view tmv.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.

/*
 * 8088flex TMV video decoder
 * Copyright (c) 2009 Daniel Verkamp <daniel at drv.nu>
 *
 * 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 Street, Fifth Floor, Boston, MA 02110-1301 USA
 */

/**
 * 8088flex TMV video decoder
 * @file libavcodec/tmv.c
 * @author Daniel Verkamp
 * @sa http://www.oldskool.org/pc/8088_Corruption
 */

#include "avcodec.h"

#include "cga_data.h"

typedef struct TMVContext {
    AVFrame pic;
} TMVContext;

static int tmv_decode_frame(AVCodecContext *avctx, void *data,
                            int *data_size, AVPacket *avpkt)
{
    TMVContext *tmv    = avctx->priv_data;
    const uint8_t *src = avpkt->data;
    uint8_t *dst, *dst_char;
    unsigned char_cols = avctx->width >> 3;
    unsigned char_rows = avctx->height >> 3;
    unsigned x, y, mask, char_y, fg, bg, c;

    if (tmv->pic.data[0])
        avctx->release_buffer(avctx, &tmv->pic);

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

    if (avpkt->size < 2*char_rows*char_cols) {
        av_log(avctx, AV_LOG_ERROR,
               "Input buffer too small, truncated sample?\n");
        *data_size = 0;
        return -1;
    }

    tmv->pic.pict_type = FF_I_TYPE;
    tmv->pic.key_frame = 1;
    dst                = tmv->pic.data[0];

    tmv->pic.palette_has_changed = 1;
    memcpy(tmv->pic.data[1], ff_cga_palette, 16 * 4);

    for (y = 0; y < char_rows; y++) {
        for (x = 0; x < char_cols; x++) {
            c  = *src++ * 8;
            bg = *src  >> 4;
            fg = *src++ & 0xF;

            dst_char = dst + x * 8;
            for (char_y = 0; char_y < 8; char_y++) {
                for (mask = 0x80; mask; mask >>= 1) {
                    *dst_char++ = ff_cga_font[c + char_y] & mask ? fg : bg;
                }
                dst_char += tmv->pic.linesize[0] - 8;
            }
        }
        dst += tmv->pic.linesize[0] * 8;
    }

    *data_size = sizeof(AVFrame);
    *(AVFrame *)data = tmv->pic;
    return avpkt->size;
}

static av_cold int tmv_decode_close(AVCodecContext *avctx)
{
    TMVContext *tmv = avctx->priv_data;

    if (tmv->pic.data[0])
        avctx->release_buffer(avctx, &tmv->pic);

    return 0;
}

AVCodec tmv_decoder = {
    .name           = "tmv",
    .type           = CODEC_TYPE_VIDEO,
    .id             = CODEC_ID_TMV,
    .priv_data_size = sizeof(TMVContext),
    .close          = tmv_decode_close,
    .decode         = tmv_decode_frame,
    .capabilities   = CODEC_CAP_DR1,
    .long_name      = NULL_IF_CONFIG_SMALL("8088flex TMV"),
};