rockbox/apps/codecs/libfaad/common.h
Andrew Mahone af5bad46d8 Libfaad pow2/log2 improvements:
Correct pow2_* to calculate correct output for input < 0.0 or between 0.0 and 1.0 instead of returning 0. This will change output, but the float version of this codec uses pow(2.0,x) in place of these functions, so this behavior was probably a bug
Replace 64-bit multiply in pow2_* with left or right shift by whole part of input, rounding if shifting right. An ARM-specific optimized version is provided as a C inline function, other targets use a C macro.
Increase constant size of table for pow2 to improve accuracy, also avoid early truncation in linear interpolation of result.
Move constant tables for pow2 and log2 to iram.

git-svn-id: svn://svn.rockbox.org/rockbox/trunk@23967 a1c6a512-1295-4272-9138-f99709370657
2009-12-13 03:45:40 +00:00

442 lines
8.8 KiB
C

/*
** 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.
**
** $Id$
**/
#ifndef __COMMON_H__
#define __COMMON_H__
#ifdef __cplusplus
extern "C" {
#endif
#include "faad_config.h"
#include "codeclib.h"
extern struct codec_api* ci;
#if defined(DEBUG) || defined(SIMULATOR)
#undef DEBUGF
#define DEBUGF ci->debugf
#else
#define DEBUGF(...)
#endif
#ifdef ROCKBOX_HAS_LOGF
#undef LOGF
#define LOGF ci->logf
#else
#define LOGF(...)
#endif
#define INLINE __inline
#if 0 //defined(_WIN32) && !defined(_WIN32_WCE)
#define ALIGN __declspec(align(16))
#else
#define ALIGN
#endif
#ifndef max
#define max(a, b) (((a) > (b)) ? (a) : (b))
#endif
#ifndef min
#define min(a, b) (((a) < (b)) ? (a) : (b))
#endif
/* COMPILE TIME DEFINITIONS */
/* use double precision */
/* #define USE_DOUBLE_PRECISION */
/* use fixed point reals */
#define FIXED_POINT
//#define BIG_IQ_TABLE
/* Use if target platform has address generators with autoincrement */
//#define PREFER_POINTERS
#ifdef _WIN32_WCE
#define FIXED_POINT
#endif
#define ERROR_RESILIENCE
/* Allow decoding of MAIN profile AAC */
#define MAIN_DEC
#if MEMORYSIZE > 2
/* Allow decoding of SSR profile AAC */
#define SSR_DEC
#endif
/* Allow decoding of LTP profile AAC */
#define LTP_DEC
/* Allow decoding of LD profile AAC */
#define LD_DEC
/* Allow decoding of scalable profiles */
//#define SCALABLE_DEC
/* Allow decoding of Digital Radio Mondiale (DRM) */
//#define DRM
//#define DRM_PS
/* LD can't do without LTP */
#ifdef LD_DEC
#ifndef ERROR_RESILIENCE
#define ERROR_RESILIENCE
#endif
#ifndef LTP_DEC
#define LTP_DEC
#endif
#endif
#define ALLOW_SMALL_FRAMELENGTH
// Define LC_ONLY_DECODER if you want a pure AAC LC decoder (independant of SBR_DEC and PS_DEC)
#define LC_ONLY_DECODER
#ifdef LC_ONLY_DECODER
#undef LD_DEC
#undef LTP_DEC
#undef MAIN_DEC
#undef SSR_DEC
#undef DRM
#undef ALLOW_SMALL_FRAMELENGTH
#undef ERROR_RESILIENCE
#endif
#if MEMORYSIZE > 2
#define SBR_DEC
//#define SBR_LOW_POWER
#define PS_DEC
#endif
/* FIXED POINT: No MAIN decoding */
#ifdef FIXED_POINT
# ifdef MAIN_DEC
# undef MAIN_DEC
# endif
#endif // FIXED_POINT
#ifdef DRM
# ifndef SCALABLE_DEC
# define SCALABLE_DEC
# endif
#endif
#ifdef FIXED_POINT
#define DIV_R(A, B) (((int64_t)A << REAL_BITS)/B)
#define DIV_C(A, B) (((int64_t)A << COEF_BITS)/B)
#else
#define DIV_R(A, B) ((A)/(B))
#define DIV_C(A, B) ((A)/(B))
#endif
#ifndef SBR_LOW_POWER
#define qmf_t complex_t
#define QMF_RE(A) RE(A)
#define QMF_IM(A) IM(A)
#else
#define qmf_t real_t
#define QMF_RE(A) (A)
#define QMF_IM(A)
#endif
/* END COMPILE TIME DEFINITIONS */
#if defined(_WIN32) && !defined(__MINGW32__)
#include <stdlib.h>
typedef unsigned __int64 uint64_t;
typedef unsigned __int32 uint32_t;
typedef unsigned __int16 uint16_t;
typedef unsigned __int8 uint8_t;
typedef __int64 int64_t;
typedef __int32 int32_t;
typedef __int16 int16_t;
typedef __int8 int8_t;
typedef float float32_t;
#else
#include <stdio.h>
#ifdef HAVE_SYS_TYPES_H
# include <sys/types.h>
#endif
#ifdef HAVE_SYS_STAT_H
# include <sys/stat.h>
#endif
#ifdef STDC_HEADERS
# include <stdlib.h>
# include <stddef.h>
#else
# ifdef HAVE_STDLIB_H
# include <stdlib.h>
# endif
#endif
#ifdef HAVE_STRING_H
