#ifndef _XMMINTRIN_H_INCLUDED
#define _XMMINTRIN_H_INCLUDED
#ifdef __x86_64__
#include "third_party/intel/mm_malloc.internal.h"
#include "third_party/intel/mmintrin.internal.h"

enum _mm_hint {
  _MM_HINT_ET0 = 7,
  _MM_HINT_ET1 = 6,
  _MM_HINT_T0 = 3,
  _MM_HINT_T1 = 2,
  _MM_HINT_T2 = 1,
  _MM_HINT_NTA = 0
};

#ifdef __OPTIMIZE__
extern __inline void
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_prefetch(const void *__P, enum _mm_hint __I) {
  __builtin_prefetch(__P, (__I & 0x4) >> 2, __I & 0x3);
}
#else
#define _mm_prefetch(P, I) __builtin_prefetch((P), ((I & 0x4) >> 2), (I & 0x3))
#endif

#ifndef __SSE__
#pragma GCC push_options
#pragma GCC target("sse")
#define __DISABLE_SSE__
#endif /* __SSE__ */

typedef float __m128 __attribute__((__vector_size__(16), __may_alias__));

typedef float __m128_u
    __attribute__((__vector_size__(16), __may_alias__, __aligned__(1)));

typedef float __v4sf __attribute__((__vector_size__(16)));

#define _MM_SHUFFLE(fp3, fp2, fp1, fp0) \
  (((fp3) << 6) | ((fp2) << 4) | ((fp1) << 2) | (fp0))

#define _MM_EXCEPT_MASK      0x003f
#define _MM_EXCEPT_INVALID   0x0001
#define _MM_EXCEPT_DENORM    0x0002
#define _MM_EXCEPT_DIV_ZERO  0x0004
#define _MM_EXCEPT_OVERFLOW  0x0008
#define _MM_EXCEPT_UNDERFLOW 0x0010
#define _MM_EXCEPT_INEXACT   0x0020

#define _MM_MASK_MASK      0x1f80
#define _MM_MASK_INVALID   0x0080
#define _MM_MASK_DENORM    0x0100
#define _MM_MASK_DIV_ZERO  0x0200
#define _MM_MASK_OVERFLOW  0x0400
#define _MM_MASK_UNDERFLOW 0x0800
#define _MM_MASK_INEXACT   0x1000

#define _MM_ROUND_MASK        0x6000
#define _MM_ROUND_NEAREST     0x0000
#define _MM_ROUND_DOWN        0x2000
#define _MM_ROUND_UP          0x4000
#define _MM_ROUND_TOWARD_ZERO 0x6000

#define _MM_FLUSH_ZERO_MASK 0x8000
#define _MM_FLUSH_ZERO_ON   0x8000
#define _MM_FLUSH_ZERO_OFF  0x0000

