1 /* 2 * Copyright (c) 2018-2020 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_LOAD_H 25 #define ARM_COMPUTE_WRAPPER_LOAD_H 26 27 #include <arm_neon.h> 28 29 namespace arm_compute 30 { 31 namespace wrapper 32 { 33 #define VLOAD_IMPL(stype, vtype, postfix) \ 34 inline vtype vload(const stype *ptr) \ 35 { \ 36 return vld1_##postfix(ptr); \ 37 } 38 39 VLOAD_IMPL(uint8_t, uint8x8_t, u8) 40 VLOAD_IMPL(int8_t, int8x8_t, s8) 41 VLOAD_IMPL(uint16_t, uint16x4_t, u16) 42 VLOAD_IMPL(int16_t, int16x4_t, s16) 43 VLOAD_IMPL(uint32_t, uint32x2_t, u32) 44 VLOAD_IMPL(int32_t, int32x2_t, s32) 45 //VLOAD_IMPL(uint64_t, uint64x1_t, u64) 46 //VLOAD_IMPL(int64_t, int64x1_t, s64) 47 VLOAD_IMPL(float, float32x2_t, f32) 48 #ifdef __ARM_FEATURE_FP16_VECTOR_ARITHMETIC 49 VLOAD_IMPL(float16_t, float16x4_t, f16) 50 #endif // __ARM_FEATURE_FP16_VECTOR_ARITHMETIC 51 52 #define VLOADQ_IMPL(stype, vtype, postfix) \ 53 inline vtype vloadq(const stype *ptr) \ 54 { \ 55 return vld1q_##postfix(ptr); \ 56 } 57 58 VLOADQ_IMPL(uint8_t, uint8x16_t, u8) 59 VLOADQ_IMPL(int8_t, int8x16_t, s8) 60 VLOADQ_IMPL(uint16_t, uint16x8_t, u16) 61 VLOADQ_IMPL(int16_t, int16x8_t, s16) 62 VLOADQ_IMPL(uint32_t, uint32x4_t, u32) 63 VLOADQ_IMPL(int32_t, int32x4_t, s32) 64 //VLOAD_IMPL(uint64_t, uint64x1_t, u64) 65 //VLOAD_IMPL(int64_t, int64x1_t, s64) 66 VLOADQ_IMPL(float, float32x4_t, f32) 67 #ifdef __ARM_FEATURE_FP16_VECTOR_ARITHMETIC 68 VLOADQ_IMPL(float16_t, float16x8_t, f16) 69 #endif // __ARM_FEATURE_FP16_VECTOR_ARITHMETIC 70 #undef VLOAD_IMPL 71 } // namespace wrapper 72 } // namespace arm_compute 73 #endif /* ARM_COMPUTE_WRAPPER_LOAD_H */ 74