]; - for (int i = 0; i < inner_tile_rows; i++) - { - for (int j = 0; j < inner_tile_cols; j++) - { - x[i][j] = vdup_n_f32(0.0f); - XTx[i][j] = vdup_n_f32(0.0f); - } - } - - // Read a 6x6 tile in the Winograd domain - for (int i = pad_top; i < cells_i; i++) - { - for (int j = pad_left; j < cells_j; j++) - { - x[i][j] = vld1_f32(x_ptrs[i][j]); - x_ptrs[i][j] += 2; - } - } - - // Compute XT . x - for (int j = pad_left; j < cells_j; j++) - { - // XTx[0][j] = 4*x[0][j] + -5*x[2][j] + 1*x[4][j]; - XTx[0][j] = vmls_n_f32(vmla_n_f32(x[4][j], x[0][j], 4.0f), x[2][j], 5.0f); - - // XTx[1][j] = -4*x[1][j] + -4*x[2][j] + 1*x[3][j] + 1*x[4][j]; - XTx[1][j] = vmls_n_f32(vadd_f32(x[3][j], x[4][j]), vadd_f32(x[1][j], x[2][j]), 4.0f); - - // XTx[2][j] = 4*x[1][j] + -4*x[2][j] + -1*x[3][j] + 1*x[4][j]; - XTx[2][j] = vmla_n_f32(vsub_f32(x[4][j], x[3][j]), vsub_f32(x[1][j], x[2][j]), 4.0f); - - // XTx[3][j] = -2*x[1][j] + -1*x[2][j] + 2*x[3][j] + 1*x[4][j]; - XTx[3][j] = vmla_n_f32(vsub_f32(x[4][j], x[2][j]), vsub_f32(x[3][j], x[1][j]), 2.0f); - - // XTx[4][j] = 2*x[1][j] + -1*x[2][j] + -2*x[3][j] + 1*x[4][j]; - XTx[4][j] = vmla_n_f32(vsub_f32(x[4][j], x[2][j]), vsub_f32(x[1][j], x[3][j]), 2.0f); - - // XTx[5][j] = 4*x[1][j] + -5*x[3][j] + 1*x[5][j]; - XTx[5][j] = vmls_n_f32(vmla_n_f32(x[5][j], x[1][j], 4.0f), x[3][j], 5.0f); - } - - // Compute U = XT . x . X - for (int i = 0; i < inner_tile_rows; i++) - { - // U[i][0] = 4*XTx[i][0] + -5*XTx[i][2] + 1*XTx[i][4]; - U[i][0] = vmls_n_f32(vmla_n_f32(XTx[i][4], XTx[i][0], 4.0f), XTx[i][2], 5.0f); - - // U[i][1] = -4*XTx[i][1] + -4*XTx[i][2] + 1*XTx[i][3] + 1*XTx[i][4]; - U[i][1] = vmls_n_f32(vadd_f32(XTx[i][3], XTx[i][4]), vadd_f32(XTx[i][1], XTx[i][2]), 4.0f); - - // U[i][2] = 4*XTx[i][1] + -4*XTx[i][2] + -1*XTx[i][3] + 1*XTx[i][4]; - U[i][2] = vmla_n_f32(vsub_f32(XTx[i][4], XTx[i][3]), vsub_f32(XTx[i][1], XTx[i][2]), 4.0f); - - // U[i][3] = -2*XTx[i][1] + -1*XTx[i][2] + 2*XTx[i][3] + 1*XTx[i][4]; - U[i][3] = vmla_n_f32(vsub_f32(XTx[i][4], XTx[i][2]), vsub_f32(XTx[i][3], XTx[i][1]), 2.0f); - - // U[i][4] = 2*XTx[i][1] + -1*XTx[i][2] + -2*XTx[i][3] + 1*XTx[i][4]; - U[i][4] = vmla_n_f32(vsub_f32(XTx[i][4], XTx[i][2]), vsub_f32(XTx[i][1], XTx[i][3]), 2.0f); - - // U[i][5] = 4*XTx[i][1] + -5*XTx[i][3] + 1*XTx[i][5]; - U[i][5] = vmls_n_f32(vmla_n_f32(XTx[i][5], XTx[i][1], 4.0f), XTx[i][3], 5.0f); - } - - // Store the transformed matrix - for (int i = 0, m = 0; i < inner_tile_rows; i++) - { - for (int j = 0; j < inner_tile_cols; j++, m++) - { - vst1_f32(outptr + m*matrix_stride, U[i][j]); - } - } - outptr += 2; - } -#endif // __arm_any__ - for (; channels_remaining; channels_remaining--) - { - // Load x - for (int i = pad_top; i < cells_i; i++) - { - for (int j = pad_left; j < cells_j; j++) - { - x[i][j] = *(x_ptrs[i][j]++); - } - } - - // Compute XT . x - for (int j = pad_left; j < cells_j; j++) - { - XTx[0][j] = 4*x[0][j] + -5*x[2][j] + 1*x[4][j]; - XTx[1][j] = -4*x[1][j] + -4*x[2][j] + 1*x[3][j] + 1*x[4][j]; - XTx[2][j] = 4*x[1][j] + -4*x[2][j] + -1*x[3][j] + 1*x[4][j]; - XTx[3][j] = -2*x[1][j] + -1*x[2][j] + 2*x[3][j] + 1*x[4][j]; - XTx[4][j] = 2*x[1][j] + -1*x[2][j] + -2*x[3][j] + 1*x[4][j]; - XTx[5][j] = 4*x[1][j] + -5*x[3][j] + 1*x[5][j]; - } - - // Compute U = XT . x . X - for (int i = 0; i < inner_tile_rows; i++) - { - U[i][0] = 4*XTx[i][0] + -5*XTx[i][2] + 1*XTx[i][4]; - U[i][1] = -4*XTx[i][1] + -4*XTx[i][2] + 1*XTx[i][3] + 1*XTx[i][4]; - U[i][2] = 4*XTx[i][1] + -4*XTx[i][2] + -1*XTx[i][3] + 1*XTx[i][4]; - U[i][3] = -2*XTx[i][1] + -1*XTx[i][2] + 2*XTx[i][3] + 1*XTx[i][4]; - U[i][4] = 2*XTx[i][1] + -1*XTx[i][2] + -2*XTx[i][3] + 1*XTx[i][4]; - U[i][5] = 4*XTx[i][1] + -5*XTx[i][3] + 1*XTx[i][5]; - } - - // Store the transformed matrix - for (int i = 0, m = 0; i < inner_tile_rows; i++) - { - for (int j = 0; j < inner_tile_cols; j++, m++) - { - *(outptr + m*matrix_stride) = U[i][j]; - } - } - outptr++; - } -} -} - -namespace winograd -{ -template -using Tiles = InputTransformImplTiles; - -template <> -const Tiles<3>::TileFn Tiles<3>::tilefn_generic = winograd_input_transform_6x6_fp32_process_tile; - -template <> -const Tiles<3>::TileFn Tiles<3>::tilefn_unpadded = winograd_input_transform_6x6_fp32_process_tile; - -template <> -const Tiles<3>::TileFn Tiles<3>::tilefn_top_padded[n_pad_top] = { - winograd_input_transform_6x6_fp32_process_tile, -}; - -template <> -const Tiles<3>::TileFn Tiles<3>::tilefn_left_padded[n_pad_left] = { - winograd_input_transform_6x6_fp32_process_tile, -}; - -template <> -const Tiles<3>::TileFn Tiles<3>::tilefn_bottom_padded[n_pad_bottom] = { - winograd_input_transform_6x6_fp32_process_tile, - winograd_input_transform_6x6_fp32_process_tile, - winograd_input_transform_6x6_fp32_process_tile, - winograd_input_transform_6x6_fp32_process_tile, - winograd_input_transform_6x6_fp32_process_tile, - winograd_input_transform_6x6_fp32_process_tile, -}; - -template <> -const Tiles<3>::TileFn Tiles<3>::tilefn_right_padded[n_pad_right] = { - winograd_input_transform_6x6_fp32_process_tile, - winograd_input_transform_6x6_fp32_process_tile, - winograd_input_transform_6x6_fp32_process_tile, - winograd_input_transform_6x6_fp32_process_tile, - winograd_input_transform_6x6_fp32_process_tile, - winograd_input_transform_6x6_fp32_process_tile, -}; - -template <> -const Tiles<5>::TileFn Tiles<5>::tilefn_generic = winograd_input_transform_6x6_fp32_process_tile; - - -template <> -const Tiles<5>::TileFn Tiles<5>::tilefn_unpadded = winograd_input_transform_6x6_fp32_process_tile; - - -template <> -const Tiles<5>::TileFn Tiles<5>::tilefn_top_padded[n_pad_top] = { - winograd_input_transform_6x6_fp32_process_tile, -}; - -template <> -const Tiles<5>::TileFn Tiles<5>::tilefn_left_padded[n_pad_left] = { - winograd_input_transform_6x6_fp32_process_tile, -}; - -template <> -const Tiles<5>::TileFn Tiles<5>::tilefn_bottom_padded[n_pad_bottom] = { - winograd_input_transform_6x6_fp32_process_tile, - winograd_input_transform_6x6_fp32_process_tile, - winograd_input_transform_6x6_fp32_process_tile, - winograd_input_transform_6x6_fp32_process_tile, - winograd_input_transform_6x6_fp32_process_tile, - winograd_input_transform_6x6_fp32_process_tile, -}; - -template <> -const Tiles<5>::TileFn Tiles<5>::tilefn_right_padded[n_pad_right] = { - winograd_input_transform_6x6_fp32_process_tile, - winograd_input_transform_6x6_fp32_process_tile, - winograd_input_transform_6x6_fp32_process_tile, - winograd_input_transform_6x6_fp32_process_tile, - winograd_input_transform_6x6_fp32_process_tile, - winograd_input_transform_6x6_fp32_process_tile, -}; - -template class InputTransform<3, 3, 6, 6, float>; -template class InputTransform<5, 5, 6, 6, float>; -} diff --git a/src/core/NEON/kernels/convolution/winograd/transforms/output_2_7_fp32.cpp b/src/core/NEON/kernels/convolution/winograd/transforms/output_2_7_fp32.cpp deleted file mode 100644 index ea842a45ee..0000000000 --- a/src/core/NEON/kernels/convolution/winograd/transforms/output_2_7_fp32.cpp +++ /dev/null @@ -1,163 +0,0 @@ -/* - * Copyright (c) 2017 ARM Limited. - * - * SPDX-License-Identifier: MIT - * - * Permission is hereby granted, free of charge, to any person obtaining a copy - * of this software and associated documentation files (the "Software"), to - * deal in the Software without restriction, including without limitation the - * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or - * sell copies of the Software, and to permit persons to whom the Software is - * furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in all - * copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE - * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER - * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, - * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. - */ - -#include "arm_compute/core/NEON/kernels/convolution/winograd/transforms/output.hpp" -#include "arm_compute/core/NEON/kernels/convolution/winograd/winograd_output_transform.hpp" -#include "arm_compute/core/NEON/kernels/convolution/common/arm.hpp" - -namespace -{ - -template -void winograd_output_transform_2_7_fp32_process_tile( - const int n_channels, - const float* const matrix_base, - const int matrix_stride, - const float* const biases, - float* const output, - const int output_row_stride, - const int output_col_stride, - const int _pad_bottom, - const int _pad_right -) -{ - (void) output_row_stride; - (void) _pad_bottom; - constexpr int output_tile_cols = 2; - constexpr int inner_tile_cols = 8; - - const int pad_right = Specialized ? PadRight : _pad_right; - const int cells_j = output_tile_cols - pad_right; - - - // Construct a map to the output cells - float *outptrs[cells_j]; - for (int j = 0; j < cells_j; j++) - { - outptrs[j] = output + j*output_col_stride; - } - const float *inptr = matrix_base; - const float *bptr = biases; - - // For each channel of the output - int channels_remaining = n_channels; -#ifdef __arm_any__ - for (; channels_remaining >= 4; channels_remaining -= 4) - { - // Matrices used and computed during this transform - float32x4_t F[inner_tile_cols], f[output_tile_cols], b = vdupq_n_f32(0.0f); - - // Read a 1x8 tile in the Winograd domain - for (int j = 0; j < inner_tile_cols; j++) - { - F[j] = vld1q_f32(inptr + j*matrix_stride); - } - inptr += 4; - - f[0] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(F[6], 1), F[5], 1), F[4], 1), F[3], 1), F[2], 1), F[1], 1), F[0], 1); - f[1] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(F[7], 1), F[2], 1), F[6], 3), F[4], 2), F[3], -2), F[5], -3), F[1], -1); - - // Write out the output tile - if (bptr != 0) - { - b = vld1q_f32(bptr); - bptr += 4; - } - for (int j = 0; j < cells_j; j++) - { - vst1q_f32(outptrs[j], f[j] + b); - outptrs[j] += 4; - } - } - for (; channels_remaining >= 2; channels_remaining -= 2) - { - // Matrices used and computed during this transform - float32x2_t F[inner_tile_cols], f[output_tile_cols], b = vdup_n_f32(0.0f); - - // Read a 1x8 tile in the Winograd domain - for (int j = 0; j < inner_tile_cols; j++) - { - F[j] = vld1_f32(inptr + j*matrix_stride); - } - inptr += 2; - - f[0] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(F[6], 1), F[5], 1), F[4], 1), F[3], 1), F[2], 1), F[1], 1), F[0], 1); - f[1] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(F[7], 1), F[2], 1), F[6], 3), F[4], 2), F[3], -2), F[5], -3), F[1], -1); - - // Write out the output tile - if (bptr != 0) - { - b = vld1_f32(bptr); - bptr += 2; - } - for (int j = 0; j < cells_j; j++) - { - vst1_f32(outptrs[j], f[j] + b); - outptrs[j] += 2; - } - } -#endif // __arm_any__ - for (; channels_remaining; channels_remaining--) - { - // Matrices used and computed during this transform - float F[inner_tile_cols], f[output_tile_cols], b = 0.0f; - - // Read a 1x8 tile in the Winograd domain - for (int j = 0; j < inner_tile_cols; j++) - { - F[j] = *(inptr + j*matrix_stride); - } - inptr++; - - f[0] = F[0]*1 + F[1]*1 + F[2]*1 + F[3]*1 + F[4]*1 + F[5]*1 + F[6]*1; - f[1] = F[1]*-1 + F[5]*-3 + F[3]*-2 + F[4]*2 + F[6]*3 + F[2]*1 + F[7]*1; - - // Write out the output tile - if (bptr != 0) - { - b = *(bptr++); - } - for (int j = 0; j < cells_j; j++) - { - *(outptrs[j]++) = f[j] + b; - } - } -} -} // namespace (anonymous) - -namespace winograd -{ -using Tiles = OutputTransformImplTiles<1, 7, 1, 8, float>; - -template <> -const Tiles::TileFn Tiles::tilefn_unpadded = winograd_output_transform_2_7_fp32_process_tile; - -template <> -const Tiles::TileFn Tiles::tilefn_right_padded[n_pad_right] = { - winograd_output_transform_2_7_fp32_process_tile -}; - -template class OutputTransform<1, 7, 1, 8, float>; -template class OutputTransform<7, 1, 8, 1, float>; -} // namespace winograd diff --git a/src/core/NEON/kernels/convolution/winograd/transforms/output_2x2_3x3_fp32.cpp b/src/core/NEON/kernels/convolution/winograd/transforms/output_2x2_3x3_fp32.cpp deleted file mode 100644 index 597b074026..0000000000 --- a/src/core/NEON/kernels/convolution/winograd/transforms/output_2x2_3x3_fp32.cpp +++ /dev/null @@ -1,375 +0,0 @@ -/* - * Copyright (c) 2017 ARM Limited. - * - * SPDX-License-Identifier: MIT - * - * Permission is hereby granted, free of charge, to any person obtaining a copy - * of this software and associated documentation files (the "Software"), to - * deal in the Software without restriction, including without limitation the - * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or - * sell copies of the Software, and to permit persons to whom the Software is - * furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in all - * copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE - * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER - * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, - * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. - */ - -#include "arm_compute/core/NEON/kernels/convolution/winograd/transforms/output.hpp" -#include "arm_compute/core/NEON/kernels/convolution/winograd/winograd_output_transform.hpp" -#include "arm_compute/core/NEON/kernels/convolution/common/arm.hpp" - -namespace -{ - -template -void winograd_output_transform_2x2_3x3_fp32_process_tile( - const int n_channels, - const float* const matrix_base, - const int matrix_stride, - const float* const biases, - float* const output, - const int output_row_stride, - const int output_col_stride, - const int _pad_bottom, - const int _pad_right -) -{ - constexpr int OutputTileRows = 2, OutputTileCols = 2; - const int pad_bottom = Specialized ? PadBottom : _pad_bottom; - const int pad_right = Specialized ? PadRight : _pad_right; - - const int cells_i = OutputTileRows - pad_bottom; - const int cells_j = OutputTileCols - pad_right; - - // Construct a map to the output cells - float *outptrs[OutputTileRows][OutputTileCols]; - for (int i = 0; i < cells_i; i++) - { - for (int j = 0; j < cells_j; j++) - { - outptrs[i][j] = output + i*output_row_stride + j*output_col_stride; - } - } - const float *inptr = matrix_base; - const float *bptr = biases; - - if (bptr) - { - // For each channel of the output - int channels_remaining = n_channels; -#ifdef __aarch64__ - for (; channels_remaining >= 4; channels_remaining -= 4) - { - // Matrices used and computed during this transform - float32x4_t F[4][4], FZ[4][2], f[2][2], b; - - // Read a 4x4 tile in the Winograd domain - for (int i = 0, m = 0; i < 4; i++) - { - for (int j = 0; j < 4; j++, m++) - { - F[i][j] = vld1q_f32(inptr + m*matrix_stride); - } - } - inptr += 4; - - // Compute the matrix F Z - for (int i = 0; i < 4; i++) - { - // FZ[i][0] = F[i][0] + F[i][1] + F[i][2]; - FZ[i][0] = vaddq_f32(vaddq_f32(F[i][0], F[i][1]), F[i][2]); - - // FZ[i][1] = F[i][1] - F[i][2] - F[i][3]; - FZ[i][1] = vsubq_f32(vsubq_f32(F[i][1], F[i][2]), F[i][3]); - } - - // Compute the output tile f = ZT F Z - for (int j = 0; j < 2; j++) - { - // f[0][j] = FZ[0][j] + FZ[1][j] + FZ[2][j]; - f[0][j] = vaddq_f32(vaddq_f32(FZ[0][j], FZ[1][j]), FZ[2][j]); - - // f[1][j] = FZ[1][j] - FZ[2][j] - FZ[3][j]; - f[1][j] = vsubq_f32(vsubq_f32(FZ[1][j], FZ[2][j]), FZ[3][j]); - } - - // Load the bias vector - b = vld1q_f32(bptr); - bptr += 4; - - // Write out the output tile - for (int i = 0; i < cells_i; i++) - { - for (int j = 0; j < cells_j; j++) - { - vst1q_f32(outptrs[i][j], vaddq_f32(f[i][j], b)); - outptrs[i][j] += 4; - } - } - } -#endif // __aarch64__ -#ifdef __arm_any__ - for (; channels_remaining >= 2; channels_remaining -= 2) - { - // Matrices used and computed during this transform - float32x2_t F[4][4], FZ[4][2], f[2][2], b; - - // Read a 4x4 tile in the Winograd domain - for (int i = 0, m = 0; i < 4; i++) - { - for (int j = 0; j < 4; j++, m++) - { - F[i][j] = vld1_f32(inptr + m*matrix_stride); - } - } - inptr += 2; - - // Compute the matrix F Z - for (int i = 0; i < 4; i++) - { - // FZ[i][0] = F[i][0] + F[i][1] + F[i][2]; - FZ[i][0] = vadd_f32(vadd_f32(F[i][0], F[i][1]), F[i][2]); - - // FZ[i][1] = F[i][1] - F[i][2] - F[i][3]; - FZ[i][1] = vsub_f32(vsub_f32(F[i][1], F[i][2]), F[i][3]); - } - - // Compute the output tile f = ZT F Z - for (int j = 0; j < 2; j++) - { - // f[0][j] = FZ[0][j] + FZ[1][j] + FZ[2][j]; - f[0][j] = vadd_f32(vadd_f32(FZ[0][j], FZ[1][j]), FZ[2][j]); - - // f[1][j] = FZ[1][j] - FZ[2][j] - FZ[3][j]; - f[1][j] = vsub_f32(vsub_f32(FZ[1][j], FZ[2][j]), FZ[3][j]); - } - - // Load the bias vector - b = vld1_f32(bptr); - bptr += 2; - - // Write out the output tile - for (int i = 0; i < cells_i; i++) - { - for (int j = 0; j < cells_j; j++) - { - vst1_f32(outptrs[i][j], vadd_f32(f[i][j], b)); - outptrs[i][j] += 2; - } - } - } -#endif // __arm_any__ - for (; channels_remaining; channels_remaining--) - { - // Matrices used and computed during this transform - float F[4][4], FZ[4][2], f[2][2], b; - - // Read a 4x4 tile in the Winograd domain - for (int i = 0, m = 0; i < 4; i++) - { - for (int j = 0; j < 4; j++, m++) - { - F[i][j] = *(inptr + m*matrix_stride); - } - } - inptr++; - - // Compute the matrix F Z - for (int i = 0; i < 4; i++) - { - FZ[i][0] = F[i][0] + F[i][1] + F[i][2]; - FZ[i][1] = F[i][1] - F[i][2] - F[i][3]; - } - - // Compute the output tile f = ZT F Z - for (int j = 0; j < 2; j++) - { - f[0][j] = FZ[0][j] + FZ[1][j] + FZ[2][j]; - f[1][j] = FZ[1][j] - FZ[2][j] - FZ[3][j]; - } - - // Load the bias - b = *(bptr++); - - // Write out the output tile - for (int i = 0; i < cells_i; i++) - { - for (int j = 0; j < cells_j; j++) - { - *(outptrs[i][j]++) = f[i][j] + b; - } - } - } - } - else - { - // For each channel of the output - int channels_remaining = n_channels; -#ifdef __aarch64__ - for (; channels_remaining >= 4; channels_remaining -= 4) - { - // Matrices used and computed during this transform - float32x4_t F[4][4], FZ[4][2], f[2][2]; - - // Read a 4x4 tile in the Winograd domain - for (int i = 0, m = 0; i < 4; i++) - { - for (int j = 0; j < 4; j++, m++) - { - F[i][j] = vld1q_f32(inptr + m*matrix_stride); - } - } - inptr += 4; - - // Compute the matrix F Z - for (int i = 0; i < 4; i++) - { - // FZ[i][0] = F[i][0] + F[i][1] + F[i][2]; - FZ[i][0] = vaddq_f32(vaddq_f32(F[i][0], F[i][1]), F[i][2]); - - // FZ[i][1] = F[i][1] - F[i][2] - F[i][3]; - FZ[i][1] = vsubq_f32(vsubq_f32(F[i][1], F[i][2]), F[i][3]); - } - - // Compute the output tile f = ZT F Z - for (int j = 0; j < 2; j++) - { - // f[0][j] = FZ[0][j] + FZ[1][j] + FZ[2][j]; - f[0][j] = vaddq_f32(vaddq_f32(FZ[0][j], FZ[1][j]), FZ[2][j]); - - // f[1][j] = FZ[1][j] - FZ[2][j] - FZ[3][j]; - f[1][j] = vsubq_f32(vsubq_f32(FZ[1][j], FZ[2][j]), FZ[3][j]); - } - - // Write out the output tile - for (int i = 0; i < cells_i; i++) - { - for (int j = 0; j < cells_j; j++) - { - vst1q_f32(outptrs[i][j], f[i][j]); - outptrs[i][j] += 4; - } - } - } -#endif // __aarch64__ -#ifdef __arm_any__ - for (; channels_remaining >= 2; channels_remaining -= 2) - { - // Matrices used and computed during this transform - float32x2_t F[4][4], FZ[4][2], f[2][2]; - - // Read a 4x4 tile in the Winograd domain - for (int i = 0, m = 0; i < 4; i++) - { - for (int j = 0; j < 4; j++, m++) - { - F[i][j] = vld1_f32(inptr + m*matrix_stride); - } - } - inptr += 2; - - // Compute the matrix F Z - for (int i = 0; i < 4; i++) - { - // FZ[i][0] = F[i][0] + F[i][1] + F[i][2]; - FZ[i][0] = vadd_f32(vadd_f32(F[i][0], F[i][1]), F[i][2]); - - // FZ[i][1] = F[i][1] - F[i][2] - F[i][3]; - FZ[i][1] = vsub_f32(vsub_f32(F[i][1], F[i][2]), F[i][3]); - } - - // Compute the output tile f = ZT F Z - for (int j = 0; j < 2; j++) - { - // f[0][j] = FZ[0][j] + FZ[1][j] + FZ[2][j]; - f[0][j] = vadd_f32(vadd_f32(FZ[0][j], FZ[1][j]), FZ[2][j]); - - // f[1][j] = FZ[1][j] - FZ[2][j] - FZ[3][j]; - f[1][j] = vsub_f32(vsub_f32(FZ[1][j], FZ[2][j]), FZ[3][j]); - } - - // Write out the output tile - for (int i = 0; i < cells_i; i++) - { - for (int j = 0; j < cells_j; j++) - { - vst1_f32(outptrs[i][j], f[i][j]); - outptrs[i][j] += 2; - } - } - } -#endif // __arm_any__ - for (; channels_remaining; channels_remaining--) - { - // Matrices used and computed during this transform - float F[4][4], FZ[4][2], f[2][2]; - - // Read a 4x4 tile in the Winograd domain - for (int i = 0, m = 0; i < 4; i++) - { - for (int j = 0; j < 4; j++, m++) - { - F[i][j] = *(inptr + m*matrix_stride); - } - } - inptr++; - - // Compute the matrix F Z - for (int i = 0; i < 4; i++) - { - FZ[i][0] = F[i][0] + F[i][1] + F[i][2]; - FZ[i][1] = F[i][1] - F[i][2] - F[i][3]; - } - - // Compute the output tile f = ZT F Z - for (int j = 0; j < 2; j++) - { - f[0][j] = FZ[0][j] + FZ[1][j] + FZ[2][j]; - f[1][j] = FZ[1][j] - FZ[2][j] - FZ[3][j]; - } - - // Write out the output tile - for (int i = 0; i < cells_i; i++) - { - for (int j = 0; j < cells_j; j++) - { - *(outptrs[i][j]++) = f[i][j]; - } - } - } - } -} - -} // namespace (anonymous) - -namespace winograd -{ -using Tiles = OutputTransformImplTiles<3, 3, 4, 4, float>; - -template <> -const Tiles::TileFn Tiles::tilefn_generic = winograd_output_transform_2x2_3x3_fp32_process_tile; - -template <> -const Tiles::TileFn Tiles::tilefn_unpadded = winograd_output_transform_2x2_3x3_fp32_process_tile; - -template <> -const Tiles::TileFn Tiles::tilefn_bottom_padded[n_pad_bottom] = { - winograd_output_transform_2x2_3x3_fp32_process_tile -}; - -template <> -const Tiles::TileFn Tiles::tilefn_right_padded[n_pad_right] = { - winograd_output_transform_2x2_3x3_fp32_process_tile -}; - -template class OutputTransform<3, 3, 4, 4, float>; -} // namespace winograd - diff --git a/src/core/NEON/kernels/convolution/winograd/transforms/output_2x2_5x5_fp32.cpp b/src/core/NEON/kernels/convolution/winograd/transforms/output_2x2_5x5_fp32.cpp deleted file mode 100644 index 60d7181d97..0000000000 --- a/src/core/NEON/kernels/convolution/winograd/transforms/output_2x2_5x5_fp32.cpp +++ /dev/null @@ -1,369 +0,0 @@ -/* - * Copyright (c) 2017 ARM Limited. - * - * SPDX-License-Identifier: MIT - * - * Permission is hereby granted, free of charge, to any person obtaining a copy - * of this software and associated documentation files (the "Software"), to - * deal in the Software without restriction, including without limitation the - * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or - * sell copies of the Software, and to permit persons to whom the Software is - * furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in all - * copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE - * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER - * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, - * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. - */ - -#include "arm_compute/core/NEON/kernels/convolution/winograd/transforms/output.hpp" -#include "arm_compute/core/NEON/kernels/convolution/winograd/winograd_output_transform.hpp" -#include "arm_compute/core/NEON/kernels/convolution/common/arm.hpp" - -namespace -{ - -template -void winograd_output_transform_2x2_5x5_fp32_process_tile( - const int n_channels, - const float* const matrix_base, - const int matrix_stride, - const float* const biases, - float* const output, - const int output_row_stride, - const int output_col_stride, - const int _pad_bottom, - const int _pad_right -) -{ - constexpr int OutputTileRows = 2, OutputTileCols = 2; - const int pad_bottom = Specialized ? PadBottom : _pad_bottom; - const int pad_right = Specialized ? PadRight : _pad_right; - - const int cells_i = 2 - pad_bottom; - const int cells_j = 2 - pad_right; - - // Construct a map to the output cells - float *outptrs[OutputTileRows][OutputTileCols]; - for (int i = 0; i < cells_i; i++) - { - for (int j = 0; j < cells_j; j++) - { - outptrs[i][j] = output + i*output_row_stride + j*output_col_stride; - } - } - const float *inptr = matrix_base; - const float *bptr = biases; - - if (bptr) - { - // For each channel of the output - int channels_remaining = n_channels; -#ifdef __aarch64__ - for (; channels_remaining >= 4; channels_remaining -= 4) - { - // Matrices used and computed during this transform - float32x4_t F[6][6], FZ[6][2], f[2][2], b; - - // Read a 6x6 tile in the Winograd domain - for (int i = 0, m = 0; i < 6; i++) - { - for (int j = 0; j < 6; j++, m++) - { - F[i][j] = vld1q_f32(inptr + m*matrix_stride); - } - } - inptr += 4; - - // Compute the matrix F Z - for (int i = 0; i < 6; i++) - { - // FZ[i][0] = 1*F[i][0] + 1*F[i][1] + 1*F[i][2] + 1*F[i][3] + 1*F[i][4]; - FZ[i][0] = vaddq_f32(vaddq_f32(vaddq_f32(F[i][0], F[i][1]), vaddq_f32(F[i][2], F[i][3])), F[i][4]); - - // FZ[i][1] = 1*F[i][1] + -1*F[i][2] + 2*F[i][3] + -2*F[i][4] + 1*F[i][5]; - FZ[i][1] = vaddq_f32(vmlaq_n_f32(vsubq_f32(F[i][1], F[i][2]), vsubq_f32(F[i][3], F[i][4]), 2.0f), F[i][5]); - } - - // Compute the output tile f = ZT F Z - for (int j = 0; j < 2; j++) - { - // f[0][j] = 1*FZ[0][j] + 1*FZ[1][j] + 1*FZ[2][j] + 1*FZ[3][j] + 1*FZ[4][j]; - f[0][j] = vaddq_f32(vaddq_f32(vaddq_f32(FZ[0][j], FZ[1][j]), vaddq_f32(FZ[2][j], FZ[3][j])), FZ[4][j]); - - // f[1][j] = 1*FZ[1][j] + -1*FZ[2][j] + 2*FZ[3][j] + -2*FZ[4][j] + 1*FZ[5][j]; - f[1][j] = vaddq_f32(vmlaq_n_f32(vsubq_f32(FZ[1][j], FZ[2][j]), vsubq_f32(FZ[3][j], FZ[4][j]), 2.0f), FZ[5][j]); - } - - // Write out the output tile - b = vld1q_f32(bptr); - bptr += 4; - for (int i = 0; i < cells_i; i++) - { - for (int j = 0; j < cells_j; j++) - { - vst1q_f32(outptrs[i][j], vaddq_f32(f[i][j], b)); - outptrs[i][j] += 4; - } - } - } -#endif // __aarch64__ -#ifdef __arm_any__ - for (; channels_remaining >= 2; channels_remaining -= 2) - { - // Matrices used and computed during this transform - float32x2_t F[6][6], FZ[6][2], f[2][2], b; - - // Read a 6x6 tile in the Winograd domain - for (int i = 0, m = 0; i < 6; i++) - { - for (int j = 0; j < 6; j++, m++) - { - F[i][j] = vld1_f32(inptr + m*matrix_stride); - } - } - inptr += 2; - - // Compute the matrix F Z - for (int i = 0; i < 6; i++) - { - // FZ[i][0] = 1*F[i][0] + 1*F[i][1] + 1*F[i][2] + 1*F[i][3] + 1*F[i][4]; - FZ[i][0] = vadd_f32(vadd_f32(vadd_f32(F[i][0], F[i][1]), vadd_f32(F[i][2], F[i][3])), F[i][4]); - - // FZ[i][1] = 1*F[i][1] + -1*F[i][2] + 2*F[i][3] + -2*F[i][4] + 1*F[i][5]; - FZ[i][1] = vadd_f32(vmla_n_f32(vsub_f32(F[i][1], F[i][2]), vsub_f32(F[i][3], F[i][4]), 2.0f), F[i][5]); - } - - // Compute the output tile f = ZT F Z - for (int j = 0; j < 2; j++) - { - // f[0][j] = 1*FZ[0][j] + 1*FZ[1][j] + 1*FZ[2][j] + 1*FZ[3][j] + 1*FZ[4][j]; - f[0][j] = vadd_f32(vadd_f32(vadd_f32(FZ[0][j], FZ[1][j]), vadd_f32(FZ[2][j], FZ[3][j])), FZ[4][j]); - - // f[1][j] = 1*FZ[1][j] + -1*FZ[2][j] + 2*FZ[3][j] + -2*FZ[4][j] + 1*FZ[5][j]; - f[1][j] = vadd_f32(vmla_n_f32(vsub_f32(FZ[1][j], FZ[2][j]), vsub_f32(FZ[3][j], FZ[4][j]), 2.0f), FZ[5][j]); - } - - // Write out the output tile - b = vld1_f32(bptr); - bptr += 2; - for (int i = 0; i < cells_i; i++) - { - for (int j = 0; j < cells_j; j++) - { - vst1_f32(outptrs[i][j], vadd_f32(f[i][j], b)); - outptrs[i][j] += 2; - } - } - } -#endif // __arm_any__ - for (; channels_remaining; channels_remaining--) - { - // Matrices used and computed during this transform - float F[6][6], FZ[6][2], f[2][2], b; - - // Read a 6x6 tile in the Winograd domain - for (int i = 0, m = 0; i < 6; i++) - { - for (int j = 0; j < 6; j++, m++) - { - F[i][j] = *(inptr + m*matrix_stride); - } - } - inptr++; - - // Compute the matrix F Z - for (int i = 0; i < 6; i++) - { - FZ[i][0] = 1*F[i][0] + 1*F[i][1] + 1*F[i][2] + 1*F[i][3] + 1*F[i][4]; - FZ[i][1] = 1*F[i][1] + -1*F[i][2] + 2*F[i][3] + -2*F[i][4] + 1*F[i][5]; - } - - // Compute the output tile f = ZT F Z - for (int j = 0; j < 2; j++) - { - f[0][j] = 1*FZ[0][j] + 1*FZ[1][j] + 1*FZ[2][j] + 1*FZ[3][j] + 1*FZ[4][j]; - f[1][j] = 1*FZ[1][j] + -1*FZ[2][j] + 2*FZ[3][j] + -2*FZ[4][j] + 1*FZ[5][j]; - } - - // Write out the output tile - b = *(bptr++); - for (int i = 0; i < cells_i; i++) - { - for (int j = 0; j < cells_j; j++) - { - *(outptrs[i][j]++) = f[i][j] + b; - } - } - } - } - else - { - // For each channel of the output - int channels_remaining = n_channels; -#ifdef __aarch64__ - for (; channels_remaining >= 4; channels_remaining -= 4) - { - // Matrices used and computed during this transform - float32x4_t F[6][6], FZ[6][2], f[2][2]; - - // Read a 6x6 tile in the Winograd domain - for (int i = 0, m = 0; i < 6; i++) - { - for (int j = 0; j < 6; j++, m++) - { - F[i][j] = vld1q_f32(inptr + m*matrix_stride); - } - } - inptr += 4; - - // Compute the matrix F Z - for (int i = 0; i < 6; i++) - { - // FZ[i][0] = 1*F[i][0] + 1*F[i][1] + 1*F[i][2] + 1*F[i][3] + 1*F[i][4]; - FZ[i][0] = vaddq_f32(vaddq_f32(vaddq_f32(F[i][0], F[i][1]), vaddq_f32(F[i][2], F[i][3])), F[i][4]); - - // FZ[i][1] = 1*F[i][1] + -1*F[i][2] + 2*F[i][3] + -2*F[i][4] + 1*F[i][5]; - FZ[i][1] = vaddq_f32(vmlaq_n_f32(vsubq_f32(F[i][1], F[i][2]), vsubq_f32(F[i][3], F[i][4]), 2.0f), F[i][5]); - } - - // Compute the output tile f = ZT F Z - for (int j = 0; j < 2; j++) - { - // f[0][j] = 1*FZ[0][j] + 1*FZ[1][j] + 1*FZ[2][j] + 1*FZ[3][j] + 1*FZ[4][j]; - f[0][j] = vaddq_f32(vaddq_f32(vaddq_f32(FZ[0][j], FZ[1][j]), vaddq_f32(FZ[2][j], FZ[3][j])), FZ[4][j]); - - // f[1][j] = 1*FZ[1][j] + -1*FZ[2][j] + 2*FZ[3][j] + -2*FZ[4][j] + 1*FZ[5][j]; - f[1][j] = vaddq_f32(vmlaq_n_f32(vsubq_f32(FZ[1][j], FZ[2][j]), vsubq_f32(FZ[3][j], FZ[4][j]), 2.0f), FZ[5][j]); - } - - // Write out the output tile - for (int i = 0; i < cells_i; i++) - { - for (int j = 0; j < cells_j; j++) - { - vst1q_f32(outptrs[i][j], f[i][j]); - outptrs[i][j] += 4; - } - } - } -#endif // __aarch64__ -#ifdef __arm_any__ - for (; channels_remaining >= 2; channels_remaining -= 2) - { - // Matrices used and computed during this transform - float32x2_t F[6][6], FZ[6][2], f[2][2]; - - // Read a 6x6 tile in the Winograd domain - for (int i = 0, m = 0; i < 6; i++) - { - for (int j = 0; j < 6; j++, m++) - { - F[i][j] = vld1_f32(inptr + m*matrix_stride); - } - } - inptr += 2; - - // Compute the matrix F Z - for (int i = 0; i < 6; i++) - { - // FZ[i][0] = 1*F[i][0] + 1*F[i][1] + 1*F[i][2] + 1*F[i][3] + 1*F[i][4]; - FZ[i][0] = vadd_f32(vadd_f32(vadd_f32(F[i][0], F[i][1]), vadd_f32(F[i][2], F[i][3])), F[i][4]); - - // FZ[i][1] = 1*F[i][1] + -1*F[i][2] + 2*F[i][3] + -2*F[i][4] + 1*F[i][5]; - FZ[i][1] = vadd_f32(vmla_n_f32(vsub_f32(F[i][1], F[i][2]), vsub_f32(F[i][3], F[i][4]), 2.0f), F[i][5]); - } - - // Compute the output tile f = ZT F Z - for (int j = 0; j < 2; j++) - { - // f[0][j] = 1*FZ[0][j] + 1*FZ[1][j] + 1*FZ[2][j] + 1*FZ[3][j] + 1*FZ[4][j]; - f[0][j] = vadd_f32(vadd_f32(vadd_f32(FZ[0][j], FZ[1][j]), vadd_f32(FZ[2][j], FZ[3][j])), FZ[4][j]); - - // f[1][j] = 1*FZ[1][j] + -1*FZ[2][j] + 2*FZ[3][j] + -2*FZ[4][j] + 1*FZ[5][j]; - f[1][j] = vadd_f32(vmla_n_f32(vsub_f32(FZ[1][j], FZ[2][j]), vsub_f32(FZ[3][j], FZ[4][j]), 2.0f), FZ[5][j]); - } - - // Write out the output tile - for (int i = 0; i < cells_i; i++) - { - for (int j = 0; j < cells_j; j++) - { - vst1_f32(outptrs[i][j], f[i][j]); - outptrs[i][j] += 2; - } - } - } -#endif // __arm_any__ - for (; channels_remaining; channels_remaining--) - { - // Matrices used and computed during this transform - float F[6][6], FZ[6][2], f[2][2]; - - // Read a 6x6 tile in the Winograd domain - for (int i = 0, m = 0; i < 6; i++) - { - for (int j = 0; j < 6; j++, m++) - { - F[i][j] = *(inptr + m*matrix_stride); - } - } - inptr++; - - // Compute the matrix F Z - for (int i = 0; i < 6; i++) - { - FZ[i][0] = 1*F[i][0] + 1*F[i][1] + 1*F[i][2] + 1*F[i][3] + 1*F[i][4]; - FZ[i][1] = 1*F[i][1] + -1*F[i][2] + 2*F[i][3] + -2*F[i][4] + 1*F[i][5]; - } - - // Compute the output tile f = ZT F Z - for (int j = 0; j < 2; j++) - { - f[0][j] = 1*FZ[0][j] + 1*FZ[1][j] + 1*FZ[2][j] + 1*FZ[3][j] + 1*FZ[4][j]; - f[1][j] = 1*FZ[1][j] + -1*FZ[2][j] + 2*FZ[3][j] + -2*FZ[4][j] + 1*FZ[5][j]; - } - - // Write out the output tile - for (int i = 0; i < cells_i; i++) - { - for (int j = 0; j < cells_j; j++) - { - *(outptrs[i][j]++) = f[i][j]; - } - } - } - } -} - -} // namespace (anonymous) - -namespace winograd -{ -using Tiles = OutputTransformImplTiles<5, 5, 6, 6, float>; - -template <> -const Tiles::TileFn Tiles::tilefn_generic = winograd_output_transform_2x2_5x5_fp32_process_tile; - -template <> -const Tiles::TileFn Tiles::tilefn_unpadded = winograd_output_transform_2x2_5x5_fp32_process_tile; - -template <> -const Tiles::TileFn Tiles::tilefn_bottom_padded[n_pad_bottom] = { - winograd_output_transform_2x2_5x5_fp32_process_tile -}; - -template <> -const Tiles::TileFn Tiles::tilefn_right_padded[n_pad_right] = { - winograd_output_transform_2x2_5x5_fp32_process_tile -}; - -template class OutputTransform<5, 5, 6, 6, float>; -} // namespace winograd - diff --git a/src/core/NEON/kernels/convolution/winograd/transforms/output_4_5_fp32.cpp b/src/core/NEON/kernels/convolution/winograd/transforms/output_4_5_fp32.cpp deleted file mode 100644 index 911759b128..0000000000 --- a/src/core/NEON/kernels/convolution/winograd/transforms/output_4_5_fp32.cpp +++ /dev/null @@ -1,171 +0,0 @@ -/* - * Copyright (c) 2017 ARM Limited. - * - * SPDX-License-Identifier: MIT - * - * Permission is hereby granted, free of charge, to any person obtaining a copy - * of this software and associated documentation files (the "Software"), to - * deal in the Software without restriction, including without limitation the - * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or - * sell copies of the Software, and to permit persons to whom the Software is - * furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in all - * copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE - * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER - * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, - * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. - */ - -#include "arm_compute/core/NEON/kernels/convolution/winograd/transforms/output.hpp" -#include "arm_compute/core/NEON/kernels/convolution/winograd/winograd_output_transform.hpp" -#include "arm_compute/core/NEON/kernels/convolution/common/arm.hpp" - -namespace -{ - -template -void winograd_output_transform_4_5_fp32_process_tile( - const int n_channels, - const float* const matrix_base, - const int matrix_stride, - const float* const biases, - float* const output, - const int output_row_stride, - const int output_col_stride, - const int _pad_bottom, - const int _pad_right -) -{ - (void) output_row_stride; - (void) _pad_bottom; - constexpr int output_tile_cols = 4; - constexpr int inner_tile_cols = 8; - - const int pad_right = Specialized ? PadRight : _pad_right; - const int cells_j = output_tile_cols - pad_right; - - // Construct a map to the output cells - float *outptrs[cells_j]; - for (int j = 0; j < cells_j; j++) - { - outptrs[j] = output + j*output_col_stride; - } - const float *inptr = matrix_base; - const float *bptr = biases; - - // For each channel of the output - int channels_remaining = n_channels; -#ifdef __arm_any__ - for (; channels_remaining >= 4; channels_remaining -= 4) - { - // Matrices used and computed during this transform - float32x4_t F[inner_tile_cols], f[output_tile_cols], b = vdupq_n_f32(0.0f); - - // Read a 1x8 tile in the Winograd domain - for (int j = 0; j < inner_tile_cols; j++) - { - F[j] = vld1q_f32(inptr + j*matrix_stride); - } - inptr += 4; - - f[0] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(F[6], 1), F[5], 1), F[4], 1), F[3], 1), F[2], 1), F[1], 1), F[0], 1); - f[1] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(F[2], 1), F[6], 3), F[4], 2), F[3], -2), F[5], -3), F[1], -1); - f[2] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(F[2], 1), F[1], 1), F[6], 9), F[5], 9), F[4], 4), F[3], 4); - f[3] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(F[7], 1), F[2], 1), F[6], 27), F[4], 8), F[3], -8), F[5], -27), F[1], -1); - - // Write out the output tile - if (bptr != 0) - { - b = vld1q_f32(bptr); - bptr += 4; - } - for (int j = 0; j < cells_j; j++) - { - vst1q_f32(outptrs[j], f[j] + b); - outptrs[j] += 4; - } - } - for (; channels_remaining >= 2; channels_remaining -= 2) - { - // Matrices used and computed during this transform - float32x2_t F[inner_tile_cols], f[output_tile_cols], b = vdup_n_f32(0.0f); - - // Read a 1x8 tile in the Winograd domain - for (int j = 0; j < inner_tile_cols; j++) - { - F[j] = vld1_f32(inptr + j*matrix_stride); - } - inptr += 2; - - f[0] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(F[6], 1), F[5], 1), F[4], 1), F[3], 1), F[2], 1), F[1], 1), F[0], 1); - f[1] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(F[2], 1), F[6], 3), F[4], 2), F[3], -2), F[5], -3), F[1], -1); - f[2] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(F[2], 1), F[1], 1), F[6], 9), F[5], 9), F[4], 4), F[3], 4); - f[3] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(F[7], 1), F[2], 1), F[6], 27), F[4], 8), F[3], -8), F[5], -27), F[1], -1); - - // Write out the output tile - if (bptr != 0) - { - b = vld1_f32(bptr); - bptr += 2; - } - for (int j = 0; j < cells_j; j++) - { - vst1_f32(outptrs[j], f[j] + b); - outptrs[j] += 2; - } - } -#endif // __arm_any__ - for (; channels_remaining; channels_remaining--) - { - // Matrices used and computed during this transform - float F[inner_tile_cols], f[output_tile_cols], b = 0.0f; - - // Read a 1x8 tile in the Winograd domain - for (int j = 0; j < inner_tile_cols; j++) - { - F[j] = *(inptr + j*matrix_stride); - } - inptr++; - - f[0] = F[0]*1 + F[1]*1 + F[2]*1 + F[3]*1 + F[4]*1 + F[5]*1 + F[6]*1; - f[1] = F[1]*-1 + F[5]*-3 + F[3]*-2 + F[4]*2 + F[6]*3 + F[2]*1; - f[2] = F[3]*4 + F[4]*4 + F[5]*9 + F[6]*9 + F[1]*1 + F[2]*1; - f[3] = F[1]*-1 + F[5]*-27 + F[3]*-8 + F[4]*8 + F[6]*27 + F[2]*1 + F[7]*1; - - // Write out the output tile - if (bptr != 0) - { - b = *(bptr++); - } - for (int j = 0; j < cells_j; j++) - { - *(outptrs[j]++) = f[j] + b; - } - } -} - -} // namespace (anonymous) - -namespace winograd -{ -using Tiles = OutputTransformImplTiles<1, 5, 1, 8, float>; - -template <> -const Tiles::TileFn Tiles::tilefn_unpadded = winograd_output_transform_4_5_fp32_process_tile; - -template <> -const Tiles::TileFn Tiles::tilefn_right_padded[n_pad_right] = { - winograd_output_transform_4_5_fp32_process_tile, - winograd_output_transform_4_5_fp32_process_tile, - winograd_output_transform_4_5_fp32_process_tile -}; - -template class OutputTransform<1, 5, 1, 8, float>; -template class OutputTransform<5, 1, 8, 1, float>; -} // namespace winograd diff --git a/src/core/NEON/kernels/convolution/winograd/transforms/output_4x4_3x3_fp32.cpp b/src/core/NEON/kernels/convolution/winograd/transforms/output_4x4_3x3_fp32.cpp deleted file mode 100644 index 15cc04b352..0000000000 --- a/src/core/NEON/kernels/convolution/winograd/transforms/output_4x4_3x3_fp32.cpp +++ /dev/null @@ -1,428 +0,0 @@ -/* - * Copyright (c) 2017 ARM Limited. - * - * SPDX-License-Identifier: MIT - * - * Permission is hereby granted, free of charge, to any person obtaining a copy - * of this software and associated documentation files (the "Software"), to - * deal in the Software without restriction, including without limitation the - * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or - * sell copies of the Software, and to permit persons to whom the Software is - * furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in all - * copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE - * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER - * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, - * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. - */ - -#include "arm_compute/core/NEON/kernels/convolution/winograd/transforms/output.hpp" -#include "arm_compute/core/NEON/kernels/convolution/winograd/winograd_output_transform.hpp" -#include "arm_compute/core/NEON/kernels/convolution/common/arm.hpp" - -namespace -{ - -template -void winograd_output_transform_4x4_3x3_fp32_process_tile( - const int n_channels, - const float* const matrix_base, - const int matrix_stride, - const float* const biases, - float* const output, - const int output_row_stride, - const int output_col_stride, - const int _pad_bottom, - const int _pad_right -) -{ - const int pad_bottom = Specialized ? PadBottom : _pad_bottom; - const int pad_right = Specialized ? PadRight : _pad_right; - constexpr int TileRows = 4, TileCols = 4; - - const int cells_i = TileRows - pad_bottom; - const int cells_j = TileCols - pad_right; - - // Construct a map to the output cells - float *outptrs[TileRows][TileCols]; - for (int i = 0; i < cells_i; i++) - { - for (int j = 0; j < cells_j; j++) - { - outptrs[i][j] = output + i*output_row_stride + j*output_col_stride; - } - } - const float *inptr = matrix_base; - const float *bptr = biases; - - if (bptr) - { - // For each channel of the output - int channels_remaining = n_channels; -#ifdef __aarch64__ - for (; channels_remaining >= 4; channels_remaining -= 4) - { - // Matrices used and computed during this transform - float32x4_t F[6][6], FZ[6][4], f[4][4], b; - - // Read a 6x6 tile in the Winograd domain - for (int i = 0, m = 0; i < 6; i++) - { - for (int j = 0; j < 6; j++, m++) - { - F[i][j] = vld1q_f32(inptr + m*matrix_stride); - } - } - inptr += 4; - - // Compute the matrix F Z - for (int i = 0; i < 6; i++) - { - // FZ[i][0] = 1*F[i][0] + 1*F[i][1] + 1*F[i][2] + 1*F[i][3] + 1*F[i][4]; - FZ[i][0] = vaddq_f32(vaddq_f32(vaddq_f32(F[i][0], F[i][1]), vaddq_f32(F[i][2], F[i][3])), F[i][4]); - - // FZ[i][1] = 1*F[i][1] + -1*F[i][2] + 2*F[i][3] + -2*F[i][4]; - FZ[i][1] = vmlaq_n_f32(vsubq_f32(F[i][1], F[i][2]), vsubq_f32(F[i][3], F[i][4]), 2.0f); - - // FZ[i][2] = 1*F[i][1] + 1*F[i][2] + 4*F[i][3] + 4*F[i][4]; - FZ[i][2] = vmlaq_n_f32(vaddq_f32(F[i][1], F[i][2]), vaddq_f32(F[i][3], F[i][4]), 4.0f); - - // FZ[i][3] = 1*F[i][1] + -1*F[i][2] + 8*F[i][3] + -8*F[i][4] + 1*F[i][5]; - FZ[i][3] = vaddq_f32(vmlaq_n_f32(vsubq_f32(F[i][1], F[i][2]), vsubq_f32(F[i][3], F[i][4]), 8.0f), F[i][5]); - } - - // Compute the output tile f = ZT F Z - for (int j = 0; j < 4; j++) - { - // f[0][j] = 1*FZ[0][j] + 1*FZ[1][j] + 1*FZ[2][j] + 1*FZ[3][j] + 1*FZ[4][j]; - f[0][j] = vaddq_f32(vaddq_f32(vaddq_f32(FZ[0][j], FZ[1][j]), vaddq_f32(FZ[2][j], FZ[3][j])), FZ[4][j]); - - // f[1][j] = 1*FZ[1][j] + -1*FZ[2][j] + 2*FZ[3][j] + -2*FZ[4][j]; - f[1][j] = vmlaq_n_f32(vsubq_f32(FZ[1][j], FZ[2][j]), vsubq_f32(FZ[3][j], FZ[4][j]), 2.0f); - - // f[2][j] = 1*FZ[1][j] + 1*FZ[2][j] + 4*FZ[3][j] + 4*FZ[4][j]; - f[2][j] = vmlaq_n_f32(vaddq_f32(FZ[1][j], FZ[2][j]), vaddq_f32(FZ[3][j], FZ[4][j]), 4.0f); - - // f[3][j] = 1*FZ[1][j] + -1*FZ[2][j] + 8*FZ[3][j] + -8*FZ[4][j] + 1*FZ[5][j]; - f[3][j] = vaddq_f32(vmlaq_n_f32(vsubq_f32(FZ[1][j], FZ[2][j]), vsubq_f32(FZ[3][j], FZ[4][j]), 8.0f), FZ[5][j]); - } - - // Write out the output tile - b = vld1q_f32(bptr); - bptr += 4; - for (int i = 0; i < cells_i; i++) - { - for (int j = 0; j < cells_j; j++) - { - vst1q_f32(outptrs[i][j], vaddq_f32(f[i][j], b)); - outptrs[i][j] += 4; - } - } - } -#endif // __aarch64__ -#ifdef __arm_any__ - for (; channels_remaining >= 2; channels_remaining -= 2) - { - // Matrices used and computed during this transform - float32x2_t F[6][6], FZ[6][4], f[4][4], b; - - // Read a 6x6 tile in the Winograd domain - for (int i = 0, m = 0; i < 6; i++) - { - for (int j = 0; j < 6; j++, m++) - { - F[i][j] = vld1_f32(inptr + m*matrix_stride); - } - } - inptr += 2; - - // Compute the matrix F Z - for (int i = 0; i < 6; i++) - { - // FZ[i][0] = 1*F[i][0] + 1*F[i][1] + 1*F[i][2] + 1*F[i][3] + 1*F[i][4]; - FZ[i][0] = vadd_f32(vadd_f32(vadd_f32(F[i][0], F[i][1]), vadd_f32(F[i][2], F[i][3])), F[i][4]); - - // FZ[i][1] = 1*F[i][1] + -1*F[i][2] + 2*F[i][3] + -2*F[i][4]; - FZ[i][1] = vmla_n_f32(vsub_f32(F[i][1], F[i][2]), vsub_f32(F[i][3], F[i][4]), 2.0f); - - // FZ[i][2] = 1*F[i][1] + 1*F[i][2] + 4*F[i][3] + 4*F[i][4]; - FZ[i][2] = vmla_n_f32(vadd_f32(F[i][1], F[i][2]), vadd_f32(F[i][3], F[i][4]), 4.0f); - - // FZ[i][3] = 1*F[i][1] + -1*F[i][2] + 8*F[i][3] + -8*F[i][4] + 1*F[i][5]; - FZ[i][3] = vadd_f32(vmla_n_f32(vsub_f32(F[i][1], F[i][2]), vsub_f32(F[i][3], F[i][4]), 8.0f), F[i][5]); - } - - // Compute the output tile f = ZT F Z - for (int j = 0; j < 4; j++) - { - // f[0][j] = 1*FZ[0][j] + 1*FZ[1][j] + 1*FZ[2][j] + 1*FZ[3][j] + 1*FZ[4][j]; - f[0][j] = vadd_f32(vadd_f32(vadd_f32(FZ[0][j], FZ[1][j]), vadd_f32(FZ[2][j], FZ[3][j])), FZ[4][j]); - - // f[1][j] = 1*FZ[1][j] + -1*FZ[2][j] + 2*FZ[3][j] + -2*FZ[4][j]; - f[1][j] = vmla_n_f32(vsub_f32(FZ[1][j], FZ[2][j]), vsub_f32(FZ[3][j], FZ[4][j]), 2.0f); - - // f[2][j] = 1*FZ[1][j] + 1*FZ[2][j] + 4*FZ[3][j] + 4*FZ[4][j]; - f[2][j] = vmla_n_f32(vadd_f32(FZ[1][j], FZ[2][j]), vadd_f32(FZ[3][j], FZ[4][j]), 4.0f); - - // f[3][j] = 1*FZ[1][j] + -1*FZ[2][j] + 8*FZ[3][j] + -8*FZ[4][j] + 1*FZ[5][j]; - f[3][j] = vadd_f32(vmla_n_f32(vsub_f32(FZ[1][j], FZ[2][j]), vsub_f32(FZ[3][j], FZ[4][j]), 8.0f), FZ[5][j]); - } - - // Write out the output tile - b = vld1_f32(bptr); - bptr += 2; - for (int i = 0; i < cells_i; i++) - { - for (int j = 0; j < cells_j; j++) - { - vst1_f32(outptrs[i][j], vadd_f32(f[i][j], b)); - outptrs[i][j] += 2; - } - } - } -#endif - for (; channels_remaining; channels_remaining--) - { - // Matrices used and computed during this transform - float F[6][6], FZ[6][4], f[4][4], b; - - // Read a 6x6 tile in the Winograd domain - for (int i = 0, m = 0; i < 6; i++) - { - for (int j = 0; j < 6; j++, m++) - { - F[i][j] = *(inptr + m*matrix_stride); - } - } - inptr++; - - // Compute the matrix F Z - for (int i = 0; i < 6; i++) - { - FZ[i][0] = 1*F[i][0] + 1*F[i][1] + 1*F[i][2] + 1*F[i][3] + 1*F[i][4]; - FZ[i][1] = 1*F[i][1] + -1*F[i][2] + 2*F[i][3] + -2*F[i][4]; - FZ[i][2] = 1*F[i][1] + 1*F[i][2] + 4*F[i][3] + 4*F[i][4]; - FZ[i][3] = 1*F[i][1] + -1*F[i][2] + 8*F[i][3] + -8*F[i][4] + 1*F[i][5]; - } - - // Compute the output tile f = ZT F Z - for (int j = 0; j < 4; j++) - { - f[0][j] = 1*FZ[0][j] + 1*FZ[1][j] + 1*FZ[2][j] + 1*FZ[3][j] + 1*FZ[4][j]; - f[1][j] = 1*FZ[1][j] + -1*FZ[2][j] + 2*FZ[3][j] + -2*FZ[4][j]; - f[2][j] = 1*FZ[1][j] + 1*FZ[2][j] + 4*FZ[3][j] + 4*FZ[4][j]; - f[3][j] = 1*FZ[1][j] + -1*FZ[2][j] + 8*FZ[3][j] + -8*FZ[4][j] + 1*FZ[5][j]; - } - - // Write out the output tile - b = *(bptr++); - for (int i = 0; i < cells_i; i++) - { - for (int j = 0; j < cells_j; j++) - { - *(outptrs[i][j]++) = f[i][j] + b; - } - } - } - } - else - { - // For each channel of the output - int channels_remaining = n_channels; -#ifdef __aarch64__ - for (; channels_remaining >= 4; channels_remaining -= 4) - { - // Matrices used and computed during this transform - float32x4_t F[6][6], FZ[6][4], f[4][4]; - - // Read a 6x6 tile in the Winograd domain - for (int i = 0, m = 0; i < 6; i++) - { - for (int j = 0; j < 6; j++, m++) - { - F[i][j] = vld1q_f32(inptr + m*matrix_stride); - } - } - inptr += 4; - - // Compute the matrix F Z - for (int i = 0; i < 6; i++) - { - // FZ[i][0] = 1*F[i][0] + 1*F[i][1] + 1*F[i][2] + 1*F[i][3] + 1*F[i][4]; - FZ[i][0] = vaddq_f32(vaddq_f32(vaddq_f32(F[i][0], F[i][1]), vaddq_f32(F[i][2], F[i][3])), F[i][4]); - - // FZ[i][1] = 1*F[i][1] + -1*F[i][2] + 2*F[i][3] + -2*F[i][4]; - FZ[i][1] = vmlaq_n_f32(vsubq_f32(F[i][1], F[i][2]), vsubq_f32(F[i][3], F[i][4]), 2.0f); - - // FZ[i][2] = 1*F[i][1] + 1*F[i][2] + 4*F[i][3] + 4*F[i][4]; - FZ[i][2] = vmlaq_n_f32(vaddq_f32(F[i][1], F[i][2]), vaddq_f32(F[i][3], F[i][4]), 4.0f); - - // FZ[i][3] = 1*F[i][1] + -1*F[i][2] + 8*F[i][3] + -8*F[i][4] + 1*F[i][5]; - FZ[i][3] = vaddq_f32(vmlaq_n_f32(vsubq_f32(F[i][1], F[i][2]), vsubq_f32(F[i][3], F[i][4]), 8.0f), F[i][5]); - } - - // Compute the output tile f = ZT F Z - for (int j = 0; j < 4; j++) - { - // f[0][j] = 1*FZ[0][j] + 1*FZ[1][j] + 1*FZ[2][j] + 1*FZ[3][j] + 1*FZ[4][j]; - f[0][j] = vaddq_f32(vaddq_f32(vaddq_f32(FZ[0][j], FZ[1][j]), vaddq_f32(FZ[2][j], FZ[3][j])), FZ[4][j]); - - // f[1][j] = 1*FZ[1][j] + -1*FZ[2][j] + 2*FZ[3][j] + -2*FZ[4][j]; - f[1][j] = vmlaq_n_f32(vsubq_f32(FZ[1][j], FZ[2][j]), vsubq_f32(FZ[3][j], FZ[4][j]), 2.0f); - - // f[2][j] = 1*FZ[1][j] + 1*FZ[2][j] + 4*FZ[3][j] + 4*FZ[4][j]; - f[2][j] = vmlaq_n_f32(vaddq_f32(FZ[1][j], FZ[2][j]), vaddq_f32(FZ[3][j], FZ[4][j]), 4.0f); - - // f[3][j] = 1*FZ[1][j] + -1*FZ[2][j] + 8*FZ[3][j] + -8*FZ[4][j] + 1*FZ[5][j]; - f[3][j] = vaddq_f32(vmlaq_n_f32(vsubq_f32(FZ[1][j], FZ[2][j]), vsubq_f32(FZ[3][j], FZ[4][j]), 8.0f), FZ[5][j]); - } - - // Write out the output tile - for (int i = 0; i < cells_i; i++) - { - for (int j = 0; j < cells_j; j++) - { - vst1q_f32(outptrs[i][j], f[i][j]); - outptrs[i][j] += 4; - } - } - } -#endif // __aarch64__ -#ifdef __arm_any__ - for (; channels_remaining >= 2; channels_remaining -= 2) - { - // Matrices used and computed during this transform - float32x2_t F[6][6], FZ[6][4], f[4][4]; - - // Read a 6x6 tile in the Winograd domain - for (int i = 0, m = 0; i < 6; i++) - { - for (int j = 0; j < 6; j++, m++) - { - F[i][j] = vld1_f32(inptr + m*matrix_stride); - } - } - inptr += 2; - - // Compute the matrix F Z - for (int i = 0; i < 6; i++) - { - // FZ[i][0] = 1*F[i][0] + 1*F[i][1] + 1*F[i][2] + 1*F[i][3] + 1*F[i][4]; - FZ[i][0] = vadd_f32(vadd_f32(vadd_f32(F[i][0], F[i][1]), vadd_f32(F[i][2], F[i][3])), F[i][4]); - - // FZ[i][1] = 1*F[i][1] + -1*F[i][2] + 2*F[i][3] + -2*F[i][4]; - FZ[i][1] = vmla_n_f32(vsub_f32(F[i][1], F[i][2]), vsub_f32(F[i][3], F[i][4]), 2.0f); - - // FZ[i][2] = 1*F[i][1] + 1*F[i][2] + 4*F[i][3] + 4*F[i][4]; - FZ[i][2] = vmla_n_f32(vadd_f32(F[i][1], F[i][2]), vadd_f32(F[i][3], F[i][4]), 4.0f); - - // FZ[i][3] = 1*F[i][1] + -1*F[i][2] + 8*F[i][3] + -8*F[i][4] + 1*F[i][5]; - FZ[i][3] = vadd_f32(vmla_n_f32(vsub_f32(F[i][1], F[i][2]), vsub_f32(F[i][3], F[i][4]), 8.0f), F[i][5]); - } - - // Compute the output tile f = ZT F Z - for (int j = 0; j < 4; j++) - { - // f[0][j] = 1*FZ[0][j] + 1*FZ[1][j] + 1*FZ[2][j] + 1*FZ[3][j] + 1*FZ[4][j]; - f[0][j] = vadd_f32(vadd_f32(vadd_f32(FZ[0][j], FZ[1][j]), vadd_f32(FZ[2][j], FZ[3][j])), FZ[4][j]); - - // f[1][j] = 1*FZ[1][j] + -1*FZ[2][j] + 2*FZ[3][j] + -2*FZ[4][j]; - f[1][j] = vmla_n_f32(vsub_f32(FZ[1][j], FZ[2][j]), vsub_f32(FZ[3][j], FZ[4][j]), 2.0f); - - // f[2][j] = 1*FZ[1][j] + 1*FZ[2][j] + 4*FZ[3][j] + 4*FZ[4][j]; - f[2][j] = vmla_n_f32(vadd_f32(FZ[1][j], FZ[2][j]), vadd_f32(FZ[3][j], FZ[4][j]), 4.0f); - - // f[3][j] = 1*FZ[1][j] + -1*FZ[2][j] + 8*FZ[3][j] + -8*FZ[4][j] + 1*FZ[5][j]; - f[3][j] = vadd_f32(vmla_n_f32(vsub_f32(FZ[1][j], FZ[2][j]), vsub_f32(FZ[3][j], FZ[4][j]), 8.0f), FZ[5][j]); - } - - // Write out the output tile - for (int i = 0; i < cells_i; i++) - { - for (int j = 0; j < cells_j; j++) - { - vst1_f32(outptrs[i][j], f[i][j]); - outptrs[i][j] += 2; - } - } - } -#endif - for (; channels_remaining; channels_remaining--) - { - // Matrices used and computed during this transform - float F[6][6], FZ[6][4], f[4][4]; - - // Read a 6x6 tile in the Winograd domain - for (int i = 0, m = 0; i < 6; i++) - { - for (int j = 0; j < 6; j++, m++) - { - F[i][j] = *(inptr + m*matrix_stride); - } - } - inptr++; - - // Compute the matrix F Z - for (int i = 0; i < 6; i++) - { - FZ[i][0] = 1*F[i][0] + 1*F[i][1] + 1*F[i][2] + 1*F[i][3] + 1*F[i][4]; - FZ[i][1] = 1*F[i][1] + -1*F[i][2] + 2*F[i][3] + -2*F[i][4]; - FZ[i][2] = 1*F[i][1] + 1*F[i][2] + 4*F[i][3] + 4*F[i][4]; - FZ[i][3] = 1*F[i][1] + -1*F[i][2] + 8*F[i][3] + -8*F[i][4] + 1*F[i][5]; - } - - // Compute the output tile f = ZT F Z - for (int j = 0; j < 4; j++) - { - f[0][j] = 1*FZ[0][j] + 1*FZ[1][j] + 1*FZ[2][j] + 1*FZ[3][j] + 1*FZ[4][j]; - f[1][j] = 1*FZ[1][j] + -1*FZ[2][j] + 2*FZ[3][j] + -2*FZ[4][j]; - f[2][j] = 1*FZ[1][j] + 1*FZ[2][j] + 4*FZ[3][j] + 4*FZ[4][j]; - f[3][j] = 1*FZ[1][j] + -1*FZ[2][j] + 8*FZ[3][j] + -8*FZ[4][j] + 1*FZ[5][j]; - } - - // Write out the output tile - for (int i = 0; i < cells_i; i++) - { - for (int j = 0; j < cells_j; j++) - { - *(outptrs[i][j]++) = f[i][j]; - } - } - } - } -} - -} // namespace (anonymous) - -namespace winograd -{ -using Tiles = OutputTransformImplTiles<3, 3, 6, 6, float>; - -template <> -const Tiles::TileFn Tiles::tilefn_generic = winograd_output_transform_4x4_3x3_fp32_process_tile; - -template <> -const Tiles::TileFn Tiles::tilefn_unpadded = winograd_output_transform_4x4_3x3_fp32_process_tile; - -template <> -const Tiles::TileFn Tiles::tilefn_bottom_padded[n_pad_bottom] = { - winograd_output_transform_4x4_3x3_fp32_process_tile, - winograd_output_transform_4x4_3x3_fp32_process_tile, - winograd_output_transform_4x4_3x3_fp32_process_tile, -}; - -template <> -const Tiles::TileFn Tiles::tilefn_right_padded[n_pad_right] = { - winograd_output_transform_4x4_3x3_fp32_process_tile, - winograd_output_transform_4x4_3x3_fp32_process_tile, - winograd_output_transform_4x4_3x3_fp32_process_tile, -}; - -template class OutputTransform<3, 3, 6, 6, float>; -} // namespace winograd diff --git a/src/core/NEON/kernels/convolution/winograd/transforms/output_6_3_fp32.cpp b/src/core/NEON/kernels/convolution/winograd/transforms/output_6_3_fp32.cpp deleted file mode 100644 index 58bed71a47..0000000000 --- a/src/core/NEON/kernels/convolution/winograd/transforms/output_6_3_fp32.cpp +++ /dev/null @@ -1,179 +0,0 @@ -/* - * Copyright (c) 2017 ARM Limited. - * - * SPDX-License-Identifier: MIT - * - * Permission is hereby granted, free of charge, to any person obtaining a copy - * of this software and associated documentation files (the "Software"), to - * deal in the Software without restriction, including without limitation the - * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or - * sell copies of the Software, and to permit persons to whom the Software is - * furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in all - * copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE - * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER - * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, - * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. - */ - -#include "arm_compute/core/NEON/kernels/convolution/winograd/transforms/output.hpp" -#include "arm_compute/core/NEON/kernels/convolution/winograd/winograd_output_transform.hpp" -#include "arm_compute/core/NEON/kernels/convolution/common/arm.hpp" - -namespace -{ - -template -void winograd_output_transform_6_3_fp32_process_tile( - const int n_channels, - const float* const matrix_base, - const int matrix_stride, - const float* const biases, - float* const output, - const int output_row_stride, - const int output_col_stride, - const int _pad_bottom, - const int _pad_right -) -{ - (void) output_row_stride; - (void) _pad_bottom; - constexpr int output_tile_cols = 6; - constexpr int inner_tile_cols = 8; - - const int pad_right = Specialized ? PadRight : _pad_right; - const int cells_j = output_tile_cols - pad_right; - - // Construct a map to the output cells - float *outptrs[cells_j]; - for (int j = 0; j < cells_j; j++) - { - outptrs[j] = output + j*output_col_stride; - } - const float *inptr = matrix_base; - const float *bptr = biases; - - // For each channel of the output - int channels_remaining = n_channels; -#ifdef __arm_any__ - for (; channels_remaining >= 4; channels_remaining -= 4) - { - // Matrices used and computed during this transform - float32x4_t F[inner_tile_cols], f[output_tile_cols], b = vdupq_n_f32(0.0f); - - // Read a 1x8 tile in the Winograd domain - for (int j = 0; j < inner_tile_cols; j++) - { - F[j] = vld1q_f32(inptr + j*matrix_stride); - } - inptr += 4; - - f[0] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(F[6], 1), F[5], 1), F[4], 1), F[3], 1), F[2], 1), F[1], 1), F[0], 1); - f[1] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(F[2], 1), F[6], 3), F[4], 2), F[3], -2), F[5], -3), F[1], -1); - f[2] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(F[2], 1), F[1], 1), F[6], 9), F[5], 9), F[4], 4), F[3], 4); - f[3] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(F[2], 1), F[6], 27), F[4], 8), F[3], -8), F[5], -27), F[1], -1); - f[4] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(F[2], 1), F[1], 1), F[6], 81), F[5], 81), F[4], 16), F[3], 16); - f[5] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(F[7], 1), F[2], 1), F[6], 243), F[4], 32), F[3], -32), F[5], -243), F[1], -1); - - // Write out the output tile - if (bptr != 0) - { - b = vld1q_f32(bptr); - bptr += 4; - } - for (int j = 0; j < cells_j; j++) - { - vst1q_f32(outptrs[j], f[j] + b); - outptrs[j] += 4; - } - } - for (; channels_remaining >= 2; channels_remaining -= 2) - { - // Matrices used and computed during this transform - float32x2_t F[inner_tile_cols], f[output_tile_cols], b = vdup_n_f32(0.0f); - - // Read a 1x8 tile in the Winograd domain - for (int j = 0; j < inner_tile_cols; j++) - { - F[j] = vld1_f32(inptr + j*matrix_stride); - } - inptr += 2; - - f[0] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(F[6], 1), F[5], 1), F[4], 1), F[3], 1), F[2], 1), F[1], 1), F[0], 1); - f[1] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(F[2], 1), F[6], 3), F[4], 2), F[3], -2), F[5], -3), F[1], -1); - f[2] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(F[2], 1), F[1], 1), F[6], 9), F[5], 9), F[4], 4), F[3], 4); - f[3] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(F[2], 1), F[6], 27), F[4], 8), F[3], -8), F[5], -27), F[1], -1); - f[4] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(F[2], 1), F[1], 1), F[6], 81), F[5], 81), F[4], 16), F[3], 16); - f[5] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(F[7], 1), F[2], 1), F[6], 243), F[4], 32), F[3], -32), F[5], -243), F[1], -1); - - // Write out the output tile - if (bptr != 0) - { - b = vld1_f32(bptr); - bptr += 2; - } - for (int j = 0; j < cells_j; j++) - { - vst1_f32(outptrs[j], f[j] + b); - outptrs[j] += 2; - } - } -#endif // __arm_any__ - for (; channels_remaining; channels_remaining--) - { - // Matrices used and computed during this transform - float F[inner_tile_cols], f[output_tile_cols], b = 0.0f; - - // Read a 1x8 tile in the Winograd domain - for (int j = 0; j < inner_tile_cols; j++) - { - F[j] = *(inptr + j*matrix_stride); - } - inptr++; - - f[0] = F[0]*1 + F[1]*1 + F[2]*1 + F[3]*1 + F[4]*1 + F[5]*1 + F[6]*1; - f[1] = F[1]*-1 + F[5]*-3 + F[3]*-2 + F[4]*2 + F[6]*3 + F[2]*1; - f[2] = F[3]*4 + F[4]*4 + F[5]*9 + F[6]*9 + F[1]*1 + F[2]*1; - f[3] = F[1]*-1 + F[5]*-27 + F[3]*-8 + F[4]*8 + F[6]*27 + F[2]*1; - f[4] = F[3]*16 + F[4]*16 + F[5]*81 + F[6]*81 + F[1]*1 + F[2]*1; - f[5] = F[1]*-1 + F[5]*-243 + F[3]*-32 + F[4]*32 + F[6]*243 + F[2]*1 + F[7]*1; - - // Write out the output tile - if (bptr != 0) - { - b = *(bptr++); - } - for (int j = 0; j < cells_j; j++) - { - *(outptrs[j]++) = f[j] + b; - } - } -} - -} // namespace (anonymous) - -namespace winograd -{ -using Tiles = OutputTransformImplTiles<1, 3, 1, 8, float>; - -template <> -const Tiles::TileFn Tiles::tilefn_unpadded = winograd_output_transform_6_3_fp32_process_tile; - -template <> -const Tiles::TileFn Tiles::tilefn_right_padded[n_pad_right] = { - winograd_output_transform_6_3_fp32_process_tile, - winograd_output_transform_6_3_fp32_process_tile, - winograd_output_transform_6_3_fp32_process_tile, - winograd_output_transform_6_3_fp32_process_tile, - winograd_output_transform_6_3_fp32_process_tile, -}; - -template class OutputTransform<1, 3, 1, 8, float>; -template class OutputTransform<3, 1, 8, 1, float>; -} // namespace winograd diff --git a/src/core/NEON/kernels/convolution/winograd/transforms/weights_2_7_fp32.cpp b/src/core/NEON/kernels/convolution/winograd/transforms/weights_2_7_fp32.cpp deleted file mode 100644 index 85cf418656..0000000000 --- a/src/core/NEON/kernels/convolution/winograd/transforms/weights_2_7_fp32.cpp +++ /dev/null @@ -1,124 +0,0 @@ -/* - * Copyright (c) 2017 ARM Limited. - * - * SPDX-License-Identifier: MIT - * - * Permission is hereby granted, free of charge, to any person obtaining a copy - * of this software and associated documentation files (the "Software"), to - * deal in the Software without restriction, including without limitation the - * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or - * sell copies of the Software, and to permit persons to whom the Software is - * furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in all - * copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE - * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER - * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, - * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. - */ - -#include "arm_compute/core/NEON/kernels/convolution/common/arm.hpp" -#include "arm_compute/core/NEON/kernels/convolution/winograd/winograd_gemm.hpp" -#include "arm_compute/core/NEON/kernels/convolution/winograd/transforms/kernel.hpp" - -namespace winograd -{ - template <> - template <> - void WinogradGEMM<1, 2, 1, 7>::WeightsTransform::execute( - const int n_output_channels, - const int n_input_channels, - const float* const input, // NOTE: Data in HWIO order - float* const output, - const int matrix_stride, - const int matrix_row_stride - ) - { - // Get pointers to each cell of the weight tensor - const auto weight_col_stride = n_input_channels * n_output_channels; - const float *inptrs[kernel_cols]; - for (int j = 0; j < kernel_cols; j++) - { - inptrs[j] = input + j*weight_col_stride; - } - - // For each input channel - for (int ic = 0; ic < n_input_channels; ic++) - { - float *outptr = output + ic * matrix_row_stride; - - // For each output channel - int channels_remaining = n_output_channels; - for (; channels_remaining; channels_remaining--) - { - // Matrices used and computed in this kernel - float w[kernel_cols], V[inner_tile_cols]; - - // Read weights - for (int j = 0; j < kernel_cols; j++) - { - w[j] = *(inptrs[j]++); - } - - // Compute V = w WT - V[0] = (w[0]*-1) / 36.0f; - V[1] = (w[1]*-1 + w[3]*-1 + w[5]*-1 + w[0]*1 + w[2]*1 + w[4]*1 + w[6]*1) / 48.0f; - V[2] = (w[0]*1 + w[1]*1 + w[2]*1 + w[3]*1 + w[4]*1 + w[5]*1 + w[6]*1) / 48.0f; - V[3] = (w[0]*-1 + w[6]*-64 + w[4]*-16 + w[2]*-4 + w[1]*2 + w[3]*8 + w[5]*32) / 120.0f; - V[4] = (w[0]*-1 + w[6]*-64 + w[5]*-32 + w[4]*-16 + w[3]*-8 + w[2]*-4 + w[1]*-2) / 120.0f; - V[5] = (w[5]*-243 + w[3]*-27 + w[1]*-3 + w[2]*9 + w[4]*81 + w[6]*729 + w[0]*1) / 720.0f; - V[6] = (w[1]*3 + w[2]*9 + w[3]*27 + w[4]*81 + w[5]*243 + w[6]*729 + w[0]*1) / 720.0f; - V[7] = (w[6]*1) / 1.0f; - - // Store the transformed weights - for (int j = 0; j < inner_tile_cols; j++) - { - *(outptr + j*matrix_stride) = V[j]; - } - outptr++; - } - } - } - - template <> - template <> - int WinogradGEMM<1, 2, 1, 7>::WeightsTransform::ops_performed(const KernelShape &shape) - { - (void) shape; - return 0; // TODO - } - - template <> - template <> - void WinogradGEMM<2, 1, 7, 1>::WeightsTransform::execute( - const int n_output_channels, - const int n_input_channels, - const float* const input, // NOTE: Data in HWIO order - float* const output, - const int matrix_stride, - const int matrix_row_stride - ) - { - // Redirect to the 1xN implementation - WinogradGEMM<1, 2, 1, 7>::template WeightsTransform::execute( - n_output_channels, n_input_channels, input, output, matrix_stride, - matrix_row_stride - ); - } - - template <> - template <> - int WinogradGEMM<2, 1, 7, 1>::WeightsTransform::ops_performed(const KernelShape &shape) - { - (void) shape; - return 0; // TODO - } - - template struct WinogradGEMM<1, 2, 1, 7>::WeightsTransform; - template struct WinogradGEMM<2, 1, 7, 1>::WeightsTransform; -} diff --git a/src/core/NEON/kernels/convolution/winograd/transforms/weights_2x2_3x3_fp32.cpp b/src/core/NEON/kernels/convolution/winograd/transforms/weights_2x2_3x3_fp32.cpp deleted file mode 100644 index 6c71461f81..0000000000 --- a/src/core/NEON/kernels/convolution/winograd/transforms/weights_2x2_3x3_fp32.cpp +++ /dev/null @@ -1,228 +0,0 @@ -/* - * Copyright (c) 2017 ARM Limited. - * - * SPDX-License-Identifier: MIT - * - * Permission is hereby granted, free of charge, to any person obtaining a copy - * of this software and associated documentation files (the "Software"), to - * deal in the Software without restriction, including without limitation the - * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or - * sell copies of the Software, and to permit persons to whom the Software is - * furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in all - * copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE - * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER - * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, - * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. - */ - -#include "arm_compute/core/NEON/kernels/convolution/common/arm.hpp" -#include "arm_compute/core/NEON/kernels/convolution/winograd/winograd_gemm.hpp" -#include "arm_compute/core/NEON/kernels/convolution/winograd/transforms/kernel.hpp" - -namespace winograd -{ - template <> - template <> - void WinogradGEMM<2, 2, 3, 3>::WeightsTransform::execute( - const int n_output_channels, - const int n_input_channels, - const float* const input, - float* const output, - const int matrix_stride, - const int matrix_row_stride - ) - { - constexpr int inner_tile_i = 4; - constexpr int inner_tile_j = 4; - - // Get pointers to each cell of the weight tensor - const auto weight_col_stride = n_input_channels * n_output_channels; - const auto weight_row_stride = 3 * weight_col_stride; - const float *inptrs[3][3]; - for (int i = 0; i < 3; i++) - { - for (int j = 0; j < 3; j++) - { - inptrs[i][j] = input + i*weight_row_stride + j*weight_col_stride; - } - } - - // For each input channel - for (int ic = 0; ic < n_input_channels; ic++) - { - float *outptr = output + ic * matrix_row_stride; - - // For each output channel - int channels_remaining = n_output_channels; -#ifdef __aarch64__ - for (; channels_remaining >= 4; channels_remaining -= 4) - { - // Matrices used and computed in this kernel - float32x4_t w[3][3], Ww[inner_tile_i][3], V[inner_tile_i][inner_tile_j]; - - // Read weights - for (int i = 0; i < 3; i++) - { - for (int j = 0; j < 3; j++) - { - w[i][j] = vld1q_f32(inptrs[i][j]); - inptrs[i][j] += 4; - } - } - - // Compute the matrix W w - for (int j = 0; j < 3; j++) - { - Ww[0][j] = w[0][j]; - - // Ww[1][j] = 0.5*(w[0][j] + w[1][j] + w[2][j]); - Ww[1][j] = vmulq_n_f32(vaddq_f32(vaddq_f32(w[0][j], w[1][j]), w[2][j]), 0.5f); - - // Ww[2][j] = 0.5*(w[0][j] - w[1][j] + w[2][j]); - Ww[2][j] = vmulq_n_f32(vaddq_f32(vsubq_f32(w[0][j], w[1][j]), w[2][j]), 0.5f); - - Ww[3][j] = w[2][j]; - } - - // Compute V = W w WT - for (int i = 0; i < inner_tile_i; i++) - { - V[i][0] = Ww[i][0]; - - // V[i][1] = 0.5*(Ww[i][0] + Ww[i][1] + Ww[i][2]); - V[i][1] = vmulq_n_f32(vaddq_f32(vaddq_f32(Ww[i][0], Ww[i][1]), Ww[i][2]), 0.5f); - - // V[i][2] = 0.5*(Ww[i][0] - Ww[i][1] + Ww[i][2]); - V[i][2] = vmulq_n_f32(vaddq_f32(vsubq_f32(Ww[i][0], Ww[i][1]), Ww[i][2]), 0.5f); - - V[i][3] = Ww[i][2]; - } - - // Store the transformed weights - for (int i = 0, m = 0; i < inner_tile_i; i++) - { - for (int j = 0; j < inner_tile_j; j++, m++) - { - vst1q_f32(outptr + m*matrix_stride, V[i][j]); - } - } - outptr += 4; - } -#endif // __aarch64__ -#ifdef __arm_any__ - for (; channels_remaining >= 2; channels_remaining -= 2) - { - // Matrices used and computed in this kernel - float32x2_t w[3][3], Ww[inner_tile_i][3], V[inner_tile_i][inner_tile_j]; - - // Read weights - for (int i = 0; i < 3; i++) - { - for (int j = 0; j < 3; j++) - { - w[i][j] = vld1_f32(inptrs[i][j]); - inptrs[i][j] += 2; - } - } - - // Compute the matrix W w - for (int j = 0; j < 3; j++) - { - Ww[0][j] = w[0][j]; - - // Ww[1][j] = 0.5*(w[0][j] + w[1][j] + w[2][j]); - Ww[1][j] = vmul_n_f32(vadd_f32(vadd_f32(w[0][j], w[1][j]), w[2][j]), 0.5f); - - // Ww[2][j] = 0.5*(w[0][j] - w[1][j] + w[2][j]); - Ww[2][j] = vmul_n_f32(vadd_f32(vsub_f32(w[0][j], w[1][j]), w[2][j]), 0.5f); - - Ww[3][j] = w[2][j]; - } - - // Compute V = W w WT - for (int i = 0; i < inner_tile_i; i++) - { - V[i][0] = Ww[i][0]; - - // V[i][1] = 0.5*(Ww[i][0] + Ww[i][1] + Ww[i][2]); - V[i][1] = vmul_n_f32(vadd_f32(vadd_f32(Ww[i][0], Ww[i][1]), Ww[i][2]), 0.5f); - - // V[i][2] = 0.5*(Ww[i][0] - Ww[i][1] + Ww[i][2]); - V[i][2] = vmul_n_f32(vadd_f32(vsub_f32(Ww[i][0], Ww[i][1]), Ww[i][2]), 0.5f); - - V[i][3] = Ww[i][2]; - } - - // Store the transformed weights - for (int i = 0, m = 0; i < inner_tile_i; i++) - { - for (int j = 0; j < inner_tile_j; j++, m++) - { - vst1_f32(outptr + m*matrix_stride, V[i][j]); - } - } - outptr += 2; - } -#endif // __arm_any__ - for (; channels_remaining; channels_remaining--) - { - // Matrices used and computed in this kernel - float w[3][3], Ww[inner_tile_i][3], V[inner_tile_i][inner_tile_j]; - - // Read weights - for (int i = 0; i < 3; i++) - { - for (int j = 0; j < 3; j++) - { - w[i][j] = *(inptrs[i][j]++); - } - } - - // Compute the matrix W w - for (int j = 0; j < 3; j++) - { - Ww[0][j] = w[0][j]; - Ww[1][j] = 0.5*(w[0][j] + w[1][j] + w[2][j]); - Ww[2][j] = 0.5*(w[0][j] - w[1][j] + w[2][j]); - Ww[3][j] = w[2][j]; - } - - // Compute V = W w WT - for (int i = 0; i < inner_tile_i; i++) - { - V[i][0] = Ww[i][0]; - V[i][1] = 0.5*(Ww[i][0] + Ww[i][1] + Ww[i][2]); - V[i][2] = 0.5*(Ww[i][0] - Ww[i][1] + Ww[i][2]); - V[i][3] = Ww[i][2]; - } - - // Store the transformed weights - for (int i = 0, m = 0; i < inner_tile_i; i++) - { - for (int j = 0; j < inner_tile_j; j++, m++) - { - *(outptr + m*matrix_stride) = V[i][j]; - } - } - outptr++; - } - } - } - - template <> - template <> - int WinogradGEMM<2, 2, 3, 3>::WeightsTransform::ops_performed(const KernelShape &shape) - { - const int channel_prod = shape.n_input_channels * shape.n_output_channels; - return 2 * 18 * channel_prod; - } - - template struct WinogradGEMM<2, 2, 3, 3>::WeightsTransform; -} // namespace winograd diff --git a/src/core/NEON/kernels/convolution/winograd/transforms/weights_2x2_5x5_fp32.cpp b/src/core/NEON/kernels/convolution/winograd/transforms/weights_2x2_5x5_fp32.cpp deleted file mode 100644 index 2f4f6e1ba2..0000000000 --- a/src/core/NEON/kernels/convolution/winograd/transforms/weights_2x2_5x5_fp32.cpp +++ /dev/null @@ -1,408 +0,0 @@ -/* - * Copyright (c) 2017 ARM Limited. - * - * SPDX-License-Identifier: MIT - * - * Permission is hereby granted, free of charge, to any person obtaining a copy - * of this software and associated documentation files (the "Software"), to - * deal in the Software without restriction, including without limitation the - * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or - * sell copies of the Software, and to permit persons to whom the Software is - * furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in all - * copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE - * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER - * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, - * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. - */ - -#include "arm_compute/core/NEON/kernels/convolution/common/arm.hpp" -#include "arm_compute/core/NEON/kernels/convolution/winograd/winograd_gemm.hpp" -#include "arm_compute/core/NEON/kernels/convolution/winograd/transforms/kernel.hpp" - -namespace winograd -{ - template <> - template <> - void WinogradGEMM<2, 2, 5, 5>::WeightsTransform::execute( - const int n_output_channels, - const int n_input_channels, - const float* const input, - float* const output, - const int matrix_stride, - const int matrix_row_stride - ) - { - // Get pointers to each cell of the weight tensor - const auto weight_col_stride = n_input_channels * n_output_channels; - const auto weight_row_stride = 5 * weight_col_stride; - const float *inptrs[5][5]; - for (int i = 0; i < 5; i++) - { - for (int j = 0; j < 5; j++) - { - inptrs[i][j] = input + i*weight_row_stride + j*weight_col_stride; - } - } - - // For each input channel - for (int ic = 0; ic < n_input_channels; ic++) - { - float *outptr = output + ic * matrix_row_stride; - - // For each output channel - int channels_remaining = n_output_channels; -#ifdef __aarch64__ - for (; channels_remaining >= 4; channels_remaining -= 4) - { - // Matrices used and computed in this kernel - float32x4_t w[5][5], Ww[6][5], V[6][6]; - - // Read weights - for (int i = 0; i < 5; i++) - { - for (int j = 0; j < 5; j++) - { - w[i][j] = vld1q_f32(inptrs[i][j]); - inptrs[i][j] += 4; - } - } - - // Compute the matrix W w - for (int j = 0; j < 5; j++) - { - // Ww[0][j] = w[0][j]/4.0f; - Ww[0][j] = vmulq_n_f32(w[0][j], 1.0f/4.0f); - - // Ww[1][j] = -( w[0][j] + w[1][j] + w[2][j] + w[3][j] + w[4][j])/6.0f; - Ww[1][j] = vmulq_n_f32( - vaddq_f32( - vaddq_f32( - vaddq_f32(w[1][j], w[0][j]), - vaddq_f32(w[3][j], w[2][j]) - ), - w[4][j] - ), - -1.0f/6.0f - ); - - // Ww[2][j] = +(-w[0][j] + w[1][j] - w[2][j] + w[3][j] - w[4][j])/6.0f; - // Ww[2][j] = ((w[1][j] - w[0][j]) + (w[3][j] - w[2][j]) - w[4][j])/6.0f; - Ww[2][j] = vmulq_n_f32( - vsubq_f32( - vaddq_f32( - vsubq_f32(w[1][j], w[0][j]), - vsubq_f32(w[3][j], w[2][j]) - ), - w[4][j] - ), - 1.0f/6.0f - ); - - // Ww[3][j] = (w[0][j]/8.0f + w[1][j]/4.0f + w[2][j]/2.0f + w[3][j] + 2*w[4][j])/3.0f; - Ww[3][j] = vmulq_n_f32( - vmlaq_n_f32( - vaddq_f32( - vaddq_f32(vmulq_n_f32(w[0][j], 1.0f/8.0f), vmulq_n_f32(w[1][j], 1.0f/4.0f)), - vaddq_f32(vmulq_n_f32(w[2][j], 1.0f/2.0f), w[3][j]) - ), - w[4][j], 2.0f - ), - 1.0f/3.0f - ); - - // Ww[4][j] = (w[0][j]/8.0f - w[1][j]/4.0f + w[2][j]/2.0f - w[3][j] + 2*w[4][j])/3.0f; - Ww[4][j] = vmulq_n_f32( - vmlaq_n_f32( - vaddq_f32( - vsubq_f32(vmulq_n_f32(w[0][j], 1.0f/8.0f), vmulq_n_f32(w[1][j], 1.0f/4.0f)), - vsubq_f32(vmulq_n_f32(w[2][j], 1.0f/2.0f), w[3][j]) - ), - w[4][j], 2.0f - ), - 1.0f/3.0f - ); - - // Ww[5][j] = w[4][j]; - Ww[5][j] = w[4][j]; - } - - // Compute V = W w WT - for (int i = 0; i < 6; i++) - { - // V[i][0] = Ww[i][0]/4.0f; - V[i][0] = vmulq_n_f32(Ww[i][0], 1.0f/4.0f); - - // V[i][1] = -( Ww[i][0] + Ww[i][1] + Ww[i][2] + Ww[i][3] + Ww[i][4])/6.0f; - V[i][1] = vmulq_n_f32( - vaddq_f32( - vaddq_f32( - vaddq_f32(Ww[i][1], Ww[i][0]), - vaddq_f32(Ww[i][3], Ww[i][2]) - ), - Ww[i][4] - ), - -1.0f/6.0f - ); - - // V[i][2] = +(-Ww[i][0] + Ww[i][1] - Ww[i][2] + Ww[i][3] - Ww[i][4])/6.0f; - // V[i][2] = ((Ww[i][1] - Ww[i][0]) + (Ww[i][3] - Ww[i][2]) - Ww[i][4])/6.0f; - V[i][2] = vmulq_n_f32( - vsubq_f32( - vaddq_f32( - vsubq_f32(Ww[i][1], Ww[i][0]), - vsubq_f32(Ww[i][3], Ww[i][2]) - ), - Ww[i][4] - ), - 1.0f/6.0f - ); - - // V[i][3] = (Ww[i][0]/8.0f + Ww[i][1]/4.0f + Ww[i][2]/2.0f + Ww[i][3] + 2*Ww[i][4])/3.0f; - V[i][3] = vmulq_n_f32( - vmlaq_n_f32( - vaddq_f32( - vaddq_f32(vmulq_n_f32(Ww[i][0], 1.0f/8.0f), vmulq_n_f32(Ww[i][1], 1.0f/4.0f)), - vaddq_f32(vmulq_n_f32(Ww[i][2], 1.0f/2.0f), Ww[i][3]) - ), - Ww[i][4], 2.0f - ), - 1.0f/3.0f - ); - - // V[i][4] = (Ww[i][0]/8.0f - Ww[i][1]/4.0f + Ww[i][2]/2.0f - Ww[i][3] + 2*Ww[i][4])/3.0f; - V[i][4] = vmulq_n_f32( - vmlaq_n_f32( - vaddq_f32( - vsubq_f32(vmulq_n_f32(Ww[i][0], 