Doug Nazar [Wed, 7 Aug 2019 14:01:34 +0000 (10:01 -0400)]
matroska: Handle interlaced field order
Amr Mahdi [Wed, 7 Aug 2019 12:09:46 +0000 (12:09 +0000)]
wavparse: Fix ignoring of last chunk in push mode
In push mode (streaming), if the last audio payload chunk is less than the segment rate buffer size, it would be ignored since the plugin waits until it has at least segment rate bufer size of audio.
The fix is to introduce a flushing flag that indicates that no more audio will be available so that the plugin can recognize this condition and flush the data is has even if it is less
than the desired segment rate buffer size.
Robert Tiemann [Tue, 6 Aug 2019 14:27:37 +0000 (16:27 +0200)]
souphttpsrc: Log any error returned by soup_session_send()
luke.lin [Wed, 7 Aug 2019 02:42:21 +0000 (11:42 +0900)]
qtdemux: enlarge the maximal atom size
For 8K content, frame size is over 25MB, and cause the negotiation failure.
Enlarge the limitation of QTDEMUX_MAX_ATOM_SIZE to 32MB.
Mathieu Duponchelle [Sat, 27 Jul 2019 02:05:01 +0000 (04:05 +0200)]
rtspsrc: expose and implement is-live property
This is useful to support the ONVIF case: when is-live is set to
FALSE and onvif-rate-control is no, the client can control the
rate of delivery and arrange for the server to block and still
keep sending when unblocked, without requiring back and forth
PAUSE / PLAY requests. This enables, amongst other things, fast
frame stepping on the client side.
When is-live is FALSE, we don't use a manager at all. This case
was actually already pretty well handled by the current code. The
standard manager, rtpbin, is simply no longer needed in this case.
Applications can instantiate a downloadbuffer after rtspsrc if
needed.
Mathieu Duponchelle [Sat, 27 Jul 2019 02:03:44 +0000 (04:03 +0200)]
rtspsrc: reset_time when flush stopping
Mathieu Duponchelle [Fri, 12 Jul 2019 20:33:08 +0000 (22:33 +0200)]
rtspsrc: expose and implement onvif-mode property
Refactor the code for parsing and generating the Range, taking
advantage of existing API in GstRtspTimeRange.
Only use the TCP protocol in that mode, as per the specification.
Generate an accurate segment when in that mode, and signal to the
depayloader that it should not generate its own segment, through
the "onvif-mode" field in the caps, see
<https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/merge_requests/328>
for more information.
Translate trickmode seek flags to their ONVIF representation
Expose an onvif-rate-control property
Mathieu Duponchelle [Mon, 1 Jul 2019 18:38:20 +0000 (20:38 +0200)]
rtspsrc: improve handling of rate in seeks
Mathieu Duponchelle [Wed, 31 Jul 2019 19:55:16 +0000 (21:55 +0200)]
rtpfunnel: forward correct segment when switching pad
Forwarding a single segment event from the pad that first gets
chained is incorrect: when that first event was sent by an element
such as x264enc, with its offset start, we end pushing out of segment
buffers for the other pad(s).
Instead, everytime the active pad changes, forward the appropriate
segment event.
Fixes https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/issues/1028
Sebastian Dröge [Mon, 5 Aug 2019 16:35:36 +0000 (19:35 +0300)]
rtspsrc: Use new GstRTSPMessage API to set message body from a buffer directly
Antonio Ospite [Thu, 4 Apr 2019 11:17:34 +0000 (13:17 +0200)]
rtpsource: fix receiver source stats to consider previously queued packets
When it is not clear yet if a packet relative to a source should be
pushed, the packet is put into a queue, this happens in two cases:
- the source is still in probation;
- there is a large jump in seqnum, and it is not clear what
the cause is, future packets will help making a guess.
In either case stats about received packets are not updated at all; and
even if they were, when init_seq() is called it resets all receiver
stats, effectively loosing any possible stat about previously received
packets.
Fix this by taking into account the queued packets and update the stats
when calling init_seq().
Antonio Ospite [Tue, 9 Apr 2019 08:46:39 +0000 (10:46 +0200)]
rtpsource: clarify meaning of the octets-sent and octets-received stats
The octets-send and octets-received stats count the payload bytes
excluding RTP and lower level headers, clarify that in the
documentation.
