path: root/gcc-4.9/gcc/config/aarch64/arm_neon.h

/* ARM NEON intrinsics include file.

   Copyright (C) 2011-2014 Free Software Foundation, Inc.
   Contributed by ARM Ltd.

   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/>.  */

#ifndef _AARCH64_NEON_H_
#define _AARCH64_NEON_H_

#include <stdint.h>

#define __AARCH64_UINT64_C(__C) ((uint64_t) __C)
#define __AARCH64_INT64_C(__C) ((int64_t) __C)

typedef __builtin_aarch64_simd_qi int8x8_t
  __attribute__ ((__vector_size__ (8)));
typedef __builtin_aarch64_simd_hi int16x4_t
  __attribute__ ((__vector_size__ (8)));
typedef __builtin_aarch64_simd_si int32x2_t
  __attribute__ ((__vector_size__ (8)));
typedef int64_t int64x1_t;
typedef double float64x1_t;
typedef __builtin_aarch64_simd_sf float32x2_t
  __attribute__ ((__vector_size__ (8)));
typedef __builtin_aarch64_simd_poly8 poly8x8_t
  __attribute__ ((__vector_size__ (8)));
typedef __builtin_aarch64_simd_poly16 poly16x4_t
  __attribute__ ((__vector_size__ (8)));
typedef __builtin_aarch64_simd_uqi uint8x8_t
  __attribute__ ((__vector_size__ (8)));
typedef __builtin_aarch64_simd_uhi uint16x4_t
  __attribute__ ((__vector_size__ (8)));
typedef __builtin_aarch64_simd_usi uint32x2_t
  __attribute__ ((__vector_size__ (8)));
typedef uint64_t uint64x1_t;
typedef __builtin_aarch64_simd_qi int8x16_t
  __attribute__ ((__vector_size__ (16)));
typedef __builtin_aarch64_simd_hi int16x8_t
  __attribute__ ((__vector_size__ (16)));
typedef __builtin_aarch64_simd_si int32x4_t
  __attribute__ ((__vector_size__ (16)));
typedef __builtin_aarch64_simd_di int64x2_t
  __attribute__ ((__vector_size__ (16)));
typedef __builtin_aarch64_simd_sf float32x4_t
  __attribute__ ((__vector_size__ (16)));
typedef __builtin_aarch64_simd_df float64x2_t
  __attribute__ ((__vector_size__ (16)));
typedef __builtin_aarch64_simd_poly8 poly8x16_t
  __attribute__ ((__vector_size__ (16)));
typedef __builtin_aarch64_simd_poly16 poly16x8_t
  __attribute__ ((__vector_size__ (16)));
typedef __builtin_aarch64_simd_poly64 poly64x2_t
  __attribute__ ((__vector_size__ (16)));
typedef __builtin_aarch64_simd_uqi uint8x16_t
  __attribute__ ((__vector_size__ (16)));
typedef __builtin_aarch64_simd_uhi uint16x8_t
  __attribute__ ((__vector_size__ (16)));
typedef __builtin_aarch64_simd_usi uint32x4_t
  __attribute__ ((__vector_size__ (16)));
typedef __builtin_aarch64_simd_udi uint64x2_t
  __attribute__ ((__vector_size__ (16)));

typedef float float32_t;
typedef double float64_t;
typedef __builtin_aarch64_simd_poly8 poly8_t;
typedef __builtin_aarch64_simd_poly16 poly16_t;
typedef __builtin_aarch64_simd_poly64 poly64_t;
typedef __builtin_aarch64_simd_poly128 poly128_t;

typedef struct int8x8x2_t
{
  int8x8_t val[2];
} int8x8x2_t;

typedef struct int8x16x2_t
{
  int8x16_t val[2];
} int8x16x2_t;

typedef struct int16x4x2_t
{
  int16x4_t val[2];
} int16x4x2_t;

typedef struct int16x8x2_t
{
  int16x8_t val[2];
} int16x8x2_t;

typedef struct int32x2x2_t
{
  int32x2_t val[2];
} int32x2x2_t;

