mmintrin.h: Mark vector intrinsics always_inline.
2005-06-29 Stuart Hastings <stuart@apple.com>
* gcc/config/i386/mmintrin.h: Mark vector intrinsics always_inline.
* gcc/config/i386/emmintrin.h: Likewise.
* gcc/config/i386/pmmintrin.h: Likewise.
* gcc/config/i386/xmmintrin.h: Likewise.
From-SVN: r101425
This commit is contained in:
Stuart Hastings
Stuart Hastings
parent
e972ccce2c
commit
972fcc764a
@ -1,3 +1,10 @@
|
||||
2005-06-29 Stuart Hastings <stuart@apple.com>
|
||||
|
||||
* gcc/config/i386/mmintrin.h: Mark vector intrinsics always_inline.
|
||||
* gcc/config/i386/emmintrin.h: Likewise.
|
||||
* gcc/config/i386/pmmintrin.h: Likewise.
|
||||
* gcc/config/i386/xmmintrin.h: Likewise.
|
||||
|
||||
2005-06-29 Steve Ellcey <sje@cup.hp.com>
|
||||
|
||||
PR middle-end/21969
|
||||
|
||||
File diff suppressed because it is too large
Load Diff
@ -42,26 +42,26 @@ typedef short __v4hi __attribute__ ((__vector_size__ (8)));
|
||||
typedef char __v8qi __attribute__ ((__vector_size__ (8)));
|
||||
|
||||
/* Empty the multimedia state. */
|
||||
static __inline void
|
||||
static __inline void __attribute__((__always_inline__))
|
||||
_mm_empty (void)
|
||||
{
|
||||
__builtin_ia32_emms ();
|
||||
}
|
||||
|
||||
static __inline void
|
||||
static __inline void __attribute__((__always_inline__))
|
||||
_m_empty (void)
|
||||
{
|
||||
_mm_empty ();
|
||||
}
|
||||
|
||||
/* Convert I to a __m64 object. The integer is zero-extended to 64-bits. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_cvtsi32_si64 (int __i)
|
||||
{
|
||||
return (__m64) __builtin_ia32_vec_init_v2si (__i, 0);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_from_int (int __i)
|
||||
{
|
||||
return _mm_cvtsi32_si64 (__i);
|
||||
@ -69,14 +69,14 @@ _m_from_int (int __i)
|
||||
|
||||
#ifdef __x86_64__
|
||||
/* Convert I to a __m64 object. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_cvtsi64x_si64 (long long __i)
|
||||
{
|
||||
return (__m64) __i;
|
||||
}
|
||||
|
||||
/* Convert I to a __m64 object. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_set_pi64x (long long __i)
|
||||
{
|
||||
return (__m64) __i;
|
||||
@ -84,13 +84,13 @@ _mm_set_pi64x (long long __i)
|
||||
#endif
|
||||
|
||||
/* Convert the lower 32 bits of the __m64 object into an integer. */
|
||||
static __inline int
|
||||
static __inline int __attribute__((__always_inline__))
|
||||
_mm_cvtsi64_si32 (__m64 __i)
|
||||
{
|
||||
return __builtin_ia32_vec_ext_v2si ((__v2si)__i, 0);
|
||||
}
|
||||
|
||||
static __inline int
|
||||
static __inline int __attribute__((__always_inline__))
|
||||
_m_to_int (__m64 __i)
|
||||
{
|
||||
return _mm_cvtsi64_si32 (__i);
|
||||
@ -98,7 +98,7 @@ _m_to_int (__m64 __i)
|
||||
|
||||
#ifdef __x86_64__
|
||||
/* Convert the lower 32 bits of the __m64 object into an integer. */
|
||||
static __inline long long
|
||||
static __inline long long __attribute__((__always_inline__))
|
||||
_mm_cvtsi64_si64x (__m64 __i)
|
||||
{
|
||||
return (long long)__i;
|
||||
@ -108,13 +108,13 @@ _mm_cvtsi64_si64x (__m64 __i)
|
||||
/* Pack the four 16-bit values from M1 into the lower four 8-bit values of
|
||||
the result, and the four 16-bit values from M2 into the upper four 8-bit
|
||||
values of the result, all with signed saturation. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_packs_pi16 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_packsswb ((__v4hi)__m1, (__v4hi)__m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_packsswb (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_packs_pi16 (__m1, __m2);
|
||||
@ -123,13 +123,13 @@ _m_packsswb (__m64 __m1, __m64 __m2)
|
||||
/* Pack the two 32-bit values from M1 in to the lower two 16-bit values of
|
||||
the result, and the two 32-bit values from M2 into the upper two 16-bit
|
||||
values of the result, all with signed saturation. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_packs_pi32 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_packssdw ((__v2si)__m1, (__v2si)__m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_packssdw (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_packs_pi32 (__m1, __m2);
|
||||
@ -138,13 +138,13 @@ _m_packssdw (__m64 __m1, __m64 __m2)
|
||||
/* Pack the four 16-bit values from M1 into the lower four 8-bit values of
|
||||
the result, and the four 16-bit values from M2 into the upper four 8-bit
|
||||
