rtcpbuffer: fix sanity check for gst_rtcp_packet_xr_get_prt_info()
The length sanity check was bogus and was removed by commit c389dbf3 of !80 (merged), but should probably be re-added in some form.
It's not quite clear to me how the check matches the spec though - I'm guesing it was supposed to use *end_seq
and *start_seq
, but I think it would also have to take into account thinning? And is the min size calculation correct? (Isn't it more then 8 bits per seqnum?)
In general it also looks to me at first glance like sizes aren't properly checked in many places in the various new XR parsing functions before advancing pointers and such, but perhaps I misread that.