1.0f/8.0f), vmulq_n_f32(Ww[i][1], 1.0f/4.0f)), - vsubq_f32(vmulq_n_f32(Ww[i][2], 1.0f/2.0f), Ww[i][3]) - ), - Ww[i][4], 2.0f - ), - 1.0f/3.0f - ); - - // V[i][5] = Ww[i][4]; - V[i][5] = Ww[i][4]; - } - - // Store the transformed weights - for (int i = 0, m = 0; i < 6; i++) - { - for (int j = 0; j < 6; j++, m++) - { - vst1q_f32(outptr + m*matrix_stride, V[i][j]); - } - } - outptr += 4; - } -#endif // __aarch64__ -#ifdef __arm_any__ - for (; channels_remaining >= 2; channels_remaining -= 2) - { - // Matrices used and computed in this kernel - float32x2_t w[5][5], Ww[6][5], V[6][6]; - - // Read weights - for (int i = 0; i < 5; i++) - { - for (int j = 0; j < 5; j++) - { - w[i][j] = vld1_f32(inptrs[i][j]); - inptrs[i][j] += 2; - } - } - - // Compute the matrix W w - for (int j = 0; j < 5; j++) - { - // Ww[0][j] = w[0][j]/4.0f; - Ww[0][j] = vmul_n_f32(w[0][j], 1.0f/4.0f); - - // Ww[1][j] = -( w[0][j] + w[1][j] + w[2][j] + w[3][j] + w[4][j])/6.0f; - Ww[1][j] = vmul_n_f32( - vadd_f32( - vadd_f32( - vadd_f32(w[1][j], w[0][j]), - vadd_f32(w[3][j], w[2][j]) - ), - w[4][j] - ), - -1.0f/6.0f - ); - - // Ww[2][j] = +(-w[0][j] + w[1][j] - w[2][j] + w[3][j] - w[4][j])/6.0f; - // Ww[2][j] = ((w[1][j] - w[0][j]) + (w[3][j] - w[2][j]) - w[4][j])/6.0f; - Ww[2][j] = vmul_n_f32( - vsub_f32( - vadd_f32( - vsub_f32(w[1][j], w[0][j]), - vsub_f32(w[3][j], w[2][j]) - ), - w[4][j] - ), - 1.0f/6.0f - ); - - // Ww[3][j] = (w[0][j]/8.0f + w[1][j]/4.0f + w[2][j]/2.0f + w[3][j] + 2*w[4][j])/3.0f; - Ww[3][j] = vmul_n_f32( - vmla_n_f32( - vadd_f32( - vadd_f32(vmul_n_f32(w[0][j], 1.0f/8.0f), vmul_n_f32(w[1][j], 1.0f/4.0f)), - vadd_f32(vmul_n_f32(w[2][j], 1.0f/2.0f), w[3][j]) - ), - w[4][j], 2.0f - ), - 1.0f/3.0f - ); - - // Ww[4][j] = (w[0][j]/8.0f - w[1][j]/4.0f + w[2][j]/2.0f - w[3][j] + 2*w[4][j])/3.0f; - Ww[4][j] = vmul_n_f32( - vmla_n_f32( - vadd_f32( - vsub_f32(vmul_n_f32(w[0][j], 1.0f/8.0f), vmul_n_f32(w[1][j], 1.0f/4.0f)), - vsub_f32(vmul_n_f32(w[2][j], 1.0f/2.0f), w[3][j]) - ), - w[4][j], 2.0f - ), - 1.0f/3.0f - ); - - // Ww[5][j] = w[4][j]; - Ww[5][j] = w[4][j]; - } - - // Compute V = W w WT - for (int i = 0; i < 6; i++) - { - // V[i][0] = Ww[i][0]/4.0f; - V[i][0] = vmul_n_f32(Ww[i][0], 1.0f/4.0f); - - // V[i][1] = -( Ww[i][0] + Ww[i][1] + Ww[i][2] + Ww[i][3] + Ww[i][4])/6.0f; - V[i][1] = vmul_n_f32( - vadd_f32( - vadd_f32( - vadd_f32(Ww[i][1], Ww[i][0]), - vadd_f32(Ww[i][3], Ww[i][2]) - ), - Ww[i][4] - ), - -1.0f/6.0f - ); - - // V[i][2] = +(-Ww[i][0] + Ww[i][1] - Ww[i][2] + Ww[i][3] - Ww[i][4])/6.0f; - // V[i][2] = ((Ww[i][1] - Ww[i][0]) + (Ww[i][3] - Ww[i][2]) - Ww[i][4])/6.0f; - V[i][2] = vmul_n_f32( - vsub_f32( - vadd_f32( - vsub_f32(Ww[i][1], Ww[i][0]), - vsub_f32(Ww[i][3], Ww[i][2]) - ), - Ww[i][4] - ), - 1.0f/6.0f - ); - - // V[i][3] = (Ww[i][0]/8.0f + Ww[i][1]/4.0f + Ww[i][2]/2.0f + Ww[i][3] + 2*Ww[i][4])/3.0f; - V[i][3] = vmul_n_f32( - vmla_n_f32( - vadd_f32( - vadd_f32(vmul_n_f32(Ww[i][0], 1.0f/8.0f), vmul_n_f32(Ww[i][1], 1.0f/4.0f)), - vadd_f32(vmul_n_f32(Ww[i][2], 1.0f/2.0f), Ww[i][3]) - ), - Ww[i][4], 2.0f - ), - 1.0f/3.0f - ); - - // V[i][4] = (Ww[i][0]/8.0f - Ww[i][1]/4.0f + Ww[i][2]/2.0f - Ww[i][3] + 2*Ww[i][4])/3.0f; - V[i][4] = vmul_n_f32( - vmla_n_f32( - vadd_f32( - vsub_f32(vmul_n_f32(Ww[i][0], 1.0f/8.0f), vmul_n_f32(Ww[i][1], 1.0f/4.0f)), - vsub_f32(vmul_n_f32(Ww[i][2], 1.0f/2.0f), Ww[i][3]) - ), - Ww[i][4], 2.0f - ), - 1.0f/3.0f - ); - - // V[i][5] = Ww[i][4]; - V[i][5] = Ww[i][4]; - } - - // Store the transformed weights - for (int i = 0, m = 0; i < 6; i++) - { - for (int j = 0; j < 6; j++, m++) - { - vst1_f32(outptr + m*matrix_stride, V[i][j]); - } - } - outptr += 2; - } -#endif // __arm_any__ - for (; channels_remaining; channels_remaining--) - { - // Matrices used and computed in this kernel - float w[5][5], Ww[6][5], V[6][6]; - - // Read weights - for (int i = 0; i < 5; i++) - { - for (int j = 0; j < 5; j++) - { - w[i][j] = *(inptrs[i][j]++); - } - } - - // Compute the matrix W w - for (int j = 0; j < 5; j++) - { - Ww[0][j] = w[0][j]/4.0f; - Ww[1][j] = -( w[0][j] + w[1][j] + w[2][j] + w[3][j] + w[4][j])/6.0f; - Ww[2][j] = +(-w[0][j] + w[1][j] - w[2][j] + w[3][j] - w[4][j])/6.0f; - Ww[3][j] = (w[0][j]/8.0f + w[1][j]/4.0f + w[2][j]/2.0f + w[3][j] + 2*w[4][j])/3.0f; - Ww[4][j] = (w[0][j]/8.0f - w[1][j]/4.0f + w[2][j]/2.0f - w[3][j] + 2*w[4][j])/3.0f; - Ww[5][j] = w[4][j]; - } - - // Compute V = W w WT - for (int i = 0; i < 6; i++) - { - V[i][0] = Ww[i][0]/4.0f; - V[i][1] = -( Ww[i][0] + Ww[i][1] + Ww[i][2] + Ww[i][3] + Ww[i][4])/6.0f; - V[i][2] = +(-Ww[i][0] + Ww[i][1] - Ww[i][2] + Ww[i][3] - Ww[i][4])/6.0f; - V[i][3] = (Ww[i][0]/8.0f + Ww[i][1]/4.0f + Ww[i][2]/2.0f + Ww[i][3] + 2*Ww[i][4])/3.0f; - V[i][4] = (Ww[i][0]/8.0f - Ww[i][1]/4.0f + Ww[i][2]/2.0f - Ww[i][3] + 2*Ww[i][4])/3.0f; - V[i][5] = Ww[i][4]; - } - - // Store the transformed weights - for (int i = 0, m = 0; i < 6; i++) - { - for (int j = 0; j < 6; j++, m++) - { - *(outptr + m*matrix_stride) = V[i][j]; - } - } - outptr++; - } - } - } - - template <> - template <> - int WinogradGEMM<2, 2, 5, 5>::WeightsTransform::ops_performed(const KernelShape &shape) - { - return 0; // TODO - } - - template class WinogradGEMM<2, 2, 5, 5>::WeightsTransform; -} // namespace winograd diff --git a/src/core/NEON/kernels/convolution/winograd/transforms/weights_4_5_fp32.cpp b/src/core/NEON/kernels/convolution/winograd/transforms/weights_4_5_fp32.cpp deleted file mode 100644 index 2f14e20142..0000000000 --- a/src/core/NEON/kernels/convolution/winograd/transforms/weights_4_5_fp32.cpp +++ /dev/null @@ -1,124 +0,0 @@ -/* - * Copyright (c) 2017 ARM Limited. - * - * SPDX-License-Identifier: MIT - * - * Permission is hereby granted, free of charge, to any person obtaining a copy - * of this software and associated documentation files (the "Software"), to - * deal in the Software without restriction, including without limitation the - * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or - * sell copies of the Software, and to permit persons to whom the Software is - * furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in all - * copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE - * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER - * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, - * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. - */ - -#include "arm_compute/core/NEON/kernels/convolution/common/arm.hpp" -#include "arm_compute/core/NEON/kernels/convolution/winograd/winograd_gemm.hpp" -#include "arm_compute/core/NEON/kernels/convolution/winograd/transforms/kernel.hpp" - -namespace winograd -{ - template <> - template <> - void WinogradGEMM<1, 4, 1, 5>::WeightsTransform::execute( - const int n_output_channels, - const int n_input_channels, - const float* const input, // NOTE: Data in HWIO order - float* const output, - const int matrix_stride, - const int matrix_row_stride - ) - { - // Get pointers to each cell of the weight tensor - const auto weight_col_stride = n_input_channels * n_output_channels; - const float *inptrs[kernel_cols]; - for (int j = 0; j < kernel_cols; j++) - { - inptrs[j] = input + j*weight_col_stride; - } - - // For each input channel - for (int ic = 0; ic < n_input_channels; ic++) - { - float *outptr = output + ic * matrix_row_stride; - - // For each output channel - int channels_remaining = n_output_channels; - for (; channels_remaining; channels_remaining--) - { - // Matrices used and computed in this kernel - float w[kernel_cols], V[inner_tile_cols]; - - // Read weights - for (int j = 0; j < kernel_cols; j++) - { - w[j] = *(inptrs[j]++); - } - - // Compute V = w WT - V[0] = (w[0]*-1) / 36; - V[1] = (w[1]*-1 + w[3]*-1 + w[0]*1 + w[2]*1 + w[4]*1) / 48; - V[2] = (w[0]*1 + w[1]*1 + w[2]*1 + w[3]*1 + w[4]*1) / 48; - V[3] = (w[0]*-1 + w[4]*-16 + w[2]*-4 + w[1]*2 + w[3]*8) / 120; - V[4] = (w[0]*-1 + w[4]*-16 + w[3]*-8 + w[2]*-4 + w[1]*-2) / 120; - V[5] = (w[3]*-27 + w[1]*-3 + w[2]*9 + w[4]*81 + w[0]*1) / 720; - V[6] = (w[1]*3 + w[2]*9 + w[3]*27 + w[4]*81 + w[0]*1) / 720; - V[7] = (w[4]*1) / 1; - - // Store the transformed weights - for (int j = 0; j < inner_tile_cols; j++) - { - *(outptr + j*matrix_stride) = V[j]; - } - outptr++; - } - } - } - - template <> - template <> - int WinogradGEMM<1, 4, 1, 5>::WeightsTransform::ops_performed(const KernelShape &shape) - { - (void) shape; - return 0; // TODO - } - - template <> - template <> - void WinogradGEMM<4, 1, 5, 1>::WeightsTransform::execute( - const int n_output_channels, - const int n_input_channels, - const float* const input, // NOTE: Data in HWIO order - float* const output, - const int matrix_stride, - const int matrix_row_stride - ) - { - // Redirect to the 1xN implementation - WinogradGEMM<1, 4, 1, 5>::template WeightsTransform::execute( - n_output_channels, n_input_channels, input, output, matrix_stride, - matrix_row_stride - ); - } - - template <> - template <> - int WinogradGEMM<4, 1, 5, 1>::WeightsTransform::ops_performed(const KernelShape &shape) - { - (void) shape; - return 0; // TODO - } - - template struct WinogradGEMM<1, 4, 1, 5>::WeightsTransform; - template struct WinogradGEMM<4, 1, 5, 1>::WeightsTransform; -} diff --git a/src/core/NEON/kernels/convolution/winograd/transforms/weights_4x4_3x3_fp32.cpp b/src/core/NEON/kernels/convolution/winograd/transforms/weights_4x4_3x3_fp32.cpp deleted file mode 100644 index a56a475fc9..0000000000 --- a/src/core/NEON/kernels/convolution/winograd/transforms/weights_4x4_3x3_fp32.cpp +++ /dev/null @@ -1,266 +0,0 @@ -/* - * Copyright (c) 2017 ARM Limited. - * - * SPDX-License-Identifier: MIT - * - * Permission is hereby granted, free of charge, to any person obtaining a copy - * of this software and associated documentation files (the "Software"), to - * deal in the Software without restriction, including without limitation the - * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or - * sell copies of the Software, and to permit persons to whom the Software is - * furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in all - * copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE - * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER - * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, - * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. - */ - -#include "arm_compute/core/NEON/kernels/convolution/common/arm.hpp" -#include "arm_compute/core/NEON/kernels/convolution/winograd/winograd_gemm.hpp" -#include "arm_compute/core/NEON/kernels/convolution/winograd/transforms/kernel.hpp" - -namespace winograd -{ - /* Float implementation for kernel transform F(4x4, 3x3) */ - template <> - template <> - void WinogradGEMM<4, 4, 3, 3>::WeightsTransform::execute( - const int n_output_channels, - const int n_input_channels, - const float* const input, // NOTE: Data in HWIO order - float* const output, - const int matrix_stride, - const int matrix_row_stride - ) - { - // Get pointers to each cell of the weight tensor - const auto weight_col_stride = n_input_channels * n_output_channels; - const auto weight_row_stride = 3 * weight_col_stride; - const float *inptrs[3][3]; - for (int i = 0; i < 3; i++) - { - for (int j = 0; j < 3; j++) - { - inptrs[i][j] = input + i*weight_row_stride + j*weight_col_stride; - } - } - - // For each input channel - for (int ic = 0; ic < n_input_channels; ic++) - { - float *outptr = output + ic * matrix_row_stride; - - // For each output channel - int channels_remaining = n_output_channels; -#ifdef __aarch64__ - for (; channels_remaining >= 4; channels_remaining -= 4) - { - // Matrices used and computed in this kernel - float32x4_t w[3][3], Ww[6][3], V[6][6]; - - // Read weights - for (int i = 0; i < 3; i++) - { - for (int j = 0; j < 3; j++) - { - w[i][j] = vld1q_f32(inptrs[i][j]); - inptrs[i][j] += 4; - } - } - - // Compute the matrix W w - for (int j = 0; j < 3; j++) - { - // Ww[0][j] = 6*w[0][j]; - Ww[0][j] = vmulq_n_f32(w[0][j], 6.0); - - // Ww[1][j] = -4*w[0][j] + -4*w[1][j] + -4*w[2][j]; - Ww[1][j] = vmulq_n_f32(vaddq_f32(vaddq_f32(w[0][j], w[1][j]), w[2][j]), -4.0); - - // Ww[2][j] = -4*w[0][j] + 4*w[1][j] + -4*w[2][j]; - Ww[2][j] = vmulq_n_f32(vsubq_f32(vsubq_f32(w[1][j], w[0][j]), w[2][j]), 4.0); - - // Ww[3][j] = 1*w[0][j] + 2*w[1][j] + 4*w[2][j]; - Ww[3][j] = vmlaq_n_f32(vmlaq_n_f32(w[0][j], w[1][j], 2.0f), w[2][j], 4.0f); - - // Ww[4][j] = 1*w[0][j] + -2*w[1][j] + 4*w[2][j]; - Ww[4][j] = vmlaq_n_f32(vmlsq_n_f32(w[0][j], w[1][j], 2.0f), w[2][j], 4.0f); - - // Ww[5][j] = 24*w[2][j]; - Ww[5][j] = vmulq_n_f32(w[2][j], 24.0f); - } - - // Compute V = W w WT - for (int i = 0; i < 6; i++) - { - const float recip576 = 1.0f / 576.0f; - - // V[i][0] = 6*Ww[i][0]; - V[i][0] = vmulq_n_f32(vmulq_n_f32(Ww[i][0], 6.0), recip576); - - // V[i][1] = -4*Ww[i][0] + -4*Ww[i][1] + -4*Ww[i][2]; - V[i][1] = vmulq_n_f32(vmulq_n_f32(vaddq_f32(vaddq_f32(Ww[i][0], Ww[i][1]), Ww[i][2]), -4.0), recip576); - - // V[i][2] = -4*Ww[i][0] + 4*Ww[i][1] + -4*Ww[i][2]; - V[i][2] = vmulq_n_f32(vmulq_n_f32(vsubq_f32(vsubq_f32(Ww[i][1], Ww[i][0]), Ww[i][2]), 4.0), recip576); - - // V[i][3] = 1*Ww[i][0] + 2*Ww[i][1] + 4*Ww[i][2]; - V[i][3] = vmulq_n_f32(vmlaq_n_f32(vmlaq_n_f32(Ww[i][0], Ww[i][1], 2.0f), Ww[i][2], 4.0f), recip576); - - // V[i][4] = 1*Ww[i][0] + -2*Ww[i][1] + 4*Ww[i][2]; - V[i][4] = vmulq_n_f32(vmlaq_n_f32(vmlsq_n_f32(Ww[i][0], Ww[i][1], 2.0f), Ww[i][2], 4.0f), recip576); - - // V[i][5] = 24*Ww[i][2]; - V[i][5] = vmulq_n_f32(vmulq_n_f32(Ww[i][2], 24.0f), recip576); - } - - // Store the transformed weights - for (int i = 0, m = 0; i < 6; i++) - { - for (int j = 0; j < 6; j++, m++) - { - vst1q_f32(outptr + m*matrix_stride, V[i][j]); - } - } - outptr += 4; - } -#endif // __aarch64__ -#ifdef __arm_any__ - for (; channels_remaining >= 2; channels_remaining -= 2) - { - // Matrices used and computed in this kernel - float32x2_t w[3][3], Ww[6][3], V[6][6]; - - // Read weights - for (int i = 0; i < 3; i++) - { - for (int j = 0; j < 3; j++) - { - w[i][j] = vld1_f32(inptrs[i][j]); - inptrs[i][j] += 2; - } - } - - // Compute the matrix W w - for (int j = 0; j < 3; j++) - { - // Ww[0][j] = 6*w[0][j]; - Ww[0][j] = vmul_n_f32(w[0][j], 6.0); - - // Ww[1][j] = -4*w[0][j] + -4*w[1][j] + -4*w[2][j]; - Ww[1][j] = vmul_n_f32(vadd_f32(vadd_f32(w[0][j], w[1][j]), w[2][j]), -4.0); - - // Ww[2][j] = -4*w[0][j] + 4*w[1][j] + -4*w[2][j]; - Ww[2][j] = vmul_n_f32(vsub_f32(vsub_f32(w[1][j], w[0][j]), w[2][j]), 4.0); - - // Ww[3][j] = 1*w[0][j] + 2*w[1][j] + 4*w[2][j]; - Ww[3][j] = vmla_n_f32(vmla_n_f32(w[0][j], w[1][j], 2.0f), w[2][j], 4.0f); - - // Ww[4][j] = 1*w[0][j] + -2*w[1][j] + 4*w[2][j]; - Ww[4][j] = vmla_n_f32(vmls_n_f32(w[0][j], w[1][j], 2.0f), w[2][j], 4.0f); - - // Ww[5][j] = 24*w[2][j]; - Ww[5][j] = vmul_n_f32(w[2][j], 24.0f); - } - - // Compute V = W w WT - for (int i = 0; i < 6; i++) - { - const float recip576 = 1.0f / 576.0f; - - // V[i][0] = 6*Ww[i][0]; - V[i][0] = vmul_n_f32(vmul_n_f32(Ww[i][0], 6.0), recip576); - - // V[i][1] = -4*Ww[i][0] + -4*Ww[i][1] + -4*Ww[i][2]; - V[i][1] = vmul_n_f32(vmul_n_f32(vadd_f32(vadd_f32(Ww[i][0], Ww[i][1]), Ww[i][2]), -4.0), recip576); - - // V[i][2] = -4*Ww[i][0] + 4*Ww[i][1] + -4*Ww[i][2]; - V[i][2] = vmul_n_f32(vmul_n_f32(vsub_f32(vsub_f32(Ww[i][1], Ww[i][0]), Ww[i][2]), 4.0), recip576); - - // V[i][3] = 1*Ww[i][0] + 2*Ww[i][1] + 4*Ww[i][2]; - V[i][3] = vmul_n_f32(vmla_n_f32(vmla_n_f32(Ww[i][0], Ww[i][1], 2.0f), Ww[i][2], 4.0f), recip576); - - // V[i][4] = 1*Ww[i][0] + -2*Ww[i][1] + 4*Ww[i][2]; - V[i][4] = vmul_n_f32(vmla_n_f32(vmls_n_f32(Ww[i][0], Ww[i][1], 2.0f), Ww[i][2], 4.0f), recip576); - - // V[i][5] = 24*Ww[i][2]; - V[i][5] = vmul_n_f32(vmul_n_f32(Ww[i][2], 24.0f), recip576); - } - - // Store the transformed weights - for (int i = 0, m = 0; i < 6; i++) - { - for (int j = 0; j < 6; j++, m++) - { - vst1_f32(outptr + m*matrix_stride, V[i][j]); - } - } - outptr += 2; - } -#endif // __arm_any__ - for (; channels_remaining; channels_remaining--) - { - // Matrices used and computed in this kernel - float w[3][3], Ww[6][3], V[6][6]; - - // Read weights - for (int i = 0; i < 3; i++) - { - for (int j = 0; j < 3; j++) - { - w[i][j] = *(inptrs[i][j]++); - } - } - - // Compute the matrix W w - for (int j = 0; j < 3; j++) - { - Ww[0][j] = 6*w[0][j]; - Ww[1][j] = -4*w[0][j] + -4*w[1][j] + -4*w[2][j]; - Ww[2][j] = -4*w[0][j] + 4*w[1][j] + -4*w[2][j]; - Ww[3][j] = 1*w[0][j] + 2*w[1][j] + 4*w[2][j]; - Ww[4][j] = 1*w[0][j] + -2*w[1][j] + 4*w[2][j]; - Ww[5][j] = 24*w[2][j]; - } - - // Compute V = W w WT - for (int i = 0; i < 6; i++) - { - V[i][0] = ( 6*Ww[i][0]) / 576.0; - V[i][1] = (-4*Ww[i][0] + -4*Ww[i][1] + -4*Ww[i][2]) / 576.0; - V[i][2] = (-4*Ww[i][0] + 4*Ww[i][1] + -4*Ww[i][2]) / 576.0; - V[i][3] = ( 1*Ww[i][0] + 2*Ww[i][1] + 4*Ww[i][2]) / 576.0; - V[i][4] = ( 1*Ww[i][0] + -2*Ww[i][1] + 4*Ww[i][2]) / 576.0; - V[i][5] = (24*Ww[i][2]) / 576.0; - } - - // Store the transformed weights - for (int i = 0, m = 0; i < 6; i++) - { - for (int j = 0; j < 6; j++, m++) - { - *(outptr + m*matrix_stride) = V[i][j]; - } - } - outptr++; - } - } - } - - template <> - template <> - int WinogradGEMM<4, 4, 3, 3>::WeightsTransform::ops_performed(const KernelShape &shape) - { - const int channel_prod = shape.n_input_channels * shape.n_output_channels; - return 9 * 16 * channel_prod; - } - - template struct WinogradGEMM<4, 4, 3, 3>::WeightsTransform; -} diff --git a/src/core/NEON/kernels/convolution/winograd/transforms/weights_6_3_fp32.cpp b/src/core/NEON/kernels/convolution/winograd/transforms/weights_6_3_fp32.cpp deleted file mode 100644 index c560aa8c8f..0000000000 --- a/src/core/NEON/kernels/convolution/winograd/transforms/weights_6_3_fp32.cpp +++ /dev/null @@ -1,125 +0,0 @@ -/* - * Copyright (c) 2017 ARM Limited. - * - * SPDX-License-Identifier: MIT - * - * Permission is hereby granted, free of charge, to any person obtaining a copy - * of this software and associated documentation files (the "Software"), to - * deal in the Software without restriction, including without limitation the - * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or - * sell copies of the Software, and to permit persons to whom the Software is - * furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in all - * copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE - * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER - * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, - * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. - */ - -#include "arm_compute/core/NEON/kernels/convolution/common/arm.hpp" -#include "arm_compute/core/NEON/kernels/convolution/winograd/winograd_gemm.hpp" -#include "arm_compute/core/NEON/kernels/convolution/winograd/transforms/kernel.hpp" - - -namespace winograd -{ - template <> - template <> - void WinogradGEMM<1, 6, 1, 3>::WeightsTransform::execute( - const int n_output_channels, - const int n_input_channels, - const float* const input, // NOTE: Data in HWIO order - float* const output, - const int matrix_stride, - const int matrix_row_stride - ) - { - // Get pointers to each cell of the weight tensor - const auto weight_col_stride = n_input_channels * n_output_channels; - const float *inptrs[3]; - for (int j = 0; j < 3; j++) - { - inptrs[j] = input + j*weight_col_stride; - } - - // For each input channel - for (int ic = 0; ic < n_input_channels; ic++) - { - float *outptr = output + ic * matrix_row_stride; - - // For each output channel - int channels_remaining = n_output_channels; - for (; channels_remaining; channels_remaining--) - { - // Matrices used and computed in this kernel - float w[3], V[inner_tile_cols]; - - // Read weights - for (int j = 0; j < 3; j++) - { - w[j] = *(inptrs[j]++); - } - - // Compute V = w WT - V[0] = (w[0]*-1) / 36.0f; - V[1] = (w[1]*-1 + w[0]*1 + w[2]*1) / 48.0f; - V[2] = (w[0]*1 + w[1]*1 + w[2]*1) / 48.0f; - V[3] = (w[0]*-1 + w[2]*-4 + w[1]*2) / 120.0f; - V[4] = (w[0]*-1 + w[2]*-4 + w[1]*-2) / 120.0f; - V[5] = (w[1]*-3 + w[2]*9 + w[0]*1) / 720.0f; - V[6] = (w[1]*3 + w[2]*9 + w[0]*1) / 720.0f; - V[7] = (w[2]*1) / 1; - - // Store the transformed weights - for (int j = 0; j < inner_tile_cols; j++) - { - *(outptr + j*matrix_stride) = V[j]; - } - outptr++; - } - } - } - - template <> - template <> - int WinogradGEMM<1, 6, 1, 3>::WeightsTransform::ops_performed(const KernelShape &shape) - { - (void) shape; - return 0; // TODO - } - - template <> - template <> - void WinogradGEMM<6, 1, 3, 1>::WeightsTransform::execute( - const int n_output_channels, - const int n_input_channels, - const float* const input, // NOTE: Data in HWIO order - float* const output, - const int matrix_stride, - const int matrix_row_stride - ) - { - // Redirect to the 1xN implementation - WinogradGEMM<1, 6, 1, 3>::template WeightsTransform::execute( - n_output_channels, n_input_channels, input, output, matrix_stride, - matrix_row_stride - ); - } - - template <> - template <> - int WinogradGEMM<6, 1, 3, 1>::WeightsTransform::ops_performed(const KernelShape &shape) - { - (void) shape; - return 0; // TODO - } - - template struct WinogradGEMM<1, 6, 1, 3>::WeightsTransform; - template struct WinogradGEMM<6, 1, 3, 1>::WeightsTransform; -} diff --git a/src/core/NEON/kernels/convolution/winograd/winograd.cpp b/src/core/NEON/kernels/convolution/winograd/winograd.cpp new file mode 100644 index 0000000000..226f303c7d --- /dev/null +++ b/src/core/NEON/kernels/convolution/winograd/winograd.cpp @@ -0,0 +1,236 @@ +/* + * Copyright (c) 2017-2019 ARM Limited. + * + * SPDX-License-Identifier: MIT + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#include +#include "winograd.hpp" +using namespace winograd; + +/** Get the output shape of a convolution. */ +template +template +Tensor4DShape WinogradGEMM::Convolution::get_output_shape( + const KernelShape &kernel_shape, + const Tensor4DShape &in_shape, + const PaddingType padding +) +{ + return Tensor4DShape { + in_shape.n_batches, + (padding == PADDING_SAME) ? in_shape.n_rows : in_shape.n_rows - (kernel_rows - 1), + (padding == PADDING_SAME) ? in_shape.n_cols : in_shape.n_cols - (kernel_cols - 1), + kernel_shape.n_output_channels, + in_shape.ordering + }; +} + +/* Get the memory required to transform the kernel. + */ +template +template +size_t WinogradGEMM::Convolution::get_kernel_transform_working_size(const KernelShape &shape) +{ + if (shape.ordering == HWIO) + { + // Kernel is already in the correct order, so no additional memory is + // required. + return 0; + } + else + { + // Need to re-order the kernel into HWIO form, require enough space to + // represent the tensor. + return sizeof(TIn) * shape.size(); + } +} + +/** Get the memory required to store the kernel transformed into the + * Winograd domain. + */ +template +template +size_t WinogradGEMM::Convolution::get_kernel_storage_size(const KernelShape &shape) +{ + return N_GEMMS * get_kernel_matrix_size(shape); +} + + +template +template +size_t WinogradGEMM::Convolution::get_input_storage_size( + const KernelShape &kernel_shape, + const Tensor4DShape &input_shape, + const PaddingType padding +) +{ + return N_GEMMS * get_input_matrix_size(kernel_shape, input_shape, padding); +} + + +template +template +size_t WinogradGEMM::Convolution::get_output_storage_size( + const KernelShape &kernel_shape, + const Tensor4DShape &input_shape, + const PaddingType padding +) +{ + return N_GEMMS * get_output_matrix_size(kernel_shape, input_shape, padding); +} + + +/** Get the memory required to apply a Winograd operator to some input. + */ +template +template +size_t WinogradGEMM::Convolution::get_working_space_size( + const KernelShape &kernel_shape, + const Tensor4DShape &input_shape, + const PaddingType padding_type +) +{ + const auto output_shape = get_output_shape(kernel_shape, input_shape, padding_type); + + // Get the memory required to store the matrices + const size_t matrix_sizes = N_GEMMS * ( + get_input_matrix_size(kernel_shape, input_shape, padding_type) + + get_output_matrix_size(kernel_shape, input_shape, padding_type) + ); + + // Add additional space to re-order the input and output if the input tensor + // is not in NHWC format. + if (input_shape.ordering == NHWC) + { + return matrix_sizes; // No extra spacing required + } + else // NCHW, must reorder the input and output tensors + { + // We only need to re-order the input or output at any one time, so request + // enough memory to do the largest of these. + const size_t extra_memory = std::max( + sizeof(TIn) * input_shape.size(), + sizeof(TOut) * output_shape.size() + ); + return matrix_sizes + extra_memory; + } +} + + +/* Get the memory required by a single "input" matrix. + */ +template +template +size_t WinogradGEMM::Convolution::get_input_matrix_size( + const KernelShape &kernel_shape, + const Tensor4DShape &input_shape, + const PaddingType padding_type +) +{ + return get_input_matrix_stride(kernel_shape, input_shape, padding_type) * sizeof(TGIn); +} + +template +template +int WinogradGEMM::Convolution::get_input_matrix_stride( + const KernelShape &kernel_shape, + const Tensor4DShape &input_shape, + const PaddingType padding_type +) +{ + // Compute shape for the GEMM + const auto output_shape = get_output_shape(kernel_shape, input_shape, padding_type); + const int tile_rows = iceildiv(output_shape.n_rows, output_tile_rows); + const int tile_cols = iceildiv(output_shape.n_cols, output_tile_cols); + const int M = roundup(input_shape.n_batches * tile_rows * tile_cols, M_BLOCK); + const int K = kernel_shape.n_input_channels; + + return M * K; +} + + +/* Get the memory required by a single "output" matrix. + */ +template +template +size_t WinogradGEMM::Convolution::get_output_matrix_size( + const KernelShape &kernel_shape, + const Tensor4DShape &input_shape, + const PaddingType padding_type +) +{ + return get_output_matrix_stride(kernel_shape, input_shape, padding_type) * sizeof(TGOut); +} + + +template +template +int WinogradGEMM::Convolution::get_output_matrix_stride( + const KernelShape &kernel_shape, + const Tensor4DShape &input_shape, + const PaddingType padding_type +) +{ + // Compute shape for the GEMM + const auto output_shape = get_output_shape(kernel_shape, input_shape, padding_type); + const int tile_rows = iceildiv(output_shape.n_rows, output_tile_rows); + const int tile_cols = iceildiv(output_shape.n_cols, output_tile_cols); + const int M = roundup(tile_rows * tile_cols, M_BLOCK); + const int N = roundup(kernel_shape.n_output_channels, N_BLOCK); + + return input_shape.n_batches * M * N; +} + + +/* Get the memory required by a single "kernel" matrix. + */ +template +template +size_t WinogradGEMM::Convolution::get_kernel_matrix_size(const KernelShape &shape) +{ + return sizeof(TGIn) * get_kernel_matrix_stride(shape); +} + +template +template +int WinogradGEMM::Convolution::get_kernel_matrix_stride(const KernelShape &shape) +{ + const int K = shape.n_input_channels; + const int N = roundup(shape.n_output_channels, N_BLOCK); + return K * N; +} + + +// Instantiate required implementations +template class WinogradGEMM<2, 2, 3, 3, WinogradRoots::Integers>::Convolution; +template class WinogradGEMM<4, 4, 3, 3, WinogradRoots::Integers>::Convolution; + +template class WinogradGEMM<1, 6, 1, 3, WinogradRoots::Integers>::Convolution; +template class WinogradGEMM<6, 1, 3, 1, WinogradRoots::Integers>::Convolution; + +template class WinogradGEMM<2, 2, 5, 5, WinogradRoots::Integers>::Convolution; + +template class WinogradGEMM<1, 4, 1, 5, WinogradRoots::Integers>::Convolution; +template class WinogradGEMM<4, 1, 5, 1, WinogradRoots::Integers>::Convolution; + +template class WinogradGEMM<1, 2, 1, 7, WinogradRoots::Integers>::Convolution; +template class WinogradGEMM<2, 1, 7, 1, WinogradRoots::Integers>::Convolution; diff --git a/src/core/NEON/kernels/convolution/winograd/winograd_gemm.cpp b/src/core/NEON/kernels/convolution/winograd/winograd_gemm.cpp deleted file mode 100644 index a7de2fd3e5..0000000000 --- a/src/core/NEON/kernels/convolution/winograd/winograd_gemm.cpp +++ /dev/null @@ -1,240 +0,0 @@ -/* - * Copyright (c) 2017 ARM Limited. - * - * SPDX-License-Identifier: MIT - * - * Permission is hereby granted, free of charge, to any person obtaining a copy - * of this software and associated documentation files (the "Software"), to - * deal in the Software without restriction, including without limitation the - * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or - * sell copies of the Software, and to permit persons to whom the Software is - * furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in all - * copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE - * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER - * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, - * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. - */ -#include -#include "arm_compute/core/NEON/kernels/convolution/winograd/winograd_gemm.hpp" -#include "arm_compute/core/NEON/kernels/convolution/winograd/batched_blocked_gemm.hpp" -using namespace winograd; - -/** Get the output shape of a convolution. */ -template -template -Tensor4DShape WinogradGEMM::Convolution::get_output_shape( - const KernelShape &kernel_shape, - const Tensor4DShape &in_shape, - const PaddingType padding -) -{ - return Tensor4DShape { - in_shape.n_batches, - (padding == PADDING_SAME) ? in_shape.n_rows : in_shape.n_rows - (kernel_rows - 1), - (padding == PADDING_SAME) ? in_shape.n_cols : in_shape.n_cols - (kernel_cols - 1), - kernel_shape.n_output_channels, - in_shape.ordering - }; -} - -/* Get the memory required to transform the kernel. - */ -template -template -size_t WinogradGEMM::Convolution::get_kernel_transform_working_size(const KernelShape &shape) -{ - if (shape.ordering == HWIO) - { - // Kernel is already in the correct order, so no additional memory is - // required. - return 0; - } - else - { - // Need to re-order the kernel into HWIO form, require enough space to - // represent the tensor. - return sizeof(TIn) * shape.size(); - } -} - -/** Get the memory required to store the kernel transformed into the - * Winograd domain. - */ -template -template -size_t WinogradGEMM::Convolution::get_kernel_storage_size(const KernelShape &shape) -{ - return N_GEMMS * get_kernel_matrix_size(shape); -} - - -template -template -size_t WinogradGEMM::Convolution::get_input_storage_size( - const KernelShape &kernel_shape, - const Tensor4DShape &input_shape, - const PaddingType padding -) -{ - return N_GEMMS * get_input_matrix_size(kernel_shape, input_shape, padding); -} - - -template -template -size_t WinogradGEMM::Convolution::get_output_storage_size( - const KernelShape &kernel_shape, - const Tensor4DShape &input_shape, - const PaddingType padding -) -{ - return N_GEMMS * get_output_matrix_size(kernel_shape, input_shape, padding); -} - - -/** Get the memory required to apply a Winograd operator to some input. - */ -template -template -size_t WinogradGEMM::Convolution::get_working_space_size( - const KernelShape &kernel_shape, - const Tensor4DShape &input_shape, - const PaddingType padding_type -) -{ - const auto output_shape = get_output_shape(kernel_shape, input_shape, padding_type); - - // Get the memory required to store the matrices - const size_t matrix_sizes = N_GEMMS * ( - get_input_matrix_size(kernel_shape, input_shape, padding_type) + - get_output_matrix_size(kernel_shape, input_shape, padding_type) - ); - - // Add additional space to re-order the input and output if the input tensor - // is not in NHWC format. - if (input_shape.ordering == NHWC) - { - return matrix_sizes; // No extra spacing required - } - else // NCHW, must reorder the input and output tensors - { - // We only need to re-order the input or output at any one time, so request - // enough memory to do the largest of these. - const size_t extra_memory = std::max( - sizeof(TIn) * input_shape.size(), - sizeof(TOut) * output_shape.size() - ); - return matrix_sizes + extra_memory; - } -} - - -/* Get the memory required by a single "input" matrix. - */ -template -template -size_t WinogradGEMM::Convolution::get_input_matrix_size( - const KernelShape &kernel_shape, - const Tensor4DShape &input_shape, - const PaddingType padding_type -) -{ - return get_input_matrix_stride(kernel_shape, input_shape, padding_type) * sizeof(TIn); -} - -template -template -int WinogradGEMM::Convolution::get_input_matrix_stride( - const KernelShape &kernel_shape, - const Tensor4DShape &input_shape, - const PaddingType padding_type -) -{ - // Compute shape for the GEMM - const auto output_shape = get_output_shape(kernel_shape, input_shape, padding_type); - const int tile_rows = iceildiv(output_shape.n_rows, output_tile_rows); - const int tile_cols = iceildiv(output_shape.n_cols, output_tile_cols); - const int M = roundup(input_shape.n_batches * tile_rows * tile_cols, M_BLOCK); - const int K = kernel_shape.n_input_channels; - - return M * K; -} - - -/* Get the memory required by a single "output" matrix. - */ -template -template -size_t WinogradGEMM::Convolution::get_output_matrix_size( - const KernelShape &kernel_shape, - const Tensor4DShape &input_shape, - const PaddingType padding_type -) -{ - return get_output_matrix_stride(kernel_shape, input_shape, padding_type) * sizeof(TOut); -} - - -template -template -int WinogradGEMM::Convolution::get_output_matrix_stride( - const KernelShape &kernel_shape, - const Tensor4DShape &input_shape, - const PaddingType padding_type -) -{ - // Compute shape for the GEMM - const auto output_shape = get_output_shape(kernel_shape, input_shape, padding_type); - const int tile_rows = iceildiv(output_shape.n_rows, output_tile_rows); - const int tile_cols = iceildiv(output_shape.n_cols, output_tile_cols); - const int M = roundup(tile_rows * tile_cols, M_BLOCK); - const int N = roundup(kernel_shape.n_output_channels, N_BLOCK); - - return input_shape.n_batches * M * N; -} - - -/* Get the memory required by a single "kernel" matrix. - */ -template -template -size_t WinogradGEMM::Convolution::get_kernel_matrix_size(const KernelShape &shape) -{ - return sizeof(TIn) * get_kernel_matrix_stride(shape); -} - -template -template -int WinogradGEMM::Convolution::get_kernel_matrix_stride(const KernelShape &shape) -{ - const int K = shape.n_input_channels; - const int N = roundup(shape.n_output_channels, N_BLOCK); - return K * N; -} - - -// Instantiate required implementations -template class WinogradGEMM<2, 2, 3, 3>::Convolution; -template class WinogradGEMM<4, 4, 3, 3>::Convolution; - -template class WinogradGEMM<1, 6, 1, 3>::Convolution; -template class WinogradGEMM<6, 1, 3, 1>::Convolution; - -template class WinogradGEMM<2, 2, 5, 5>::Convolution; - -template class WinogradGEMM<1, 4, 1, 5>::Convolution; -template class WinogradGEMM<4, 1, 5, 1>::Convolution; - -template class WinogradGEMM<1, 2, 1, 7>::Convolution; -template class WinogradGEMM<2, 1, 7, 1>::Convolution; - - - diff --git a/src/core/NEON/kernels/convolution/winograd/winograd_transforms/input.hpp b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/input.hpp new file mode 100644 index 0000000000..fcbd21fcd0 --- /dev/null +++ b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/input.hpp @@ -0,0 +1,265 @@ +/* + * Copyright (c) 2019 ARM Limited. + * + * SPDX-License-Identifier: MIT + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#pragma once + +#include "winograd.hpp" +#include "padding.hpp" + +#define MEMBERFN(RTYPE) template <\ + int InnerTileRows, int InnerTileCols,\ + typename TIn, typename TOut, WinogradRoots Roots\ +> RTYPE InputTransform + + +#define Nx1MEMBERFN(RTYPE) template <\ + int InnerTileRows, typename TIn, typename TOut, WinogradRoots Roots\ +> RTYPE InputTransform + +namespace winograd +{ + +MEMBERFN()::InputTransform( + const int kernel_rows, + const int kernel_cols, + const int n_batches, + const int n_rows, + const int n_cols, + const int n_channels, + const int padding_top, + const int padding_left, + const int padding_bottom, + const int padding_right +) : _n_batches(n_batches), _n_rows(n_rows), _n_cols(n_cols), _n_channels(n_channels), + _inptr(nullptr), _outptr(nullptr), + _overlap_rows(kernel_rows - 1), _overlap_cols(kernel_cols - 1), + _padding_top(padding_top), _padding_left(padding_left), _padding_bottom(padding_bottom), _padding_right(padding_right), + _tiles_M(iceildiv(padding_top + n_rows + padding_bottom - kernel_rows + 1, InnerTileRows - kernel_rows + 1)), + _tiles_N(iceildiv(padding_left + n_cols + padding_right - kernel_cols + 1, InnerTileCols - kernel_cols + 1)), + _matrix_stride(0), _matrix_row_stride(0), _matrix_batch_stride(0), + _in_col_stride(0), _in_row_stride(0), _in_batch_stride(0), + _working_space_col_stride(n_channels), + _working_space_row_stride(InnerTileCols * _working_space_col_stride), + _working_space(nullptr) +{ +} + +MEMBERFN(void)::set_input_tensor(const void* const inptr) +{ + set_input_tensor(inptr, _n_channels); +} + +MEMBERFN(void)::set_input_tensor(const void* const inptr, const int ldcol) +{ + set_input_tensor(inptr, _n_cols * ldcol, ldcol); +} + +MEMBERFN(void)::set_input_tensor(const void* const inptr, const int ldrow, const int ldcol) +{ + set_input_tensor(inptr, _n_rows * ldrow, ldrow, ldcol); +} + +MEMBERFN(void)::set_input_tensor(const