/* * Copyright (c) 2017-2021 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 "src/gpu/cl/kernels/ClIm2ColKernel.h" #include "arm_compute/core/CL/CLHelpers.h" #include "arm_compute/core/CL/CLKernelLibrary.h" #include "arm_compute/core/CL/ICLTensor.h" #include "arm_compute/core/CL/OpenCL.h" #include "arm_compute/core/Helpers.h" #include "arm_compute/core/TensorInfo.h" #include "arm_compute/core/Validate.h" #include "arm_compute/core/utils/misc/ShapeCalculator.h" #include "src/core/AccessWindowStatic.h" #include "src/core/CL/CLValidate.h" #include "src/core/helpers/AutoConfiguration.h" #include "src/core/helpers/WindowHelpers.h" #include "support/Cast.h" #include "support/StringSupport.h" #include #include #include namespace arm_compute { using namespace misc::shape_calculator; namespace opencl { namespace kernels { namespace { struct Im2ColConfiguration { std::string kernel_name{}; std::set build_options{}; unsigned int num_elems_processed_per_iteration{}; bool is_padding_required_nchw{}; }; Status validate_arguments(const ITensorInfo *src, const ITensorInfo *dst, const Size2D &kernel_dims, const PadStrideInfo &conv_info, bool has_bias, const Size2D &dilation, unsigned int num_groups) { const unsigned int channel_idx = get_data_layout_dimension_index(src->data_layout(), DataLayoutDimension::CHANNEL); ARM_COMPUTE_RETURN_ERROR_ON_F16_UNSUPPORTED(src); ARM_COMPUTE_RETURN_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(src, 1, DataType::QASYMM8, DataType::QASYMM8_SIGNED, DataType::F16, DataType::F32); ARM_COMPUTE_RETURN_ERROR_ON(is_data_type_quantized(src->data_type()) && has_bias); ARM_COMPUTE_RETURN_ERROR_ON_NULLPTR(dst); ARM_COMPUTE_RETURN_ERROR_ON((dilation.x() < 1) || (dilation.y() < 1)); ARM_COMPUTE_RETURN_ERROR_ON(src->data_layout() == DataLayout::UNKNOWN); ARM_COMPUTE_RETURN_ERROR_ON(num_groups == 0); ARM_COMPUTE_RETURN_ERROR_ON(src->data_layout() == DataLayout::NHWC && num_groups > 1); ARM_COMPUTE_RETURN_ERROR_ON((src->dimension(channel_idx) % num_groups) != 0); // Since there's no implicit padding added, check the total input spatial dimensions (with conv paddings) are big enough for the kernel dimensions const unsigned int width_idx = get_data_layout_dimension_index(src->data_layout(), DataLayoutDimension::WIDTH); const unsigned int height_idx = get_data_layout_dimension_index(src->data_layout(), DataLayoutDimension::HEIGHT); const unsigned total_width = src->dimension(width_idx) + conv_info.pad_left() + conv_info.pad_right(); const unsigned total_height = src->dimension(height_idx) + conv_info.pad_top() + conv_info.pad_bottom(); ARM_COMPUTE_RETURN_ERROR_ON((total_width < kernel_dims.width) || (total_height < kernel_dims.height)); if(dst->total_size() > 0) { const TensorInfo tensor_info_output = dst->clone()->set_tensor_shape(compute_im2col_conv_shape(src, kernel_dims, conv_info, has_bias, dilation, num_groups == 1, num_groups)); ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_SHAPES(dst, &tensor_info_output); ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_TYPES(src, dst); ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_QUANTIZATION_INFO(src, dst); } return Status{}; } std::pair validate_and_configure_window(ITensorInfo *src, ITensorInfo *dst, const Size2D &kernel_dims, const PadStrideInfo &conv_info, bool has_bias, const Size2D &dilation, unsigned int num_elems_processed_per_iteration, bool is_padding_required_nchw, unsigned int num_groups) { ARM_COMPUTE_ERROR_ON_NULLPTR(src, dst); // Output tensor auto initialization if not yet initialized TensorShape expected_output_shape = compute_im2col_conv_shape(src, kernel_dims, conv_info, has_bias, dilation, num_groups == 