values of the result, all with unsigned saturation. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_packs_pu16 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_packuswb ((__v4hi)__m1, (__v4hi)__m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_packuswb (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_packs_pu16 (__m1, __m2);
|
||||
@ -152,13 +152,13 @@ _m_packuswb (__m64 __m1, __m64 __m2)
|
||||
|
||||
/* Interleave the four 8-bit values from the high half of M1 with the four
|
||||
8-bit values from the high half of M2. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_unpackhi_pi8 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_punpckhbw ((__v8qi)__m1, (__v8qi)__m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_punpckhbw (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_unpackhi_pi8 (__m1, __m2);
|
||||
@ -166,13 +166,13 @@ _m_punpckhbw (__m64 __m1, __m64 __m2)
|
||||
|
||||
/* Interleave the two 16-bit values from the high half of M1 with the two
|
||||
16-bit values from the high half of M2. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_unpackhi_pi16 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_punpckhwd ((__v4hi)__m1, (__v4hi)__m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_punpckhwd (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_unpackhi_pi16 (__m1, __m2);
|
||||
@ -180,13 +180,13 @@ _m_punpckhwd (__m64 __m1, __m64 __m2)
|
||||
|
||||
/* Interleave the 32-bit value from the high half of M1 with the 32-bit
|
||||
value from the high half of M2. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_unpackhi_pi32 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_punpckhdq ((__v2si)__m1, (__v2si)__m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_punpckhdq (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_unpackhi_pi32 (__m1, __m2);
|
||||
@ -194,13 +194,13 @@ _m_punpckhdq (__m64 __m1, __m64 __m2)
|
||||
|
||||
/* Interleave the four 8-bit values from the low half of M1 with the four
|
||||
8-bit values from the low half of M2. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_unpacklo_pi8 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_punpcklbw ((__v8qi)__m1, (__v8qi)__m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_punpcklbw (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_unpacklo_pi8 (__m1, __m2);
|
||||
@ -208,13 +208,13 @@ _m_punpcklbw (__m64 __m1, __m64 __m2)
|
||||
|
||||
/* Interleave the two 16-bit values from the low half of M1 with the two
|
||||
16-bit values from the low half of M2. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_unpacklo_pi16 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_punpcklwd ((__v4hi)__m1, (__v4hi)__m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_punpcklwd (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_unpacklo_pi16 (__m1, __m2);
|
||||
@ -222,59 +222,59 @@ _m_punpcklwd (__m64 __m1, __m64 __m2)
|
||||
|
||||
/* Interleave the 32-bit value from the low half of M1 with the 32-bit
|
||||
value from the low half of M2. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_unpacklo_pi32 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_punpckldq ((__v2si)__m1, (__v2si)__m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_punpckldq (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_unpacklo_pi32 (__m1, __m2);
|
||||
}
|
||||
|
||||
/* Add the 8-bit values in M1 to the 8-bit values in M2. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_add_pi8 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_paddb ((__v8qi)__m1, (__v8qi)__m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_paddb (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_add_pi8 (__m1, __m2);
|
||||
}
|
||||
|
||||
/* Add the 16-bit values in M1 to the 16-bit values in M2. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_add_pi16 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_paddw ((__v4hi)__m1, (__v4hi)__m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_paddw (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_add_pi16 (__m1, __m2);
|
||||
}
|
||||
|
||||
/* Add the 32-bit values in M1 to the 32-bit values in M2. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_add_pi32 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_paddd ((__v2si)__m1, (__v2si)__m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_paddd (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_add_pi32 (__m1, __m2);
|
||||
}
|
||||