void* const inptr, const int ldbatch, const int ldrow, const int ldcol) +{ + _inptr = static_cast(inptr); + _in_batch_stride = ldbatch; + _in_row_stride = ldrow; + _in_col_stride = ldcol; +} + +MEMBERFN(void)::set_output_matrices(void * const mptr, const int ldmatrix, const int ldrow) +{ + _outptr = static_cast(mptr); + _matrix_stride = ldmatrix; + _matrix_row_stride = ldrow; + _matrix_batch_stride = _tiles_M * _tiles_N * ldrow; +} + +Nx1MEMBERFN()::InputTransform( + const int kernel_rows, + const int kernel_cols, + const int n_batches, + const int n_rows, + const int n_cols, + const int n_channels, + const int padding_top, + const int padding_left, + const int padding_bottom, + const int padding_right +) : InputTransform<1, InnerTileRows, TIn, TOut, Roots>::InputTransform( + /* Transpose rows and columns */ + kernel_cols, kernel_rows, n_batches, n_cols, n_rows, n_channels, + padding_left, padding_top, padding_right, padding_bottom + ) +{ +} + +Nx1MEMBERFN(void)::set_input_tensor(const void* const inptr) +{ + set_input_tensor(inptr, this->_n_channels); +} + +Nx1MEMBERFN(void)::set_input_tensor(const void* const inptr, const int ldcol) +{ + set_input_tensor(inptr, this->_n_cols * ldcol, ldcol); +} + +Nx1MEMBERFN(void)::set_input_tensor(const void* const inptr, const int ldrow, const int ldcol) +{ + set_input_tensor(inptr, this->_n_rows * ldrow, ldrow, ldcol); +} + +Nx1MEMBERFN(void)::set_input_tensor(const void* const inptr, const int ldbatch, const int ldrow, const int ldcol) +{ + // Transpose row and column strides + Base::set_input_tensor(inptr, ldbatch, ldcol, ldrow); +} + +MEMBERFN(size_t)::get_working_space_size(const unsigned int nthreads) const +{ + return sizeof(TIn) * InnerTileRows * _working_space_row_stride * nthreads; +} + +MEMBERFN(void)::set_working_space(void * const buffer) +{ + _working_space = static_cast(buffer); +} + +MEMBERFN(unsigned int)::get_window(void) const +{ + return iceildiv(_n_channels, WINDOW_BLOCK); +} + +MEMBERFN(void)::run( + const unsigned int start, + const unsigned int stop, + const unsigned int threadid +) +{ + // Determine the channels on which to work + if (start >= get_window()) + { + return; // No work to do beyond the end of the window + } + const unsigned int start_channel = start * WINDOW_BLOCK; + const unsigned int stop_channel = std::min(_n_channels , stop * WINDOW_BLOCK); + const unsigned int n_channels = stop_channel - start_channel; + + // Loop over batches + for (int batch = 0; batch < _n_batches; batch++) + { + const TIn* const inptr_batch = _inptr + start_channel + batch*_in_batch_stride; + TOut* const outptr_batch = _outptr + start_channel + batch*_matrix_batch_stride; + + // Loop over rows of tiles + for (int tile_i = 0; tile_i < _tiles_M; tile_i++) + { + // Compute the starting and ending row of pixels within the row of tiles, + // hence compute the padding to apply to the top and bottom of each tile. + const int row_top = tile_i * (InnerTileRows - _overlap_rows) - _padding_top; + const int row_bottom = row_top + InnerTileRows; + const int row_pad_top = std::max(0, _padding_top - tile_i * (InnerTileRows - _overlap_rows)); + const int row_pad_bottom = std::max(0, row_bottom - _n_rows); + + // Get a pointer to the start of the row. + const int row_offset = std::min(0, row_pad_top - _padding_top); + const TIn* const inptr_row = inptr_batch + _in_row_stride*(row_offset + tile_i*(InnerTileRows - _overlap_rows)); + TOut* const outptr_row = outptr_batch + tile_i*_tiles_N*_matrix_row_stride; + + // Loop over tiles within the row + for (int tile_j = 0; tile_j < _tiles_N; tile_j++) + { + // Compute the starting and ending column of pixels within the tile, + // hence compute the padding to apply to the left and right of the + // tile. + const int tile_left = tile_j * (InnerTileCols - _overlap_cols) - _padding_left; + const int tile_right = tile_left + InnerTileCols; + const int tile_pad_left = std::max(0, _padding_left - tile_j * (InnerTileCols - _overlap_cols)); + const int tile_pad_right = std::max(0, tile_right - _n_cols); + + // Get a pointer to the start of the tile. + const int col_offset = std::min(0, tile_pad_left - _padding_left); + const TIn* const inptr_tile = inptr_row + _in_col_stride*(col_offset + tile_j*(InnerTileCols - _overlap_cols)); + TOut* const outptr_tile = outptr_row + tile_j * _matrix_row_stride; + + // Transform the tile, applying padding if necessary. + if (row_pad_top || tile_pad_left || row_pad_bottom || tile_pad_right) + { + transform_padded_tile( + threadid, n_channels, outptr_tile, inptr_tile, + row_pad_top, tile_pad_left, row_pad_bottom, tile_pad_right + ); + } + else + { + transform_unpadded_tile(threadid, n_channels, outptr_tile, inptr_tile); + } + } + } + } +} + +MEMBERFN(void)::transform_unpadded_tile( + const unsigned int /* threadid unused */, + const int n_channels, + TOut * const outptr, + const TIn * const inptr +) +{ + transform_tile( + n_channels, inptr, _in_row_stride, _in_col_stride, outptr, _matrix_stride + ); +} + +MEMBERFN(void)::transform_padded_tile( + const unsigned int threadid, + const int n_channels, + TOut * const outptr, + const TIn * const inptr, + const int padding_top, + const int padding_left, + const int padding_bottom, + const int padding_right +) +{ + padding::copy_and_pad_tile( + InnerTileRows, InnerTileCols, n_channels, + inptr, _in_row_stride, _in_col_stride, + static_cast(get_working_space(threadid)), _working_space_row_stride, _working_space_col_stride, + padding_top, padding_left, padding_bottom, padding_right + ); + + transform_tile( + n_channels, static_cast(get_working_space(threadid)), + _working_space_row_stride, _working_space_col_stride, + outptr, _matrix_stride + ); +} + +MEMBERFN(void *)::get_working_space(const unsigned int threadid) const +{ + return _working_space + InnerTileRows * _working_space_row_stride * threadid; +} + +} // namespace winograd diff --git a/src/core/NEON/kernels/convolution/winograd/winograd_transforms/input_1x8_fp32_fp32_integers.cpp b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/input_1x8_fp32_fp32_integers.cpp new file mode 100644 index 0000000000..5040ec1bd4 --- /dev/null +++ b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/input_1x8_fp32_fp32_integers.cpp @@ -0,0 +1,158 @@ +/* + * Copyright (c) 2019 ARM Limited. + * + * SPDX-License-Identifier: MIT + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "arm.hpp" +#include "input.hpp" + +namespace winograd +{ + +template <> +void InputTransform<1, 8, float, float, WinogradRoots::Integers>::transform_tile( + const int n_channels, + const float* const input_base, + const int, // We don't need to stride over rows + const int input_col_stride, + float* outptr, + const int matrix_stride +) +{ + constexpr int inner_tile_cols = 8; + + // Get pointers into the input tile + const float *x_ptrs[inner_tile_cols]; + for (int j = 0, xj = 0; j < inner_tile_cols; j++, xj++) + { + x_ptrs[j] = input_base + xj*input_col_stride; + } + + // Vectors used/computed in this kernel. + float x[inner_tile_cols]; + float U[inner_tile_cols]; + + for (int j = 0; j < inner_tile_cols; j++) + { + x[j] = 0.0f; + } + + // Perform the Winograd input transformation for each channel in the input + // tensor. + int channels_remaining = n_channels; +#ifdef _arm_any_ + for (; channels_remaining >= 4; channels_remaining -= 4) + { + float32x4_t x[inner_tile_cols], U[inner_tile_cols]; + for (int j = 0; j < inner_tile_cols; j++) + { + x[j] = vdupq_n_f32(0.0f); + } + + // Load x + for (int j = 0; j < inner_tile_cols; j++) + { + x[j] = vld1q_f32(x_ptrs[j]); + x_ptrs[j] += 4; + } + + // Compute U = x . X + U[0] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(x[6], 1), x[2], 49), x[4], -14), x[0], -36); + U[1] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(x[6], 1), x[2], 36), x[3], 13), x[4], -13), x[1], -36), x[5], -1); + U[2] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(x[6], 1), x[5], 1), x[2], 36), x[1], 36), x[4], -13), x[3], -13); + U[3] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(x[6], 1), x[3], 20), x[2], 9), x[5], -2), x[4], -10), x[1], -18); + U[4] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(x[6], 1), x[1], 18), x[2], 9), x[5], 2), x[4], -10), x[3], -20); + U[5] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(x[6], 1), x[3], 15), x[2], 4), x[5], -3), x[4], -5), x[1], -12); + U[6] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(x[6], 1), x[1], 12), x[2], 4), x[5], 3), x[4], -5), x[3], -15); + U[7] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(x[7], 1), x[3], 49), x[5], -14), x[1], -36); + + // Store the transformed vector + for (int j = 0; j < inner_tile_cols; j++) + { + vst1q_f32(outptr + j*matrix_stride, U[j]); + } + outptr += 4; + } + for (; channels_remaining >= 2; channels_remaining -= 2) + { + float32x2_t x[inner_tile_cols], U[inner_tile_cols]; + for (int j = 0; j < inner_tile_cols; j++) + { + x[j] = vdup_n_f32(0.0f); + } + + // Load x + for (int j = 0; j < inner_tile_cols; j++) + { + x[j] = vld1_f32(x_ptrs[j]); + x_ptrs[j] += 2; + } + + // Compute U = x . X + U[0] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(x[6], 1), x[2], 49), x[4], -14), x[0], -36); + U[1] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(x[6], 1), x[2], 36), x[3], 13), x[4], -13), x[1], -36), x[5], -1); + U[2] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(x[6], 1), x[5], 1), x[2], 36), x[1], 36), x[4], -13), x[3], -13); + U[3] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(x[6], 1), x[3], 20), x[2], 9), x[5], -2), x[4], -10), x[1], -18); + U[4] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(x[6], 1), x[1], 18), x[2], 9), x[5], 2), x[4], -10), x[3], -20); + U[5] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(x[6], 1), x[3], 15), x[2], 4), x[5], -3), x[4], -5), x[1], -12); + U[6] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(x[6], 1), x[1], 12), x[2], 4), x[5], 3), x[4], -5), x[3], -15); + U[7] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(x[7], 1), x[3], 49), x[5], -14), x[1], -36); + + // Store the transformed vector + for (int j = 0; j < inner_tile_cols; j++) + { + vst1_f32(outptr + j*matrix_stride, U[j]); + } + outptr += 2; + } +#endif // _arm_any_ + for (; channels_remaining; channels_remaining--) + { + // Load x + for (int j = 0; j < inner_tile_cols; j++) + { + x[j] = *(x_ptrs[j]++); + } + + // Compute U = x . X + U[0] = x[0]*-36 + x[4]*-14 + x[2]*49 + x[6]*1; + U[1] = x[5]*-1 + x[1]*-36 + x[4]*-13 + x[3]*13 + x[2]*36 + x[6]*1; + U[2] = x[3]*-13 + x[4]*-13 + x[1]*36 + x[2]*36 + x[5]*1 + x[6]*1; + U[3] = x[1]*-18 + x[4]*-10 + x[5]*-2 + x[2]*9 + x[3]*20 + x[6]*1; + U[4] = x[3]*-20 + x[4]*-10 + x[5]*2 + x[2]*9 + x[1]*18 + x[6]*1; + U[5] = x[1]*-12 + x[4]*-5 + x[5]*-3 + x[2]*4 + x[3]*15 + x[6]*1; + U[6] = x[3]*-15 + x[4]*-5 + x[5]*3 + x[2]*4 + x[1]*12 + x[6]*1; + U[7] = x[1]*-36 + x[5]*-14 + x[3]*49 + x[7]*1; + + // Store the transformed vector + for (int j = 0; j < inner_tile_cols; j++) + { + *(outptr + j*matrix_stride) = U[j]; + } + outptr++; + } +} + +template class InputTransform<1, 8, float, float, WinogradRoots::Integers>; +template class InputTransform<8, 1, float, float, WinogradRoots::Integers>; + +} // namespace winograd diff --git a/src/core/NEON/kernels/convolution/winograd/winograd_transforms/input_4x4_fp32_fp32_integers.cpp b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/input_4x4_fp32_fp32_integers.cpp new file mode 100644 index 0000000000..9393785dfc --- /dev/null +++ b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/input_4x4_fp32_fp32_integers.cpp @@ -0,0 +1,255 @@ +/* + * Copyright (c) 2019 ARM Limited. + * + * SPDX-License-Identifier: MIT + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "input.hpp" +#include "arm.hpp" + +namespace winograd +{ + +template <> +void InputTransform<4, 4, float, float, WinogradRoots::Integers>::transform_tile( + const int n_channels, + const float* const input_base, + const int input_row_stride, + const int input_col_stride, + float* outptr, + const int matrix_stride +) +{ + constexpr int inner_tile_rows = 4, inner_tile_cols = 4; + + // Get pointers into the input tile + const float *x_ptrs[inner_tile_rows][inner_tile_cols]; + for (int i = 0, xi = 0; i < inner_tile_rows; i++, xi++) + { + // Get a pointer into the row + const float* const row_ptr = input_base + xi*input_row_stride; + + for (int j = 0, xj = 0; j < inner_tile_cols; j++, xj++) + { + x_ptrs[i][j] = row_ptr + xj*input_col_stride; + } + } + + // Matrices used/computed in this kernel. + float x[inner_tile_rows][inner_tile_cols]; + float XTx[inner_tile_rows][inner_tile_cols]; + float U[inner_tile_rows][inner_tile_cols]; + + for (int i = 0; i < inner_tile_rows; i++) + { + for (int j = 0; j < inner_tile_cols; j++) + { + x[i][j] = XTx[i][j] = 0.0f; + } + } + + // Perform the Winograd input transformation for each channel in the input + // tensor. + int channels_remaining = n_channels; +#ifdef __aarch64__ + for (; channels_remaining >= 4; channels_remaining -= 4) + { + // Matrices used/computed in this kernel. + float32x4_t x[inner_tile_rows][inner_tile_cols]; + float32x4_t XTx[inner_tile_rows][inner_tile_cols]; + float32x4_t U[inner_tile_rows][inner_tile_cols]; + + for (int i = 0; i < inner_tile_rows; i++) + { + for (int j = 0; j < inner_tile_cols; j++) + { + x[i][j] = vdupq_n_f32(0.0f); + XTx[i][j] = vdupq_n_f32(0.0f); + } + } + + // Load x + for (int i = 0; i < inner_tile_rows; i++) + { + for (int j = 0; j < inner_tile_cols; j++) + { + x[i][j] = vld1q_f32(x_ptrs[i][j]); + x_ptrs[i][j] += 4; + } + } + + // Compute XT . x + for (int j = 0; j < inner_tile_cols; j++) + { + // XTx[0][j] = x[0][j] - x[2][j]; + XTx[0][j] = vsubq_f32(x[0][j], x[2][j]); + + // XTx[1][j] = x[1][j] + x[2][j]; + XTx[1][j] = vaddq_f32(x[1][j], x[2][j]); + + // XTx[2][j] = x[2][j] - x[1][j]; + XTx[2][j] = vsubq_f32(x[2][j], x[1][j]); + + // XTx[3][j] = x[1][j] - x[3][j]; + XTx[3][j] = vsubq_f32(x[1][j], x[3][j]); + } + + // Compute U = XT . x . X + for (int i = 0; i < inner_tile_rows; i++) + { + // U[i][0] = XTx[i][0] - XTx[i][2]; + U[i][0] = vsubq_f32(XTx[i][0], XTx[i][2]); + + // U[i][1] = XTx[i][1] + XTx[i][2]; + U[i][1] = vaddq_f32(XTx[i][1], XTx[i][2]); + + // U[i][2] = XTx[i][2] - XTx[i][1]; + U[i][2] = vsubq_f32(XTx[i][2], XTx[i][1]); + + // U[i][3] = XTx[i][1] - XTx[i][3]; + U[i][3] = vsubq_f32(XTx[i][1], XTx[i][3]); + } + + // Store the transformed matrix + for (int i = 0, m = 0; i < inner_tile_rows; i++) + { + for (int j = 0; j < inner_tile_cols; j++, m++) + { + vst1q_f32(outptr + m*matrix_stride, U[i][j]); + } + } + outptr += 4; + } +#endif // __aarch64__ +#ifdef __arm_any__ + for (; channels_remaining >= 2; channels_remaining -= 2) + { + // Matrices used/computed in this kernel. + float32x2_t x[inner_tile_rows][inner_tile_cols]; + float32x2_t XTx[inner_tile_rows][inner_tile_cols]; + float32x2_t U[inner_tile_rows][inner_tile_cols]; + + for (int i = 0; i < inner_tile_rows; i++) + { + for (int j = 0; j < inner_tile_cols; j++) + { + x[i][j] = vdup_n_f32(0.0f); + XTx[i][j] = vdup_n_f32(0.0f); + } + } + + // Load x + for (int i = 0; i < inner_tile_rows; i++) + { + for (int j = 0; j < inner_tile_cols; j++) + { + x[i][j] = vld1_f32(x_ptrs[i][j]); + x_ptrs[i][j] += 2; + } + } + + // Compute XT . x + for (int j = 0; j < inner_tile_cols; j++) + { + // XTx[0][j] = x[0][j] - x[2][j]; + XTx[0][j] = vsub_f32(x[0][j], x[2][j]); + + // XTx[1][j] = x[1][j] + x[2][j]; + XTx[1][j] = vadd_f32(x[1][j], x[2][j]); + + // XTx[2][j] = x[2][j] - x[1][j]; + XTx[2][j] = vsub_f32(x[2][j], x[1][j]); + + // XTx[3][j] = x[1][j] - x[3][j]; + XTx[3][j] = vsub_f32(x[1][j], x[3][j]); + } + + // Compute U = XT . x . X + for (int i = 0; i < inner_tile_rows; i++) + { + // U[i][0] = XTx[i][0] - XTx[i][2]; + U[i][0] = vsub_f32(XTx[i][0], XTx[i][2]); + + // U[i][1] = XTx[i][1] + XTx[i][2]; + U[i][1] = vadd_f32(XTx[i][1], XTx[i][2]); + + // U[i][2] = XTx[i][2] - XTx[i][1]; + U[i][2] = vsub_f32(XTx[i][2], XTx[i][1]); + + // U[i][3] = XTx[i][1] - XTx[i][3]; + U[i][3] = vsub_f32(XTx[i][1], XTx[i][3]); + } + + // Store the transformed matrix + for (int i = 0, m = 0; i < inner_tile_rows; i++) + { + for (int j = 0; j < inner_tile_cols; j++, m++) + { + vst1_f32(outptr + m*matrix_stride, U[i][j]); + } + } + outptr += 2; + } +#endif // __arm_any__ + for (; channels_remaining; channels_remaining--) + { + // Load x + for (int i = 0; i < inner_tile_rows; i++) + { + for (int j = 0; j < inner_tile_cols; j++) + { + x[i][j] = *(x_ptrs[i][j]++); + } + } + + // Compute XT . x + for (int j = 0; j < inner_tile_cols; j++) + { + XTx[0][j] = x[0][j] - x[2][j]; + XTx[1][j] = x[1][j] + x[2][j]; + XTx[2][j] = x[2][j] - x[1][j]; + XTx[3][j] = x[1][j] - x[3][j]; + } + + // Compute U = XT . x . X + for (int i = 0; i < inner_tile_rows; i++) + { + U[i][0] = XTx[i][0] - XTx[i][2]; + U[i][1] = XTx[i][1] + XTx[i][2]; + U[i][2] = XTx[i][2] - XTx[i][1]; + U[i][3] = XTx[i][1] - XTx[i][3]; + } + + // Store the transformed matrix + for (int i = 0, m = 0; i < inner_tile_rows; i++) + { + for (int j = 0; j < inner_tile_cols; j++, m++) + { + *(outptr + m*matrix_stride) = U[i][j]; + } + } + outptr++; + } +} + +template class InputTransform<4, 4, float, float, WinogradRoots::Integers>; + +} // namespace diff --git a/src/core/NEON/kernels/convolution/winograd/winograd_transforms/input_6x6_fp32_fp32_integers.cpp b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/input_6x6_fp32_fp32_integers.cpp new file mode 100644 index 0000000000..908fc8292a --- /dev/null +++ b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/input_6x6_fp32_fp32_integers.cpp @@ -0,0 +1,1308 @@ +/* + * Copyright (c) 2019 ARM Limited. + * + * SPDX-License-Identifier: MIT + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "arm.hpp" +#include "input.hpp" + +namespace winograd +{ + +#ifdef __aarch64__ + +template <> +void InputTransform<6, 6, float, float, WinogradRoots::Integers>::transform_tile( + int n_channels, + const float* input_base, + const int input_row_stride, + const int input_col_stride, + float* matrix_base, + const int matrix_stride +) +{ + const float pcoeffs[4] = {1.0f, 2.0f, 4.0f, 5.0f}; + __asm__ __volatile__( + "ldr q0, [%[pcoeffs]]\n" + "add x25, %[inptr0], %[input_row_stride]\n" + "add x18, %[input_col_stride1], %[input_col_stride1]\n" + "add x16, x25, %[input_row_stride]\n" + "add x19, x18, %[input_col_stride1]\n" + "add x26, x16, %[input_row_stride]\n" + "add x20, x19, %[input_col_stride1]\n" + "add x17, x26, %[input_row_stride]\n" + "add x21, x20, %[input_col_stride1]\n" + "add x27, x17, %[input_row_stride]\n" + "add x28, %[outptr0], %[output_row_stride]\n" + "add x11, %[output_col_stride1], %[output_col_stride1]\n" + "add x22, x28, %[output_row_stride]\n" + "add x13, x11, %[output_col_stride1]\n" + "add x12, x22, %[output_row_stride]\n" + "add x23, x13, %[output_col_stride1]\n" + "add x14, x12, %[output_row_stride]\n" + "add x15, x23, %[output_col_stride1]\n" + "add x24, x14, %[output_row_stride]\n" + "cmp %w[n_channels], #4\n" + "blt 2f\n" + "1:\n" + "ldr q8, [%[inptr0], x20]\n" + "ldr q2, [%[inptr0], x18]\n" + "mov v14.16b, v8.16b\n" + "ldr q9, [%[inptr0]]\n" + "mov v10.16b, v8.16b\n" + "ldr q1, [%[inptr0], x21]\n" + "fmla v14.4s, v9.4s, v0.s[2]\n" + "ldr q4, [%[inptr0], x19]\n" + "mov v9.16b, v8.16b\n" + "ldr q12, [%[inptr0], %[input_col_stride1]]\n" + "fmls v10.4s, v12.4s, v0.s[2]\n" + "ldr q5, [x16, x20]\n" + "fmls v14.4s, v2.4s, v0.s[3]\n" + "ldr q20, [x16, x18]\n" + "fmla v9.4s, v12.4s, v0.s[2]\n" + "ldr q3, [x16]\n" + "fmls v10.4s, v2.4s, v0.s[2]\n" + "ldr q6, [x16, x21]\n" + "mov v7.16b, v8.16b\n" + "ldr q16, [x16, x19]\n" + "fmls v9.4s, v2.4s, v0.s[2]\n" + "ldr q22, [x16, %[input_col_stride1]]\n" + "fadd v10.4s, v10.4s, v4.4s\n" + "ldr q17, [x17, x20]\n" + "fmls v7.4s, v12.4s, v0.s[1]\n" + "ldr q15, [x17, x18]\n" + "fsub v9.4s, v9.4s, v4.4s\n" + "ldr q19, [x17]\n" + "mov v8.16b, v8.16b\n" + "ldr q18, [x17, x21]\n" + "fsub v7.4s, v7.4s, v2.4s\n" + "ldr q13, [x17, x19]\n" + "fmla v7.4s, v4.4s, v0.s[1]\n" + "ldr q21, [x17, %[input_col_stride1]]\n" + "fmla v8.4s, v12.4s, v0.s[1]\n" + "add %[inptr0], %[inptr0], #16\n" + "mov v11.16b, v1.16b\n" + "add x16, x16, #16\n" + "mov v1.16b, v5.16b\n" + "add x17, x17, #16\n" + "fsub v8.4s, v8.4s, v2.4s\n" + "fmla v11.4s, v12.4s, v0.s[2]\n" + "fmls v8.4s, v4.4s, v0.s[1]\n" + "fmla v1.4s, v3.4s, v0.s[2]\n" + "mov v2.16b, v5.16b\n" + "mov v3.16b, v5.16b\n" + "fmls v11.4s, v4.4s, v0.s[3]\n" + "mov v4.16b, v5.16b\n" + "fmls v1.4s, v20.4s, v0.s[3]\n" + "fmls v2.4s, v22.4s, v0.s[2]\n" + "fmla v3.4s, v22.4s, v0.s[2]\n" + "fmls v4.4s, v22.4s, v0.s[1]\n" + "mov v5.16b, v5.16b\n" + "mov v6.16b, v6.16b\n" + "fmls v2.4s, v20.4s, v0.s[2]\n" + "mov v12.16b, v17.16b\n" + "fmls v3.4s, v20.4s, v0.s[2]\n" + "fsub v4.4s, v4.4s, v20.4s\n" + "fmla v4.4s, v16.4s, v0.s[1]\n" + "fmla v5.4s, v22.4s, v0.s[1]\n" + "fadd v2.4s, v2.4s, v16.4s\n" + "fmla v6.4s, v22.4s, v0.s[2]\n" + "fsub v3.4s, v3.4s, v16.4s\n" + "fmla v12.4s, v19.4s, v0.s[2]\n" + "fsub v5.4s, v5.4s, v20.4s\n" + "mov v19.16b, v17.16b\n" + "fmls v5.4s, v16.4s, v0.s[1]\n" + "fmls v6.4s, v16.4s, v0.s[3]\n" + "fmls v12.4s, v15.4s, v0.s[3]\n" + "fmls v19.4s, v21.4s, v0.s[2]\n" + "mov v20.16b, v17.16b\n" + "mov v16.16b, v17.16b\n" + "mov v17.16b, v17.16b\n" + "mov v18.16b, v18.16b\n" + "fmls v19.4s, v15.4s, v0.s[2]\n" + "fmla v20.4s, v21.4s, v0.s[2]\n" + "fmls v16.4s, v21.4s, v0.s[1]\n" + "fmla v17.4s, v21.4s, v0.s[1]\n" + "fmla v18.4s, v21.4s, v0.s[2]\n" + "mov v23.16b, v12.16b\n" + "fadd v19.4s, v19.4s, v13.4s\n" + "fmls v20.4s, v15.4s, v0.s[2]\n" + "fsub v16.4s, v16.4s, v15.4s\n" + "fsub v17.4s, v17.4s, v15.4s\n" + "fmla v16.4s, v13.4s, v0.s[1]\n" + "fmls v17.4s, v13.4s, v0.s[1]\n" + "fsub v20.4s, v20.4s, v13.4s\n" + "fmls v18.4s, v13.4s, v0.s[3]\n" + "fmla v23.4s, v14.4s, v0.s[2]\n" + "mov v15.16b, v19.16b\n" + "mov v14.16b, v20.16b\n" + "mov v24.16b, v16.16b\n" + "fmla v15.4s, v10.4s, v0.s[2]\n" + "mov v10.16b, v17.16b\n" + "fmls v23.4s, v1.4s, v0.s[3]\n" + "fmla v14.4s, v9.4s, v0.s[2]\n" + "fmla v24.4s, v7.4s, v0.s[2]\n" + "fmla v10.4s, v8.4s, v0.s[2]\n" + "fmls v15.4s, v2.4s, v0.s[3]\n" + "mov v7.16b, v18.16b\n" + "str q23, [%[outptr0]]\n" + "fmls v14.4s, v3.4s, v0.s[3]\n" + "fmls v24.4s, v4.4s, v0.s[3]\n" + "fmls v10.4s, v5.4s, v0.s[3]\n" + "str q15, [%[outptr0], %[output_col_stride1]]\n" + "fmla v7.4s, v11.4s, v0.s[2]\n" + "str q14, [%[outptr0], x11]\n" + "str q24, [%[outptr0], x13]\n" + "str q10, [%[outptr0], x23]\n" + "fmls v7.4s, v6.4s, v0.s[3]\n" + "str q7, [%[outptr0], x15]\n" + "add %[outptr0], %[outptr0], #16\n" + "mov v26.16b, v12.16b\n" + "mov v25.16b, v19.16b\n" + "ldr q11, [x25, x20]\n" + "mov v10.16b, v11.16b\n" + "ldr q23, [x25, x18]\n" + "mov v9.16b, v11.16b\n" + "ldr q7, [x25]\n" + "fmla v10.4s, v7.4s, v0.s[2]\n" + "ldr q13, [x25, x21]\n" + "mov v7.16b, v11.16b\n" + "ldr q31, [x25, x19]\n" + "mov v8.16b, v11.16b\n" + "ldr q21, [x25, %[input_col_stride1]]\n" + "fmls v10.4s, v23.4s, v0.s[3]\n" + "ldr q30, [x26, x20]\n" + "fmls v9.4s, v21.4s, v0.s[2]\n" + "ldr q29, [x26, x18]\n" + "fmla v7.4s, v21.4s, v0.s[2]\n" + "ldr q22, [x26]\n" + "fmls v8.4s, v21.4s, v0.s[1]\n" + "ldr q24, [x26, x21]\n" + "fmls v9.4s, v23.4s, v0.s[2]\n" + "ldr q27, [x26, x19]\n" + "fmls v7.4s, v23.4s, v0.s[2]\n" + "ldr q28, [x26, %[input_col_stride1]]\n" + "fsub v8.4s, v8.4s, v23.4s\n" + "add x25, x25, #16\n" + "fadd v9.4s, v9.4s, v31.4s\n" + "add x26, x26, #16\n" + "fsub v7.4s, v7.4s, v31.4s\n" + "fmla v8.4s, v31.4s, v0.s[1]\n" + "mov v11.16b, v11.16b\n" + "mov v15.16b, v13.16b\n" + "mov v14.16b, v30.16b\n" + "mov v13.16b, v30.16b\n" + "fmla v11.4s, v21.4s, v0.s[1]\n" + "fmla v15.4s, v21.4s, v0.s[2]\n" + "fmla v14.4s, v22.4s, v0.s[2]\n" + "fmls v13.4s, v28.4s, v0.s[2]\n" + "mov v21.16b, v30.16b\n" + "mov v22.16b, v30.16b\n" + "fsub v11.4s, v11.4s, v23.4s\n" + "fmls v15.4s, v31.4s, v0.s[3]\n" + "fmls v11.4s, v31.4s, v0.s[1]\n" + "fmls v14.4s, v29.4s, v0.s[3]\n" + "fmls v13.4s, v29.4s, v0.s[2]\n" + "fmla v21.4s, v28.4s, v0.s[2]\n" + "fmls v22.4s, v28.4s, v0.s[1]\n" + "mov v23.16b, v30.16b\n" + "mov v24.16b, v24.16b\n" + "fmls v26.4s, v10.4s, v0.s[2]\n" + "fadd v13.4s, v13.4s, v27.4s\n" + "fmls v21.4s, v29.4s, v0.s[2]\n" + "fsub v22.4s, v22.4s, v29.4s\n" + "fmla v23.4s, v28.4s, v0.s[1]\n" + "fmla v22.4s, v27.4s, v0.s[1]\n" + "fmla v24.4s, v28.4s, v0.s[2]\n" + "fsub v21.4s, v21.4s, v27.4s\n" + "fmls v26.4s, v1.4s, v0.s[2]\n" + "fsub v23.4s, v23.4s, v29.4s\n" + "fmls v25.4s, v9.4s, v0.s[2]\n" + "fmls v23.4s, v27.4s, v0.s[1]\n" + "fmls v24.4s, v27.4s, v0.s[3]\n" + "fadd v26.4s, v26.4s, v14.4s\n" + "mov v27.16b, v20.16b\n" + "str q26, [x28]\n" + "fmls v25.4s, v2.4s, v0.s[2]\n" + "fmls v27.4s, v7.4s, v0.s[2]\n" + "mov v31.16b, v16.16b\n" + "mov v30.16b, v17.16b\n" + "mov v29.16b, v18.16b\n" + "fadd v25.4s, v25.4s, v13.4s\n" + "fmls v31.4s, v8.4s, v0.s[2]\n" + "str q25, [x28, %[output_col_stride1]]\n" + "fmls v27.4s, v3.4s, v0.s[2]\n" + "fmls v30.4s, v11.4s, v0.s[2]\n" + "fmls v29.4s, v15.4s, v0.s[2]\n" + "fmls v31.4s, v4.4s, v0.s[2]\n" + "mov v26.16b, v12.16b\n" + "fadd v27.4s, v27.4s, v21.4s\n" + "mov v25.16b, v19.16b\n" + "str q27, [x28, x11]\n" + "fmls v30.4s, v5.4s, v0.s[2]\n" + "fadd v31.4s, v31.4s, v22.4s\n" + "fmls v29.4s, v6.4s, v0.s[2]\n" + "str q31, [x28, x13]\n" + "fmla v26.4s, v10.4s, v0.s[2]\n" + "fadd v30.4s, v30.4s, v23.4s\n" + "fmla v25.4s, v9.4s, v0.s[2]\n" + "str q30, [x28, x23]\n" + "fadd v29.4s, v29.4s, v24.4s\n" + "str q29, [x28, x15]\n" + "fmls v26.4s, v1.4s, v0.s[2]\n" + "fmls v25.4s, v2.4s, v0.s[2]\n" + "add x28, x28, #16\n" + "mov v30.16b, v20.16b\n" + "mov v29.16b, v16.16b\n" + "fsub v26.4s, v26.4s, v14.4s\n" + "mov v28.16b, v17.16b\n" + "str q26, [x22]\n" + "fsub v25.4s, v25.4s, v13.4s\n" + "str q25, [x22, %[output_col_stride1]]\n" + "fmla v30.4s, v7.4s, v0.s[2]\n" + "fmla v29.4s, v8.4s, v0.s[2]\n" + "fmla v28.4s, v11.4s, v0.s[2]\n" + "mov v26.16b, v18.16b\n" + "mov v25.16b, v12.16b\n" + "fmls v30.4s, v3.4s, v0.s[2]\n" + "mov v31.16b, v19.16b\n" + "fmls v29.4s, v4.4s, v0.s[2]\n" + "fmls v28.4s, v5.4s, v0.s[2]\n" + "fmla v26.4s, v15.4s, v0.s[2]\n" + "fmls v25.4s, v10.4s, v0.s[1]\n" + "fsub v30.4s, v30.4s, v21.4s\n" + "fmls v31.4s, v9.4s, v0.s[1]\n" + "str q30, [x22, x11]\n" + "fsub v29.4s, v29.4s, v22.4s\n" + "str q29, [x22, x13]\n" + "fsub v28.4s, v28.4s, v23.4s\n" + "str q28, [x22, x23]\n" + "fmls v26.4s, v6.4s, v0.s[2]\n" + "fsub v25.4s, v25.4s, v1.4s\n" + "fsub v31.4s, v31.4s, v2.4s\n" + "fmla v25.4s, v14.4s, v0.s[1]\n" + "fmla v31.4s, v13.4s, v0.s[1]\n" + "fsub v26.4s, v26.4s, v24.4s\n" + "mov v27.16b, v20.16b\n" + "str q26, [x22, x15]\n" + "mov v26.16b, v16.16b\n" + "str q25, [x12]\n" + "fmls v27.4s, v7.4s, v0.s[1]\n" + "str q31, [x12, %[output_col_stride1]]\n" + "fmls v26.4s, v8.4s, v0.s[1]\n" + "mov v25.16b, v17.16b\n" + "add x22, x22, #16\n" + "fsub v27.4s, v27.4s, v3.4s\n" + "mov v28.16b, v18.16b\n" + "fmla v27.4s, v21.4s, v0.s[1]\n" + "fsub v26.4s, v26.4s, v4.4s\n" + "fmla v26.4s, v22.4s, v0.s[1]\n" + "fmls v25.4s, v11.4s, v0.s[1]\n" + "fmls v28.4s, v15.4s, v0.s[1]\n" + "mov v12.16b, v12.16b\n" + "str q27, [x12, x11]\n" + "mov v19.16b, v19.16b\n" + "str q26, [x12, x13]\n" + "fsub v25.4s, v25.4s, v5.4s\n" + "fmla v25.4s, v23.4s, v0.s[1]\n" + "fsub v28.4s, v28.4s, v6.4s\n" + "fmla v28.4s, v24.4s, v0.s[1]\n" + "fmla v12.4s, v10.4s, v0.s[1]\n" + "fmla v19.4s, v9.4s, v0.s[1]\n" + "mov v20.16b, v20.16b\n" + "str q25, [x12, x23]\n" + "mov v16.16b, v16.16b\n" + "str q28, [x12, x15]\n" + "fsub v12.4s, v12.4s, v1.4s\n" + "fmls v12.4s, v14.4s, v0.s[1]\n" + "add x12, x12, #16\n" + "fsub v19.4s, v19.4s, v2.4s\n" + "fmla v20.4s, v7.4s, v0.s[1]\n" + "fmls v19.4s, v13.4s, v0.s[1]\n" + "fmla v16.4s, v8.4s, v0.s[1]\n" + "str q12, [x14]\n" + "mov v1.16b, v17.16b\n" + "fsub v20.4s, v20.4s, v3.4s\n" + "mov v17.16b, v18.16b\n" + "str q19, [x14, %[output_col_stride1]]\n" + "fmls v20.4s, v21.4s, v0.s[1]\n" + "fsub v16.4s, v16.4s, v4.4s\n" + "fmla v1.4s, v11.4s, v0.s[1]\n" + "fmls v16.4s, v22.4s, v0.s[1]\n" + "fmla v17.4s, v15.4s, v0.s[1]\n" + "str q20, [x14, x11]\n" + "fsub v1.4s, v1.4s, v5.4s\n" + "str q16, [x14, x13]\n" + "fmls v1.4s, v23.4s, v0.s[1]\n" + "fsub v17.4s, v17.4s, v6.4s\n" + "fmls v17.4s, v24.4s, v0.s[1]\n" + "str q1, [x14, x23]\n" + "str q17, [x14, x15]\n" + "add x14, x14, #16\n" + "ldr q2, [x27, x20]\n" + "mov v4.16b, v2.16b\n" + "ldr q17, [x27, x18]\n" + "mov v12.16b, v2.16b\n" + "ldr q18, [x27]\n" + "fmla v4.4s, v18.4s, v0.s[2]\n" + "ldr q3, [x27, x21]\n" + "mov v6.16b, v2.16b\n" + "ldr q5, [x27, x19]\n" + "mov v1.16b, v2.16b\n" + "ldr q18, [x27, %[input_col_stride1]]\n" + "fmls v4.4s, v17.4s, v0.s[3]\n" + "add x27, x27, #16\n" + "fmls v12.4s, v18.4s, v0.s[2]\n" + "sub %w[n_channels], %w[n_channels], #4\n" + "fmla v6.4s, v18.4s, v0.s[2]\n" + "cmp %w[n_channels], #4\n" + "fmls v1.4s, v18.4s, v0.s[1]\n" + "mov v2.16b, v2.16b\n" + "fmls v12.4s, v17.4s, v0.s[2]\n" + "mov v3.16b, v3.16b\n" + "fmls v6.4s, v17.4s, v0.s[2]\n" + "fmla v2.4s, v18.4s, v0.s[1]\n" + "fsub v1.4s, v1.4s, v17.4s\n" + "fmla v3.4s, v18.4s, v0.s[2]\n" + "fadd v12.4s, v12.4s, v5.4s\n" + "fmla v1.4s, v5.4s, v0.s[1]\n" + "fsub v6.4s, v6.4s, v5.4s\n" + "fsub v2.4s, v2.4s, v17.4s\n" + "fmls v2.4s, v5.4s, v0.s[1]\n" + "fmls v3.4s, v5.4s, v0.s[3]\n" + "mov v4.16b, v4.16b\n" + "mov v16.16b, v12.16b\n" + "mov v5.16b, v6.16b\n" + "mov v6.16b, v1.16b\n" + "fmla v4.4s, v10.4s, v0.s[2]\n" + "fmla v16.4s, v9.4s, v0.s[2]\n" + "fmla v5.4s, v7.4s, v0.s[2]\n" + "fmla v6.4s, v8.4s, v0.s[2]\n" + "mov v9.16b, v2.16b\n" + "mov v10.16b, v3.16b\n" + "fmls v4.4s, v14.4s, v0.s[3]\n" + "fmls v16.4s, v13.4s, v0.s[3]\n" + "fmls v5.4s, v21.4s, v0.s[3]\n" + "fmls v6.4s, v22.4s, v0.s[3]\n" + "fmla v9.4s, v11.4s, v0.s[2]\n" + "fmla v10.4s, v15.4s, v0.s[2]\n" + "str q4, [x24]\n" + "str q16, [x24, %[output_col_stride1]]\n" + "str q5, [x24, x11]\n" + "str q6, [x24, x13]\n" + "fmls v9.4s, v23.4s, v0.s[3]\n" + "fmls v10.4s, v24.4s, v0.s[3]\n" + "str q9, [x24, x23]\n" + "str q10, [x24, x15]\n" + "add x24, x24, #16\n" + "bge 1b\n" + "2:\n" + "cmp %w[n_channels], #2\n" + "blt 3f\n" + "ldr d8, [%[inptr0], x20]\n" + "mov v14.16b, v8.16b\n" + "ldr d2, [%[inptr0], x18]\n" + "mov v10.16b, v8.16b\n" + "ldr d9, [%[inptr0]]\n" + "fmla v14.4s, v9.4s, v0.s[2]\n" + "ldr d1, [%[inptr0], x21]\n" + "mov v9.16b, v8.16b\n" + "ldr d4, [%[inptr0], x19]\n" + "mov v7.16b, v8.16b\n" + "ldr d12, [%[inptr0], %[input_col_stride1]]\n" + "fmls v14.4s, v2.4s, v0.s[3]\n" + "ldr d5, [x16, x20]\n" + "fmls v10.4s, v12.4s, v0.s[2]\n" + "ldr d20, [x16, x18]\n" + "fmla v9.4s, v12.4s, v0.s[2]\n" + "ldr d3, [x16]\n" + "fmls v7.4s, v12.4s, v0.s[1]\n" + "ldr d6, [x16, x21]\n" + "fmls v10.4s, v2.4s, v0.s[2]\n" + "ldr d16, [x16, x19]\n" + "fmls v9.4s, v2.4s, v0.s[2]\n" + "ldr d22, [x16, %[input_col_stride1]]\n" + "fsub v7.4s, v7.4s, v2.4s\n" + "ldr d17, [x17, x20]\n" + "fadd v10.4s, v10.4s, v4.4s\n" + "ldr d15, [x17, x18]\n" + "fsub v9.4s, v9.4s, v4.4s\n" + "ldr d19, [x17]\n" + "fmla v7.4s, v4.4s, v0.s[1]\n" + "ldr d18, [x17, x21]\n" + "mov v8.16b, v8.16b\n" + "ldr d13, [x17, x19]\n" + "mov v11.16b, v1.16b\n" + "ldr d21, [x17, %[input_col_stride1]]\n" + "fmla v8.4s, v12.4s, v0.s[1]\n" + "add %[inptr0], %[inptr0], #8\n" + "fmla v11.4s, v12.4s, v0.s[2]\n" + "add x16, x16, #8\n" + "mov v1.16b, v5.16b\n" + "add x17, x17, #8\n" + "fsub v8.4s, v8.4s, v2.4s\n" + "mov v2.16b, v5.16b\n" + "fmls v8.4s, v4.4s, v0.s[1]\n" + "fmls v11.4s, v4.4s, v0.s[3]\n" + "fmla v1.4s, v3.4s, v0.s[2]\n" + "fmls v2.4s, v22.4s, v0.s[2]\n" + "mov v3.16b, v5.16b\n" + "mov v4.16b, v5.16b\n" + "mov v5.16b, v5.16b\n" + "mov v6.16b, v6.16b\n" + "fmls v1.4s, v20.4s, v0.s[3]\n" + "fmls v2.4s, v20.4s, v0.s[2]\n" + "fmla v3.4s, v22.4s, v0.s[2]\n" + "fmls v4.4s, v22.4s, v0.s[1]\n" + "fmla v5.4s, v22.4s, v0.s[1]\n" + "fmla v6.4s, v22.4s, v0.s[2]\n" + "fadd v2.4s, v2.4s, v16.4s\n" + "mov v12.16b, v17.16b\n" + "fmls v3.4s, v20.4s, v0.s[2]\n" + "fsub v4.4s, v4.4s, v20.4s\n" + "fmla v4.4s, v16.4s, v0.s[1]\n" + "fsub v5.4s, v5.4s, v20.4s\n" + "fmls v5.4s, v16.4s, v0.s[1]\n" + "fmls v6.4s, v16.4s, v0.s[3]\n" + "fsub v3.4s, v3.4s, v16.4s\n" + "fmla v12.4s, v19.4s, v0.s[2]\n" + "mov v19.16b, v17.16b\n" + "mov v20.16b, v17.16b\n" + "mov v16.16b, v17.16b\n" + "mov v17.16b, v17.16b\n" + "fmls v12.4s, v15.4s, v0.s[3]\n" + "fmls v19.4s, v21.4s, v0.s[2]\n" + "fmla v20.4s, v21.4s, v0.s[2]\n" + "fmls v16.4s, v21.4s, v0.s[1]\n" + "fmla v17.4s, v21.4s, v0.s[1]\n" + "mov v18.16b, v18.16b\n" + "fmls v19.4s, v15.4s, v0.s[2]\n" + "mov v23.16b, v12.16b\n" + "fmls v20.4s, v15.4s, v0.s[2]\n" + "fsub v16.4s, v16.4s, v15.4s\n" + "fmla v16.4s, v13.4s, v0.s[1]\n" + "fsub v17.4s, v17.4s, v15.4s\n" + "fadd v19.4s, v19.4s, v13.4s\n" + "fmls v17.4s, v13.4s, v0.s[1]\n" + "fsub v20.4s, v20.4s, v13.4s\n" + "fmla v18.4s, v21.4s, v0.s[2]\n" + "fmla v23.4s, v14.4s, v0.s[2]\n" + "mov v15.16b, v19.16b\n" + "mov v14.16b, v20.16b\n" + "mov v24.16b, v16.16b\n" + "fmls v18.4s, v13.4s, v0.s[3]\n" + "fmla v15.4s, v10.4s, v0.s[2]\n" + "fmls v23.4s, v1.4s, v0.s[3]\n" + "fmla v14.4s, v9.4s, v0.s[2]\n" + "fmla v24.4s, v7.4s, v0.s[2]\n" + "mov v10.16b, v17.16b\n" + "fmls v15.4s, v2.4s, v0.s[3]\n" + "mov v7.16b, v18.16b\n" + "str d23, [%[outptr0]]\n" + "fmls v14.4s, v3.4s, v0.s[3]\n" + "fmls v24.4s, v4.4s, v0.s[3]\n" + "fmla v10.4s, v8.4s, v0.s[2]\n" + "str d15, [%[outptr0], %[output_col_stride1]]\n" + "fmla v7.4s, v11.4s, v0.s[2]\n" + "str d14, [%[outptr0], x11]\n" + "fmls v10.4s, v5.4s, v0.s[3]\n" + "str d24, [%[outptr0], x13]\n" + "fmls v7.4s, v6.4s, v0.s[3]\n" + "str d10, [%[outptr0], x23]\n" + "str d7, [%[outptr0], x15]\n" + "add %[outptr0], %[outptr0], #8\n" + "mov v26.16b, v12.16b\n" + "mov v25.16b, v19.16b\n" + "ldr d11, [x25, x20]\n" + "mov v10.16b, v11.16b\n" + "ldr d23, [x25, x18]\n" + "mov v9.16b, v11.16b\n" + "ldr d7, [x25]\n" + "fmla v10.4s, v7.4s, v0.s[2]\n" + "ldr d13, [x25, x21]\n" + "mov v7.16b, v11.16b\n" + "ldr d31, [x25, x19]\n" + "mov v8.16b, v11.16b\n" + "ldr d21, [x25, %[input_col_stride1]]\n" + "fmls v10.4s, v23.4s, v0.s[3]\n" + "ldr d30, [x26, x20]\n" + "fmls v9.4s, v21.4s, v0.s[2]\n" + "ldr d29, [x26, x18]\n" + "fmla v7.4s, v21.4s, v0.s[2]\n" + "ldr d22, [x26]\n" + "fmls v8.4s, v21.4s, v0.s[1]\n" + "ldr d24, [x26, x21]\n" + "fmls v9.4s, v23.4s, v0.s[2]\n" + "ldr d27, [x26, x19]\n" + "fmls v7.4s, v23.4s, v0.s[2]\n" + "ldr d28, [x26, %[input_col_stride1]]\n" + "fsub v8.4s, v8.4s, v23.4s\n" + "add x25, x25, #8\n" + "fadd v9.4s, v9.4s, v31.4s\n" + "add x26, x26, #8\n" + "fsub v7.4s, v7.4s, v31.4s\n" + "fmla v8.4s, v31.4s, v0.s[1]\n" + "mov v11.16b, v11.16b\n" + "mov v15.16b, v13.16b\n" + "mov v14.16b, v30.16b\n" + "mov v13.16b, v30.16b\n" + "fmla v11.4s, v21.4s, v0.s[1]\n" + "fmla v15.4s, v21.4s, v0.s[2]\n" + "fmla v14.4s, v22.4s, v0.s[2]\n" + "fmls v13.4s, v28.4s, v0.s[2]\n" + "mov v21.16b, v30.16b\n" + "mov v22.16b, v30.16b\n" + "fsub v11.4s, v11.4s, v23.4s\n" + "fmls v15.4s, v31.4s, v0.s[3]\n" + "fmls v11.4s, v31.4s, v0.s[1]\n" + "fmls v14.4s, v29.4s, v0.s[3]\n" + "fmls v13.4s, v29.4s, v0.s[2]\n" + "fmla v21.4s, v28.4s, v0.s[2]\n" + "fmls v22.4s, v28.4s, v0.s[1]\n" + "mov v23.16b, v30.16b\n" + "mov v24.16b, v24.16b\n" + "fmls v26.4s, v10.4s, v0.s[2]\n" + "fadd v13.4s, v13.4s, v27.4s\n" + "fmls v21.4s, v29.4s, v0.s[2]\n" + "fsub v22.4s, v22.4s, v29.4s\n" + "fmla v23.4s, v28.4s, v0.s[1]\n" + "fmla v22.4s, v27.4s, v0.s[1]\n" + "fmla v24.4s, v28.4s, v0.s[2]\n" + "fsub v21.4s, v21.4s, v27.4s\n" + "fmls v26.4s, v1.4s, v0.s[2]\n" + "fsub v23.4s, v23.4s, v29.4s\n" + "fmls v25.4s, v9.4s, v0.s[2]\n" + "fmls v23.4s, v27.4s, v0.s[1]\n" + "fmls v24.4s, v27.4s, v0.s[3]\n" + "fadd v26.4s, v26.4s, v14.4s\n" + "mov v27.16b, v20.16b\n" + "str d26, [x28]\n" + "fmls v25.4s, v2.4s, v0.s[2]\n" + "fmls v27.4s, v7.4s, v0.s[2]\n" + "mov v31.16b, v16.16b\n" + "mov v30.16b, v17.16b\n" + "mov v29.16b, v18.16b\n" + "fadd v25.4s, v25.4s, v13.4s\n" + "fmls v31.4s, v8.4s, v0.s[2]\n" + "str d25, [x28, %[output_col_stride1]]\n" + "fmls v27.4s, v3.4s, v0.s[2]\n" + "fmls v30.4s, v11.4s, v0.s[2]\n" + "fmls v29.4s, v15.4s, v0.s[2]\n" + "fmls v31.4s, v4.4s, v0.s[2]\n" + "mov v26.16b, v12.16b\n" + "fadd v27.4s, v27.4s, v21.4s\n" + "mov v25.16b, v19.16b\n" + "str d27, [x28, x11]\n" + "fmls v30.4s, v5.4s, v0.s[2]\n" + "fadd v31.4s, v31.4s, v22.4s\n" + "fmls v29.4s, v6.4s, v0.s[2]\n" + "str d31, [x28, x13]\n" + "fmla v26.4s, v10.4s, v0.s[2]\n" + "fadd v30.4s, v30.4s, v23.4s\n" + "fmla v25.4s, v9.4s, v0.s[2]\n" + "str d30, [x28, x23]\n" + "fadd v29.4s, v29.4s, v24.4s\n" + "str d29, [x28, x15]\n" + "fmls v26.4s, v1.4s, v0.s[2]\n" + "fmls v25.4s, v2.4s, v0.s[2]\n" + "add x28, x28, #8\n" + "mov v30.16b, v20.16b\n" + "mov v29.16b, v16.16b\n" + "fsub v26.4s, v26.4s, v14.4s\n" + "mov v28.16b, v17.16b\n" + "str d26, [x22]\n" + "fsub v25.4s, v25.4s, v13.4s\n" + "str d25, [x22, %[output_col_stride1]]\n" + "fmla v30.4s, v7.4s, v0.s[2]\n" + "fmla v29.4s, v8.4s, v0.s[2]\n" + "fmla v28.4s, v11.4s, v0.s[2]\n" + "mov v26.16b, v18.16b\n" + "mov v25.16b, v12.16b\n" + "fmls v30.4s, v3.4s, v0.s[2]\n" + "mov v31.16b, v19.16b\n" + "fmls v29.4s, v4.4s, v0.s[2]\n" + "fmls v28.4s, v5.4s, v0.s[2]\n" + "fmla v26.4s, v15.4s, v0.s[2]\n" + "fmls v25.4s, v10.4s, v0.s[1]\n" + "fsub v30.4s, v30.4s, v21.4s\n" + "fmls