Commit graph

119 commits

Author SHA1 Message Date
Sebastian Dröge
4ffa6350e8 webrtc: In all blocking pad probes except for sink pads also handle serialized events
Otherwise it can happen that e.g. the stream-start event is tried to be
sent as part of pushing the first buffer. Downstream might not be in
PAUSED/PLAYING yet, so the event is rejected with GST_FLOW_FLUSHING and
because it's an event would not cause the blocking pad probe to trigger
first. This would then return GST_FLOW_FLUSHING for the buffer and shut
down all of upstream.

To solve this we return GST_PAD_PROBE_DROP for all events. In case of
sticky events they would be resent again later once we unblocked after
blocking on the buffer and everything works fine.

Don't handle events specifically in sink pad blocking pad probes as here
downstream is not linked yet and we are actually waiting for the
following CAPS event before unblocking can happen.

Fixes https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/issues/1172
2020-02-11 00:49:51 +00:00
Sebastian Dröge
c16d4d2c33 webrtcbin: Add a blocking pad probe for the receivebin -> sctpdec connection
Without this it might happen that received data from the DTLS transport
is already passed to sctpdec before its state was set to PLAYING. This
would cause the data to be dropped, GST_FLOW_FLUSHING to be returned and
the whole DTLS transport to shut down.

Fixes https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/issues/1172
among other things.
2020-02-11 00:49:51 +00:00
Mathieu Duponchelle
7cc185bd86 webrtcbin: connect rtp funnel after updating ptmaps
We need the streams' pt maps updated before requesting pads
on rtpbin, because this is what will trigger the requesting
of FEC encoders, and our handler for this request looks for
the payload types in the relevant stream's pt map.

Fixes #1187
2020-01-21 11:17:38 +00:00
Sebastian Dröge
0c39068c89 webrtcbin: Start datachannel SCTP elements only after the DTLS connection is established
Otherwise we would start sending data to the DTLS connection before, and
the DTLS elements consider this an error.

Also RFC 8261 mentions:
  o A DTLS connection MUST be established before an SCTP association can
    be set up.
2020-01-19 11:16:34 +00:00
Sebastian Dröge
2798a80ebe webrtcbin: Add handling of unspecified peer-connection-state situation
For us it can happen that the DTLS transports are still in the process
of connecting while the ICE transport is already completed. This
situation is not specified in the spec but conceptually that means it is
still in the process of connecting.

Fixes https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/issues/758
2020-01-19 11:16:34 +00:00
Sebastian Dröge
4b73322333 webrtcbin: Return the old state if we ended up being in an unspecified situation
Previously we would've returned NEW, which is usually more wrong.
2020-01-19 11:16:34 +00:00
Sebastian Dröge
22869356db webrtcbin: Fix transitions for the peer connection state
They're now mapping exactly to what
  https://www.w3.org/TR/webrtc/#rtcpeerconnectionstate-enum
actually specifies.

Related to https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/issues/758
2020-01-19 11:16:34 +00:00
Sebastian Dröge
41175f4ebe webrtcbin: Fix transitions for the connection state
They're now mapping exactly to what
  https://www.w3.org/TR/webrtc/#dom-rtciceconnectionstate
actually specifies.

Related to https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/issues/758
2020-01-19 11:16:34 +00:00
Sebastian Dröge
47ce34d32c webrtcbin: Don't consider transceivers without mid as inactive during ICE gathering state updates
We don't have any mid before parsing the SDP, which happens after we
handled the SDP answer and that usually happens long after ICE candidate
gathering is finished.

Without this all transceivers are considered inactive and as such ICE
gathering is for active transceiver was considered complete from the
very beginning.

Fixes https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/issues/1126
2020-01-19 10:47:59 +00:00
Sebastian Dröge
de0f803d56 webrtcbin: Don't consider RTP receivers stopped
We don't support stopping RTP receivers currently so let's not consider
them all stopped all the time. This fixes some of the ICE/DTLS state
change handling and specifically fixes the ICE gathering state.