|
||||
/* Add the 64-bit values in M1 to the 64-bit values in M2. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_add_si64 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_paddq ((long long)__m1, (long long)__m2);
|
||||
@ -282,13 +282,13 @@ _mm_add_si64 (__m64 __m1, __m64 __m2)
|
||||
|
||||
/* Add the 8-bit values in M1 to the 8-bit values in M2 using signed
|
||||
saturated arithmetic. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_adds_pi8 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_paddsb ((__v8qi)__m1, (__v8qi)__m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_paddsb (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_adds_pi8 (__m1, __m2);
|
||||
@ -296,13 +296,13 @@ _m_paddsb (__m64 __m1, __m64 __m2)
|
||||
|
||||
/* Add the 16-bit values in M1 to the 16-bit values in M2 using signed
|
||||
saturated arithmetic. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_adds_pi16 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_paddsw ((__v4hi)__m1, (__v4hi)__m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_paddsw (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_adds_pi16 (__m1, __m2);
|
||||
@ -310,13 +310,13 @@ _m_paddsw (__m64 __m1, __m64 __m2)
|
||||
|
||||
/* Add the 8-bit values in M1 to the 8-bit values in M2 using unsigned
|
||||
saturated arithmetic. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_adds_pu8 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_paddusb ((__v8qi)__m1, (__v8qi)__m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_paddusb (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_adds_pu8 (__m1, __m2);
|
||||
@ -324,59 +324,59 @@ _m_paddusb (__m64 __m1, __m64 __m2)
|
||||
|
||||
/* Add the 16-bit values in M1 to the 16-bit values in M2 using unsigned
|
||||
saturated arithmetic. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_adds_pu16 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_paddusw ((__v4hi)__m1, (__v4hi)__m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_paddusw (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_adds_pu16 (__m1, __m2);
|
||||
}
|
||||
|
||||
/* Subtract the 8-bit values in M2 from the 8-bit values in M1. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_sub_pi8 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_psubb ((__v8qi)__m1, (__v8qi)__m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_psubb (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_sub_pi8 (__m1, __m2);
|
||||
}
|
||||
|
||||
/* Subtract the 16-bit values in M2 from the 16-bit values in M1. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_sub_pi16 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_psubw ((__v4hi)__m1, (__v4hi)__m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_psubw (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_sub_pi16 (__m1, __m2);
|
||||
}
|
||||
|
||||
/* Subtract the 32-bit values in M2 from the 32-bit values in M1. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_sub_pi32 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_psubd ((__v2si)__m1, (__v2si)__m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_psubd (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_sub_pi32 (__m1, __m2);
|
||||
}
|
||||
|
||||
/* Add the 64-bit values in M1 to the 64-bit values in M2. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_sub_si64 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_psubq ((long long)__m1, (long long)__m2);
|
||||
@ -384,13 +384,13 @@ _mm_sub_si64 (__m64 __m1, __m64 __m2)
|
||||
|
||||
/* Subtract the 8-bit values in M2 from the 8-bit values in M1 using signed
|
||||
saturating arithmetic. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_subs_pi8 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_psubsb ((__v8qi)__m1, (__v8qi)__m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_psubsb (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_subs_pi8 (__m1, __m2);
|
||||
@ -398,13 +398,13 @@ _m_psubsb (__m64 __m1, __m64 __m2)
|
||||
|
||||
/* Subtract the 16-bit values in M2 from the 16-bit values in M1 using
|
||||
signed saturating arithmetic. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_subs_pi16 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_psubsw ((__v4hi)__m1, (__v4hi)__m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_psubsw (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_subs_pi16 (__m1, __m2);
|
||||
