Sebastian Dröge [Fri, 10 Jan 2020 12:54:26 +0000 (14:54 +0200)]
imagefreeze: Add support for replacing the output buffer
By default imagefreeze will still reject new buffers after the first one
and immediately return GST_FLOW_EOS but the new allow-replace property
allows to change this.
Whenever updating the buffer we now also keep track of the configured
caps of the buffer and from the source pad task negotiate correctly
based on the potentially updated caps.
Only the very first time negotiation of a framerate with downstream is
performed, afterwards only the caps themselves apart from the framerate
are updated.
Alicia Boya García [Thu, 9 Jan 2020 18:43:02 +0000 (18:43 +0000)]
qtdemux: Fix race on pad reconnection
Elements emitting frames through several srcpads should use a
flow combiner to aggregate the chain returns and therefore only return
GST_FLOW_NOT_LINKED to upstream when all the downstream pads have
received GST_FLOW_NOT_LINKED.
In addition to that, in order to handle pads being relinked downstream,
the flow combiner should be reset in response to RECONFIGURE events.
This ensures that a both srcpads process a chain operation before a
GST_FLOW_NOT_LINKED can be propagated upstream (which would usually stop
the pipeline).
Otherwise, in a configuration with two srcpads, only one linked at a
time, after the relink the element could chain data through the now
unlinked pad and the flow combiner would resolve as GST_FLOW_NOT_LINKED
(stopping the pipeline) just because the now linked pad has not been
chained yet to update the flow combiner.
This patch adds handling of RECONFIGURE events to qtdemux. Also, since
this event handling causes the flow combiner to be used from a thread
other than the qtdemux streaming thread, usages of the flow combiner
has been guarded by the object lock.
Seungha Yang [Mon, 6 Jan 2020 16:20:24 +0000 (01:20 +0900)]
splitmuxsink: Fix assertion failure on set_property()
GValue might have null object.
(gst-inspect-1.0:10304): GStreamer-CRITICAL ...
gst_object_ref_sink: assertion 'object != NULL' failed
Daniel Molkentin [Fri, 3 Jan 2020 14:16:02 +0000 (15:16 +0100)]
videocrop: allow properties to be animated by GstController
Aaron Boxer [Tue, 24 Dec 2019 13:24:51 +0000 (08:24 -0500)]
rtspsrc: improved handling of control concatenation with base
Also, `control_url` variable has been renamed to `control_path`,
as it is actually a path.
Aaron Boxer [Fri, 6 Dec 2019 17:34:15 +0000 (12:34 -0500)]
rtspsrc: append aggregate control string to base URL before query string
Appending control string to end of query changes meaning of query string
Fixes #650
Eric Marks [Sat, 28 Dec 2019 23:01:19 +0000 (23:01 +0000)]
aasink & cacasink: add filter aatv & cacatv
Add transform filter capabilities to aasink and cacasink in the form of new elements aatv and cacatv.
Niels De Graef [Thu, 6 Jun 2019 09:03:34 +0000 (11:03 +0200)]
alpha: Cleanup using G_DECLARE_FINAL_TYPE
We started depending on GLib 2.44, so we can clean up all the GObject
boilerplate macros.
Stéphane Cerveau [Wed, 18 Dec 2019 15:07:18 +0000 (16:07 +0100)]
good: use of g_value_dup_string
Use helper method to get string from GValue.
Havard Graff [Thu, 19 Dec 2019 22:48:09 +0000 (23:48 +0100)]
rtpbin: fix shutdown crash in rtpbin
The key is to make sure the jitterbuffer is set to NULL *before* the
ptdemux.
The race that existed would basically happen when ptdemux had reached
READY, and the jitterbuffer would then push a buffer, triggering a new
pad with a new payloadtype being added and ghosted to the rtpbin itself.
However, the srcpad of the ptdemux would now be inactive, and all the
sticky-event pushed on it would be swallowed, not allowing any to reach
the ghost-pad. Then the buffer in-flight would come to the ghostpad,
and we would assert that a buffer arrived before the necessary
events.
By simply re-ordering the state-changes, we ensure that there will be
no buffer racing into the ptdemux while its state is being changed,
and the problem disappears completely.
