/* * 4XM codec * Copyright (c) 2003 Michael Niedermayer * * 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 */ /** * @file * 4XM codec. */ #include <inttypes.h> #include "libavutil/avassert.h"#include "libavutil/frame.h"#include "libavutil/imgutils.h"#include "libavutil/intreadwrite.h"#include "libavutil/mem.h"#include "libavutil/mem_internal.h"#include "libavutil/thread.h"#include "avcodec.h"#include "blockdsp.h"#include "bswapdsp.h"#include "bytestream.h"#include "codec_internal.h"#include "decode.h"#include "get_bits.h"  #define BLOCK_TYPE_VLC_BITS 5#define ACDC_VLC_BITS 9 #define CFRAME_BUFFER_COUNT 100 static const uint8_t block_type_tab[2][4][8][2] = {    {        {    // { 8, 4, 2 } x { 8, 4, 2}            { 0, 1 }, { 2, 2 }, { 6, 3 }, { 14, 4 }, { 30, 5 }, { 31, 5 }, { 0, 0 }        }, { // { 8, 4 } x 1            { 0, 1 }, { 0, 0 }, { 2, 2 }, { 6, 3 }, { 14, 4 }, { 15, 4 }, { 0, 0 }        }, { // 1 x { 8, 4 }            { 0, 1 }, { 2, 2 }, { 0, 0 }, { 6, 3 }, { 14, 4 }, { 15, 4 }, { 0, 0 }        }, { // 1 x 2, 2 x 1            { 0, 1 }, { 0, 0 }, { 0, 0 }, { 2, 2 }, { 6, 3 }, { 14, 4 }, { 15, 4 }        }    }, {        {   // { 8, 4, 2 } x { 8, 4, 2}            { 1, 2 }, { 4, 3 }, { 5, 3 }, { 0, 2 }, { 6, 3 }, { 7, 3 }, { 0, 0 }        }, {// { 8, 4 } x 1            { 1, 2 }, { 0, 0 }, { 2, 2 }, { 0, 2 }, { 6, 3 }, { 7, 3 }, { 0, 0 }        }, {// 1 x { 8, 4 }            { 1, 2 }, { 2, 2 }, { 0, 0 }, { 0, 2 }, { 6, 3 }, { 7, 3 }, { 0, 0 }        }, {// 1 x 2, 2 x 1            { 1, 2 }, { 0, 0 }, { 0, 0 }, { 0, 2 }, { 2, 2 }, { 6, 3 }, { 7, 3 }      }    }}; static const uint8_t size2index[4][4] = {    { -1, 3, 1, 1 },    {  3, 0, 0, 0 },    {  2, 0, 0, 0 },    {  2, 0, 0, 0 },}; static const int8_t mv[256][2] = {    {   0,   0 }, {   0,  -1 }, {  -1,   0 }, {   1,   0 }, {   0,   1 }, {  -1,  -1 }, {   1,  -1 }, {  -1,   1 },    {   1,   1 }, {   0,  -2 }, {  -2,   0 }, {   2,   0 }, {   0,   2 }, {  -1,  -2 }, {   1,  -2 }, {  -2,  -1 },    {   2,  -1 }, {  -2,   1 }, {   2,   1 }, {  -1,   2 }, {   1,   2 }, {  -2,  -2 }, {   2,  -2 }, {  -2,   2 },    {   2,   2 }, {   0,  -3 }, {  -3,   0 }, {   3,   0 }, {   0,   3 }, {  -1,  -3 }, {   1,  -3 }, {  -3,  -1 },    {   3,  -1 }, {  -3,   1 }, {   3,   1 }, {  -1,   3 }, {   1,   3 }, {  -2,  -3 }, {   2,  -3 }, {  -3,  -2 },    {   3,  -2 }, {  -3,   2 }, {   3,   2 }, {  -2,   3 }, {   2,   3 }, {   0,  -4 }, {  -4,   0 }, {   4,   0 },    {   0,   4 }, {  -1,  -4 }, {   1,  -4 }, {  -4,  -1 }, {   4,  -1 }, {   4,   1 }, {  -1,   4 }, {   1,   4 },    {  -3,  -3 }, {  -3,   3 }, {   3,   3 }, {  -2,  -4 }, {  -4,  -2 }, {   4,  -2 }, {  -4,   2 }, {  -2,   4 },    {   2,   4 }, {  -3,  -4 }, {   3,  -4 }, {   4,  -3 }, {  -5,   0 }, {  -4,   3 }, {  -3,   4 }, {   3,   4 },    {  -1,  -5 }, {  -5,  -1 }, {  -5,   1 }, {  -1,   5 }, {  -2,  -5 }, {   2,  -5 }, {   5,  -2 }, {   5,   2 },    {  -4,  -4 }, {  -4,   4 }, {  -3,  -5 }, {  -5,  -3 }, {  -5,   3 }, {   3,   5 }, {  -6,   0 }, {   0,   6 },    {  -6,  -1 }, {  -6,   1 }, {   1,   6 }, {   2,  -6 }, {  -6,   2 }, {   2,   6 }, {  -5,  -4 }, {   5,   4 },    {   4,   5 }, {  -6,  -3 }, {   6,   3 }, {  -7,   0 }, {  -1,  -7 }, {   5,  -5 }, {  -7,   1 }, {  -1,   7 },    {   4,  -6 }, {   6,   4 }, {  -2,  -7 }, {  -7,   2 }, {  -3,  -7 }, {   7,  -3 }, {   3,   7 }, {   6,  -5 },    {   0,  -8 }, {  -1,  -8 }, {  -7,  -4 }, {  -8,   1 }, {   4,   7 }, {   2,  -8 }, {  -2,   8 }, {   6,   6 },    {  -8,   3 }, {   5,  -7 }, {  -5,   7 }, {   8,  -4 }, {   0,  -9 }, {  -9,  -1 }, {   1,   9 }, {   7,  -6 },    {  -7,   6 }, {  -5,  -8 }, {  -5,   8 }, {  -9,   3 }, {   9,  -4 }, {   7,  -7 }, {   8,  -6 }, {   6,   8 },    {  10,   1 }, { -10,   2 }, {   9,  -5 }, {  10,  -3 }, {  -8,  -7 }, { -10,  -4 }, {   6,  -9 }, { -11,   0 },    {  11,   1 }, { -11,  -2 }, {  -2,  11 }, {   7,  -9 }, {  -7,   9 }, {  10,   6 }, {  -4,  11 }, {   8,  -9 },    {   8,   9 }, {   5,  11 }, {   7, -10 }, {  12,  -3 }, {  11,   6 }, {  -9,  -9 }, {   8,  10 }, {   5,  12 },    { -11,   7 }, {  13,   2 }, {   6, -12 }, {  10,   9 }, { -11,   8 }, {  -7,  12 }, {   0,  14 }, {  14,  -2 },    {  -9,  11 }, {  -6,  13 }, { -14,  -4 }, {  -5, -14 }, {   5,  14 }, { -15,  -1 }, { -14,  -6 }, {   3, -15 },    {  11, -11 }, {  -7,  14 }, {  -5,  15 }, {   8, -14 }, {  15,   6 }, {   3,  16 }, {   7, -15 }, { -16,   5 },    {   0,  17 }, { -16,  -6 }, { -10,  14 }, { -16,   7 }, {  12,  13 }, { -16,   8 }, { -17,   6 }, { -18,   3 },    {  -7,  17 }, {  15,  11 }, {  16,  10 }, {   2, -19 }, {   3, -19 }, { -11, -16 }, { -18,   8 }, { -19,  -6 },    {   2, -20 }, { -17, -11 }, { -10, -18 }, {   8,  19 }, { -21,  -1 }, { -20,   7 }, {  -4,  21 }, {  21,   5 },    {  15,  16 }, {   2, -22 }, { -10, -20 }, { -22,   5 }, {  20, -11 }, {  -7, -22 }, { -12,  20 }, {  23,  -5 },    {  13, -20 }, {  24,  -2 }, { -15,  19 }, { -11,  22 }, {  16,  19 }, {  23, -10 }, { -18, -18 }, {  -9, -24 },    {  24, -10 }, {  -3,  26 }, { -23,  13 }, { -18, -20 }, {  17,  21 }, {  -4,  27 }, {  27,   6 }, {   1, -28 },    { -11,  26 }, { -17, -23 }, {   7,  28 }, {  11, -27 }, {  29,   5 }, { -23, -19 }, { -28, -11 }, { -21,  22 },    { -30,   7 }, { -17,  26 }, { -27,  16 }, {  13,  29 }, {  19, -26 }, {  10, -31 }, { -14, -30 }, {  20, -27 },    { -29,  18 }, { -16, -31 }, { -28, -22 }, {  21, -30 }, { -25,  28 }, {  26, -29 }, {  25, -32 }, { -32, -32 }}; /* This is simply the scaled down elementwise product of the standard JPEG * quantizer table and the AAN premul table. */static const uint8_t dequant_table[64] = {    16, 15, 13, 19, 24, 31, 28, 17,    17, 23, 25, 31, 36, 63, 45, 21,    18, 24, 27, 37, 52, 59, 49, 20,    16, 28, 34, 40, 60, 80, 51, 20,    18, 31, 48, 66, 68, 86, 56, 21,    19, 38, 56, 59, 64, 64, 48, 20,    27, 48, 55, 55, 56, 51, 35, 15,    20, 35, 34, 32, 31, 22, 15,  8,}; static VLCElem block_type_vlc[2][4][32];  typedef struct CFrameBuffer {    unsigned int allocated_size;    unsigned int size;    int id;    uint8_t *data;} CFrameBuffer; typedef struct FourXContext {    AVCodecContext *avctx;    BlockDSPContext bdsp;    BswapDSPContext bbdsp;    uint16_t *frame_buffer;    uint16_t *last_frame_buffer;    GetBitContext pre_gb;          ///< ac/dc prefix    GetBitContext gb;    GetByteContext g;    GetByteContext g2;    int mv[256];    VLC pre_vlc;    int last_dc;    DECLARE_ALIGNED(32, int16_t, block)[6][64];    void *bitstream_buffer;    unsigned int bitstream_buffer_size;    int version;    CFrameBuffer cfrm[CFRAME_BUFFER_COUNT];} FourXContext;  #define FIX_1_082392200  70936#define FIX_1_414213562  92682#define FIX_1_847759065 121095#define FIX_2_613125930 171254 #define MULTIPLY(var, const) ((int)((var) * (unsigned)(const)) >> 16) static void idct(int16_t block[64]){    int tmp0, tmp1, tmp2, tmp3, tmp4, tmp5, tmp6, tmp7;    int tmp10, tmp11, tmp12, tmp13;    int z5, z10, z11, z12, z13;    int i;    