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_undefined_ps(void) {
  __m128 __Y = __Y;
  return __Y;
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_setzero_ps(void) {
  return __extension__(__m128){0.0f, 0.0f, 0.0f, 0.0f};
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_add_ss(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_addss((__v4sf)__A, (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_sub_ss(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_subss((__v4sf)__A, (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_mul_ss(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_mulss((__v4sf)__A, (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_div_ss(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_divss((__v4sf)__A, (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_sqrt_ss(__m128 __A) {
  return (__m128)__builtin_ia32_sqrtss((__v4sf)__A);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_rcp_ss(__m128 __A) {
  return (__m128)__builtin_ia32_rcpss((__v4sf)__A);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_rsqrt_ss(__m128 __A) {
  return (__m128)__builtin_ia32_rsqrtss((__v4sf)__A);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_min_ss(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_minss((__v4sf)__A, (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_max_ss(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_maxss((__v4sf)__A, (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_add_ps(__m128 __A, __m128 __B) {
  return (__m128)((__v4sf)__A + (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_sub_ps(__m128 __A, __m128 __B) {
  return (__m128)((__v4sf)__A - (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_mul_ps(__m128 __A, __m128 __B) {
  return (__m128)((__v4sf)__A * (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_div_ps(__m128 __A, __m128 __B) {
  return (__m128)((__v4sf)__A / (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_sqrt_ps(__m128 __A) {
  return (__m128)__builtin_ia32_sqrtps((__v4sf)__A);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_rcp_ps(__m128 __A) {
  return (__m128)__builtin_ia32_rcpps((__v4sf)__A);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_rsqrt_ps(__m128 __A) {
  return (__m128)__builtin_ia32_rsqrtps((__v4sf)__A);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_min_ps(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_minps((__v4sf)__A, (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_max_ps(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_maxps((__v4sf)__A, (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_and_ps(__m128 __A, __m128 __B) {
  return __builtin_ia32_andps(__A, __B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_andnot_ps(__m128 __A, __m128 __B) {
  return __builtin_ia32_andnps(__A, __B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_or_ps(__m128 __A, __m128 __B) {
  return __builtin_ia32_orps(__A, __B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_xor_ps(__m128 __A, __m128 __B) {
  return __builtin_ia32_xorps(__A, __B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cmpeq_ss(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_cmpeqss((__v4sf)__A, (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cmplt_ss(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_cmpltss((__v4sf)__A, (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cmple_ss(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_cmpless((__v4sf)__A, (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cmpgt_ss(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_movss(
      (__v4sf)__A, (__v4sf)__builtin_ia32_cmpltss((__v4sf)__B, (__v4sf)__A));
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cmpge_ss(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_movss(
      (__v4sf)__A, (__v4sf)__builtin_ia32_cmpless((__v4sf)__B, (__v4sf)__A));
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cmpneq_ss(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_cmpneqss((__v4sf)__A, (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cmpnlt_ss(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_cmpnltss((__v4sf)__A, (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cmpnle_ss(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_cmpnless((__v4sf)__A, (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cmpngt_ss(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_movss(
      (__v4sf)__A, (__v4sf)__builtin_ia32_cmpnltss((__v4sf)__B, (__v4sf)__A));
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cmpnge_ss(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_movss(
      (__v4sf)__A, (__v4sf)__builtin_ia32_cmpnless((__v4sf)__B, (__v4sf)__A));
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cmpord_ss(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_cmpordss((__v4sf)__A, (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cmpunord_ss(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_cmpunordss((__v4sf)__A, (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cmpeq_ps(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_cmpeqps((__v4sf)__A, (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cmplt_ps(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_cmpltps((__v4sf)__A, (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cmple_ps(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_cmpleps((__v4sf)__A, (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cmpgt_ps(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_cmpgtps((__v4sf)__A, (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cmpge_ps(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_cmpgeps((__v4sf)__A, (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cmpneq_ps(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_cmpneqps((__v4sf)__A, (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cmpnlt_ps(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_cmpnltps((__v4sf)__A, (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cmpnle_ps(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_cmpnleps((__v4sf)__A, (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cmpngt_ps(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_cmpngtps((__v4sf)__A, (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cmpnge_ps(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_cmpngeps((__v4sf)__A, (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cmpord_ps(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_cmpordps((__v4sf)__A, (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cmpunord_ps(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_cmpunordps((__v4sf)__A, (__v4sf)__B);
}

extern __inline int
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_comieq_ss(__m128 __A, __m128 __B) {
  return __builtin_ia32_comieq((__v4sf)__A, (__v4sf)__B);
}

extern __inline int
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_comilt_ss(__m128 __A, __m128 __B) {
  return __builtin_ia32_comilt((__v4sf)__A, (__v4sf)__B);
}

extern __inline int
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_comile_ss(__m128 __A, __m128 __B) {
  return __builtin_ia32_comile((__v4sf)__A, (__v4sf)__B);
}

extern __inline int
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_comigt_ss(__m128 __A, __m128 __B) {
  return __builtin_ia32_comigt((__v4sf)__A, (__v4sf)__B);
}

extern __inline int
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_comige_ss(__m128 __A, __m128 __B) {
  return __builtin_ia32_comige((__v4sf)__A, (__v4sf)__B);
}