Antonio Ospite [Thu, 4 Apr 2019 11:16:36 +0000 (13:16 +0200)]
rtpsource: expose field bytes_received in RTPSourceStats
Since commit
c971d1a9a (rtpsource: refactor bitrate estimation,
2010-03-02) bytes_received filed in RTPSourceStats is set but then never
used again, expose it so that it can be used by user code to verify how
many bytes have been received.
Antonio Ospite [Fri, 21 Jun 2019 15:46:36 +0000 (17:46 +0200)]
rtpmanager: consider UDP and IP headers in bandwidth calculation
According to RFC3550 lower-level headers should be considered for
bandwidth calculation.
See https://tools.ietf.org/html/rfc3550#section-6.2 paragraph 4:
Bandwidth calculations for control and data traffic include
lower-layer transport and network protocols (e.g., UDP and IP) since
that is what the resource reservation system would need to know.
Fix the source data to accommodate that.
Assume UDPv4 over IP for now, this is a simplification but it's good
enough for now.
While at it define a constant and use that instead of a magic number.
NOTE: this change basically reverts the logic of commit
529f443a6
(rtpsource: use payload size to estimate bitrate, 2010-03-02)
Seungha Yang [Thu, 1 Aug 2019 06:02:23 +0000 (15:02 +0900)]
qtdemux: Use empty-array safe way to cleanup GPtrArray
Fix assertion fail
GLib-CRITICAL **: g_ptr_array_remove_range: assertion 'index_ < rarray->len' failed
Marc Leeman [Thu, 1 Aug 2019 14:28:04 +0000 (14:28 +0000)]
rtpmp4vpay: config-interval -1 send at idr
adjust/port from rtph264pay and allow sending the configuration data at
every IDR
The payloader was stripping the configuration data when the
config-interval was set to 0. The code was written in such a way !(a >
0) that it stripped the config when it was set at -1 (send config_data
as soon as possible).
This resulted in some MPEG4 streams where no GOP/VOP-I was detected to
be sent out without configuration.
Doug Nazar [Sat, 27 Jul 2019 18:21:34 +0000 (14:21 -0400)]
matroskademux: Ignore crc32 element while peeking at cluster.
Guillaume Desmottes [Thu, 25 Jul 2019 15:51:26 +0000 (21:21 +0530)]
gtkglsink: fix crash when widget is resized after element destruction
Prevent _size_changed_cb() to be called after gtkglsink has been finalized.
Fix #632
Mathieu Duponchelle [Fri, 26 Jul 2019 00:45:51 +0000 (02:45 +0200)]
qtdemux: fix reverse playback EOS conditions
In reverse playback, we don't want to rely on the position of the current
keyframe to decide a stream is EOS: the last GOP we push will start with
a keyframe, which position is likely to be outside of the segment.
Instead, let the normal seek_to_previous_keyframe mechanism do its job,
it works just fine.
Mathieu Duponchelle [Mon, 22 Jul 2019 23:42:02 +0000 (01:42 +0200)]
qtdemux: fix key unit seek corner case
If a key unit seek is performed with a time position that matches
the offset of a keyframe, but not its actual PTS, we need to
adjust the segment nevertheless.
For example consider the following case:
* stream starts with a keyframe at 0 nanosecond, lasting 40 milliseconds
* user does a key unit seek at 20 milliseconds
* we don't adjust the segment as the time position is "over" a keyframe
* we push a segment that starts at 20 milliseconds
* we push a buffer with PTS == 0
* an element downstream (eg rtponviftimestamp) tries to calculate the
stream time of the buffer, fails to do so and drops it
Sebastian Dröge [Thu, 25 Jul 2019 12:08:54 +0000 (15:08 +0300)]
jpegdec: Don't dereference NULL input state if we have no caps in TIME segments
Simply assume that the JPEG frame is not going to be interlaced instead
of crashing.
Knut Andre Tidemann [Mon, 22 Jul 2019 08:28:50 +0000 (10:28 +0200)]
rtp: opuspay: fix memory leak in gst_rtp_opus_pay_setcaps.
The src caps were never dereferenced, causing a memory leak.