# if !defined(STDC_HEADERS) && defined(HAVE_MEMORY_H)
# include <memory.h>
# endif
//# include <string.h>
#endif
#ifdef HAVE_STRINGS_H
//# include <strings.h>
#endif
#ifdef HAVE_INTTYPES_H
# include <inttypes.h>
#else
# ifdef HAVE_STDINT_H
# include <stdint.h>
# else
/* we need these... */
typedef unsigned long long uint64_t;
typedef unsigned long uint32_t;
typedef unsigned short uint16_t;
typedef unsigned char uint8_t;
typedef long long int64_t;
typedef long int32_t;
typedef short int16_t;
typedef char int8_t;
# endif
#endif
#ifdef HAVE_UNISTD_H
# include <unistd.h>
#endif
#ifndef HAVE_FLOAT32_T
typedef float float32_t;
#endif
#if STDC_HEADERS
//# include <string.h>
#else
# if !HAVE_STRCHR
# define strchr index
# define strrchr rindex
# endif
char *strchr(), *strrchr();
# if !HAVE_MEMCPY
# define memcpy(d, s, n) bcopy((s), (d), (n))
# define memmove(d, s, n) bcopy((s), (d), (n))
# endif
#endif
#endif
#ifdef WORDS_BIGENDIAN
#define ARCH_IS_BIG_ENDIAN
#endif
/* FIXED_POINT doesn't work with MAIN and SSR yet */
#ifdef FIXED_POINT
#undef MAIN_DEC
#undef SSR_DEC
#endif
#if defined(FIXED_POINT)
#include "fixed.h"
#elif defined(USE_DOUBLE_PRECISION)
typedef double real_t;
#include <math.h>
#define MUL_R(A,B) ((A)*(B))
#define MUL_C(A,B) ((A)*(B))
#define MUL_F(A,B) ((A)*(B))
/* 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);
}
#define REAL_CONST(A) ((real_t)(A))
#define COEF_CONST(A) ((real_t)(A))
#define Q2_CONST(A) ((real_t)(A))
#define FRAC_CONST(A) ((real_t)(A)) /* pure fractional part */
#else /* Normal floating point operation */
typedef float real_t;
#define MUL_R(A,B) ((A)*(B))
#define MUL_C(A,B) ((A)*(B))
#define MUL_F(A,B) ((A)*(B))
#define REAL_CONST(A) ((real_t)(A))
#define COEF_CONST(A) ((real_t)(A))
#define Q2_CONST(A) ((real_t)(A))
#define FRAC_CONST(A) ((real_t)(A)) /* pure fractional part */
/* 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);
}
#if defined(_WIN32) && !defined(__MINGW32__)
#define HAS_LRINTF
static INLINE int lrintf(float f)
{
int i;
__asm
{
fld f
fistp i
}
return i;
}
#elif (defined(__i386__) && defined(__GNUC__))
#define HAS_LRINTF
// from http://www.stereopsis.com/FPU.html
static INLINE int lrintf(float f)
{
int i;
__asm__ __volatile__ (
"flds %1 \n\t"
"fistpl %0 \n\t"
: "=m" (i)
: "m" (f));
return i;
}
#endif
#ifdef __ICL /* only Intel C compiler has fmath ??? */
#include <mathf.h>
#define sin sinf
#define cos cosf
#define log logf
#define floor floorf
#define ceil ceilf
#define sqrt sqrtf
#else
#ifdef HAVE_LRINTF
# define HAS_LRINTF
# define _ISOC9X_SOURCE 1
# define _ISOC99_SOURCE 1
# define __USE_ISOC9X 1
# define __USE_ISOC99 1
#endif
#include <math.h>
#ifdef HAVE_SINF
# define sin sinf
#error
#endif
#ifdef HAVE_COSF
# define cos cosf
#endif
#ifdef HAVE_LOGF
# define log logf
#endif
#ifdef HAVE_EXPF
# define exp expf
#endif
#ifdef HAVE_FLOORF
# define floor floorf
#endif
#ifdef HAVE_CEILF
# define ceil ceilf
#endif
#ifdef HAVE_SQRTF
# define sqrt sqrtf
#endif
#endif
#endif
#ifndef HAS_LRINTF
/* standard cast */
#define lrintf(f) ((int32_t)(f))
#endif
typedef real_t complex_t[2];
#define RE(A) A[0]
#define IM(A) A[1]
/* common functions */
uint8_t cpu_has_sse(void);
uint32_t random_int(void);
#define wl_min_lzc(x) bs_generic(x, BS_LOG2|BS_0_0)
#ifdef FIXED_POINT
#define LOG2_MIN_INF REAL_CONST(-10000)
int32_t log2_int(uint32_t val);
int32_t log2_fix(uint32_t val);
uint32_t pow2_int(real_t val);
uint32_t pow2_fix(real_t val);
#endif
uint8_t get_sr_index(const uint32_t samplerate);
uint8_t max_pred_sfb(const uint8_t sr_index);
uint8_t max_tns_sfb(const uint8_t sr_index, const uint8_t object_type,
const uint8_t is_short);
uint32_t get_sample_rate(const uint8_t sr_index);
int8_t can_decode_ot(const uint8_t object_type);
void *faad_malloc(size_t size);
void faad_free(void *b);
//#define PROFILE
#ifdef PROFILE
static int64_t faad_get_ts()
{
__asm
{
rdtsc
}
}
#endif
#ifndef M_PI
#define M_PI 3.14159265358979323846
#endif
#ifndef M_PI_2 /* PI/2 */
#define M_PI_2 1.57079632679489661923
#endif
#ifdef __cplusplus
}
#endif
#endif