v31.4s, v9.4s, v0.s[1]\n" + "str d30, [x22, x11]\n" + "fsub v29.4s, v29.4s, v22.4s\n" + "str d29, [x22, x13]\n" + "fsub v28.4s, v28.4s, v23.4s\n" + "str d28, [x22, x23]\n" + "fmls v26.4s, v6.4s, v0.s[2]\n" + "fsub v25.4s, v25.4s, v1.4s\n" + "fsub v31.4s, v31.4s, v2.4s\n" + "fmla v25.4s, v14.4s, v0.s[1]\n" + "fmla v31.4s, v13.4s, v0.s[1]\n" + "fsub v26.4s, v26.4s, v24.4s\n" + "mov v27.16b, v20.16b\n" + "str d26, [x22, x15]\n" + "mov v26.16b, v16.16b\n" + "str d25, [x12]\n" + "fmls v27.4s, v7.4s, v0.s[1]\n" + "str d31, [x12, %[output_col_stride1]]\n" + "fmls v26.4s, v8.4s, v0.s[1]\n" + "mov v25.16b, v17.16b\n" + "add x22, x22, #8\n" + "fsub v27.4s, v27.4s, v3.4s\n" + "mov v28.16b, v18.16b\n" + "fmla v27.4s, v21.4s, v0.s[1]\n" + "fsub v26.4s, v26.4s, v4.4s\n" + "fmla v26.4s, v22.4s, v0.s[1]\n" + "fmls v25.4s, v11.4s, v0.s[1]\n" + "fmls v28.4s, v15.4s, v0.s[1]\n" + "mov v12.16b, v12.16b\n" + "str d27, [x12, x11]\n" + "mov v19.16b, v19.16b\n" + "str d26, [x12, x13]\n" + "fsub v25.4s, v25.4s, v5.4s\n" + "fmla v25.4s, v23.4s, v0.s[1]\n" + "fsub v28.4s, v28.4s, v6.4s\n" + "fmla v28.4s, v24.4s, v0.s[1]\n" + "fmla v12.4s, v10.4s, v0.s[1]\n" + "fmla v19.4s, v9.4s, v0.s[1]\n" + "mov v20.16b, v20.16b\n" + "str d25, [x12, x23]\n" + "mov v16.16b, v16.16b\n" + "str d28, [x12, x15]\n" + "fsub v12.4s, v12.4s, v1.4s\n" + "fmls v12.4s, v14.4s, v0.s[1]\n" + "add x12, x12, #8\n" + "fsub v19.4s, v19.4s, v2.4s\n" + "fmla v20.4s, v7.4s, v0.s[1]\n" + "fmls v19.4s, v13.4s, v0.s[1]\n" + "fmla v16.4s, v8.4s, v0.s[1]\n" + "str d12, [x14]\n" + "mov v1.16b, v17.16b\n" + "fsub v20.4s, v20.4s, v3.4s\n" + "mov v17.16b, v18.16b\n" + "str d19, [x14, %[output_col_stride1]]\n" + "fmls v20.4s, v21.4s, v0.s[1]\n" + "fsub v16.4s, v16.4s, v4.4s\n" + "fmla v1.4s, v11.4s, v0.s[1]\n" + "fmls v16.4s, v22.4s, v0.s[1]\n" + "fmla v17.4s, v15.4s, v0.s[1]\n" + "str d20, [x14, x11]\n" + "fsub v1.4s, v1.4s, v5.4s\n" + "str d16, [x14, x13]\n" + "fmls v1.4s, v23.4s, v0.s[1]\n" + "fsub v17.4s, v17.4s, v6.4s\n" + "fmls v17.4s, v24.4s, v0.s[1]\n" + "str d1, [x14, x23]\n" + "str d17, [x14, x15]\n" + "add x14, x14, #8\n" + "ldr d2, [x27, x20]\n" + "mov v4.16b, v2.16b\n" + "ldr d17, [x27, x18]\n" + "mov v12.16b, v2.16b\n" + "ldr d18, [x27]\n" + "fmla v4.4s, v18.4s, v0.s[2]\n" + "ldr d3, [x27, x21]\n" + "mov v6.16b, v2.16b\n" + "ldr d5, [x27, x19]\n" + "mov v1.16b, v2.16b\n" + "ldr d18, [x27, %[input_col_stride1]]\n" + "fmls v4.4s, v17.4s, v0.s[3]\n" + "add x27, x27, #8\n" + "fmls v12.4s, v18.4s, v0.s[2]\n" + "sub %w[n_channels], %w[n_channels], #2\n" + "fmla v6.4s, v18.4s, v0.s[2]\n" + "fmls v1.4s, v18.4s, v0.s[1]\n" + "mov v2.16b, v2.16b\n" + "mov v3.16b, v3.16b\n" + "fmls v12.4s, v17.4s, v0.s[2]\n" + "mov v4.16b, v4.16b\n" + "fmls v6.4s, v17.4s, v0.s[2]\n" + "fsub v1.4s, v1.4s, v17.4s\n" + "fmla v1.4s, v5.4s, v0.s[1]\n" + "fmla v2.4s, v18.4s, v0.s[1]\n" + "fadd v12.4s, v12.4s, v5.4s\n" + "fmla v3.4s, v18.4s, v0.s[2]\n" + "fsub v6.4s, v6.4s, v5.4s\n" + "fmla v4.4s, v10.4s, v0.s[2]\n" + "fsub v2.4s, v2.4s, v17.4s\n" + "mov v16.16b, v12.16b\n" + "fmls v2.4s, v5.4s, v0.s[1]\n" + "fmls v3.4s, v5.4s, v0.s[3]\n" + "fmls v4.4s, v14.4s, v0.s[3]\n" + "fmla v16.4s, v9.4s, v0.s[2]\n" + "mov v5.16b, v6.16b\n" + "mov v6.16b, v1.16b\n" + "mov v9.16b, v2.16b\n" + "mov v10.16b, v3.16b\n" + "str d4, [x24]\n" + "fmls v16.4s, v13.4s, v0.s[3]\n" + "fmla v5.4s, v7.4s, v0.s[2]\n" + "fmla v6.4s, v8.4s, v0.s[2]\n" + "fmla v9.4s, v11.4s, v0.s[2]\n" + "fmla v10.4s, v15.4s, v0.s[2]\n" + "str d16, [x24, %[output_col_stride1]]\n" + "fmls v5.4s, v21.4s, v0.s[3]\n" + "fmls v6.4s, v22.4s, v0.s[3]\n" + "fmls v9.4s, v23.4s, v0.s[3]\n" + "fmls v10.4s, v24.4s, v0.s[3]\n" + "str d5, [x24, x11]\n" + "str d6, [x24, x13]\n" + "str d9, [x24, x23]\n" + "str d10, [x24, x15]\n" + "add x24, x24, #8\n" + "3:\n" + "cbz %w[n_channels], 4f\n" + "ldr s8, [%[inptr0], x20]\n" + "mov v14.16b, v8.16b\n" + "ldr s2, [%[inptr0], x18]\n" + "mov v10.16b, v8.16b\n" + "ldr s9, [%[inptr0]]\n" + "fmla v14.4s, v9.4s, v0.s[2]\n" + "ldr s1, [%[inptr0], x21]\n" + "mov v9.16b, v8.16b\n" + "ldr s4, [%[inptr0], x19]\n" + "mov v7.16b, v8.16b\n" + "ldr s12, [%[inptr0], %[input_col_stride1]]\n" + "fmls v14.4s, v2.4s, v0.s[3]\n" + "ldr s5, [x16, x20]\n" + "fmls v10.4s, v12.4s, v0.s[2]\n" + "ldr s20, [x16, x18]\n" + "fmla v9.4s, v12.4s, v0.s[2]\n" + "ldr s3, [x16]\n" + "fmls v7.4s, v12.4s, v0.s[1]\n" + "ldr s6, [x16, x21]\n" + "fmls v10.4s, v2.4s, v0.s[2]\n" + "ldr s16, [x16, x19]\n" + "fmls v9.4s, v2.4s, v0.s[2]\n" + "ldr s22, [x16, %[input_col_stride1]]\n" + "fsub v7.4s, v7.4s, v2.4s\n" + "ldr s17, [x17, x20]\n" + "fadd v10.4s, v10.4s, v4.4s\n" + "ldr s15, [x17, x18]\n" + "fsub v9.4s, v9.4s, v4.4s\n" + "ldr s19, [x17]\n" + "fmla v7.4s, v4.4s, v0.s[1]\n" + "ldr s18, [x17, x21]\n" + "mov v8.16b, v8.16b\n" + "ldr s13, [x17, x19]\n" + "mov v11.16b, v1.16b\n" + "ldr s21, [x17, %[input_col_stride1]]\n" + "fmla v8.4s, v12.4s, v0.s[1]\n" + "add %[inptr0], %[inptr0], #4\n" + "fmla v11.4s, v12.4s, v0.s[2]\n" + "add x16, x16, #4\n" + "mov v1.16b, v5.16b\n" + "add x17, x17, #4\n" + "fsub v8.4s, v8.4s, v2.4s\n" + "mov v2.16b, v5.16b\n" + "fmls v8.4s, v4.4s, v0.s[1]\n" + "fmls v11.4s, v4.4s, v0.s[3]\n" + "fmla v1.4s, v3.4s, v0.s[2]\n" + "fmls v2.4s, v22.4s, v0.s[2]\n" + "mov v3.16b, v5.16b\n" + "mov v4.16b, v5.16b\n" + "mov v5.16b, v5.16b\n" + "mov v6.16b, v6.16b\n" + "fmls v1.4s, v20.4s, v0.s[3]\n" + "fmls v2.4s, v20.4s, v0.s[2]\n" + "fmla v3.4s, v22.4s, v0.s[2]\n" + "fmls v4.4s, v22.4s, v0.s[1]\n" + "fmla v5.4s, v22.4s, v0.s[1]\n" + "fmla v6.4s, v22.4s, v0.s[2]\n" + "fadd v2.4s, v2.4s, v16.4s\n" + "mov v12.16b, v17.16b\n" + "fmls v3.4s, v20.4s, v0.s[2]\n" + "fsub v4.4s, v4.4s, v20.4s\n" + "fmla v4.4s, v16.4s, v0.s[1]\n" + "fsub v5.4s, v5.4s, v20.4s\n" + "fmls v5.4s, v16.4s, v0.s[1]\n" + "fmls v6.4s, v16.4s, v0.s[3]\n" + "fsub v3.4s, v3.4s, v16.4s\n" + "fmla v12.4s, v19.4s, v0.s[2]\n" + "mov v19.16b, v17.16b\n" + "mov v20.16b, v17.16b\n" + "mov v16.16b, v17.16b\n" + "mov v17.16b, v17.16b\n" + "fmls v12.4s, v15.4s, v0.s[3]\n" + "fmls v19.4s, v21.4s, v0.s[2]\n" + "fmla v20.4s, v21.4s, v0.s[2]\n" + "fmls v16.4s, v21.4s, v0.s[1]\n" + "fmla v17.4s, v21.4s, v0.s[1]\n" + "mov v18.16b, v18.16b\n" + "fmls v19.4s, v15.4s, v0.s[2]\n" + "mov v23.16b, v12.16b\n" + "fmls v20.4s, v15.4s, v0.s[2]\n" + "fsub v16.4s, v16.4s, v15.4s\n" + "fmla v16.4s, v13.4s, v0.s[1]\n" + "fsub v17.4s, v17.4s, v15.4s\n" + "fadd v19.4s, v19.4s, v13.4s\n" + "fmls v17.4s, v13.4s, v0.s[1]\n" + "fsub v20.4s, v20.4s, v13.4s\n" + "fmla v18.4s, v21.4s, v0.s[2]\n" + "fmla v23.4s, v14.4s, v0.s[2]\n" + "mov v15.16b, v19.16b\n" + "mov v14.16b, v20.16b\n" + "mov v24.16b, v16.16b\n" + "fmls v18.4s, v13.4s, v0.s[3]\n" + "fmla v15.4s, v10.4s, v0.s[2]\n" + "fmls v23.4s, v1.4s, v0.s[3]\n" + "fmla v14.4s, v9.4s, v0.s[2]\n" + "fmla v24.4s, v7.4s, v0.s[2]\n" + "mov v10.16b, v17.16b\n" + "fmls v15.4s, v2.4s, v0.s[3]\n" + "mov v7.16b, v18.16b\n" + "str s23, [%[outptr0]]\n" + "fmls v14.4s, v3.4s, v0.s[3]\n" + "fmls v24.4s, v4.4s, v0.s[3]\n" + "fmla v10.4s, v8.4s, v0.s[2]\n" + "str s15, [%[outptr0], %[output_col_stride1]]\n" + "fmla v7.4s, v11.4s, v0.s[2]\n" + "str s14, [%[outptr0], x11]\n" + "fmls v10.4s, v5.4s, v0.s[3]\n" + "str s24, [%[outptr0], x13]\n" + "fmls v7.4s, v6.4s, v0.s[3]\n" + "str s10, [%[outptr0], x23]\n" + "str s7, [%[outptr0], x15]\n" + "add %[outptr0], %[outptr0], #4\n" + "mov v26.16b, v12.16b\n" + "mov v25.16b, v19.16b\n" + "ldr s11, [x25, x20]\n" + "mov v10.16b, v11.16b\n" + "ldr s23, [x25, x18]\n" + "mov v9.16b, v11.16b\n" + "ldr s7, [x25]\n" + "fmla v10.4s, v7.4s, v0.s[2]\n" + "ldr s13, [x25, x21]\n" + "mov v7.16b, v11.16b\n" + "ldr s31, [x25, x19]\n" + "mov v8.16b, v11.16b\n" + "ldr s21, [x25, %[input_col_stride1]]\n" + "fmls v10.4s, v23.4s, v0.s[3]\n" + "ldr s30, [x26, x20]\n" + "fmls v9.4s, v21.4s, v0.s[2]\n" + "ldr s29, [x26, x18]\n" + "fmla v7.4s, v21.4s, v0.s[2]\n" + "ldr s22, [x26]\n" + "fmls v8.4s, v21.4s, v0.s[1]\n" + "ldr s24, [x26, x21]\n" + "fmls v9.4s, v23.4s, v0.s[2]\n" + "ldr s27, [x26, x19]\n" + "fmls v7.4s, v23.4s, v0.s[2]\n" + "ldr s28, [x26, %[input_col_stride1]]\n" + "fsub v8.4s, v8.4s, v23.4s\n" + "add x25, x25, #4\n" + "fadd v9.4s, v9.4s, v31.4s\n" + "add x26, x26, #4\n" + "fsub v7.4s, v7.4s, v31.4s\n" + "fmla v8.4s, v31.4s, v0.s[1]\n" + "mov v11.16b, v11.16b\n" + "mov v15.16b, v13.16b\n" + "mov v14.16b, v30.16b\n" + "mov v13.16b, v30.16b\n" + "fmla v11.4s, v21.4s, v0.s[1]\n" + "fmla v15.4s, v21.4s, v0.s[2]\n" + "fmla v14.4s, v22.4s, v0.s[2]\n" + "fmls v13.4s, v28.4s, v0.s[2]\n" + "mov v21.16b, v30.16b\n" + "mov v22.16b, v30.16b\n" + "fsub v11.4s, v11.4s, v23.4s\n" + "fmls v15.4s, v31.4s, v0.s[3]\n" + "fmls v11.4s, v31.4s, v0.s[1]\n" + "fmls v14.4s, v29.4s, v0.s[3]\n" + "fmls v13.4s, v29.4s, v0.s[2]\n" + "fmla v21.4s, v28.4s, v0.s[2]\n" + "fmls v22.4s, v28.4s, v0.s[1]\n" + "mov v23.16b, v30.16b\n" + "mov v24.16b, v24.16b\n" + "fmls v26.4s, v10.4s, v0.s[2]\n" + "fadd v13.4s, v13.4s, v27.4s\n" + "fmls v21.4s, v29.4s, v0.s[2]\n" + "fsub v22.4s, v22.4s, v29.4s\n" + "fmla v23.4s, v28.4s, v0.s[1]\n" + "fmla v22.4s, v27.4s, v0.s[1]\n" + "fmla v24.4s, v28.4s, v0.s[2]\n" + "fsub v21.4s, v21.4s, v27.4s\n" + "fmls v26.4s, v1.4s, v0.s[2]\n" + "fsub v23.4s, v23.4s, v29.4s\n" + "fmls v25.4s, v9.4s, v0.s[2]\n" + "fmls v23.4s, v27.4s, v0.s[1]\n" + "fmls v24.4s, v27.4s, v0.s[3]\n" + "fadd v26.4s, v26.4s, v14.4s\n" + "mov v27.16b, v20.16b\n" + "str s26, [x28]\n" + "fmls v25.4s, v2.4s, v0.s[2]\n" + "fmls v27.4s, v7.4s, v0.s[2]\n" + "mov v31.16b, v16.16b\n" + "mov v30.16b, v17.16b\n" + "mov v29.16b, v18.16b\n" + "fadd v25.4s, v25.4s, v13.4s\n" + "fmls v31.4s, v8.4s, v0.s[2]\n" + "str s25, [x28, %[output_col_stride1]]\n" + "fmls v27.4s, v3.4s, v0.s[2]\n" + "fmls v30.4s, v11.4s, v0.s[2]\n" + "fmls v29.4s, v15.4s, v0.s[2]\n" + "fmls v31.4s, v4.4s, v0.s[2]\n" + "mov v26.16b, v12.16b\n" + "fadd v27.4s, v27.4s, v21.4s\n" + "mov v25.16b, v19.16b\n" + "str s27, [x28, x11]\n" + "fmls v30.4s, v5.4s, v0.s[2]\n" + "fadd v31.4s, v31.4s, v22.4s\n" + "fmls v29.4s, v6.4s, v0.s[2]\n" + "str s31, [x28, x13]\n" + "fmla v26.4s, v10.4s, v0.s[2]\n" + "fadd v30.4s, v30.4s, v23.4s\n" + "fmla v25.4s, v9.4s, v0.s[2]\n" + "str s30, [x28, x23]\n" + "fadd v29.4s, v29.4s, v24.4s\n" + "str s29, [x28, x15]\n" + "fmls v26.4s, v1.4s, v0.s[2]\n" + "fmls v25.4s, v2.4s, v0.s[2]\n" + "add x28, x28, #4\n" + "mov v30.16b, v20.16b\n" + "mov v29.16b, v16.16b\n" + "fsub v26.4s, v26.4s, v14.4s\n" + "mov v28.16b, v17.16b\n" + "str s26, [x22]\n" + "fsub v25.4s, v25.4s, v13.4s\n" + "str s25, [x22, %[output_col_stride1]]\n" + "fmla v30.4s, v7.4s, v0.s[2]\n" + "fmla v29.4s, v8.4s, v0.s[2]\n" + "fmla v28.4s, v11.4s, v0.s[2]\n" + "mov v26.16b, v18.16b\n" + "mov v25.16b, v12.16b\n" + "fmls v30.4s, v3.4s, v0.s[2]\n" + "mov v31.16b, v19.16b\n" + "fmls v29.4s, v4.4s, v0.s[2]\n" + "fmls v28.4s, v5.4s, v0.s[2]\n" + "fmla v26.4s, v15.4s, v0.s[2]\n" + "fmls v25.4s, v10.4s, v0.s[1]\n" + "fsub v30.4s, v30.4s, v21.4s\n" + "fmls v31.4s, v9.4s, v0.s[1]\n" + "str s30, [x22, x11]\n" + "fsub v29.4s, v29.4s, v22.4s\n" + "str s29, [x22, x13]\n" + "fsub v28.4s, v28.4s, v23.4s\n" + "str s28, [x22, x23]\n" + "fmls v26.4s, v6.4s, v0.s[2]\n" + "fsub v25.4s, v25.4s, v1.4s\n" + "fsub v31.4s, v31.4s, v2.4s\n" + "fmla v25.4s, v14.4s, v0.s[1]\n" + "fmla v31.4s, v13.4s, v0.s[1]\n" + "fsub v26.4s, v26.4s, v24.4s\n" + "mov v27.16b, v20.16b\n" + "str s26, [x22, x15]\n" + "mov v26.16b, v16.16b\n" + "str s25, [x12]\n" + "fmls v27.4s, v7.4s, v0.s[1]\n" + "str s31, [x12, %[output_col_stride1]]\n" + "fmls v26.4s, v8.4s, v0.s[1]\n" + "mov v25.16b, v17.16b\n" + "add x22, x22, #4\n" + "fsub v27.4s, v27.4s, v3.4s\n" + "mov v28.16b, v18.16b\n" + "fmla v27.4s, v21.4s, v0.s[1]\n" + "fsub v26.4s, v26.4s, v4.4s\n" + "fmla v26.4s, v22.4s, v0.s[1]\n" + "fmls v25.4s, v11.4s, v0.s[1]\n" + "fmls v28.4s, v15.4s, v0.s[1]\n" + "mov v12.16b, v12.16b\n" + "str s27, [x12, x11]\n" + "mov v19.16b, v19.16b\n" + "str s26, [x12, x13]\n" + "fsub v25.4s, v25.4s, v5.4s\n" + "fmla v25.4s, v23.4s, v0.s[1]\n" + "fsub v28.4s, v28.4s, v6.4s\n" + "fmla v28.4s, v24.4s, v0.s[1]\n" + "fmla v12.4s, v10.4s, v0.s[1]\n" + "fmla v19.4s, v9.4s, v0.s[1]\n" + "mov v20.16b, v20.16b\n" + "str s25, [x12, x23]\n" + "mov v16.16b, v16.16b\n" + "str s28, [x12, x15]\n" + "fsub v12.4s, v12.4s, v1.4s\n" + "fmls v12.4s, v14.4s, v0.s[1]\n" + "add x12, x12, #4\n" + "fsub v19.4s, v19.4s, v2.4s\n" + "fmla v20.4s, v7.4s, v0.s[1]\n" + "fmls v19.4s, v13.4s, v0.s[1]\n" + "fmla v16.4s, v8.4s, v0.s[1]\n" + "str s12, [x14]\n" + "mov v1.16b, v17.16b\n" + "fsub v20.4s, v20.4s, v3.4s\n" + "mov v17.16b, v18.16b\n" + "str s19, [x14, %[output_col_stride1]]\n" + "fmls v20.4s, v21.4s, v0.s[1]\n" + "fsub v16.4s, v16.4s, v4.4s\n" + "fmla v1.4s, v11.4s, v0.s[1]\n" + "fmls v16.4s, v22.4s, v0.s[1]\n" + "fmla v17.4s, v15.4s, v0.s[1]\n" + "str s20, [x14, x11]\n" + "fsub v1.4s, v1.4s, v5.4s\n" + "str s16, [x14, x13]\n" + "fmls v1.4s, v23.4s, v0.s[1]\n" + "fsub v17.4s, v17.4s, v6.4s\n" + "fmls v17.4s, v24.4s, v0.s[1]\n" + "str s1, [x14, x23]\n" + "str s17, [x14, x15]\n" + "add x14, x14, #4\n" + "ldr s2, [x27, x20]\n" + "mov v4.16b, v2.16b\n" + "ldr s17, [x27, x18]\n" + "mov v12.16b, v2.16b\n" + "ldr s18, [x27]\n" + "fmla v4.4s, v18.4s, v0.s[2]\n" + "ldr s3, [x27, x21]\n" + "mov v6.16b, v2.16b\n" + "ldr s5, [x27, x19]\n" + "mov v1.16b, v2.16b\n" + "ldr s18, [x27, %[input_col_stride1]]\n" + "fmls v4.4s, v17.4s, v0.s[3]\n" + "add x27, x27, #4\n" + "fmls v12.4s, v18.4s, v0.s[2]\n" + "fmla v6.4s, v18.4s, v0.s[2]\n" + "fmls v1.4s, v18.4s, v0.s[1]\n" + "mov v2.16b, v2.16b\n" + "mov v3.16b, v3.16b\n" + "mov v4.16b, v4.16b\n" + "fmls v12.4s, v17.4s, v0.s[2]\n" + "fmls v6.4s, v17.4s, v0.s[2]\n" + "fsub v1.4s, v1.4s, v17.4s\n" + "fmla v2.4s, v18.4s, v0.s[1]\n" + "fmla v1.4s, v5.4s, v0.s[1]\n" + "fmla v3.4s, v18.4s, v0.s[2]\n" + "fadd v12.4s, v12.4s, v5.4s\n" + "fsub v6.4s, v6.4s, v5.4s\n" + "fsub v2.4s, v2.4s, v17.4s\n" + "fmla v4.4s, v10.4s, v0.s[2]\n" + "fmls v2.4s, v5.4s, v0.s[1]\n" + "fmls v3.4s, v5.4s, v0.s[3]\n" + "mov v16.16b, v12.16b\n" + "mov v5.16b, v6.16b\n" + "fmls v4.4s, v14.4s, v0.s[3]\n" + "mov v6.16b, v1.16b\n" + "fmla v16.4s, v9.4s, v0.s[2]\n" + "fmla v5.4s, v7.4s, v0.s[2]\n" + "fmla v6.4s, v8.4s, v0.s[2]\n" + "mov v9.16b, v2.16b\n" + "str s4, [x24]\n" + "mov v10.16b, v3.16b\n" + "fmls v16.4s, v13.4s, v0.s[3]\n" + "fmls v5.4s, v21.4s, v0.s[3]\n" + "fmls v6.4s, v22.4s, v0.s[3]\n" + "fmla v9.4s, v11.4s, v0.s[2]\n" + "fmla v10.4s, v15.4s, v0.s[2]\n" + "str s16, [x24, %[output_col_stride1]]\n" + "str s5, [x24, x11]\n" + "fmls v9.4s, v23.4s, v0.s[3]\n" + "str s6, [x24, x13]\n" + "fmls v10.4s, v24.4s, v0.s[3]\n" + "str s9, [x24, x23]\n" + "str s10, [x24, x15]\n" + "add x24, x24, #4\n" + "4:\n" + : [outptr0] "+r" (matrix_base), + [n_channels] "+r" (n_channels), + [inptr0] "+r" (input_base) + : [pcoeffs] "r" (pcoeffs), + [output_row_stride] "r" (6 * matrix_stride * sizeof(float)), + [output_col_stride1] "r" (matrix_stride * sizeof(float)), + [input_row_stride] "r" (input_row_stride * sizeof(float)), + [input_col_stride1] "r" (input_col_stride * sizeof(float)) + : "cc", "v0", "v1", "v10", "v11", "v12", "v13", "v14", "v15", "v16", "v17", + "v18", "v19", "v2", "v20", "v21", "v22", "v23", "v24", "v25", "v26", + "v27", "v28", "v29", "v3", "v30", "v31", "v4", "v5", "v6", "v7", "v8", + "v9", "x11", "x12", "x13", "x14", "x15", "x16", "x17", "x18", "x19", + "x20", "x21", "x22", "x23", "x24", "x25", "x26", "x27", "x28", "memory" + ); +} + +#else // __arm__ not __aarch64__ + +template <> +void InputTransform<6, 6, float, float, WinogradRoots::Integers>::transform_tile( + const int n_channels, + const float* const input_base, + const int input_row_stride, + const int input_col_stride, + float* outptr, + const int matrix_stride +) +{ + constexpr int inner_tile_rows = 6; + constexpr int inner_tile_cols = 6; + + // Get pointers into the input tile + const float *x_ptrs[inner_tile_rows][inner_tile_cols]; + for (int i = 0, xi = 0; i < inner_tile_rows; i++, xi++) + { + // Get a pointer into the row + const float* const row_ptr = input_base + xi*input_row_stride; + + for (int j = 0, xj = 0; j < inner_tile_cols; j++, xj++) + { + x_ptrs[i][j] = row_ptr + xj*input_col_stride; + } + } + + // Matrices used/computed in this kernel. + float x[inner_tile_rows][inner_tile_cols]; + float XTx[inner_tile_rows][inner_tile_cols]; + float U[inner_tile_rows][inner_tile_cols]; + for (int i = 0; i < inner_tile_rows; i++) + { + for (int j = 0; j < inner_tile_cols; j++) + { + x[i][j] = XTx[i][j] = 0.0f; + } + } + + // Perform the Winograd input transformation for each channel in the input + // tensor. + int channels_remaining = n_channels; + for (; channels_remaining >= 2; channels_remaining -= 2) + { + // Matrices used/computed in this kernel + float32x2_t x[inner_tile_rows][inner_tile_cols]; + float32x2_t XTx[inner_tile_rows][inner_tile_cols]; + float32x2_t U[inner_tile_rows][inner_tile_cols]; + for (int i = 0; i < inner_tile_rows; i++) + { + for (int j = 0; j < inner_tile_cols; j++) + { + x[i][j] = vdup_n_f32(0.0f); + XTx[i][j] = vdup_n_f32(0.0f); + } + } + + // Read a 6x6 tile in the Winograd domain + for (int i = 0; i < inner_tile_rows; i++) + { + for (int j = 0; j < inner_tile_cols; j++) + { + x[i][j] = vld1_f32(x_ptrs[i][j]); + x_ptrs[i][j] += 2; + } + } + + // Compute XT . x + for (int j = 0; j < inner_tile_cols; j++) + { + // XTx[0][j] = 4*x[0][j] + -5*x[2][j] + 1*x[4][j]; + XTx[0][j] = vmls_n_f32(vmla_n_f32(x[4][j], x[0][j], 4.0f), x[2][j], 5.0f); + + // XTx[1][j] = -4*x[1][j] + -4*x[2][j] + 1*x[3][j] + 1*x[4][j]; + XTx[1][j] = vmls_n_f32(vadd_f32(x[3][j], x[4][j]), vadd_f32(x[1][j], x[2][j]), 4.0f); + + // XTx[2][j] = 4*x[1][j] + -4*x[2][j] + -1*x[3][j] + 1*x[4][j]; + XTx[2][j] = vmla_n_f32(vsub_f32(x[4][j], x[3][j]), vsub_f32(x[1][j], x[2][j]), 4.0f); + + // XTx[3][j] = -2*x[1][j] + -1*x[2][j] + 2*x[3][j] + 1*x[4][j]; + XTx[3][j] = vmla_n_f32(vsub_f32(x[4][j], x[2][j]), vsub_f32(x[3][j], x[1][j]), 2.0f); + + // XTx[4][j] = 2*x[1][j] + -1*x[2][j] + -2*x[3][j] + 1*x[4][j]; + XTx[4][j] = vmla_n_f32(vsub_f32(x[4][j], x[2][j]), vsub_f32(x[1][j], x[3][j]), 2.0f); + + // XTx[5][j] = 4*x[1][j] + -5*x[3][j] + 1*x[5][j]; + XTx[5][j] = vmls_n_f32(vmla_n_f32(x[5][j], x[1][j], 4.0f), x[3][j], 5.0f); + } + + // Compute U = XT . x . X + for (int i = 0; i < inner_tile_rows; i++) + { + // U[i][0] = 4*XTx[i][0] + -5*XTx[i][2] + 1*XTx[i][4]; + U[i][0] = vmls_n_f32(vmla_n_f32(XTx[i][4], XTx[i][0], 4.0f), XTx[i][2], 5.0f); + + // U[i][1] = -4*XTx[i][1] + -4*XTx[i][2] + 1*XTx[i][3] + 1*XTx[i][4]; + U[i][1] = vmls_n_f32(vadd_f32(XTx[i][3], XTx[i][4]), vadd_f32(XTx[i][1], XTx[i][2]), 4.0f); + + // U[i][2] = 4*XTx[i][1] + -4*XTx[i][2] + -1*XTx[i][3] + 1*XTx[i][4]; + U[i][2] = vmla_n_f32(vsub_f32(XTx[i][4], XTx[i][3]), vsub_f32(XTx[i][1], XTx[i][2]), 4.0f); + + // U[i][3] = -2*XTx[i][1] + -1*XTx[i][2] + 2*XTx[i][3] + 1*XTx[i][4]; + U[i][3] = vmla_n_f32(vsub_f32(XTx[i][4], XTx[i][2]), vsub_f32(XTx[i][3], XTx[i][1]), 2.0f); + + // U[i][4] = 2*XTx[i][1] + -1*XTx[i][2] + -2*XTx[i][3] + 1*XTx[i][4]; + U[i][4] = vmla_n_f32(vsub_f32(XTx[i][4], XTx[i][2]), vsub_f32(XTx[i][1], XTx[i][3]), 2.0f); + + // U[i][5] = 4*XTx[i][1] + -5*XTx[i][3] + 1*XTx[i][5]; + U[i][5] = vmls_n_f32(vmla_n_f32(XTx[i][5], XTx[i][1], 4.0f), XTx[i][3], 5.0f); + } + + // Store the transformed matrix + for (int i = 0, m = 0; i < inner_tile_rows; i++) + { + for (int j = 0; j < inner_tile_cols; j++, m++) + { + vst1_f32(outptr + m*matrix_stride, U[i][j]); + } + } + outptr += 2; + } + for (; channels_remaining; channels_remaining--) + { + // Load x + for (int i = 0; i < inner_tile_rows; i++) + { + for (int j = 0; j < inner_tile_cols; j++) + { + x[i][j] = *(x_ptrs[i][j]++); + } + } + + // Compute XT . x + for (int j = 0; j < inner_tile_cols; j++) + { + XTx[0][j] = 4*x[0][j] + -5*x[2][j] + 1*x[4][j]; + XTx[1][j] = -4*x[1][j] + -4*x[2][j] + 1*x[3][j] + 1*x[4][j]; + XTx[2][j] = 4*x[1][j] + -4*x[2][j] + -1*x[3][j] + 1*x[4][j]; + XTx[3][j] = -2*x[1][j] + -1*x[2][j] + 2*x[3][j] + 1*x[4][j]; + XTx[4][j] = 2*x[1][j] + -1*x[2][j] + -2*x[3][j] + 1*x[4][j]; + XTx[5][j] = 4*x[1][j] + -5*x[3][j] + 1*x[5][j]; + } + + // Compute U = XT . x . X + for (int i = 0; i < inner_tile_rows; i++) + { + U[i][0] = 4*XTx[i][0] + -5*XTx[i][2] + 1*XTx[i][4]; + U[i][1] = -4*XTx[i][1] + -4*XTx[i][2] + 1*XTx[i][3] + 1*XTx[i][4]; + U[i][2] = 4*XTx[i][1] + -4*XTx[i][2] + -1*XTx[i][3] + 1*XTx[i][4]; + U[i][3] = -2*XTx[i][1] + -1*XTx[i][2] + 2*XTx[i][3] + 1*XTx[i][4]; + U[i][4] = 2*XTx[i][1] + -1*XTx[i][2] + -2*XTx[i][3] + 1*XTx[i][4]; + U[i][5] = 4*XTx[i][1] + -5*XTx[i][3] + 1*XTx[i][5]; + } + + // Store the transformed matrix + for (int i = 0, m = 0; i < inner_tile_rows; i++) + { + for (int j = 0; j < inner_tile_cols; j++, m++) + { + *(outptr + m*matrix_stride) = U[i][j]; + } + } + outptr++; + } +} + +#endif + +template class InputTransform<6, 6, float, float, WinogradRoots::Integers>; + +} // namespace winograd diff --git a/src/core/NEON/kernels/convolution/winograd/winograd_transforms/kernel.hpp b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/kernel.hpp new file mode 100644 index 0000000000..e45f1863e3 --- /dev/null +++ b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/kernel.hpp @@ -0,0 +1,78 @@ +/* + * Copyright (c) 2019 ARM Limited. + * + * SPDX-License-Identifier: MIT + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#pragma once +#include "winograd.hpp" +using namespace winograd; + +#define MEMBERFN(RTYPE) template <\ + int KernelRows, int KernelCols, int InnerTileRows, int InnerTileCols, typename TIn, typename TOut, WinogradRoots Roots\ +> RTYPE WeightTransform + +MEMBERFN()::WeightTransform( + const int n_output_channels, + const int n_input_channels +) : _n_output_channels(n_output_channels), _n_input_channels(n_input_channels), + _matrices(nullptr), _matrix_stride(0), _matrix_row_stride(0), _weights(nullptr) +{ + +} + +MEMBERFN(void)::set_weight_tensor(const void * const weights) +{ + _weights = static_cast(weights); +} + +MEMBERFN(void)::set_output_matrices(void * const mptr, const int ldmatrix, const int ldrow) +{ + _matrices = static_cast(mptr); + _matrix_stride = ldmatrix; + _matrix_row_stride = ldrow; +} + +MEMBERFN(size_t)::get_working_space_size(unsigned int) const +{ + return 0; +} + +MEMBERFN(void)::set_working_space(void *) +{ +} + +MEMBERFN(unsigned int)::get_window(void) const +{ + // TODO When the weights transform supports multithreading, return the number + // of output channels. For now we return 1 to indicate that the weights must + // be transformed as a single block. + // return n_output_channels; + return 1; +} + +MEMBERFN(void)::run(const unsigned int, const unsigned int, unsigned int) +{ + execute( + _n_output_channels, _n_input_channels, _weights, + _matrices, _matrix_stride, _matrix_row_stride + ); +} diff --git a/src/core/NEON/kernels/convolution/winograd/winograd_transforms/output.hpp b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/output.hpp new file mode 100644 index 0000000000..d97af21a43 --- /dev/null +++ b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/output.hpp @@ -0,0 +1,249 @@ +/* + * Copyright (c) 2019 ARM Limited. + * + * SPDX-License-Identifier: MIT + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#pragma once + +#include +#include "winograd.hpp" +#include "padding.hpp" +#include "utils.hpp" + +#define MEMBERFN(RTYPE) template<\ + int KernelRows, int KernelCols, int InnerTileRows, int InnerTileCols,\ + typename TIn, typename TOut, WinogradRoots Roots\ +> RTYPE OutputTransform + +#define Nx1MEMBERFN(RTYPE) template<\ + int KernelRows, int InnerTileRows, typename TIn, typename TOut, WinogradRoots Roots\ +> RTYPE OutputTransform + +namespace winograd +{ + +MEMBERFN()::OutputTransform( + const int n_batches, + const int n_rows, + const int n_cols, + const int n_channels +) : _n_batches(n_batches), _n_rows(n_rows), _n_cols(n_cols), _n_channels(n_channels), + _matrix_base(nullptr), + _biases(nullptr), + _matrix_stride(0), _matrix_row_stride(0), _matrix_batch_stride(0), + _outptr(nullptr), + _tiles_M(iceildiv(n_rows, output_tile_rows)), + _tiles_N(iceildiv(n_cols, output_tile_cols)), + _out_col_stride(0), _out_row_stride(0), _out_batch_stride(0), + _working_space_col_stride(n_channels), + _working_space_row_stride(output_tile_cols * _working_space_col_stride), + _working_space(nullptr) +{ +} + +MEMBERFN(void)::set_input_matrices(const void * const mptr, const int ldmatrix, const int ldrow) +{ + _matrix_base = static_cast(mptr); + _matrix_stride = ldmatrix; + _matrix_row_stride = ldrow; + _matrix_batch_stride = _tiles_M * _tiles_N * ldrow; +} + +MEMBERFN(void)::set_bias(const void * const bias) +{ + _biases = static_cast(bias); +} + +MEMBERFN(void)::set_output_tensor(void * const outptr) +{ + set_output_tensor(outptr, _n_channels); +} + +MEMBERFN(void)::set_output_tensor(void * const outptr, const int ldcol) +{ + set_output_tensor(outptr, _n_cols * ldcol, ldcol); +} + +MEMBERFN(void)::set_output_tensor(void * const outptr, const int ldrow, const int ldcol) +{ + set_output_tensor(outptr, _n_rows * ldrow, ldrow, ldcol); +} + +MEMBERFN(void)::set_output_tensor(void * const outptr, const int ldbatch, const int ldrow, const int ldcol) +{ + _outptr = static_cast(outptr); + _out_batch_stride = ldbatch; + _out_row_stride = ldrow; + _out_col_stride = ldcol; +} + +Nx1MEMBERFN()::OutputTransform( + const int n_batches, + const int n_rows, + const int n_cols, + const int n_channels +) : OutputTransform<1, KernelRows, 1, InnerTileRows, TIn, TOut, Roots>::OutputTransform( + n_batches, n_cols, n_rows, n_channels /* Transpose rows and columns */ + ) +{ +} + +Nx1MEMBERFN(void)::set_output_tensor(void * const outptr) +{ + set_output_tensor(outptr, this->_n_channels); +} + +Nx1MEMBERFN(void)::set_output_tensor(void * const outptr, const int ldcol) +{ + set_output_tensor(outptr, this->_n_cols * ldcol, ldcol); +} + +Nx1MEMBERFN(void)::set_output_tensor(void * const outptr, const int ldrow, const int ldcol) +{ + set_output_tensor(outptr, this->_n_rows * ldrow, ldrow, ldcol); +} + +Nx1MEMBERFN(void)::set_output_tensor(void * const outptr, const int ldbatch, const int ldrow, const int ldcol) +{ + // Transpose rows and columns + Base::set_output_tensor(outptr, ldbatch, ldcol, ldrow); +} + +MEMBERFN(size_t)::get_working_space_size(const unsigned int nthreads) const +{ + return sizeof(TOut) * output_tile_rows * _working_space_row_stride * nthreads; +} + +MEMBERFN(void)::set_working_space(void * const buffer) +{ + _working_space = static_cast(buffer); +} + +MEMBERFN(unsigned int)::get_window(void) const +{ + return iceildiv(_n_channels, WINDOW_BLOCK); +} + +MEMBERFN(void)::run( + const unsigned int start, + const unsigned int stop, + const unsigned int threadid +) +{ + // Determine the channels on which to work + if (start >= get_window()) + { + return; // No work to do beyond the end of the window + } + const unsigned int start_channel = start * WINDOW_BLOCK; + const unsigned int stop_channel = std::min(_n_channels, stop * WINDOW_BLOCK); + const unsigned int n_channels = stop_channel - start_channel; + + const auto matrix_tile_col_stride = _matrix_row_stride; + const auto matrix_tile_row_stride = _tiles_N * matrix_tile_col_stride; + + const TOut* const bptr = (_biases == nullptr) ? nullptr : _biases + start_channel; + + // Loop over batches + for (int batch = 0; batch < _n_batches; batch++) + { + const TIn* const matrix_batch = _matrix_base + start_channel + batch * _matrix_batch_stride; + TOut* const outptr_batch = _outptr + start_channel + batch * _out_batch_stride; + + for (int tile_i = 0; tile_i < _tiles_M; tile_i++) + { + // Compute properties of the row of output tiles + const int row_pad_bottom = std::max(0, (tile_i + 1)*output_tile_rows - _n_rows); + const TIn* const matrix_tile_row = matrix_batch + tile_i * matrix_tile_row_stride; + TOut* const outptr_row = outptr_batch + tile_i * output_tile_rows * _out_row_stride; + + for (int tile_j = 0; tile_j < _tiles_N; tile_j++) + { + // Compute property of this specific tile + const int tile_pad_right = std::max(0, (tile_j + 1)*output_tile_cols - _n_cols); + const TIn* const matrix_tile = matrix_tile_row + tile_j * matrix_tile_col_stride; + TOut* const outptr_tile = outptr_row + tile_j * output_tile_cols * _out_col_stride; + + // Perform the transformation + if (row_pad_bottom || tile_pad_right) + { + transform_cropped_tile( + threadid, n_channels, outptr_tile, matrix_tile, bptr, + row_pad_bottom, tile_pad_right + ); + } + else + { + transform_uncropped_tile( + threadid, n_channels, outptr_tile, matrix_tile, bptr + ); + } + } + } + } +} + +MEMBERFN(void)::transform_uncropped_tile( + const unsigned int /* threadid unused */, + const int n_channels, + TOut * const outptr, + const TIn * const inptr, + const TOut * const biases +) +{ + transform_tile( + n_channels, inptr, _matrix_stride, biases, + outptr, _out_row_stride, _out_col_stride + ); +} + +MEMBERFN(void)::transform_cropped_tile( + const unsigned int threadid, + const int n_channels, + TOut * const outptr, + const TIn * const inptr, + const TOut * const biases, + const int pad_bottom, + const int pad_right +) +{ + // Transform into working space and then copy the relevant section out. + TOut *wsptr = static_cast(get_working_space(threadid)); + transform_tile( + n_channels, inptr, _matrix_stride, biases, + wsptr, _working_space_row_stride, _working_space_col_stride + ); + + padding::crop_and_copy_tile( + output_tile_rows, output_tile_cols, n_channels, + wsptr, _working_space_row_stride, _working_space_col_stride, + outptr, _out_row_stride, _out_col_stride, + 0u, 0u, pad_bottom, pad_right + ); +} + +MEMBERFN(void *)::get_working_space(const unsigned int threadid) const +{ + return _working_space + output_tile_rows * _working_space_row_stride * threadid; +} + +} // namespace winograd diff --git a/src/core/NEON/kernels/convolution/winograd/winograd_transforms/output_2_7_fp32_fp32_integers.cpp b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/output_2_7_fp32_fp32_integers.cpp new file mode 100644 index 0000000000..c32d7f2f58 --- /dev/null +++ b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/output_2_7_fp32_fp32_integers.cpp @@ -0,0 +1,137 @@ +/* + * Copyright (c) 2019 ARM Limited. + * + * SPDX-License-Identifier: MIT + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "arm.hpp" +#include "output.hpp" + +namespace winograd +{ + +template <> +void OutputTransform<1, 7, 1, 8, float, float, WinogradRoots::Integers>::transform_tile( + const int n_channels, + const float* inptr, + const int matrix_stride, + const float* bptr, + float* const output, + const int, // No need to stride across rows + const int output_col_stride +) +{ + // Construct a map to the output cells + float *outptrs[output_tile_cols]; + for (int j = 0; j < output_tile_cols; j++) + { + outptrs[j] = output + j*output_col_stride; + } + + // For each channel of the output + int channels_remaining = n_channels; +#ifdef __arm_any__ + for (; channels_remaining >= 4; channels_remaining -= 4) + { + // Matrices used and computed during this transform + float32x4_t F[inner_tile_cols], f[output_tile_cols], b = vdupq_n_f32(0.0f); + + // Read a 1x8 tile in the Winograd domain + for (int j = 0; j < inner_tile_cols; j++) + { + F[j] = vld1q_f32(inptr + j*matrix_stride); + } + inptr += 4; + + f[0] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(F[6], 1), F[5], 1), F[4], 1), F[3], 1), F[2], 1), F[1], 1), F[0], 1); + f[1] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(F[7], 1), F[2], 1), F[6], 3), F[4], 2), F[3], -2), F[5], -3), F[1], -1); + + // Write out the output tile + if (bptr != 0) + { + b = vld1q_f32(bptr); + bptr += 4; + } + for (int j = 0; j < output_tile_cols; j++) + { + vst1q_f32(outptrs[j], f[j] + b); + outptrs[j] += 4; + } + } + for (; channels_remaining >= 2; channels_remaining -= 2) + { + // Matrices used and computed during this transform + float32x2_t F[inner_tile_cols], f[output_tile_cols], b = vdup_n_f32(0.0f); + + // Read a 1x8 tile in the Winograd domain + for (int j = 0; j < inner_tile_cols; j++) + { + F[j] = vld1_f32(inptr + j*matrix_stride); + } + inptr += 2; + + f[0] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(F[6], 1), F[5], 1), F[4], 1), F[3], 1), F[2], 1), F[1], 1), F[0], 1); + f[1] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(F[7], 1), F[2], 1), F[6], 3), F[4], 2), F[3], -2), F[5], -3), F[1], -1); + + // Write out the output tile + if (bptr != 0) + { + b = vld1_f32(bptr); + bptr += 2; + } + for (int j = 0; j < output_tile_cols; j++) + { + vst1_f32(outptrs[j], f[j] + b); + outptrs[j] += 2; + } + } +#endif // __arm_any__ + for (; channels_remaining; channels_remaining--) + { + // Matrices used and computed during this transform + float F[inner_tile_cols], f[output_tile_cols], b = 0.0f; + + // Read a 1x8 tile in the Winograd domain + for (int j = 0; j < inner_tile_cols; j++) + { + F[j] = *(inptr + j*matrix_stride); + } + inptr++; + + f[0] = F[0]*1 + F[1]*1 + F[2]*1 + F[3]*1 + F[4]*1 + F[5]*1 + F[6]*1; + f[1] = F[1]*-1 + F[5]*-3 + F[3]*-2 + F[4]*2 + F[6]*3 + F[2]*1 + F[7]*1; + + // Write out the output tile + if (bptr != 0) + { + b = *(bptr++); + } + for (int j = 0; j < output_tile_cols; j++) + { + *(outptrs[j]++) = f[j] + b; + } + } +} + +template class OutputTransform<1, 7, 1, 8, float, float, WinogradRoots::Integers>; +template class OutputTransform<7, 1, 8, 1, float, float, WinogradRoots::Integers>; + +} // namespace winograd diff --git a/src/core/NEON/kernels/convolution/winograd/winograd_transforms/output_2x2_3x3_fp32_fp32_integers.cpp b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/output_2x2_3x3_fp32_fp32_integers.cpp new file mode 100644 index 0000000000..d6ebf44f41 --- /dev/null +++ b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/output_2x2_3x3_fp32_fp32_integers.cpp @@ -0,0 +1,222 @@ +/* + * Copyright (c) 2019 ARM Limited. + * + * SPDX-License-Identifier: MIT + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "arm.hpp" +#include "output.hpp" + +namespace winograd +{ + +template <> +void OutputTransform<3, 3, 4, 4, float, float, WinogradRoots::Integers>::transform_tile( + const int n_channels, + const float* inptr, + const int matrix_stride, + const float* bptr, + float* const output, + const int output_row_stride, + const int output_col_stride +) +{ + // Construct a map to the output cells + float *outptrs[output_tile_rows][output_tile_cols]; + for (int i = 0; i < output_tile_rows; i++) + { + for (int j = 0; j < output_tile_cols; j++) + { + outptrs[i][j] = output + i*output_row_stride + j*output_col_stride; + } + } + + // For each channel of the output + int channels_remaining = n_channels; +#ifdef __aarch64__ + for (; channels_remaining >= 4; channels_remaining -= 4) + { + // Matrices used and computed during this transform + float32x4_t F[4][4], FZ[4][2], f[2][2], b; + + // Read a 4x4 tile in the Winograd domain + for (int i = 0, m = 0; i < 4; i++) + { + for (int j = 0; j < 4; j++, m++) + { + F[i][j] = vld1q_f32(inptr + m*matrix_stride); + } + } + inptr += 4; + + // Compute the matrix F Z + for (int i = 0; i < 4; i++) + { + // FZ[i][0] = F[i][0] + F[i][1] + F[i][2]; + FZ[i][0] = vaddq_f32(vaddq_f32(F[i][0], F[i][1]), F[i][2]); + + // FZ[i][1] = F[i][1] - F[i][2] - F[i][3]; + FZ[i][1] = vsubq_f32(vsubq_f32(F[i][1], F[i][2]), F[i][3]); + } + + // Compute the output tile f = ZT F Z + for (int j = 0; j < 2; j++) + { + // f[0][j] = FZ[0][j] + FZ[1][j] + FZ[2][j]; + f[0][j] = vaddq_f32(vaddq_f32(FZ[0][j], FZ[1][j]), FZ[2][j]); + + // f[1][j] = FZ[1][j] - FZ[2][j] - FZ[3][j]; + f[1][j] = vsubq_f32(vsubq_f32(FZ[1][j], FZ[2][j]), FZ[3][j]); + } + + // Load the bias vector + if (bptr != nullptr) + { + b = vld1q_f32(bptr); + bptr += 4; + } + else + { + b = vdupq_n_f32(0.0f); + } + + // Write out the output tile + for (int i = 0; i < output_tile_rows; i++) + { + for (int j = 0; j < output_tile_cols; j++) + { + vst1q_f32(outptrs[i][j], vaddq_f32(f[i][j], b)); + outptrs[i][j] += 4; + } + } + } +#endif // __aarch64__ +#ifdef __arm_any__ + for (; channels_remaining >= 2; channels_remaining -= 2) + { + // Matrices used and computed during this transform + float32x2_t F[4][4], FZ[4][2], f[2][2], b; + + // Read a 4x4 tile in the Winograd domain + for (int i = 0, m = 0; i < 4; i++) + { + for (int j = 0; j < 4; j++, m++) + { + F[i][j] = vld1_f32(inptr + m*matrix_stride); + } + } + inptr += 2; + + // Compute the matrix F Z + for (int i = 0; i < 4; i++) + { + // FZ[i][0] = F[i][0] + F[i][1] + F[i][2]; + FZ[i][0] = vadd_f32(vadd_f32(F[i][0], F[i][1]), F[i][2]); + + // FZ[i][1] = F[i][1] - F[i][2] - F[i][3]; + FZ[i][1] = vsub_f32(vsub_f32(F[i][1], F[i][2]), F[i][3]); + } + + // Compute the output tile f = ZT F Z + for (int j = 0; j < 2; j++) + { + // f[0][j] = FZ[0][j] + FZ[1][j] + FZ[2][j]; + f[0][j] = vadd_f32(vadd_f32(FZ[0][j], FZ[1][j]), FZ[2][j]); + + // f[1][j] = FZ[1][j] - FZ[2][j] - FZ[3][j]; + f[1][j] = vsub_f32(vsub_f32(FZ[1][j], FZ[2][j]), FZ[3][j]); + } + + // Load the bias vector + if (bptr != nullptr) + { + b = vld1_f32(bptr); + bptr += 2; + } + else + { + b = vdup_n_f32(0.0f); + } + + // Write out the output tile + for (int i = 0; i < output_tile_rows; i++) + { + for (int