typedef struct int32x4x2_t
{
  int32x4_t val[2];
} int32x4x2_t;

typedef struct int64x1x2_t
{
  int64x1_t val[2];
} int64x1x2_t;

typedef struct int64x2x2_t
{
  int64x2_t val[2];
} int64x2x2_t;

typedef struct uint8x8x2_t
{
  uint8x8_t val[2];
} uint8x8x2_t;

typedef struct uint8x16x2_t
{
  uint8x16_t val[2];
} uint8x16x2_t;

typedef struct uint16x4x2_t
{
  uint16x4_t val[2];
} uint16x4x2_t;

typedef struct uint16x8x2_t
{
  uint16x8_t val[2];
} uint16x8x2_t;

typedef struct uint32x2x2_t
{
  uint32x2_t val[2];
} uint32x2x2_t;

typedef struct uint32x4x2_t
{
  uint32x4_t val[2];
} uint32x4x2_t;

typedef struct uint64x1x2_t
{
  uint64x1_t val[2];
} uint64x1x2_t;

typedef struct uint64x2x2_t
{
  uint64x2_t val[2];
} uint64x2x2_t;

typedef struct float32x2x2_t
{
  float32x2_t val[2];
} float32x2x2_t;

typedef struct float32x4x2_t
{
  float32x4_t val[2];
} float32x4x2_t;

typedef struct float64x2x2_t
{
  float64x2_t val[2];
} float64x2x2_t;

typedef struct float64x1x2_t
{
  float64x1_t val[2];
} float64x1x2_t;

typedef struct poly8x8x2_t
{
  poly8x8_t val[2];
} poly8x8x2_t;

typedef struct poly8x16x2_t
{
  poly8x16_t val[2];
} poly8x16x2_t;

typedef struct poly16x4x2_t
{
  poly16x4_t val[2];
} poly16x4x2_t;

typedef struct poly16x8x2_t
{
  poly16x8_t val[2];
} poly16x8x2_t;

typedef struct int8x8x3_t
{
  int8x8_t val[3];
} int8x8x3_t;

typedef struct int8x16x3_t
{
  int8x16_t val[3];
} int8x16x3_t;

typedef struct int16x4x3_t
{
  int16x4_t val[3];
} int16x4x3_t;

typedef struct int16x8x3_t
{
  int16x8_t val[3];
} int16x8x3_t;

typedef struct int32x2x3_t
{
  int32x2_t val[3];
} int32x2x3_t;

typedef struct int32x4x3_t
{
  int32x4_t val[3];
} int32x4x3_t;

typedef struct int64x1x3_t
{
  int64x1_t val[3];
} int64x1x3_t;

typedef struct int64x2x3_t
{
  int64x2_t val[3];
} int64x2x3_t;

typedef struct uint8x8x3_t
{
  uint8x8_t val[3];
} uint8x8x3_t;

typedef struct uint8x16x3_t
{
  uint8x16_t val[3];
} uint8x16x3_t;

typedef struct uint16x4x3_t
{
  uint16x4_t val[3];
} uint16x4x3_t;

typedef struct uint16x8x3_t
{
  uint16x8_t val[3];
} uint16x8x3_t;

typedef struct uint32x2x3_t
{
  uint32x2_t val[3];
} uint32x2x3_t;

typedef struct uint32x4x3_t
{
  uint32x4_t val[3];
} uint32x4x3_t;

typedef struct uint64x1x3_t
{
  uint64x1_t val[3];
} uint64x1x3_t;

typedef struct uint64x2x3_t
{
  uint64x2_t val[3];
} uint64x2x3_t;

typedef struct float32x2x3_t
{
  float32x2_t val[3];
} float32x2x3_t;

typedef struct float32x4x3_t
{
  float32x4_t val[3];
} float32x4x3_t;

typedef struct float64x2x3_t
{
  float64x2_t val[3];
} float64x2x3_t;

typedef struct float64x1x3_t
{
  float64x1_t val[3];
} float64x1x3_t;

typedef struct poly8x8x3_t
{
  poly8x8_t val[3];
} poly8x8x3_t;

