/* 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. * * Copyright: * 2021 Atharva Nimbalkar */ #if !defined(SIMDE_ARM_NEON_SQADD_H) #define SIMDE_ARM_NEON_SQADD_H #include "types.h" #include // Workaround on ARM64 windows due to windows SDK bug // https://developercommunity.visualstudio.com/t/In-arm64_neonh-vsqaddb_u8-vsqaddh_u16/10271747?sort=newest #if (defined _MSC_VER) && (defined SIMDE_ARM_NEON_A64V8_NATIVE) && (_MSC_VER < 1938) #pragma message ("Due to msvc bug, current version of msvc is supported by workaround. Recommend to update msvc") #undef vsqaddb_u8 #define vsqaddb_u8(src1, src2) neon_usqadds8(__uint8ToN8_v(src1), __int8ToN8_v(src2)).n8_u8[0] #undef vsqaddh_u16 #define vsqaddh_u16(src1, src2) neon_usqadds16(__uint16ToN16_v(src1), __int16ToN16_v(src2)).n16_u16[0] #undef vsqadds_u32 #define vsqadds_u32(src1, src2) _CopyUInt32FromFloat(neon_usqadds32(_CopyFloatFromUInt32(src1), _CopyFloatFromInt32(src2))) #undef vsqaddd_u64 #define vsqaddd_u64(src1, src2) neon_usqadds64(__uint64ToN64_v(src1), __int64ToN64_v(src2)).n64_u64[0] #endif HEDLEY_DIAGNOSTIC_PUSH SIMDE_DISABLE_UNWANTED_DIAGNOSTICS SIMDE_BEGIN_DECLS_ SIMDE_FUNCTION_ATTRIBUTES uint8_t simde_vsqaddb_u8(uint8_t a, int8_t b) { #if defined(SIMDE_ARM_NEON_A64V8_NATIVE) #if defined(SIMDE_BUG_CLANG_REV_365298) return vsqaddb_u8(a, HEDLEY_STATIC_CAST(uint8_t, b)); #else return vsqaddb_u8(a, b); #endif #else int16_t r_ = HEDLEY_STATIC_CAST(int16_t, a) + HEDLEY_STATIC_CAST(int16_t, b); return (r_ < 0) ? 0 : ((r_ > UINT8_MAX) ? UINT8_MAX : HEDLEY_STATIC_CAST(uint8_t, r_)); #endif } #if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES) #undef vsqaddb_u8 #define vsqaddb_u8(a, b) simde_vsqaddb_u8((a), (b)) #endif SIMDE_FUNCTION_ATTRIBUTES uint16_t simde_vsqaddh_u16(uint16_t a, int16_t b) { #if defined(SIMDE_ARM_NEON_A64V8_NATIVE) #if defined(SIMDE_BUG_CLANG_REV_365298) return vsqaddh_u16(a, HEDLEY_STATIC_CAST(uint16_t, b)); #else return vsqaddh_u16(a, b); #endif #else int32_t r_ = HEDLEY_STATIC_CAST(int32_t, a) + HEDLEY_STATIC_CAST(int32_t, b); return (r_ < 0) ? 0 : ((r_ > UINT16_MAX) ? UINT16_MAX : HEDLEY_STATIC_CAST(uint16_t, r_)); #endif } #if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES) #undef vsqaddh_u16 #define vsqaddh_u16(a, b) simde_vsqaddh_u16((a), (b)) #endif SIMDE_FUNCTION_ATTRIBUTES uint32_t simde_vsqadds_u32(uint32_t a, int32_t b) { #if defined(SIMDE_ARM_NEON_A64V8_NATIVE) #if defined(SIMDE_BUG_CLANG_REV_365298) return vsqadds_u32(a, HEDLEY_STATIC_CAST(uint32_t, b)); #else return vsqadds_u32(a, b); #endif #else int64_t r_ = HEDLEY_STATIC_CAST(int64_t, a) + HEDLEY_STATIC_CAST(int64_t, b); return (r_ < 0) ? 