source: Daodan/MSYS2/mingw32/lib/gcc/i686-w64-mingw32/11.2.0/include/mmintrin.h@ 1194

/* Copyright (C) 2002-2021 Free Software Foundation, Inc.

   This file is part of GCC.

   GCC is free software; you can redistribute it and/or modify
   it under the terms of the GNU General Public License as published by
   the Free Software Foundation; either version 3, or (at your option)
   any later version.

   GCC is distributed in the hope that it will be useful,
   but WITHOUT ANY WARRANTY; without even the implied warranty of
   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
   GNU General Public License for more details.

   Under Section 7 of GPL version 3, you are granted additional
   permissions described in the GCC Runtime Library Exception, version
   3.1, as published by the Free Software Foundation.

   You should have received a copy of the GNU General Public License and
   a copy of the GCC Runtime Library Exception along with this program;
   see the files COPYING3 and COPYING.RUNTIME respectively.  If not, see
   <http://www.gnu.org/licenses/>.  */

/* Implemented from the specification included in the Intel C++ Compiler
   User Guide and Reference, version 9.0.  */

#ifndef _MMINTRIN_H_INCLUDED
#define _MMINTRIN_H_INCLUDED

#if defined __x86_64__ && !defined __SSE__ || !defined __MMX__
#pragma GCC push_options
#ifdef __MMX_WITH_SSE__
#pragma GCC target("sse2")
#elif defined __x86_64__
#pragma GCC target("sse,mmx")
#else
#pragma GCC target("mmx")
#endif
#define __DISABLE_MMX__
#endif /* __MMX__ */

/* The Intel API is flexible enough that we must allow aliasing with other
   vector types, and their scalar components.  */
typedef int __m64 __attribute__ ((__vector_size__ (8), __may_alias__));
typedef int __m32 __attribute__ ((__vector_size__ (4), __may_alias__));
typedef short __m16 __attribute__ ((__vector_size__ (2), __may_alias__));

/* Unaligned version of the same type  */
typedef int __m64_u __attribute__ ((__vector_size__ (8), __may_alias__, __aligned__ (1)));
typedef int __m32_u __attribute__ ((__vector_size__ (4), \
                                    __may_alias__, __aligned__ (1)));
typedef short __m16_u __attribute__ ((__vector_size__ (2), \
                                      __may_alias__, __aligned__ (1)));

/* Internal data types for implementing the intrinsics.  */
typedef int __v2si __attribute__ ((__vector_size__ (8)));
typedef short __v4hi __attribute__ ((__vector_size__ (8)));
typedef char __v8qi __attribute__ ((__vector_size__ (8)));
typedef long long __v1di __attribute__ ((__vector_size__ (8)));
typedef float __v2sf __attribute__ ((__vector_size__ (8)));

/* Empty the multimedia state.  */
extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_empty (void)
{
  __builtin_ia32_emms ();
}

extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_empty (void)
{
  _mm_empty ();
}

/* Convert I to a __m64 object.  The integer is zero-extended to 64-bits.  */
extern __inline __m64  __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_cvtsi32_si64 (int __i)
{
  return (__m64) __builtin_ia32_vec_init_v2si (__i, 0);
}

extern __inline __m64  __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_from_int (int __i)
{
  return _mm_cvtsi32_si64 (__i);
}

#ifdef __x86_64__
/* Convert I to a __m64 object.  */

/* Intel intrinsic.  */
extern __inline __m64  __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_from_int64 (long long __i)
{
  return (__m64) __i;
}

extern __inline __m64  __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_cvtsi64_m64 (long long __i)
{
  return (__m64) __i;
}

/* Microsoft intrinsic.  */
extern __inline __m64  __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_cvtsi64x_si64 (long long __i)
{
  return (__m64) __i;
}

extern __inline __m64  __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_set_pi64x (long long __i)
{
  return (__m64) __i;
}
#endif

/* Convert the lower 32 bits of the __m64 object into an integer.  */
extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_cvtsi64_si32 (__m64 __i)
{
  return __builtin_ia32_vec_ext_v2si ((__v2si)__i, 0);
}

extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_to_int (__m64 __i)
{
  return _mm_cvtsi64_si32 (__i);
}