typedef struct poly8x16x3_t
{
  poly8x16_t val[3];
} poly8x16x3_t;

typedef struct poly16x4x3_t
{
  poly16x4_t val[3];
} poly16x4x3_t;

typedef struct poly16x8x3_t
{
  poly16x8_t val[3];
} poly16x8x3_t;

typedef struct int8x8x4_t
{
  int8x8_t val[4];
} int8x8x4_t;

typedef struct int8x16x4_t
{
  int8x16_t val[4];
} int8x16x4_t;

typedef struct int16x4x4_t
{
  int16x4_t val[4];
} int16x4x4_t;

typedef struct int16x8x4_t
{
  int16x8_t val[4];
} int16x8x4_t;

typedef struct int32x2x4_t
{
  int32x2_t val[4];
} int32x2x4_t;

typedef struct int32x4x4_t
{
  int32x4_t val[4];
} int32x4x4_t;

typedef struct int64x1x4_t
{
  int64x1_t val[4];
} int64x1x4_t;

typedef struct int64x2x4_t
{
  int64x2_t val[4];
} int64x2x4_t;

typedef struct uint8x8x4_t
{
  uint8x8_t val[4];
} uint8x8x4_t;

typedef struct uint8x16x4_t
{
  uint8x16_t val[4];
} uint8x16x4_t;

typedef struct uint16x4x4_t
{
  uint16x4_t val[4];
} uint16x4x4_t;

typedef struct uint16x8x4_t
{
  uint16x8_t val[4];
} uint16x8x4_t;

typedef struct uint32x2x4_t
{
  uint32x2_t val[4];
} uint32x2x4_t;

typedef struct uint32x4x4_t
{
  uint32x4_t val[4];
} uint32x4x4_t;

typedef struct uint64x1x4_t
{
  uint64x1_t val[4];
} uint64x1x4_t;

typedef struct uint64x2x4_t
{
  uint64x2_t val[4];
} uint64x2x4_t;

typedef struct float32x2x4_t
{
  float32x2_t val[4];
} float32x2x4_t;

typedef struct float32x4x4_t
{
  float32x4_t val[4];
} float32x4x4_t;

typedef struct float64x2x4_t
{
  float64x2_t val[4];
} float64x2x4_t;

typedef struct float64x1x4_t
{
  float64x1_t val[4];
} float64x1x4_t;

typedef struct poly8x8x4_t
{
  poly8x8_t val[4];
} poly8x8x4_t;

typedef struct poly8x16x4_t
{
  poly8x16_t val[4];
} poly8x16x4_t;

typedef struct poly16x4x4_t
{
  poly16x4_t val[4];
} poly16x4x4_t;

typedef struct poly16x8x4_t
{
  poly16x8_t val[4];
} poly16x8x4_t;

/* vget_lane internal macros.  */

#define __aarch64_vget_lane_any(__size, __cast_ret, __cast_a, __a, __b) \
  (__cast_ret								\
     __builtin_aarch64_be_checked_get_lane##__size (__cast_a __a, __b))

#define __aarch64_vget_lane_f32(__a, __b) \
  __aarch64_vget_lane_any (v2sf, , , __a, __b)
#define __aarch64_vget_lane_f64(__a, __b) (__a)

#define __aarch64_vget_lane_p8(__a, __b) \
  __aarch64_vget_lane_any (v8qi, (poly8_t), (int8x8_t), __a, __b)
#define __aarch64_vget_lane_p16(__a, __b) \
  __aarch64_vget_lane_any (v4hi, (poly16_t), (int16x4_t), __a, __b)

#define __aarch64_vget_lane_s8(__a, __b) \
  __aarch64_vget_lane_any (v8qi, , ,__a, __b)
#define __aarch64_vget_lane_s16(__a, __b) \
  __aarch64_vget_lane_any (v4hi, , ,__a, __b)
#define __aarch64_vget_lane_s32(__a, __b) \
  __aarch64_vget_lane_any (v2si, , ,__a, __b)
#define __aarch64_vget_lane_s64(__a, __b) (__a)