@ -412,13 +412,13 @@ _m_psubsw (__m64 __m1, __m64 __m2)
|
||||
|
||||
/* Subtract the 8-bit values in M2 from the 8-bit values in M1 using
|
||||
unsigned saturating arithmetic. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_subs_pu8 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_psubusb ((__v8qi)__m1, (__v8qi)__m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_psubusb (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_subs_pu8 (__m1, __m2);
|
||||
@ -426,13 +426,13 @@ _m_psubusb (__m64 __m1, __m64 __m2)
|
||||
|
||||
/* Subtract the 16-bit values in M2 from the 16-bit values in M1 using
|
||||
unsigned saturating arithmetic. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_subs_pu16 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_psubusw ((__v4hi)__m1, (__v4hi)__m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_psubusw (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_subs_pu16 (__m1, __m2);
|
||||
@ -441,13 +441,13 @@ _m_psubusw (__m64 __m1, __m64 __m2)
|
||||
/* Multiply four 16-bit values in M1 by four 16-bit values in M2 producing
|
||||
four 32-bit intermediate results, which are then summed by pairs to
|
||||
produce two 32-bit results. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_madd_pi16 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_pmaddwd ((__v4hi)__m1, (__v4hi)__m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_pmaddwd (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_madd_pi16 (__m1, __m2);
|
||||
@ -455,13 +455,13 @@ _m_pmaddwd (__m64 __m1, __m64 __m2)
|
||||
|
||||
/* Multiply four signed 16-bit values in M1 by four signed 16-bit values in
|
||||
M2 and produce the high 16 bits of the 32-bit results. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_mulhi_pi16 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_pmulhw ((__v4hi)__m1, (__v4hi)__m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_pmulhw (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_mulhi_pi16 (__m1, __m2);
|
||||
@ -469,226 +469,226 @@ _m_pmulhw (__m64 __m1, __m64 __m2)
|
||||
|
||||
/* Multiply four 16-bit values in M1 by four 16-bit values in M2 and produce
|
||||
the low 16 bits of the results. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_mullo_pi16 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_pmullw ((__v4hi)__m1, (__v4hi)__m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_pmullw (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_mullo_pi16 (__m1, __m2);
|
||||
}
|
||||
|
||||
/* Shift four 16-bit values in M left by COUNT. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_sll_pi16 (__m64 __m, __m64 __count)
|
||||
{
|
||||
return (__m64) __builtin_ia32_psllw ((__v4hi)__m, (long long)__count);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_psllw (__m64 __m, __m64 __count)
|
||||
{
|
||||
return _mm_sll_pi16 (__m, __count);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_slli_pi16 (__m64 __m, int __count)
|
||||
{
|
||||
return (__m64) __builtin_ia32_psllw ((__v4hi)__m, __count);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_psllwi (__m64 __m, int __count)
|
||||
{
|
||||
return _mm_slli_pi16 (__m, __count);
|
||||
}
|
||||
|
||||
/* Shift two 32-bit values in M left by COUNT. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_sll_pi32 (__m64 __m, __m64 __count)
|
||||
{
|
||||
return (__m64) __builtin_ia32_pslld ((__v2si)__m, (long long)__count);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_pslld (__m64 __m, __m64 __count)
|
||||
{
|
||||
return _mm_sll_pi32 (__m, __count);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_slli_pi32 (__m64 __m, int __count)
|
||||
{
|
||||
return (__m64) __builtin_ia32_pslld ((__v2si)__m, __count);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_pslldi (__m64 __m, int __count)
|
||||
{
|
||||
return _mm_slli_pi32 (__m, __count);
|
||||
}
|
||||
|
||||
/* Shift the 64-bit value in M left by COUNT. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_sll_si64 (__m64 __m, __m64 __count)
|
||||
{
|
||||
return (__m64) __builtin_ia32_psllq ((long long)__m, (long long)__count);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_psllq (__m64 __m, __m64 __count)
|
||||
{
|
||||
return _mm_sll_si64 (__m, __count);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_slli_si64 (__m64 __m, int __count)
|
||||
{
|
||||