Notice also that there is not point in disconnecting the signals on the
ptdemux before this point, since we need the push-thread to settle
down before we can do this in a non-racy way.
Aaron Boxer [Thu, 12 Sep 2019 20:22:10 +0000 (14:22 -0600)]
rtspsrc: avoid seek DISCONT when only rate changes in same direction
Not setting DISCONT avoids a noticable delay when seeking
with only rate changing, in the same direction as current
rate.
Olivier Crête [Tue, 10 Dec 2019 23:13:11 +0000 (18:13 -0500)]
rtspsrc: Remove deprecated GTimeVal
GTimeVal won't work past 2038
Olivier Crête [Tue, 10 Dec 2019 22:13:45 +0000 (17:13 -0500)]
osxaudio: Remove deprecated GTimeVal
Sebastian Dröge [Wed, 18 Dec 2019 10:19:27 +0000 (12:19 +0200)]
avimux: Add support for S24LE and S32LE raw audio
avidemux already handles this correctly.
Sebastian Dröge [Mon, 16 Dec 2019 19:07:08 +0000 (21:07 +0200)]
avimux: Allow muxing v210 video into AVI
avidemux already handles this.
Vivia Nikolaidou [Mon, 16 Dec 2019 16:43:44 +0000 (18:43 +0200)]
flvdemux: Don't replace video codec data when we receive a PAR
Receiving a pixel-aspect-ratio should trigger a caps change, but not
replace the existing video codec tag
Mathieu Duponchelle [Thu, 12 Dec 2019 19:20:35 +0000 (20:20 +0100)]
qtmux: protect access to GstElement.sinkpads
Mathieu Duponchelle [Tue, 3 Dec 2019 14:30:06 +0000 (15:30 +0100)]
qtmux: port to GstAggregator
Joakim Johansson [Mon, 16 Dec 2019 12:03:51 +0000 (13:03 +0100)]
gstrtspsrc: Add missing lock on free set_get_param_q
Otherwise is it possible to get a crash in gst_rtspsrc_set_parameter.
Sebastian Dröge [Thu, 12 Dec 2019 16:53:00 +0000 (18:53 +0200)]
splitmuxsink: Increment fragment_id even if no fragment location was provided
Applications might handle locations and generally configuration of the
sink by themselves instead of having splitmuxsink set the location on
the sink. Nonetheless it makes sense to increment the fragment_id that
is passed to the signal so that applications know which fragment is
requested.
Jan Alexander Steffens (heftig) [Thu, 12 Dec 2019 09:59:35 +0000 (10:59 +0100)]
flvmux: Use the last DTS for the metadata timestamp
This avoids creating a timestamp regression during a stream.
https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/merge_requests/429
Mathieu Duponchelle [Wed, 11 Dec 2019 16:30:50 +0000 (17:30 +0100)]
qtdemux: send GAP events for lagging audio and video streams too
The logic is taken straight from matroskademux, see
77403d0afee635f2de6c2e53a23e1f50ad0d00fa
Seungha Yang [Tue, 10 Dec 2019 14:48:35 +0000 (23:48 +0900)]
flvmux: Use thread-safe gmtime_r if available
gmtime on *nix is not thread-safe.
Stéphane Cerveau [Thu, 5 Dec 2019 14:58:40 +0000 (14:58 +0000)]
splitmuxsink: provides a start-index property
Allow to change the fragment-id start index.
Philipp Zabel [Tue, 3 Dec 2019 10:36:07 +0000 (11:36 +0100)]
qmlglsink: fix build on EGL platform without X11 headers
If Mesa is built without X11 headers, building against Mesa EGL headers
requires a dependency on egl.pc, to define MESA_EGL_NO_X11_HEADERS.
This fixes a build error when compiling ext/qt/gstqtglutility.cc:
In file included from /usr/include/EGL/egl.h:39,
from /usr/include/gstreamer-1.0/gst/gl/egl/gstegl.h:44,
from ../gst-plugins-good-1.16.1/ext/qt/gstqtglutility.cc:43:
/usr/include/EGL/eglplatform.h:124:10: fatal error: X11/Xlib.h: No such file or directory
Tim-Philipp Müller [Wed, 4 Dec 2019 01:03:49 +0000 (01:03 +0000)]
rtpjpegdepay: outputs framed jpeg
Add parsed=true to output caps, as we always output
whole frames, timestamped and all. Means also that
the output can be decoded by avdec_mjpeg wihout
plugging an extra parser (which has no rank).