Mathieu Duponchelle [Fri, 12 Jul 2019 18:51:44 +0000 (20:51 +0200)]
qtdemux: implement support for trickmode interval
When the seek event contains a (newly-added) trickmode interval,
and TRICKMODE_KEY_UNITS was requested, only let through keyframes
separated with the required interval
Nirbheek Chauhan [Wed, 17 Jul 2019 13:42:19 +0000 (19:12 +0530)]
meson: Don't generate doc cache when no plugins are enabled
Fixes gst-build with -Dauto-features=disabled
Seungha Yang [Mon, 15 Jul 2019 14:24:05 +0000 (23:24 +0900)]
matroska: Port to color_{primaries,transfer,matrix}_to_iso
... and remove duplicated code.
Jan Schmidt [Sat, 25 May 2019 12:08:05 +0000 (22:08 +1000)]
splitmuxsink: add the ability to mux auxilliary video streams
The primary video stream is used to select fragment cut points
at keyframe boundaries. Auxilliary video streams may be
broken up at any packet - so fragments may not start with a keyframe
for those streams.
Jan Schmidt [Tue, 11 Jun 2019 13:17:30 +0000 (23:17 +1000)]
splitmuxsrc: Add video_%d pad template.
splitmuxsrc actually supports multiple video pads. Make that clear,
especially since it was already creating pads named "video_0" etc.
Mathieu Duponchelle [Tue, 9 Jul 2019 21:12:45 +0000 (23:12 +0200)]
qtdemux: fix conditions for end of segment in reverse playback
The time_position field of the stream is offset by the media_start
of its QtDemuxSegment compared to the start of the GstSegment of
the demuxer, take it into account when making comparisons.
Seungha Yang [Tue, 9 Jul 2019 14:06:12 +0000 (23:06 +0900)]
matroskademux: Fix mismatched transfer characteristic
TransferCharacteristics(18) should be ARIB STD-B67 (HLG)
See https://www.webmproject.org/docs/container/#TransferCharacteristics
Also map more color primaries indexes which have been handled by matroska-mux.
Seungha Yang [Tue, 9 Jul 2019 10:49:57 +0000 (19:49 +0900)]
v4l2: Remove misleading comments
gst_pad_template_new() does not take ownership of the caps
Olivier Crête [Wed, 23 Jan 2019 23:27:06 +0000 (18:27 -0500)]
rtp session: Add test for collision loopback detection
Ignore further collisions if the remote SSRC change with ours, it's
probably because someone is sending us back the packets we send out.
Olivier Crête [Wed, 23 Jan 2019 23:14:23 +0000 (18:14 -0500)]
rtpsession tests: Add test for third-party collision detection
Add tests to validate the code that ignores the same packets coming
from 2 different sources (an third-party collision).
Olivier Crête [Wed, 23 Jan 2019 22:19:15 +0000 (17:19 -0500)]
rtpsession: Add test for collision on incoming packets
Make sure that the collision is properly detected on incoming packets.
Olivier Crête [Wed, 23 Jan 2019 22:09:27 +0000 (17:09 -0500)]
rtpsession test: Verify that on-ssrc-collision message is emitted
Olivier Crête [Wed, 23 Jan 2019 21:58:22 +0000 (16:58 -0500)]
rtpsession: Also send conflict event when sending packet
If the conflict is detected when sending a packet, then also send an
upstream event to tell the source to reconfigure itself.
Also ignore the collision if we see more than one collision from the same
remote source to avoid problems on loops.
Song Bing [Mon, 15 Apr 2019 23:32:03 +0000 (16:32 -0700)]
v4l2transform: set right buffer count.
Set right buffer count to avoid one buffer.
Olivier Crête [Thu, 27 Jun 2019 23:47:41 +0000 (19:47 -0400)]
rtph265pay: Also immediately send packet if it is a suffix NAL
Immediately send packet if it contains any suffix NAL, this is required
in case they come after the VCL nal to not have to wait until the next frame.
Olivier Crête [Thu, 27 Jun 2019 23:46:01 +0000 (19:46 -0400)]
rtph265pay: Don't drop second byte of NAL header
At least keep 2 bytes per NAL even if the second one is 0, the
second byte of the NAL header could very well be 0.