#ifdef __x86_64__
/* Convert the __m64 object to a 64bit integer.  */

/* Intel intrinsic.  */
extern __inline long long __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_to_int64 (__m64 __i)
{
  return (long long)__i;
}

extern __inline long long __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_cvtm64_si64 (__m64 __i)
{
  return (long long)__i;
}

/* Microsoft intrinsic.  */
extern __inline long long __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_cvtsi64_si64x (__m64 __i)
{
  return (long long)__i;
}
#endif

/* 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.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_packs_pi16 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_packsswb ((__v4hi)__m1, (__v4hi)__m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_packsswb (__m64 __m1, __m64 __m2)
{
  return _mm_packs_pi16 (__m1, __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.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_packs_pi32 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_packssdw ((__v2si)__m1, (__v2si)__m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_packssdw (__m64 __m1, __m64 __m2)
{
  return _mm_packs_pi32 (__m1, __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.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_packs_pu16 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_packuswb ((__v4hi)__m1, (__v4hi)__m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_packuswb (__m64 __m1, __m64 __m2)
{
  return _mm_packs_pu16 (__m1, __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.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_unpackhi_pi8 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_punpckhbw ((__v8qi)__m1, (__v8qi)__m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_punpckhbw (__m64 __m1, __m64 __m2)
{
  return _mm_unpackhi_pi8 (__m1, __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.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_unpackhi_pi16 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_punpckhwd ((__v4hi)__m1, (__v4hi)__m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_punpckhwd (__m64 __m1, __m64 __m2)
{
  return _mm_unpackhi_pi16 (__m1, __m2);
}

/* Interleave the 32-bit value from the high half of M1 with the 32-bit
   value from the high half of M2.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_unpackhi_pi32 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_punpckhdq ((__v2si)__m1, (__v2si)__m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_punpckhdq (__m64 __m1, __m64 __m2)
{
  return _mm_unpackhi_pi32 (__m1, __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.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_unpacklo_pi8 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_punpcklbw ((__v8qi)__m1, (__v8qi)__m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_punpcklbw (__m64 __m1, __m64 __m2)
{
  return _mm_unpacklo_pi8 (__m1, __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.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_unpacklo_pi16 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_punpcklwd ((__v4hi)__m1, (__v4hi)__m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_punpcklwd (__m64 __m1, __m64 __m2)
{
  return _mm_unpacklo_pi16 (__m1, __m2);
}

/* Interleave the 32-bit value from the low half of M1 with the 32-bit
   value from the low half of M2.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_unpacklo_pi32 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_punpckldq ((__v2si)__m1, (__v2si)__m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_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.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_add_pi8 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_paddb ((__v8qi)__m1, (__v8qi)__m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_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.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_add_pi16 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_paddw ((__v4hi)__m1, (__v4hi)__m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_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.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_add_pi32 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_paddd ((__v2si)__m1, (__v2si)__m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_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.  */
#ifndef __SSE2__
#pragma GCC push_options
#ifdef __MMX_WITH_SSE__
#pragma GCC target("sse2")
#else
#pragma GCC target("sse2,mmx")
#endif
#define __DISABLE_SSE2__
#endif /* __SSE2__ */

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_add_si64 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_paddq ((__v1di)__m1, (__v1di)__m2);
}
#ifdef __DISABLE_SSE2__
#undef __DISABLE_SSE2__
#pragma GCC pop_options
#endif /* __DISABLE_SSE2__ */

/* Add the 8-bit values in M1 to the 8-bit values in M2 using signed
   saturated arithmetic.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_adds_pi8 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_paddsb ((__v8qi)__m1, (__v8qi)__m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_paddsb (__m64 __m1, __m64 __m2)
{
  return _mm_adds_pi8 (__m1, __m2);
}

/* Add the 16-bit values in M1 to the 16-bit values in M2 using signed
   saturated arithmetic.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_adds_pi16 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_paddsw ((__v4hi)__m1, (__v4hi)__m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_paddsw (__m64 __m1, __m64 __m2)
{
  return _mm_adds_pi16 (__m1, __m2);
}

