1 /* 2 * Copyright (c) 2019-2020 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_WRAPPER_DIV_H 25 #define ARM_COMPUTE_WRAPPER_DIV_H 26 27 #include "src/core/NEON/NEMath.h" 28 #include <arm_neon.h> 29 30 namespace arm_compute 31 { 32 namespace wrapper 33 { 34 #ifdef __aarch64__ 35 36 #define VDIV_IMPL(stype, vtype, prefix, postfix) \ 37 inline vtype vdiv(const vtype &a, const vtype &b) \ 38 { \ 39 return prefix##_##postfix(a, b); \ 40 } 41 VDIV_IMPL(float32x2_t, float32x2_t, vdiv, f32) 42 #ifdef __ARM_FEATURE_FP16_VECTOR_ARITHMETIC 43 VDIV_IMPL(float16x4_t, float16x4_t, vdiv, f16) 44 #endif // __ARM_FEATURE_FP16_VECTOR_ARITHMETIC 45 46 VDIV_IMPL(float32x4_t, float32x4_t, vdivq, f32) 47 #ifdef __ARM_FEATURE_FP16_VECTOR_ARITHMETIC 48 VDIV_IMPL(float16x8_t, float16x8_t, vdivq, f16) 49 #endif // __ARM_FEATURE_FP16_VECTOR_ARITHMETIC 50 51 #else // __aarch64__ 52 53 #define VDIV_IMPL(stype, vtype, mul_prefix, inv_prefix, postfix) \ 54 inline vtype vdiv(const vtype &a, const vtype &b) \ 55 { \ 56 return mul_prefix##_##postfix(a, inv_prefix##_##postfix(b)); \ 57 } 58 VDIV_IMPL(float32x2_t, float32x2_t, vmul, vinv, f32) 59 #ifdef __ARM_FEATURE_FP16_VECTOR_ARITHMETIC 60 VDIV_IMPL(float16x4_t, float16x4_t, vmul, vinv, f16) 61 #endif // __ARM_FEATURE_FP16_VECTOR_ARITHMETIC 62 63 VDIV_IMPL(float32x4_t, float32x4_t, vmulq, vinvq, f32) 64 #ifdef __ARM_FEATURE_FP16_VECTOR_ARITHMETIC 65 VDIV_IMPL(float16x8_t, float16x8_t, vmulq, vinvq, f16) 66 #endif // __ARM_FEATURE_FP16_VECTOR_ARITHMETIC 67 68 #endif // __aarch64__ 69 70 #undef VDIV_IMPL 71 } // namespace wrapper 72 } // namespace arm_compute 73 #endif /* ARM_COMPUTE_WRAPPER_DIV_H */ 74