extern __inline int
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_comineq_ss(__m128 __A, __m128 __B) {
  return __builtin_ia32_comineq((__v4sf)__A, (__v4sf)__B);
}

extern __inline int
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_ucomieq_ss(__m128 __A, __m128 __B) {
  return __builtin_ia32_ucomieq((__v4sf)__A, (__v4sf)__B);
}

extern __inline int
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_ucomilt_ss(__m128 __A, __m128 __B) {
  return __builtin_ia32_ucomilt((__v4sf)__A, (__v4sf)__B);
}

extern __inline int
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_ucomile_ss(__m128 __A, __m128 __B) {
  return __builtin_ia32_ucomile((__v4sf)__A, (__v4sf)__B);
}

extern __inline int
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_ucomigt_ss(__m128 __A, __m128 __B) {
  return __builtin_ia32_ucomigt((__v4sf)__A, (__v4sf)__B);
}

extern __inline int
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_ucomige_ss(__m128 __A, __m128 __B) {
  return __builtin_ia32_ucomige((__v4sf)__A, (__v4sf)__B);
}

extern __inline int
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_ucomineq_ss(__m128 __A, __m128 __B) {
  return __builtin_ia32_ucomineq((__v4sf)__A, (__v4sf)__B);
}

extern __inline int
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cvtss_si32(__m128 __A) {
  return __builtin_ia32_cvtss2si((__v4sf)__A);
}

extern __inline int
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cvt_ss2si(__m128 __A) {
  return _mm_cvtss_si32(__A);
}

#ifdef __x86_64__

extern __inline long long
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cvtss_si64(__m128 __A) {
  return __builtin_ia32_cvtss2si64((__v4sf)__A);
}

extern __inline long long
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cvtss_si64x(__m128 __A) {
  return __builtin_ia32_cvtss2si64((__v4sf)__A);
}
#endif

extern __inline __m64
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cvtps_pi32(__m128 __A) {
  return (__m64)__builtin_ia32_cvtps2pi((__v4sf)__A);
}

extern __inline __m64
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cvt_ps2pi(__m128 __A) {
  return _mm_cvtps_pi32(__A);
}

extern __inline int
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cvttss_si32(__m128 __A) {
  return __builtin_ia32_cvttss2si((__v4sf)__A);
}

extern __inline int
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cvtt_ss2si(__m128 __A) {
  return _mm_cvttss_si32(__A);
}

#ifdef __x86_64__

extern __inline long long
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cvttss_si64(__m128 __A) {
  return __builtin_ia32_cvttss2si64((__v4sf)__A);
}

extern __inline long long
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cvttss_si64x(__m128 __A) {
  return __builtin_ia32_cvttss2si64((__v4sf)__A);
}
#endif

extern __inline __m64
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cvttps_pi32(__m128 __A) {
  return (__m64)__builtin_ia32_cvttps2pi((__v4sf)__A);
}

extern __inline __m64
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cvtt_ps2pi(__m128 __A) {
  return _mm_cvttps_pi32(__A);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cvtsi32_ss(__m128 __A, int __B) {
  return (__m128)__builtin_ia32_cvtsi2ss((__v4sf)__A, __B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cvt_si2ss(__m128 __A, int __B) {
  return _mm_cvtsi32_ss(__A, __B);
}

#ifdef __x86_64__

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cvtsi64_ss(__m128 __A, long long __B) {
  return (__m128)__builtin_ia32_cvtsi642ss((__v4sf)__A, __B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cvtsi64x_ss(__m128 __A, long long __B) {
  return (__m128)__builtin_ia32_cvtsi642ss((__v4sf)__A, __B);
}
#endif

