Tim-Philipp Müller [Sun, 21 May 2017 10:44:37 +0000 (11:44 +0100)]
gl: x11: fix compiler warning
Matthew Waters [Sun, 21 May 2017 10:03:01 +0000 (12:03 +0200)]
glcontext: add public swap_buffers function
That simply calls the implementation
Tim-Philipp Müller [Sun, 21 May 2017 09:57:18 +0000 (10:57 +0100)]
examples: gl: remove ancient bitrotten .vcproj files
Graham Leggett [Sat, 20 May 2017 16:22:10 +0000 (16:22 +0000)]
adaptivedemux: fix debug message printf format
Match gsize to G_GSIZE_FORMAT in the debug message.
https://bugzilla.gnome.org/show_bug.cgi?id=782873
Tim-Philipp Müller [Sat, 20 May 2017 16:04:52 +0000 (17:04 +0100)]
win32: update .def file for API removal
Olivier Crête [Sat, 20 May 2017 15:47:04 +0000 (17:47 +0200)]
audiointerleave: Take object lock while modifying channel count
Olivier Crête [Sat, 20 May 2017 13:56:16 +0000 (15:56 +0200)]
aggregator: Remove unused GST_FLOW_NOT_HANDLED
Matthew Waters [Sat, 20 May 2017 12:24:57 +0000 (14:24 +0200)]
aggregator: add simple support for caps handling
Modelled off the videoaggregator caps handling as that seems the most
mature aggregtor-using implementation that has caps handling there is.
https://bugzilla.gnome.org/show_bug.cgi?id=776931
Nicolas Dufresne [Sat, 20 May 2017 11:10:53 +0000 (13:10 +0200)]
aggregator: Reset upstream latency on first buffer
In the case an aggregator is created and pads are requested but only
linked later, we end up never updating the upstream latency.
This was because latency queries on pads that are not linked succeed,
so we never did a new query once a live source has been linked, so the
thread was never started.
https://bugzilla.gnome.org/show_bug.cgi?id=757548
George Kiagiadakis [Tue, 4 Apr 2017 08:25:43 +0000 (11:25 +0300)]
videoaggregator: delay using new caps from a sink pad until the next buffer in the queue is taken
When caps changes while streaming, the new caps was getting processed
immediately in videoaggregator, but the next buffer in the queue that
corresponds to this new caps was not necessarily being used immediately,
which resulted sometimes in using an old buffer with new caps. Of course
there used to be a separate buffer_vinfo for mapping the buffer with its
own caps, but in compositor the GstVideoConverter was still using wrong
info and resulted in invalid reads and corrupt output.
This approach here is more safe. We delay using the new caps
until we actually select the next buffer in the queue for use.
This way we also eliminate the need for buffer_vinfo, since the
pad->info is always in sync with the format of the selected buffer.
https://bugzilla.gnome.org/show_bug.cgi?id=780682
Olivier Crête [Sat, 14 May 2016 13:52:37 +0000 (15:52 +0200)]
aggregator: Always handle sync'ed events on output thread
Having all synchronized events always be handled on the output
thread should make synchronization easier.
https://bugzilla.gnome.org/show_bug.cgi?id=781673
Olivier Crête [Wed, 6 Jul 2016 20:39:17 +0000 (16:39 -0400)]
aggregator: Delay clipping to output thread
This is required because the synchronized events like caps or segments
may only be processed on the output thread.
https://bugzilla.gnome.org/show_bug.cgi?id=781673
Olivier Crête [Thu, 7 Jul 2016 20:13:57 +0000 (16:13 -0400)]
aggregator: Make pad eos as soon as all buffers are processed, dont way for events
https://bugzilla.gnome.org/show_bug.cgi?id=781673
Olivier Crête [Thu, 7 Jul 2016 15:47:40 +0000 (11:47 -0400)]
aggregator: Only count buffers when declaring queue full
https://bugzilla.gnome.org/show_bug.cgi?id=781673
Olivier Crête [Wed, 6 Jul 2016 21:28:11 +0000 (17:28 -0400)]
tests: Test caps using query
Sending an event can accepted event if the caps were rejected
because the event could be queued and processed later.