Olivier Crête [Wed, 26 Jun 2019 20:42:44 +0000 (16:42 -0400)]
rtph26xpay: Avoid print when there is no bundle at end of packet
Olivier Crête [Wed, 26 Jun 2019 20:25:01 +0000 (16:25 -0400)]
rtph26xpay: Wait until there is a VCL or suffix NAL to send
With unit tests.
Olivier Crête [Wed, 19 Jun 2019 21:16:03 +0000 (17:16 -0400)]
rtph265pay test: Add unit tests for aggregation
Olivier Crête [Tue, 18 Jun 2019 23:07:38 +0000 (19:07 -0400)]
rtph265pay: Implement Aggregation packets
Align with rtph264pay
Olivier Crête [Tue, 18 Jun 2019 19:03:09 +0000 (15:03 -0400)]
rtph264pay test: Add unit tests for aggregation
Olivier Crête [Tue, 18 Jun 2019 17:45:15 +0000 (13:45 -0400)]
rtph264pay: Report latency when in maximal aggregation mode
Olivier Crête [Mon, 17 Jun 2019 15:31:53 +0000 (11:31 -0400)]
rtph264pay: Default to not adding latency when aggregating
Send the bundle as soon as there is one VCL unit in the packet at
the end of an incoming buffer.
The DELTA_UNIT flag is not reliable, so ignore it.
Olivier Crête [Fri, 14 Jun 2019 20:54:23 +0000 (16:54 -0400)]
rtp-payloading test: Fix working to 1.0 buffers instead of groups
Olivier Crête [Thu, 13 Jun 2019 22:07:35 +0000 (18:07 -0400)]
rtph265pay: Replace fragmentation while-loop with for-loop
Align with rtph264pay
Olivier Crête [Thu, 13 Jun 2019 21:42:05 +0000 (17:42 -0400)]
rtph265pay: Rename payload_len to max_fragment_size
Align to rtph264pay
Olivier Crête [Thu, 13 Jun 2019 21:30:08 +0000 (17:30 -0400)]
rtph265pay: Clean up _payload_nal
Move determining whether we need to fragment at all into the
fragmenter.
Align with rtph264pay
Olivier Crête [Thu, 13 Jun 2019 21:23:26 +0000 (17:23 -0400)]
rtph265pay: Extract sending fragments into _payload_nal_fragment
Align with rtph264pay
Olivier Crête [Thu, 13 Jun 2019 20:22:57 +0000 (16:22 -0400)]
rtph265pay: Extract sending a single packet into _payload_nal_single
Align with rtph264pay
Olivier Crête [Thu, 13 Jun 2019 20:14:31 +0000 (16:14 -0400)]
rtph265pay: Define and use FU_A_TYPE_ID
Align with rtph264pay
Olivier Crête [Thu, 13 Jun 2019 20:08:37 +0000 (16:08 -0400)]
rtph265pay: Use snake_case variables
Align with rtph264pay
Olivier Crête [Thu, 13 Jun 2019 20:04:39 +0000 (16:04 -0400)]
rtph265pay: Clean up whitespace and syntax
Align with rtph264pay
Jan Alexander Steffens (heftig) [Tue, 3 Jul 2018 17:39:25 +0000 (19:39 +0200)]
rtph264pay: Support STAP-A bundling
Add a new property "do-aggregate"* to the H.264 RTP payloader which
enables STAP-A aggregation as per [RFC-6184][1]. With aggregation enabled,
packets are bundled instead of sent immediately, up until the MTU size.
Bundles also end at access unit boundaries or when packets have to be
fragmented.
*: The property-name is kept generic since it might apply more widely,
e.g. STAP-B or MTAP.
[1]: https://tools.ietf.org/html/rfc6184#section-5.7
Closes https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/issues/434
Jan Alexander Steffens (heftig) [Mon, 5 Nov 2018 16:15:39 +0000 (17:15 +0100)]
rtph264pay: Fix delta-unit/discont handling when injecting SPS/PPS
Apply the wanted delta-unit and discont to the first packet; following
packets for this frame are always delta units and not discont.