#define __aarch64_vget_lane_u8(__a, __b) \
  __aarch64_vget_lane_any (v8qi, (uint8_t), (int8x8_t), __a, __b)
#define __aarch64_vget_lane_u16(__a, __b) \
  __aarch64_vget_lane_any (v4hi, (uint16_t), (int16x4_t), __a, __b)
#define __aarch64_vget_lane_u32(__a, __b) \
  __aarch64_vget_lane_any (v2si, (uint32_t), (int32x2_t), __a, __b)
#define __aarch64_vget_lane_u64(__a, __b) (__a)

#define __aarch64_vgetq_lane_f32(__a, __b) \
  __aarch64_vget_lane_any (v4sf, , , __a, __b)
#define __aarch64_vgetq_lane_f64(__a, __b) \
  __aarch64_vget_lane_any (v2df, , , __a, __b)

#define __aarch64_vgetq_lane_p8(__a, __b) \
  __aarch64_vget_lane_any (v16qi, (poly8_t), (int8x16_t), __a, __b)
#define __aarch64_vgetq_lane_p16(__a, __b) \
  __aarch64_vget_lane_any (v8hi, (poly16_t), (int16x8_t), __a, __b)

#define __aarch64_vgetq_lane_s8(__a, __b) \
  __aarch64_vget_lane_any (v16qi, , ,__a, __b)
#define __aarch64_vgetq_lane_s16(__a, __b) \
  __aarch64_vget_lane_any (v8hi, , ,__a, __b)
#define __aarch64_vgetq_lane_s32(__a, __b) \
  __aarch64_vget_lane_any (v4si, , ,__a, __b)
#define __aarch64_vgetq_lane_s64(__a, __b) \
  __aarch64_vget_lane_any (v2di, , ,__a, __b)

#define __aarch64_vgetq_lane_u8(__a, __b) \
  __aarch64_vget_lane_any (v16qi, (uint8_t), (int8x16_t), __a, __b)
#define __aarch64_vgetq_lane_u16(__a, __b) \
  __aarch64_vget_lane_any (v8hi, (uint16_t), (int16x8_t), __a, __b)
#define __aarch64_vgetq_lane_u32(__a, __b) \
  __aarch64_vget_lane_any (v4si, (uint32_t), (int32x4_t), __a, __b)
#define __aarch64_vgetq_lane_u64(__a, __b) \
  __aarch64_vget_lane_any (v2di, (uint64_t), (int64x2_t), __a, __b)

/* __aarch64_vdup_lane internal macros.  */
#define __aarch64_vdup_lane_any(__size, __q1, __q2, __a, __b) \
  vdup##__q1##_n_##__size (__aarch64_vget##__q2##_lane_##__size (__a, __b))

#define __aarch64_vdup_lane_f32(__a, __b) \
   __aarch64_vdup_lane_any (f32, , , __a, __b)
#define __aarch64_vdup_lane_f64(__a, __b) (__a)
#define __aarch64_vdup_lane_p8(__a, __b) \
   __aarch64_vdup_lane_any (p8, , , __a, __b)
#define __aarch64_vdup_lane_p16(__a, __b) \
   __aarch64_vdup_lane_any (p16, , , __a, __b)
#define __aarch64_vdup_lane_s8(__a, __b) \
   __aarch64_vdup_lane_any (s8, , , __a, __b)
#define __aarch64_vdup_lane_s16(__a, __b) \
   __aarch64_vdup_lane_any (s16, , , __a, __b)
#define __aarch64_vdup_lane_s32(__a, __b) \
   __aarch64_vdup_lane_any (s32, , , __a, __b)
#define __aarch64_vdup_lane_s64(__a, __b) (__a)
#define __aarch64_vdup_lane_u8(__a, __b) \
   __aarch64_vdup_lane_any (u8, , , __a, __b)
#define __aarch64_vdup_lane_u16(__a, __b) \
   __aarch64_vdup_lane_any (u16, , , __a, __b)
#define __aarch64_vdup_lane_u32(__a, __b) \
   __aarch64_vdup_lane_any (u32, , , __a, __b)
#define __aarch64_vdup_lane_u64(__a, __b) (__a)