int temp[64];     for (i = 0; i < 8; i++) {        tmp10 = block[8 * 0 + i] + block[8 * 4 + i];        tmp11 = block[8 * 0 + i] - block[8 * 4 + i];         tmp13 = block[8 * 2 + i] + block[8 * 6 + i];        tmp12 = MULTIPLY(block[8 * 2 + i] - block[8 * 6 + i], FIX_1_414213562) - tmp13;         tmp0 = tmp10 + tmp13;        tmp3 = tmp10 - tmp13;        tmp1 = tmp11 + tmp12;        tmp2 = tmp11 - tmp12;         z13 = block[8 * 5 + i] + block[8 * 3 + i];        z10 = block[8 * 5 + i] - block[8 * 3 + i];        z11 = block[8 * 1 + i] + block[8 * 7 + i];        z12 = block[8 * 1 + i] - block[8 * 7 + i];         tmp7  =          z11 + z13;        tmp11 = MULTIPLY(z11 - z13, FIX_1_414213562);         z5    = MULTIPLY(z10 + z12, FIX_1_847759065);        tmp10 = MULTIPLY(z12,  FIX_1_082392200) - z5;        tmp12 = MULTIPLY(z10, -FIX_2_613125930) + z5;         tmp6 = tmp12 - tmp7;        tmp5 = tmp11 - tmp6;        tmp4 = tmp10 + tmp5;         temp[8 * 0 + i] = tmp0 + tmp7;        temp[8 * 7 + i] = tmp0 - tmp7;        temp[8 * 1 + i] = tmp1 + tmp6;        temp[8 * 6 + i] = tmp1 - tmp6;        temp[8 * 2 + i] = tmp2 + tmp5;        temp[8 * 5 + i] = tmp2 - tmp5;        temp[8 * 4 + i] = tmp3 + tmp4;        temp[8 * 3 + i] = tmp3 - tmp4;    }     for (i = 0; i < 8 * 8; i += 8) {        tmp10 = temp[0 + i] + temp[4 + i];        tmp11 = temp[0 + i] - temp[4 + i];         tmp13 = temp[2 + i] + temp[6 + i];        tmp12 = MULTIPLY(temp[2 + i] - temp[6 + i], FIX_1_414213562) - tmp13;         tmp0 = tmp10 + tmp13;        tmp3 = tmp10 - tmp13;        tmp1 = tmp11 + tmp12;        tmp2 = tmp11 - tmp12;         z13 = temp[5 + i] + temp[3 + i];        z10 = temp[5 + i] - temp[3 + i];        z11 = temp[1 + i] + temp[7 + i];        z12 = temp[1 + i] - temp[7 + i];         tmp7  = z11 + z13;        tmp11 = MULTIPLY(z11 - z13, FIX_1_414213562);         z5    = MULTIPLY(z10 + z12, FIX_1_847759065);        tmp10 = MULTIPLY(z12,  FIX_1_082392200) - z5;        tmp12 = MULTIPLY(z10, -FIX_2_613125930) + z5;         tmp6 = tmp12 - tmp7;        tmp5 = tmp11 - tmp6;        tmp4 = tmp10 + tmp5;         block[0 + i] = (tmp0 + tmp7) >> 6;        block[7 + i] = (tmp0 - tmp7) >> 6;        block[1 + i] = (tmp1 + tmp6) >> 6;        block[6 + i] = (tmp1 - tmp6) >> 6;        block[2 + i] = (tmp2 + tmp5) >> 6;        block[5 + i] = (tmp2 - tmp5) >> 6;        block[4 + i] = (tmp3 + tmp4) >> 6;        block[3 + i] = (tmp3 - tmp4) >> 6;    }} static av_cold void init_vlcs(void){    int i, j;     for (i = 0; i < 2; i++) {        for (j = 0; j < 4; j++) {            ff_vlc_init_table_sparse(block_type_vlc[i][j], FF_ARRAY_ELEMS(block_type_vlc[i][j]),                                     BLOCK_TYPE_VLC_BITS, 7,                                     &block_type_tab[i][j][0][1], 2, 1,                                     &block_type_tab[i][j][0][0], 2, 1,                                     NULL, 0, 0, 0);        }    }} static void init_mv(FourXContext *f, int linesize){    int i;     for (i = 0; i < 256; i++) {        if (f->version > 1)            f->mv[i] = mv[i][0] + mv[i][1] * linesize / 2;        else            f->mv[i] = (i & 15) - 8 + ((i >> 4) - 8) * linesize / 2;    }} #if HAVE_BIGENDIAN#define LE_CENTRIC_MUL(dst, src, scale, dc)             \    {                                                   \        unsigned tmpval = AV_RN32(src);                 \        tmpval = (tmpval << 16) | (tmpval >> 16);       \        tmpval = tmpval * (scale) + (dc);               \        tmpval = (tmpval << 16) | (tmpval >> 16);       \        AV_WN32A(dst, tmpval);                          \    }#else#define LE_CENTRIC_MUL(dst, src, scale, dc)              \    {                                                    \        unsigned tmpval = AV_RN32(src) * (scale) + (dc); \        AV_WN32A(dst, tmpval);                           \    }#endif static inline void mcdc(uint16_t *dst, const uint16_t *src, int log2w,                        int h, int stride, int scale, unsigned dc){    int i;    dc *= 0x10001;     switch (log2w) {    case 0:        for (i = 0; i < h; i++) {            dst[0] = scale * src[0] + dc;            if (scale)                src += stride;            dst += stride;        }        break;    case 1:        for (i = 0; i < h; i++) {            LE_CENTRIC_MUL(dst, src, scale, dc);            if (scale)                src += stride;            dst += stride;        }        break;    case 2:        for (i = 0; i < h; i++) {            LE_CENTRIC_MUL(dst, src, scale, dc);            LE_CENTRIC_MUL(dst + 2, src + 2, scale, dc);            if (scale)                src += stride;            dst += stride;        }        break;    case 3:        for (i = 0; i < h; i++) {            LE_CENTRIC_MUL(dst,     src,     scale, dc);            LE_CENTRIC_MUL(dst + 2, src + 2, scale, dc);            LE_CENTRIC_MUL(dst + 4, src + 4, scale, dc);            LE_CENTRIC_MUL(dst + 6, src + 6, scale, dc);            if (scale)                src += stride;            dst += stride;        }        break;    default:        av_assert0(0);    }} static int decode_p_block(FourXContext *f, uint16_t *dst, const uint16_t *src,                          int log2w, int log2h, int stride){    int index, h, code, ret, scale = 1;    uint16_t *start, *end;    unsigned dc = 0;     av_assert0(log2w >= 0 && log2h >= 0);     index = size2index[log2h][log2w];    av_assert0(index >= 0);     if (get_bits_left(&f->gb) < 1)        return AVERROR_INVALIDDATA;    h     = 1 << log2h;    code  = get_vlc2(&f->gb, block_type_vlc[1 - (f->version > 1)][index],                     BLOCK_TYPE_VLC_BITS, 1);    av_assert0(code >= 0 && code <= 6);     start = f->last_frame_buffer;    end   = start + stride * (f->avctx->height - h + 1) - (1 << log2w);     if (code == 1) {        log2h--;        if ((ret = decode_p_block(f, dst, src, log2w, log2h, stride)) < 0)            return ret;        return decode_p_block(f, dst + (stride << log2h),                              src + (stride << log2h),                              log2w, log2h, stride);    } else if (code == 2) {        log2w--;        if ((ret = decode_p_block(f, dst , src, log2w, log2h, stride)) < 0)            return ret;        return decode_p_block(f, dst + (1 << log2w),                              src + (1 << log2w),                              log2w, log2h, stride);    } else if (code == 6) {        if (bytestream2_get_bytes_left(&f->g2) < 4) {            av_log(f->avctx, AV_LOG_ERROR, "wordstream overread\n");            return AVERROR_INVALIDDATA;        }        if (log2w) {            dst[0]      = bytestream2_get_le16u(&f->g2);            dst[1]      = bytestream2_get_le16u(&f->g2);        } else {            dst[0]      = bytestream2_get_le16u(&f->g2);            dst[stride] = bytestream2_get_le16u(&f->g2);        }        return 0;    }     if ((code&3)==0 && bytestream2_get_bytes_left(&f->g) < 1) {        av_log(f->avctx, AV_LOG_ERROR, "bytestream