1 |
709 |
jeremybenn |
/* Copyright (C) 2003, 2004, 2005, 2006, 2007, 2008, 2009
|
2 |
|
|
Free Software Foundation, Inc.
|
3 |
|
|
|
4 |
|
|
This file is part of GCC.
|
5 |
|
|
|
6 |
|
|
GCC is free software; you can redistribute it and/or modify
|
7 |
|
|
it under the terms of the GNU General Public License as published by
|
8 |
|
|
the Free Software Foundation; either version 3, or (at your option)
|
9 |
|
|
any later version.
|
10 |
|
|
|
11 |
|
|
GCC is distributed in the hope that it will be useful,
|
12 |
|
|
but WITHOUT ANY WARRANTY; without even the implied warranty of
|
13 |
|
|
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
14 |
|
|
GNU General Public License for more details.
|
15 |
|
|
|
16 |
|
|
Under Section 7 of GPL version 3, you are granted additional
|
17 |
|
|
permissions described in the GCC Runtime Library Exception, version
|
18 |
|
|
3.1, as published by the Free Software Foundation.
|
19 |
|
|
|
20 |
|
|
You should have received a copy of the GNU General Public License and
|
21 |
|
|
a copy of the GCC Runtime Library Exception along with this program;
|
22 |
|
|
see the files COPYING3 and COPYING.RUNTIME respectively. If not, see
|
23 |
|
|
<http://www.gnu.org/licenses/>. */
|
24 |
|
|
|
25 |
|
|
/* Implemented from the specification included in the Intel C++ Compiler
|
26 |
|
|
User Guide and Reference, version 9.0. */
|
27 |
|
|
|
28 |
|
|
#ifndef _PMMINTRIN_H_INCLUDED
|
29 |
|
|
#define _PMMINTRIN_H_INCLUDED
|
30 |
|
|
|
31 |
|
|
#ifndef __SSE3__
|
32 |
|
|
# error "SSE3 instruction set not enabled"
|
33 |
|
|
#else
|
34 |
|
|
|
35 |
|
|
/* We need definitions from the SSE2 and SSE header files*/
|
36 |
|
|
#include <emmintrin.h>
|
37 |
|
|
|
38 |
|
|
/* Additional bits in the MXCSR. */
|
39 |
|
|
#define _MM_DENORMALS_ZERO_MASK 0x0040
|
40 |
|
|
#define _MM_DENORMALS_ZERO_ON 0x0040
|
41 |
|
|
#define _MM_DENORMALS_ZERO_OFF 0x0000
|
42 |
|
|
|
43 |
|
|
#define _MM_SET_DENORMALS_ZERO_MODE(mode) \
|
44 |
|
|
_mm_setcsr ((_mm_getcsr () & ~_MM_DENORMALS_ZERO_MASK) | (mode))
|
45 |
|
|
#define _MM_GET_DENORMALS_ZERO_MODE() \
|
46 |
|
|
(_mm_getcsr() & _MM_DENORMALS_ZERO_MASK)
|
47 |
|
|
|
48 |
|
|
extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
|
49 |
|
|
_mm_addsub_ps (__m128 __X, __m128 __Y)
|
50 |
|
|
{
|
51 |
|
|
return (__m128) __builtin_ia32_addsubps ((__v4sf)__X, (__v4sf)__Y);
|
52 |
|
|
}
|
53 |
|
|
|
54 |
|
|
extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
|
55 |
|
|
_mm_hadd_ps (__m128 __X, __m128 __Y)
|
56 |
|
|
{
|
57 |
|
|
return (__m128) __builtin_ia32_haddps ((__v4sf)__X, (__v4sf)__Y);
|
58 |
|
|
}
|
59 |
|
|
|
60 |
|
|
extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
|
61 |
|
|
_mm_hsub_ps (__m128 __X, __m128 __Y)
|
62 |
|
|
{
|
63 |
|
|
return (__m128) __builtin_ia32_hsubps ((__v4sf)__X, (__v4sf)__Y);
|
64 |
|
|
}
|
65 |
|
|
|
66 |
|
|
extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
|
67 |
|
|
_mm_movehdup_ps (__m128 __X)
|
68 |
|
|
{
|
69 |
|
|
return (__m128) __builtin_ia32_movshdup ((__v4sf)__X);
|
70 |
|
|
}
|
71 |
|
|
|
72 |
|
|
extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
|
73 |
|
|
_mm_moveldup_ps (__m128 __X)
|
74 |
|
|
{
|
75 |
|
|
return (__m128) __builtin_ia32_movsldup ((__v4sf)__X);
|
76 |
|
|
}
|
77 |
|
|
|
78 |
|
|
extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
|
79 |
|
|
_mm_addsub_pd (__m128d __X, __m128d __Y)
|
80 |
|
|
{
|
81 |
|
|
return (__m128d) __builtin_ia32_addsubpd ((__v2df)__X, (__v2df)__Y);
|
82 |
|
|
}
|
83 |
|
|
|
84 |
|
|
extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
|
85 |
|
|
_mm_hadd_pd (__m128d __X, __m128d __Y)
|
86 |
|
|
{
|
87 |
|
|
return (__m128d) __builtin_ia32_haddpd ((__v2df)__X, (__v2df)__Y);
|
88 |
|
|
}
|
89 |
|
|
|
90 |
|
|
extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
|
91 |
|
|
_mm_hsub_pd (__m128d __X, __m128d __Y)
|
92 |
|
|
{
|
93 |
|
|
return (__m128d) __builtin_ia32_hsubpd ((__v2df)__X, (__v2df)__Y);
|
94 |
|
|
}
|
95 |
|
|
|
96 |
|
|
extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
|
97 |
|
|
_mm_loaddup_pd (double const *__P)
|
98 |
|
|
{
|
99 |
|
|
return _mm_load1_pd (__P);
|
100 |
|
|
}
|
101 |
|
|
|
102 |
|
|
extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
|
103 |
|
|
_mm_movedup_pd (__m128d __X)
|
104 |
|
|
{
|
105 |
|
|
return _mm_shuffle_pd (__X, __X, _MM_SHUFFLE2 (0,0));
|
106 |
|
|
}
|
107 |
|
|
|
108 |
|
|
extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
|
109 |
|
|
_mm_lddqu_si128 (__m128i const *__P)
|
110 |
|
|
{
|
111 |
|
|
return (__m128i) __builtin_ia32_lddqu ((char const *)__P);
|
112 |
|
|
}
|
113 |
|
|
|
114 |
|
|
extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
|
115 |
|
|
_mm_monitor (void const * __P, unsigned int __E, unsigned int __H)
|
116 |
|
|
{
|
117 |
|
|
__builtin_ia32_monitor (__P, __E, __H);
|
118 |
|
|
}
|
119 |
|
|
|
120 |
|
|
extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
|
121 |
|
|
_mm_mwait (unsigned int __E, unsigned int __H)
|
122 |
|
|
{
|
123 |
|
|
__builtin_ia32_mwait (__E, __H);
|
124 |
|
|
}
|
125 |
|
|
|
126 |
|
|
#endif /* __SSE3__ */
|
127 |
|
|
|
128 |
|
|
#endif /* _PMMINTRIN_H_INCLUDED */
|