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cvtpi32_ps(__m128 __A, __m64 __B) {
  return (__m128)__builtin_ia32_cvtpi2ps((__v4sf)__A, (__v2si)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cvt_pi2ps(__m128 __A, __m64 __B) {
  return _mm_cvtpi32_ps(__A, __B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cvtpi16_ps(__m64 __A) {
  __v4hi __sign;
  __v2si __hisi, __losi;
  __v4sf __zero, __ra, __rb;
  __sign = __builtin_ia32_pcmpgtw((__v4hi)0LL, (__v4hi)__A);
  __losi = (__v2si)__builtin_ia32_punpcklwd((__v4hi)__A, __sign);
  __hisi = (__v2si)__builtin_ia32_punpckhwd((__v4hi)__A, __sign);
  __zero = (__v4sf)_mm_setzero_ps();
  __ra = __builtin_ia32_cvtpi2ps(__zero, __losi);
  __rb = __builtin_ia32_cvtpi2ps(__ra, __hisi);
  return (__m128)__builtin_ia32_movlhps(__ra, __rb);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cvtpu16_ps(__m64 __A) {
  __v2si __hisi, __losi;
  __v4sf __zero, __ra, __rb;
  __losi = (__v2si)__builtin_ia32_punpcklwd((__v4hi)__A, (__v4hi)0LL);
  __hisi = (__v2si)__builtin_ia32_punpckhwd((__v4hi)__A, (__v4hi)0LL);
  __zero = (__v4sf)_mm_setzero_ps();
  __ra = __builtin_ia32_cvtpi2ps(__zero, __losi);
  __rb = __builtin_ia32_cvtpi2ps(__ra, __hisi);
  return (__m128)__builtin_ia32_movlhps(__ra, __rb);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cvtpi8_ps(__m64 __A) {
  __v8qi __sign;

  __sign = __builtin_ia32_pcmpgtb((__v8qi)0LL, (__v8qi)__A);

  __A = (__m64)__builtin_ia32_punpcklbw((__v8qi)__A, __sign);

  return _mm_cvtpi16_ps(__A);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cvtpu8_ps(__m64 __A) {
  __A = (__m64)__builtin_ia32_punpcklbw((__v8qi)__A, (__v8qi)0LL);
  return _mm_cvtpu16_ps(__A);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cvtpi32x2_ps(__m64 __A, __m64 __B) {
  __v4sf __zero = (__v4sf)_mm_setzero_ps();
  __v4sf __sfa = __builtin_ia32_cvtpi2ps(__zero, (__v2si)__A);
  __v4sf __sfb = __builtin_ia32_cvtpi2ps(__sfa, (__v2si)__B);
  return (__m128)__builtin_ia32_movlhps(__sfa, __sfb);
}

extern __inline __m64
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cvtps_pi16(__m128 __A) {
  __v4sf __hisf = (__v4sf)__A;
  __v4sf __losf = __builtin_ia32_movhlps(__hisf, __hisf);
  __v2si __hisi = __builtin_ia32_cvtps2pi(__hisf);
  __v2si __losi = __builtin_ia32_cvtps2pi(__losf);
  return (__m64)__builtin_ia32_packssdw(__hisi, __losi);
}

extern __inline __m64
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cvtps_pi8(__m128 __A) {
  __v4hi __tmp = (__v4hi)_mm_cvtps_pi16(__A);
  return (__m64)__builtin_ia32_packsswb(__tmp, (__v4hi)0LL);
}

#ifdef __OPTIMIZE__
extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_shuffle_ps(__m128 __A, __m128 __B, int const __mask) {
  return (__m128)__builtin_ia32_shufps((__v4sf)__A, (__v4sf)__B, __mask);
}
#else
#define _mm_shuffle_ps(A, B, MASK)                                         \
  ((__m128)__builtin_ia32_shufps((__v4sf)(__m128)(A), (__v4sf)(__m128)(B), \
                                 (int)(MASK)))
#endif