Jan Alexander Steffens (heftig) [Tue, 3 Dec 2019 12:47:22 +0000 (13:47 +0100)]
flvmux: Correct metadata handling in file and stream mode
In file mode, only push one onMetaData at the start of the stream.
In stream mode, always push complete onMetaData. They get replaced, not
merged.
https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/merge_requests/418
Jan Alexander Steffens (heftig) [Tue, 3 Dec 2019 12:46:09 +0000 (13:46 +0100)]
flvmux: Don't calculate duration in streamable mode
There's no header to rewrite, so the duration is left unused.
https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/merge_requests/418
Havard Graff [Wed, 30 Nov 2016 14:55:01 +0000 (15:55 +0100)]
rtpL16depay: don't crash if data is not modulo channels*width
Tim-Philipp Müller [Mon, 2 Dec 2019 19:00:45 +0000 (19:00 +0000)]
pkgconfig: remove gst-plugins-good-1.0-uninstalled.pc
This was never installed and it was only used by the uninstalled
autotools dev environment to locate the -good plugins for use
in unit tests in gstreamer modules higher up the stack.
It is no longer needed now that we no longer have an autotools build.
Håvard Graff [Tue, 10 Oct 2017 13:45:28 +0000 (15:45 +0200)]
meson.build: use join_paths() on prefix
So that "/" are correct on Windows.
Havard Graff [Fri, 30 Jun 2017 07:48:58 +0000 (09:48 +0200)]
rtpopuspay: use baseclass allocator for buffers
That way we get some of the meta -> rtp-extension goodies.
Seungha Yang [Fri, 29 Nov 2019 11:46:26 +0000 (20:46 +0900)]
vp9dec: Fix broken 4:4:4 8bits decoding
VPX_IMG_FMT_I444 pixel format with sRGB colorspace means
GBR data.
Fixes: https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/issues/651
Havard Graff [Fri, 18 Oct 2019 15:45:43 +0000 (17:45 +0200)]
rtpsession: add test for requesting FIR after having requested PLI
Havard Graff [Tue, 26 Nov 2019 14:00:18 +0000 (15:00 +0100)]
rtpjitterbuffer: make test more stable
Havard Graff [Fri, 29 Nov 2019 13:23:49 +0000 (14:23 +0100)]
rtpsession: add locking for clear-pt-map
...or it will segfault from time to time...
Linus Svensson [Thu, 31 May 2018 08:29:43 +0000 (10:29 +0200)]
matroskamux: Add property to set DateUTC
Add a property that makes it possible for an application to set the
DateUTC header field in matroska files. This is useful for live feeds,
where the DateUTC header can be set to a UTC timestamp, matching the
beginning of the file.
Needs gstreamer!323
Fixes https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/issues/481
Linus Svensson [Thu, 31 May 2018 09:20:36 +0000 (11:20 +0200)]
matroskamux: Use nanosecond precision for DateUTC
DateUTC is specified with nanosecond precision in matroska, make use of
that.
Nicolas Dufresne [Wed, 17 Oct 2018 02:28:13 +0000 (02:28 +0000)]
v4l2bufferpool: Queue number of allocated buffers to capture
Before we do streamon, we queue all capture buffers by calling
resurrect. When the driver supports CREATE_BUFS, this would lead
to buffers being allocated till the maximum of 32 is reached.
Instead, we now save the number of allocated buffers and queue this
amount.
Jan Alexander Steffens (heftig) [Tue, 19 Nov 2019 13:23:48 +0000 (14:23 +0100)]
matroskamux: Pass the right size to gst_collect_pads_add_pad
We were lucky that GstMatroskamuxPad is larger than GstMatroskaPad.
https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/merge_requests/393
Nicolas Dufresne [Mon, 18 Nov 2019 18:27:42 +0000 (13:27 -0500)]
v4l2object: Workaround bad TRY_FMT colorimetry implementation
libv4l2 reset the colorpace to 0 and does not do any request to the
driver. This yields an invalid colorspace which currently cause a
negotiation failure. This workaround by ignoring bad values during the
TRY_FMT step.
aogun [Mon, 4 Nov 2019 09:18:30 +0000 (17:18 +0800)]
aacparse: fix wrong offset of adts channel
Seungha Yang [Mon, 7 Oct 2019 03:45:00 +0000 (12:45 +0900)]
splitmuxsink: Don't take lock during posting message
An application might try to access splitmuxsink from sync message handler
by g_object_{get,set} which takes lock also. In general, we don't
take lock around message handler.