/* Add the 8-bit values in M1 to the 8-bit values in M2 using unsigned
   saturated arithmetic.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_adds_pu8 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_paddusb ((__v8qi)__m1, (__v8qi)__m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_paddusb (__m64 __m1, __m64 __m2)
{
  return _mm_adds_pu8 (__m1, __m2);
}

/* Add the 16-bit values in M1 to the 16-bit values in M2 using unsigned
   saturated arithmetic.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_adds_pu16 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_paddusw ((__v4hi)__m1, (__v4hi)__m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_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.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_sub_pi8 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_psubb ((__v8qi)__m1, (__v8qi)__m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_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.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_sub_pi16 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_psubw ((__v4hi)__m1, (__v4hi)__m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_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.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_sub_pi32 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_psubd ((__v2si)__m1, (__v2si)__m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_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.  */
#ifndef __SSE2__
#pragma GCC push_options
#ifdef __MMX_WITH_SSE__
#pragma GCC target("sse2")
#else
#pragma GCC target("sse2,mmx")
#endif
#define __DISABLE_SSE2__
#endif /* __SSE2__ */

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_sub_si64 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_psubq ((__v1di)__m1, (__v1di)__m2);
}
#ifdef __DISABLE_SSE2__
#undef __DISABLE_SSE2__
#pragma GCC pop_options
#endif /* __DISABLE_SSE2__ */

/* Subtract the 8-bit values in M2 from the 8-bit values in M1 using signed
   saturating arithmetic.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_subs_pi8 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_psubsb ((__v8qi)__m1, (__v8qi)__m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_psubsb (__m64 __m1, __m64 __m2)
{
  return _mm_subs_pi8 (__m1, __m2);
}

/* Subtract the 16-bit values in M2 from the 16-bit values in M1 using
   signed saturating arithmetic.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_subs_pi16 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_psubsw ((__v4hi)__m1, (__v4hi)__m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_psubsw (__m64 __m1, __m64 __m2)
{
  return _mm_subs_pi16 (__m1, __m2);
}

/* Subtract the 8-bit values in M2 from the 8-bit values in M1 using
   unsigned saturating arithmetic.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_subs_pu8 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_psubusb ((__v8qi)__m1, (__v8qi)__m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_psubusb (__m64 __m1, __m64 __m2)
{
  return _mm_subs_pu8 (__m1, __m2);
}

/* Subtract the 16-bit values in M2 from the 16-bit values in M1 using
   unsigned saturating arithmetic.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_subs_pu16 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_psubusw ((__v4hi)__m1, (__v4hi)__m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_psubusw (__m64 __m1, __m64 __m2)
{
  return _mm_subs_pu16 (__m1, __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.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_madd_pi16 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_pmaddwd ((__v4hi)__m1, (__v4hi)__m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_pmaddwd (__m64 __m1, __m64 __m2)
{
  return _mm_madd_pi16 (__m1, __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.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_mulhi_pi16 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_pmulhw ((__v4hi)__m1, (__v4hi)__m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_pmulhw (__m64 __m1, __m64 __m2)
{
  return _mm_mulhi_pi16 (__m1, __m2);
}

/* Multiply four 16-bit values in M1 by four 16-bit values in M2 and produce
   the low 16 bits of the results.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_mullo_pi16 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_pmullw ((__v4hi)__m1, (__v4hi)__m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_pmullw (__m64 __m1, __m64 __m2)
{
  return _mm_mullo_pi16 (__m1, __m2);
}

/* Shift four 16-bit values in M left by COUNT.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_sll_pi16 (__m64 __m, __m64 __count)
{
  return (__m64) __builtin_ia32_psllw ((__v4hi)__m, (__v4hi)__count);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_psllw (__m64 __m, __m64 __count)
{
  return _mm_sll_pi16 (__m, __count);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_slli_pi16 (__m64 __m, int __count)
{
  return (__m64) __builtin_ia32_psllwi ((__v4hi)__m, __count);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_psllwi (__m64 __m, int __count)
{
  return _mm_slli_pi16 (__m, __count);
}