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_unpackhi_ps(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_unpckhps((__v4sf)__A, (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_unpacklo_ps(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_unpcklps((__v4sf)__A, (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_loadh_pi(__m128 __A, __m64 const *__P) {
  return (__m128)__builtin_ia32_loadhps((__v4sf)__A, (const __v2sf *)__P);
}

extern __inline void
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_storeh_pi(__m64 *__P, __m128 __A) {
  __builtin_ia32_storehps((__v2sf *)__P, (__v4sf)__A);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_movehl_ps(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_movhlps((__v4sf)__A, (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_movelh_ps(__m128 __A, __m128 __B) {
  return (__m128)__builtin_ia32_movlhps((__v4sf)__A, (__v4sf)__B);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_loadl_pi(__m128 __A, __m64 const *__P) {
  return (__m128)__builtin_ia32_loadlps((__v4sf)__A, (const __v2sf *)__P);
}

extern __inline void
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_storel_pi(__m64 *__P, __m128 __A) {
  __builtin_ia32_storelps((__v2sf *)__P, (__v4sf)__A);
}

extern __inline int
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_movemask_ps(__m128 __A) {
  return __builtin_ia32_movmskps((__v4sf)__A);
}

extern __inline unsigned int
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_getcsr(void) {
  return __builtin_ia32_stmxcsr();
}

extern __inline unsigned int
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _MM_GET_EXCEPTION_STATE(void) {
  return _mm_getcsr() & _MM_EXCEPT_MASK;
}

extern __inline unsigned int
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _MM_GET_EXCEPTION_MASK(void) {
  return _mm_getcsr() & _MM_MASK_MASK;
}

extern __inline unsigned int
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _MM_GET_ROUNDING_MODE(void) {
  return _mm_getcsr() & _MM_ROUND_MASK;
}

extern __inline unsigned int
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _MM_GET_FLUSH_ZERO_MODE(void) {
  return _mm_getcsr() & _MM_FLUSH_ZERO_MASK;
}

extern __inline void
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_setcsr(unsigned int __I) {
  __builtin_ia32_ldmxcsr(__I);
}

extern __inline void
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _MM_SET_EXCEPTION_STATE(unsigned int __mask) {
  _mm_setcsr((_mm_getcsr() & ~_MM_EXCEPT_MASK) | __mask);
}

extern __inline void
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _MM_SET_EXCEPTION_MASK(unsigned int __mask) {
  _mm_setcsr((_mm_getcsr() & ~_MM_MASK_MASK) | __mask);
}

extern __inline void
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _MM_SET_ROUNDING_MODE(unsigned int __mode) {
  _mm_setcsr((_mm_getcsr() & ~_MM_ROUND_MASK) | __mode);
}

extern __inline void
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _MM_SET_FLUSH_ZERO_MODE(unsigned int __mode) {
  _mm_setcsr((_mm_getcsr() & ~_MM_FLUSH_ZERO_MASK) | __mode);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_set_ss(float __F) {
  return __extension__(__m128)(__v4sf){__F, 0.0f, 0.0f, 0.0f};
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_set1_ps(float __F) {
  return __extension__(__m128)(__v4sf){__F, __F, __F, __F};
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_set_ps1(float __F) {
  return _mm_set1_ps(__F);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_load_ss(float const *__P) {
  return _mm_set_ss(*__P);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_load1_ps(float const *__P) {
  return _mm_set1_ps(*__P);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_load_ps1(float const *__P) {
  return _mm_load1_ps(__P);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_load_ps(float const *__P) {
  return *(__m128 *)__P;
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_loadu_ps(float const *__P) {
  return *(__m128_u *)__P;
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_loadr_ps(float const *__P) {
  __v4sf __tmp = *(__v4sf *)__P;
  return (__m128)__builtin_ia32_shufps(__tmp, __tmp, _MM_SHUFFLE(0, 1, 2, 3));
}

extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__,
                                      __artificial__))
_mm_set_ps(const float __Z, const float __Y, const float __X, const float __W) {
  return __extension__(__m128)(__v4sf){__W, __X, __Y, __Z};
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_setr_ps(float __Z, float __Y, float __X, float __W) {
  return __extension__(__m128)(__v4sf){__Z, __Y, __X, __W};
}

extern __inline void
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_store_ss(float *__P, __m128 __A) {
  *__P = ((__v4sf)__A)[0];
}

extern __inline float
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_cvtss_f32(__m128 __A) {
  return ((__v4sf)__A)[0];
}

extern __inline void
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_store_ps(float *__P, __m128 __A) {
  *(__m128 *)__P = __A;
}