Jan Alexander Steffens (heftig) [Mon, 5 Nov 2018 18:03:45 +0000 (19:03 +0100)]
rtph264pay: Replace fragmentation while-loop with for-loop
Jan Alexander Steffens (heftig) [Mon, 5 Nov 2018 17:57:38 +0000 (18:57 +0100)]
rtph264pay: Calculate the right max_fragments
Jan Alexander Steffens (heftig) [Mon, 5 Nov 2018 17:36:35 +0000 (18:36 +0100)]
rtph264pay: Rename payload_len to max_fragment_size
Jan Alexander Steffens (heftig) [Mon, 5 Nov 2018 17:34:40 +0000 (18:34 +0100)]
rtph264pay: Clean up _payload_nal_fragment
Jan Alexander Steffens (heftig) [Mon, 5 Nov 2018 17:06:19 +0000 (18:06 +0100)]
rtph264pay: Clean up _payload_nal
Move determining whether we need to fragment at all into the fragmenter.
Jan Alexander Steffens (heftig) [Mon, 5 Nov 2018 17:04:13 +0000 (18:04 +0100)]
rtph264pay: Clean up _payload_nal_single
Jan Alexander Steffens (heftig) [Mon, 5 Nov 2018 16:55:23 +0000 (17:55 +0100)]
rtph264pay: Extract sending fragments into _payload_nal_fragment
Jan Alexander Steffens (heftig) [Mon, 5 Nov 2018 16:49:52 +0000 (17:49 +0100)]
rtph264pay: Extract sending a single packet into _payload_nal_single
Jan Alexander Steffens (heftig) [Mon, 5 Nov 2018 16:10:03 +0000 (17:10 +0100)]
rtph264pay: Define and use FU_A_TYPE_ID
Jan Alexander Steffens (heftig) [Mon, 5 Nov 2018 16:07:06 +0000 (17:07 +0100)]
rtph264pay: Use snake_case variables
Jan Alexander Steffens (heftig) [Mon, 5 Nov 2018 16:04:14 +0000 (17:04 +0100)]
rtph264pay: Clean up whitespace and syntax
Olivier Crête [Thu, 6 Jun 2019 20:05:31 +0000 (16:05 -0400)]
rtpjitterbuffer: Unlock output if the queue is full
Thomas Bluemel [Sun, 30 Jun 2019 05:17:28 +0000 (23:17 -0600)]
rtpjitterbuffer: Ignore unsolicited rtx packets.
If an rtx packet arrives that hasn't been requested (it might
have been requested from prior to a reset), ignore it so that
it doesn't inadvertently trigger a clock skew.
Havard Graff [Sun, 30 Jun 2019 05:16:44 +0000 (23:16 -0600)]
rtpjitterbuffer: Add unit test for unsolicited rtx affecting skew
Thomas Bluemel [Thu, 13 Jun 2019 21:45:28 +0000 (15:45 -0600)]
rtpjitterbuffer: Only calculate skew or reset if no gap.
In the case of reordered packets, calculating skew would cause
pts values to be off. Only calculate skew when packets come
in as expected. Also, late RTX packets should not trigger
clock skew adjustments.
Fixes #612
Mart Raudsepp [Tue, 2 Jul 2019 18:21:05 +0000 (21:21 +0300)]
qtdemux: Provide a 30 frames lead-in for MP3
mpegaudioparse suggests MP3 needs 10 or 30 frames of lead-in (depending on
mpegaudioversion, which we don't know here), thus provide at least 30 frames
lead-in for such cases as a followup to commit
cbfa4531ee5ef.
Olivier Crête [Fri, 24 May 2019 14:31:39 +0000 (10:31 -0400)]
rtpjitterbuffer: max-dropout-time gets cast to int32
So any value over MAXINT32 gets considered as negative and is silently ignored.
Mathieu Duponchelle [Tue, 2 Jul 2019 11:00:32 +0000 (13:00 +0200)]
qtdemux: do_seek can never be called with a NULL event
Mathieu Duponchelle [Mon, 1 Jul 2019 20:38:41 +0000 (22:38 +0200)]
qtdemux: only adjust segment time when adjusting segment start
We ended up setting segment.time to segment.position when doing
reverse playback, which is obviously wrong.
Mathieu Duponchelle [Mon, 1 Jul 2019 11:54:13 +0000 (13:54 +0200)]
rtspsrc: unref the event in element seek handler
Mathieu Duponchelle [Fri, 28 Jun 2019 22:25:26 +0000 (00:25 +0200)]
rtspsrc: handle seek event on the element
Without this, the user has to wait for rtspsrc to have sent a PLAY
request and exposed its pads before seeking it.
