view libfaad2/ssr_fb.c @ 26625:5b89b42f6d50

Only compile and use libmpeg2 AltiVec code when AltiVec is available. The AltiVec code needs -maltivec to compile, but then AltiVec instructions appear in other places of the code causing MPlayer to sigill. Somehow upstream libmpeg2 manages not to sigill under what appear to be the same circumstances. Enlightenment welcome.

/*
** 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 [e-mail].
**
** $Id: ssr_fb.c,v 1.13 2004/09/04 14:56:29 menno Exp $
**/

#include "common.h"
#include "structs.h"

#ifdef SSR_DEC

#include <string.h>
#include <stdlib.h>
#include "syntax.h"
#include "filtbank.h"
#include "mdct.h"
#include "ssr_fb.h"
#include "ssr_win.h"

fb_info *ssr_filter_bank_init(uint16_t frame_len)
{
    uint16_t nshort = frame_len/8;

    fb_info *fb = (fb_info*)faad_malloc(sizeof(fb_info));
    memset(fb, 0, sizeof(fb_info));

    /* normal */
    fb->mdct256 = faad_mdct_init(2*nshort);
    fb->mdct2048 = faad_mdct_init(2*frame_len);

    fb->long_window[0]  = sine_long_256;
    fb->short_window[0] = sine_short_32;
    fb->long_window[1]  = kbd_long_256;
    fb->short_window[1] = kbd_short_32;

    return fb;
}

void ssr_filter_bank_end(fb_info *fb)
{
    faad_mdct_end(fb->mdct256);
    faad_mdct_end(fb->mdct2048);

    if (fb) faad_free(fb);
}

static INLINE void imdct_ssr(fb_info *fb, real_t *in_data,
                             real_t *out_data, uint16_t len)
{
    mdct_info *mdct;

    switch (len)
    {
    case 512:
        mdct = fb->mdct2048;
        break;
    case 64:
        mdct = fb->mdct256;
        break;
    }

    faad_imdct(mdct, in_data, out_data);
}

/* NON-overlapping inverse filterbank for use with SSR */
void ssr_ifilter_bank(fb_info *fb, uint8_t window_sequence, uint8_t window_shape,
                      uint8_t window_shape_prev, real_t *freq_in,
                      real_t *time_out, uint16_t frame_len)
{
    int16_t i;
    real_t *transf_buf;

    real_t *window_long;
    real_t *window_long_prev;
    real_t *window_short;
    real_t *window_short_prev;

    uint16_t nlong = frame_len;
    uint16_t nshort = frame_len/8;
    uint16_t trans = nshort/2;

    uint16_t nflat_ls = (nlong-nshort)/2;

    transf_buf = (real_t*)faad_malloc(2*nlong*sizeof(real_t));

    window_long       = fb->long_window[window_shape];
    window_long_prev  = fb->long_window[window_shape_prev];
    window_short      = fb->short_window[window_shape];
    window_short_prev = fb->short_window[window_shape_prev];

    switch (window_sequence)
    {
    case ONLY_LONG_SEQUENCE:
        imdct_ssr(fb, freq_in, transf_buf, 2*nlong);
        for (i = nlong-1; i >= 0; i--)
        {
            time_out[i] = MUL_R_C(transf_buf[i],window_long_prev[i]);
            time_out[nlong+i] = MUL_R_C(transf_buf[nlong+i],window_long[nlong-1-i]);
        }
        break;

    case LONG_START_SEQUENCE:
        imdct_ssr(fb, freq_in, transf_buf, 2*nlong);
        for (i = 0; i < nlong; i++)
            time_out[i] = MUL_R_C(transf_buf[i],window_long_prev[i]);
        for (i = 0; i < nflat_ls; i++)
            time_out[nlong+i] = transf_buf[nlong+i];
        for (i = 0; i < nshort; i++)
            time_out[nlong+nflat_ls+i] = MUL_R_C(transf_buf[nlong+nflat_ls+i],window_short[nshort-i-1]);
        for (i = 0; i < nflat_ls; i++)
            time_out[nlong+nflat_ls+nshort+i] = 0;
        break;

    case EIGHT_SHORT_SEQUENCE:
        imdct_ssr(fb, freq_in+0*nshort, transf_buf+2*nshort*0, 2*nshort);
        imdct_ssr(fb, freq_in+1*nshort, transf_buf+2*nshort*1, 2*nshort);
        imdct_ssr(fb, freq_in+2*nshort, transf_buf+2*nshort*2, 2*nshort);
        imdct_ssr(fb, freq_in+3*nshort, transf_buf+2*nshort*3, 2*nshort);
        imdct_ssr(fb, freq_in+4*nshort, transf_buf+2*nshort*4, 2*nshort);
        imdct_ssr(fb, freq_in+5*nshort, transf_buf+2*nshort*5, 2*nshort);
        imdct_ssr(fb, freq_in+6*nshort, transf_buf+2*nshort*6, 2*nshort);
        imdct_ssr(fb, freq_in+7*nshort, transf_buf+2*nshort*7, 2*nshort);
        for(i = nshort-1; i >= 0; i--)
        {
            time_out[i+0*nshort] = MUL_R_C(transf_buf[nshort*0+i],window_short_prev[i]);
            time_out[i+1*nshort] = MUL_R_C(transf_buf[nshort*1+i],window_short[i]);
            time_out[i+2*nshort] = MUL_R_C(transf_buf[nshort*2+i],window_short_prev[i]);
            time_out[i+3*nshort] = MUL_R_C(transf_buf[nshort*3+i],window_short[i]);
            time_out[i+4*nshort] = MUL_R_C(transf_buf[nshort*4+i],window_short_prev[i]);
            time_out[i+5*nshort] = MUL_R_C(transf_buf[nshort*5+i],window_short[i]);
            time_out[i+6*nshort] = MUL_R_C(transf_buf[nshort*6+i],window_short_prev[i]);
            time_out[i+7*nshort] = MUL_R_C(transf_buf[nshort*7+i],window_short[i]);
            time_out[i+8*nshort] = MUL_R_C(transf_buf[nshort*8+i],window_short_prev[i]);
            time_out[i+9*nshort] = MUL_R_C(transf_buf[nshort*9+i],window_short[i]);
            time_out[i+10*nshort] = MUL_R_C(transf_buf[nshort*10+i],window_short_prev[i]);
            time_out[i+11*nshort] = MUL_R_C(transf_buf[nshort*11+i],window_short[i]);
            time_out[i+12*nshort] = MUL_R_C(transf_buf[nshort*12+i],window_short_prev[i]);
            time_out[i+13*nshort] = MUL_R_C(transf_buf[nshort*13+i],window_short[i]);
            time_out[i+14*nshort] = MUL_R_C(transf_buf[nshort*14+i],window_short_prev[i]);
            time_out[i+15*nshort] = MUL_R_C(transf_buf[nshort*15+i],window_short[i]);
        }
        break;

    case LONG_STOP_SEQUENCE:
        imdct_ssr(fb, freq_in, transf_buf, 2*nlong);
        for (i = 0; i < nflat_ls; i++)
            time_out[i] = 0;
        for (i = 0; i < nshort; i++)
            time_out[nflat_ls+i] = MUL_R_C(transf_buf[nflat_ls+i],window_short_prev[i]);
        for (i = 0; i < nflat_ls; i++)
            time_out[nflat_ls+nshort+i] = transf_buf[nflat_ls+nshort+i];
        for (i = 0; i < nlong; i++)
            time_out[nlong+i] = MUL_R_C(transf_buf[nlong+i],window_long[nlong-1-i]);
		break;
    }

    faad_free(transf_buf);
}


#endif