[FFmpeg-devel] [PATCH 2/3] avformat/oggdec: Do not hardcode arbitrary and sometimes unavailable size
Lynne
dev at lynne.ee
Sat Jun 6 21:39:15 EEST 2020
Jun 6, 2020, 19:08 by michael at niedermayer.cc:
> Fixes: regression since e983197cbc93420b67aa7e811be47d7278c2c8a2
> Fixes: out of array read
> Fixes: 22185/clusterfuzz-testcase-minimized-ffmpeg_DEMUXER_fuzzer-5662069073641472
>
> Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
> Signed-off-by: Michael Niedermayer <michael at niedermayer.cc>
> ---
> libavformat/oggdec.c | 6 +++---
> 1 file changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/libavformat/oggdec.c b/libavformat/oggdec.c
> index f65013f55e..bd4311bb41 100644
> --- a/libavformat/oggdec.c
> +++ b/libavformat/oggdec.c
> @@ -206,7 +206,7 @@ static const struct ogg_codec *ogg_find_codec(uint8_t *buf, int size)
> * situation where a new audio stream spawn (identified with a new serial) and
> * must replace the previous one (track switch).
> */
> -static int ogg_replace_stream(AVFormatContext *s, uint32_t serial, char *magic,
> +static int ogg_replace_stream(AVFormatContext *s, uint32_t serial, char *magic, int magic_size,
> int probing)
> {
> struct ogg *ogg = s->priv_data;
> @@ -220,7 +220,7 @@ static int ogg_replace_stream(AVFormatContext *s, uint32_t serial, char *magic,
> }
>
> /* Check for codecs */
> - codec = ogg_find_codec(magic, 8);
> + codec = ogg_find_codec(magic, magic_size);
> if (!codec && !probing) {
> av_log(s, AV_LOG_ERROR, "Cannot identify new stream\n");
> return AVERROR_INVALIDDATA;
> @@ -430,7 +430,7 @@ static int ogg_read_page(AVFormatContext *s, int *sid, int probing)
> /* CRC is correct so we can be 99% sure there's an actual change here */
> if (idx < 0) {
> if (data_packets_seen(ogg))
> - idx = ogg_replace_stream(s, serial, readout_buf, probing);
> + idx = ogg_replace_stream(s, serial, readout_buf, size, probing);
> else
> idx = ogg_new_stream(s, serial);
>
Makes sense.
Just rename it from magic_size to page_size. Apart from that LGTM.
More information about the ffmpeg-devel
mailing list