Scott Kanowitz [Thu, 12 Sep 2019 19:21:24 +0000 (15:21 -0400)]
jpegdec: Fix incorrect logic in EOI tag detection
This change fixes the reversed logic in the EOI tag detection
code.
Niels De Graef [Mon, 26 Aug 2019 06:03:24 +0000 (08:03 +0200)]
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.
Nicolas Dufresne [Thu, 14 Nov 2019 22:33:08 +0000 (17:33 -0500)]
rtpjitterbuffer: Check the exit condition after executing timers
The do_expected_timeout() function may release the JBUF_LOCK, so we need
to check if nothing wanted the timer thread to exit after this call.
The side effect was that we may endup going back into waiting for a timer
which will cause arbitrary delay on tear down (or deadlock when test
clock is used).
Fixes #653
Nicolas Dufresne [Thu, 14 Nov 2019 22:20:51 +0000 (17:20 -0500)]
rtpjitterbuffer: Check exit condition immediately after JBUF_WAIT
JBUF_WAIT_QUEUE drops the JBUF_LOCK, which means the stop condition
for the chain function may have changed (change_state to NULL). Check
this immediately after the wait so that we don't delay shutting down.
Nicolas Dufresne [Tue, 12 Nov 2019 22:28:22 +0000 (17:28 -0500)]
videocrop: Also update the coordinate when in-place
This update is needed when the output caps is not changed (e.g. we are
moving a viewport around).
Fixes #669
Nicolas Dufresne [Mon, 11 Nov 2019 18:19:08 +0000 (13:19 -0500)]
videocrop: Don't always re-run the allocation query
When in-place, running an allocation is not useful since videocrop
is not implicated in the allocation. So only force the allocation
query for the case it was in passthrough. This is needed since the
change in the crop region will likely pull us out of this mode. For the
case we where neither in passthrough or in-place, the allocation query
is already ran by the baseclass, so nothing special is needed.
This fixes performance issues when changing the crop region per frame.
This was reproduced using videocrop2-test.
Nicolas Dufresne [Mon, 11 Nov 2019 18:18:52 +0000 (13:18 -0500)]
videocrop: Cleanup spurious assignment
These are just writing the same thing a second time.
Michael Olbrich [Wed, 7 Nov 2018 08:00:02 +0000 (09:00 +0100)]
jpegdec: don't overwrite the last valid line
If the the height is not a multiple of the macro block size then the memory
of the last line is reused for all extra lines. This is no problem if the
last line is duplicated properly. However, if the extra lines are not
initialized properly during encoding, then the last visible line is
overwritten with undefined data.
Use a extra buffer to avoid this problem.
Stéphane Cerveau [Thu, 7 Nov 2019 11:28:58 +0000 (12:28 +0100)]
splitmuxsink: add fakesink support
fakesink does not support "location" property and was generating
a warning.
Sergey Nazaryev [Wed, 12 Dec 2018 16:07:39 +0000 (19:07 +0300)]
multiudpsink: don't lose scope_id
Nirbheek Chauhan [Tue, 5 Nov 2019 16:11:55 +0000 (21:41 +0530)]
vpx: Error out if enabled and no features found
Seee: https://gitlab.freedesktop.org/gstreamer/cerbero/issues/200
Guillaume Desmottes [Sat, 25 May 2019 19:19:21 +0000 (21:19 +0200)]
v4l2object: update match_buffer_layout() debug messages
It's no longer used only to try importing buffers.
Guillaume Desmottes [Thu, 23 May 2019 08:49:39 +0000 (10:49 +0200)]
v4l2object: try matching buffer layout from downstream
Ask v4l2 to produce buffers matching the buffer layout requested
downstream.
