[FFmpeg-devel] [PATCH] matroskadec: Fix overflow introduced in a569a7b3

[James Almer]

On 7/6/2019 1:59 PM, Andreas Rheinhardt wrote:
> This commit fixes an overflow introduced in a569a7b3 that affected EBML
> elements that the Matroska demuxer doesn't want to parse like CRC-32
> elements. The return value of avio_skip (the new position on success or
> an AVERROR on failure) has been assigned to an integer which meant that
> new positions in the range of 2GB to 4GB-1 etc. were considered errors.
> 
> Fixes ticket #8001.
> 
> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt at gmail.com>
> ---
>  libavformat/matroskadec.c | 6 ++++--
>  1 file changed, 4 insertions(+), 2 deletions(-)
> 
> diff --git a/libavformat/matroskadec.c b/libavformat/matroskadec.c
> index bc73bfed11..4d7076fa26 100644
> --- a/libavformat/matroskadec.c
> +++ b/libavformat/matroskadec.c
> @@ -1259,12 +1259,13 @@ static int ebml_parse_elem(MatroskaDemuxContext *matroska,
>          return 1;
>      default:
>          if (length) {
> +            int64_t res2;
>              if (ffio_limit(pb, length) != length) {
>                  // ffio_limit emits its own error message,
>                  // so we don't have to.
>                  return AVERROR(EIO);
>              }
> -            if ((res = avio_skip(pb, length - 1)) >= 0) {
> +            if ((res2 = avio_skip(pb, length - 1)) >= 0) {
>                  // avio_skip might take us past EOF. We check for this
>                  // by skipping only length - 1 bytes, reading a byte and
>                  // checking the error flags. This is done in order to check
> @@ -1272,7 +1273,8 @@ static int ebml_parse_elem(MatroskaDemuxContext *matroska,
>                  // no filesize (that ffio_limit relies on) is available.
>                  avio_r8(pb);
>                  res = NEEDS_CHECKING;
> -            }
> +            } else
> +                res = res2;
>          } else
>              res = 0;
>      }

Pushed.