j = 0; j < output_tile_cols; j++) + { + vst1_f32(outptrs[i][j], vadd_f32(f[i][j], b)); + outptrs[i][j] += 2; + } + } + } +#endif // __arm_any__ + for (; channels_remaining; channels_remaining--) + { + // Matrices used and computed during this transform + float F[4][4], FZ[4][2], f[2][2], b; + + // Read a 4x4 tile in the Winograd domain + for (int i = 0, m = 0; i < 4; i++) + { + for (int j = 0; j < 4; j++, m++) + { + F[i][j] = *(inptr + m*matrix_stride); + } + } + inptr++; + + // Compute the matrix F Z + for (int i = 0; i < 4; i++) + { + FZ[i][0] = F[i][0] + F[i][1] + F[i][2]; + FZ[i][1] = F[i][1] - F[i][2] - F[i][3]; + } + + // Compute the output tile f = ZT F Z + for (int j = 0; j < 2; j++) + { + f[0][j] = FZ[0][j] + FZ[1][j] + FZ[2][j]; + f[1][j] = FZ[1][j] - FZ[2][j] - FZ[3][j]; + } + + // Load the bias + if (bptr != nullptr) + { + b = *(bptr++); + } + else + { + b = 0.0f; + } + + // Write out the output tile + for (int i = 0; i < output_tile_rows; i++) + { + for (int j = 0; j < output_tile_cols; j++) + { + *(outptrs[i][j]++) = f[i][j] + b; + } + } + } +} + +template class OutputTransform<3, 3, 4, 4, float, float, WinogradRoots::Integers>; + +} // namespace diff --git a/src/core/NEON/kernels/convolution/winograd/winograd_transforms/output_2x2_5x5_fp32_fp32_integers.cpp b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/output_2x2_5x5_fp32_fp32_integers.cpp new file mode 100644 index 0000000000..d93d9e234a --- /dev/null +++ b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/output_2x2_5x5_fp32_fp32_integers.cpp @@ -0,0 +1,216 @@ +/* + * Copyright (c) 2019 ARM Limited. + * + * SPDX-License-Identifier: MIT + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "output.hpp" +#include "arm.hpp" + +namespace winograd +{ + +template <> +void OutputTransform<5, 5, 6, 6, float, float, WinogradRoots::Integers>::transform_tile( + const int n_channels, + const float* inptr, + const int matrix_stride, + const float* bptr, + float* const output, + const int output_row_stride, + const int output_col_stride +) +{ + // Construct a map to the output cells + float *outptrs[output_tile_rows][output_tile_cols]; + for (int i = 0; i < output_tile_rows; i++) + { + for (int j = 0; j < output_tile_cols; j++) + { + outptrs[i][j] = output + i*output_row_stride + j*output_col_stride; + } + } + + // For each channel of the output + int channels_remaining = n_channels; +#ifdef __aarch64__ + for (; channels_remaining >= 4; channels_remaining -= 4) + { + // Matrices used and computed during this transform + float32x4_t F[6][6], FZ[6][2], f[2][2], b; + + // Read a 6x6 tile in the Winograd domain + for (int i = 0, m = 0; i < 6; i++) + { + for (int j = 0; j < 6; j++, m++) + { + F[i][j] = vld1q_f32(inptr + m*matrix_stride); + } + } + inptr += 4; + + // Compute the matrix F Z + for (int i = 0; i < 6; i++) + { + // FZ[i][0] = 1*F[i][0] + 1*F[i][1] + 1*F[i][2] + 1*F[i][3] + 1*F[i][4]; + FZ[i][0] = vaddq_f32(vaddq_f32(vaddq_f32(F[i][0], F[i][1]), vaddq_f32(F[i][2], F[i][3])), F[i][4]); + + // FZ[i][1] = 1*F[i][1] + -1*F[i][2] + 2*F[i][3] + -2*F[i][4] + 1*F[i][5]; + FZ[i][1] = vaddq_f32(vmlaq_n_f32(vsubq_f32(F[i][1], F[i][2]), vsubq_f32(F[i][3], F[i][4]), 2.0f), F[i][5]); + } + + // Compute the output tile f = ZT F Z + for (int j = 0; j < 2; j++) + { + // f[0][j] = 1*FZ[0][j] + 1*FZ[1][j] + 1*FZ[2][j] + 1*FZ[3][j] + 1*FZ[4][j]; + f[0][j] = vaddq_f32(vaddq_f32(vaddq_f32(FZ[0][j], FZ[1][j]), vaddq_f32(FZ[2][j], FZ[3][j])), FZ[4][j]); + + // f[1][j] = 1*FZ[1][j] + -1*FZ[2][j] + 2*FZ[3][j] + -2*FZ[4][j] + 1*FZ[5][j]; + f[1][j] = vaddq_f32(vmlaq_n_f32(vsubq_f32(FZ[1][j], FZ[2][j]), vsubq_f32(FZ[3][j], FZ[4][j]), 2.0f), FZ[5][j]); + } + + // Write out the output tile + if (bptr != nullptr) + { + b = vld1q_f32(bptr); + bptr += 4; + } + else + { + b = vdupq_n_f32(0.0f); + } + for (int i = 0; i < output_tile_rows; i++) + { + for (int j = 0; j < output_tile_cols; j++) + { + vst1q_f32(outptrs[i][j], vaddq_f32(f[i][j], b)); + outptrs[i][j] += 4; + } + } + } +#endif // __aarch64__ +#ifdef __arm_any__ + for (; channels_remaining >= 2; channels_remaining -= 2) + { + // Matrices used and computed during this transform + float32x2_t F[6][6], FZ[6][2], f[2][2], b; + + // Read a 6x6 tile in the Winograd domain + for (int i = 0, m = 0; i < 6; i++) + { + for (int j = 0; j < 6; j++, m++) + { + F[i][j] = vld1_f32(inptr + m*matrix_stride); + } + } + inptr += 2; + + // Compute the matrix F Z + for (int i = 0; i < 6; i++) + { + // FZ[i][0] = 1*F[i][0] + 1*F[i][1] + 1*F[i][2] + 1*F[i][3] + 1*F[i][4]; + FZ[i][0] = vadd_f32(vadd_f32(vadd_f32(F[i][0], F[i][1]), vadd_f32(F[i][2], F[i][3])), F[i][4]); + + // FZ[i][1] = 1*F[i][1] + -1*F[i][2] + 2*F[i][3] + -2*F[i][4] + 1*F[i][5]; + FZ[i][1] = vadd_f32(vmla_n_f32(vsub_f32(F[i][1], F[i][2]), vsub_f32(F[i][3], F[i][4]), 2.0f), F[i][5]); + } + + // Compute the output tile f = ZT F Z + for (int j = 0; j < 2; j++) + { + // f[0][j] = 1*FZ[0][j] + 1*FZ[1][j] + 1*FZ[2][j] + 1*FZ[3][j] + 1*FZ[4][j]; + f[0][j] = vadd_f32(vadd_f32(vadd_f32(FZ[0][j], FZ[1][j]), vadd_f32(FZ[2][j], FZ[3][j])), FZ[4][j]); + + // f[1][j] = 1*FZ[1][j] + -1*FZ[2][j] + 2*FZ[3][j] + -2*FZ[4][j] + 1*FZ[5][j]; + f[1][j] = vadd_f32(vmla_n_f32(vsub_f32(FZ[1][j], FZ[2][j]), vsub_f32(FZ[3][j], FZ[4][j]), 2.0f), FZ[5][j]); + } + + // Write out the output tile + if (bptr != nullptr) + { + b = vld1_f32(bptr); + bptr += 2; + } + else + { + b = vdup_n_f32(0.0f); + } + for (int i = 0; i < output_tile_rows; i++) + { + for (int j = 0; j < output_tile_cols; j++) + { + vst1_f32(outptrs[i][j], vadd_f32(f[i][j], b)); + outptrs[i][j] += 2; + } + } + } +#endif // __arm_any__ + for (; channels_remaining; channels_remaining--) + { + // Matrices used and computed during this transform + float F[6][6], FZ[6][2], f[2][2], b; + + // Read a 6x6 tile in the Winograd domain + for (int i = 0, m = 0; i < 6; i++) + { + for (int j = 0; j < 6; j++, m++) + { + F[i][j] = *(inptr + m*matrix_stride); + } + } + inptr++; + + // Compute the matrix F Z + for (int i = 0; i < 6; i++) + { + FZ[i][0] = 1*F[i][0] + 1*F[i][1] + 1*F[i][2] + 1*F[i][3] + 1*F[i][4]; + FZ[i][1] = 1*F[i][1] + -1*F[i][2] + 2*F[i][3] + -2*F[i][4] + 1*F[i][5]; + } + + // Compute the output tile f = ZT F Z + for (int j = 0; j < 2; j++) + { + f[0][j] = 1*FZ[0][j] + 1*FZ[1][j] + 1*FZ[2][j] + 1*FZ[3][j] + 1*FZ[4][j]; + f[1][j] = 1*FZ[1][j] + -1*FZ[2][j] + 2*FZ[3][j] + -2*FZ[4][j] + 1*FZ[5][j]; + } + + // Write out the output tile + if (bptr != nullptr) + { + b = *(bptr++); + } + else + { + b = 0.0f; + } + for (int i = 0; i < output_tile_rows; i++) + { + for (int j = 0; j < output_tile_cols; j++) + { + *(outptrs[i][j]++) = f[i][j] + b; + } + } + } +} + +template class OutputTransform<5, 5, 6, 6, float, float, WinogradRoots::Integers>; + +} // namespace diff --git a/src/core/NEON/kernels/convolution/winograd/winograd_transforms/output_4_5_fp32_fp32_integers.cpp b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/output_4_5_fp32_fp32_integers.cpp new file mode 100644 index 0000000000..7187ef2d20 --- /dev/null +++ b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/output_4_5_fp32_fp32_integers.cpp @@ -0,0 +1,143 @@ +/* + * Copyright (c) 2019 ARM Limited. + * + * SPDX-License-Identifier: MIT + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "output.hpp" +#include "arm.hpp" + +namespace winograd +{ + +template <> +void OutputTransform<1, 5, 1, 8, float, float, WinogradRoots::Integers>::transform_tile( + const int n_channels, + const float* inptr, + const int matrix_stride, + const float* bptr, + float* const output, + const int, // No need to stride across rows + const int output_col_stride +) +{ + // Construct a map to the output cells + float *outptrs[output_tile_cols]; + for (int j = 0; j < output_tile_cols; j++) + { + outptrs[j] = output + j*output_col_stride; + } + + // For each channel of the output + int channels_remaining = n_channels; +#ifdef __arm_any__ + for (; channels_remaining >= 4; channels_remaining -= 4) + { + // Matrices used and computed during this transform + float32x4_t F[inner_tile_cols], f[output_tile_cols], b = vdupq_n_f32(0.0f); + + // Read a 1x8 tile in the Winograd domain + for (int j = 0; j < inner_tile_cols; j++) + { + F[j] = vld1q_f32(inptr + j*matrix_stride); + } + inptr += 4; + + f[0] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(F[6], 1), F[5], 1), F[4], 1), F[3], 1), F[2], 1), F[1], 1), F[0], 1); + f[1] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(F[2], 1), F[6], 3), F[4], 2), F[3], -2), F[5], -3), F[1], -1); + f[2] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(F[2], 1), F[1], 1), F[6], 9), F[5], 9), F[4], 4), F[3], 4); + f[3] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(F[7], 1), F[2], 1), F[6], 27), F[4], 8), F[3], -8), F[5], -27), F[1], -1); + + // Write out the output tile + if (bptr != 0) + { + b = vld1q_f32(bptr); + bptr += 4; + } + for (int j = 0; j < output_tile_cols; j++) + { + vst1q_f32(outptrs[j], f[j] + b); + outptrs[j] += 4; + } + } + for (; channels_remaining >= 2; channels_remaining -= 2) + { + // Matrices used and computed during this transform + float32x2_t F[inner_tile_cols], f[output_tile_cols], b = vdup_n_f32(0.0f); + + // Read a 1x8 tile in the Winograd domain + for (int j = 0; j < inner_tile_cols; j++) + { + F[j] = vld1_f32(inptr + j*matrix_stride); + } + inptr += 2; + + f[0] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(F[6], 1), F[5], 1), F[4], 1), F[3], 1), F[2], 1), F[1], 1), F[0], 1); + f[1] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(F[2], 1), F[6], 3), F[4], 2), F[3], -2), F[5], -3), F[1], -1); + f[2] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(F[2], 1), F[1], 1), F[6], 9), F[5], 9), F[4], 4), F[3], 4); + f[3] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(F[7], 1), F[2], 1), F[6], 27), F[4], 8), F[3], -8), F[5], -27), F[1], -1); + + // Write out the output tile + if (bptr != 0) + { + b = vld1_f32(bptr); + bptr += 2; + } + for (int j = 0; j < output_tile_cols; j++) + { + vst1_f32(outptrs[j], f[j] + b); + outptrs[j] += 2; + } + } +#endif // __arm_any__ + for (; channels_remaining; channels_remaining--) + { + // Matrices used and computed during this transform + float F[inner_tile_cols], f[output_tile_cols], b = 0.0f; + + // Read a 1x8 tile in the Winograd domain + for (int j = 0; j < inner_tile_cols; j++) + { + F[j] = *(inptr + j*matrix_stride); + } + inptr++; + + f[0] = F[0]*1 + F[1]*1 + F[2]*1 + F[3]*1 + F[4]*1 + F[5]*1 + F[6]*1; + f[1] = F[1]*-1 + F[5]*-3 + F[3]*-2 + F[4]*2 + F[6]*3 + F[2]*1; + f[2] = F[3]*4 + F[4]*4 + F[5]*9 + F[6]*9 + F[1]*1 + F[2]*1; + f[3] = F[1]*-1 + F[5]*-27 + F[3]*-8 + F[4]*8 + F[6]*27 + F[2]*1 + F[7]*1; + + // Write out the output tile + if (bptr != 0) + { + b = *(bptr++); + } + for (int j = 0; j < output_tile_cols; j++) + { + *(outptrs[j]++) = f[j] + b; + } + } +} + +template class OutputTransform<1, 5, 1, 8, float, float, WinogradRoots::Integers>; +template class OutputTransform<5, 1, 8, 1, float, float, WinogradRoots::Integers>; + +} // namespace winograd diff --git a/src/core/NEON/kernels/convolution/winograd/winograd_transforms/output_4x4_3x3_fp32_fp32_integers.cpp b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/output_4x4_3x3_fp32_fp32_integers.cpp new file mode 100644 index 0000000000..fd16a4df1c --- /dev/null +++ b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/output_4x4_3x3_fp32_fp32_integers.cpp @@ -0,0 +1,1855 @@ +/* + * Copyright (c) 2019 ARM Limited. + * + * SPDX-License-Identifier: MIT + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "arm.hpp" +#include "output.hpp" + +namespace winograd +{ + +#ifdef __aarch64__ + +template <> +void OutputTransform<3, 3, 6, 6, float, float, winograd::WinogradRoots::Integers>::transform_tile( + int n_channels, + const float* inptr, + const int matrix_stride, + const float* bptr, + float* output, + const int output_row_stride, + const int output_col_stride +) +{ + const float coeffs[2] = {2.0f, 4.0f}; + if (bptr != nullptr) + { + __asm__ __volatile__ ( + "ldr d0, [%[pcoeffs]]\n" + "add x21, %[in_col_stride1], %[in_col_stride1]\n" + "add x22, x21, %[in_col_stride1]\n" + "add x25, %[inptr0], %[in_row_stride]\n" + "add x15, %[output_col_stride1], %[output_col_stride1]\n" + "add x23, x22, %[in_col_stride1]\n" + "add x13, x25, %[in_row_stride]\n" + "add x16, x15, %[output_col_stride1]\n" + "add x24, x23, %[in_col_stride1]\n" + "add x26, x13, %[in_row_stride]\n" + "add x17, %[outptr0], %[output_row_stride]\n" + "add x14, x26, %[in_row_stride]\n" + "add x28, x17, %[output_row_stride]\n" + "lsr x19, %[n_channels], #2\n" + "add x27, x14, %[in_row_stride]\n" + "add x18, x28, %[output_row_stride]\n" + "and x20, %[n_channels], #3\n" + "cbz x19, 4f\n" + "1:\n" + "ldr q19, [%[inptr0]]\n" + "subs x19, x19, #1\n" + "ldr q20, [%[inptr0], %[in_col_stride1]]\n" + "ldr q4, [%[inptr0], x21]\n" + "fadd v1.4s, v20.4s, v4.4s\n" + "ldr q17, [%[inptr0], x22]\n" + "fsub v7.4s, v20.4s, v4.4s\n" + "ldr q22, [%[inptr0], x23]\n" + "fadd v5.4s, v17.4s, v22.4s\n" + "ldr q18, [%[inptr0], x24]\n" + "fsub v10.4s, v17.4s, v22.4s\n" + "ldr q25, [x25]\n" + "fadd v8.4s, v19.4s, v1.4s\n" + "ldr q12, [x25, %[in_col_stride1]]\n" + "mov v4.16b, v1.16b\n" + "ldr q23, [x25, x21]\n" + "mov v1.16b, v7.16b\n" + "ldr q9, [x25, x22]\n" + "fmul v10.4s, v10.4s, v0.s[0]\n" + "ldr q11, [x25, x23]\n" + "fadd v8.4s, v8.4s, v5.4s\n" + "ldr q6, [x25, x24]\n" + "fmla v4.4s, v5.4s, v0.s[1]\n" + "fadd v7.4s, v7.4s, v10.4s\n" + "fmla v1.4s, v10.4s, v0.s[1]\n" + "fadd v1.4s, v1.4s, v18.4s\n" + "beq 3f\n" + "2:\n" + "fadd v3.4s, v12.4s, v23.4s\n" + "ldr q2, [x13]\n" + "fadd v27.4s, v9.4s, v11.4s\n" + "ldr q21, [x13, %[in_col_stride1]]\n" + "fsub v16.4s, v12.4s, v23.4s\n" + "ldr q26, [x13, x21]\n" + "fsub v9.4s, v9.4s, v11.4s\n" + "ldr q17, [x13, x22]\n" + "fadd v14.4s, v25.4s, v3.4s\n" + "ldr q19, [x13, x23]\n" + "mov v11.16b, v3.16b\n" + "ldr q10, [x13, x24]\n" + "mov v3.16b, v16.16b\n" + "ldr q15, [x26]\n" + "fmul v9.4s, v9.4s, v0.s[0]\n" + "ldr q12, [x26, %[in_col_stride1]]\n" + "fadd v14.4s, v14.4s, v27.4s\n" + "ldr q20, [x26, x21]\n" + "fmla v11.4s, v27.4s, v0.s[1]\n" + "ldr q24, [x26, x22]\n" + "fadd v23.4s, v21.4s, v26.4s\n" + "ldr q29, [x26, x23]\n" + "fadd v13.4s, v16.4s, v9.4s\n" + "ldr q5, [x26, x24]\n" + "fmla v3.4s, v9.4s, v0.s[1]\n" + "ldr q18, [x14]\n" + "fadd v30.4s, v17.4s, v19.4s\n" + "add %[inptr0], %[inptr0], #16\n" + "fadd v16.4s, v2.4s, v23.4s\n" + "add x25, x25, #16\n" + "fsub v21.4s, v21.4s, v26.4s\n" + "ldr q22, [x14, %[in_col_stride1]]\n" + "fadd v3.4s, v3.4s, v6.4s\n" + "ldr q28, [x14, x21]\n" + "fsub v19.4s, v17.4s, v19.4s\n" + "add x13, x13, #16\n" + "fadd v16.4s, v16.4s, v30.4s\n" + "add x26, x26, #16\n" + "mov v17.16b, v23.16b\n" + "subs x19, x19, #1\n" + "fadd v26.4s, v12.4s, v20.4s\n" + "fsub v9.4s, v12.4s, v20.4s\n" + "fmul v19.4s, v19.4s, v0.s[0]\n" + "ldr q20, [x14, x22]\n" + "fmla v17.4s, v30.4s, v0.s[1]\n" + "fadd v25.4s, v24.4s, v29.4s\n" + "fsub v12.4s, v24.4s, v29.4s\n" + "fadd v24.4s, v22.4s, v28.4s\n" + "fadd v23.4s, v15.4s, v26.4s\n" + "mov v15.16b, v26.16b\n" + "fsub v22.4s, v22.4s, v28.4s\n" + "fadd v29.4s, v14.4s, v16.4s\n" + "fsub v16.4s, v14.4s, v16.4s\n" + "ldr q28, [x14, x23]\n" + "fmul v12.4s, v12.4s, v0.s[0]\n" + "fmla v15.4s, v25.4s, v0.s[1]\n" + "fadd v23.4s, v23.4s, v25.4s\n" + "mov v6.16b, v21.16b\n" + "fadd v30.4s, v21.4s, v19.4s\n" + "fadd v26.4s, v18.4s, v24.4s\n" + "mov v25.16b, v24.16b\n" + "fadd v18.4s, v8.4s, v29.4s\n" + "fmla v6.4s, v19.4s, v0.s[1]\n" + "fadd v27.4s, v20.4s, v28.4s\n" + "fsub v21.4s, v20.4s, v28.4s\n" + "mov v19.16b, v29.16b\n" + "fadd v29.4s, v13.4s, v30.4s\n" + "fsub v8.4s, v13.4s, v30.4s\n" + "fadd v14.4s, v9.4s, v12.4s\n" + "fadd v6.4s, v6.4s, v10.4s\n" + "ldr q20, [x14, x24]\n" + "fadd v26.4s, v26.4s, v27.4s\n" + "add x14, x14, #16\n" + "fmla v9.4s, v12.4s, v0.s[1]\n" + "ldr q24, [x27]\n" + "fmul v21.4s, v21.4s, v0.s[0]\n" + "fmla v25.4s, v27.4s, v0.s[1]\n" + "fadd v10.4s, v7.4s, v29.4s\n" + "ldr q2, [%[bptr]]\n" + "mov v7.16b, v29.16b\n" + "add %[bptr], %[bptr], #16\n" + "fadd v9.4s, v9.4s, v5.4s\n" + "fadd v13.4s, v23.4s, v26.4s\n" + "fsub v23.4s, v23.4s, v26.4s\n" + "fadd v27.4s, v11.4s, v17.4s\n" + "fsub v11.4s, v11.4s, v17.4s\n" + "fadd v30.4s, v15.4s, v25.4s\n" + "fsub v15.4s, v15.4s, v25.4s\n" + "ldr q28, [x27, %[in_col_stride1]]\n" + "fadd v18.4s, v18.4s, v13.4s\n" + "fmla v19.4s, v13.4s, v0.s[1]\n" + "fadd v26.4s, v22.4s, v21.4s\n" + "mov v12.16b, v22.16b\n" + "fmul v23.4s, v23.4s, v0.s[0]\n" + "fadd v17.4s, v4.4s, v27.4s\n" + "fmul v15.4s, v15.4s, v0.s[0]\n" + "mov v4.16b, v27.16b\n" + "fmla v12.4s, v21.4s, v0.s[1]\n" + "ldr q22, [x27, x21]\n" + "fadd v18.4s, v18.4s, v2.4s\n" + "fadd v19.4s, v19.4s, v2.4s\n" + "fadd v17.4s, v17.4s, v30.4s\n" + "fmla v4.4s, v30.4s, v0.s[1]\n" + "fadd v25.4s, v28.4s, v22.4s\n" + "fsub v27.4s, v28.4s, v22.4s\n" + "fadd v12.4s, v12.4s, v20.4s\n" + "ldr q29, [x27, x22]\n" + "str q18, [%[outptr0]]\n" + "fadd v22.4s, v16.4s, v23.4s\n" + "str q19, [x28]\n" + "fadd v28.4s, v24.4s, v25.4s\n" + "ldr q30, [x27, x23]\n" + "fadd v20.4s, v29.4s, v30.4s\n" + "fsub v18.4s, v29.4s, v30.4s\n" + "mov v21.16b, v25.16b\n" + "ldr q25, [x27, x24]\n" + "fmla v16.4s, v23.4s, v0.s[1]\n" + "ldr q19, [%[inptr0]]\n" + "fadd v17.4s, v17.4s, v2.4s\n" + "add x27, x27, #16\n" + "fadd v28.4s, v28.4s, v20.4s\n" + "fmul v18.4s, v18.4s, v0.s[0]\n" + "fmla v21.4s, v20.4s, v0.s[1]\n" + "ldr q20, [%[inptr0], %[in_col_stride1]]\n" + "fadd v22.4s, v22.4s, v2.4s\n" + "fadd v4.4s, v4.4s, v2.4s\n" + "str q17, [%[outptr0], x15]\n" + "mov v24.16b, v27.16b\n" + "fadd v23.4s, v27.4s, v18.4s\n" + "fadd v16.4s, v16.4s, v28.4s\n" + "fadd v13.4s, v14.4s, v26.4s\n" + "fsub v30.4s, v14.4s, v26.4s\n" + "str q22, [x17]\n" + "fmla v24.4s, v18.4s, v0.s[1]\n" + "str q4, [x28, x15]\n" + "mov v14.16b, v8.16b\n" + "fadd v29.4s, v11.4s, v15.4s\n" + "ldr q4, [%[inptr0], x21]\n" + "fadd v10.4s, v10.4s, v13.4s\n" + "ldr q17, [%[inptr0], x22]\n" + "fadd v24.4s, v24.4s, v25.4s\n" + "ldr q22, [%[inptr0], x23]\n" + "fmul v30.4s, v30.4s, v0.s[0]\n" + "fmla v7.4s, v13.4s, v0.s[1]\n" + "mov v26.16b, v11.16b\n" + "fadd v13.4s, v3.4s, v6.4s\n" + "fsub v3.4s, v3.4s, v6.4s\n" + "ldr q18, [%[inptr0], x24]\n" + "fadd v10.4s, v10.4s, v2.4s\n" + "fadd v29.4s, v29.4s, v2.4s\n" + "fadd v8.4s, v8.4s, v30.4s\n" + "fmla v14.4s, v30.4s, v0.s[1]\n" + "fmla v26.4s, v15.4s, v0.s[1]\n" + "ldr q25, [x25]\n" + "fadd v27.4s, v9.4s, v12.4s\n" + "fadd v1.4s, v1.4s, v13.4s\n" + "str q10, [%[outptr0], %[output_col_stride1]]\n" + "fsub v6.4s, v9.4s, v12.4s\n" + "str q29, [x17, x15]\n" + "fadd v14.4s, v14.4s, v23.4s\n" + "fadd v26.4s, v26.4s, v21.4s\n" + "ldr q12, [x25, %[in_col_stride1]]\n" + "fadd v1.4s, v1.4s, v27.4s\n" + "ldr q23, [x25, x21]\n" + "fmul v6.4s, v6.4s, v0.s[0]\n" + "ldr q9, [x25, x22]\n" + "mov v5.16b, v13.16b\n" + "ldr q11, [x25, x23]\n" + "mov v13.16b, v3.16b\n" + "fadd v8.4s, v8.4s, v2.4s\n" + "fadd v1.4s, v1.4s, v2.4s\n" + "fadd v7.4s, v7.4s, v2.4s\n" + "fadd v10.4s, v3.4s, v6.4s\n" + "fmla v5.4s, v27.4s, v0.s[1]\n" + "fmla v13.4s, v6.4s, v0.s[1]\n" + "ldr q6, [x25, x24]\n" + "str q8, [x17, %[output_col_stride1]]\n" + "fadd v16.4s, v16.4s, v2.4s\n" + "str q1, [%[outptr0], x16]\n" + "fadd v14.4s, v14.4s, v2.4s\n" + "str q7, [x28, %[output_col_stride1]]\n" + "fadd v10.4s, v10.4s, v2.4s\n" + "fadd v13.4s, v13.4s, v24.4s\n" + "add %[outptr0], %[outptr0], #16\n" + "str q16, [x18]\n" + "fadd v5.4s, v5.4s, v2.4s\n" + "str q14, [x18, %[output_col_stride1]]\n" + "fadd v26.4s, v26.4s, v2.4s\n" + "str q10, [x17, x16]\n" + "fadd v1.4s, v20.4s, v4.4s\n" + "fadd v13.4s, v13.4s, v2.4s\n" + "add x17, x17, #16\n" + "str q5, [x28, x16]\n" + "fadd v5.4s, v17.4s, v22.4s\n" + "str q26, [x18, x15]\n" + "fsub v7.4s, v20.4s, v4.4s\n" + "fadd v8.4s, v19.4s, v1.4s\n" + "add x28, x28, #16\n" + "str q13, [x18, x16]\n" + "mov v4.16b, v1.16b\n" + "fsub v10.4s, v17.4s, v22.4s\n" + "add x18, x18, #16\n" + "mov v1.16b, v7.16b\n" + "fadd v8.4s, v8.4s, v5.4s\n" + "fmla v4.4s, v5.4s, v0.s[1]\n" + "fmul v10.4s, v10.4s, v0.s[0]\n" + "fadd v7.4s, v7.4s, v10.4s\n" + "fmla v1.4s, v10.4s, v0.s[1]\n" + "fadd v1.4s, v1.4s, v18.4s\n" + "bne 2b\n" + "3:\n" + "fadd v3.4s, v12.4s, v23.4s\n" + "ldr q2, [x13]\n" + "fadd v27.4s, v9.4s, v11.4s\n" + "ldr q21, [x13, %[in_col_stride1]]\n" + "fsub v16.4s, v12.4s, v23.4s\n" + "ldr q26, [x13, x21]\n" + "fsub v9.4s, v9.4s, v11.4s\n" + "ldr q17, [x13, x22]\n" + "fadd v14.4s, v25.4s, v3.4s\n" + "ldr q19, [x13, x23]\n" + "mov v11.16b, v3.16b\n" + "ldr q10, [x13, x24]\n" + "mov v3.16b, v16.16b\n" + "ldr q15, [x26]\n" + "fmul v9.4s, v9.4s, v0.s[0]\n" + "ldr q12, [x26, %[in_col_stride1]]\n" + "fadd v14.4s, v14.4s, v27.4s\n" + "ldr q20, [x26, x21]\n" + "fmla v11.4s, v27.4s, v0.s[1]\n" + "ldr q24, [x26, x22]\n" + "fadd v23.4s, v21.4s, v26.4s\n" + "ldr q29, [x26, x23]\n" + "fadd v13.4s, v16.4s, v9.4s\n" + "ldr q5, [x26, x24]\n" + "fmla v3.4s, v9.4s, v0.s[1]\n" + "ldr q18, [x14]\n" + "fadd v30.4s, v17.4s, v19.4s\n" + "add %[inptr0], %[inptr0], #16\n" + "fadd v16.4s, v2.4s, v23.4s\n" + "add x25, x25, #16\n" + "fsub v21.4s, v21.4s, v26.4s\n" + "ldr q22, [x14, %[in_col_stride1]]\n" + "fadd v3.4s, v3.4s, v6.4s\n" + "ldr q28, [x14, x21]\n" + "fsub v19.4s, v17.4s, v19.4s\n" + "add x13, x13, #16\n" + "fadd v16.4s, v16.4s, v30.4s\n" + "add x26, x26, #16\n" + "mov v17.16b, v23.16b\n" + "fadd v26.4s, v12.4s, v20.4s\n" + "fsub v9.4s, v12.4s, v20.4s\n" + "ldr q2, [%[bptr]]\n" + "fmul v19.4s, v19.4s, v0.s[0]\n" + "add %[bptr], %[bptr], #16\n" + "fmla v17.4s, v30.4s, v0.s[1]\n" + "fadd v25.4s, v24.4s, v29.4s\n" + "fadd v23.4s, v15.4s, v26.4s\n" + "fsub v12.4s, v24.4s, v29.4s\n" + "mov v15.16b, v26.16b\n" + "fadd v24.4s, v22.4s, v28.4s\n" + "fsub v22.4s, v22.4s, v28.4s\n" + "fadd v29.4s, v14.4s, v16.4s\n" + "fsub v16.4s, v14.4s, v16.4s\n" + "ldr q20, [x14, x22]\n" + "fadd v23.4s, v23.4s, v25.4s\n" + "fmul v12.4s, v12.4s, v0.s[0]\n" + "fmla v15.4s, v25.4s, v0.s[1]\n" + "mov v6.16b, v21.16b\n" + "fadd v30.4s, v21.4s, v19.4s\n" + "fadd v26.4s, v18.4s, v24.4s\n" + "mov v25.16b, v24.16b\n" + "fadd v18.4s, v8.4s, v29.4s\n" + "fmla v6.4s, v19.4s, v0.s[1]\n" + "mov v19.16b, v29.16b\n" + "fadd v27.4s, v11.4s, v17.4s\n" + "fsub v11.4s, v11.4s, v17.4s\n" + "fadd v29.4s, v13.4s, v30.4s\n" + "fsub v8.4s, v13.4s, v30.4s\n" + "fadd v14.4s, v9.4s, v12.4s\n" + "fadd v6.4s, v6.4s, v10.4s\n" + "ldr q28, [x14, x23]\n" + "fadd v17.4s, v4.4s, v27.4s\n" + "mov v4.16b, v27.16b\n" + "fmla v9.4s, v12.4s, v0.s[1]\n" + "fadd v27.4s, v20.4s, v28.4s\n" + "fsub v21.4s, v20.4s, v28.4s\n" + "fadd v10.4s, v7.4s, v29.4s\n" + "mov v7.16b, v29.16b\n" + "fadd v13.4s, v3.4s, v6.4s\n" + "fsub v3.4s, v3.4s, v6.4s\n" + "ldr q20, [x14, x24]\n" + "fadd v9.4s, v9.4s, v5.4s\n" + "fadd v26.4s, v26.4s, v27.4s\n" + "fmul v21.4s, v21.4s, v0.s[0]\n" + "add x14, x14, #16\n" + "fmla v25.4s, v27.4s, v0.s[1]\n" + "mov v12.16b, v22.16b\n" + "fadd v1.4s, v1.4s, v13.4s\n" + "mov v5.16b, v13.16b\n" + "fadd v13.4s, v23.4s, v26.4s\n" + "fsub v23.4s, v23.4s, v26.4s\n" + "fadd v26.4s, v22.4s, v21.4s\n" + "ldr q24, [x27]\n" + "fmla v12.4s, v21.4s, v0.s[1]\n" + "fadd v30.4s, v15.4s, v25.4s\n" + "fsub v15.4s, v15.4s, v25.4s\n" + "ldr q28, [x27, %[in_col_stride1]]\n" + "fadd v18.4s, v18.4s, v13.4s\n" + "fmul v23.4s, v23.4s, v0.s[0]\n" + "fmla v19.4s, v13.4s, v0.s[1]\n" + "ldr q22, [x27, x21]\n" + "fadd v12.4s, v12.4s, v20.4s\n" + "ldr q29, [x27, x22]\n" + "fadd v17.4s, v17.4s, v30.4s\n" + "fmul v15.4s, v15.4s, v0.s[0]\n" + "fmla v4.4s, v30.4s, v0.s[1]\n" + "fadd v25.4s, v28.4s, v22.4s\n" + "fsub v27.4s, v28.4s, v22.4s\n" + "fadd v22.4s, v16.4s, v23.4s\n" + "fadd v18.4s, v18.4s, v2.4s\n" + "fadd v17.4s, v17.4s, v2.4s\n" + "fadd v19.4s, v19.4s, v2.4s\n" + "fadd v28.4s, v24.4s, v25.4s\n" + "mov v21.16b, v25.16b\n" + "fmla v16.4s, v23.4s, v0.s[1]\n" + "ldr q30, [x27, x23]\n" + "str q18, [%[outptr0]]\n" + "fadd v20.4s, v29.4s, v30.4s\n" + "str q17, [%[outptr0], x15]\n" + "fsub v18.4s, v29.4s, v30.4s\n" + "str q19, [x28]\n" + "mov v24.16b, v27.16b\n" + "fadd v13.4s, v14.4s, v26.4s\n" + "ldr q25, [x27, x24]\n" + "fadd v28.4s, v28.4s, v20.4s\n" + "add x27, x27, #16\n" + "fmul v18.4s, v18.4s, v0.s[0]\n" + "fmla v21.4s, v20.4s, v0.s[1]\n" + "fsub v30.4s, v14.4s, v26.4s\n" + "mov v14.16b, v8.16b\n" + "fadd v10.4s, v10.4s, v13.4s\n" + "fmla v7.4s, v13.4s, v0.s[1]\n" + "fadd v16.4s, v16.4s, v28.4s\n" + "fadd v29.4s, v11.4s, v15.4s\n" + "fadd v23.4s, v27.4s, v18.4s\n" + "fmla v24.4s, v18.4s, v0.s[1]\n" + "fmul v30.4s, v30.4s, v0.s[0]\n" + "mov v26.16b, v11.16b\n" + "fadd v27.4s, v9.4s, v12.4s\n" + "fsub v6.4s, v9.4s, v12.4s\n" + "mov v13.16b, v3.16b\n" + "fadd v10.4s, v10.4s, v2.4s\n" + "fadd v24.4s, v24.4s, v25.4s\n" + "fmla v26.4s, v15.4s, v0.s[1]\n" + "fadd v8.4s, v8.4s, v30.4s\n" + "fmla v14.4s, v30.4s, v0.s[1]\n" + "fadd v1.4s, v1.4s, v27.4s\n" + "fmul v6.4s, v6.4s, v0.s[0]\n" + "str q10, [%[outptr0], %[output_col_stride1]]\n" + "fmla v5.4s, v27.4s, v0.s[1]\n" + "fadd v26.4s, v26.4s, v21.4s\n" + "fadd v22.4s, v22.4s, v2.4s\n" + "fadd v14.4s, v14.4s, v23.4s\n" + "fadd v8.4s, v8.4s, v2.4s\n" + "fadd v10.4s, v3.4s, v6.4s\n" + "fmla v13.4s, v6.4s, v0.s[1]\n" + "fadd v1.4s, v1.4s, v2.4s\n" + "fadd v29.4s, v29.4s, v2.4s\n" + "str q22, [x17]\n" + "fadd v7.4s, v7.4s, v2.4s\n" + "str q8, [x17, %[output_col_stride1]]\n" + "fadd v4.4s, v4.4s, v2.4s\n" + "fadd v13.4s, v13.4s, v24.4s\n" + "fadd v10.4s, v10.4s, v2.4s\n" + "str q1, [%[outptr0], x16]\n" + "fadd v5.4s, v5.4s, v2.4s\n" + "str q29, [x17, x15]\n" + "fadd v16.4s, v16.4s, v2.4s\n" + "str q7, [x28, %[output_col_stride1]]\n" + "fadd v14.4s, v14.4s, v2.4s\n" + "str q10, [x17, x16]\n" + "fadd v26.4s, v26.4s, v2.4s\n" + "str q4, [x28, x15]\n" + "fadd v13.4s, v13.4s, v2.4s\n" + "str q5, [x28, x16]\n" + "add %[outptr0], %[outptr0], #16\n" + "str q16, [x18]\n" + "add x17, x17, #16\n" + "str q14, [x18, %[output_col_stride1]]\n" + "add x28, x28, #16\n" + "str q26, [x18, x15]\n" + "str q13, [x18, x16]\n" + "add x18, x18, #16\n" + "4:\n" + "cmp x20, #2\n" + "blt 5f\n" + "ldr d19, [%[inptr0]]\n" + "ldr d20, [%[inptr0], %[in_col_stride1]]\n" + "sub x20, x20, #2\n" + "ldr d4, [%[inptr0], x21]\n" + "ldr d17, [%[inptr0], x22]\n" + "fadd v1.4s, v20.4s, v4.4s\n" + "ldr d22, [%[inptr0], x23]\n" + "fadd v5.4s, v17.4s, v22.4s\n" + "ldr d18, [%[inptr0], x24]\n" + "fsub v7.4s, v20.4s, v4.4s\n" + "ldr d25, [x25]\n" + "fsub v10.4s, v17.4s, v22.4s\n" + "ldr d12, [x25, %[in_col_stride1]]\n" + "fadd v8.4s, v19.4s, v1.4s\n" + "ldr d23, [x25, x21]\n" + "mov v4.16b, v1.16b\n" + "ldr d9, [x25, x22]\n" + "mov v1.16b, v7.16b\n" + "ldr d11, [x25, x23]\n" + "fmul v10.4s, v10.4s, v0.s[0]\n" + "ldr d6, [x25, x24]\n" + "fadd v8.4s, v8.4s, v5.4s\n" + "ldr d2, [x13]\n" + "fmla v4.4s, v5.4s, v0.s[1]\n" + "ldr d21, [x13, %[in_col_stride1]]\n" + "fadd v3.4s, v12.4s, v23.4s\n" + "ldr d26, [x13, x21]\n" + "fadd v7.4s, v7.4s, v10.4s\n" + "ldr d17, [x13, x22]\n" + "fmla v1.4s, v10.4s, v0.s[1]\n" + "ldr d19, [x13, x23]\n" + "fadd v27.4s, v9.4s, v11.4s\n" + "ldr d10, [x13, x24]\n" + "fadd v14.4s, v25.4s, v3.4s\n" + "ldr d15, [x26]\n" + "fsub v16.4s, v12.4s, v23.4s\n" + "ldr d12, [x26, %[in_col_stride1]]\n" + "fadd v1.4s, v1.4s, v18.4s\n" + "ldr d20, [x26, x21]\n" + "fsub v9.4s, v9.4s, v11.4s\n" + "ldr d24, [x26, x22]\n" + "fadd v14.4s, v14.4s, v27.4s\n" + "ldr d29, [x26, x23]\n" + "mov v11.16b, v3.16b\n" + "ldr d5, [x26, x24]\n" + "mov v3.16b, v16.16b\n" + "ldr d18, [x14]\n" + "fmul v9.4s, v9.4s, v0.s[0]\n" + "add %[inptr0], %[inptr0], #8\n" + "fmla v11.4s, v27.4s, v0.s[1]\n" + "add x25, x25, #8\n" + "fadd v23.4s, v21.4s, v26.4s\n" + "add x13, x13, #8\n" + "fsub v21.4s, v21.4s, v26.4s\n" + "ldr d22, [x14, %[in_col_stride1]]\n" + "fadd v13.4s, v16.4s, v9.4s\n" + "add x26, x26, #8\n" + "fmla v3.4s, v9.4s, v0.s[1]\n" + "fadd v30.4s, v17.4s, v19.4s\n" + "fadd v16.4s, v2.4s, v23.4s\n" + "fsub v19.4s, v17.4s, v19.4s\n" + "mov v17.16b, v23.16b\n" + "fadd v26.4s, v12.4s, v20.4s\n" + "fsub v9.4s, v12.4s, v20.4s\n" + "ldr d28, [x14, x21]\n" + "fadd v3.4s, v3.4s, v6.4s\n" + "ldr d20, [x14, x22]\n" + "fadd v16.4s, v16.4s, v30.4s\n" + "fmul v19.4s, v19.4s, v0.s[0]\n" + "fmla v17.4s, v30.4s, v0.s[1]\n" + "fadd v25.4s, v24.4s, v29.4s\n" + "fadd v23.4s, v15.4s, v26.4s\n" + "fsub v12.4s, v24.4s, v29.4s\n" + "mov v15.16b, v26.16b\n" + "fadd v24.4s, v22.4s, v28.4s\n" + "fsub v22.4s, v22.4s, v28.4s\n" + "fadd v29.4s, v14.4s, v16.4s\n" + "fsub v16.4s, v14.4s, v16.4s\n" + "ldr d28, [x14, x23]\n" + "fadd v23.4s, v23.4s, v25.4s\n" + "fmul v12.4s, v12.4s, v0.s[0]\n" + "fmla v15.4s, v25.4s, v0.s[1]\n" + "mov v6.16b, v21.16b\n" + "fadd v30.4s, v21.4s, v19.4s\n" + "fadd v26.4s, v18.4s, v24.4s\n" + "mov v25.16b, v24.16b\n" + "fadd v18.4s, v8.4s, v29.4s\n" + "fmla v6.4s, v19.4s, v0.s[1]\n" + "fadd v27.4s, v20.4s, v28.4s\n" + "fsub v21.4s, v20.4s, v28.4s\n" + "mov v19.16b, v29.16b\n" + "fadd v29.4s, v13.4s, v30.4s\n" + "fsub v8.4s, v13.4s, v30.4s\n" + "fadd v14.4s, v9.4s, v12.4s\n" + "fadd v6.4s, v6.4s, v10.4s\n" + "ldr d20, [x14, x24]\n" + "fadd v26.4s, v26.4s, v27.4s\n" + "add x14, x14, #8\n" + "fmla v9.4s, v12.4s, v0.s[1]\n" + "ldr d24, [x27]\n" + "fmul v21.4s, v21.4s, v0.s[0]\n" + "fmla v25.4s, v27.4s, v0.s[1]\n" + "fadd v10.4s, v7.4s, v29.4s\n" + "ldr d2, [%[bptr]]\n" + "mov v7.16b, v29.16b\n" + "add %[bptr], %[bptr], #8\n" + "fadd v9.4s, v9.4s, v5.4s\n" + "fadd v13.4s, v23.4s, v26.4s\n" + "fsub v23.4s, v23.4s, v26.4s\n" + "fadd v27.4s, v11.4s, v17.4s\n" + "fsub v11.4s, v11.4s, v17.4s\n" + "fadd v30.4s, v15.4s, v25.4s\n" + "fsub v15.4s, v15.4s, v25.4s\n" + "ldr d28, [x27, %[in_col_stride1]]\n" + "fadd v18.4s, v18.4s, v13.4s\n" + "fmla v19.4s, v13.4s, v0.s[1]\n" + "fadd v26.4s, v22.4s, v21.4s\n" + "mov v12.16b, v22.16b\n" + "fmul v23.4s, v23.4s, v0.s[0]\n" + "fadd v17.4s, v4.4s, v27.4s\n" + "fmul v15.4s, v15.4s, v0.s[0]\n" + "mov v4.16b, v27.16b\n" + "fmla v12.4s, v21.4s, v0.s[1]\n" + "ldr d22, [x27, x21]\n" + "fadd v18.4s, v18.4s, v2.4s\n" + "fadd v19.4s, v19.4s, v2.4s\n" + "fadd v17.4s, v17.4s, v30.4s\n" + "fmla v4.4s, v30.4s, v0.s[1]\n" + "fadd v25.4s, v28.4s, v22.4s\n" + "fsub v27.4s, v28.4s, v22.4s\n" + "fadd v12.4s, v12.4s, v20.4s\n" + "ldr d29, [x27, x22]\n" + "str d18, [%[outptr0]]\n" + "fadd v22.4s, v16.4s, v23.4s\n" + "str d19, [x28]\n" + "fadd v28.4s, v24.4s, v25.4s\n" + "ldr d30, [x27, x23]\n" + "fadd v20.4s, v29.4s, v30.4s\n" + "fsub v18.4s, v29.4s, v30.4s\n" + "mov v21.16b, v25.16b\n" + "ldr d25, [x27, x24]\n" + "fmla v16.4s, v23.4s, v0.s[1]\n" + "add x27, x27, #8\n" + "mov v24.16b, v27.16b\n" + "fadd v17.4s, v17.4s, v2.4s\n" + "fadd v28.4s, v28.4s, v20.4s\n" + "fmul v18.4s, v18.4s, v0.s[0]\n" + "fmla v21.4s, v20.4s, v0.s[1]\n" + "fadd v13.4s, v14.4s, v26.4s\n" + "fsub v30.4s, v14.4s, v26.4s\n" + "mov v14.16b, v8.16b\n" + "str d17, [%[outptr0], x15]\n" + "fadd v29.4s, v11.4s, v15.4s\n" + "fadd v23.4s, v27.4s, v18.4s\n" + "fmla v24.4s, v18.4s, v0.s[1]\n" + "fadd v16.4s, v16.4s, v28.4s\n" + "fadd v10.4s, v10.4s, v13.4s\n" + "fmul v30.4s, v30.4s, v0.s[0]\n" + "fmla v7.4s, v13.4s, v0.s[1]\n" + "mov v26.16b, v11.16b\n" + "fadd v13.4s, v3.4s, v6.4s\n" + "fadd v24.4s, v24.4s, v25.4s\n" + "fadd v27.4s, v9.4s, v12.4s\n" + "fsub v3.4s, v3.4s, v6.4s\n" + "fsub v6.4s, v9.4s, v12.4s\n" + "fadd v8.4s, v8.4s, v30.4s\n" + "fmla v14.4s, v30.4s, v0.s[1]\n" + "fmla v26.4s, v15.4s, v0.s[1]\n" + "fadd v1.4s, v1.4s, v13.4s\n" + "mov v5.16b, v13.16b\n" + "fadd v10.4s, v10.4s, v2.4s\n" + "fmul v6.4s, v6.4s, v0.s[0]\n" + "mov v13.16b, v3.16b\n" + "fadd v14.4s, v14.4s, v23.4s\n" + "fadd v22.4s, v22.4s, v2.4s\n" + "fadd v26.4s, v26.4s, v21.4s\n" + "fadd v1.4s, v1.4s, v27.4s\n" + "str d10, [%[outptr0], %[output_col_stride1]]\n" + "fmla v5.4s, v27.4s, v0.s[1]\n" + "fadd v10.4s, v3.4s, v6.4s\n" + "fmla v13.4s, v6.4s, v0.s[1]\n" + "str d22, [x17]\n" + "fadd v8.4s, v8.4s, v2.4s\n" + "fadd v1.4s, v1.4s, v2.4s\n" + "fadd v29.4s, v29.4s, v2.4s\n" + "fadd v7.4s, v7.4s, v2.4s\n" + "fadd v4.4s, v4.4s, v2.4s\n" + "fadd v13.4s, v13.4s, v24.4s\n" + "fadd v10.4s, v10.4s, v2.4s\n" + "str d8, [x17, %[output_col_stride1]]\n" + "fadd v5.4s, v5.4s, v2.4s\n" + "str d1, [%[outptr0], x16]\n" + "fadd v16.4s, v16.4s, v2.4s\n" + "str d29, [x17, x15]\n" + "fadd v14.4s, v14.4s, v2.4s\n" + "str d10, [x17, x16]\n" + "fadd v26.4s, v26.4s, v2.4s\n" + "str d7, [x28, %[output_col_stride1]]\n" + "fadd v13.4s, v13.4s, v2.4s\n" + "str d4, [x28, x15]\n" + "add %[outptr0], %[outptr0], #8\n" + "str d5, [x28, x16]\n" + "add x17, x17, #8\n" + "str d16, [x18]\n" + "add x28, x28, #8\n" + "str d14, [x18, %[output_col_stride1]]\n" + "str d26, [x18, x15]\n" + "str d13, [x18, x16]\n" + "add x18, x18, #8\n" + "5:\n" + "cbz x20, 6f\n" + "ldr s19, [%[inptr0]]\n" + "ldr s20, [%[inptr0], %[in_col_stride1]]\n" + "ldr s4, [%[inptr0], x21]\n" + "fadd v1.4s, v20.4s, v4.4s\n" + "ldr s17, [%[inptr0], x22]\n" + "fsub v7.4s, v20.4s, v4.4s\n" + "ldr s22, [%[inptr0], x23]\n" + "fadd v5.4s, v17.4s, v22.4s\n" + "ldr s18, [%[inptr0], x24]\n" + "fsub v10.4s, v17.4s, v22.4s\n" + "ldr s25, [x25]\n" + "fadd v8.4s, v19.4s, v1.4s\n" + "ldr s12, [x25, %[in_col_stride1]]\n" + "mov v4.16b, v1.16b\n" + "ldr s23, [x25, x21]\n" + "mov v1.16b, v7.16b\n" + "ldr s9, [x25, x22]\n" + "fmul v10.4s, v10.4s, v0.s[0]\n" + "ldr s11, [x25, x23]\n" + "fadd v8.4s, v8.4s, v5.4s\n" + "ldr s6, [x25, x24]\n" + "fmla v4.4s, v5.4s, v0.s[1]\n" + "ldr s2, [x13]\n" + "fadd v3.4s, v12.4s, v23.4s\n" + "ldr s21, [x13, %[in_col_stride1]]\n" + "fadd v7.4s, v7.4s, v10.4s\n" + "ldr s26, [x13, x21]\n" + "fmla v1.4s, v10.4s, v0.s[1]\n" + "ldr s17, [x13, x22]\n" + "fadd v27.4s, v9.4s, v11.4s\n" + "ldr s19, [x13, x23]\n" + "fadd v14.4s, v25.4s, v3.4s\n" + "ldr s10, [x13, x24]\n" + "fsub v16.4s, v12.4s, v23.4s\n" + "ldr s15, [x26]\n" + "fadd v1.4s, v1.4s, v18.4s\n" + "ldr s12, [x26, %[in_col_stride1]]\n" + "fsub v9.4s, v9.4s, v11.4s\n" + "ldr s20, [x26, x21]\n" + "fadd v14.4s, v14.4s, v27.4s\n" + "ldr s24, [x26, x22]\n" + "mov v11.16b, v3.16b\n" + "ldr s29, [x26, x23]\n" + "mov v3.16b, v16.16b\n" + "ldr s5, [x26, x24]\n" + "fmul v9.4s, v9.4s, v0.s[0]\n" + "ldr s18, [x14]\n" + "fmla v11.4s, v27.4s, v0.s[1]\n" + "fadd v23.4s, v21.4s, v26.4s\n" + "fsub v21.4s, v21.4s, v26.4s\n" + "fadd v30.4s, v17.4s, v19.4s\n" + "fsub v19.4s, v17.4s, v19.4s\n" + "ldr s22, [x14, %[in_col_stride1]]\n" + "fadd v13.4s, v16.4s, v9.4s\n" + "fmla v3.4s, v9.4s, v0.s[1]\n" + "fadd v16.4s, v2.4s, v23.4s\n" + "mov v17.16b, v23.16b\n" + "fadd v26.4s, v12.4s, v20.4s\n" + "fsub v9.4s, v12.4s, v20.4s\n" + "fmul v19.4s, v19.4s, v0.s[0]\n" + "ldr s28, [x14, x21]\n" + "fadd v3.4s, v3.4s, v6.4s\n" + "ldr s20, [x14, x22]\n" + "fadd v16.4s, v16.4s, v30.4s\n" + "fmla v17.4s, v30.4s, v0.s[1]\n" + "fadd v25.4s, v24.4s, v29.4s\n" + "fadd v23.4s, v15.4s, v26.4s\n" + "fsub v12.4s, v24.4s, v29.4s\n" + "mov v15.16b, v26.16b\n" + "fadd v24.4s, v22.4s, v28.4s\n" + "fsub v22.4s, v22.4s, v28.4s\n" + "fadd v30.4s, v21.4s, v19.4s\n" + "mov v6.16b, v21.16b\n" + "fadd v23.4s, v23.4s, v25.4s\n" + "fmla v15.4s, v25.4s, v0.s[1]\n" + "fmul v12.4s, v12.4s, v0.s[0]\n" + "ldr s28, [x14, x23]\n" + "fmla v6.4s, v19.4s, v0.s[1]\n" + "fadd v27.4s, v20.4s, v28.4s\n" + "fadd v26.4s, v18.4s, v24.4s\n" + "fsub v21.4s, v20.4s, v28.4s\n" + "mov v25.16b, v24.16b\n" + "fadd v29.4s, v14.4s, v16.4s\n" + "fsub v16.4s, v14.4s, v16.4s\n" + "ldr s20, [x14, x24]\n" + "fadd v6.4s, v6.4s, v10.4s\n" + "ldr s24, [x27]\n" + "fadd v26.4s, v26.4s, v27.4s\n" + "fmul v21.4s, v21.4s, v0.s[0]\n" + "fmla v25.4s, v27.4s, v0.s[1]\n" + "fadd v18.4s, v8.4s, v29.4s\n" + "mov v19.16b, v29.16b\n" + "fadd v29.4s, v13.4s, v30.4s\n" + "fsub v8.4s, v13.4s, v30.4s\n" + "fadd v27.4s, v11.4s, v17.4s\n" + "fsub v11.4s, v11.4s, v17.4s\n" + "fadd v13.4s, v23.4s, v26.4s\n" + "fsub v23.4s, v23.4s, v26.4s\n" + "ldr s28, [x27, %[in_col_stride1]]\n" + "fadd v10.4s, v7.4s, v29.4s\n" + "mov v7.16b, v29.16b\n" + "fadd v17.4s, v4.4s, v27.4s\n" + "mov v4.16b, v27.16b\n" + "fadd v18.4s, v18.4s, v13.4s\n" + "fmla v19.4s, v13.4s, v0.s[1]\n" + "fmul v23.4s, v23.4s, v0.s[0]\n" + "fadd v30.4s, v15.4s, v25.4s\n" + "fsub v15.4s, v15.4s, v25.4s\n" + "fadd v13.4s, v3.4s, v6.4s\n" + "fsub v3.4s, v3.4s, v6.4s\n" + "ldr s2, [%[bptr]]\n" + "fadd v18.4s, v18.4s, v2.4s\n" + "fadd v19.4s, v19.4s, v2.4s\n" + "fadd v17.4s, v17.4s, v30.4s\n" + "fmla v4.4s, v30.4s, v0.s[1]\n" + "fadd v14.4s, v9.4s, v12.4s\n" + "fmul v15.4s, v15.4s, v0.s[0]\n" + "fadd v1.4s, v1.4s, v13.4s\n" + "str s18, [%[outptr0]]\n" + "fadd v26.4s, v22.4s, v21.4s\n" + "str s19, [x28]\n" + "fmla v9.4s, v12.4s, v0.s[1]\n" + "mov v12.16b, v22.16b\n" + "ldr s22, [x27, x21]\n" + "fadd v25.4s, v28.4s, v22.4s\n" + "fsub v27.4s, v28.4s, v22.4s\n" + "fadd v22.4s, v16.4s, v23.4s\n" + "fadd v9.4s, v9.4s, v5.4s\n" + "ldr s29, [x27, x22]\n" + "fmla v12.4s, v21.4s, v0.s[1]\n" + "ldr s30, [x27, x23]\n" + "fadd v28.4s, v24.4s, v25.4s\n" + "mov v21.16b, v25.16b\n" + "fmla v16.4s, v23.4s, v0.s[1]\n" + "ldr s25, [x27, x24]\n" + "mov v5.16b, v13.16b\n" + "fadd v17.4s, v17.4s, v2.4s\n" + "fadd v12.4s, v12.4s, v20.4s\n" + "fadd v20.4s, v29.4s, v30.4s\n" + "fsub v18.4s, v29.4s, v30.4s\n" + "mov v24.16b, v27.16b\n" + "fadd v22.4s, v22.4s, v2.4s\n" + "fadd v4.4s, v4.4s, v2.4s\n" + "str s17, [%[outptr0], x15]\n" + "fadd v13.4s, v14.4s, v26.4s\n" + "fadd v28.4s, v28.4s, v20.4s\n" + "fmla v21.4s, v20.4s, v0.s[1]\n" + "fmul