extern __inline void
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_storeu_ps(float *__P, __m128 __A) {
  *(__m128_u *)__P = __A;
}

extern __inline void
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_store1_ps(float *__P, __m128 __A) {
  __v4sf __va = (__v4sf)__A;
  __v4sf __tmp = __builtin_ia32_shufps(__va, __va, _MM_SHUFFLE(0, 0, 0, 0));
  _mm_storeu_ps(__P, __tmp);
}

extern __inline void
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_store_ps1(float *__P, __m128 __A) {
  _mm_store1_ps(__P, __A);
}

extern __inline void
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_storer_ps(float *__P, __m128 __A) {
  __v4sf __va = (__v4sf)__A;
  __v4sf __tmp = __builtin_ia32_shufps(__va, __va, _MM_SHUFFLE(0, 1, 2, 3));
  _mm_store_ps(__P, __tmp);
}

extern __inline __m128
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_move_ss(__m128 __A, __m128 __B) {
  return (__m128)__builtin_shuffle(
      (__v4sf)__A, (__v4sf)__B,
      __extension__(__attribute__((__vector_size__(16))) int){4, 1, 2, 3});
}

#ifdef __OPTIMIZE__
extern __inline int
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_extract_pi16(__m64 const __A, int const __N) {
  return __builtin_ia32_vec_ext_v4hi((__v4hi)__A, __N);
}

extern __inline int
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _m_pextrw(__m64 const __A, int const __N) {
  return _mm_extract_pi16(__A, __N);
}
#else
#define _mm_extract_pi16(A, N) \
  ((int)__builtin_ia32_vec_ext_v4hi((__v4hi)(__m64)(A), (int)(N)))

#define _m_pextrw(A, N) _mm_extract_pi16(A, N)
#endif

#ifdef __OPTIMIZE__
extern __inline __m64
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_insert_pi16(__m64 const __A, int const __D, int const __N) {
  return (__m64)__builtin_ia32_vec_set_v4hi((__v4hi)__A, __D, __N);
}

extern __inline __m64
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _m_pinsrw(__m64 const __A, int const __D, int const __N) {
  return _mm_insert_pi16(__A, __D, __N);
}
#else
#define _mm_insert_pi16(A, D, N) \
  ((__m64)__builtin_ia32_vec_set_v4hi((__v4hi)(__m64)(A), (int)(D), (int)(N)))

#define _m_pinsrw(A, D, N) _mm_insert_pi16(A, D, N)
#endif

extern __inline __m64
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_max_pi16(__m64 __A, __m64 __B) {
  return (__m64)__builtin_ia32_pmaxsw((__v4hi)__A, (__v4hi)__B);
}

extern __inline __m64
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _m_pmaxsw(__m64 __A, __m64 __B) {
  return _mm_max_pi16(__A, __B);
}

extern __inline __m64
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_max_pu8(__m64 __A, __m64 __B) {
  return (__m64)__builtin_ia32_pmaxub((__v8qi)__A, (__v8qi)__B);
}

extern __inline __m64
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _m_pmaxub(__m64 __A, __m64 __B) {
  return _mm_max_pu8(__A, __B);
}

extern __inline __m64
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_min_pi16(__m64 __A, __m64 __B) {
  return (__m64)__builtin_ia32_pminsw((__v4hi)__A, (__v4hi)__B);
}

extern __inline __m64
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _m_pminsw(__m64 __A, __m64 __B) {
  return _mm_min_pi16(__A, __B);
}

extern __inline __m64
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_min_pu8(__m64 __A, __m64 __B) {
  return (__m64)__builtin_ia32_pminub((__v8qi)__A, (__v8qi)__B);
}

extern __inline __m64
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _m_pminub(__m64 __A, __m64 __B) {
  return _mm_min_pu8(__A, __B);
}

extern __inline int
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_movemask_pi8(__m64 __A) {
  return __builtin_ia32_pmovmskb((__v8qi)__A);
}

extern __inline int
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _m_pmovmskb(__m64 __A) {
  return _mm_movemask_pi8(__A);
}

extern __inline __m64
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_mulhi_pu16(__m64 __A, __m64 __B) {
  return (__m64)__builtin_ia32_pmulhuw((__v4hi)__A, (__v4hi)__B);
}

extern __inline __m64
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _m_pmulhuw(__m64 __A, __m64 __B) {
  return _mm_mulhi_pu16(__A, __B);
}