/* __aarch64_vdup_laneq internal macros.  */
#define __aarch64_vdup_laneq_f32(__a, __b) \
   __aarch64_vdup_lane_any (f32, , q, __a, __b)
#define __aarch64_vdup_laneq_f64(__a, __b) \
   __aarch64_vdup_lane_any (f64, , q, __a, __b)
#define __aarch64_vdup_laneq_p8(__a, __b) \
   __aarch64_vdup_lane_any (p8, , q, __a, __b)
#define __aarch64_vdup_laneq_p16(__a, __b) \
   __aarch64_vdup_lane_any (p16, , q, __a, __b)
#define __aarch64_vdup_laneq_s8(__a, __b) \
   __aarch64_vdup_lane_any (s8, , q, __a, __b)
#define __aarch64_vdup_laneq_s16(__a, __b) \
   __aarch64_vdup_lane_any (s16, , q, __a, __b)
#define __aarch64_vdup_laneq_s32(__a, __b) \
   __aarch64_vdup_lane_any (s32, , q, __a, __b)
#define __aarch64_vdup_laneq_s64(__a, __b) \
   __aarch64_vdup_lane_any (s64, , q, __a, __b)
#define __aarch64_vdup_laneq_u8(__a, __b) \
   __aarch64_vdup_lane_any (u8, , q, __a, __b)
#define __aarch64_vdup_laneq_u16(__a, __b) \
   __aarch64_vdup_lane_any (u16, , q, __a, __b)
#define __aarch64_vdup_laneq_u32(__a, __b) \
   __aarch64_vdup_lane_any (u32, , q, __a, __b)
#define __aarch64_vdup_laneq_u64(__a, __b) \
   __aarch64_vdup_lane_any (u64, , q, __a, __b)

/* __aarch64_vdupq_lane internal macros.  */
#define __aarch64_vdupq_lane_f32(__a, __b) \
   __aarch64_vdup_lane_any (f32, q, , __a, __b)
#define __aarch64_vdupq_lane_f64(__a, __b) (vdupq_n_f64 (__a))
#define __aarch64_vdupq_lane_p8(__a, __b) \
   __aarch64_vdup_lane_any (p8, q, , __a, __b)
#define __aarch64_vdupq_lane_p16(__a, __b) \
   __aarch64_vdup_lane_any (p16, q, , __a, __b)
#define __aarch64_vdupq_lane_s8(__a, __b) \
   __aarch64_vdup_lane_any (s8, q, , __a, __b)
#define __aarch64_vdupq_lane_s16(__a, __b) \
   __aarch64_vdup_lane_any (s16, q, , __a, __b)
#define __aarch64_vdupq_lane_s32(__a, __b) \
   __aarch64_vdup_lane_any (s32, q, , __a, __b)
#define __aarch64_vdupq_lane_s64(__a, __b) (vdupq_n_s64 (__a))
#define __aarch64_vdupq_lane_u8(__a, __b) \
   __aarch64_vdup_lane_any (u8, q, , __a, __b)
#define __aarch64_vdupq_lane_u16(__a, __b) \
   __aarch64_vdup_lane_any (u16, q, , __a, __b)
#define __aarch64_vdupq_lane_u32(__a, __b) \
   __aarch64_vdup_lane_any (u32, q, , __a, __b)
#define __aarch64_vdupq_lane_u64(__a, __b) (vdupq_n_u64 (__a))

