| /* |
| * 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/GLES_COMPUTE/kernels/GCPixelWiseMultiplicationKernel.h" |
| |
| #include "arm_compute/core/Error.h" |
| #include "arm_compute/core/GLES_COMPUTE/GCHelpers.h" |
| #include "arm_compute/core/GLES_COMPUTE/GCKernelLibrary.h" |
| #include "arm_compute/core/GLES_COMPUTE/IGCTensor.h" |
| #include "arm_compute/core/GLES_COMPUTE/OpenGLES.h" |
| #include "arm_compute/core/Helpers.h" |
| #include "arm_compute/core/TensorInfo.h" |
| #include "arm_compute/core/Validate.h" |
| #include "arm_compute/core/Window.h" |
| |
| #include <cmath> |
| #include <cstdlib> |
| #include <set> |
| #include <string> |
| using namespace arm_compute; |
| |
| GCPixelWiseMultiplicationKernel::GCPixelWiseMultiplicationKernel() |
| : _input1(nullptr), _input2(nullptr), _output(nullptr) |
| { |
| } |
| |
| void GCPixelWiseMultiplicationKernel::configure(const IGCTensor *input1, const IGCTensor *input2, IGCTensor *output, float scale) |
| { |
| ARM_COMPUTE_ERROR_ON_NULLPTR(input1, input2, output); |
| ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(input1, 1, DataType::F32); |
| ARM_COMPUTE_ERROR_ON_MISMATCHING_DATA_TYPES(input1, input2); |
| ARM_COMPUTE_ERROR_ON_NULLPTR(output); |
| ARM_COMPUTE_ERROR_ON_MSG(scale < 0, "Scale cannot be negative. "); |
| |
| // Auto initialize output if not initialized |
| { |
| set_shape_if_empty(*output->info(), input1->info()->tensor_shape()); |
| set_format_if_unknown(*output->info(), Format::F32); |
| } |
| |
| ARM_COMPUTE_ERROR_ON_MISMATCHING_SHAPES(input1, input2, output); |
| ARM_COMPUTE_ERROR_ON_MISMATCHING_DATA_TYPES(input1, input2, output); |
| ARM_COMPUTE_ERROR_ON_MSG(scale < 0, "Scale cannot be negative. "); |
| |
| _input1 = input1; |
| _input2 = input2; |
| _output = output; |
| |
| std::string data_type; |
| std::string compute_type; |
| |
| // Set kernel build options |
| std::set<std::string> build_opts; |
| build_opts.emplace("#define LOCAL_SIZE_X " + support::cpp11::to_string(1)); |
| build_opts.emplace("#define LOCAL_SIZE_Y " + support::cpp11::to_string(1)); |
| build_opts.emplace("#define LOCAL_SIZE_Z " + support::cpp11::to_string(1)); |
| |
| build_opts.emplace("#define SCALE " + support::cpp11::to_string(scale)); |
| |
| // Create kernel |
| _kernel = static_cast<GCKernel>(GCKernelLibrary::get().create_kernel("pixelwise_mul_float", build_opts)); |
| |
| // Configure kernel window |
| constexpr unsigned int num_elems_processed_per_iteration = 1; |
| |
| Window win = calculate_max_window(*input1->info(), Steps(num_elems_processed_per_iteration)); |
| |
| AccessWindowHorizontal input1_access(input1->info(), 0, num_elems_processed_per_iteration); |
| AccessWindowHorizontal input2_access(input2->info(), 0, num_elems_processed_per_iteration); |
| AccessWindowHorizontal output_access(output->info(), 0, num_elems_processed_per_iteration); |
| |
| update_window_and_padding(win, input1_access, input2_access, output_access); |
| |
| ValidRegion valid_region = intersect_valid_regions(input1->info()->valid_region(), |
| input2->info()->valid_region()); |
| output_access.set_valid_region(win, valid_region); |
| |
| IGCKernel::configure(win); |
| } |
| |
| void GCPixelWiseMultiplicationKernel::run(const Window &window) |
| { |
| ARM_COMPUTE_ERROR_ON_UNCONFIGURED_KERNEL(this); |
| ARM_COMPUTE_ERROR_ON_INVALID_SUBWINDOW(IGCKernel::window(), window); |
| |
| _kernel.use(); |
| |
| Window slice = window.first_slice_window_3D(); |
| |
| do |
| { |
| unsigned int idx = 0; |
| unsigned int binding = 1; |
| add_3D_tensor_argument(idx, _input1, binding++, slice); |
| add_3D_tensor_argument(idx, _input2, binding++, slice); |
| add_3D_tensor_argument(idx, _output, binding++, slice); |
| |
| _kernel.update_shader_params(); |
| enqueue(*this, slice); |
| } |
| while(window.slide_window_slice_3D(slice)); |
| } |