[FFmpeg-devel] [PATCH 2/2] avformat/matroskadec: Fix lzo decompression

[James Almer]

On 12/27/2019 11:40 PM, Andreas Rheinhardt wrote:
> When a Matroska Block is only stored in compressed form, the size of
> the uncompressed block is not explicitly coded and therefore not known
> before decompressing it. Therefore the demuxer uses a guess for the
> uncompressed size: The first guess is three times the compressed size
> and if this is not enough, it is repeatedly incremented by a factor of
> three. But when this happens with lzo, the decompression is neither
> resumed nor started again. Instead when av_lzo1x_decode indicates that x
> bytes of input data could not be decoded, because the output buffer is
> already full, the first (not the last) x bytes of the input buffer are
> resent for decoding in the next try; they overwrite already decoded
> data.
> 
> This commit fixes this by instead restarting the decompression anew,
> just with a bigger buffer.
> 
> This seems to be a regression since 935ec5a1.
> 
> A FATE-test for this has been added.
> 
> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt at gmail.com>
> ---
>  libavformat/matroskadec.c                 |  3 ++-
>  tests/fate/matroska.mak                   |  4 ++++
>  tests/ref/fate/matroska-lzo-decompression | 10 ++++++++++
>  3 files changed, 16 insertions(+), 1 deletion(-)
>  create mode 100644 tests/ref/fate/matroska-lzo-decompression
> 
> diff --git a/libavformat/matroskadec.c b/libavformat/matroskadec.c
> index 72624dc3f1..75f72d330c 100644
> --- a/libavformat/matroskadec.c
> +++ b/libavformat/matroskadec.c
> @@ -1599,6 +1599,7 @@ static int matroska_decode_buffer(uint8_t **buf, int *buf_size,
>  #if CONFIG_LZO
>      case MATROSKA_TRACK_ENCODING_COMP_LZO:
>          do {
> +            int insize = isize;
>              olen       = pkt_size *= 3;
>              newpktdata = av_realloc(pkt_data, pkt_size + AV_LZO_OUTPUT_PADDING
>                                                         + AV_INPUT_BUFFER_PADDING_SIZE);
> @@ -1607,7 +1608,7 @@ static int matroska_decode_buffer(uint8_t **buf, int *buf_size,
>                  goto failed;
>              }
>              pkt_data = newpktdata;
> -            result   = av_lzo1x_decode(pkt_data, &olen, data, &isize);
> +            result   = av_lzo1x_decode(pkt_data, &olen, data, &insize);
>          } while (result == AV_LZO_OUTPUT_FULL && pkt_size < 10000000);
>          if (result) {
>              result = AVERROR_INVALIDDATA;
> diff --git a/tests/fate/matroska.mak b/tests/fate/matroska.mak
> index 597be7346c..4aca4dc8eb 100644
> --- a/tests/fate/matroska.mak
> +++ b/tests/fate/matroska.mak
> @@ -22,6 +22,10 @@ fate-matroska-xiph-lacing: CMD = framecrc -i $(TARGET_SAMPLES)/mkv/xiph_lacing.m
>  FATE_MATROSKA-$(call ALLYES, MATROSKA_DEMUXER ZLIB) += fate-matroska-zlib-decompression
>  fate-matroska-zlib-decompression: CMD = framecrc -i $(TARGET_SAMPLES)/mkv/subtitle_zlib.mks -c:s copy
>  
> +# This tests that the matroska demuxer can decompress lzo compressed tracks.
> +FATE_MATROSKA-$(call ALLYES, MATROSKA_DEMUXER LZO) += fate-matroska-lzo-decompression
> +fate-matroska-lzo-decompression: CMD = framecrc -i $(TARGET_SAMPLES)/mkv/lzo.mka -c copy
> +
>  # This tests that the matroska demuxer correctly propagates
>  # the channel layout contained in vorbis comments in the CodecPrivate
>  # of flac tracks. It also tests header removal compression.
> diff --git a/tests/ref/fate/matroska-lzo-decompression b/tests/ref/fate/matroska-lzo-decompression
> new file mode 100644
> index 0000000000..241d5adf44
> --- /dev/null
> +++ b/tests/ref/fate/matroska-lzo-decompression
> @@ -0,0 +1,10 @@
> +#tb 0: 11337/500000000
> +#media_type 0: audio
> +#codec_id 0: pcm_s16le
> +#sample_rate 0: 44100
> +#channel_layout 0: 3
> +#channel_layout_name 0: stereo
> +0,          0,          0,     4096,    16384, 0x00000000
> +0,       4096,       4096,     4096,    16384, 0xad7eebf4
> +0,       8192,       8192,     4096,    16384, 0x1d1ff9d9
> +0,      12288,      12288,     4097,    16384, 0xf1d9e2e2

Pushed alongside the sample. Thanks.