/* __aarch64_vdupq_laneq internal macros.  */
#define __aarch64_vdupq_laneq_f32(__a, __b) \
   __aarch64_vdup_lane_any (f32, q, q, __a, __b)
#define __aarch64_vdupq_laneq_f64(__a, __b) \
   __aarch64_vdup_lane_any (f64, q, q, __a, __b)
#define __aarch64_vdupq_laneq_p8(__a, __b) \
   __aarch64_vdup_lane_any (p8, q, q, __a, __b)
#define __aarch64_vdupq_laneq_p16(__a, __b) \
   __aarch64_vdup_lane_any (p16, q, q, __a, __b)
#define __aarch64_vdupq_laneq_s8(__a, __b) \
   __aarch64_vdup_lane_any (s8, q, q, __a, __b)
#define __aarch64_vdupq_laneq_s16(__a, __b) \
   __aarch64_vdup_lane_any (s16, q, q, __a, __b)
#define __aarch64_vdupq_laneq_s32(__a, __b) \
   __aarch64_vdup_lane_any (s32, q, q, __a, __b)
#define __aarch64_vdupq_laneq_s64(__a, __b) \
   __aarch64_vdup_lane_any (s64, q, q, __a, __b)
#define __aarch64_vdupq_laneq_u8(__a, __b) \
   __aarch64_vdup_lane_any (u8, q, q, __a, __b)
#define __aarch64_vdupq_laneq_u16(__a, __b) \
   __aarch64_vdup_lane_any (u16, q, q, __a, __b)
#define __aarch64_vdupq_laneq_u32(__a, __b) \
   __aarch64_vdup_lane_any (u32, q, q, __a, __b)
#define __aarch64_vdupq_laneq_u64(__a, __b) \
   __aarch64_vdup_lane_any (u64, q, q, __a, __b)

/* vadd  */
__extension__ static __inline int8x8_t __attribute__ ((__always_inline__))
vadd_s8 (int8x8_t __a, int8x8_t __b)
{
  return __a + __b;
}

__extension__ static __inline int16x4_t __attribute__ ((__always_inline__))
vadd_s16 (int16x4_t __a, int16x4_t __b)
{
  return __a + __b;
}

__extension__ static __inline int32x2_t __attribute__ ((__always_inline__))
vadd_s32 (int32x2_t __a, int32x2_t __b)
{
  return __a + __b;
}

__extension__ static __inline float32x2_t __attribute__ ((__always_inline__))
vadd_f32 (float32x2_t __a, float32x2_t __b)
{
  return __a + __b;
}

__extension__ static __inline float64x1_t __attribute__ ((__always_inline__))
vadd_f64 (float64x1_t __a, float64x1_t __b)
{
  return __a + __b;
}

__extension__ static __inline uint8x8_t __attribute__ ((__always_inline__))
vadd_u8 (uint8x8_t __a, uint8x8_t __b)
{
  return __a + __b;
}

__extension__ static __inline uint16x4_t __attribute__ ((__always_inline__))
vadd_u16 (uint16x4_t __a, uint16x4_t __b)
{
  return __a + __b;
}

__extension__ static __inline uint32x2_t __attribute__ ((__always_inline__))
vadd_u32 (uint32x2_t __a, uint32x2_t __b)
{
  return __a + __b;
}

__extension__ static __inline int64x1_t __attribute__ ((__always_inline__))
vadd_s64 (int64x1_t __a, int64x1_t __b)
{
  return __a + __b;
}

__extension__ static __inline uint64x1_t __attribute__ ((__always_inline__))
vadd_u64 (uint64x1_t __a, uint64x1_t __b)
{
  return __a + __b;
}

__extension__ static __inline int8x16_t __attribute__ ((__always_inline__))
vaddq_s8 (int8x16_t __a, int8x16_t __b)
{
  return __a + __b;
}

__extension__ static __inline int16x8_t __attribute__ ((__always_inline__))
vaddq_s16 (int16x8_t __a, int16x8_t __b)
{
  return __a + __b;
}

__extension__ static __inline int32x4_t __attribute__ ((__always_inline__))
vaddq_s32 (int32x4_t __a, int32x4_t __b)
{
  return __a + __b;
}

__extension__ static __inline int64x2_t __attribute__ ((__always_inline__))
vaddq_s64 (int64x2_t __a, int64x2_t __b)
{
  return __a + __b;
}

__extension__ static __inline float32x4_t __attribute__ ((__always_inline__))
vaddq_f32 (float32x4_t __a, float32x4_t __b)
{
  return __a + __b;
}

__extension__ static __inline float64x2_t __attribute__ ((__always_inline__))
vaddq_f64 (float64x2_t __a, float64x2_t __b)
{
  return __a + __b;
}