0 : ((r_ > UINT32_MAX) ? UINT32_MAX : HEDLEY_STATIC_CAST(uint32_t, r_)); #endif } #if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES) #undef vsqadds_u32 #define vsqadds_u32(a, b) simde_vsqadds_u32((a), (b)) #endif SIMDE_FUNCTION_ATTRIBUTES uint64_t simde_vsqaddd_u64(uint64_t a, int64_t b) { #if defined(SIMDE_ARM_NEON_A64V8_NATIVE) #if defined(SIMDE_BUG_CLANG_REV_365298) return vsqaddd_u64(a, HEDLEY_STATIC_CAST(uint64_t, b)); #else return vsqaddd_u64(a, b); #endif #else uint64_t r_; if (b > 0) { uint64_t ub = HEDLEY_STATIC_CAST(uint64_t, b); r_ = ((UINT64_MAX - a) < ub) ? UINT64_MAX : a + ub; } else { uint64_t nb = HEDLEY_STATIC_CAST(uint64_t, -b); r_ = (nb > a) ? 0 : a - nb; } return r_; #endif } #if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES) #undef vsqaddd_u64 #define vsqaddd_u64(a, b) simde_vsqaddd_u64((a), (b)) #endif SIMDE_FUNCTION_ATTRIBUTES simde_uint8x8_t simde_vsqadd_u8(simde_uint8x8_t a, simde_int8x8_t b) { #if defined(SIMDE_ARM_NEON_A64V8_NATIVE) return vsqadd_u8(a, b); #else simde_uint8x8_private r_, a_ = simde_uint8x8_to_private(a); simde_int8x8_private b_ = simde_int8x8_to_private(b); SIMDE_VECTORIZE for (size_t i = 0 ; i < (sizeof(r_.values) / sizeof(r_.values[0])) ; i++) { r_.values[i] = simde_vsqaddb_u8(a_.values[i], b_.values[i]); } return simde_uint8x8_from_private(r_); #endif } #if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES) #undef vsqadd_u8 #define vsqadd_u8(a, b) simde_vsqadd_u8((a), (b)) #endif SIMDE_FUNCTION_ATTRIBUTES simde_uint16x4_t simde_vsqadd_u16(simde_uint16x4_t a, simde_int16x4_t b) { #if defined(SIMDE_ARM_NEON_A64V8_NATIVE) return vsqadd_u16(a, b); #else simde_uint16x4_private r_, a_ = simde_uint16x4_to_private(a); simde_int16x4_private b_ = simde_int16x4_to_private(b); SIMDE_VECTORIZE for (size_t i = 0 ; i < (sizeof(r_.values) / sizeof(r_.values[0])) ; i++) { r_.values[i] = simde_vsqaddh_u16(a_.values[i], b_.values[i]); } return simde_uint16x4_from_private(r_); #endif } #if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES) #undef vsqadd_u16 #define vsqadd_u16(a, b) simde_vsqadd_u16((a), (b)) #endif SIMDE_FUNCTION_ATTRIBUTES simde_uint32x2_t simde_vsqadd_u32(simde_uint32x2_t a, simde_int32x2_t b) { #if defined(SIMDE_ARM_NEON_A64V8_NATIVE) return vsqadd_u32(a, b); #else simde_uint32x2_private r_, a_ = simde_uint32x2_to_private(a); simde_int32x2_private b_ = simde_int32x2_to_private(b); SIMDE_VECTORIZE for (size_t i = 0 ; i < (sizeof(r_.values) / sizeof(r_.values[0])) ; i++) { r_.values[i] = simde_vsqadds_u32(a_.values[i], b_.values[i]); } return simde_uint32x2_from_private(r_); #endif } #if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES) #undef vsqadd_u32 #define vsqadd_u32(a, b) simde_vsqadd_u32((a), (b)) #endif SIMDE_FUNCTION_ATTRIBUTES simde_uint64x1_t simde_vsqadd_u64(simde_uint64x1_t a, simde_int64x1_t b) { #if defined(SIMDE_ARM_NEON_A64V8_NATIVE) return vsqadd_u64(a, b); #else simde_uint64x1_private r_, a_ = simde_uint64x1_to_private(a); simde_int64x1_private b_ = simde_int64x1_to_private(b); SIMDE_VECTORIZE for (size_t i = 0 ; i < (sizeof(r_.values) / sizeof(r_.values[0])) ; i++) { r_.values[i] = simde_vsqaddd_u64(a_.values[i], b_.values[i]); } return simde_uint64x1_from_private(r_); #endif } #if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES) #undef vsqadd_u64 #define vsqadd_u64(a, b) simde_vsqadd_u64((a), (b)) #endif SIMDE_FUNCTION_ATTRIBUTES simde_uint8x16_t simde_vsqaddq_u8(simde_uint8x16_t a, simde_int8x16_t b) { #if defined(SIMDE_ARM_NEON_A64V8_NATIVE) return vsqaddq_u8(a, b); #else simde_uint8x16_private r_, a_ = simde_uint8x16_to_private(a); simde_int8x16_private b_ = simde_int8x16_to_private(b); SIMDE_VECTORIZE for (size_t i = 0 ; i < (sizeof(r_.values) / sizeof(r_.values[0])) ; i++) { r_.values[i] = simde_vsqaddb_u8(a_.values[i], b_.values[i]); } return simde_uint8x16_from_private(r_); #endif } #if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES) #undef vsqaddq_u8 #define vsqaddq_u8(a, b) simde_vsqaddq_u8((a), (b)) #endif SIMDE_FUNCTION_ATTRIBUTES simde_uint16x8_t simde_vsqaddq_u16(simde_uint16x8_t a, simde_int16x8_t b) { #if defined(SIMDE_ARM_NEON_A64V8_NATIVE) return vsqaddq_u16(a, b); #else simde_uint16x8_private r_, a_ = simde_uint16x8_to_private(a); simde_int16x8_private b_ = simde_int16x8_to_private(b); SIMDE_VECTORIZE for (size_t i = 0 ; i < (sizeof(r_.values) / sizeof(r_.values[0])) ; i++) { r_.values[i] = simde_vsqaddh_u16(a_.values[i], b_.values[i]); } return simde_uint16x8_from_private(r_); #endif } #if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES) #undef vsqaddq_u16 #define vsqaddq_u16(a, b) simde_vsqaddq_u16((a), (b)) #endif SIMDE_FUNCTION_ATTRIBUTES simde_uint32x4_t simde_vsqaddq_u32(simde_uint32x4_t a, simde_int32x4_t b) { #if defined(SIMDE_ARM_NEON_A64V8_NATIVE) return vsqaddq_u32(a, b); #else simde_uint32x4_private r_, a_ = simde_uint32x4_to_private(a); simde_int32x4_private b_ = simde_int32x4_to_private(b); SIMDE_VECTORIZE for (size_t i = 0 ; i < (sizeof(r_.values) / sizeof(r_.values[0])) ; i++) { r_.values[i] = simde_vsqadds_u32(a_.values[i], b_.values[i]); } return simde_uint32x4_from_private(r_); #endif } #if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES) #undef vsqaddq_u32 #define vsqaddq_u32(a, b) simde_vsqaddq_u32((a), (b)) #endif SIMDE_FUNCTION_ATTRIBUTES simde_uint64x2_t simde_vsqaddq_u64(simde_uint64x2_t a, simde_int64x2_t b) { #if defined(SIMDE_ARM_NEON_A64V8_NATIVE) return vsqaddq_u64(a, b); #else simde_uint64x2_private r_, a_ = simde_uint64x2_to_private(a); simde_int64x2_private b_ = simde_int64x2_to_private(b); SIMDE_VECTORIZE for (size_t i = 0 ; i < (sizeof(r_.values) / sizeof(r_.values[0])) ; i++) { r_.values[i] = simde_vsqaddd_u64(a_.values[i], b_.values[i]); } return simde_uint64x2_from_private(r_); #endif } #if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES) #undef vsqaddq_u64 #define vsqaddq_u64(a, b) simde_vsqaddq_u64((a), (b)) #endif SIMDE_END_DECLS_ HEDLEY_DIAGNOSTIC_POP #endif /* !defined(SIMDE_ARM_NEON_SQADD_H) */