return (__m64) __builtin_ia32_psllq ((long long)__m, (long long)__count);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_psllqi (__m64 __m, int __count)
|
||||
{
|
||||
return _mm_slli_si64 (__m, __count);
|
||||
}
|
||||
|
||||
/* Shift four 16-bit values in M right by COUNT; shift in the sign bit. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_sra_pi16 (__m64 __m, __m64 __count)
|
||||
{
|
||||
return (__m64) __builtin_ia32_psraw ((__v4hi)__m, (long long)__count);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_psraw (__m64 __m, __m64 __count)
|
||||
{
|
||||
return _mm_sra_pi16 (__m, __count);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_srai_pi16 (__m64 __m, int __count)
|
||||
{
|
||||
return (__m64) __builtin_ia32_psraw ((__v4hi)__m, __count);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_psrawi (__m64 __m, int __count)
|
||||
{
|
||||
return _mm_srai_pi16 (__m, __count);
|
||||
}
|
||||
|
||||
/* Shift two 32-bit values in M right by COUNT; shift in the sign bit. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_sra_pi32 (__m64 __m, __m64 __count)
|
||||
{
|
||||
return (__m64) __builtin_ia32_psrad ((__v2si)__m, (long long)__count);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_psrad (__m64 __m, __m64 __count)
|
||||
{
|
||||
return _mm_sra_pi32 (__m, __count);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_srai_pi32 (__m64 __m, int __count)
|
||||
{
|
||||
return (__m64) __builtin_ia32_psrad ((__v2si)__m, __count);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_psradi (__m64 __m, int __count)
|
||||
{
|
||||
return _mm_srai_pi32 (__m, __count);
|
||||
}
|
||||
|
||||
/* Shift four 16-bit values in M right by COUNT; shift in zeros. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_srl_pi16 (__m64 __m, __m64 __count)
|
||||
{
|
||||
return (__m64) __builtin_ia32_psrlw ((__v4hi)__m, (long long)__count);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_psrlw (__m64 __m, __m64 __count)
|
||||
{
|
||||
return _mm_srl_pi16 (__m, __count);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_srli_pi16 (__m64 __m, int __count)
|
||||
{
|
||||
return (__m64) __builtin_ia32_psrlw ((__v4hi)__m, __count);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_psrlwi (__m64 __m, int __count)
|
||||
{
|
||||
return _mm_srli_pi16 (__m, __count);
|
||||
}
|
||||
|
||||
/* Shift two 32-bit values in M right by COUNT; shift in zeros. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_srl_pi32 (__m64 __m, __m64 __count)
|
||||
{
|
||||
return (__m64) __builtin_ia32_psrld ((__v2si)__m, (long long)__count);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_psrld (__m64 __m, __m64 __count)
|
||||
{
|
||||
return _mm_srl_pi32 (__m, __count);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_srli_pi32 (__m64 __m, int __count)
|
||||
{
|
||||
return (__m64) __builtin_ia32_psrld ((__v2si)__m, __count);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_psrldi (__m64 __m, int __count)
|
||||
{
|
||||
return _mm_srli_pi32 (__m, __count);
|
||||
}
|
||||
|
||||
/* Shift the 64-bit value in M left by COUNT; shift in zeros. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_srl_si64 (__m64 __m, __m64 __count)
|
||||
{
|
||||
return (__m64) __builtin_ia32_psrlq ((long long)__m, (long long)__count);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_psrlq (__m64 __m, __m64 __count)
|
||||
{
|
||||
return _mm_srl_si64 (__m, __count);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_srli_si64 (__m64 __m, int __count)
|
||||
{
|
||||
return (__m64) __builtin_ia32_psrlq ((long long)__m, (long long)__count);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_psrlqi (__m64 __m, int __count)
|
||||
{
|
||||
return _mm_srli_si64 (__m, __count);
|
||||
}
|
||||
|
||||
/* Bit-wise AND the 64-bit values in M1 and M2. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_and_si64 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return __builtin_ia32_pand (__m1, __m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_pand (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_and_si64 (__m1, __m2);
|
||||
@ -696,39 +696,39 @@ _m_pand (__m64 __m1, __m64 __m2)
|
||||
|
||||
/* Bit-wise complement the 64-bit value in M1 and bit-wise AND it with the
|
||||
64-bit value in M2. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_andnot_si64 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return __builtin_ia32_pandn (__m1, __m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_pandn (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_andnot_si64 (__m1, __m2);