#ifdef __OPTIMIZE__
extern __inline __m64
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_shuffle_pi16(__m64 __A, int const __N) {
  return (__m64)__builtin_ia32_pshufw((__v4hi)__A, __N);
}

extern __inline __m64
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _m_pshufw(__m64 __A, int const __N) {
  return _mm_shuffle_pi16(__A, __N);
}
#else
#define _mm_shuffle_pi16(A, N) \
  ((__m64)__builtin_ia32_pshufw((__v4hi)(__m64)(A), (int)(N)))

#define _m_pshufw(A, N) _mm_shuffle_pi16(A, N)
#endif

extern __inline void
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_maskmove_si64(__m64 __A, __m64 __N, char *__P) {
  __builtin_ia32_maskmovq((__v8qi)__A, (__v8qi)__N, __P);
}

extern __inline void
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _m_maskmovq(__m64 __A, __m64 __N, char *__P) {
  _mm_maskmove_si64(__A, __N, __P);
}

extern __inline __m64
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_avg_pu8(__m64 __A, __m64 __B) {
  return (__m64)__builtin_ia32_pavgb((__v8qi)__A, (__v8qi)__B);
}

extern __inline __m64
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _m_pavgb(__m64 __A, __m64 __B) {
  return _mm_avg_pu8(__A, __B);
}

extern __inline __m64
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_avg_pu16(__m64 __A, __m64 __B) {
  return (__m64)__builtin_ia32_pavgw((__v4hi)__A, (__v4hi)__B);
}

extern __inline __m64
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _m_pavgw(__m64 __A, __m64 __B) {
  return _mm_avg_pu16(__A, __B);
}

extern __inline __m64
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_sad_pu8(__m64 __A, __m64 __B) {
  return (__m64)__builtin_ia32_psadbw((__v8qi)__A, (__v8qi)__B);
}

extern __inline __m64
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _m_psadbw(__m64 __A, __m64 __B) {
  return _mm_sad_pu8(__A, __B);
}

extern __inline void
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_stream_pi(__m64 *__P, __m64 __A) {
  __builtin_ia32_movntq((unsigned long long *)__P, (unsigned long long)__A);
}

extern __inline void
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_stream_ps(float *__P, __m128 __A) {
  __builtin_ia32_movntps(__P, (__v4sf)__A);
}

extern __inline void
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_sfence(void) {
  __builtin_ia32_sfence();
}

#define _MM_TRANSPOSE4_PS(row0, row1, row2, row3)                      \
  do {                                                                 \
    __v4sf __r0 = (row0), __r1 = (row1), __r2 = (row2), __r3 = (row3); \
    __v4sf __t0 = __builtin_ia32_unpcklps(__r0, __r1);                 \
    __v4sf __t1 = __builtin_ia32_unpcklps(__r2, __r3);                 \
    __v4sf __t2 = __builtin_ia32_unpckhps(__r0, __r1);                 \
    __v4sf __t3 = __builtin_ia32_unpckhps(__r2, __r3);                 \
    (row0) = __builtin_ia32_movlhps(__t0, __t1);                       \
    (row1) = __builtin_ia32_movhlps(__t1, __t0);                       \
    (row2) = __builtin_ia32_movlhps(__t2, __t3);                       \
    (row3) = __builtin_ia32_movhlps(__t3, __t2);                       \
  } while (0)

#include "third_party/intel/emmintrin.internal.h"

#ifdef __DISABLE_SSE__
#undef __DISABLE_SSE__
#pragma GCC pop_options
#endif /* __DISABLE_SSE__ */

extern __inline void
    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
    _mm_pause(void) {
  __builtin_ia32_pause();
}

#endif /* __x86_64__ */
#endif /* _XMMINTRIN_H_INCLUDED */