Guillaume Desmottes [Tue, 21 May 2019 08:31:46 +0000 (10:31 +0200)]
v4l2object: factor out gst_v4l2_object_match_buffer_layout()
No semantic change.
Havard Graff [Sun, 20 Oct 2019 10:17:25 +0000 (12:17 +0200)]
rtpjitterbuffer: make sure not to drop packets based on skew
One of the jitterbuffers functions is to try and make sense of weird
network behavior.
It is quite unhelpful for the jitterbuffer to start dropping packets
itself when what you are trying to achieve is better network resilience.
In the case of a skew, this could often mean the sender has restarted
in some fashion, and then dropping the very first buffer of this "new"
stream could often mean missing valuable information, like in the case
of video and I-frames.
This patch simply reverts back to the old behavior, prior to https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/commit/
8d955fc32b552b2db933c67f3cfa31d987f36b81
and includes the simplest test I could write to demonstrate the behavior,
where a single packet arrives "perfectly", then a 50ms gap happens,
and then two more packets arrive in perfect order after that.
# Conflicts:
# tests/check/elements/rtpjitterbuffer.c
Guillaume Desmottes [Wed, 17 Apr 2019 07:10:22 +0000 (12:40 +0530)]
v4l2transform: use alignments from upstream when importing on sink
Try configuring the v4l2 output with the alignments from upstream when
importing its buffers. This allows us to support importing with
non-standard strides and/or heights if supported by the driver.
Guillaume Desmottes [Wed, 17 Apr 2019 06:55:14 +0000 (12:25 +0530)]
v4l2object: add support for vertical padding when importing buffers
We were already supporting horizontal padding by setting bytesperline to
the buffer stride but not vertical one.
We are now updating the format height with the padded height and crop to
the actual video resolution if needed.
Guillaume Desmottes [Wed, 17 Apr 2019 06:16:10 +0000 (11:46 +0530)]
v4l2object: fix debug message if driver rejects stride
The 'want' and 'got' strides were inversed.
Guillaume Desmottes [Mon, 15 Apr 2019 06:13:41 +0000 (11:43 +0530)]
v4l2: improve logs when importing buffers
Log strides and offsets from upstream.
Also fix a typo.
James Cowgill [Tue, 29 Oct 2019 14:05:48 +0000 (14:05 +0000)]
v4l2videodec: ensure pool exists before orphaning it
In commit
e2ff87732d0b ("v4l2videodec: support orphaning") support for
orphaning the capture buffer pool was added when the format is
renegotiated. However, the commit forgot to check that a pool existed
before doing this. This is needed because it's possible for the format
to be renegotiated before a capture pool is allocated, which would
result in trying to orphan a NULL pool and lead to a NULL pointer
dereference.
Fix this by checking a pool exists first. If the pool doesn't exist,
there are no buffers to be reclaimed, so skip the allocation query in
that case.
Matthew Waters [Fri, 25 Oct 2019 11:03:18 +0000 (22:03 +1100)]
qmlglsrc: read from the back buffer when use-default-fbo = TRUE
glReadBuffer(GL_COLOR_ATTACHMENT0) on the default framebuffer (0) is
invalid GL API usage and would result in a GL error being thrown.
Matthew Waters [Fri, 25 Oct 2019 10:47:01 +0000 (21:47 +1100)]
qmlglsrc: fix vertical flip matrix
Some time ago libgstgl defined the majorness of matrices it uses.
The majorness used by qmlglsrc was incompatible with the libgstgl.
Patricia Muscalu [Tue, 30 Jul 2019 10:07:18 +0000 (12:07 +0200)]
qtmux: Fix memory leak while pushing fragmented data
The memory leak occurs in the case when the buffer has been
added to the fragment_buffers array of the current pad and
never been sent because of the push failure of the previous
buffers: moof or mdat header or fragmented buffer(s).
Edward Hervey [Fri, 11 Oct 2019 12:20:15 +0000 (14:20 +0200)]
good: Avoid usage of deprecated API
GTimeval and related functions are now deprecated in glib.