|
||||
}
|
||||
|
||||
/* Bit-wise inclusive OR the 64-bit values in M1 and M2. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_or_si64 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return __builtin_ia32_por (__m1, __m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_por (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_or_si64 (__m1, __m2);
|
||||
}
|
||||
|
||||
/* Bit-wise exclusive OR the 64-bit values in M1 and M2. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_xor_si64 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return __builtin_ia32_pxor (__m1, __m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_pxor (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_xor_si64 (__m1, __m2);
|
||||
@ -736,25 +736,25 @@ _m_pxor (__m64 __m1, __m64 __m2)
|
||||
|
||||
/* Compare eight 8-bit values. The result of the comparison is 0xFF if the
|
||||
test is true and zero if false. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_cmpeq_pi8 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_pcmpeqb ((__v8qi)__m1, (__v8qi)__m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_pcmpeqb (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_cmpeq_pi8 (__m1, __m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_cmpgt_pi8 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_pcmpgtb ((__v8qi)__m1, (__v8qi)__m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_pcmpgtb (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_cmpgt_pi8 (__m1, __m2);
|
||||
@ -762,25 +762,25 @@ _m_pcmpgtb (__m64 __m1, __m64 __m2)
|
||||
|
||||
/* Compare four 16-bit values. The result of the comparison is 0xFFFF if
|
||||
the test is true and zero if false. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_cmpeq_pi16 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_pcmpeqw ((__v4hi)__m1, (__v4hi)__m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_pcmpeqw (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_cmpeq_pi16 (__m1, __m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_cmpgt_pi16 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_pcmpgtw ((__v4hi)__m1, (__v4hi)__m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_pcmpgtw (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_cmpgt_pi16 (__m1, __m2);
|
||||
@ -788,53 +788,53 @@ _m_pcmpgtw (__m64 __m1, __m64 __m2)
|
||||
|
||||
/* Compare two 32-bit values. The result of the comparison is 0xFFFFFFFF if
|
||||
the test is true and zero if false. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_cmpeq_pi32 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_pcmpeqd ((__v2si)__m1, (__v2si)__m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_pcmpeqd (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_cmpeq_pi32 (__m1, __m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_cmpgt_pi32 (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return (__m64) __builtin_ia32_pcmpgtd ((__v2si)__m1, (__v2si)__m2);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_m_pcmpgtd (__m64 __m1, __m64 __m2)
|
||||
{
|
||||
return _mm_cmpgt_pi32 (__m1, __m2);
|
||||
}
|
||||
|
||||
/* Creates a 64-bit zero. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_setzero_si64 (void)
|
||||
{
|
||||
return (__m64)0LL;
|
||||
}
|
||||
|
||||
/* Creates a vector of two 32-bit values; I0 is least significant. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_set_pi32 (int __i1, int __i0)
|
||||
{
|
||||
return (__m64) __builtin_ia32_vec_init_v2si (__i0, __i1);
|
||||
}
|
||||
|
||||
/* Creates a vector of four 16-bit values; W0 is least significant. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_set_pi16 (short __w3, short __w2, short __w1, short __w0)
|
||||
{
|
||||
return (__m64) __builtin_ia32_vec_init_v4hi (__w0, __w1, __w2, __w3);
|
||||
}
|
||||
|
||||
/* Creates a vector of eight 8-bit values; B0 is least significant. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_set_pi8 (char __b7, char __b6, char __b5, char __b4,
|
||||
char __b3, char __b2, char __b1, char __b0)
|
||||
{
|
||||
@ -843,19 +843,19 @@ _mm_set_pi8 (char __b7, char __b6, char __b5, char __b4,
|
||||
}
|
||||
|
||||
/* Similar, but with the arguments in reverse order. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_setr_pi32 (int __i0, int __i1)
|
||||
{
|
||||
