view libfaad2/fixed.h @ 19506:564163b2c84c

Remove useless define from MACH64 CFLAGS.
author diego
date Wed, 23 Aug 2006 10:35:55 +0000
parents 59b6fa5b4201
children e83eef58b30a
line wrap: on
line source

/*
** FAAD2 - Freeware Advanced Audio (AAC) Decoder including SBR decoding
** Copyright (C) 2003-2004 M. Bakker, Ahead Software AG, http://www.nero.com
**  
** This program 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 2 of the License, or
** (at your option) any later version.
** 
** This program 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.
** 
** You should have received a copy of the GNU General Public License
** along with this program; if not, write to the Free Software 
** Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
**
** Any non-GPL usage of this software or parts of this software is strictly
** forbidden.
**
** Commercial non-GPL licensing of this software is possible.
** For more info contact Ahead Software through Mpeg4AAClicense@nero.com.
**
** $Id: fixed.h,v 1.26 2004/09/04 14:56:28 menno Exp $
**/

#ifndef __FIXED_H__
#define __FIXED_H__

#ifdef __cplusplus
extern "C" {
#endif

#if defined(_WIN32_WCE) && defined(_ARM_)
#include <cmnintrin.h>
#endif

#define COEF_BITS 28
#define COEF_PRECISION (1 << COEF_BITS)
#define REAL_BITS 14 // MAXIMUM OF 14 FOR FIXED POINT SBR
#define REAL_PRECISION (1 << REAL_BITS)

/* FRAC is the fractional only part of the fixed point number [0.0..1.0) */
#define FRAC_SIZE 32 /* frac is a 32 bit integer */
#define FRAC_BITS 31
#define FRAC_PRECISION ((uint32_t)(1 << FRAC_BITS))
#define FRAC_MAX 0x7FFFFFFF

typedef int32_t real_t;


#define REAL_CONST(A) (((A) >= 0) ? ((real_t)((A)*(REAL_PRECISION)+0.5)) : ((real_t)((A)*(REAL_PRECISION)-0.5)))
#define COEF_CONST(A) (((A) >= 0) ? ((real_t)((A)*(COEF_PRECISION)+0.5)) : ((real_t)((A)*(COEF_PRECISION)-0.5)))
#define FRAC_CONST(A) (((A) == 1.00) ? ((real_t)FRAC_MAX) : (((A) >= 0) ? ((real_t)((A)*(FRAC_PRECISION)+0.5)) : ((real_t)((A)*(FRAC_PRECISION)-0.5))))
//#define FRAC_CONST(A) (((A) >= 0) ? ((real_t)((A)*(FRAC_PRECISION)+0.5)) : ((real_t)((A)*(FRAC_PRECISION)-0.5)))

#define Q2_BITS 22
#define Q2_PRECISION (1 << Q2_BITS)
#define Q2_CONST(A) (((A) >= 0) ? ((real_t)((A)*(Q2_PRECISION)+0.5)) : ((real_t)((A)*(Q2_PRECISION)-0.5)))

#if defined(_WIN32) && !defined(_WIN32_WCE)

/* multiply with real shift */
static INLINE real_t MUL_R(real_t A, real_t B)
{
    _asm {
        mov eax,A
        imul B
        shrd eax,edx,REAL_BITS
    }
}

/* multiply with coef shift */
static INLINE real_t MUL_C(real_t A, real_t B)
{
    _asm {
        mov eax,A
        imul B
        shrd eax,edx,COEF_BITS
    }
}

static INLINE real_t MUL_Q2(real_t A, real_t B)
{
    _asm {
        mov eax,A
        imul B
        shrd eax,edx,Q2_BITS
    }
}

static INLINE real_t MUL_SHIFT6(real_t A, real_t B)
{
    _asm {
        mov eax,A
        imul B
        shrd eax,edx,6
    }
}

static INLINE real_t MUL_SHIFT23(real_t A, real_t B)
{
    _asm {
        mov eax,A
        imul B
        shrd eax,edx,23
    }
}

#if 1
static INLINE real_t _MulHigh(real_t A, real_t B)
{
    _asm {
        mov eax,A
        imul B
        mov eax,edx
    }
}

/* multiply with fractional shift */
static INLINE real_t MUL_F(real_t A, real_t B)
{
    return _MulHigh(A,B) << (FRAC_SIZE-FRAC_BITS);
}

/* Complex multiplication */
static INLINE void ComplexMult(real_t *y1, real_t *y2,
    real_t x1, real_t x2, real_t c1, real_t c2)
{
    *y1 = (_MulHigh(x1, c1) + _MulHigh(x2, c2))<<(FRAC_SIZE-FRAC_BITS);
    *y2 = (_MulHigh(x2, c1) - _MulHigh(x1, c2))<<(FRAC_SIZE-FRAC_BITS);
}
#else
static INLINE real_t MUL_F(real_t A, real_t B)
{
    _asm {
        mov eax,A
        imul B
        shrd eax,edx,FRAC_BITS
    }
}