/* Shift two 32-bit values in M left by COUNT.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_sll_pi32 (__m64 __m, __m64 __count)
{
  return (__m64) __builtin_ia32_pslld ((__v2si)__m, (__v2si)__count);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_pslld (__m64 __m, __m64 __count)
{
  return _mm_sll_pi32 (__m, __count);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_slli_pi32 (__m64 __m, int __count)
{
  return (__m64) __builtin_ia32_pslldi ((__v2si)__m, __count);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_pslldi (__m64 __m, int __count)
{
  return _mm_slli_pi32 (__m, __count);
}

/* Shift the 64-bit value in M left by COUNT.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_sll_si64 (__m64 __m, __m64 __count)
{
  return (__m64) __builtin_ia32_psllq ((__v1di)__m, (__v1di)__count);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_psllq (__m64 __m, __m64 __count)
{
  return _mm_sll_si64 (__m, __count);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_slli_si64 (__m64 __m, int __count)
{
  return (__m64) __builtin_ia32_psllqi ((__v1di)__m, __count);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_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.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_sra_pi16 (__m64 __m, __m64 __count)
{
  return (__m64) __builtin_ia32_psraw ((__v4hi)__m, (__v4hi)__count);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_psraw (__m64 __m, __m64 __count)
{
  return _mm_sra_pi16 (__m, __count);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_srai_pi16 (__m64 __m, int __count)
{
  return (__m64) __builtin_ia32_psrawi ((__v4hi)__m, __count);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_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.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_sra_pi32 (__m64 __m, __m64 __count)
{
  return (__m64) __builtin_ia32_psrad ((__v2si)__m, (__v2si)__count);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_psrad (__m64 __m, __m64 __count)
{
  return _mm_sra_pi32 (__m, __count);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_srai_pi32 (__m64 __m, int __count)
{
  return (__m64) __builtin_ia32_psradi ((__v2si)__m, __count);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_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.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_srl_pi16 (__m64 __m, __m64 __count)
{
  return (__m64) __builtin_ia32_psrlw ((__v4hi)__m, (__v4hi)__count);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_psrlw (__m64 __m, __m64 __count)
{
  return _mm_srl_pi16 (__m, __count);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_srli_pi16 (__m64 __m, int __count)
{
  return (__m64) __builtin_ia32_psrlwi ((__v4hi)__m, __count);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_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.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_srl_pi32 (__m64 __m, __m64 __count)
{
  return (__m64) __builtin_ia32_psrld ((__v2si)__m, (__v2si)__count);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_psrld (__m64 __m, __m64 __count)
{
  return _mm_srl_pi32 (__m, __count);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_srli_pi32 (__m64 __m, int __count)
{
  return (__m64) __builtin_ia32_psrldi ((__v2si)__m, __count);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_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.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_srl_si64 (__m64 __m, __m64 __count)
{
  return (__m64) __builtin_ia32_psrlq ((__v1di)__m, (__v1di)__count);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_psrlq (__m64 __m, __m64 __count)
{
  return _mm_srl_si64 (__m, __count);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_srli_si64 (__m64 __m, int __count)
{
  return (__m64) __builtin_ia32_psrlqi ((__v1di)__m, __count);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_psrlqi (__m64 __m, int __count)
{
  return _mm_srli_si64 (__m, __count);
}

/* Bit-wise AND the 64-bit values in M1 and M2.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_and_si64 (__m64 __m1, __m64 __m2)
{
  return __builtin_ia32_pand (__m1, __m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_pand (__m64 __m1, __m64 __m2)
{
  return _mm_and_si64 (__m1, __m2);
}

/* Bit-wise complement the 64-bit value in M1 and bit-wise AND it with the
   64-bit value in M2.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_andnot_si64 (__m64 __m1, __m64 __m2)
{
  return __builtin_ia32_pandn (__m1, __m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_pandn (__m64 __m1, __m64 __m2)
{
  return _mm_andnot_si64 (__m1, __m2);
}

