1 /* 2 * Copyright (c) 2022 Arm Limited. 3 * 4 * SPDX-License-Identifier: MIT 5 * 6 * Permission is hereby granted, free of charge, to any person obtaining a copy 7 * of this software and associated documentation files (the "Software"), to 8 * deal in the Software without restriction, including without limitation the 9 * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or 10 * sell copies of the Software, and to permit persons to whom the Software is 11 * furnished to do so, subject to the following conditions: 12 * 13 * The above copyright notice and this permission notice shall be included in all 14 * copies or substantial portions of the Software. 15 * 16 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 17 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 18 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE 19 * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER 20 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, 21 * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE 22 * SOFTWARE. 23 */ 24 #ifndef ARM_COMPUTE_CL_GEMMLOWP_MATRIXMULTIPLY_RESHAPED_ONLY_RHS_MMUL_KERNEL_H 25 #define ARM_COMPUTE_CL_GEMMLOWP_MATRIXMULTIPLY_RESHAPED_ONLY_RHS_MMUL_KERNEL_H 26 27 #include "arm_compute/core/KernelDescriptors.h" 28 #include "src/core/common/Macros.h" 29 #include "src/gpu/cl/IClKernel.h" 30 31 namespace arm_compute 32 { 33 namespace opencl 34 { 35 namespace kernels 36 { 37 /** OpenCL kernel to multiply matrices with QASYMM8/QASYMM8_SIGNED data types when only the input matrix RHS (src1) has been reshaped using the MMUL instruction 38 * 39 * @note The input matrix src1 must be reshaped through @ref opencl::kernels::ClGemmReshapeRhsMatrixKernel 40 * @note For fused output stage, only GEMMLowpOutputStageType::QUANTIZE_DOWN_FIXEDPOINT type is supported 41 */ 42 class ClGemmLowpMatrixMultiplyReshapedOnlyRhsMMULKernel : public IClKernel 43 { 44 public: 45 ClGemmLowpMatrixMultiplyReshapedOnlyRhsMMULKernel(); 46 ARM_COMPUTE_DISALLOW_COPY_ALLOW_MOVE(ClGemmLowpMatrixMultiplyReshapedOnlyRhsMMULKernel); 47 /** Initialise the kernel's source and destination. 48 * 49 * @param[in] compile_context The compile context to be used. 50 * @param[in] src0 Input tensor containing the LHS matrix. Data type supported: QASYMM8/QASYMM8_SIGNED 51 * @param[in] src1 Input tensor containing the RHS reshaped matrix. Data type supported: same as @p src0 52 * @param[out] dst Destination tensor. Data type supported: QASYMM8/QASYMM8_SIGNED/S32. 53 * @param[in] gemm_info GEMM information used to retrieve the original dimensions of the input matrices, output stage information and RHS/LHS info. 54 * lhs_info.m0: 1,2,4 55 * Only the following values are supported for RHS info: 56 * rhs_info.n0: 1,4,8 57 * rhs_info.k0: same as lhs_info.k0: 4 58 * rhs_info.transpose: true 59 * @param[in] vector_sum_col (Optional) Input row-vector of sums of all the entries in each column of matrix B. 60 * Note: vector_sum_col can be a nullptr in case a_offset = 0. Data type supported: S32 61 * @param[in] vector_sum_row (Optional) Input row-vector of sums of all the entries in each row of matrix A. 62 * Note: vector_sum_row can be a nullptr in case b_offset = 0. Data type supported: S32 63 * @param[in] bias (Optional) Biases tensor. Can be a nullptr if the addition of biases is not required. 64 * Biases are 1D tensor with dimensions [OFM] or same dimensionality as dst if gemm_info.broadcast_bias is false. Data type supported: S32. 65 * @param[in] output_multipliers (Optional) Output multipliers tensor. Supported data types: S32. 66 * @param[in] output_shifts (Optional) Output shifts tensor. Supported data types: S32. 67 */ 68 void configure(const CLCompileContext &compile_context, const ITensorInfo *src0, const ITensorInfo *src1, ITensorInfo *dst, const GEMMKernelInfo &gemm_info, 69 ITensorInfo *vector_sum_col = nullptr, const ITensorInfo *vector_sum_row = nullptr, ITensorInfo *bias = nullptr, 70 ITensorInfo *output_multipliers = nullptr, ITensorInfo *output_shifts = nullptr); 71 /** Static function to check if given info will lead to a valid configuration 72 * 73 * Similar to @ref ClGemmLowpMatrixMultiplyReshapedOnlyRhsMMULKernel::configure() 74 * 75 * @return a status 76 */ 77 static Status validate(const ITensorInfo *src0, const ITensorInfo *src1, const ITensorInfo *dst, const GEMMKernelInfo &gemm_info, 78 const ITensorInfo *vector_sum_col = nullptr, const ITensorInfo *vector_sum_row = nullptr, const ITensorInfo *bias = nullptr, 79 const ITensorInfo *output_multipliers = nullptr, const ITensorInfo *output_shifts = nullptr); 80 81 // Inherited methods overridden: 82 void run_op(ITensorPack &tensors, const Window &window, cl::CommandQueue &queue) override; 83 84 private: 85 bool _fuse_output_stage{ false }; 86 signed int _m{ 1 }; 87 signed int _n{ 1 }; 88 signed int _k{ 1 }; 89 }; 90 } // namespace kernels 91 } // namespace opencl 92 } // namespace arm_compute 93 #endif /* ARM_COMPUTE_CL_GEMMLOWP_MATRIXMULTIPLY_RESHAPED_ONLY_RHS_MMULKERNEL_H */