Previously the ICE gathering state was immediately going from NEW to
COMPLETE because it considered all transceivers stopped and as such all
activate transceivers were finished gathering ICE candidates.

Fixes https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/issues/1126
2020-01-19 10:47:59 +00:00
Sebastian Dröge
57c982a1dd webrtcbin: Improve logging related to ICE/DTLS state changes 2020-01-19 10:47:59 +00:00
Jan Schmidt
8e87fe42ad WebRTC: Support non-trickle ICE candidates in the SDP
Add any ICE candidates from the SDP before adding pending
trickle ICE candidates to support non-trickle peers

Fixes https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/issues/678
2020-01-13 02:30:44 +11:00
Edward Hervey
c026522084 webrtcbin: Fix memory leak
The structure is not used after this block

CID: 1455481
2019-11-21 16:25:21 +01:00
Niels De Graef
d8f61515d8 Don't pass default GLib marshallers for signals
By passing NULL to `g_signal_new` instead of a marshaller, GLib will
actually internally optimize the signal (if the marshaller is available
in GLib itself) by also setting the valist marshaller. This makes the
signal emission a bit more performant than the regular marshalling,
which still needs to box into `GValue` and call libffi in case of a
generic marshaller.

Note that for custom marshallers, one would use
`g_signal_set_va_marshaller()` with the valist marshaller instead.
2019-11-06 14:27:46 +00:00
Aaron Boxer
6d3429af34 documentation: fixed a heap o' typos 2019-11-05 09:11:25 -05:00
David Gunzinger
e2e86658f2 webrtc: fix type of max-retransmits, make it work 2019-08-13 12:17:13 +02:00
Jakub Adam
831b124976 webrtcbin: Support data channel SDP offers from Chrome
When negotiating a data channel, Chrome as recent as 75 still uses SDP
based on version 05 of the SCTP SDP draft, for example:

 m=application 9 DTLS/SCTP 5000
 a=sctpmap:5000 webrtc-datachannel 1024

Implement support for parsing SCTP port out of SDP message with sctpmap
attribute. Fixes data channel negotiation with Chrome browser.
2019-07-29 22:04:08 +00:00
Sebastian Dröge
329b2d3a6a webrtcbin: Don't assert if an SDP media can't be converted to caps
Fixes https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/issues/1008
2019-07-08 07:18:41 +00:00
Matthew Waters
cef839533e webrtcbin: use the latest self-generated SDP as the basis for renegotiations
Fixes multiple errors when a webrtcbin renegotiation can switch between the
offerer and the answerer.
2019-07-03 23:44:15 +00:00
Matthew Waters
95488812b2 webrtc: fix the location of signalling-state change notification
1. The spec indicates that the notification should occur near the end of
   'setting the description' processing
2. The current location with the drop of the lock could cause the 'check
   if negotiation is needed' logic to execute and become confused about
   the state of the webrtcbin's current local descriptions.
   In the bad case, the following assertions could be hit:
   g_assert (trans->mline < gst_sdp_message_medias_len (webrtc->current_local_description->sdp));
   g_assert (trans->mline < gst_sdp_message_medias_len (webrtc->current_remote_description->sdp));