v18.4s, v18.4s, v0.s[0]\n" + "fsub v30.4s, v14.4s, v26.4s\n" + "str s22, [x17]\n" + "mov v14.16b, v8.16b\n" + "str s4, [x28, x15]\n" + "fadd v10.4s, v10.4s, v13.4s\n" + "fadd v16.4s, v16.4s, v28.4s\n" + "fmla v7.4s, v13.4s, v0.s[1]\n" + "fadd v23.4s, v27.4s, v18.4s\n" + "fmla v24.4s, v18.4s, v0.s[1]\n" + "fmul v30.4s, v30.4s, v0.s[0]\n" + "fadd v29.4s, v11.4s, v15.4s\n" + "mov v26.16b, v11.16b\n" + "fadd v27.4s, v9.4s, v12.4s\n" + "fsub v6.4s, v9.4s, v12.4s\n" + "mov v13.16b, v3.16b\n" + "fadd v24.4s, v24.4s, v25.4s\n" + "fadd v10.4s, v10.4s, v2.4s\n" + "fadd v8.4s, v8.4s, v30.4s\n" + "fmla v14.4s, v30.4s, v0.s[1]\n" + "fmla v26.4s, v15.4s, v0.s[1]\n" + "fadd v1.4s, v1.4s, v27.4s\n" + "fmul v6.4s, v6.4s, v0.s[0]\n" + "fmla v5.4s, v27.4s, v0.s[1]\n" + "str s10, [%[outptr0], %[output_col_stride1]]\n" + "fadd v29.4s, v29.4s, v2.4s\n" + "fadd v14.4s, v14.4s, v23.4s\n" + "fadd v8.4s, v8.4s, v2.4s\n" + "fadd v26.4s, v26.4s, v21.4s\n" + "fadd v1.4s, v1.4s, v2.4s\n" + "fadd v10.4s, v3.4s, v6.4s\n" + "fmla v13.4s, v6.4s, v0.s[1]\n" + "str s29, [x17, x15]\n" + "fadd v7.4s, v7.4s, v2.4s\n" + "str s8, [x17, %[output_col_stride1]]\n" + "fadd v5.4s, v5.4s, v2.4s\n" + "str s1, [%[outptr0], x16]\n" + "fadd v16.4s, v16.4s, v2.4s\n" + "fadd v13.4s, v13.4s, v24.4s\n" + "fadd v10.4s, v10.4s, v2.4s\n" + "str s7, [x28, %[output_col_stride1]]\n" + "fadd v14.4s, v14.4s, v2.4s\n" + "str s5, [x28, x16]\n" + "fadd v26.4s, v26.4s, v2.4s\n" + "str s16, [x18]\n" + "fadd v13.4s, v13.4s, v2.4s\n" + "str s10, [x17, x16]\n" + "str s14, [x18, %[output_col_stride1]]\n" + "str s26, [x18, x15]\n" + "str s13, [x18, x16]\n" + "6:\n" + : [bptr] "+r" (bptr), [outptr0] "+r" (output), [inptr0] "+r" (inptr) + : [output_row_stride] "r" (output_row_stride * sizeof(float)), [output_col_stride1] "r" (output_col_stride * sizeof(float)), [pcoeffs] "r" (coeffs), [n_channels] "r" ((long) n_channels), [in_row_stride] "r" (6 * matrix_stride * sizeof(float)), [in_col_stride1] "r" (matrix_stride * sizeof(float)) + : "cc", "v0", "v1", "v10", "v11", "v12", "v13", "v14", "v15", "v16", "v17", "v18", "v19", "v2", "v20", "v21", "v22", "v23", "v24", "v25", "v26", "v27", "v28", "v29", "v3", "v30", "v4", "v5", "v6", "v7", "v8", "v9", "x13", "x14", "x15", "x16", "x17", "x18", "x19", "x20", "x21", "x22", "x23", "x24", "x25", "x26", "x27", "x28", "memory" + ); + } + else + { + __asm__ __volatile__ ( + "ldr d0, [%[pcoeffs]]\n" + "add x21, %[in_col_stride1], %[in_col_stride1]\n" // Compute input column stride 2 + "add x22, x21, %[in_col_stride1]\n" // Compute input column stride 3 + "add x25, %[inptr0], %[in_row_stride]\n" // Compute input row pointers + "add x15, %[output_col_stride1], %[output_col_stride1]\n" // Compute output column stride 2 + "add x23, x22, %[in_col_stride1]\n" // Compute input column stride 4 + "add x13, x25, %[in_row_stride]\n" // Compute input row pointers + "add x16, x15, %[output_col_stride1]\n" // Compute output column stride 3 + "add x24, x23, %[in_col_stride1]\n" // Compute input column stride 5 + "add x26, x13, %[in_row_stride]\n" // Compute input row pointers + "add x17, %[outptr0], %[output_row_stride]\n" // Compute output row pointer 1 + "add x14, x26, %[in_row_stride]\n" // Compute input row pointers + "add x28, x17, %[output_row_stride]\n" // Compute output row pointer 2 + "lsr x19, %[n_channels], #2\n" + "add x27, x14, %[in_row_stride]\n" // Compute input row pointers + "add x18, x28, %[output_row_stride]\n" // Compute output row pointer 3 + "and x20, %[n_channels], #3\n" + "cbz x19, 4f\n" + "1:\n" // Quad head + "ldr q17, [%[inptr0]]\n" + "subs x19, x19, #1\n" + "ldr q23, [%[inptr0], %[in_col_stride1]]\n" + "ldr q27, [%[inptr0], x21]\n" + "fadd v4.4s, v23.4s, v27.4s\n" + "ldr q24, [%[inptr0], x22]\n" + "fsub v13.4s, v23.4s, v27.4s\n" + "ldr q11, [%[inptr0], x23]\n" + "fadd v10.4s, v24.4s, v11.4s\n" + "ldr q12, [%[inptr0], x24]\n" + "fsub v11.4s, v24.4s, v11.4s\n" + "ldr q20, [x25]\n" + "fadd v7.4s, v17.4s, v4.4s\n" + "ldr q19, [x25, %[in_col_stride1]]\n" + "mov v4.16b, v4.16b\n" + "ldr q22, [x25, x21]\n" + "mov v1.16b, v13.16b\n" + "ldr q14, [x25, x22]\n" + "fmul v11.4s, v11.4s, v0.s[0]\n" + "ldr q18, [x25, x23]\n" + "fadd v7.4s, v7.4s, v10.4s\n" + "ldr q3, [x25, x24]\n" + "fmla v4.4s, v10.4s, v0.s[1]\n" + "fadd v8.4s, v13.4s, v11.4s\n" + "fmla v1.4s, v11.4s, v0.s[1]\n" + "fadd v1.4s, v1.4s, v12.4s\n" + "beq 3f\n" + "2:\n" // Quad loop + "fadd v2.4s, v19.4s, v22.4s\n" + "ldr q16, [x13]\n" + "fadd v23.4s, v14.4s, v18.4s\n" + "ldr q21, [x13, %[in_col_stride1]]\n" + "fsub v15.4s, v19.4s, v22.4s\n" + "ldr q24, [x13, x21]\n" + "fsub v31.4s, v14.4s, v18.4s\n" + "ldr q25, [x13, x22]\n" + "fadd v11.4s, v20.4s, v2.4s\n" + "ldr q17, [x13, x23]\n" + "mov v13.16b, v2.16b\n" + "ldr q9, [x13, x24]\n" + "mov v2.16b, v15.16b\n" + "ldr q6, [x26]\n" + "fmul v31.4s, v31.4s, v0.s[0]\n" + "ldr q19, [x26, %[in_col_stride1]]\n" + "fadd v11.4s, v11.4s, v23.4s\n" + "ldr q22, [x26, x21]\n" + "fmla v13.4s, v23.4s, v0.s[1]\n" + "ldr q12, [x26, x22]\n" + "fadd v29.4s, v21.4s, v24.4s\n" + "ldr q26, [x26, x23]\n" + "fadd v15.4s, v15.4s, v31.4s\n" + "ldr q5, [x26, x24]\n" + "fmla v2.4s, v31.4s, v0.s[1]\n" + "ldr q10, [x14]\n" + "fadd v18.4s, v25.4s, v17.4s\n" + "add %[inptr0], %[inptr0], #16\n" + "fadd v27.4s, v16.4s, v29.4s\n" + "add x25, x25, #16\n" + "fsub v14.4s, v21.4s, v24.4s\n" + "ldr q30, [x14, %[in_col_stride1]]\n" + "fadd v2.4s, v2.4s, v3.4s\n" + "ldr q31, [x14, x21]\n" + "fsub v28.4s, v25.4s, v17.4s\n" + "add x13, x13, #16\n" + "fadd v27.4s, v27.4s, v18.4s\n" + "add x26, x26, #16\n" + "mov v21.16b, v29.16b\n" + "subs x19, x19, #1\n" + "fadd v20.4s, v19.4s, v22.4s\n" + "fsub v17.4s, v19.4s, v22.4s\n" + "fmul v28.4s, v28.4s, v0.s[0]\n" + "ldr q23, [x14, x22]\n" + "fmla v21.4s, v18.4s, v0.s[1]\n" + "fadd v29.4s, v12.4s, v26.4s\n" + "fsub v16.4s, v12.4s, v26.4s\n" + "fadd v25.4s, v30.4s, v31.4s\n" + "fadd v24.4s, v6.4s, v20.4s\n" + "mov v6.16b, v20.16b\n" + "fsub v22.4s, v30.4s, v31.4s\n" + "fadd v31.4s, v11.4s, v27.4s\n" + "fsub v12.4s, v11.4s, v27.4s\n" + "ldr q26, [x14, x23]\n" + "fmul v16.4s, v16.4s, v0.s[0]\n" + "fmla v6.4s, v29.4s, v0.s[1]\n" + "fadd v24.4s, v24.4s, v29.4s\n" + "mov v3.16b, v14.16b\n" + "fadd v20.4s, v14.4s, v28.4s\n" + "fadd v29.4s, v10.4s, v25.4s\n" + "mov v10.16b, v25.16b\n" + "fadd v25.4s, v7.4s, v31.4s\n" + "fmla v3.4s, v28.4s, v0.s[1]\n" + "fadd v14.4s, v23.4s, v26.4s\n" + "fsub v23.4s, v23.4s, v26.4s\n" + "mov v26.16b, v31.16b\n" + "fadd v31.4s, v15.4s, v20.4s\n" + "fsub v11.4s, v15.4s, v20.4s\n" + "fadd v20.4s, v17.4s, v16.4s\n" + "mov v7.16b, v17.16b\n" + "fadd v3.4s, v3.4s, v9.4s\n" + "ldr q18, [x14, x24]\n" + "fadd v29.4s, v29.4s, v14.4s\n" + "add x14, x14, #16\n" + "fmla v7.4s, v16.4s, v0.s[1]\n" + "ldr q19, [x27]\n" + "fmul v23.4s, v23.4s, v0.s[0]\n" + "fmla v10.4s, v14.4s, v0.s[1]\n" + "fadd v15.4s, v8.4s, v31.4s\n" + "mov v14.16b, v31.16b\n" + "fadd v28.4s, v24.4s, v29.4s\n" + "fsub v24.4s, v24.4s, v29.4s\n" + "fadd v7.4s, v7.4s, v5.4s\n" + "ldr q27, [x27, %[in_col_stride1]]\n" + "fadd v30.4s, v13.4s, v21.4s\n" + "fsub v9.4s, v13.4s, v21.4s\n" + "fadd v17.4s, v22.4s, v23.4s\n" + "mov v8.16b, v22.16b\n" + "fadd v25.4s, v25.4s, v28.4s\n" + "fmul v24.4s, v24.4s, v0.s[0]\n" + "fmla v26.4s, v28.4s, v0.s[1]\n" + "ldr q29, [x27, x21]\n" + "fmla v8.4s, v23.4s, v0.s[1]\n" + "ldr q28, [x27, x22]\n" + "fadd v13.4s, v4.4s, v30.4s\n" + "mov v4.16b, v30.16b\n" + "str q25, [%[outptr0]]\n" // Store output (0, 0) + "fadd v16.4s, v27.4s, v29.4s\n" + "str q26, [x28]\n" // Store output (2, 0) + "fsub v29.4s, v27.4s, v29.4s\n" + "fadd v8.4s, v8.4s, v18.4s\n" + "ldr q23, [x27, x23]\n" + "fadd v30.4s, v28.4s, v23.4s\n" + "ldr q25, [x27, x24]\n" + "fadd v19.4s, v19.4s, v16.4s\n" + "add x27, x27, #16\n" + "fsub v27.4s, v28.4s, v23.4s\n" + "mov v16.16b, v16.16b\n" + "fadd v22.4s, v20.4s, v17.4s\n" + "fsub v20.4s, v20.4s, v17.4s\n" + "fadd v21.4s, v12.4s, v24.4s\n" + "mov v26.16b, v12.16b\n" + "fadd v19.4s, v19.4s, v30.4s\n" + "fmla v16.4s, v30.4s, v0.s[1]\n" + "fmul v27.4s, v27.4s, v0.s[0]\n" + "ldr q17, [%[inptr0]]\n" + "fmla v26.4s, v24.4s, v0.s[1]\n" + "ldr q23, [%[inptr0], %[in_col_stride1]]\n" + "str q21, [x17]\n" // Store output (1, 0) + "mov v5.16b, v29.16b\n" + "fadd v15.4s, v15.4s, v22.4s\n" + "fmul v20.4s, v20.4s, v0.s[0]\n" + "fadd v18.4s, v29.4s, v27.4s\n" + "fmla v14.4s, v22.4s, v0.s[1]\n" + "fmla v5.4s, v27.4s, v0.s[1]\n" + "ldr q27, [%[inptr0], x21]\n" + "fadd v26.4s, v26.4s, v19.4s\n" + "ldr q24, [%[inptr0], x22]\n" + "str q15, [%[outptr0], %[output_col_stride1]]\n" // Store output (0, 1) + "fadd v12.4s, v11.4s, v20.4s\n" + "str q14, [x28, %[output_col_stride1]]\n" // Store output (2, 1) + "mov v28.16b, v11.16b\n" + "fadd v5.4s, v5.4s, v25.4s\n" + "ldr q11, [%[inptr0], x23]\n" + "str q26, [x18]\n" // Store output (3, 0) + "fadd v21.4s, v6.4s, v10.4s\n" + "str q12, [x17, %[output_col_stride1]]\n" // Store output (1, 1) + "fmla v28.4s, v20.4s, v0.s[1]\n" + "fsub v10.4s, v6.4s, v10.4s\n" + "ldr q12, [%[inptr0], x24]\n" + "mov v15.16b, v9.16b\n" + "ldr q20, [x25]\n" + "fadd v13.4s, v13.4s, v21.4s\n" + "ldr q19, [x25, %[in_col_stride1]]\n" + "fadd v28.4s, v28.4s, v18.4s\n" + "ldr q22, [x25, x21]\n" + "fmul v10.4s, v10.4s, v0.s[0]\n" + "ldr q14, [x25, x22]\n" + "fmla v4.4s, v21.4s, v0.s[1]\n" + "ldr q18, [x25, x23]\n" + "str q13, [%[outptr0], x15]\n" // Store output (0, 2) + "fadd v6.4s, v2.4s, v3.4s\n" + "str q28, [x18, %[output_col_stride1]]\n" // Store output (3, 1) + "fadd v30.4s, v7.4s, v8.4s\n" + "fadd v13.4s, v9.4s, v10.4s\n" + "fmla v15.4s, v10.4s, v0.s[1]\n" + "str q4, [x28, x15]\n" // Store output (2, 2) + "fsub v2.4s, v2.4s, v3.4s\n" + "fadd v1.4s, v1.4s, v6.4s\n" + "ldr q3, [x25, x24]\n" + "fsub v8.4s, v7.4s, v8.4s\n" + "mov v6.16b, v6.16b\n" + "str q13, [x17, x15]\n" // Store output (1, 2) + "fadd v15.4s, v15.4s, v16.4s\n" + "mov v9.16b, v2.16b\n" + "fadd v4.4s, v23.4s, v27.4s\n" + "fadd v1.4s, v1.4s, v30.4s\n" + "fmla v6.4s, v30.4s, v0.s[1]\n" + "fmul v8.4s, v8.4s, v0.s[0]\n" + "fadd v10.4s, v24.4s, v11.4s\n" + "str q15, [x18, x15]\n" // Store output (3, 2) + "fsub v13.4s, v23.4s, v27.4s\n" + "fadd v7.4s, v17.4s, v4.4s\n" + "fsub v11.4s, v24.4s, v11.4s\n" + "str q1, [%[outptr0], x16]\n" // Store output (0, 3) + "mov v4.16b, v4.16b\n" + "str q6, [x28, x16]\n" // Store output (2, 3) + "fadd v2.4s, v2.4s, v8.4s\n" + "fmla v9.4s, v8.4s, v0.s[1]\n" + "add %[outptr0], %[outptr0], #16\n" + "fadd v7.4s, v7.4s, v10.4s\n" + "add x28, x28, #16\n" + "fmul v11.4s, v11.4s, v0.s[0]\n" + "fmla v4.4s, v10.4s, v0.s[1]\n" + "str q2, [x17, x16]\n" // Store output (1, 3) + "mov v1.16b, v13.16b\n" + "fadd v9.4s, v9.4s, v5.4s\n" + "add x17, x17, #16\n" + "fadd v8.4s, v13.4s, v11.4s\n" + "fmla v1.4s, v11.4s, v0.s[1]\n" + "str q9, [x18, x16]\n" // Store output (3, 3) + "add x18, x18, #16\n" + "fadd v1.4s, v1.4s, v12.4s\n" + "bne 2b\n" + "3:\n" // Quad tail + "fadd v2.4s, v19.4s, v22.4s\n" + "ldr q16, [x13]\n" + "fadd v23.4s, v14.4s, v18.4s\n" + "ldr q21, [x13, %[in_col_stride1]]\n" + "fsub v15.4s, v19.4s, v22.4s\n" + "ldr q24, [x13, x21]\n" + "fsub v31.4s, v14.4s, v18.4s\n" + "ldr q25, [x13, x22]\n" + "fadd v11.4s, v20.4s, v2.4s\n" + "ldr q17, [x13, x23]\n" + "mov v13.16b, v2.16b\n" + "ldr q9, [x13, x24]\n" + "mov v2.16b, v15.16b\n" + "ldr q6, [x26]\n" + "fmul v31.4s, v31.4s, v0.s[0]\n" + "ldr q19, [x26, %[in_col_stride1]]\n" + "fadd v11.4s, v11.4s, v23.4s\n" + "ldr q22, [x26, x21]\n" + "fmla v13.4s, v23.4s, v0.s[1]\n" + "ldr q12, [x26, x22]\n" + "fadd v29.4s, v21.4s, v24.4s\n" + "ldr q26, [x26, x23]\n" + "fadd v15.4s, v15.4s, v31.4s\n" + "ldr q5, [x26, x24]\n" + "fmla v2.4s, v31.4s, v0.s[1]\n" + "ldr q10, [x14]\n" + "fadd v18.4s, v25.4s, v17.4s\n" + "add %[inptr0], %[inptr0], #16\n" + "fadd v27.4s, v16.4s, v29.4s\n" + "add x25, x25, #16\n" + "fsub v14.4s, v21.4s, v24.4s\n" + "ldr q30, [x14, %[in_col_stride1]]\n" + "fadd v2.4s, v2.4s, v3.4s\n" + "ldr q31, [x14, x21]\n" + "fsub v28.4s, v25.4s, v17.4s\n" + "add x13, x13, #16\n" + "fadd v27.4s, v27.4s, v18.4s\n" + "add x26, x26, #16\n" + "mov v21.16b, v29.16b\n" + "fadd v20.4s, v19.4s, v22.4s\n" + "fsub v17.4s, v19.4s, v22.4s\n" + "fadd v29.4s, v12.4s, v26.4s\n" + "fmul v28.4s, v28.4s, v0.s[0]\n" + "fsub v16.4s, v12.4s, v26.4s\n" + "fmla v21.4s, v18.4s, v0.s[1]\n" + "ldr q23, [x14, x22]\n" + "fadd v24.4s, v6.4s, v20.4s\n" + "mov v6.16b, v20.16b\n" + "fadd v25.4s, v30.4s, v31.4s\n" + "fsub v22.4s, v30.4s, v31.4s\n" + "fadd v20.4s, v14.4s, v28.4s\n" + "mov v3.16b, v14.16b\n" + "fmul v16.4s, v16.4s, v0.s[0]\n" + "fmla v6.4s, v29.4s, v0.s[1]\n" + "fadd v24.4s, v24.4s, v29.4s\n" + "ldr q26, [x14, x23]\n" + "fmla v3.4s, v28.4s, v0.s[1]\n" + "fadd v14.4s, v23.4s, v26.4s\n" + "fadd v29.4s, v10.4s, v25.4s\n" + "fsub v23.4s, v23.4s, v26.4s\n" + "mov v10.16b, v25.16b\n" + "fadd v31.4s, v11.4s, v27.4s\n" + "fsub v12.4s, v11.4s, v27.4s\n" + "ldr q18, [x14, x24]\n" + "fadd v3.4s, v3.4s, v9.4s\n" + "ldr q19, [x27]\n" + "fadd v29.4s, v29.4s, v14.4s\n" + "add x14, x14, #16\n" + "fmul v23.4s, v23.4s, v0.s[0]\n" + "fmla v10.4s, v14.4s, v0.s[1]\n" + "fadd v25.4s, v7.4s, v31.4s\n" + "mov v26.16b, v31.16b\n" + "fadd v31.4s, v15.4s, v20.4s\n" + "fsub v11.4s, v15.4s, v20.4s\n" + "fadd v28.4s, v24.4s, v29.4s\n" + "fsub v24.4s, v24.4s, v29.4s\n" + "fadd v30.4s, v13.4s, v21.4s\n" + "fsub v9.4s, v13.4s, v21.4s\n" + "fadd v20.4s, v17.4s, v16.4s\n" + "mov v7.16b, v17.16b\n" + "fadd v15.4s, v8.4s, v31.4s\n" + "mov v14.16b, v31.16b\n" + "fadd v25.4s, v25.4s, v28.4s\n" + "fmul v24.4s, v24.4s, v0.s[0]\n" + "fmla v7.4s, v16.4s, v0.s[1]\n" + "ldr q27, [x27, %[in_col_stride1]]\n" + "fmla v26.4s, v28.4s, v0.s[1]\n" + "ldr q29, [x27, x21]\n" + "fadd v13.4s, v4.4s, v30.4s\n" + "mov v4.16b, v30.16b\n" + "str q25, [%[outptr0]]\n" // Store output (0, 0) + "fadd v17.4s, v22.4s, v23.4s\n" + "fadd v7.4s, v7.4s, v5.4s\n" + "ldr q28, [x27, x22]\n" + "str q26, [x28]\n" // Store output (2, 0) + "mov v8.16b, v22.16b\n" + "fadd v16.4s, v27.4s, v29.4s\n" + "fsub v29.4s, v27.4s, v29.4s\n" + "fadd v21.4s, v12.4s, v24.4s\n" + "mov v26.16b, v12.16b\n" + "fmla v8.4s, v23.4s, v0.s[1]\n" + "fadd v22.4s, v20.4s, v17.4s\n" + "fsub v20.4s, v20.4s, v17.4s\n" + "ldr q23, [x27, x23]\n" + "fadd v19.4s, v19.4s, v16.4s\n" + "mov v16.16b, v16.16b\n" + "str q21, [x17]\n" // Store output (1, 0) + "fadd v30.4s, v28.4s, v23.4s\n" + "fadd v8.4s, v8.4s, v18.4s\n" + "ldr q25, [x27, x24]\n" + "fsub v27.4s, v28.4s, v23.4s\n" + "add x27, x27, #16\n" + "mov v5.16b, v29.16b\n" + "fmla v26.4s, v24.4s, v0.s[1]\n" + "fadd v19.4s, v19.4s, v30.4s\n" + "fmla v16.4s, v30.4s, v0.s[1]\n" + "fadd v15.4s, v15.4s, v22.4s\n" + "fmul v20.4s, v20.4s, v0.s[0]\n" + "fmul v27.4s, v27.4s, v0.s[0]\n" + "fmla v14.4s, v22.4s, v0.s[1]\n" + "mov v28.16b, v11.16b\n" + "fadd v21.4s, v6.4s, v10.4s\n" + "fadd v26.4s, v26.4s, v19.4s\n" + "fsub v10.4s, v6.4s, v10.4s\n" + "str q15, [%[outptr0], %[output_col_stride1]]\n" // Store output (0, 1) + "fadd v12.4s, v11.4s, v20.4s\n" + "str q14, [x28, %[output_col_stride1]]\n" // Store output (2, 1) + "fadd v18.4s, v29.4s, v27.4s\n" + "fmla v5.4s, v27.4s, v0.s[1]\n" + "fmla v28.4s, v20.4s, v0.s[1]\n" + "str q26, [x18]\n" // Store output (3, 0) + "fadd v13.4s, v13.4s, v21.4s\n" + "str q12, [x17, %[output_col_stride1]]\n" // Store output (1, 1) + "fmul v10.4s, v10.4s, v0.s[0]\n" + "fmla v4.4s, v21.4s, v0.s[1]\n" + "mov v15.16b, v9.16b\n" + "fadd v5.4s, v5.4s, v25.4s\n" + "fadd v28.4s, v28.4s, v18.4s\n" + "str q13, [%[outptr0], x15]\n" // Store output (0, 2) + "fadd v6.4s, v2.4s, v3.4s\n" + "fadd v13.4s, v9.4s, v10.4s\n" + "fmla v15.4s, v10.4s, v0.s[1]\n" + "str q4, [x28, x15]\n" // Store output (2, 2) + "fadd v30.4s, v7.4s, v8.4s\n" + "str q28, [x18, %[output_col_stride1]]\n" // Store output (3, 1) + "fsub v2.4s, v2.4s, v3.4s\n" + "fadd v1.4s, v1.4s, v6.4s\n" + "fsub v8.4s, v7.4s, v8.4s\n" + "str q13, [x17, x15]\n" // Store output (1, 2) + "fadd v15.4s, v15.4s, v16.4s\n" + "mov v6.16b, v6.16b\n" + "mov v9.16b, v2.16b\n" + "fadd v1.4s, v1.4s, v30.4s\n" + "fmul v8.4s, v8.4s, v0.s[0]\n" + "str q15, [x18, x15]\n" // Store output (3, 2) + "fmla v6.4s, v30.4s, v0.s[1]\n" + "str q1, [%[outptr0], x16]\n" // Store output (0, 3) + "fadd v2.4s, v2.4s, v8.4s\n" + "str q6, [x28, x16]\n" // Store output (2, 3) + "fmla v9.4s, v8.4s, v0.s[1]\n" + "add %[outptr0], %[outptr0], #16\n" + "add x28, x28, #16\n" + "str q2, [x17, x16]\n" // Store output (1, 3) + "fadd v9.4s, v9.4s, v5.4s\n" + "add x17, x17, #16\n" + "str q9, [x18, x16]\n" // Store output (3, 3) + "add x18, x18, #16\n" + "4:\n" // Double + "cmp x20, #2\n" + "blt 5f\n" + "ldr d17, [%[inptr0]]\n" + "ldr d23, [%[inptr0], %[in_col_stride1]]\n" + "sub x20, x20, #2\n" + "ldr d27, [%[inptr0], x21]\n" + "ldr d24, [%[inptr0], x22]\n" + "fadd v4.4s, v23.4s, v27.4s\n" + "ldr d11, [%[inptr0], x23]\n" + "fadd v10.4s, v24.4s, v11.4s\n" + "ldr d12, [%[inptr0], x24]\n" + "fsub v13.4s, v23.4s, v27.4s\n" + "ldr d20, [x25]\n" + "fsub v11.4s, v24.4s, v11.4s\n" + "ldr d19, [x25, %[in_col_stride1]]\n" + "fadd v7.4s, v17.4s, v4.4s\n" + "ldr d22, [x25, x21]\n" + "mov v4.16b, v4.16b\n" + "ldr d14, [x25, x22]\n" + "mov v1.16b, v13.16b\n" + "ldr d18, [x25, x23]\n" + "fmul v11.4s, v11.4s, v0.s[0]\n" + "ldr d3, [x25, x24]\n" + "fadd v7.4s, v7.4s, v10.4s\n" + "ldr d16, [x13]\n" + "fmla v4.4s, v10.4s, v0.s[1]\n" + "ldr d21, [x13, %[in_col_stride1]]\n" + "fadd v2.4s, v19.4s, v22.4s\n" + "ldr d24, [x13, x21]\n" + "fadd v8.4s, v13.4s, v11.4s\n" + "ldr d25, [x13, x22]\n" + "fmla v1.4s, v11.4s, v0.s[1]\n" + "ldr d17, [x13, x23]\n" + "fadd v23.4s, v14.4s, v18.4s\n" + "ldr d9, [x13, x24]\n" + "fadd v11.4s, v20.4s, v2.4s\n" + "ldr d6, [x26]\n" + "fsub v15.4s, v19.4s, v22.4s\n" + "ldr d19, [x26, %[in_col_stride1]]\n" + "fadd v1.4s, v1.4s, v12.4s\n" + "ldr d22, [x26, x21]\n" + "fsub v31.4s, v14.4s, v18.4s\n" + "ldr d12, [x26, x22]\n" + "fadd v11.4s, v11.4s, v23.4s\n" + "ldr d26, [x26, x23]\n" + "mov v13.16b, v2.16b\n" + "ldr d5, [x26, x24]\n" + "mov v2.16b, v15.16b\n" + "ldr d10, [x14]\n" + "fmul v31.4s, v31.4s, v0.s[0]\n" + "add %[inptr0], %[inptr0], #8\n" + "fmla v13.4s, v23.4s, v0.s[1]\n" + "add x25, x25, #8\n" + "fadd v29.4s, v21.4s, v24.4s\n" + "add x13, x13, #8\n" + "fsub v14.4s, v21.4s, v24.4s\n" + "ldr d30, [x14, %[in_col_stride1]]\n" + "fadd v15.4s, v15.4s, v31.4s\n" + "add x26, x26, #8\n" + "fmla v2.4s, v31.4s, v0.s[1]\n" + "fadd v18.4s, v25.4s, v17.4s\n" + "fadd v27.4s, v16.4s, v29.4s\n" + "fsub v28.4s, v25.4s, v17.4s\n" + "mov v21.16b, v29.16b\n" + "fadd v20.4s, v19.4s, v22.4s\n" + "fsub v17.4s, v19.4s, v22.4s\n" + "ldr d31, [x14, x21]\n" + "fadd v2.4s, v2.4s, v3.4s\n" + "ldr d23, [x14, x22]\n" + "fadd v27.4s, v27.4s, v18.4s\n" + "fmul v28.4s, v28.4s, v0.s[0]\n" + "fmla v21.4s, v18.4s, v0.s[1]\n" + "fadd v29.4s, v12.4s, v26.4s\n" + "fadd v24.4s, v6.4s, v20.4s\n" + "fsub v16.4s, v12.4s, v26.4s\n" + "mov v6.16b, v20.16b\n" + "fadd v25.4s, v30.4s, v31.4s\n" + "fsub v22.4s, v30.4s, v31.4s\n" + "fadd v31.4s, v11.4s, v27.4s\n" + "fsub v12.4s, v11.4s, v27.4s\n" + "ldr d26, [x14, x23]\n" + "fadd v24.4s, v24.4s, v29.4s\n" + "fmul v16.4s, v16.4s, v0.s[0]\n" + "fmla v6.4s, v29.4s, v0.s[1]\n" + "mov v3.16b, v14.16b\n" + "fadd v20.4s, v14.4s, v28.4s\n" + "fadd v29.4s, v10.4s, v25.4s\n" + "mov v10.16b, v25.16b\n" + "fadd v25.4s, v7.4s, v31.4s\n" + "fmla v3.4s, v28.4s, v0.s[1]\n" + "fadd v14.4s, v23.4s, v26.4s\n" + "fsub v23.4s, v23.4s, v26.4s\n" + "mov v26.16b, v31.16b\n" + "fadd v31.4s, v15.4s, v20.4s\n" + "fsub v11.4s, v15.4s, v20.4s\n" + "fadd v20.4s, v17.4s, v16.4s\n" + "mov v7.16b, v17.16b\n" + "fadd v3.4s, v3.4s, v9.4s\n" + "ldr d18, [x14, x24]\n" + "fadd v29.4s, v29.4s, v14.4s\n" + "add x14, x14, #8\n" + "fmla v7.4s, v16.4s, v0.s[1]\n" + "ldr d19, [x27]\n" + "fmul v23.4s, v23.4s, v0.s[0]\n" + "fmla v10.4s, v14.4s, v0.s[1]\n" + "fadd v15.4s, v8.4s, v31.4s\n" + "mov v14.16b, v31.16b\n" + "fadd v28.4s, v24.4s, v29.4s\n" + "fsub v24.4s, v24.4s, v29.4s\n" + "fadd v7.4s, v7.4s, v5.4s\n" + "ldr d27, [x27, %[in_col_stride1]]\n" + "fadd v30.4s, v13.4s, v21.4s\n" + "fsub v9.4s, v13.4s, v21.4s\n" + "fadd v17.4s, v22.4s, v23.4s\n" + "mov v8.16b, v22.16b\n" + "fadd v25.4s, v25.4s, v28.4s\n" + "fmul v24.4s, v24.4s, v0.s[0]\n" + "fmla v26.4s, v28.4s, v0.s[1]\n" + "ldr d29, [x27, x21]\n" + "fmla v8.4s, v23.4s, v0.s[1]\n" + "ldr d28, [x27, x22]\n" + "fadd v13.4s, v4.4s, v30.4s\n" + "mov v4.16b, v30.16b\n" + "str d25, [%[outptr0]]\n" // Store output (0, 0) + "fadd v16.4s, v27.4s, v29.4s\n" + "str d26, [x28]\n" // Store output (2, 0) + "fsub v29.4s, v27.4s, v29.4s\n" + "fadd v8.4s, v8.4s, v18.4s\n" + "ldr d23, [x27, x23]\n" + "fadd v30.4s, v28.4s, v23.4s\n" + "ldr d25, [x27, x24]\n" + "fadd v19.4s, v19.4s, v16.4s\n" + "add x27, x27, #8\n" + "fsub v27.4s, v28.4s, v23.4s\n" + "mov v16.16b, v16.16b\n" + "fadd v22.4s, v20.4s, v17.4s\n" + "fsub v20.4s, v20.4s, v17.4s\n" + "fadd v21.4s, v12.4s, v24.4s\n" + "mov v26.16b, v12.16b\n" + "fadd v19.4s, v19.4s, v30.4s\n" + "fmla v16.4s, v30.4s, v0.s[1]\n" + "fmul v27.4s, v27.4s, v0.s[0]\n" + "mov v5.16b, v29.16b\n" + "fmla v26.4s, v24.4s, v0.s[1]\n" + "fadd v15.4s, v15.4s, v22.4s\n" + "str d21, [x17]\n" // Store output (1, 0) + "fmul v20.4s, v20.4s, v0.s[0]\n" + "fmla v14.4s, v22.4s, v0.s[1]\n" + "mov v28.16b, v11.16b\n" + "fadd v18.4s, v29.4s, v27.4s\n" + "fmla v5.4s, v27.4s, v0.s[1]\n" + "str d15, [%[outptr0], %[output_col_stride1]]\n" // Store output (0, 1) + "fadd v26.4s, v26.4s, v19.4s\n" + "fadd v12.4s, v11.4s, v20.4s\n" + "fmla v28.4s, v20.4s, v0.s[1]\n" + "str d14, [x28, %[output_col_stride1]]\n" // Store output (2, 1) + "fadd v21.4s, v6.4s, v10.4s\n" + "fadd v5.4s, v5.4s, v25.4s\n" + "fsub v10.4s, v6.4s, v10.4s\n" + "str d26, [x18]\n" // Store output (3, 0) + "mov v15.16b, v9.16b\n" + "str d12, [x17, %[output_col_stride1]]\n" // Store output (1, 1) + "fadd v28.4s, v28.4s, v18.4s\n" + "fadd v13.4s, v13.4s, v21.4s\n" + "fmla v4.4s, v21.4s, v0.s[1]\n" + "fmul v10.4s, v10.4s, v0.s[0]\n" + "fadd v6.4s, v2.4s, v3.4s\n" + "fadd v30.4s, v7.4s, v8.4s\n" + "fsub v2.4s, v2.4s, v3.4s\n" + "str d28, [x18, %[output_col_stride1]]\n" // Store output (3, 1) + "fsub v8.4s, v7.4s, v8.4s\n" + "str d13, [%[outptr0], x15]\n" // Store output (0, 2) + "str d4, [x28, x15]\n" // Store output (2, 2) + "fadd v13.4s, v9.4s, v10.4s\n" + "fmla v15.4s, v10.4s, v0.s[1]\n" + "fadd v1.4s, v1.4s, v6.4s\n" + "mov v6.16b, v6.16b\n" + "fmul v8.4s, v8.4s, v0.s[0]\n" + "mov v9.16b, v2.16b\n" + "str d13, [x17, x15]\n" // Store output (1, 2) + "fadd v15.4s, v15.4s, v16.4s\n" + "fadd v1.4s, v1.4s, v30.4s\n" + "fmla v6.4s, v30.4s, v0.s[1]\n" + "fadd v2.4s, v2.4s, v8.4s\n" + "fmla v9.4s, v8.4s, v0.s[1]\n" + "str d15, [x18, x15]\n" // Store output (3, 2) + "str d1, [%[outptr0], x16]\n" // Store output (0, 3) + "str d2, [x17, x16]\n" // Store output (1, 3) + "fadd v9.4s, v9.4s, v5.4s\n" + "str d6, [x28, x16]\n" // Store output (2, 3) + "add %[outptr0], %[outptr0], #8\n" + "add x17, x17, #8\n" + "add x28, x28, #8\n" + "str d9, [x18, x16]\n" // Store output (3, 3) + "add x18, x18, #8\n" + "5:\n" // Scalar + "cbz x20, 6f\n" + "ldr s17, [%[inptr0]]\n" + "ldr s23, [%[inptr0], %[in_col_stride1]]\n" + "ldr s27, [%[inptr0], x21]\n" + "fadd v4.4s, v23.4s, v27.4s\n" + "ldr s24, [%[inptr0], x22]\n" + "fsub v13.4s, v23.4s, v27.4s\n" + "ldr s11, [%[inptr0], x23]\n" + "fadd v10.4s, v24.4s, v11.4s\n" + "ldr s12, [%[inptr0], x24]\n" + "fsub v11.4s, v24.4s, v11.4s\n" + "ldr s20, [x25]\n" + "fadd v7.4s, v17.4s, v4.4s\n" + "ldr s19, [x25, %[in_col_stride1]]\n" + "mov v4.16b, v4.16b\n" + "ldr s22, [x25, x21]\n" + "mov v1.16b, v13.16b\n" + "ldr s14, [x25, x22]\n" + "fmul v11.4s, v11.4s, v0.s[0]\n" + "ldr s18, [x25, x23]\n" + "fadd v7.4s, v7.4s, v10.4s\n" + "ldr s3, [x25, x24]\n" + "fmla v4.4s, v10.4s, v0.s[1]\n" + "ldr s16, [x13]\n" + "fadd v2.4s, v19.4s, v22.4s\n" + "ldr s21, [x13, %[in_col_stride1]]\n" + "fadd v8.4s, v13.4s, v11.4s\n" + "ldr s24, [x13, x21]\n" + "fmla v1.4s, v11.4s, v0.s[1]\n" + "ldr s25, [x13, x22]\n" + "fadd v23.4s, v14.4s, v18.4s\n" + "ldr s17, [x13, x23]\n" + "fadd v11.4s, v20.4s, v2.4s\n" + "ldr s9, [x13, x24]\n" + "fsub v15.4s, v19.4s, v22.4s\n" + "ldr s6, [x26]\n" + "fadd v1.4s, v1.4s, v12.4s\n" + "ldr s19, [x26, %[in_col_stride1]]\n" + "fsub v31.4s, v14.4s, v18.4s\n" + "ldr s22, [x26, x21]\n" + "fadd v11.4s, v11.4s, v23.4s\n" + "ldr s12, [x26, x22]\n" + "mov v13.16b, v2.16b\n" + "ldr s26, [x26, x23]\n" + "mov v2.16b, v15.16b\n" + "ldr s5, [x26, x24]\n" + "fmul v31.4s, v31.4s, v0.s[0]\n" + "ldr s10, [x14]\n" + "fmla v13.4s, v23.4s, v0.s[1]\n" + "fadd v29.4s, v21.4s, v24.4s\n" + "fsub v14.4s, v21.4s, v24.4s\n" + "fadd v18.4s, v25.4s, v17.4s\n" + "fsub v28.4s, v25.4s, v17.4s\n" + "ldr s30, [x14, %[in_col_stride1]]\n" + "fadd v15.4s, v15.4s, v31.4s\n" + "fmla v2.4s, v31.4s, v0.s[1]\n" + "fadd v27.4s, v16.4s, v29.4s\n" + "mov v21.16b, v29.16b\n" + "fadd v20.4s, v19.4s, v22.4s\n" + "fsub v17.4s, v19.4s, v22.4s\n" + "fmul v28.4s, v28.4s, v0.s[0]\n" + "ldr s31, [x14, x21]\n" + "fadd v2.4s, v2.4s, v3.4s\n" + "ldr s23, [x14, x22]\n" + "fadd v27.4s, v27.4s, v18.4s\n" + "fmla v21.4s, v18.4s, v0.s[1]\n" + "fadd v29.4s, v12.4s, v26.4s\n" + "fadd v24.4s, v6.4s, v20.4s\n" + "fsub v16.4s, v12.4s, v26.4s\n" + "mov v6.16b, v20.16b\n" + "fadd v25.4s, v30.4s, v31.4s\n" + "fsub v22.4s, v30.4s, v31.4s\n" + "fadd v20.4s, v14.4s, v28.4s\n" + "mov v3.16b, v14.16b\n" + "fadd v24.4s, v24.4s, v29.4s\n" + "fmla v6.4s, v29.4s, v0.s[1]\n" + "fmul v16.4s, v16.4s, v0.s[0]\n" + "ldr s26, [x14, x23]\n" + "fmla v3.4s, v28.4s, v0.s[1]\n" + "fadd v14.4s, v23.4s, v26.4s\n" + "fadd v29.4s, v10.4s, v25.4s\n" + "fsub v23.4s, v23.4s, v26.4s\n" + "mov v10.16b, v25.16b\n" + "fadd v31.4s, v11.4s, v27.4s\n" + "fsub v12.4s, v11.4s, v27.4s\n" + "ldr s18, [x14, x24]\n" + "fadd v3.4s, v3.4s, v9.4s\n" + "ldr s19, [x27]\n" + "fadd v29.4s, v29.4s, v14.4s\n" + "fmul v23.4s, v23.4s, v0.s[0]\n" + "fmla v10.4s, v14.4s, v0.s[1]\n" + "fadd v25.4s, v7.4s, v31.4s\n" + "mov v26.16b, v31.16b\n" + "fadd v31.4s, v15.4s, v20.4s\n" + "fsub v11.4s, v15.4s, v20.4s\n" + "fadd v30.4s, v13.4s, v21.4s\n" + "fsub v9.4s, v13.4s, v21.4s\n" + "fadd v28.4s, v24.4s, v29.4s\n" + "fsub v24.4s, v24.4s, v29.4s\n" + "ldr s27, [x27, %[in_col_stride1]]\n" + "fadd v15.4s, v8.4s, v31.4s\n" + "mov v14.16b, v31.16b\n" + "fadd v13.4s, v4.4s, v30.4s\n" + "mov v4.16b, v30.16b\n" + "fadd v25.4s, v25.4s, v28.4s\n" + "fmla v26.4s, v28.4s, v0.s[1]\n" + "fmul v24.4s, v24.4s, v0.s[0]\n" + "fadd v21.4s, v6.4s, v10.4s\n" + "fsub v10.4s, v6.4s, v10.4s\n" + "fadd v6.4s, v2.4s, v3.4s\n" + "fsub v2.4s, v2.4s, v3.4s\n" + "ldr s29, [x27, x21]\n" + "str s25, [%[outptr0]]\n" // Store output (0, 0) + "fadd v20.4s, v17.4s, v16.4s\n" + "str s26, [x28]\n" // Store output (2, 0) + "mov v7.16b, v17.16b\n" + "fadd v17.4s, v22.4s, v23.4s\n" + "mov v8.16b, v22.16b\n" + "fadd v13.4s, v13.4s, v21.4s\n" + "fmul v10.4s, v10.4s, v0.s[0]\n" + "fmla v7.4s, v16.4s, v0.s[1]\n" + "ldr s28, [x27, x22]\n" + "fmla v8.4s, v23.4s, v0.s[1]\n" + "ldr s23, [x27, x23]\n" + "fmla v4.4s, v21.4s, v0.s[1]\n" + "ldr s25, [x27, x24]\n" + "str s13, [%[outptr0], x15]\n" // Store output (0, 2) + "fadd v16.4s, v27.4s, v29.4s\n" + "fadd v7.4s, v7.4s, v5.4s\n" + "fadd v30.4s, v28.4s, v23.4s\n" + "fadd v8.4s, v8.4s, v18.4s\n" + "fsub v29.4s, v27.4s, v29.4s\n" + "str s4, [x28, x15]\n" // Store output (2, 2) + "fsub v27.4s, v28.4s, v23.4s\n" + "fadd v19.4s, v19.4s, v16.4s\n" + "mov v16.16b, v16.16b\n" + "fadd v21.4s, v12.4s, v24.4s\n" + "mov v26.16b, v12.16b\n" + "mov v5.16b, v29.16b\n" + "fadd v22.4s, v20.4s, v17.4s\n" + "fmul v27.4s, v27.4s, v0.s[0]\n" + "fmla v16.4s, v30.4s, v0.s[1]\n" + "fadd v19.4s, v19.4s, v30.4s\n" + "fmla v26.4s, v24.4s, v0.s[1]\n" + "str s21, [x17]\n" // Store output (1, 0) + "fsub v20.4s, v20.4s, v17.4s\n" + "fadd v15.4s, v15.4s, v22.4s\n" + "fmla v14.4s, v22.4s, v0.s[1]\n" + "fadd v18.4s, v29.4s, v27.4s\n" + "fmla v5.4s, v27.4s, v0.s[1]\n" + "fadd v26.4s, v26.4s, v19.4s\n" + "mov v28.16b, v11.16b\n" + "fmul v20.4s, v20.4s, v0.s[0]\n" + "fadd v13.4s, v9.4s, v10.4s\n" + "str s15, [%[outptr0], %[output_col_stride1]]\n" // Store output (0, 1) + "mov v15.16b, v9.16b\n" + "str s14, [x28, %[output_col_stride1]]\n" // Store output (2, 1) + "fadd v5.4s, v5.4s, v25.4s\n" + "str s26, [x18]\n" // Store output (3, 0) + "fadd v30.4s, v7.4s, v8.4s\n" + "str s13, [x17, x15]\n" // Store output (1, 2) + "fadd v12.4s, v11.4s, v20.4s\n" + "fmla v28.4s, v20.4s, v0.s[1]\n" + "fmla v15.4s, v10.4s, v0.s[1]\n" + "fadd v1.4s, v1.4s, v6.4s\n" + "fsub v8.4s, v7.4s, v8.4s\n" + "mov v6.16b, v6.16b\n" + "mov v9.16b, v2.16b\n" + "str s12, [x17, %[output_col_stride1]]\n" // Store output (1, 1) + "fadd v28.4s, v28.4s, v18.4s\n" + "fadd v15.4s, v15.4s, v16.4s\n" + "fadd v1.4s, v1.4s, v30.4s\n" + "fmul v8.4s, v8.4s, v0.s[0]\n" + "fmla v6.4s, v30.4s, v0.s[1]\n" + "str s28, [x18, %[output_col_stride1]]\n" // Store output (3, 1) + "str s1, [%[outptr0], x16]\n" // Store output (0, 3) + "str s6, [x28, x16]\n" // Store output (2, 3) + "fadd v2.4s, v2.4s, v8.4s\n" + "str s15, [x18, x15]\n" // Store output (3, 2) + "fmla v9.4s, v8.4s, v0.s[1]\n" + "str s2, [x17, x16]\n" // Store output (1, 3) + "fadd v9.4s, v9.4s, v5.4s\n" + "str s9, [x18, x16]\n" // Store output (3, 3) + "6:\n" // End + : [outptr0] "+r" (output), [inptr0] "+r" (inptr) + : [output_col_stride1] "r" (output_col_stride * sizeof(float)), [pcoeffs] "r" (coeffs), [n_channels] "r" ((long) n_channels), [in_row_stride] "r" (6 * matrix_stride * sizeof(float)), [in_col_stride1] "r" (matrix_stride * sizeof(float)), [output_row_stride] "r" (output_row_stride * sizeof(float)) + : "cc", "v0", "v1", "v10", "v11", "v12", "v13", "v14", "v15", "v16", "v17", "v18", "v19", "v2", "v20", "v21", "v22", "v23", "v24", "v25", "v26", "v27", "v28", "v29", "v3", "v30", "v31", "v4", "v5", "v6", "v7", "v8", "v9", "x13", "x14", "x15", "x16", "x17", "x18", "x19", "x20", "x21", "x22", "x23", "x24", "x25", "x26", "x27", "x28", "memory" + ); + } +} + +#else + +template <> +void winograd::OutputTransform<3, 3, 6, 6, float, float, winograd::WinogradRoots::Integers>::transform_tile( + const int n_channels, + const float* inptr, + const int matrix_stride, + const float* bptr, + float* const output, + const int output_row_stride, + const int output_col_stride +) +{ + // Construct a map to the output cells + float *outptrs[output_tile_rows][output_tile_cols]; + for (int i = 0; i < output_tile_rows; i++) + { + for (int j = 0; j < output_tile_cols; j++) + { + outptrs[i][j] = output + i*output_row_stride + j*output_col_stride; + } + } + + // For each channel of the output + int channels_remaining = n_channels; +#ifdef __arm__ + for (; channels_remaining >= 2; channels_remaining -= 2) + { + // Matrices used and computed during this transform + float32x2_t F[6][6], FZ[6][4], f[4][4], b; + + // Read a 6x6 tile in the Winograd domain + for (int i = 0, m = 0; i < 6; i++) + { + for (int j = 0; j < 6; j++, m++) + { + F[i][j] = vld1_f32(inptr + m*matrix_stride); + } + } + inptr += 2; + + // Compute the matrix F Z + for (int i = 0; i < 6; i++) + { + // FZ[i][0] = 1*F[i][0] + 1*F[i][1] + 1*F[i][2] + 1*F[i][3] + 1*F[i][4]; + FZ[i][0] = vadd_f32(vadd_f32(vadd_f32(F[i][0], F[i][1]), vadd_f32(F[i][2], F[i][3])), F[i][4]); + + // FZ[i][1] = 1*F[i][1] + -1*F[i][2] + 2*F[i][3] + -2*F[i][4]; + FZ[i][1] = vmla_n_f32(vsub_f32(F[i][1], F[i][2]), vsub_f32(F[i][3], F[i][4]), 2.0f); + + // FZ[i][2] = 1*F[i][1] + 1*F[i][2] + 4*F[i][3] + 4*F[i][4]; + FZ[i][2] = vmla_n_f32(vadd_f32(F[i][1], F[i][2]), vadd_f32(F[i][3], F[i][4]), 4.0f); + + // FZ[i][3] = 1*F[i][1] + -1*F[i][2] + 8*F[i][3] + -8*F[i][4] + 1*F[i][5]; + FZ[i][3] = vadd_f32(vmla_n_f32(vsub_f32(F[i][1], F[i][2]), vsub_f32(F[i][3], F[i][4]), 8.0f), F[i][5]); + } + + // Compute the output tile f = ZT F Z + for (int j = 0; j < 4; j++) + { + // f[0][j] = 1*FZ[0][j] + 1*FZ[1][j] + 1*FZ[2][j] + 1*FZ[3][j] + 1*FZ[4][j]; + f[0][j] = vadd_f32(vadd_f32(vadd_f32(FZ[0][j], FZ[1][j]), vadd_f32(FZ[2][j], FZ[3][j])), FZ[4][j]); + + // f[1][j] = 1*FZ[1][j] + -1*FZ[2][j] + 2*FZ[3][j] + -2*FZ[4][j]; + f[1][j] = vmla_n_f32(vsub_f32(FZ[1][j], FZ[2][j]), vsub_f32(FZ[3][j], FZ[4][j]), 2.0f); + + // f[2][j] = 1*FZ[1][j] + 1*FZ[2][j] + 4*FZ[3][j] + 4*FZ[4][j]; + f[2][j] = vmla_n_f32(vadd_f32(FZ[1][j], FZ[2][j]), vadd_f32(FZ[3][j], FZ[4][j]), 4.0f); + + // f[3][j] = 1*FZ[1][j] + -1*FZ[2][j] + 8*FZ[3][j] + -8*FZ[4][j] + 1*FZ[5][j]; + f[3][j] = vadd_f32(vmla_n_f32(vsub_f32(FZ[1][j], FZ[2][j]), vsub_f32(FZ[3][j], FZ[4][j]), 8.0f), FZ[5][j]); + } + + // Write out the output tile + if (bptr != nullptr) + { + b = vld1_f32(bptr); + bptr += 2; + } + else + { + b = vdup_n_f32(0.0f); + } + for (int i = 0; i < output_tile_rows; i++) + { + for (int j = 0; j < output_tile_cols; j++) + { + vst1_f32(outptrs[i][j], vadd_f32(f[i][j], b)); + outptrs[i][j] += 2; + } + } + } +#endif // __arm__ + for (; channels_remaining; channels_remaining--) + { + // Matrices used and computed during this transform + float F[6][6], FZ[6][4], f[4][4], b; + + // Read a 6x6 tile in the Winograd domain + for (int i = 0, m = 0; i < 6; i++) + { + for (int j = 0; j < 6; j++, m++) + { + F[i][j] = *(inptr + m*matrix_stride); + } + } + inptr++; + + // Compute the matrix F Z + for (int i = 0; i < 6; i++) + { + FZ[i][0] = 1*F[i][0] + 1*F[i][1] + 1*F[i][2] + 1*F[i][3] + 1*F[i][4]; + FZ[i][1] = 1*F[i][1] + -1*F[i][2] + 2*F[i][3] + -2*F[i][4]; + FZ[i][2] = 1*F[i][1] + 1*F[i][2] + 4*F[i][3] + 4*F[i][4]; + FZ[i][3] = 1*F[i][1] + -1*F[i][2] + 8*F[i][3] + -8*F[i][4] + 1*F[i][5]; + } + + // Compute the output tile f = ZT F Z + for (int j = 0; j < 4; j++) + { + f[0][j] = 1*FZ[0][j] + 1*FZ[1][j] + 1*FZ[2][j] + 1*FZ[3][j] + 1*FZ[4][j]; + f[1][j] = 1*FZ[1][j] + -1*FZ[2][j] + 2*FZ[3][j] + -2*FZ[4][j]; + f[2][j] = 1*FZ[1][j] + 1*FZ[2][j] + 4*FZ[3][j] + 4*FZ[4][j]; + f[3][j] = 1*FZ[1][j] + -1*FZ[2][j] + 8*FZ[3][j] + -8*FZ[4][j] + 1*FZ[5][j]; + } + + // Write out the output tile + if (bptr != nullptr) + { + b = *(bptr++); + } + else + { + b = 0.0f; + } + for (int i = 0; i < output_tile_rows; i++) + { + for (int j = 0; j < output_tile_cols; j++) + { + *(outptrs[i][j]++) = f[i][j] + b; + } + } + } +} + +#endif + +template class OutputTransform<3, 3, 6, 6, float, float, winograd::WinogradRoots::Integers>; + +} // namespace winograd diff --git a/src/core/NEON/kernels/convolution/winograd/winograd_transforms/output_6_3_fp32_fp32_integers.cpp b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/output_6_3_fp32_fp32_integers.cpp new file mode 100644 index 0000000000..ce921cea01 --- /dev/null +++ b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/output_6_3_fp32_fp32_integers.cpp @@ -0,0 +1,149 @@ +/* + * Copyright (c) 2019 ARM Limited. + * + * SPDX-License-Identifier: MIT + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "output.hpp" +#include "arm.hpp" + +namespace winograd +{ + +template <> +void OutputTransform<1, 3, 1, 8, float, float, WinogradRoots::Integers>::transform_tile( + const int n_channels, + const float* inptr, + const int matrix_stride, + const float* bptr, + float* const output, + const int, // No need to stride across rows + const int output_col_stride +) +{ + // Construct a map to the output cells + float *outptrs[output_tile_cols]; + for (int j = 0; j < output_tile_cols; j++) + { + outptrs[j] = output + j*output_col_stride; + } + + // For each channel of the output + int channels_remaining = n_channels; +#ifdef __arm_any__ + for (; channels_remaining >= 4; channels_remaining -= 4) + { + // Matrices used and computed during this transform + float32x4_t F[inner_tile_cols], f[output_tile_cols], b = vdupq_n_f32(0.0f); + + // Read a 1x8 tile in the Winograd domain + for (int j = 0; j < inner_tile_cols; j++) + { + F[j] = vld1q_f32(inptr + j*matrix_stride); + } + inptr += 4; + + f[0] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(F[6], 1), F[5], 1), F[4], 1), F[3], 1), F[2], 1), F[1], 1), F[0], 1); + f[1] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(F[2], 1), F[6], 3), F[4], 2), F[3], -2), F[5], -3), F[1], -1); + f[2] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(F[2], 1), F[1], 1), F[6], 9), F[5], 9), F[4], 4), F[3], 4); + f[3] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(F[2], 1), F[6], 27), F[4], 8), F[3], -8), F[5], -27), F[1], -1); + f[4] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(F[2], 1), F[1], 1), F[6], 81), F[5], 81), F[4], 16), F[3], 16); + f[5] = vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmlaq_n_f32(vmulq_n_f32(F[7], 1), F[2], 1), F[6], 243), F[4], 32), F[3], -32), F[5], -243), F[1], -1); + + // Write out the output tile + if (bptr != 0) + { + b = vld1q_f32(bptr); + bptr += 4; + } + for (int j = 0; j < output_tile_cols; j++) + { + vst1q_f32(outptrs[j], f[j] + b); + outptrs[j] += 4; + } + } + for (; channels_remaining >= 2; channels_remaining -= 2) + { + // Matrices used and computed during this transform + float32x2_t F[inner_tile_cols], f[output_tile_cols], b = vdup_n_f32(0.0f); + + // Read a 1x8 tile in the Winograd domain + for (int j = 0; j < inner_tile_cols; j++) + { + F[j] = vld1_f32(inptr + j*matrix_stride); + } + inptr += 2; + + f[0] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(F[6], 1), F[5], 1), F[4], 1), F[3], 1), F[2], 1), F[1], 1), F[0], 1); + f[1] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(F[2], 1), F[6], 3), F[4], 