/* * Copyright (c) 2017-2018 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. */ #ifdef __arm__ #include #include "../../asmlib.hpp" // Kernel implementation. // // Assume that "Apanel" points to a chunk of A blocks (each size 6xK) in read-order. // Assume that "Bpanel" points to a chunk of B blocks (each size 8xK) in read-order. // Assume that "Cpanel" points to a chunk of C output blocks (each size // 8x6), the chunks being arranged in a row major fashion. // // Note that the intent of this is that either ablocks or bblocks will be 1 // - this construction allows the output loop to proceed in either order. namespace arm_gemm { void a32_sgemm_8x6(const float *Apanel, const float *Bpanel, float *Cpanel, int ablocks, int bblocks, int K) { const float *a_ptr = Apanel; float *c_ptr = Cpanel; for (int yb=0; yb