Moving the signalling state change later in the set description task
means that checking for a renegotiation will early abort as the
signalling state is not STABLE before the session description and
transceivers have been updated.
2019-06-04 05:43:43 +00:00
Matthew Waters
f8911deccf webrtc: only set sctp ports if they are different
SCTPassociation will complain if we do that while running and resetting
is not something we support at the moment
2019-05-30 21:33:09 +10:00
Matthew Waters
be011d2086 webrtc/dc: move some code from webrtcbin into the datachannel 2019-05-30 21:33:09 +10:00
Matthew Waters
a51db86ac4 webrtc: hold onto any unknown ICE candidates until the next SDP set
It is very possible for badly behaving signalling or peers to send
us ICE candidates before we receive an SDP.  While we had consideration
for that on the first set SDP, subsequent SDP's could result in
misconfigured ICE transports.  Expand the previous code to also take
into account reconfigurations.
2019-05-30 21:33:09 +10:00
Matthew Waters
177aa22bcd webrtc: Initial support for stream addition/removal
Limitations:
- No transport changes at all (ICE, DTLS)
- Codec changes are untested and probably don't work
- Stream removal doesn't remove transports (i.e. non-bundled transports
  will stay around until webrtcbin is shutdown)
- Unified Plan SDP only. No Plan-B support.
2019-05-30 21:33:09 +10:00
Matthew Waters
033e55695f webrtcbin: expose the transceiver as a pad property 2019-05-30 21:33:09 +10:00
Matthew Waters
6ad0edbe92 webrtc: track and log more rtpbin state
like bye's timeouts, validation, activation, etc
2019-05-30 21:33:09 +10:00
Matthew Waters
5ea7031bd0 webrtc: mark remote/local-description as readonly 2019-05-30 21:32:06 +10:00
Matthew Waters
19b3d744d8 webrtc: don't reuse stopped transceivers at all 2019-05-30 21:26:46 +10:00
Matthew Waters
4d34fe7617 webrtc: also check for a null mid to signify an unassociated transceiver
We always give our transceivers an mline on creation so that check is
not useful by itself
2019-05-30 21:26:46 +10:00
Matthew Waters
00977f263a webrtc: only check sink pads for a 'sink pads have caps' check 2019-05-30 21:26:46 +10:00
Matthew Waters
bd92b2f7c4 webrtc: fix answer creation with multiple streams and similar caps 2019-05-30 21:26:46 +10:00
Philippe Normand
9595a7a721 webrtcbin: Expose current and pending local/remote description properties
They are already handled in the property getter and setter functions but were
not formally declared in the GObject class.
2019-05-30 10:35:58 +01:00
Niels De Graef
7cd4064425 webrtc: Fix some signals' GIR annotations
This will lead to wrong bindings otherwise (and creates more correct
expectations for developers).
2019-05-17 15:28:54 +02:00
Thibault Saunier
7fe3f36ac8 Minor documentation fixes 2019-05-13 11:36:27 -04:00
Niels De Graef
ce92cb81a0 webrtc: Fix signals documentation
Some GIR annotations were incorrect or even missing. The former isn't
good for bindings, while the latter is especially annoying for signal
handlers, as that means your arguments will get the wrong names in the
rendered documentation.
2019-05-09 14:19:01 +02:00
Mathieu Duponchelle
a2779ef366 webrtcbin: fix pt selection for FEC and RTX when BUNDLE
When we offer bundled media, payload types must be unique
across all bundled media, as they will be multiplexed in the
same session.
2019-03-15 18:37:51 +01:00
Mathieu Duponchelle
08858d753c webrtcbin: add get-transceiver signal
get-transceivers is not introspectable, and a method to get a
transceiver by index is convenient.
2019-03-12 21:04:48 +00:00
Matthew Waters
2a1176973a webrtc: fix rtx + bundle
If bundle was used in combination with rtx, only the bundled transport
stream would have correctly configured rtx parameters.

Iterate over the payloads upfront in the bundled case to ensure the
correct payload mapping is set for the RTX elements.
2019-02-15 08:19:51 +00:00
Mathieu Duponchelle
85c75bb23b webrtc: expose ice-transport-policy property
This is the equivalent of iceTransportPolicy in the RTCConfiguration
dictionary.