Nicolas Dufresne [Wed, 26 Jun 2019 22:03:29 +0000 (18:03 -0400)]
multiudpsink: Add missing socket.h include
Without this include, macro like SO_BINDTODEVICE is not visible and
associated feature gets out-compiled. This also affects the support for
SO_SNDBUF.
Jan Alexander Steffens (heftig) [Mon, 24 Jun 2019 15:35:15 +0000 (17:35 +0200)]
flvmux: Clear new_tags if sending metadata in header
This avoids sending an additional metadata object right after the
headers.
Song Bing [Wed, 13 Jun 2018 21:55:29 +0000 (14:55 -0700)]
v4l2videodec: Fix drain() function return type
Return right type for drain() function.
Mart Raudsepp [Mon, 24 Jun 2019 11:28:39 +0000 (14:28 +0300)]
audioparsers: add back segment clipping to parsers that have lost it
The pre_push_frame default clipping behaviour was introduced in 2010
with commit
30be03004e82 and modified with commit
4163969a2422 in 2011,
when most parsers didn't implement a pre_push_frame yet. Not having it
meant that clipping was done by default. Those that did implement a
pre_push_frame (flacparse and mpegaudioparse) at the time, had the flag
adjusted as part of the 2011 refactor work.
All other parsers got a pre_push_frame vfunc implementation only in
2013, but seem to have forgot to keep the clipping behaviour, as
was done automatically when a pre_push_frame implementation doesn't
exist for the parser. aacparse lost it with commit
91d4abcea in
July 2013; the others in Dec 2013 as part of AUDIO_CODEC tag posting
in commits
6f89b430e,
d2ab5199b,
29f2cae12,
753d3c23a and
292780574.
Tim-Philipp Müller [Mon, 24 Jun 2019 09:42:31 +0000 (09:42 +0000)]
v4l2: fix compiler warning due to c99-ism
Jan Alexander Steffens (heftig) [Wed, 19 Jun 2019 12:28:28 +0000 (14:28 +0200)]
test: flvmux: Test changing caps with one sinkpad
These tests segfault without the preceding crash fix.
Jan Alexander Steffens (heftig) [Wed, 19 Jun 2019 12:08:06 +0000 (14:08 +0200)]
test: flvmux: Use gst_harness_sink_push_many
And check its return value.
Jan Alexander Steffens (heftig) [Wed, 19 Jun 2019 10:31:46 +0000 (12:31 +0200)]
flvmux: Simplify an if-else chain
Merge the identical branches and turn the condition around to make it
easier to read.
Jan Alexander Steffens (heftig) [Wed, 19 Jun 2019 10:28:22 +0000 (12:28 +0200)]
flvmux: Avoid crash when changing caps without both streams
mux->video_pad and mux->audio_pad can be NULL if the corresponding pad
has not been requested.
Sebastian Dröge [Wed, 12 Jun 2019 12:57:48 +0000 (15:57 +0300)]
rtpgstpay: Send caps anyway if caps are pending in the adapter but are different from the new ones
Otherwise it can happen that we receive a caps event, then another caps
event and only then buffers. We would then send out the first caps event
in the stream but mark buffers with the caps version of the second caps
event.
Sebastian Dröge [Wed, 12 Jun 2019 11:57:24 +0000 (14:57 +0300)]
rtpgstdepay: Only store the current caps and drop old caps immediately
Otherwise it can happen that we already collected 7 caps, miss the 8th
caps packet (packet loss) and then re-use the 1st caps for the following
buffers instead of the 8th caps which will likely cause errors further
downstream unless both caps are accidentally the same.
Keeping old caps around does not seem to have any value other than
potentially causing errors. We would always receive new caps whenever
they change (even if they were previous ones) and it's very unlikely
that they happen to be exactly the same as the previous ones.
Also after having received new caps or a buffer with a next caps
version, no buffers with old caps version will arrive anymore.
Jan Schmidt [Fri, 14 Jun 2019 16:00:43 +0000 (02:00 +1000)]
rtpjitterbuffer: Clear clock master before unreffing
Make sure to clear any master clock on the media_clock
before unreffing it to release the timer callback that's
updating the clock and keeping it reffed.