return _mm_set_pi32 (__i1, __i0);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_setr_pi16 (short __w0, short __w1, short __w2, short __w3)
|
||||
{
|
||||
return _mm_set_pi16 (__w3, __w2, __w1, __w0);
|
||||
}
|
||||
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_setr_pi8 (char __b0, char __b1, char __b2, char __b3,
|
||||
char __b4, char __b5, char __b6, char __b7)
|
||||
{
|
||||
@ -863,21 +863,21 @@ _mm_setr_pi8 (char __b0, char __b1, char __b2, char __b3,
|
||||
}
|
||||
|
||||
/* Creates a vector of two 32-bit values, both elements containing I. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_set1_pi32 (int __i)
|
||||
{
|
||||
return _mm_set_pi32 (__i, __i);
|
||||
}
|
||||
|
||||
/* Creates a vector of four 16-bit values, all elements containing W. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_set1_pi16 (short __w)
|
||||
{
|
||||
return _mm_set_pi16 (__w, __w, __w, __w);
|
||||
}
|
||||
|
||||
/* Creates a vector of eight 8-bit values, all elements containing B. */
|
||||
static __inline __m64
|
||||
static __inline __m64 __attribute__((__always_inline__))
|
||||
_mm_set1_pi8 (char __b)
|
||||
{
|
||||
return _mm_set_pi8 (__b, __b, __b, __b, __b, __b, __b, __b);
|
||||
|
||||
@ -44,80 +44,80 @@
|
||||
#define _MM_GET_DENORMALS_ZERO_MODE() \
|
||||
(_mm_getcsr() & _MM_DENORMALS_ZERO_MASK)
|
||||
|
||||
static __inline __m128
|
||||
static __inline __m128 __attribute__((__always_inline__))
|
||||
_mm_addsub_ps (__m128 __X, __m128 __Y)
|
||||
{
|
||||
return (__m128) __builtin_ia32_addsubps ((__v4sf)__X, (__v4sf)__Y);
|
||||
}
|
||||
|
||||
static __inline __m128
|
||||
static __inline __m128 __attribute__((__always_inline__))
|
||||
_mm_hadd_ps (__m128 __X, __m128 __Y)
|
||||
{
|
||||
return (__m128) __builtin_ia32_haddps ((__v4sf)__X, (__v4sf)__Y);
|
||||
}
|
||||
|
||||
static __inline __m128
|
||||
static __inline __m128 __attribute__((__always_inline__))
|
||||
_mm_hsub_ps (__m128 __X, __m128 __Y)
|
||||
{
|
||||
return (__m128) __builtin_ia32_hsubps ((__v4sf)__X, (__v4sf)__Y);
|
||||
}
|
||||
|
||||
static __inline __m128
|
||||
static __inline __m128 __attribute__((__always_inline__))
|
||||
_mm_movehdup_ps (__m128 __X)
|
||||
{
|
||||
return (__m128) __builtin_ia32_movshdup ((__v4sf)__X);
|
||||
}
|
||||
|
||||
static __inline __m128
|
||||
static __inline __m128 __attribute__((__always_inline__))
|
||||
_mm_moveldup_ps (__m128 __X)
|
||||
{
|
||||
return (__m128) __builtin_ia32_movsldup ((__v4sf)__X);
|
||||
}
|
||||
|
||||
static __inline __m128d
|
||||
static __inline __m128d __attribute__((__always_inline__))
|
||||
_mm_addsub_pd (__m128d __X, __m128d __Y)
|
||||
{
|
||||
return (__m128d) __builtin_ia32_addsubpd ((__v2df)__X, (__v2df)__Y);
|
||||
}
|
||||
|
||||
static __inline __m128d
|
||||
static __inline __m128d __attribute__((__always_inline__))
|
||||
_mm_hadd_pd (__m128d __X, __m128d __Y)
|
||||
{
|
||||
return (__m128d) __builtin_ia32_haddpd ((__v2df)__X, (__v2df)__Y);
|
||||
}
|
||||
|
||||
static __inline __m128d
|
||||
static __inline __m128d __attribute__((__always_inline__))
|
||||
_mm_hsub_pd (__m128d __X, __m128d __Y)
|
||||
{
|
||||
return (__m128d) __builtin_ia32_hsubpd ((__v2df)__X, (__v2df)__Y);
|
||||
}
|
||||
|
||||
static __inline __m128d
|
||||
static __inline __m128d __attribute__((__always_inline__))
|
||||
_mm_loaddup_pd (double const *__P)
|
||||
{
|
||||
return _mm_load1_pd (__P);
|
||||
}
|
||||
|
||||
static __inline __m128d
|
||||
static __inline __m128d __attribute__((__always_inline__))
|
||||
_mm_movedup_pd (__m128d __X)
|
||||
{
|
||||
return _mm_shuffle_pd (__X, __X, _MM_SHUFFLE2 (0,0));
|
||||
}
|
||||
|
||||
static __inline __m128i
|
||||
static __inline __m128i __attribute__((__always_inline__))
|
||||
_mm_lddqu_si128 (__m128i const *__P)
|
||||
{
|
||||
return (__m128i) __builtin_ia32_lddqu ((char const *)__P);
|
||||
}
|
||||
|
||||
#if 0
|
||||
static __inline void
|
||||
static __inline void __attribute__((__always_inline__))
|
||||
_mm_monitor (void const * __P, unsigned int __E, unsigned int __H)
|
||||
{
|
||||
__builtin_ia32_monitor (__P, __E, __H);
|
||||
}
|
||||
|
||||
static __inline void
|
||||
static __inline void __attribute__((__always_inline__))
|
||||
_mm_mwait (unsigned int __E, unsigned int __H)
|
||||
{
|
||||
__builtin_ia32_mwait (__E, __H);
|
||||
|
||||
File diff suppressed because it is too large
Load Diff
Loading…
Reference in New Issue
Block a user