[FFmpeg-devel] [PATCH v6] Fix integer parameters size check in SDP fmtp line

[Olivier MAIGNIAL]

You are right, some values are expected to be <= 32.
However, it doesn't add a bug: Without my patch fail will occur at parsing
time; With my patch it will occur a bit later on check performed by
get_bits_long().
Maybe we can add a specific max value to each parameter to make logs more
accurate on what happens.

On Wed, Jul 17, 2019 at 10:28 PM Michael Niedermayer <michael at niedermayer.cc>
wrote:

> On Fri, Jul 12, 2019 at 08:40:40AM +0200, Olivier Maignial wrote:
> > === PROBLEM ===
> >
> > I was trying to record h264 + aac streams from an RTSP server to mp4
> file. using this command line:
> >     ffmpeg -v verbose -y -i "rtsp://<ip>/my_resources" -codec copy
> -bsf:a aac_adtstoasc test.mp4
> >
> > FFmpeg then fail to record audio and output this logs:
> >     [rtsp @ 0xcda1f0] The profile-level-id field size is invalid (40)
> >     [rtsp @ 0xcda1f0] Error parsing AU headers
> >     ...
> >     [rtsp @ 0xcda1f0] Could not find codec parameters for stream 1
> (Audio: aac, 48000 Hz, 1 channels): unspecified sample format
> >
> > In SDP provided by my RTSP server I had this fmtp line:
> >     a=fmtp:98 streamType=5; profile-level-id=40; mode=AAC-hbr;
> config=1188; sizeLength=13; indexLength=3; indexDeltaLength=3;
> >
> > In FFmpeg code, I found a check introduced by commit
> 24130234cd9dd733116d17b724ea4c8e12ce097a. It disallows values greater than
> 32 for fmtp line parameters.
> > However, In RFC-6416 (RTP Payload Format for MPEG-4 Audio/Visual
> Streams) give examples of "profile-level-id" values for AAC, up to 55.
> > Furthermore, RFC-4566 (SDP: Session Description Protocol) do not give
> any limit of size on interger parameters given in fmtp line.
> >
> > === FIX ===
> >
> > Instead of prohibit values over 32, I propose to check the possible
> integer overflow.
> > The use of strtoll allow to check the string validity and the possible
> overflow.
> > Value is then checked against INT32_MIN and INT32_MAX. Using
> INT32_MIN/MAX ensure to have the same behavior on 32 or 64 bits platforms.
> >
> > This patch fix my problem and I now can record my RTSP AAC stream to mp4.
> > It has passed the full fate tests suite sucessfully.
> >
> > Signed-off-by: Olivier Maignial <olivier.maignial at smile.fr>
> > ---
> >
> > Changes V6 --> V7:
> >     - Use long long int and strtoll. LLONG_MAX is always greather than
> INT32_MIN while LONG_MAX can be equal to INT32_MAX. It allows to accept
> full INT32 range.
> >     - Avoid to use errno
> >
> >  libavformat/rtpdec_mpeg4.c | 13 ++++++++-----
> >  1 file changed, 8 insertions(+), 5 deletions(-)
> >
> > diff --git a/libavformat/rtpdec_mpeg4.c b/libavformat/rtpdec_mpeg4.c
> > index 4f70599..f1cbedf 100644
> > --- a/libavformat/rtpdec_mpeg4.c
> > +++ b/libavformat/rtpdec_mpeg4.c
> > @@ -289,15 +289,18 @@ static int parse_fmtp(AVFormatContext *s,
> >          for (i = 0; attr_names[i].str; ++i) {
> >              if (!av_strcasecmp(attr, attr_names[i].str)) {
> >                  if (attr_names[i].type == ATTR_NAME_TYPE_INT) {
> > -                    int val = atoi(value);
> > -                    if (val > 32) {
> > +                    char *end_ptr = NULL;
> > +                    long long int val = strtoll(value, &end_ptr, 10);
> > +                    if (end_ptr == value || end_ptr[0] != '\0' ||
> > +                        val < INT32_MIN || val > INT32_MAX) {
> >                          av_log(s, AV_LOG_ERROR,
> > -                               "The %s field size is invalid (%d)\n",
> > -                               attr, val);
> > +                               "The %s field value is not a valid
> number, or overflows int32: %s\n",
> > +                               attr, value);
> >                          return AVERROR_INVALIDDATA;
> >                      }
> > +
> >                      *(int *)((char *)data+
> > -                        attr_names[i].offset) = val;
> > +                        attr_names[i].offset) = (int) val;
>
> Looking at this a bit deeper, this reads several specific values
> for example sizelength which is then used to read with get_bits_long()
> that is limited to 32bit. Maybe iam missing some check but if not
> that is adding a bug.
> I did not check what the valid range of the other cases is but
> they may have similar problems
>
> [...]
> --
> Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB
>
> Why not whip the teacher when the pupil misbehaves? -- Diogenes of Sinope
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel at ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>
> To unsubscribe, visit link above, or email
> ffmpeg-devel-request at ffmpeg.org with subject "unsubscribe".