/* Complex multiplication */
static INLINE void ComplexMult(real_t *y1, real_t *y2,
    real_t x1, real_t x2, real_t c1, real_t c2)
{
    *y1 = MUL_F(x1, c1) + MUL_F(x2, c2);
    *y2 = MUL_F(x2, c1) - MUL_F(x1, c2);
}
#endif

#elif defined(__GNUC__) && defined (__arm__)

/* taken from MAD */
#define arm_mul(x, y, SCALEBITS) \
({ \
    uint32_t __hi; \
    uint32_t __lo; \
    uint32_t __result; \
    asm("smull  %0, %1, %3, %4\n\t" \
        "movs   %0, %0, lsr %5\n\t" \
        "adc    %2, %0, %1, lsl %6" \
        : "=&r" (__lo), "=&r" (__hi), "=r" (__result) \
        : "%r" (x), "r" (y), \
        "M" (SCALEBITS), "M" (32 - (SCALEBITS)) \
        : "cc"); \
        __result; \
})

static INLINE real_t MUL_R(real_t A, real_t B)
{
    return arm_mul(A, B, REAL_BITS);
}

static INLINE real_t MUL_C(real_t A, real_t B)
{
    return arm_mul(A, B, COEF_BITS);
}

static INLINE real_t MUL_Q2(real_t A, real_t B)
{
    return arm_mul(A, B, Q2_BITS);
}

static INLINE real_t MUL_SHIFT6(real_t A, real_t B)
{
    return arm_mul(A, B, 6);
}

static INLINE real_t MUL_SHIFT23(real_t A, real_t B)
{
    return arm_mul(A, B, 23);
}

static INLINE real_t _MulHigh(real_t x, real_t y)
{
    uint32_t __lo;
    uint32_t __hi;
    asm("smull\t%0, %1, %2, %3"
        : "=&r"(__lo),"=&r"(__hi)
        : "%r"(x),"r"(y)
        : "cc");
    return __hi;
}

static INLINE real_t MUL_F(real_t A, real_t B)
{
    return _MulHigh(A, B) << (FRAC_SIZE-FRAC_BITS);
}

/* Complex multiplication */
static INLINE void ComplexMult(real_t *y1, real_t *y2,
    real_t x1, real_t x2, real_t c1, real_t c2)
{
    int32_t tmp, yt1, yt2;
    asm("smull %0, %1, %4, %6\n\t"
        "smlal %0, %1, %5, %7\n\t"
        "rsb   %3, %4, #0\n\t"
        "smull %0, %2, %5, %6\n\t"
        "smlal %0, %2, %3, %7"
        : "=&r" (tmp), "=&r" (yt1), "=&r" (yt2), "=r" (x1)
        : "3" (x1), "r" (x2), "r" (c1), "r" (c2)
        : "cc" );
    *y1 = yt1 << (FRAC_SIZE-FRAC_BITS);
    *y2 = yt2 << (FRAC_SIZE-FRAC_BITS);
}

#else

  /* multiply with real shift */
  #define MUL_R(A,B) (real_t)(((int64_t)(A)*(int64_t)(B)+(1 << (REAL_BITS-1))) >> REAL_BITS)
  /* multiply with coef shift */
  #define MUL_C(A,B) (real_t)(((int64_t)(A)*(int64_t)(B)+(1 << (COEF_BITS-1))) >> COEF_BITS)
  /* multiply with fractional shift */
#if defined(_WIN32_WCE) && defined(_ARM_)
  /* eVC for PocketPC has an intrinsic function that returns only the high 32 bits of a 32x32 bit multiply */
  static INLINE real_t MUL_F(real_t A, real_t B)
  {
      return _MulHigh(A,B) << (32-FRAC_BITS);
  }
#else
  #define _MulHigh(A,B) (real_t)(((int64_t)(A)*(int64_t)(B)+(1 << (FRAC_SIZE-1))) >> FRAC_SIZE)
  #define MUL_F(A,B) (real_t)(((int64_t)(A)*(int64_t)(B)+(1 << (FRAC_BITS-1))) >> FRAC_BITS)
#endif
  #define MUL_Q2(A,B) (real_t)(((int64_t)(A)*(int64_t)(B)+(1 << (Q2_BITS-1))) >> Q2_BITS)
  #define MUL_SHIFT6(A,B) (real_t)(((int64_t)(A)*(int64_t)(B)+(1 << (6-1))) >> 6)
  #define MUL_SHIFT23(A,B) (real_t)(((int64_t)(A)*(int64_t)(B)+(1 << (23-1))) >> 23)

/* Complex multiplication */
static INLINE void ComplexMult(real_t *y1, real_t *y2,
    real_t x1, real_t x2, real_t c1, real_t c2)
{
    *y1 = (_MulHigh(x1, c1) + _MulHigh(x2, c2))<<(FRAC_SIZE-FRAC_BITS);
    *y2 = (_MulHigh(x2, c1) - _MulHigh(x1, c2))<<(FRAC_SIZE-FRAC_BITS);
}

#endif



#ifdef __cplusplus
}
#endif
#endif