Replacement APIs have been present since 2.26
Javier Celaya [Mon, 15 Jul 2019 05:46:56 +0000 (07:46 +0200)]
osxaudio: misspelled dependency
When building osxaudio, the required 'AudioToolbox' dependency is
misspelled as 'AudioToolBox', which crashes the build with error:
ld: framework not found AudioToolBox
Tim-Philipp Müller [Sat, 8 Jun 2019 23:43:00 +0000 (00:43 +0100)]
Remove autotools build system
Tim-Philipp Müller [Sun, 13 Oct 2019 11:46:58 +0000 (12:46 +0100)]
v4l2videoenc: fix wrong type cast
Follow-up to commit
1b752c0f !361
HuQian [Wed, 25 Sep 2019 12:36:32 +0000 (12:36 +0000)]
is a typo here? gstv4l2object.c
Kevin Song [Fri, 11 Oct 2019 12:27:12 +0000 (12:27 +0000)]
v4l2videodec: Check stop in flush() to avoid race condition.
Backward playback will drain and flush every frame. Stop playback
when backward playback have race condition between exit thread and
streaming thread flush. Add one check to avoid it.
Fixes #639
Fuwei Tang [Fri, 11 Oct 2019 02:33:20 +0000 (10:33 +0800)]
v4l2videoenc: fix type conversion errors
Aaron Boxer [Mon, 2 Sep 2019 12:27:35 +0000 (08:27 -0400)]
documentation: fix a number of typos
Simon Arnling Bååth [Fri, 4 Oct 2019 20:31:56 +0000 (20:31 +0000)]
gstrtpjitterbuffer: Custom messages when dropping packets
This commit adds custom element messages for when gstrtpjitterbuffer
drops an incoming rtp packets due to for example arriving too late.
Applications can listen to these messages on the bus which enables
actions to be taken when packets are dropped due to for example high
network jitter.
Two properties has been added, one to enable posting drop messages and
one to set a minimum time between each message to enable throttling the
posting of messages as high drop rates.
Thibault Saunier [Tue, 3 Sep 2019 20:46:30 +0000 (16:46 -0400)]
qtdemux: Specify REDIRECT information in error message
There are in the wild (mp4) streams that basically contain no tracks
but do have a redirect info[0], in which case, we won't be able
to expose any pad (there are no tracks) so we can't post anything but
an error on the bus, as:
- it can't send EOS downstream, it has no pad,
- posting an EOS message will be useless as PAUSED state can't be
reached and there is no sink in the pipeline meaning GstBin will
simply ignore it
The approach here is to to add details to the ERROR message with a
`redirect-location` field which elements like playbin handle and use right
away.
[0]: http://movietrailers.apple.com/movies/paramount/terminator-dark-fate/terminator-dark-fate-trailer-2_480p.mov
Olivier Crête [Thu, 26 Sep 2019 22:39:48 +0000 (18:39 -0400)]
rtpjitterbuffer: Cancel timers instead of just unlocking loop thread
When the queue is full (and adding more packets would risk a seqnum
roll-over), the best approach is to just start pushing out packets
from the other side. Just pushing out the packets results in the
timers being left hanging with old seqnums, so it's safer to just
execute them immediately in this case. It does limit the timer space
to the time it takes to receiver about 32k packets, but without
extended sequence number, this is the best RTP can do.
This also results in the test no longer needed to have timeouts or
timers as pushing packets in drives everything.
Fixes #619
Nicolas Dufresne [Fri, 27 Sep 2019 18:04:28 +0000 (14:04 -0400)]
rtpjitterbuffer: Optimize offset update
As we are applying the same offset over all timers, there timer
ordering won't change, so we can safely skip time-reordering.
Nicolas Dufresne [Fri, 27 Sep 2019 20:21:22 +0000 (16:21 -0400)]
rtptimerqueue: Optimize reschedule optations
This basically add ability to choose between inserting from head, tail
or in-place in order to try and minimize the distance to walk through in
the timer queue. This removes an overhead we had seen on high drop rate.
Nicolas Dufresne [Fri, 27 Sep 2019 18:04:03 +0000 (14:04 -0400)]
rtpjitterbuffer: Fix a typo in comment
Nicolas Dufresne [Tue, 2 Jul 2019 19:52:25 +0000 (15:52 -0400)]
rtpjitterbuffer: Don't use stats timer on the timers queue
The timer passed to update_timers may be from the stats timer. At the
moment, we could endup rescheduling (reusing) that timer onto the normal
timer queue, unschedul it as if it was from the normal timer queue or
duplicate it into the stats timer queue again. This was protected before
as the with the fact the stats timer didn't have a valid idx.
