1/*===---- pmmintrin.h - SSE3 intrinsics ------------------------------------=== 2 * 3 * Permission is hereby granted, free of charge, to any person obtaining a copy 4 * of this software and associated documentation files (the "Software"), to deal 5 * in the Software without restriction, including without limitation the rights 6 * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell 7 * copies of the Software, and to permit persons to whom the Software is 8 * furnished to do so, subject to the following conditions: 9 * 10 * The above copyright notice and this permission notice shall be included in 11 * all copies or substantial portions of the Software. 12 * 13 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 14 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 15 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE 16 * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER 17 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, 18 * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN 19 * THE SOFTWARE. 20 * 21 *===-----------------------------------------------------------------------=== 22 */ 23 24#ifndef __PMMINTRIN_H 25#define __PMMINTRIN_H 26 27#ifndef __SSE3__ 28#error "SSE3 instruction set not enabled" 29#else 30 31#include <emmintrin.h> 32 33static __inline__ __m128i __attribute__((__always_inline__, __nodebug__)) 34_mm_lddqu_si128(__m128i const *__p) 35{ 36 return (__m128i)__builtin_ia32_lddqu((char const *)__p); 37} 38 39static __inline__ __m128 __attribute__((__always_inline__, __nodebug__)) 40_mm_addsub_ps(__m128 __a, __m128 __b) 41{ 42 return __builtin_ia32_addsubps(__a, __b); 43} 44 45static __inline__ __m128 __attribute__((__always_inline__, __nodebug__)) 46_mm_hadd_ps(__m128 __a, __m128 __b) 47{ 48 return __builtin_ia32_haddps(__a, __b); 49} 50 51static __inline__ __m128 __attribute__((__always_inline__, __nodebug__)) 52_mm_hsub_ps(__m128 __a, __m128 __b) 53{ 54 return __builtin_ia32_hsubps(__a, __b); 55} 56 57static __inline__ __m128 __attribute__((__always_inline__, __nodebug__)) 58_mm_movehdup_ps(__m128 __a) 59{ 60 return __builtin_shufflevector(__a, __a, 1, 1, 3, 3); 61} 62 63static __inline__ __m128 __attribute__((__always_inline__, __nodebug__)) 64_mm_moveldup_ps(__m128 __a) 65{ 66 return __builtin_shufflevector(__a, __a, 0, 0, 2, 2); 67} 68 69static __inline__ __m128d __attribute__((__always_inline__, __nodebug__)) 70_mm_addsub_pd(__m128d __a, __m128d __b) 71{ 72 return __builtin_ia32_addsubpd(__a, __b); 73} 74 75static __inline__ __m128d __attribute__((__always_inline__, __nodebug__)) 76_mm_hadd_pd(__m128d __a, __m128d __b) 77{ 78 return __builtin_ia32_haddpd(__a, __b); 79} 80 81static __inline__ __m128d __attribute__((__always_inline__, __nodebug__)) 82_mm_hsub_pd(__m128d __a, __m128d __b) 83{ 84 return __builtin_ia32_hsubpd(__a, __b); 85} 86 87#define _mm_loaddup_pd(dp) _mm_load1_pd(dp) 88 89static __inline__ __m128d __attribute__((__always_inline__, __nodebug__)) 90_mm_movedup_pd(__m128d __a) 91{ 92 return __builtin_shufflevector(__a, __a, 0, 0); 93} 94 95#define _MM_DENORMALS_ZERO_ON (0x0040) 96#define _MM_DENORMALS_ZERO_OFF (0x0000) 97 98#define _MM_DENORMALS_ZERO_MASK (0x0040) 99 100#define _MM_GET_DENORMALS_ZERO_MODE() (_mm_getcsr() & _MM_DENORMALS_ZERO_MASK) 101#define _MM_SET_DENORMALS_ZERO_MODE(x) (_mm_setcsr((_mm_getcsr() & ~_MM_DENORMALS_ZERO_MASK) | (x))) 102 103static __inline__ void __attribute__((__always_inline__, __nodebug__)) 104_mm_monitor(void const *__p, unsigned __extensions, unsigned __hints) 105{ 106 __builtin_ia32_monitor((void *)__p, __extensions, __hints); 107} 108 109static __inline__ void __attribute__((__always_inline__, __nodebug__)) 110_mm_mwait(unsigned __extensions, unsigned __hints) 111{ 112 __builtin_ia32_mwait(__extensions, __hints); 113} 114 115#endif /* __SSE3__ */ 116 117#endif /* __PMMINTRIN_H */ 118