Only two values are implemented:

* all: default behaviour
* relay: only gather relay candidates

The third member of the iceTransportPolicy enum, "public", is
obsolete.
2019-01-23 22:47:51 +00:00
Matthew Waters
26a5cbddbb webrtcbin: only change the receive state after setting the dtls-client
Doing so before will cause SSL errors with fast-start implementations
like Chrome or if media data arrives before an answer.
2018-12-19 00:44:06 +00:00
Matthew Waters
0a3f662ed6 webrtc: A couple of documentation fixes
set-*-description only takes the combined GstWebRTCSessionDescription
object
2018-12-19 00:44:06 +00:00
Matthew Waters
57a006d8a5 tests/webrtc: use the existing functions in the plugin
Instead of redefining our own, use the function implementations in
webrtcsdp.c and utils.c
2018-11-26 17:13:08 +11:00
Matthew Waters
14ee6f9d35 webrtc: fix typo in RTCRemoteOutboundRTPStreamStats 2018-11-26 16:21:58 +11:00
Matthew Waters
a42fdbb012 webrtc: add a few comments on bundle and src pad exposure 2018-11-26 16:21:19 +11:00
Matthew Waters
6f91a191de webrtcbin: factor out dtls fingerprint setting 2018-11-26 16:20:02 +11:00
Matthew Waters
3a2566c61f webrtc: remove extra 'pad' from log line 2018-11-26 16:12:03 +11:00
Matthew Waters
5ecca0bb22 webrtc: move some functions to the appropriate files 2018-11-26 16:07:57 +11:00
Luis de Bethencourt
16c1eee36f webrtcbin: options is a placeholder argument
Make it clear this unused argument is there for planned future use.
2018-10-26 15:15:57 +01:00
Luis de Bethencourt
83b29b813e webrtcbin: ws-semantic is not supported
Don't offer something that isn't supported.
2018-10-26 14:36:47 +01:00
Mathieu Duponchelle
9f684a2f81 webrtcbin: implement support for group: BUNDLE 2018-10-15 14:17:35 +02:00
Matthew Waters
7bf18ad258 webrtc: start in the closed state
This means that we will reject all operations before we've transitioned
into READY.

This also fixes the tests using the default GMainContext in the NULL
state instead of the webrtcbin internal GMainContext and thread.  Also
removes a potential ordering race where on the element transitioning to
READY, an operations could have been queued on two different threads and
removing a guarentee on operation ordering.
2018-10-08 21:56:31 +11:00
Aleix Conchillo Flaqué
c4fe52395b webrtcbin: start and stop thread when changing state
It might be possible that if we set webrtcbin to the NULL state some
tasks (idle sources) are still executed and they might even freeze. The freeze
is caused because the webrtcbin tasks don't hold a reference to webrtcbin and
if it's last unref inside the idle source itself this will not allow the main
loop to finish because the main loop is waiting on the idle source to finish.

We now start and stop webrtcbin thread when changing states. This will allow
the idle sources to finish properly.

https://bugzilla.gnome.org/show_bug.cgi?id=797251
2018-10-08 13:46:55 +11:00
Matthew Waters
07e9374eff webrtcbin: add support for data channels based on SCTP
Mostly follows the W3C specification
https://www.w3.org/TR/webrtc/#peer-to-peer-data-api

With contributions from:
Mathieu Duponchelle <mathieu@centricular.com>

https://bugzilla.gnome.org/show_bug.cgi?id=794351
2018-09-21 19:45:12 +10:00
Matthew Waters
cf46d49b1e webrtcbin: functionify dependent element checks 2018-09-21 19:36:52 +10:00
Mathieu Duponchelle
1d6160d59c webrtcbin: New add-turn-server API
It is possible and often desirable to pass multiple ICE relays
to libnice agents, the "turn-server" property, while convenient
to use from the command line, does not allow that.