Jan Schmidt [Sun, 16 Jun 2019 01:07:31 +0000 (11:07 +1000)]
matroska: Initialise a video_context field to satisfy valgrind
Clear the mastering_display_info_present field explicitly
after reallocating the track context into a video context
to avoid uninitialised warnings in valgrind
Thibault Saunier [Fri, 14 Jun 2019 21:34:31 +0000 (17:34 -0400)]
docs: Fix link to strings
We can't link to #gchar* this way.
Mathieu Duponchelle [Thu, 13 Jun 2019 22:17:22 +0000 (00:17 +0200)]
jitterbuffer: unset DTS on output buffers
Mathieu Duponchelle [Wed, 22 May 2019 19:40:52 +0000 (21:40 +0200)]
splitmuxsink: set the same seqnum on flush_start / flush_stop
It's currently not made mandatory by aggregator, but it might
eventually be, and is more consistent behaviour
See https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/issues/977
Mikhail Fludkov [Thu, 13 Jun 2019 09:55:04 +0000 (11:55 +0200)]
rtpjitterbuffer: late packets shouldn't affect PTS of the following packet
If, say, a rtx-packet arrives really late, this can have a dramatic
effect on the jitterbuffer clock-skew logic, having it being reset
and losing track of the current dts-to-pts calculations, directly affecting
the packets that arrive later.
This is demonstrated in the test, where a RTX packet is pushed in really
late, and without this patch the last packet will have its PTS affected
by this, where as a late RTX packet should be redundant information, and
not affect anything.
Mikhail Fludkov [Wed, 12 Jun 2019 08:47:39 +0000 (10:47 +0200)]
rtpjitterbuffer: fix rtx delay calulation when large packet spacing
Stian Selnes [Thu, 24 Nov 2016 17:18:01 +0000 (18:18 +0100)]
rtpjitterbuffer: Fix delay for EXPECTED timers added by gaps
This patch corrects the delay set on EXPECTED timers that are added when
processing gaps. Previously the delay could be too small so that
'timout + delay' was much less than 'now', causing the following retries
to be scheduled too early. (They were sent earlier than
rtx-retry-timeout after the previous timeout.)
Havard Graff [Tue, 20 Nov 2018 15:11:12 +0000 (16:11 +0100)]
rtpjitterbuffer: don't try and calculate packet-rate if seqnum are jumping
Turns out that the "big-gap"-logic of the jitterbuffer has been horribly
broken.
For people using lost-events, an RTP-stream with a gap in sequencenumbers,
would produce exactly that many lost-events immediately.
So if your sequence-numbers jumped 20000, you would get 20000 lost-events
in your pipeline...
The test that looks after this logic "test_push_big_gap", basically
incremented the DTS of the buffer equal to the gap that was introduced,
so that in fact this would be more of a "large pause" test, than an
actual gap/discontinuity in the sequencenumbers.
Once the test was modified to not increment DTS (buffer arrival time) with
a similar gap, all sorts of crazy started happening, including adding
thousands of timers, and the logic that should have kicked in, the
"handle_big_gap_buffer"-logic, was not called at all, why?
Because the number max_dropout is calculated using the packet-rate, and
the packet-rate logic would, in this particular test, report that
the new packet rate was over 400000 packets per second!!!
I believe the right fix is to don't try and update the packet-rate if
there is any jumps in the sequence-numbers, and only do these calculations
for nice, sequential streams.
Havard Graff [Wed, 12 Jun 2019 09:16:22 +0000 (11:16 +0200)]
rtpjitterbuffer: fix unused variables
Jan Schmidt [Tue, 11 Jun 2019 16:42:42 +0000 (02:42 +1000)]
splitmuxsrc: Protect initial pad configuration with the object lock
gst_splitmux_src_activate_part() configures the pad information
before starting the pad task, but occasionally the changes it makes
to the pad are not seen in the pad task because they're not
protected by the right locking. Use the pad's object lock to
protect those variables.
Jan Schmidt [Tue, 11 Jun 2019 15:42:20 +0000 (01:42 +1000)]
splitmuxsrc: Restart pad task on a reconfigure
On a reconfigure event, restart streaming on the pad so
that switching tracks in playbin works cleanly