path: root/lib/Headers/xmmintrin.h

/*===---- xmmintrin.h - SSE intrinsics -------------------------------------===
 *
 * Permission is hereby granted, free of charge, to any person obtaining a copy
 * of this software and associated documentation files (the "Software"), to deal
 * in the Software without restriction, including without limitation the rights
 * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
 * copies of the Software, and to permit persons to whom the Software is
 * furnished to do so, subject to the following conditions:
 *
 * The above copyright notice and this permission notice shall be included in
 * all copies or substantial portions of the Software.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
 * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
 * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
 * THE SOFTWARE.
 *
 *===-----------------------------------------------------------------------===
 */
 
#ifndef __XMMINTRIN_H
#define __XMMINTRIN_H
 
#ifndef __SSE__
#error "SSE instruction set not enabled"
#else

#include <mmintrin.h>

typedef int __v4si __attribute__((__vector_size__(16)));
typedef float __v4sf __attribute__((__vector_size__(16)));
typedef float __m128 __attribute__((__vector_size__(16)));

// This header should only be included in a hosted environment as it depends on
// a standard library to provide allocation routines.
#if __STDC_HOSTED__
#include <mm_malloc.h>
#endif

static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
_mm_add_ss(__m128 __a, __m128 __b)
{
  __a[0] += __b[0];
  return __a;
}

static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
_mm_add_ps(__m128 __a, __m128 __b)
{
  return __a + __b;
}

static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
_mm_sub_ss(__m128 __a, __m128 __b)
{
  __a[0] -= __b[0];
  return __a;
}

static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
_mm_sub_ps(__m128 __a, __m128 __b)
{
  return __a - __b;
}

static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
_mm_mul_ss(__m128 __a, __m128 __b)
{
  __a[0] *= __b[0];
  return __a;
}

static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
_mm_mul_ps(__m128 __a, __m128 __b)
{
  return __a * __b;
}

static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
_mm_div_ss(__m128 __a, __m128 __b)
{
  __a[0] /= __b[0];
  return __a;
}

static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
_mm_div_ps(__m128 __a, __m128 __b)
{
  return __a / __b;
}

static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
_mm_sqrt_ss(__m128 __a)
{
  __m128 __c = __builtin_ia32_sqrtss(__a);
  return (__m128) { __c[0], __a[1], __a[2], __a[3] };
}

static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
_mm_sqrt_ps(__m128 __a)
{
  return __builtin_ia32_sqrtps(__a);
}

static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
_mm_rcp_ss(__m128 __a)
{
  __m128 __c = __builtin_ia32_rcpss(__a);
  return (__m128) { __c[0], __a[1], __a[2], __a[3] };
}

static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
_mm_rcp_ps(__m128 __a)
{
  return __builtin_ia32_rcpps(__a);
}

static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
_mm_rsqrt_ss(__m128 __a)
{
  __m128 __c = __builtin_ia32_rsqrtss(__a);
  return (__m128) { __c[0], __a[1], __a[2], __a[3] };
}

static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
_mm_rsqrt_ps(__m128 __a)
{
  return __builtin_ia32_rsqrtps(__a);
}

static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
_mm_min_ss(__m128 __a, __m128 __b)
{
  return __builtin_ia32_minss(__a, __b);
}

static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
_mm_min_ps(__m128 __a, __m128 __b)
{
  return __builtin_ia32_minps(__a, __b);
}

static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
_mm_max_ss(__m128 __a, __m128 __b)
{
  return __builtin_ia32_maxss(__a, __b);
}

static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
_mm_max_ps(__m128 __a, __m128 __b)
{
  return __builtin_ia32_maxps(__a, __b);
}

static __inline__ __m128 __attribute__((__always_inline__,