This adds a new action signal, "add-turn-server" to address that.

https://bugzilla.gnome.org/show_bug.cgi?id=797012
2018-09-19 16:17:24 +02:00
Sam Gigliotti
1b095e9609 webrtc: fix memory leak
When it parses SDP, it doesn't free the error object.

https://bugzilla.gnome.org/show_bug.cgi?id=796830
2018-07-19 14:30:55 +01:00
Jan Schmidt
e6a564216d webrtc: Add a warning in sdp_media_from_transceiver()
When generating caps with no ssrc, at least throw a
warning instead of using an uninitialised stack variable

https://bugzilla.gnome.org/show_bug.cgi?id=796810
2018-07-15 23:07:21 +10:00
Jan Schmidt
76a93da2a7 webrtc: Fix memory leak
Fix a leaked string when building RTX info.
2018-07-14 23:20:13 +10:00
Mathieu Duponchelle
6fd3e2a400 find_codec_preferences: use received caps
When negotiation is triggered by receiving caps on our sink pad
probes, we could encounter a race condition where need-negotiation
is emitted and the application requires the creation of an offer
before the current caps were actually updated.

This led to retrieving incomplete caps when creating the offer,
using find_codec_preferences -> pad_get_current_caps.

Instead, as we save the caps in the probe callback anyway, it is better
and thread safe to use these if they were set.

https://bugzilla.gnome.org/show_bug.cgi?id=796801
2018-07-12 21:39:41 +02:00
Jan Schmidt
27b28f3aec webrtc: Explicitly initialise mutex and condition
Fixes random crashes when an allocated webrtcbin isn't
given fresh 0-filled memory in its allocation. It works
mostly because GMutex and GCond are automatically initialised
in that case.
2018-07-01 10:44:45 +10:00
Tim-Philipp Müller
6f46792f0f webrtc: Update for g_type_class_add_private() deprecation in recent GLib 2018-06-24 00:17:26 +02:00
Mathieu Duponchelle
33c7af8845 webrtcbin: copy sticky events on our ghostpads
This lets users call gst_pad_get_current_caps on newly-added
pads to easily determine what to plug them into.

We cannot copy sticky events unconditionally in core,
see #719437

https://bugzilla.gnome.org/show_bug.cgi?id=796387
2018-05-29 13:07:30 +02:00
Tim-Philipp Müller
ed7a98d45b webrtcbin: rtpstorage takes a 64-bit integer for "size-time" property
https://bugzilla.gnome.org/show_bug.cgi?id=796429
2018-05-28 10:43:37 +01:00
Mathieu Duponchelle
5c450c5992 webrtcbin: implement support for FEC and RTX
https://bugzilla.gnome.org/show_bug.cgi?id=795044
2018-05-09 14:46:14 +02:00
Sebastian Dröge
8375e33965 webrtcbin: Remove parameter from gst_webrtc_rtp_sender_new() 2018-03-16 11:07:01 +02:00
Matthew Waters
6f50d35246 webrtc: Fix ffeb09e4 conditional
Fixes ffeb09e4ab

if (sscanf(...)) {  // != 0
  error;
}

Is not correct where != 0 indicates some kind of success.

Check instead that the correct number of elements were slurped.
2018-02-08 15:51:35 +11:00
Matthew Waters
7e6b4dcb49 webrtc: change dead code to an assert
CID #1429140
2018-02-08 15:47:33 +11:00
Matthew Waters
ffeb09e4ab webrtc: bail on invalid rtpbin names
If we fail parsing rtpbin pad names, someone has screwed up so critical
and return.

CID #1429142
2018-02-08 15:29:56 +11:00
Matthew Waters
1894293d63 webrtcbin: an element that handles the transport aspects of webrtc connections
SDP's are generated and consumed according to the W3C PeerConnection API
available from https://www.w3.org/TR/webrtc/

The SDP is either created initially from the connected
sink pads/attached transceivers as in the case of generating an offer or
intersected with the connected sink pads/attached transceivers as in
the case for creating an answer.  In both cases, the rtp payloaded streams
sent by the peer are exposed as separate src pads.

The implementation supports trickle ICE, RTCP muxing, reduced size RTCP.

With contributions from:
Nirbheek Chauhan <nirbheek@centricular.com>
Mathieu Duponchelle <mathieu@centricular.com>
Edward Hervey <edward@centricular.com>

https://bugzilla.gnome.org/show_bug.cgi?id=792523
2018-02-02 15:02:21 +11:00