overread\n");        return AVERROR_INVALIDDATA;    }     if (code == 0) {        src  += f->mv[bytestream2_get_byte(&f->g)];    } else if (code == 3 && f->version >= 2) {        return 0;    } else if (code == 4) {        src  += f->mv[bytestream2_get_byte(&f->g)];        if (bytestream2_get_bytes_left(&f->g2) < 2){            av_log(f->avctx, AV_LOG_ERROR, "wordstream overread\n");            return AVERROR_INVALIDDATA;        }        dc    = bytestream2_get_le16(&f->g2);    } else if (code == 5) {        if (bytestream2_get_bytes_left(&f->g2) < 2){            av_log(f->avctx, AV_LOG_ERROR, "wordstream overread\n");            return AVERROR_INVALIDDATA;        }        av_assert0(start <= src && src <= end);        scale = 0;        dc    = bytestream2_get_le16(&f->g2);    }     if (start > src || src > end) {        av_log(f->avctx, AV_LOG_ERROR, "mv out of pic\n");        return AVERROR_INVALIDDATA;    }     mcdc(dst, src, log2w, h, stride, scale, dc);     return 0;} static int decode_p_frame(FourXContext *f, const uint8_t *buf, int length){    int x, y;    const int width  = f->avctx->width;    const int height = f->avctx->height;    uint16_t *dst    = f->frame_buffer;    uint16_t *src;    unsigned int bitstream_size, bytestream_size, wordstream_size, extra,                 bytestream_offset, wordstream_offset;    int ret;     src = f->last_frame_buffer;     if (f->version > 1) {        extra           = 20;        if (length < extra)            return AVERROR_INVALIDDATA;        bitstream_size  = AV_RL32(buf + 8);        wordstream_size = AV_RL32(buf + 12);        bytestream_size = AV_RL32(buf + 16);    } else {        extra           = 0;        bitstream_size  = AV_RL16(buf - 4);        wordstream_size = AV_RL16(buf - 2);        bytestream_size = FFMAX(length - bitstream_size - wordstream_size, 0);    }     if (bitstream_size > length || bitstream_size >= INT_MAX/8 ||        bytestream_size > length - bitstream_size ||        wordstream_size > length - bytestream_size - bitstream_size ||        extra > length - bytestream_size - bitstream_size - wordstream_size) {        av_log(f->avctx, AV_LOG_ERROR, "lengths %d %d %d %d\n", bitstream_size, bytestream_size, wordstream_size,        bitstream_size+ bytestream_size+ wordstream_size - length);        return AVERROR_INVALIDDATA;    }     av_fast_padded_malloc(&f->bitstream_buffer, &f->bitstream_buffer_size,                          bitstream_size);    if (!f->bitstream_buffer)        return AVERROR(ENOMEM);    f->bbdsp.bswap_buf(f->bitstream_buffer, (const uint32_t *) (buf + extra),                       bitstream_size / 4);    init_get_bits(&f->gb, f->bitstream_buffer, 8 * bitstream_size);     wordstream_offset = extra + bitstream_size;    bytestream_offset = extra + bitstream_size + wordstream_size;    bytestream2_init(&f->g2, buf + wordstream_offset,                     length - wordstream_offset);    bytestream2_init(&f->g, buf + bytestream_offset,                     length - bytestream_offset);     init_mv(f, width * 2);     for (y = 0; y < height; y += 8) {        for (x = 0; x < width; x += 8)            if ((ret = decode_p_block(f, dst + x, src + x, 3, 3, width)) < 0)                return ret;        src += 8 * width;        dst += 8 * width;    }     return 0;} /** * decode block and dequantize. * Note this is almost identical to MJPEG. */static int decode_i_block(FourXContext *f, int16_t *block){

Lines 1–500 of 1044. Load more