The mapping between an RTP session and the SDP m= line is not always the
same, especially when BUNDLEing is used.
This causes a failure in a specific case where if when bundling,
if mline 0 is a data channel, and mline 1 an audio/video section,
then retrieving the transceiver at mline 0 (rtp session used) will fail
and cause an assertion.
This fix is actually potentially a regression for cases where the remote
part does not provide the a=ssrc: media level SDP attributes as is now
becoming common, especially when simulcast is involved.
The correct fix actually requires reading out header extensions as used
with bundle for signalling in the actual data, what media and therefore
transceiver is being used.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2467>
If they're not dropped, they can be blocked in the queue even if it is
leaky in the case where there is a buffer being pushed downstream. Since
in webrtc, it's unlikely that there will be a special allocator to
receive RTP packets, there is almost no downside to just ignoring the
queries.
Also drop queries if they get caught in the pad probe after the queue.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2363>
By using the clocksync inside the dtlssrtpenc, all streams inside a
bundled are synchronized together. This will cause problems if their
buffers are not already arriving synchronized: clocksync would wait for
a buffer on one stream and then buffers from the other stream(s) with
lower timestamps would all be sent out too late.
Placing the clocksync before the rtpbin and rtpfunnel synchronizes each
stream individually and they will be send out more smoothly as a result.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2355>
On the sink pad, we want the caps of the current stream, those
are the "received_caps" field. If we haven't received caps yet, then
we only care about the caps that the next element can accept, that is
the caps from the peer pad (and the preferences). Otherwise, we prevent
re-negotiation to a better codec when possible.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2183>
Now intersect against pads on both sides if they are available.
If the intersection fails, we now just reject the creation of the offer
or answer as it means that the codec_preferences are too restrictive or
that the caps on both sides the webrtcbin are not compatible.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2183>
When receiving an sctp message for a stream that not yet has an
sctpdec pad associated with it means we end up in
_on_sctpdec_pad_added. At this point we're holding the sctpassocation
lock. Then it's not possible to take the pc_lock because then code
executing under the pc_lock (which means anything in the webrtc
thread) may not take the sctpassociation lock. For example, running
the data channel close procedure from the webrtc thread means we
eventually end up sending a SCTP_RESET_STREAMS packet which needs to
grab the sctpassociation lock.
This means _on_sctpdec_pad_added simply cannot take the pc_lock and
also it is not possible to postpone the channel creation as we need to
link the pads right there. The solution is to introduce a more
granular dc_lock that protects only the things that needs to be done
to create the datachannel.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2186>
Support for closing WebRTC data channels as described in RFC
8831 (section 6.7) now fully supported. This means that we can now
reuse data channels that have been closed properly. Previously, an
application that created a lot of short-lived on-demand data channels
would quickly exhaust resources held by lingering non-closed data
channels.
We now use a one-to-one style socket interface to SCTP just like the
Google implementation (i.e. SOCK_STREAM instead of SOCK_SEQPACKET, see
RFC 6458). For some reason the socket interface to use was made
optional through a property "use-sock-stream" even though code wasn't
written to handle the SOCK_SEQPACKET style. Specifically the
SCTP_RESET_STREAMS command wouldn't work without passing the correct
assocation id. Changing the default interface to use from
SOCK_SEQPACKET to SOCK_STREAM now means we don't have to bother about
the association id as there is only one association per socket. For
the SCTP_RESET_STREAMS command we set it to SCTP_ALL_ASSOC just to
match the Google implementation.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2186>