Nicolas Dufresne [Fri, 21 Jun 2019 18:08:26 +0000 (14:08 -0400)]
rtpjitterbuffer: Update timers on ts-offset changes
As the offset is already applied now, we need to update and reschedule
all timers each time the offset is changed. I'm not sure who expect this
to be retro-actively applied, but there was a unit test for it.
Nicolas Dufresne [Thu, 20 Jun 2019 19:59:48 +0000 (15:59 -0400)]
rtpjitterbuffer: No need to wake the timer thread on head changes
If the jitterbuffer head change, there is no need to systematically
wakeup the timer thread. The timer thread will be waken up on if
an earlier timeout has been pushed. This prevent some more spurious
wakeup when the system is loaded. As a side effect, cranking the clock
may set the clock at an earlier position.
Nicolas Dufresne [Tue, 18 Jun 2019 23:07:29 +0000 (19:07 -0400)]
rtpjittterbuffer: Port timers array to RtpTimerQueue
In this patch we now make use of the new RtpTimerQueue instead of the
old GArray. This required a lot of changes all over the place, some of
the important changes are that `timer->timeout` is no longer a PTS but
the actual timeout. This was required to get the RtpTimerQueue sorting
right. The applied offset is saved as `timer->offset`, this allow
retreiving back the PTS when needed.
The clockid updates only happens once per incoming packet. If the
currently schedule timer is before the earliest timer in the queue, we
no longer wakeup the thread. This way, if other timers get setup in the
meantime, this will reduce the number of wakup.
The timer loop code has been mostly rewritten, though the behaviour of
running the lost timers first has been kept (even though there is no
test to show what would be the side effect of doing this differently).
Fixes #608
Nicolas Dufresne [Fri, 14 Jun 2019 18:29:36 +0000 (14:29 -0400)]
rtpjittterbuffer: Port from TimerQueue to RtpTimerQueue
Nicolas Dufresne [Thu, 13 Jun 2019 21:08:31 +0000 (17:08 -0400)]
rtpjitterbuffer: Port use the new RtpTimer structure
First iteration toward porting to the new timer queue.
Nicolas Dufresne [Wed, 12 Jun 2019 13:59:31 +0000 (09:59 -0400)]
rtptimerqueue: Consolidate a data structure for timers
Implement a single timer queue for all timers. The goal is to always use
ordered queues for storing timers. This way, extracting timers for
execution becomes O(1). This also allow separating the clock wait
scheduling from the timer itself and ensure that we only wake up the
timer thread when strictly needed.
The knew data structure is still O(n) on insertions and reschedule,
but we now use proximity optimization so that normal cases should be
really fast. The GList structure is also embeded intot he RtpTimer
structure to reduce the number of allocations.
Nicolas Dufresne [Mon, 10 Jun 2019 20:46:05 +0000 (16:46 -0400)]
tests: jitterbuffer: Demacroify some helpers
There is no reason for these to be macros anymore. This makes the
test helper much more readable.
Nicolas Dufresne [Thu, 6 Jun 2019 18:44:27 +0000 (14:44 -0400)]
rtpjitterbuffer: Move item structure outside of the element
This moves the RtpJitterBufferStructure type, alloc, free into
rtpjitterbuffer.c/h implementation. jitterbuffer.c strictly rely on
the fact this structure is compatible with GList, and so it make more
sense to keep encapsulate it. Also, anything that could possibly
reduce the amount of code in the element is a win.
In order to support that move, a function pointer to free the data
was added. This also allow making the free function option when
flushing the jitterbuffer.
Nicolas Dufresne [Thu, 6 Jun 2019 17:09:29 +0000 (13:09 -0400)]
rtpjitterbuffer: Constify timer pointers where possible
This helps understanding which function modify the Timerdata
and which one does not. This is not always obvious from thelper
name considering recalculate_timer() does not.
Philipp Zabel [Fri, 27 Sep 2019 06:46:22 +0000 (08:46 +0200)]
v4l2: Add MPEG-2 profile and level support
Add support for V4L2 MPEG-2 decoders reporting supported profiles and
levels.
