view eatqi.c @ 11032:01bd040f8607 libavcodec

Unroll main loop so the edge==0 case is seperate. This allows many things to be simplified away. h264 decoder is overall 1% faster with a mbaff sample and 0.1% slower with the cathedral sample, probably because the slow loop filter code must be loaded into the code cache for each first MB of each row but isnt used for the following MBs.

/*
 * Electronic Arts TQI Video Decoder
 * Copyright (c) 2007-2009 Peter Ross <[e-mail]>
 *
 * This file is part of FFmpeg.
 *
 * FFmpeg is free software; you can redistribute it and/or
 * modify it under the terms of the GNU Lesser General Public
 * License as published by the Free Software Foundation; either
 * version 2.1 of the License, or (at your option) any later version.
 *
 * FFmpeg 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
 * Lesser General Public License for more details.
 *
 * You should have received a copy of the GNU Lesser General Public
 * License along with FFmpeg; if not, write to the Free Software
 * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
 */

/**
 * @file eatqi.c
 * Electronic Arts TQI Video Decoder
 * by Peter Ross <[e-mail]>
 *
 * Technical details here:
 * http://wiki.multimedia.cx/index.php?title=Electronic_Arts_TQI
 */

#include "avcodec.h"
#include "get_bits.h"
#include "dsputil.h"
#include "aandcttab.h"
#include "mpeg12.h"
#include "mpegvideo.h"

typedef struct TqiContext {
    MpegEncContext s;
    AVFrame frame;
    void *bitstream_buf;
    unsigned int bitstream_buf_size;
    DECLARE_ALIGNED_16(DCTELEM, block)[6][64];
} TqiContext;

static av_cold int tqi_decode_init(AVCodecContext *avctx)
{
    TqiContext *t = avctx->priv_data;
    MpegEncContext *s = &t->s;
    s->avctx = avctx;
    if(avctx->idct_algo==FF_IDCT_AUTO)
        avctx->idct_algo=FF_IDCT_EA;
    dsputil_init(&s->dsp, avctx);
    ff_init_scantable(s->dsp.idct_permutation, &s->intra_scantable, ff_zigzag_direct);
    s->qscale = 1;
    avctx->time_base = (AVRational){1, 15};
    avctx->pix_fmt = PIX_FMT_YUV420P;
    ff_mpeg12_init_vlcs();
    return 0;
}

static void tqi_decode_mb(MpegEncContext *s, DCTELEM (*block)[64])
{
    int n;
    s->dsp.clear_blocks(block[0]);
    for (n=0; n<6; n++)
        ff_mpeg1_decode_block_intra(s, block[n], n);
}

static inline void tqi_idct_put(TqiContext *t, DCTELEM (*block)[64])
{
    MpegEncContext *s = &t->s;
    int linesize= t->frame.linesize[0];
    uint8_t *dest_y  = t->frame.data[0] + (s->mb_y * 16* linesize            ) + s->mb_x * 16;
    uint8_t *dest_cb = t->frame.data[1] + (s->mb_y * 8 * t->frame.linesize[1]) + s->mb_x * 8;
    uint8_t *dest_cr = t->frame.data[2] + (s->mb_y * 8 * t->frame.linesize[2]) + s->mb_x * 8;

    s->dsp.idct_put(dest_y                 , linesize, block[0]);
    s->dsp.idct_put(dest_y              + 8, linesize, block[1]);
    s->dsp.idct_put(dest_y + 8*linesize    , linesize, block[2]);
    s->dsp.idct_put(dest_y + 8*linesize + 8, linesize, block[3]);
    if(!(s->avctx->flags&CODEC_FLAG_GRAY)) {
        s->dsp.idct_put(dest_cb, t->frame.linesize[1], block[4]);
        s->dsp.idct_put(dest_cr, t->frame.linesize[2], block[5]);
    }
}

static void tqi_calculate_qtable(MpegEncContext *s, int quant)
{
    const int qscale = (215 - 2*quant)*5;
    int i;
    if (s->avctx->idct_algo==FF_IDCT_EA) {
        s->intra_matrix[0] = (ff_inv_aanscales[0]*ff_mpeg1_default_intra_matrix[0])>>11;
        for(i=1; i<64; i++)
            s->intra_matrix[i] = (ff_inv_aanscales[i]*ff_mpeg1_default_intra_matrix[i]*qscale + 32)>>14;
    }else{
        s->intra_matrix[0] = ff_mpeg1_default_intra_matrix[0];
        for(i=1; i<64; i++)
            s->intra_matrix[i] = (ff_mpeg1_default_intra_matrix[i]*qscale + 32)>>3;
    }
}

static int tqi_decode_frame(AVCodecContext *avctx,
                            void *data, int *data_size,
                            AVPacket *avpkt)
{
    const uint8_t *buf = avpkt->data;
    int buf_size = avpkt->size;
    const uint8_t *buf_end = buf+buf_size;
    TqiContext *t = avctx->priv_data;
    MpegEncContext *s = &t->s;

    s->width  = AV_RL16(&buf[0]);
    s->height = AV_RL16(&buf[2]);
    tqi_calculate_qtable(s, buf[4]);
    buf += 8;

    if (t->frame.data[0])
        avctx->release_buffer(avctx, &t->frame);

    if (s->avctx->width!=s->width || s->avctx->height!=s->height)
        avcodec_set_dimensions(s->avctx, s->width, s->height);

    if(avctx->get_buffer(avctx, &t->frame) < 0) {
        av_log(avctx, AV_LOG_ERROR, "get_buffer() failed\n");
        return -1;
    }

    av_fast_malloc(&t->bitstream_buf, &t->bitstream_buf_size, (buf_end-buf) + FF_INPUT_BUFFER_PADDING_SIZE);
    if (!t->bitstream_buf)
        return AVERROR(ENOMEM);
    s->dsp.bswap_buf(t->bitstream_buf, (const uint32_t*)buf, (buf_end-buf)/4);
    init_get_bits(&s->gb, t->bitstream_buf, 8*(buf_end-buf));

    s->last_dc[0] = s->last_dc[1] = s->last_dc[2] = 0;
    for (s->mb_y=0; s->mb_y<(avctx->height+15)/16; s->mb_y++)
    for (s->mb_x=0; s->mb_x<(avctx->width+15)/16; s->mb_x++)
    {
        tqi_decode_mb(s, t->block);
        tqi_idct_put(t, t->block);
    }

    *data_size = sizeof(AVFrame);
    *(AVFrame*)data = t->frame;
    return buf_size;
}

static av_cold int tqi_decode_end(AVCodecContext *avctx)
{
    TqiContext *t = avctx->priv_data;
    if(t->frame.data[0])
        avctx->release_buffer(avctx, &t->frame);
    av_free(t->bitstream_buf);
    return 0;
}

AVCodec eatqi_decoder = {
    "eatqi",
    CODEC_TYPE_VIDEO,
    CODEC_ID_TQI,
    sizeof(TqiContext),
    tqi_decode_init,
    NULL,
    tqi_decode_end,
    tqi_decode_frame,
    CODEC_CAP_DR1,
    .long_name = NULL_IF_CONFIG_SMALL("Electronic Arts TQI Video"),
};