__extension__ static __inline uint8x16_t __attribute__ ((__always_inline__))
vaddq_u8 (uint8x16_t __a, uint8x16_t __b)
{
  return __a + __b;
}

__extension__ static __inline uint16x8_t __attribute__ ((__always_inline__))
vaddq_u16 (uint16x8_t __a, uint16x8_t __b)
{
  return __a + __b;
}

__extension__ static __inline uint32x4_t __attribute__ ((__always_inline__))
vaddq_u32 (uint32x4_t __a, uint32x4_t __b)
{
  return __a + __b;
}

__extension__ static __inline uint64x2_t __attribute__ ((__always_inline__))
vaddq_u64 (uint64x2_t __a, uint64x2_t __b)
{
  return __a + __b;
}

__extension__ static __inline int16x8_t __attribute__ ((__always_inline__))
vaddl_s8 (int8x8_t __a, int8x8_t __b)
{
  return (int16x8_t) __builtin_aarch64_saddlv8qi (__a, __b);
}

__extension__ static __inline int32x4_t __attribute__ ((__always_inline__))
vaddl_s16 (int16x4_t __a, int16x4_t __b)
{
  return (int32x4_t) __builtin_aarch64_saddlv4hi (__a, __b);
}

__extension__ static __inline int64x2_t __attribute__ ((__always_inline__))
vaddl_s32 (int32x2_t __a, int32x2_t __b)
{
  return (int64x2_t) __builtin_aarch64_saddlv2si (__a, __b);
}

__extension__ static __inline uint16x8_t __attribute__ ((__always_inline__))
vaddl_u8 (uint8x8_t __a, uint8x8_t __b)
{
  return (uint16x8_t) __builtin_aarch64_uaddlv8qi ((int8x8_t) __a,
						   (int8x8_t) __b);
}

__extension__ static __inline uint32x4_t __attribute__ ((__always_inline__))
vaddl_u16 (uint16x4_t __a, uint16x4_t __b)
{
  return (uint32x4_t) __builtin_aarch64_uaddlv4hi ((int16x4_t) __a,
						   (int16x4_t) __b);
}

__extension__ static __inline uint64x2_t __attribute__ ((__always_inline__))
vaddl_u32 (uint32x2_t __a, uint32x2_t __b)
{
  return (uint64x2_t) __builtin_aarch64_uaddlv2si ((int32x2_t) __a,
						   (int32x2_t) __b);
}

__extension__ static __inline int16x8_t __attribute__ ((__always_inline__))
vaddl_high_s8 (int8x16_t __a, int8x16_t __b)
{
  return (int16x8_t) __builtin_aarch64_saddl2v16qi (__a, __b);
}

__extension__ static __inline int32x4_t __attribute__ ((__always_inline__))
vaddl_high_s16 (int16x8_t __a, int16x8_t __b)
{
  return (int32x4_t) __builtin_aarch64_saddl2v8hi (__a, __b);
}

__extension__ static __inline int64x2_t __attribute__ ((__always_inline__))
vaddl_high_s32 (int32x4_t __a, int32x4_t __b)
{
  return (int64x2_t) __builtin_aarch64_saddl2v4si (__a, __b);
}

__extension__ static __inline uint16x8_t __attribute__ ((__always_inline__))
vaddl_high_u8 (uint8x16_t __a, uint8x16_t __b)
{
  return (uint16x8_t) __builtin_aarch64_uaddl2v16qi ((int8x16_t) __a,
						     (int8x16_t) __b);
}

__extension__ static __inline uint32x4_t __attribute__ ((__always_inline__))
vaddl_high_u16 (uint16x8_t __a, uint16x8_t __b)
{
  return (uint32x4_t) __builtin_aarch64_uaddl2v8hi ((int16x8_t) __a,
						    (int16x8_t) __b);
}

__extension__ static __inline uint64x2_t __attribute__ ((__always_inline__))
vaddl_high_u32 (uint32x4_t __a, uint32x4_t __b)
{
  return (uint64x2_t) __builtin_aarch64_uaddl2v4si ((int32x4_t) __a,
						    (int32x4_t) __b);
}

__extensi