Philipp Zabel [Mon, 23 Sep 2019 12:34:20 +0000 (14:34 +0200)]
v4l2object: add support for ABGR, xBGR, RGBA, and RGBx formats
Map them to the new V4L2_PIX_FMT_{BGRA32,BGRX32,RGBA32,RGBX32} pixel
formats.
Philipp Zabel [Mon, 23 Sep 2019 12:10:15 +0000 (14:10 +0200)]
v4l2: update kernel headers to latest from media tree
Update to the latest installed headers (output of make headers_install)
from the media tree, keeping the slight modifications to the includes.
This includes typo fixes in enum v4l2_mpeg_video_multi_slice_mode,
MPEG-2 level and profile enums, new FWHT and H.264 Qp controls, new
RGB(A) formats, and new continuous bytestream and dynamic resolution
format flags.
Mathieu Duponchelle [Tue, 19 Dec 2017 17:23:16 +0000 (18:23 +0100)]
rtpbin: add request-jitterbuffer signal
This can be used to pass the threadsharing jitterbuffer from
gst-plugins-rs for example.
Matthew Waters [Mon, 23 Sep 2019 08:46:16 +0000 (18:46 +1000)]
build: fix werror build with newer gcc
In file included from ../../../../dist/linux_x86_64/include/gstreamer-1.0/gst/gst.h:55,
from ../../../../dist/linux_x86_64/include/gstreamer-1.0/gst/tag/tag.h:25,
from ../gst/isomp4/qtdemux.c:56:
In function ‘qtdemux_inspect_transformation_matrix’,
inlined from ‘qtdemux_parse_trak’ at ../gst/isomp4/qtdemux.c:10676:5,
inlined from ‘qtdemux_parse_tree’ at ../gst/isomp4/qtdemux.c:14210:5:
../../../../dist/linux_x86_64/include/gstreamer-1.0/gst/gstinfo.h:645:5: error: ‘%s’ directive argument is null [-Werror=format-overflow=]
645 | gst_debug_log ((cat), (level), __FILE__, GST_FUNCTION, __LINE__, \
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
646 | (GObject *) (object), __VA_ARGS__); \
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
../../../../dist/linux_x86_64/include/gstreamer-1.0/gst/gstinfo.h:1062:35: note: in expansion of macro ‘GST_CAT_LEVEL_LOG’
1062 | #define GST_DEBUG_OBJECT(obj,...) GST_CAT_LEVEL_LOG (GST_CAT_DEFAULT, GST_LEVEL_DEBUG, obj, __VA_ARGS__)
| ^~~~~~~~~~~~~~~~~
../gst/isomp4/qtdemux.c:10294:5: note: in expansion of macro ‘GST_DEBUG_OBJECT’
10294 | GST_DEBUG_OBJECT (qtdemux, "Transformation matrix rotation %s",
| ^~~~~~~~~~~~~~~~
../gst/isomp4/qtdemux.c: In function ‘qtdemux_parse_tree’:
../gst/isomp4/qtdemux.c:10294:64: note: format string is defined here
10294 | GST_DEBUG_OBJECT (qtdemux, "Transformation matrix rotation %s",
| ^~
Sebastian Dröge [Wed, 18 Sep 2019 15:31:27 +0000 (18:31 +0300)]
qtmux: Use the new helper functions for mapping the colr atom values to colorimetry
Sebastian Dröge [Wed, 18 Sep 2019 15:29:27 +0000 (18:29 +0300)]
qtdemux: Use the new helper functions for mapping the colr atom values to colorimetry
Mathieu Duponchelle [Tue, 10 Sep 2019 20:44:20 +0000 (22:44 +0200)]
docs: update plugin cache
Mathieu Duponchelle [Tue, 10 Sep 2019 20:43:49 +0000 (22:43 +0200)]
smpte: don't register transition types twice
Doug Nazar [Mon, 9 Sep 2019 00:43:17 +0000 (20:43 -0400)]
alpha: Fix one_over_kc calculation
On arm/aarch64, converting from float directly to unsigned int uses
a different opcode and negative numbers result in 0. Cast to
signed int first.