1, num_groups); auto_init_if_empty(*dst, src->clone()->set_tensor_shape(expected_output_shape)); const DataLayout data_layout = src->data_layout(); const unsigned int width_idx = get_data_layout_dimension_index(data_layout, DataLayoutDimension::WIDTH); const unsigned int height_idx = get_data_layout_dimension_index(data_layout, DataLayoutDimension::HEIGHT); const unsigned int input_width = src->dimension(width_idx); const unsigned int input_height = src->dimension(height_idx); // Configure the execute window based on the selected optimal OpenCL kernel bool window_changed = false; Window win; if(data_layout == DataLayout::NHWC) { win = calculate_max_window(*src, Steps(num_elems_processed_per_iteration)); } else { if(is_padding_required_nchw) { const BorderSize border(conv_info.pad_top(), conv_info.pad_right(), conv_info.pad_bottom(), conv_info.pad_left()); win = calculate_max_window(*src, Steps(num_elems_processed_per_iteration * conv_info.stride().first, conv_info.stride().second)); AccessWindowStatic input_access(src, -border.left, -border.top, ceil_to_multiple(input_width + border.right, kernel_dims.width * num_elems_processed_per_iteration), input_height + border.bottom); window_changed = window_changed || update_window_and_padding(win, input_access); } else { // For the generic case, CLIm2ColKernel doesn't need padding (we do not read out-of-bounds elements) so // update_window_and_padding() can be skipped win = calculate_max_window(*src, Steps()); } } // set the Z dimension's step same size as the whole dimension so that one can't split across the Z dimension win.set_dimension_step(Window::DimZ, win[Window::DimZ].end() - win[Window::DimZ].start()); Status err = (window_changed) ? ARM_COMPUTE_CREATE_ERROR(ErrorCode::RUNTIME_ERROR, "Insufficient Padding!") : Status{}; return std::make_pair(err, win); } Im2ColConfiguration configure_opencl_kernel(const ITensorInfo *src, const Size2D &kernel_dims, const PadStrideInfo &conv_info, bool has_bias, const Size2D &dilation, unsigned int num_groups) { const DataLayout data_layout = src->data_layout(); const DataType data_type = src->data_type(); const unsigned int width_idx = get_data_layout_dimension_index(data_layout, DataLayoutDimension::WIDTH); const unsigned int height_idx = get_data_layout_dimension_index(data_layout, DataLayoutDimension::HEIGHT); const unsigned int channel_idx = get_data_layout_dimension_index(data_layout, DataLayoutDimension::CHANNEL); const unsigned int input_width = src->dimension(width_idx); const unsigned int input_height = src->dimension(height_idx); const unsigned int input_channel = src->dimension(channel_idx); const std::pair convolved_dims = scaled_dimensions(input_width, input_height, kernel_dims.width, kernel_dims.height, conv_info, dilation); // Im2Col configuration std::string kernel_name = "im2col_generic_"; CLBuildOptions build_opts; unsigned int num_elems_processed_per_iteration = 1; bool is_padding_required_nchw = false; const UniformQuantizationInfo qinfo = src->quantization_info().uniform(); build_opts.add_option("-DDATA_TYPE=" + get_cl_type_from_data_type(data_type)); build_opts.add_option("-DELEMENT_SIZE=" + support::cpp11::to_string(src->element_size())); build_opts.add_option("-DKERNEL_WIDTH=" + support::cpp11::to_string(kernel_dims.width)); build_opts.add_option("-DKERNEL_HEIGHT=" + support::cpp11::to_string(kernel_dims.height)); build_opts.add_option("-DCONVOLVED_WIDTH=" + support::cpp11::to_string(convolved_dims.first)); build_opts.add_option("-DCONVOLVED_HEIGHT=" + support::cpp11::to_string(convolved_dims.second)); build_opts.add_option("-DSTRIDE_X=" + support::cpp11::to_string(conv_info.stride().first)); build_opts.add_option("-DSTRIDE_Y=" + support::cpp11::to_string(conv_info.stride().second)); build_opts.add_option("-DPAD_LEFT=" + support::cpp11::to_string(conv_info.pad_left())); build_opts.add_option("-DPAD_TOP=" + support::cpp11::to_string(conv_info.pad_top())); build_opts.add_option("-DPAD_RIGHT=" + support::cpp11::to_string(conv_info.pad_right())); build_opts.add_option("-DPAD_BOTTOM=" + support::cpp11::to_string(conv_info.pad_bottom())); build_opts.add_option("-DSRC_WIDTH=" + support::cpp11::to_string(input_width)); build_opts.add_option("-DSRC_HEIGHT=" + support::cpp11::to_string(input_height)); build_opts.add_option("-DSRC_DEPTH=" + support::cpp11::to_string(input_channel)); build_opts.add_option("-DDILATION_X=" + support::cpp11::to_string(dilation.x())); build_opts.add_option("-DDILATION_Y=" + support::cpp11::to_string(dilation.y())); build_opts.add_option_if(num_groups > 1, "-DNUM_GROUPS=" + support::cpp11::to_string(num_groups)); build_opts.add_option_if_else(is_data_type_quantized(data_type), "-DPAD_VALUE=" + support::cpp11::to_string(qinfo.offset), "-DPAD_VALUE=0"); build_opts.add_option_if(has_bias, "-DHAS_BIAS"); if(data_layout == DataLayout::NHWC) { num_elems_processed_per_iteration = std::min(2U, input_channel); is_padding_required_nchw = false; // Only the 3x3 and 9x9 cases are optimized for NHWC if(kernel_dims == Size2D(3U, 3U)) { kernel_name = "im2col3x3_"; build_opts.add_option("-DIM2COL_3X3"); } else if(kernel_dims == Size2D(9U, 9U)) { kernel_name = "im2col9x9_"; build_opts.add_option("-DIM2COL_9X9"); } else { build_opts.add_option("-DIM2COL_GENERIC"); } // Get boundary vector (the first/last vector with potentially a partial vector size) size // If input_channel is a multiple of num_elems_processed_per_iteration, the boundary vec size is the (full) vector size // otherwise, the boundary vec size is the (partial) remainder vector size const unsigned int vec_size = num_elems_processed_per_iteration; const unsigned int partial_vec_size = input_channel % vec_size; const unsigned int boundary_vec_size = vec_size - ((vec_size - partial_vec_size) % vec_size); build_opts.add_option("-DVECTOR_SIZE=" + support::cpp11::to_string(vec_size)); build_opts.add_option("-DBOUNDARY_VECTOR_SIZE=" + support::cpp11::to_string(boundary_vec_size)); } else { if(dilation == Size2D(1U, 1U)) { const bool squared_im2col = kernel_dims.width == kernel_dims.height; if(squared_im2col) { // Check if we can run an optimized im2col for NCHW switch(kernel_dims.width) { case 1: // Optimized im2col1x1 if stride_x = 1 and conv_info.has_padding() = false if(conv_info.stride().first == 1 && !conv_info.has_padding()) { kernel_name = "im2col1x1_stridex1_"; num_elems_processed_per_iteration = 4; is_padding_required_nchw = true; } break; case 3: kernel_name = "im2col3x3_"; num_elems_processed_per_iteration = 1; is_padding_required_nchw = true; break; case 5: kernel_name = "im2col5x5_"; num_elems_processed_per_iteration = 1; is_padding_required_nchw = true; break; case 11: // Optimized im2col11x11 if pad_x = pad_y = 0 if(!conv_info.has_padding()) { kernel_name = "im2col11x11_padx0_pady0_"; num_elems_processed_per_iteration = 1; is_padding_required_nchw = true; } break; default: kernel_name = "im2col_generic_"; num_elems_processed_per_iteration = 1; is_padding_required_nchw = false; break; } } else if(kernel_dims.width > 1 && !conv_info.has_padding()) { kernel_name = "im2col_generic_padx0_pady0_"; num_elems_processed_per_iteration = 1; is_padding_required_nchw = false; // Optimized im2col is performed using one or more vector operations with the specified vector size // and a remainder. For example, for 5x5 convolutions, im2col is performed using vectors of size 4 // and scalars; for 7x7 convolutions, using vectors of size 