Mercurial > libavcodec.hg
view cscd.c @ 10874:bcfe2acbf190 libavcodec
AAC: Compress codebook tables and optimise sign bit handling
The codebooks each consist of small number of values repeated in
groups of 2 or 4. Storing the codebooks as a packed list of 2- or
4-bit indexes into a table reduces their size substantially (from 7.5k
to 1.5k), resulting in less cache pressure.
For the band types with sign bits in the bitstream, storing the number
and position of non-zero codebook values using a few bits avoids
multiple get_bits() calls and floating-point comparisons which gcc
handles miserably.
Some float/int type punning also avoids gcc brain damage.
Overall speedup 20-35% on Cortex-A8, 20% on Core i7.
| author | mru |
|---|---|
| date | Wed, 13 Jan 2010 16:46:28 +0000 |
| parents | d7ed9dcc78e3 |
| children | 8a4984c5cacc |
line wrap: on
line source
/* * CamStudio decoder * Copyright (c) 2006 Reimar Doeffinger * * 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 */ #include <stdio.h> #include <stdlib.h> #include "avcodec.h" #if CONFIG_ZLIB #include <zlib.h> #endif #include "libavutil/lzo.h" typedef struct { AVFrame pic; int linelen, height, bpp; unsigned int decomp_size; unsigned char* decomp_buf; } CamStudioContext; static void copy_frame_default(AVFrame *f, const uint8_t *src, int linelen, int height) { int i; uint8_t *dst = f->data[0]; dst += (height - 1) * f->linesize[0]; for (i = height; i; i--) { memcpy(dst, src, linelen); src += linelen; dst -= f->linesize[0]; } } static void add_frame_default(AVFrame *f, const uint8_t *src, int linelen, int height) { int i, j; uint8_t *dst = f->data[0]; dst += (height - 1) * f->linesize[0]; for (i = height; i; i--) { for (j = linelen; j; j--) *dst++ += *src++; dst -= f->linesize[0] + linelen; } } #if !HAVE_BIGENDIAN #define copy_frame_16 copy_frame_default #define copy_frame_32 copy_frame_default #define add_frame_16 add_frame_default #define add_frame_32 add_frame_default #else static void copy_frame_16(AVFrame *f, const uint8_t *src, int linelen, int height) { int i, j; uint8_t *dst = f->data[0]; dst += (height - 1) * f->linesize[0]; for (i = height; i; i--) { for (j = linelen / 2; j; j--) { dst[0] = src[1]; dst[1] = src[0]; src += 2; dst += 2; } dst -= f->linesize[0] + linelen; } } static void copy_frame_32(AVFrame *f, const uint8_t *src, int linelen, int height) { int i, j; uint8_t *dst = f->data[0]; dst += (height - 1) * f->linesize[0]; for (i = height; i; i--) { for (j = linelen / 4; j; j--) { dst[0] = src[3]; dst[1] = src[2]; dst[2] = src[1]; dst[3] = src[0]; src += 4; dst += 4; } dst -= f->linesize[0] + linelen; } } static void add_frame_16(AVFrame *f, const uint8_t *src, int linelen, int height) { int i, j; uint8_t *dst = f->data[0]; dst += (height - 1) * f->linesize[0]; for (i = height; i; i--) { for (j = linelen / 2; j; j--) { dst[0] += src[1]; dst[1] += src[0]; src += 2; dst += 2; } dst -= f->linesize[0] + linelen; } } static void add_frame_32(AVFrame *f, const uint8_t *src, int linelen, int height) { int i, j; uint8_t *dst = f->data[0]; dst += (height - 1) * f->linesize[0]; for (i = height; i; i--) { for (j = linelen / 4; j; j--) { dst[0] += src[3]; dst[1] += src[2]; dst[2] += src[1]; dst[3] += src[0]; src += 4; dst += 4; } dst -= f->linesize[0] + linelen; } } #endif static int decode_frame(AVCodecContext *avctx, void *data, int *data_size, AVPacket *avpkt) { const uint8_t *buf = avpkt->data; int buf_size = avpkt->size; CamStudioContext *c = avctx->priv_data; AVFrame *picture = data; if (buf_size < 2) { av_log(avctx, AV_LOG_ERROR, "coded frame too small\n"); return -1; } if (c->pic.data[0]) avctx->release_buffer(avctx, &c->pic); c->pic.reference = 1; c->pic.buffer_hints = FF_BUFFER_HINTS_VALID | FF_BUFFER_HINTS_READABLE | FF_BUFFER_HINTS_PRESERVE | FF_BUFFER_HINTS_REUSABLE; if (avctx->get_buffer(avctx, &c->pic) < 0) { av_log(avctx, AV_LOG_ERROR, "get_buffer() failed\n"); return -1; } // decompress data switch ((buf[0] >> 1) & 7) { case 0: { // lzo compression int outlen = c->decomp_size, inlen = buf_size - 2; if (av_lzo1x_decode(c->decomp_buf, &outlen, &buf[2], &inlen)) av_log(avctx, AV_LOG_ERROR, "error during lzo decompression\n"); break; } case 1: { // zlib compression #if CONFIG_ZLIB unsigned long dlen = c->decomp_size; if (uncompress(c->decomp_buf, &dlen, &buf[2], buf_size - 2) != Z_OK) av_log(avctx, AV_LOG_ERROR, "error during zlib decompression\n"); break; #else av_log(avctx, AV_LOG_ERROR, "compiled without zlib support\n"); return -1; #endif } default: av_log(avctx, AV_LOG_ERROR, "unknown compression\n"); return -1; } // flip upside down, add difference frame if (buf[0] & 1) { // keyframe c->pic.pict_type = FF_I_TYPE; c->pic.key_frame = 1; switch (c->bpp) { case 16: copy_frame_16(&c->pic, c->decomp_buf, c->linelen, c->height); break; case 32: copy_frame_32(&c->pic, c->decomp_buf, c->linelen, c->height); break; default: copy_frame_default(&c->pic, c->decomp_buf, c->linelen, c->height); } } else { c->pic.pict_type = FF_P_TYPE; c->pic.key_frame = 0; switch (c->bpp) { case 16: add_frame_16(&c->pic, c->decomp_buf, c->linelen, c->height); break; case 32: add_frame_32(&c->pic, c->decomp_buf, c->linelen, c->height); break; default: add_frame_default(&c->pic, c->decomp_buf, c->linelen, c->height); } } *picture = c->pic; *data_size = sizeof(AVFrame); return buf_size; } static av_cold int decode_init(AVCodecContext *avctx) { CamStudioContext *c = avctx->priv_data; switch (avctx->bits_per_coded_sample) { case 16: avctx->pix_fmt = PIX_FMT_RGB555; break; case 24: avctx->pix_fmt = PIX_FMT_BGR24; break; case 32: avctx->pix_fmt = PIX_FMT_RGB32; break; default: av_log(avctx, AV_LOG_ERROR, "CamStudio codec error: invalid depth %i bpp\n", avctx->bits_per_coded_sample); return 1; } c->bpp = avctx->bits_per_coded_sample; c->pic.data[0] = NULL; c->linelen = avctx->width * avctx->bits_per_coded_sample / 8; c->height = avctx->height; c->decomp_size = c->height * c->linelen; c->decomp_buf = av_malloc(c->decomp_size + AV_LZO_OUTPUT_PADDING); if (!c->decomp_buf) { av_log(avctx, AV_LOG_ERROR, "Can't allocate decompression buffer.\n"); return 1; } return 0; } static av_cold int decode_end(AVCodecContext *avctx) { CamStudioContext *c = avctx->priv_data; av_freep(&c->decomp_buf); if (c->pic.data[0]) avctx->release_buffer(avctx, &c->pic); return 0; } AVCodec cscd_decoder = { "camstudio", CODEC_TYPE_VIDEO, CODEC_ID_CSCD, sizeof(CamStudioContext), decode_init, NULL, decode_end, decode_frame, CODEC_CAP_DR1, .long_name = NULL_IF_CONFIG_SMALL("CamStudio"), };