/* Bit-wise inclusive OR the 64-bit values in M1 and M2.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_or_si64 (__m64 __m1, __m64 __m2)
{
  return __builtin_ia32_por (__m1, __m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_por (__m64 __m1, __m64 __m2)
{
  return _mm_or_si64 (__m1, __m2);
}

/* Bit-wise exclusive OR the 64-bit values in M1 and M2.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_xor_si64 (__m64 __m1, __m64 __m2)
{
  return __builtin_ia32_pxor (__m1, __m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_pxor (__m64 __m1, __m64 __m2)
{
  return _mm_xor_si64 (__m1, __m2);
}

/* Compare eight 8-bit values.  The result of the comparison is 0xFF if the
   test is true and zero if false.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_cmpeq_pi8 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_pcmpeqb ((__v8qi)__m1, (__v8qi)__m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_pcmpeqb (__m64 __m1, __m64 __m2)
{
  return _mm_cmpeq_pi8 (__m1, __m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_cmpgt_pi8 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_pcmpgtb ((__v8qi)__m1, (__v8qi)__m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_pcmpgtb (__m64 __m1, __m64 __m2)
{
  return _mm_cmpgt_pi8 (__m1, __m2);
}

/* Compare four 16-bit values.  The result of the comparison is 0xFFFF if
   the test is true and zero if false.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_cmpeq_pi16 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_pcmpeqw ((__v4hi)__m1, (__v4hi)__m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_pcmpeqw (__m64 __m1, __m64 __m2)
{
  return _mm_cmpeq_pi16 (__m1, __m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_cmpgt_pi16 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_pcmpgtw ((__v4hi)__m1, (__v4hi)__m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_pcmpgtw (__m64 __m1, __m64 __m2)
{
  return _mm_cmpgt_pi16 (__m1, __m2);
}

/* Compare two 32-bit values.  The result of the comparison is 0xFFFFFFFF if
   the test is true and zero if false.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_cmpeq_pi32 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_pcmpeqd ((__v2si)__m1, (__v2si)__m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_pcmpeqd (__m64 __m1, __m64 __m2)
{
  return _mm_cmpeq_pi32 (__m1, __m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_cmpgt_pi32 (__m64 __m1, __m64 __m2)
{
  return (__m64) __builtin_ia32_pcmpgtd ((__v2si)__m1, (__v2si)__m2);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_m_pcmpgtd (__m64 __m1, __m64 __m2)
{
  return _mm_cmpgt_pi32 (__m1, __m2);
}

/* Creates a 64-bit zero.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_setzero_si64 (void)
{
  return (__m64)0LL;
}

/* Creates a vector of two 32-bit values; I0 is least significant.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_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.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_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.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_set_pi8 (char __b7, char __b6, char __b5, char __b4,
             char __b3, char __b2, char __b1, char __b0)
{
  return (__m64) __builtin_ia32_vec_init_v8qi (__b0, __b1, __b2, __b3,
                                               __b4, __b5, __b6, __b7);
}

/* Similar, but with the arguments in reverse order.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_setr_pi32 (int __i0, int __i1)
{
  return _mm_set_pi32 (__i1, __i0);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_setr_pi16 (short __w0, short __w1, short __w2, short __w3)
{
  return _mm_set_pi16 (__w3, __w2, __w1, __w0);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_setr_pi8 (char __b0, char __b1, char __b2, char __b3,
              char __b4, char __b5, char __b6, char __b7)
{
  return _mm_set_pi8 (__b7, __b6, __b5, __b4, __b3, __b2, __b1, __b0);
}

/* Creates a vector of two 32-bit values, both elements containing I.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_set1_pi32 (int __i)
{
  return _mm_set_pi32 (__i, __i);
}

/* Creates a vector of four 16-bit values, all elements containing W.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_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.  */
extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_set1_pi8 (char __b)
{
  return _mm_set_pi8 (__b, __b, __b, __b, __b, __b, __b, __b);
}
#ifdef __DISABLE_MMX__
#undef __DISABLE_MMX__
#pragma GCC pop_options
#endif /* __DISABLE_MMX__ */

#endif /* _MMINTRIN_H_INCLUDED */