2), F[3], -2), F[5], -3), F[1], -1); + f[2] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(F[2], 1), F[1], 1), F[6], 9), F[5], 9), F[4], 4), F[3], 4); + f[3] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(F[2], 1), F[6], 27), F[4], 8), F[3], -8), F[5], -27), F[1], -1); + f[4] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(F[2], 1), F[1], 1), F[6], 81), F[5], 81), F[4], 16), F[3], 16); + f[5] = vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmla_n_f32(vmul_n_f32(F[7], 1), F[2], 1), F[6], 243), F[4], 32), F[3], -32), F[5], -243), F[1], -1); + + // Write out the output tile + if (bptr != 0) + { + b = vld1_f32(bptr); + bptr += 2; + } + for (int j = 0; j < output_tile_cols; j++) + { + vst1_f32(outptrs[j], f[j] + b); + outptrs[j] += 2; + } + } +#endif // __arm_any__ + for (; channels_remaining; channels_remaining--) + { + // Matrices used and computed during this transform + float F[inner_tile_cols], f[output_tile_cols], b = 0.0f; + + // Read a 1x8 tile in the Winograd domain + for (int j = 0; j < inner_tile_cols; j++) + { + F[j] = *(inptr + j*matrix_stride); + } + inptr++; + + f[0] = F[0]*1 + F[1]*1 + F[2]*1 + F[3]*1 + F[4]*1 + F[5]*1 + F[6]*1; + f[1] = F[1]*-1 + F[5]*-3 + F[3]*-2 + F[4]*2 + F[6]*3 + F[2]*1; + f[2] = F[3]*4 + F[4]*4 + F[5]*9 + F[6]*9 + F[1]*1 + F[2]*1; + f[3] = F[1]*-1 + F[5]*-27 + F[3]*-8 + F[4]*8 + F[6]*27 + F[2]*1; + f[4] = F[3]*16 + F[4]*16 + F[5]*81 + F[6]*81 + F[1]*1 + F[2]*1; + f[5] = F[1]*-1 + F[5]*-243 + F[3]*-32 + F[4]*32 + F[6]*243 + F[2]*1 + F[7]*1; + + // Write out the output tile + if (bptr != 0) + { + b = *(bptr++); + } + for (int j = 0; j < output_tile_cols; j++) + { + *(outptrs[j]++) = f[j] + b; + } + } +} + +template class OutputTransform<1, 3, 1, 8, float, float, WinogradRoots::Integers>; +template class OutputTransform<3, 1, 8, 1, float, float, WinogradRoots::Integers>; + +} // namespace diff --git a/src/core/NEON/kernels/convolution/winograd/winograd_transforms/weights_2_7_fp32_fp32_integers.cpp b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/weights_2_7_fp32_fp32_integers.cpp new file mode 100644 index 0000000000..37ae43fdb0 --- /dev/null +++ b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/weights_2_7_fp32_fp32_integers.cpp @@ -0,0 +1,90 @@ +/* + * Copyright (c) 2019 ARM Limited. + * + * SPDX-License-Identifier: MIT + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "arm.hpp" +#include "kernel.hpp" + +namespace winograd +{ + +template <> +void WeightTransform<1, 7, 1, 8, float, float, WinogradRoots::Integers>::execute( + const int n_output_channels, + const int n_input_channels, + const float* const input, // NOTE: Data in HWIO order + float* const output, + const int matrix_stride, + const int matrix_row_stride +) +{ + // Get pointers to each cell of the weight tensor + const auto weight_col_stride = n_input_channels * n_output_channels; + const float *inptrs[kernel_cols]; + for (int j = 0; j < kernel_cols; j++) + { + inptrs[j] = input + j*weight_col_stride; + } + + // For each input channel + for (int ic = 0; ic < n_input_channels; ic++) + { + float *outptr = output + ic * matrix_row_stride; + + // For each output channel + int channels_remaining = n_output_channels; + for (; channels_remaining; channels_remaining--) + { + // Matrices used and computed in this kernel + float w[kernel_cols], V[inner_tile_cols]; + + // Read weights + for (int j = 0; j < kernel_cols; j++) + { + w[j] = *(inptrs[j]++); + } + + // Compute V = w WT + V[0] = (w[0]*-1) / 36.0f; + V[1] = (w[1]*-1 + w[3]*-1 + w[5]*-1 + w[0]*1 + w[2]*1 + w[4]*1 + w[6]*1) / 48.0f; + V[2] = (w[0]*1 + w[1]*1 + w[2]*1 + w[3]*1 + w[4]*1 + w[5]*1 + w[6]*1) / 48.0f; + V[3] = (w[0]*-1 + w[6]*-64 + w[4]*-16 + w[2]*-4 + w[1]*2 + w[3]*8 + w[5]*32) / 120.0f; + V[4] = (w[0]*-1 + w[6]*-64 + w[5]*-32 + w[4]*-16 + w[3]*-8 + w[2]*-4 + w[1]*-2) / 120.0f; + V[5] = (w[5]*-243 + w[3]*-27 + w[1]*-3 + w[2]*9 + w[4]*81 + w[6]*729 + w[0]*1) / 720.0f; + V[6] = (w[1]*3 + w[2]*9 + w[3]*27 + w[4]*81 + w[5]*243 + w[6]*729 + w[0]*1) / 720.0f; + V[7] = (w[6]*1) / 1.0f; + + // Store the transformed weights + for (int j = 0; j < inner_tile_cols; j++) + { + *(outptr + j*matrix_stride) = V[j]; + } + outptr++; + } + } +} + +template class WeightTransform<1, 7, 1, 8, float, float, WinogradRoots::Integers>; +template class WeightTransform<7, 1, 8, 1, float, float, WinogradRoots::Integers>; + +} // namespace winograd diff --git a/src/core/NEON/kernels/convolution/winograd/winograd_transforms/weights_2x2_3x3_fp32_fp32_integers.cpp b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/weights_2x2_3x3_fp32_fp32_integers.cpp new file mode 100644 index 0000000000..8fab6db1ba --- /dev/null +++ b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/weights_2x2_3x3_fp32_fp32_integers.cpp @@ -0,0 +1,220 @@ +/* + * Copyright (c) 2019 ARM Limited. + * + * SPDX-License-Identifier: MIT + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "arm.hpp" +#include "kernel.hpp" + +namespace winograd +{ + +template <> +void WeightTransform<3, 3, 4, 4, float, float, WinogradRoots::Integers>::execute( + const int n_output_channels, + const int n_input_channels, + const float* const input, + float* const output, + const int matrix_stride, + const int matrix_row_stride +) +{ + constexpr int inner_tile_i = 4; + constexpr int inner_tile_j = 4; + + // Get pointers to each cell of the weight tensor + const auto weight_col_stride = n_input_channels * n_output_channels; + const auto weight_row_stride = 3 * weight_col_stride; + const float *inptrs[3][3]; + for (int i = 0; i < 3; i++) + { + for (int j = 0; j < 3; j++) + { + inptrs[i][j] = input + i*weight_row_stride + j*weight_col_stride; + } + } + + // For each input channel + for (int ic = 0; ic < n_input_channels; ic++) + { + float *outptr = output + ic * matrix_row_stride; + + // For each output channel + int channels_remaining = n_output_channels; +#ifdef __aarch64__ + for (; channels_remaining >= 4; channels_remaining -= 4) + { + // Matrices used and computed in this kernel + float32x4_t w[3][3], Ww[inner_tile_i][3], V[inner_tile_i][inner_tile_j]; + + // Read weights + for (int i = 0; i < 3; i++) + { + for (int j = 0; j < 3; j++) + { + w[i][j] = vld1q_f32(inptrs[i][j]); + inptrs[i][j] += 4; + } + } + + // Compute the matrix W w + for (int j = 0; j < 3; j++) + { + Ww[0][j] = w[0][j]; + + // Ww[1][j] = 0.5*(w[0][j] + w[1][j] + w[2][j]); + Ww[1][j] = vmulq_n_f32(vaddq_f32(vaddq_f32(w[0][j], w[1][j]), w[2][j]), 0.5f); + + // Ww[2][j] = 0.5*(w[0][j] - w[1][j] + w[2][j]); + Ww[2][j] = vmulq_n_f32(vaddq_f32(vsubq_f32(w[0][j], w[1][j]), w[2][j]), 0.5f); + + Ww[3][j] = w[2][j]; + } + + // Compute V = W w WT + for (int i = 0; i < inner_tile_i; i++) + { + V[i][0] = Ww[i][0]; + + // V[i][1] = 0.5*(Ww[i][0] + Ww[i][1] + Ww[i][2]); + V[i][1] = vmulq_n_f32(vaddq_f32(vaddq_f32(Ww[i][0], Ww[i][1]), Ww[i][2]), 0.5f); + + // V[i][2] = 0.5*(Ww[i][0] - Ww[i][1] + Ww[i][2]); + V[i][2] = vmulq_n_f32(vaddq_f32(vsubq_f32(Ww[i][0], Ww[i][1]), Ww[i][2]), 0.5f); + + V[i][3] = Ww[i][2]; + } + + // Store the transformed weights + for (int i = 0, m = 0; i < inner_tile_i; i++) + { + for (int j = 0; j < inner_tile_j; j++, m++) + { + vst1q_f32(outptr + m*matrix_stride, V[i][j]); + } + } + outptr += 4; + } +#endif // __aarch64__ +#ifdef __arm_any__ + for (; channels_remaining >= 2; channels_remaining -= 2) + { + // Matrices used and computed in this kernel + float32x2_t w[3][3], Ww[inner_tile_i][3], V[inner_tile_i][inner_tile_j]; + + // Read weights + for (int i = 0; i < 3; i++) + { + for (int j = 0; j < 3; j++) + { + w[i][j] = vld1_f32(inptrs[i][j]); + inptrs[i][j] += 2; + } + } + + // Compute the matrix W w + for (int j = 0; j < 3; j++) + { + Ww[0][j] = w[0][j]; + + // Ww[1][j] = 0.5*(w[0][j] + w[1][j] + w[2][j]); + Ww[1][j] = vmul_n_f32(vadd_f32(vadd_f32(w[0][j], w[1][j]), w[2][j]), 0.5f); + + // Ww[2][j] = 0.5*(w[0][j] - w[1][j] + w[2][j]); + Ww[2][j] = vmul_n_f32(vadd_f32(vsub_f32(w[0][j], w[1][j]), w[2][j]), 0.5f); + + Ww[3][j] = w[2][j]; + } + + // Compute V = W w WT + for (int i = 0; i < inner_tile_i; i++) + { + V[i][0] = Ww[i][0]; + + // V[i][1] = 0.5*(Ww[i][0] + Ww[i][1] + Ww[i][2]); + V[i][1] = vmul_n_f32(vadd_f32(vadd_f32(Ww[i][0], Ww[i][1]), Ww[i][2]), 0.5f); + + // V[i][2] = 0.5*(Ww[i][0] - Ww[i][1] + Ww[i][2]); + V[i][2] = vmul_n_f32(vadd_f32(vsub_f32(Ww[i][0], Ww[i][1]), Ww[i][2]), 0.5f); + + V[i][3] = Ww[i][2]; + } + + // Store the transformed weights + for (int i = 0, m = 0; i < inner_tile_i; i++) + { + for (int j = 0; j < inner_tile_j; j++, m++) + { + vst1_f32(outptr + m*matrix_stride, V[i][j]); + } + } + outptr += 2; + } +#endif // __arm_any__ + for (; channels_remaining; channels_remaining--) + { + // Matrices used and computed in this kernel + float w[3][3], Ww[inner_tile_i][3], V[inner_tile_i][inner_tile_j]; + + // Read weights + for (int i = 0; i < 3; i++) + { + for (int j = 0; j < 3; j++) + { + w[i][j] = *(inptrs[i][j]++); + } + } + + // Compute the matrix W w + for (int j = 0; j < 3; j++) + { + Ww[0][j] = w[0][j]; + Ww[1][j] = 0.5*(w[0][j] + w[1][j] + w[2][j]); + Ww[2][j] = 0.5*(w[0][j] - w[1][j] + w[2][j]); + Ww[3][j] = w[2][j]; + } + + // Compute V = W w WT + for (int i = 0; i < inner_tile_i; i++) + { + V[i][0] = Ww[i][0]; + V[i][1] = 0.5*(Ww[i][0] + Ww[i][1] + Ww[i][2]); + V[i][2] = 0.5*(Ww[i][0] - Ww[i][1] + Ww[i][2]); + V[i][3] = Ww[i][2]; + } + + // Store the transformed weights + for (int i = 0, m = 0; i < inner_tile_i; i++) + { + for (int j = 0; j < inner_tile_j; j++, m++) + { + *(outptr + m*matrix_stride) = V[i][j]; + } + } + outptr++; + } + } +} + +template class WeightTransform<3, 3, 4, 4, float, float, WinogradRoots::Integers>; + +} // namespace diff --git a/src/core/NEON/kernels/convolution/winograd/winograd_transforms/weights_2x2_5x5_fp32_fp32_integers.cpp b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/weights_2x2_5x5_fp32_fp32_integers.cpp new file mode 100644 index 0000000000..79f4fa30c4 --- /dev/null +++ b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/weights_2x2_5x5_fp32_fp32_integers.cpp @@ -0,0 +1,401 @@ +/* + * Copyright (c) 2019 ARM Limited. + * + * SPDX-License-Identifier: MIT + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "arm.hpp" +#include "kernel.hpp" + +namespace winograd +{ + +template <> +void WeightTransform<5, 5, 6, 6, float, float, WinogradRoots::Integers>::execute( + const int n_output_channels, + const int n_input_channels, + const float* const input, + float* const output, + const int matrix_stride, + const int matrix_row_stride +) +{ + // Get pointers to each cell of the weight tensor + const auto weight_col_stride = n_input_channels * n_output_channels; + const auto weight_row_stride = 5 * weight_col_stride; + const float *inptrs[5][5]; + for (int i = 0; i < 5; i++) + { + for (int j = 0; j < 5; j++) + { + inptrs[i][j] = input + i*weight_row_stride + j*weight_col_stride; + } + } + + // For each input channel + for (int ic = 0; ic < n_input_channels; ic++) + { + float *outptr = output + ic * matrix_row_stride; + + // For each output channel + int channels_remaining = n_output_channels; +#ifdef __aarch64__ + for (; channels_remaining >= 4; channels_remaining -= 4) + { + // Matrices used and computed in this kernel + float32x4_t w[5][5], Ww[6][5], V[6][6]; + + // Read weights + for (int i = 0; i < 5; i++) + { + for (int j = 0; j < 5; j++) + { + w[i][j] = vld1q_f32(inptrs[i][j]); + inptrs[i][j] += 4; + } + } + + // Compute the matrix W w + for (int j = 0; j < 5; j++) + { + // Ww[0][j] = w[0][j]/4.0f; + Ww[0][j] = vmulq_n_f32(w[0][j], 1.0f/4.0f); + + // Ww[1][j] = -( w[0][j] + w[1][j] + w[2][j] + w[3][j] + w[4][j])/6.0f; + Ww[1][j] = vmulq_n_f32( + vaddq_f32( + vaddq_f32( + vaddq_f32(w[1][j], w[0][j]), + vaddq_f32(w[3][j], w[2][j]) + ), + w[4][j] + ), + -1.0f/6.0f + ); + + // Ww[2][j] = +(-w[0][j] + w[1][j] - w[2][j] + w[3][j] - w[4][j])/6.0f; + // Ww[2][j] = ((w[1][j] - w[0][j]) + (w[3][j] - w[2][j]) - w[4][j])/6.0f; + Ww[2][j] = vmulq_n_f32( + vsubq_f32( + vaddq_f32( + vsubq_f32(w[1][j], w[0][j]), + vsubq_f32(w[3][j], w[2][j]) + ), + w[4][j] + ), + 1.0f/6.0f + ); + + // Ww[3][j] = (w[0][j]/8.0f + w[1][j]/4.0f + w[2][j]/2.0f + w[3][j] + 2*w[4][j])/3.0f; + Ww[3][j] = vmulq_n_f32( + vmlaq_n_f32( + vaddq_f32( + vaddq_f32(vmulq_n_f32(w[0][j], 1.0f/8.0f), vmulq_n_f32(w[1][j], 1.0f/4.0f)), + vaddq_f32(vmulq_n_f32(w[2][j], 1.0f/2.0f), w[3][j]) + ), + w[4][j], 2.0f + ), + 1.0f/3.0f + ); + + // Ww[4][j] = (w[0][j]/8.0f - w[1][j]/4.0f + w[2][j]/2.0f - w[3][j] + 2*w[4][j])/3.0f; + Ww[4][j] = vmulq_n_f32( + vmlaq_n_f32( + vaddq_f32( + vsubq_f32(vmulq_n_f32(w[0][j], 1.0f/8.0f), vmulq_n_f32(w[1][j], 1.0f/4.0f)), + vsubq_f32(vmulq_n_f32(w[2][j], 1.0f/2.0f), w[3][j]) + ), + w[4][j], 2.0f + ), + 1.0f/3.0f + ); + + // Ww[5][j] = w[4][j]; + Ww[5][j] = w[4][j]; + } + + // Compute V = W w WT + for (int i = 0; i < 6; i++) + { + // V[i][0] = Ww[i][0]/4.0f; + V[i][0] = vmulq_n_f32(Ww[i][0], 1.0f/4.0f); + + // V[i][1] = -( Ww[i][0] + Ww[i][1] + Ww[i][2] + Ww[i][3] + Ww[i][4])/6.0f; + V[i][1] = vmulq_n_f32( + vaddq_f32( + vaddq_f32( + vaddq_f32(Ww[i][1], Ww[i][0]), + vaddq_f32(Ww[i][3], Ww[i][2]) + ), + Ww[i][4] + ), + -1.0f/6.0f + ); + + // V[i][2] = +(-Ww[i][0] + Ww[i][1] - Ww[i][2] + Ww[i][3] - Ww[i][4])/6.0f; + // V[i][2] = ((Ww[i][1] - Ww[i][0]) + (Ww[i][3] - Ww[i][2]) - Ww[i][4])/6.0f; + V[i][2] = vmulq_n_f32( + vsubq_f32( + vaddq_f32( + vsubq_f32(Ww[i][1], Ww[i][0]), + vsubq_f32(Ww[i][3], Ww[i][2]) + ), + Ww[i][4] + ), + 1.0f/6.0f + ); + + // V[i][3] = (Ww[i][0]/8.0f + Ww[i][1]/4.0f + Ww[i][2]/2.0f + Ww[i][3] + 2*Ww[i][4])/3.0f; + V[i][3] = vmulq_n_f32( + vmlaq_n_f32( + vaddq_f32( + vaddq_f32(vmulq_n_f32(Ww[i][0], 1.0f/8.0f), vmulq_n_f32(Ww[i][1], 1.0f/4.0f)), + vaddq_f32(vmulq_n_f32(Ww[i][2], 1.0f/2.0f), Ww[i][3]) + ), + Ww[i][4], 2.0f + ), + 1.0f/3.0f + ); + + // V[i][4] = (Ww[i][0]/8.0f - Ww[i][1]/4.0f + Ww[i][2]/2.0f - Ww[i][3] + 2*Ww[i][4])/3.0f; + V[i][4] = vmulq_n_f32( + vmlaq_n_f32( + vaddq_f32( + vsubq_f32(vmulq_n_f32(Ww[i][0], 1.0f/8.0f), vmulq_n_f32(Ww[i][1], 1.0f/4.0f)), + vsubq_f32(vmulq_n_f32(Ww[i][2], 1.0f/2.0f), Ww[i][3]) + ), + Ww[i][4], 2.0f + ), + 1.0f/3.0f + ); + + // V[i][5] = Ww[i][4]; + V[i][5] = Ww[i][4]; + } + + // Store the transformed weights + for (int i = 0, m = 0; i < 6; i++) + { + for (int j = 0; j < 6; j++, m++) + { + vst1q_f32(outptr + m*matrix_stride, V[i][j]); + } + } + outptr += 4; + } +#endif // __aarch64__ +#ifdef __arm_any__ + for (; channels_remaining >= 2; channels_remaining -= 2) + { + // Matrices used and computed in this kernel + float32x2_t w[5][5], Ww[6][5], V[6][6]; + + // Read weights + for (int i = 0; i < 5; i++) + { + for (int j = 0; j < 5; j++) + { + w[i][j] = vld1_f32(inptrs[i][j]); + inptrs[i][j] += 2; + } + } + + // Compute the matrix W w + for (int j = 0; j < 5; j++) + { + // Ww[0][j] = w[0][j]/4.0f; + Ww[0][j] = vmul_n_f32(w[0][j], 1.0f/4.0f); + + // Ww[1][j] = -( w[0][j] + w[1][j] + w[2][j] + w[3][j] + w[4][j])/6.0f; + Ww[1][j] = vmul_n_f32( + vadd_f32( + vadd_f32( + vadd_f32(w[1][j], w[0][j]), + vadd_f32(w[3][j], w[2][j]) + ), + w[4][j] + ), + -1.0f/6.0f + ); + + // Ww[2][j] = +(-w[0][j] + w[1][j] - w[2][j] + w[3][j] - w[4][j])/6.0f; + // Ww[2][j] = ((w[1][j] - w[0][j]) + (w[3][j] - w[2][j]) - w[4][j])/6.0f; + Ww[2][j] = vmul_n_f32( + vsub_f32( + vadd_f32( + vsub_f32(w[1][j], w[0][j]), + vsub_f32(w[3][j], w[2][j]) + ), + w[4][j] + ), + 1.0f/6.0f + ); + + // Ww[3][j] = (w[0][j]/8.0f + w[1][j]/4.0f + w[2][j]/2.0f + w[3][j] + 2*w[4][j])/3.0f; + Ww[3][j] = vmul_n_f32( + vmla_n_f32( + vadd_f32( + vadd_f32(vmul_n_f32(w[0][j], 1.0f/8.0f), vmul_n_f32(w[1][j], 1.0f/4.0f)), + vadd_f32(vmul_n_f32(w[2][j], 1.0f/2.0f), w[3][j]) + ), + w[4][j], 2.0f + ), + 1.0f/3.0f + ); + + // Ww[4][j] = (w[0][j]/8.0f - w[1][j]/4.0f + w[2][j]/2.0f - w[3][j] + 2*w[4][j])/3.0f; + Ww[4][j] = vmul_n_f32( + vmla_n_f32( + vadd_f32( + vsub_f32(vmul_n_f32(w[0][j], 1.0f/8.0f), vmul_n_f32(w[1][j], 1.0f/4.0f)), + vsub_f32(vmul_n_f32(w[2][j], 1.0f/2.0f), w[3][j]) + ), + w[4][j], 2.0f + ), + 1.0f/3.0f + ); + + // Ww[5][j] = w[4][j]; + Ww[5][j] = w[4][j]; + } + + // Compute V = W w WT + for (int i = 0; i < 6; i++) + { + // V[i][0] = Ww[i][0]/4.0f; + V[i][0] = vmul_n_f32(Ww[i][0], 1.0f/4.0f); + + // V[i][1] = -( Ww[i][0] + Ww[i][1] + Ww[i][2] + Ww[i][3] + Ww[i][4])/6.0f; + V[i][1] = vmul_n_f32( + vadd_f32( + vadd_f32( + vadd_f32(Ww[i][1], Ww[i][0]), + vadd_f32(Ww[i][3], Ww[i][2]) + ), + Ww[i][4] + ), + -1.0f/6.0f + ); + + // V[i][2] = +(-Ww[i][0] + Ww[i][1] - Ww[i][2] + Ww[i][3] - Ww[i][4])/6.0f; + // V[i][2] = ((Ww[i][1] - Ww[i][0]) + (Ww[i][3] - Ww[i][2]) - Ww[i][4])/6.0f; + V[i][2] = vmul_n_f32( + vsub_f32( + vadd_f32( + vsub_f32(Ww[i][1], Ww[i][0]), + vsub_f32(Ww[i][3], Ww[i][2]) + ), + Ww[i][4] + ), + 1.0f/6.0f + ); + + // V[i][3] = (Ww[i][0]/8.0f + Ww[i][1]/4.0f + Ww[i][2]/2.0f + Ww[i][3] + 2*Ww[i][4])/3.0f; + V[i][3] = vmul_n_f32( + vmla_n_f32( + vadd_f32( + vadd_f32(vmul_n_f32(Ww[i][0], 1.0f/8.0f), vmul_n_f32(Ww[i][1], 1.0f/4.0f)), + vadd_f32(vmul_n_f32(Ww[i][2], 1.0f/2.0f), Ww[i][3]) + ), + Ww[i][4], 2.0f + ), + 1.0f/3.0f + ); + + // V[i][4] = (Ww[i][0]/8.0f - Ww[i][1]/4.0f + Ww[i][2]/2.0f - Ww[i][3] + 2*Ww[i][4])/3.0f; + V[i][4] = vmul_n_f32( + vmla_n_f32( + vadd_f32( + vsub_f32(vmul_n_f32(Ww[i][0], 1.0f/8.0f), vmul_n_f32(Ww[i][1], 1.0f/4.0f)), + vsub_f32(vmul_n_f32(Ww[i][2], 1.0f/2.0f), Ww[i][3]) + ), + Ww[i][4], 2.0f + ), + 1.0f/3.0f + ); + + // V[i][5] = Ww[i][4]; + V[i][5] = Ww[i][4]; + } + + // Store the transformed weights + for (int i = 0, m = 0; i < 6; i++) + { + for (int j = 0; j < 6; j++, m++) + { + vst1_f32(outptr + m*matrix_stride, V[i][j]); + } + } + outptr += 2; + } +#endif // __arm_any__ + for (; channels_remaining; channels_remaining--) + { + // Matrices used and computed in this kernel + float w[5][5], Ww[6][5], V[6][6]; + + // Read weights + for (int i = 0; i < 5; i++) + { + for (int j = 0; j < 5; j++) + { + w[i][j] = *(inptrs[i][j]++); + } + } + + // Compute the matrix W w + for (int j = 0; j < 5; j++) + { + Ww[0][j] = w[0][j]/4.0f; + Ww[1][j] = -( w[0][j] + w[1][j] + w[2][j] + w[3][j] + w[4][j])/6.0f; + Ww[2][j] = +(-w[0][j] + w[1][j] - w[2][j] + w[3][j] - w[4][j])/6.0f; + Ww[3][j] = (w[0][j]/8.0f + w[1][j]/4.0f + w[2][j]/2.0f + w[3][j] + 2*w[4][j])/3.0f; + Ww[4][j] = (w[0][j]/8.0f - w[1][j]/4.0f + w[2][j]/2.0f - w[3][j] + 2*w[4][j])/3.0f; + Ww[5][j] = w[4][j]; + } + + // Compute V = W w WT + for (int i = 0; i < 6; i++) + { + V[i][0] = Ww[i][0]/4.0f; + V[i][1] = -( Ww[i][0] + Ww[i][1] + Ww[i][2] + Ww[i][3] + Ww[i][4])/6.0f; + V[i][2] = +(-Ww[i][0] + Ww[i][1] - Ww[i][2] + Ww[i][3] - Ww[i][4])/6.0f; + V[i][3] = (Ww[i][0]/8.0f + Ww[i][1]/4.0f + Ww[i][2]/2.0f + Ww[i][3] + 2*Ww[i][4])/3.0f; + V[i][4] = (Ww[i][0]/8.0f - Ww[i][1]/4.0f + Ww[i][2]/2.0f - Ww[i][3] + 2*Ww[i][4])/3.0f; + V[i][5] = Ww[i][4]; + } + + // Store the transformed weights + for (int i = 0, m = 0; i < 6; i++) + { + for (int j = 0; j < 6; j++, m++) + { + *(outptr + m*matrix_stride) = V[i][j]; + } + } + outptr++; + } + } +} + +template class WeightTransform<5, 5, 6, 6, float, float, WinogradRoots::Integers>; + +} // namespace winograd diff --git a/src/core/NEON/kernels/convolution/winograd/winograd_transforms/weights_4_5_fp32_fp32_integers.cpp b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/weights_4_5_fp32_fp32_integers.cpp new file mode 100644 index 0000000000..fb3d712954 --- /dev/null +++ b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/weights_4_5_fp32_fp32_integers.cpp @@ -0,0 +1,90 @@ +/* + * Copyright (c) 2019 ARM Limited. + * + * SPDX-License-Identifier: MIT + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "arm.hpp" +#include "kernel.hpp" + +namespace winograd +{ + +template <> +void WeightTransform<1, 5, 1, 8, float, float, WinogradRoots::Integers>::execute( + const int n_output_channels, + const int n_input_channels, + const float* const input, // NOTE: Data in HWIO order + float* const output, + const int matrix_stride, + const int matrix_row_stride +) +{ + // Get pointers to each cell of the weight tensor + const auto weight_col_stride = n_input_channels * n_output_channels; + const float *inptrs[kernel_cols]; + for (int j = 0; j < kernel_cols; j++) + { + inptrs[j] = input + j*weight_col_stride; + } + + // For each input channel + for (int ic = 0; ic < n_input_channels; ic++) + { + float *outptr = output + ic * matrix_row_stride; + + // For each output channel + int channels_remaining = n_output_channels; + for (; channels_remaining; channels_remaining--) + { + // Matrices used and computed in this kernel + float w[kernel_cols], V[inner_tile_cols]; + + // Read weights + for (int j = 0; j < kernel_cols; j++) + { + w[j] = *(inptrs[j]++); + } + + // Compute V = w WT + V[0] = (w[0]*-1) / 36; + V[1] = (w[1]*-1 + w[3]*-1 + w[0]*1 + w[2]*1 + w[4]*1) / 48; + V[2] = (w[0]*1 + w[1]*1 + w[2]*1 + w[3]*1 + w[4]*1) / 48; + V[3] = (w[0]*-1 + w[4]*-16 + w[2]*-4 + w[1]*2 + w[3]*8) / 120; + V[4] = (w[0]*-1 + w[4]*-16 + w[3]*-8 + w[2]*-4 + w[1]*-2) / 120; + V[5] = (w[3]*-27 + w[1]*-3 + w[2]*9 + w[4]*81 + w[0]*1) / 720; + V[6] = (w[1]*3 + w[2]*9 + w[3]*27 + w[4]*81 + w[0]*1) / 720; + V[7] = (w[4]*1) / 1; + + // Store the transformed weights + for (int j = 0; j < inner_tile_cols; j++) + { + *(outptr + j*matrix_stride) = V[j]; + } + outptr++; + } + } +} + +template class WeightTransform<1, 5, 1, 8, float, float, WinogradRoots::Integers>; +template class WeightTransform<5, 1, 8, 1, float, float, WinogradRoots::Integers>; + +} // namespace winograd diff --git a/src/core/NEON/kernels/convolution/winograd/winograd_transforms/weights_4x4_3x3_fp32_fp32_integers.cpp b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/weights_4x4_3x3_fp32_fp32_integers.cpp new file mode 100644 index 0000000000..9e7040bca8 --- /dev/null +++ b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/weights_4x4_3x3_fp32_fp32_integers.cpp @@ -0,0 +1,257 @@ +/* + * Copyright (c) 2019 ARM Limited. + * + * SPDX-License-Identifier: MIT + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "arm.hpp" +#include "kernel.hpp" + +namespace winograd +{ + +template <> +void WeightTransform<3, 3, 6, 6, float, float, WinogradRoots::Integers>::execute( + const int n_output_channels, + const int n_input_channels, + const float* const input, // NOTE: Data in HWIO order + float* const output, + const int matrix_stride, + const int matrix_row_stride +) +{ + // Get pointers to each cell of the weight tensor + const auto weight_col_stride = n_input_channels * n_output_channels; + const auto weight_row_stride = 3 * weight_col_stride; + const float *inptrs[3][3]; + for (int i = 0; i < 3; i++) + { + for (int j = 0; j < 3; j++) + { + inptrs[i][j] = input + i*weight_row_stride + j*weight_col_stride; + } + } + + // For each input channel + for (int ic = 0; ic < n_input_channels; ic++) + { + float *outptr = output + ic * matrix_row_stride; + + // For each output channel + int channels_remaining = n_output_channels; +#ifdef __aarch64__ + for (; channels_remaining >= 4; channels_remaining -= 4) + { + // Matrices used and computed in this kernel + float32x4_t w[3][3], Ww[6][3], V[6][6]; + + // Read weights + for (int i = 0; i < 3; i++) + { + for (int j = 0; j < 3; j++) + { + w[i][j] = vld1q_f32(inptrs[i][j]); + inptrs[i][j] += 4; + } + } + + // Compute the matrix W w + for (int j = 0; j < 3; j++) + { + // Ww[0][j] = 6*w[0][j]; + Ww[0][j] = vmulq_n_f32(w[0][j], 6.0); + + // Ww[1][j] = -4*w[0][j] + -4*w[1][j] + -4*w[2][j]; + Ww[1][j] = vmulq_n_f32(vaddq_f32(vaddq_f32(w[0][j], w[1][j]), w[2][j]), -4.0); + + // Ww[2][j] = -4*w[0][j] + 4*w[1][j] + -4*w[2][j]; + Ww[2][j] = vmulq_n_f32(vsubq_f32(vsubq_f32(w[1][j], w[0][j]), w[2][j]), 4.0); + + // Ww[3][j] = 1*w[0][j] + 2*w[1][j] + 4*w[2][j]; + Ww[3][j] = vmlaq_n_f32(vmlaq_n_f32(w[0][j], w[1][j], 2.0f), w[2][j], 4.0f); + + // Ww[4][j] = 1*w[0][j] + -2*w[1][j] + 4*w[2][j]; + Ww[4][j] = vmlaq_n_f32(vmlsq_n_f32(w[0][j], w[1][j], 2.0f), w[2][j], 4.0f); + + // Ww[5][j] = 24*w[2][j]; + Ww[5][j] = vmulq_n_f32(w[2][j], 24.0f); + } + + // Compute V = W w WT + for (int i = 0; i < 6; i++) + { + const float recip576 = 1.0f / 576.0f; + + // V[i][0] = 6*Ww[i][0]; + V[i][0] = vmulq_n_f32(vmulq_n_f32(Ww[i][0], 6.0), recip576); + + // V[i][1] = -4*Ww[i][0] + -4*Ww[i][1] + -4*Ww[i][2]; + V[i][1] = vmulq_n_f32(vmulq_n_f32(vaddq_f32(vaddq_f32(Ww[i][0], Ww[i][1]), Ww[i][2]), -4.0), recip576); + + // V[i][2] = -4*Ww[i][0] + 4*Ww[i][1] + -4*Ww[i][2]; + V[i][2] = vmulq_n_f32(vmulq_n_f32(vsubq_f32(vsubq_f32(Ww[i][1], Ww[i][0]), Ww[i][2]), 4.0), recip576); + + // V[i][3] = 1*Ww[i][0] + 2*Ww[i][1] + 4*Ww[i][2]; + V[i][3] = vmulq_n_f32(vmlaq_n_f32(vmlaq_n_f32(Ww[i][0], Ww[i][1], 2.0f), Ww[i][2], 4.0f), recip576); + + // V[i][4] = 1*Ww[i][0] + -2*Ww[i][1] + 4*Ww[i][2]; + V[i][4] = vmulq_n_f32(vmlaq_n_f32(vmlsq_n_f32(Ww[i][0], Ww[i][1], 2.0f), Ww[i][2], 4.0f), recip576); + + // V[i][5] = 24*Ww[i][2]; + V[i][5] = vmulq_n_f32(vmulq_n_f32(Ww[i][2], 24.0f), recip576); + } + + // Store the transformed weights + for (int i = 0, m = 0; i < 6; i++) + { + for (int j = 0; j < 6; j++, m++) + { + vst1q_f32(outptr + m*matrix_stride, V[i][j]); + } + } + outptr += 4; + } +#endif // __aarch64__ +#ifdef __arm_any__ + for (; channels_remaining >= 2; channels_remaining -= 2) + { + // Matrices used and computed in this kernel + float32x2_t w[3][3], Ww[6][3], V[6][6]; + + // Read weights + for (int i = 0; i < 3; i++) + { + for (int j = 0; j < 3; j++) + { + w[i][j] = vld1_f32(inptrs[i][j]); + inptrs[i][j] += 2; + } + } + + // Compute the matrix W w + for (int j = 0; j < 3; j++) + { + // Ww[0][j] = 6*w[0][j]; + Ww[0][j] = vmul_n_f32(w[0][j], 6.0); + + // Ww[1][j] = -4*w[0][j] + -4*w[1][j] + -4*w[2][j]; + Ww[1][j] = vmul_n_f32(vadd_f32(vadd_f32(w[0][j], w[1][j]), w[2][j]), -4.0); + + // Ww[2][j] = -4*w[0][j] + 4*w[1][j] + -4*w[2][j]; + Ww[2][j] = vmul_n_f32(vsub_f32(vsub_f32(w[1][j], w[0][j]), w[2][j]), 4.0); + + // Ww[3][j] = 1*w[0][j] + 2*w[1][j] + 4*w[2][j]; + Ww[3][j] = vmla_n_f32(vmla_n_f32(w[0][j], w[1][j], 2.0f), w[2][j], 4.0f); + + // Ww[4][j] = 1*w[0][j] + -2*w[1][j] + 4*w[2][j]; + Ww[4][j] = vmla_n_f32(vmls_n_f32(w[0][j], w[1][j], 2.0f), w[2][j], 4.0f); + + // Ww[5][j] = 24*w[2][j]; + Ww[5][j] = vmul_n_f32(w[2][j], 24.0f); + } + + // Compute V = W w WT + for (int i = 0; i < 6; i++) + { + const float recip576 = 1.0f / 576.0f; + + // V[i][0] = 6*Ww[i][0]; + V[i][0] = vmul_n_f32(vmul_n_f32(Ww[i][0], 6.0), recip576); + + // V[i][1] = -4*Ww[i][0] + -4*Ww[i][1] + -4*Ww[i][2]; + V[i][1] = vmul_n_f32(vmul_n_f32(vadd_f32(vadd_f32(Ww[i][0], Ww[i][1]), Ww[i][2]), -4.0), recip576); + + // V[i][2] = -4*Ww[i][0] + 4*Ww[i][1] + -4*Ww[i][2]; + V[i][2] = vmul_n_f32(vmul_n_f32(vsub_f32(vsub_f32(Ww[i][1], Ww[i][0]), Ww[i][2]), 4.0), recip576); + + // V[i][3] = 1*Ww[i][0] + 2*Ww[i][1] + 4*Ww[i][2]; + V[i][3] = vmul_n_f32(vmla_n_f32(vmla_n_f32(Ww[i][0], Ww[i][1], 2.0f), Ww[i][2], 4.0f), recip576); + + // V[i][4] = 1*Ww[i][0] + -2*Ww[i][1] + 4*Ww[i][2]; + V[i][4] = vmul_n_f32(vmla_n_f32(vmls_n_f32(Ww[i][0], Ww[i][1], 2.0f), Ww[i][2], 4.0f), recip576); + + // V[i][5] = 24*Ww[i][2]; + V[i][5] = vmul_n_f32(vmul_n_f32(Ww[i][2], 24.0f), recip576); + } + + // Store the transformed weights + for (int i = 0, m = 0; i < 6; i++) + { + for (int j = 0; j < 6; j++, m++) + { + vst1_f32(outptr + m*matrix_stride, V[i][j]); + } + } + outptr += 2; + } +#endif // __arm_any__ + for (; channels_remaining; channels_remaining--) + { + // Matrices used and computed in this kernel + float w[3][3], Ww[6][3], V[6][6]; + + // Read weights + for (int i = 0; i < 3; i++) + { + for (int j = 0; j < 3; j++) + { + w[i][j] = *(inptrs[i][j]++); + } + } + + // Compute the matrix W w + for (int j = 0; j < 3; j++) + { + Ww[0][j] = 6*w[0][j]; + Ww[1][j] = -4*w[0][j] + -4*w[1][j] + -4*w[2][j]; + Ww[2][j] = -4*w[0][j] + 4*w[1][j] + -4*w[2][j]; + Ww[3][j] = 1*w[0][j] + 2*w[1][j] + 4*w[2][j]; + Ww[4][j] = 1*w[0][j] + -2*w[1][j] + 4*w[2][j]; + Ww[5][j] = 24*w[2][j]; + } + + // Compute V = W w WT + for (int i = 0; i < 6; i++) + { + V[i][0] = ( 6*Ww[i][0]) / 576.0; + V[i][1] = (-4*Ww[i][0] + -4*Ww[i][1] + -4*Ww[i][2]) / 576.0; + V[i][2] = (-4*Ww[i][0] + 4*Ww[i][1] + -4*Ww[i][2]) / 576.0; + V[i][3] = ( 1*Ww[i][0] + 2*Ww[i][1] + 4*Ww[i][2]) / 576.0; + V[i][4] = ( 1*Ww[i][0] + -2*Ww[i][1] + 4*Ww[i][2]) / 576.0; + V[i][5] = (24*Ww[i][2]) / 576.0; + } + + // Store the transformed weights + for (int i = 0, m = 0; i < 6; i++) + { + for (int j = 0; j < 6; j++, m++) + { + *(outptr + m*matrix_stride) = V[i][j]; + } + } + outptr++; + } + } +} + +template class WeightTransform<3, 3, 6, 6, float, float, WinogradRoots::Integers>; + +} // namespace diff --git a/src/core/NEON/kernels/convolution/winograd/winograd_transforms/weights_6_3_fp32_fp32_integers.cpp b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/weights_6_3_fp32_fp32_integers.cpp new file mode 100644 index 0000000000..45723482a2 --- /dev/null +++ b/src/core/NEON/kernels/convolution/winograd/winograd_transforms/weights_6_3_fp32_fp32_integers.cpp @@ -0,0 +1,90 @@ +/* + * Copyright (c) 2019 ARM Limited. + * + * SPDX-License-Identifier: MIT + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "arm.hpp" +#include "kernel.hpp" + +namespace winograd +{ + +template <> +void WeightTransform<1, 3, 1, 8, float, float, WinogradRoots::Integers>::execute( + const int n_output_channels, + const int n_input_channels, + const float* const input, // NOTE: Data in HWIO order + float* const output, + const int matrix_stride, + const int matrix_row_stride +) +{ + // Get pointers to each cell of the weight tensor + const auto weight_col_stride = n_input_channels * n_output_channels; + const float *inptrs[3]; + for (int j = 0; j < 3; j++) + { + inptrs[j] = input + j*weight_col_stride; + } + + // For each input channel + for (int ic = 0; ic < n_input_channels; ic++) + { + float *outptr = output + ic * matrix_row_stride; + + // For each output channel + int channels_remaining = n_output_channels; + for (; channels_remaining; channels_remaining--) + { + // Matrices used and computed in this kernel + float w[3], V[inner_tile_cols]; + + // Read weights + for (int j = 0; j < 3; j++) + { + w[j] = *(inptrs[j]++); + } + + // Compute V = w WT + V[0] = (w[0]*-1) / 36.0f; + V[1] = (w[1]*-1 + w[0]*1 + w[2]*1) / 48.0f; + V[2] = (w[0]*1 + w[1]*1 + w[2]*1) / 48.0f; + V[3] = (w[0]*-1 + w[2]*-4 + w[1]*2) / 120.0f; + V[4] = (w[0]*-1 + w[2]*-4 + w[1]*-2) / 120.0f; + V[5] = (w[1]*-3 + w[2]*9 + w[0]*1) / 720.0f; + V[6] = (w[1]*3 + w[2]*9 + w[0]*1) / 720.0f; + V[7] = (w[2]*1) / 1; + + // Store the transformed weights + for (int j = 0; j < inner_tile_cols; j++) + { + *(outptr + j*matrix_stride) = V[j]; + } + outptr++; + } + } +} + +template class WeightTransform<1, 3, 1, 8, float, float, WinogradRoots::Integers>; +template class WeightTransform<3, 1, 8, 1, float, float, WinogradRoots::Integers>; + +} // namespace diff --git a/src/runtime/NEON/functions/NEWinogradConvolutionLayer.cpp b/src/runtime/NEON/functions/NEWinogradConvolutionLayer.cpp index 1d92471162..b4247be6cb 100644 --- a/src/runtime/NEON/functions/NEWinogradConvolutionLayer.cpp +++ b/src/runtime/NEON/functions/NEWinogradConvolutionLayer.cpp @@ -33,7 +33,7 @@ #include "arm_compute/runtime/NEON/functions/NEGEMMAssemblyDispatch.h" #include "support/ToolchainSupport.h" -#include "arm_compute/core/NEON/kernels/convolution/winograd/winograd_gemm.hpp" +#include "arm_compute/core/NEON/kernels/convolution/winograd/winograd.hpp" namespace arm_compute { @@ -236,10 +236,10 @@ bool check_support_fast_math(const Size2D &output_tile, const Size2D &kernel_siz NEWinogradConvolutionLayer::NEWinogradConvolutionLayer(std::shared_ptr memory_manager) : _memory_group(memory_manager), _gemm_function(memory_manager), _transform_input_kernel(nullptr), _transform_output_kernel(nullptr), _transform_weights_kernel(nullptr), _activationlayer_function(), - _permute_input(), _permute_weights(), _permute_output(), _input_workspace(), _output_workspace(), _kernel_storage(), _input_nhwc(), _output_nhwc(), _weights_hwio(), _input(), _weights(), _output(), - _is_prepared(false), _is_activationlayer_enabled(false) + _permute_input(), _permute_weights(), _permute_output(), _input_transformed(), _output_transformed(), _input_workspace(), _output_workspace(), _kernel_storage(), _input_nhwc(), _output_nhwc(), + _weights_hwio(), _input(), _weights(), _output(), _is_prepared(false), _is_activationlayer_enabled(false) { -} /* arm_compute */ +} void NEWinogradConvolutionLayer::configure(const ITensor *input, const ITensor *weights, const ITensor *biases, ITensor *output, const PadStrideInfo &conv_info, const ActivationLayerInfo &act_info, bool enable_fast_math) @@ -436,9 +436,9 @@ void NEWinogradConvolutionLayer::configure(const ITensor *input, const ITensor * b_info.init(b_shape, 1, data_type, b_strides, 0, kernel_storage_size); d_info.init(d_shape, 1, data_type, d_strides, 0, output_storage_size); - _input_workspace.allocator()->init(a_info, storage_alignment); + _input_transformed.allocator()->init(a_info, storage_alignment); _kernel_storage.allocator()->init(b_info, storage_alignment); - _output_workspace.allocator()->init(d_info, storage_alignment); + _output_transformed.allocator()->init(d_info, storage_alignment); // configure and allocate dst tensor to be used to convert from winograd domain to spatial domain when calling to reshape_output() TensorInfo info(TensorShape(_output->info()->dimension(2), _output->info()->dimension(0), @@ -447,6 +447,8 @@ void NEWinogradConvolutionLayer::configure(const ITensor *input, const ITensor * _output_nhwc.allocator()->init(info); // Configure the InputTransform + _memory_group.manage(&_input_transformed); + _memory_group.manage(&_output_transformed); _memory_group.manage(&_input_workspace); _memory_group.manage(&_output_workspace); @@ -456,7 +458,7 @@ void NEWinogradConvolutionLayer::configure(const ITensor *input, const ITensor * _permute_input.configure(input, &_input_nhwc, PermutationVector(2U, 0U, 1U)); _input_nhwc.allocator()->allocate(); transform_input_kernel->configure(&_input_nhwc, in_shape.n_batches, in_shape.n_rows, in_shape.n_cols, in_shape.n_channels, use_padding_type, - &_input_workspace, input_matrix_stride); + &_input_transformed, input_matrix_stride, &_input_workspace); // Re-order a weight tensor from [Output feature map x Input feature map x Height x Width] to [Height x Width x Input feature map x Output feature map] _permute_weights.configure(weights, &_weights_hwio, PermutationVector(3U, 2U, 0U, 1U)); @@ -465,26 +467,39 @@ void NEWinogradConvolutionLayer::configure(const ITensor *input, const ITensor * //The biases tensor has not been allocated at this point in time, the output transform will add the biases to the final result in the run() method _memory_group.manage(&_output_nhwc); - transform_output_kernel->configure(biases, &_output_workspace, + transform_output_kernel->configure(biases, &_output_transformed, output_matrix_stride, &_output_nhwc, - in_shape.n_batches, output_shape.n_rows, output_shape.n_cols, out_channels); + in_shape.n_batches, output_shape.n_rows, output_shape.n_cols, out_channels, &_output_workspace); } else { transform_input_kernel->configure(_input, in_shape.n_batches, in_shape.n_rows, in_shape.n_cols, in_shape.n_channels, use_padding_type, - &_input_workspace, input_matrix_stride); + &_input_transformed, input_matrix_stride, &_input_workspace); // Re-order a weight tensor from [Output feature map x Input feature map x Height x Width] to [Height x Width x Input feature map x Output feature map] _permute_weights.configure(weights, &_weights_hwio, PermutationVector(3U, 0U, 1U, 2U)); transform_weights_kernel->configure(&_weights_hwio, &_kernel_storage, kernel_matrix_stride, out_channels, in_channels); - transform_output_kernel->configure(biases, &_output_workspace, + transform_output_kernel->configure(biases, &_output_transformed, output_matrix_stride, _output, - in_shape.n_batches, output_shape.n_rows, output_shape.n_cols, out_channels); + in_shape.n_batches, output_shape.n_rows, output_shape.n_cols, out_channels, &_output_workspace); } - _gemm_function.configure(&_input_workspace, &_kernel_storage, nullptr, &_output_workspace, 1.0f, 0.f); + //Configure input/output workspaces, get_working_space_size() must be called after configure() + const unsigned int max_num_threads = NEScheduler::get().num_threads_hint(); + const size_t input_workspace_size = transform_input_kernel->get_working_space_size(max_num_threads); + const size_t output_workspace_size = transform_output_kernel->get_working_space_size(max_num_threads); + + TensorInfo input_workspace_info(TensorShape(input_workspace_size), 1, _input->info()->data_type()); + _input_workspace.allocator()->init(input_workspace_info); + + TensorInfo output_workspace_info(TensorShape(output_workspace_size), 1, _output->info()->data_type()); + _output_workspace.allocator()->init(output_workspace_info); + + _gemm_function.configure(&_input_transformed, &_kernel_storage, nullptr, &_output_transformed, 1.0f, 0.f); + _input_transformed.allocator()->allocate(); + _output_transformed.allocator()->allocate(); _input_workspace.allocator()->allocate(); _output_workspace.allocator()->allocate(); -- cgit v1.2.1