4 and vectors of size 3. // Using the vector size of 4 is always safe since OpenCL supports vectors of size 2 and 3. // Using the vector size of 8, however, may be faster. // For 2x2 convolutions, use vectors of size 2. (For 3x3 convolutions, im2col_kernel3x3_padx0_pady0 // is used instead.) const size_t vector_size = std::min(static_cast(4), kernel_dims.width); const size_t width_mod_vector_size = kernel_dims.width % vector_size; build_opts.add_option("-DVECTOR_SIZE=" + support::cpp11::to_string(vector_size)); build_opts.add_option("-DWIDTH_MOD_VECTOR_SIZE=" + support::cpp11::to_string(width_mod_vector_size)); } } } // Append the data layout to the kernel_name kernel_name += lower_string(string_from_data_layout(data_layout)); Im2ColConfiguration im2col_config; im2col_config.kernel_name = kernel_name; im2col_config.build_options = build_opts.options(); im2col_config.num_elems_processed_per_iteration = num_elems_processed_per_iteration; im2col_config.is_padding_required_nchw = is_padding_required_nchw; return im2col_config; } } // namespace ClIm2ColKernel::ClIm2ColKernel() : _data_layout(DataLayout::UNKNOWN), _convolved_dims(), _num_elems_processed_per_iteration(1), _kernel_dims(), _conv_info(), _num_groups() { _type = CLKernelType::ELEMENTWISE; } void ClIm2ColKernel::configure(const ClCompileContext &compile_context, ITensorInfo *src, ITensorInfo *dst, const Size2D &kernel_dims, const PadStrideInfo &conv_info, bool has_bias, const Size2D &dilation, unsigned int num_groups) { ARM_COMPUTE_ERROR_ON_NULLPTR(src, dst); ARM_COMPUTE_ERROR_THROW_ON(validate_arguments(src, dst, kernel_dims, conv_info, has_bias, dilation, num_groups)); auto padding_info = get_padding_info({ src, dst }); _data_layout = src->data_layout(); const unsigned int width_idx = get_data_layout_dimension_index(_data_layout, DataLayoutDimension::WIDTH); const unsigned int height_idx = get_data_layout_dimension_index(_data_layout, DataLayoutDimension::HEIGHT); const unsigned int input_width = src->dimension(width_idx); const unsigned int input_height = src->dimension(height_idx); // Select and configure the optimal OpenCL kernel to run. // This function returns the OpenCL kernel's name, the arguments to pass at compile time, the number of elements processed per iteration // and the padding requirement flag Im2ColConfiguration im2col_config = configure_opencl_kernel(src, kernel_dims, conv_info, has_bias, dilation, num_groups); // Create kernel _kernel = create_kernel(compile_context, im2col_config.kernel_name, im2col_config.build_options); _convolved_dims = scaled_dimensions(input_width, input_height, kernel_dims.width, kernel_dims.height, conv_info, dilation); _num_elems_processed_per_iteration = im2col_config.num_elems_processed_per_iteration; _kernel_dims = kernel_dims; // Only needed by the Tuner _conv_info = conv_info; // Only needed by the Tuner _num_groups = num_groups; // Configure kernel window auto win_config = validate_and_configure_window(src, dst, kernel_dims, conv_info, has_bias, dilation, im2col_config.num_elems_processed_per_iteration, im2col_config.is_padding_required_nchw, num_groups); ARM_COMPUTE_ERROR_THROW_ON(win_config.first); IClKernel::configure_internal(win_config.second); // Set config_id for enabling LWS tuning _config_id = im2col_config.kernel_name; _config_id += "_"; _config_id += lower_string(string_from_data_type(src->data_type())); _config_id += "_"; _config_id += support::cpp11::to_string(num_groups); _config_id += "_"; _config_id += support::cpp11::to_string(dst->dimension(0)); _config_id += "_"; _config_id += support::cpp11::to_string(dst->dimension(1)); _config_id += "_"; _config_id += lower_string(string_from_data_layout(_data_layout)); ARM_COMPUTE_ERROR_ON(src->data_layout() == DataLayout::NHWC && has_padding_changed(padding_info)); } Status ClIm2ColKernel::validate(const ITensorInfo *src, const ITensorInfo *dst, const Size2D &kernel_dims, const PadStrideInfo &conv_info, bool has_bias, const Size2D &dilation, unsigned int num_groups) { ARM_COMPUTE_RETURN_ON_ERROR(validate_arguments(src, dst, kernel_dims, conv_info, has_bias, dilation, num_groups)); Im2ColConfiguration im2col_config = configure_opencl_kernel(src, kernel_dims, conv_info, has_bias, dilation, num_groups); ARM_COMPUTE_RETURN_ON_ERROR(validate_and_configure_window(src->clone().get(), dst->clone().get(), kernel_dims, conv_info, has_bias, dilation, im2col_config.num_elems_processed_per_iteration, im2col_config.is_padding_required_nchw, num_groups) .first); return Status{}; } void ClIm2ColKernel::run_op(ITensorPack &tensors, const Window &window, cl::CommandQueue &queue) { ARM_COMPUTE_ERROR_ON_UNCONFIGURED_KERNEL(this); ARM_COMPUTE_ERROR_ON_MISMATCHING_WINDOWS(IClKernel::window(), window); ARM_COMPUTE_ERROR_ON(tensors.empty()); // Get initial windows // Collapse in order to have (SRC_DEPTH * BATCH_SIZE) on the 3rd dimension Window window_collapsed = window.collapse_if_possible(ICLKernel::window(), Window::DimZ); window_collapsed.set_dimension_step(Window::DimZ, 1); auto src = utils::cast::polymorphic_downcast(tensors.get_const_tensor(TensorType::ACL_SRC)); auto dst = utils::cast::polymorphic_downcast(tensors.get_tensor(TensorType::ACL_DST)); ARM_COMPUTE_ERROR_ON_NULLPTR(src, dst); Window window_output; window_output.use_tensor_dimensions(dst->info()->tensor_shape()); const Window first_slice_3d = window_collapsed.first_slice_window_3D(); Window slice = first_slice_3d; Window slice_in = first_slice_3d; Window slice_out = window_output.first_slice_window_2D(); if(_data_layout == DataLayout::NHWC) { const Window tmp_win = window.collapse_if_possible(ICLKernel::window(), 3); const int num_batches = tmp_win[3].end(); slice.set(1, Window::Dimension(0, static_cast(dst->info()->tensor_shape()[1]), 1)); slice.set(2, Window::Dimension(0, static_cast(num_batches), 1)); } else { slice.set(0, Window::Dimension(0, static_cast(ceil_to_multiple(_convolved_dims.first, _num_elems_processed_per_iteration)), _num_elems_processed_per_iteration)); slice.set(1, Window::Dimension(0, static_cast(_convolved_dims.second), 1)); // Note: In case of NCHW the 3rd dimension is already set collapsing the input window } // Setup input slice // The dimensions of the input are increased within the OpenCL kernel slice_in.set(Window::DimX, Window::Dimension(0, 0, 0)); slice_in.set(Window::DimY, Window::Dimension(0, 0, 0)); slice_in.set(Window::DimZ, Window::Dimension(0, 0, 0)); // Setup output slice // The dimensions of the output are increased within the OpenCL kernel slice_out.set(Window::DimX, Window::Dimension(0, 0, 0)); slice_out.set(Window::DimY, Window::Dimension(0, 0, 0)); unsigned int idx = num_arguments_per_3D_tensor() + (_num_groups == 1 ? num_arguments_per_2D_tensor() : num_arguments_per_3D_tensor()); _kernel.setArg(idx++, static_cast(src->info()->strides_in_bytes()[3])); _kernel.setArg(idx++, static_cast(dst->info()->strides_in_bytes()[((_num_groups == 1) ? 2 : 3)])); do { unsigned int idx = 0; add_3D_tensor_argument(idx, src, slice_in); if(_num_groups == 1) { add_2D_tensor_argument(idx, dst, slice_out); } else { add_3D_tensor_argument(idx, dst, slice_out); } enqueue(queue, *this, slice, lws_hint()); } while(window_collapsed.slide_window_slice_3D(slice) && window_output.slide_window_slice_2D(slice_out) && window_collapsed.slide_window_slice_3D(slice_in)); } } // namespace kernels } // namespace opencl } // namespace arm_compute