Also send a drain query in the caps test to make sure that the
event has been processed.
https://bugzilla.gnome.org/show_bug.cgi?id=781673
Olivier Crête [Wed, 6 Jul 2016 20:41:44 +0000 (16:41 -0400)]
aggregator: Simplify clip function
The return value was ignored anyway
https://bugzilla.gnome.org/show_bug.cgi?id=781673
Olivier Crête [Sun, 15 May 2016 13:04:58 +0000 (16:04 +0300)]
aggregator: Only declare first buffer on actual buffer
The function needs to be unlocked if any data is received, but only
end the first buffer processing on an actual buffer, synchronized events
don't matter on the first buffer processing.
https://bugzilla.gnome.org/show_bug.cgi?id=781673
Olivier Crête [Wed, 10 May 2017 00:20:07 +0000 (20:20 -0400)]
aggregator: Set initial position on first buffer
Set the initial position on the first buffer, otherwise the queue
will grow without limits before the output thread is started.
https://bugzilla.gnome.org/show_bug.cgi?id=781673
Olivier Crête [Wed, 10 May 2017 00:06:29 +0000 (20:06 -0400)]
aggregator: Reset the pad's first buffer flag with the rest
There is not reason to have separate code to reset this one.
https://bugzilla.gnome.org/show_bug.cgi?id=781673
Olivier Crête [Wed, 10 May 2017 00:05:55 +0000 (20:05 -0400)]
aggregator: Reset pad on init
Factor out the pad reset code from the flushing and use it on init as well
https://bugzilla.gnome.org/show_bug.cgi?id=781673
Olivier Crête [Wed, 10 May 2017 00:13:58 +0000 (20:13 -0400)]
aggregator: Fix indentation
https://bugzilla.gnome.org/show_bug.cgi?id=781673
fvanzile [Thu, 23 Feb 2017 23:42:08 +0000 (15:42 -0800)]
glcontext: keep a ref to the active thread
With the macOS/iOS implementations, the active thread can change
multiple times over the life of a pipeline which would expose a race in
the thread tracking.
Fix by taking a ref on the active thread while the context is active.
https://bugzilla.gnome.org/show_bug.cgi?id=779202
Tim-Philipp Müller [Sat, 20 May 2017 14:04:45 +0000 (15:04 +0100)]
meson: add options to set package name and origin
https://bugzilla.gnome.org/show_bug.cgi?id=782172
Josep Torra [Sat, 20 May 2017 10:34:27 +0000 (12:34 +0200)]
examples: fix macOS 9.12 deprecation warnings
Add #defines to allow older versions of macOS to use the new constant names.
Josep Torra [Sat, 20 May 2017 10:19:08 +0000 (12:19 +0200)]
cocoa: fix a recently introduced typo
Fixes gstgldisplay_cocoa.m:175:26: error: use of undeclared identifier 'singletone'.
Josep Torra [Sat, 20 May 2017 10:16:50 +0000 (12:16 +0200)]
cocoa: fix macOS 10.12 deprecation warnings
Add #defines to allow older versions of macOS to use the new constant names.
Sebastian Dröge [Mon, 17 Apr 2017 11:43:49 +0000 (14:43 +0300)]
dashdemux: Use correct (sub) fragment start when clipping chunk size
If we didn't download anything yet, we shouldn't use fragment.start but
the start position of the current sidx entry.
Sebastian Dröge [Mon, 17 Apr 2017 11:25:10 +0000 (14:25 +0300)]
dashdemux: Make sure to pass the next buffer after a seek with DISCONT flag
When we manually seek for skipping ahead in keyunit-only trickmode, we
have to enforce that ourselves as adaptivedemux does not know about the
seeks.
Sebastian Dröge [Sat, 15 Apr 2017 08:55:34 +0000 (11:55 +0300)]
dashdemux: Ensure that we never ever download over the current SIDX entry
Even if downloading more than needed to try to get the moof and first
keyframe both together.
Sebastian Dröge [Thu, 13 Apr 2017 19:49:29 +0000 (22:49 +0300)]
dashdemux: Take SIDX entry timestamp/duration into account for keyunit-only mode
And not the whole segment's timestamp/duration
Sebastian Dröge [Thu, 13 Apr 2017 14:12:26 +0000 (17:12 +0300)]
dashdemux: Reset parsing state correctly in various places
Sebastian Dröge [Thu, 13 Apr 2017 11:51:07 +0000 (14:51 +0300)]
dashdemux: Reset previous target_time when seeking
Sebastian Dröge [Mon, 3 Apr 2017 13:47:58 +0000 (16:47 +0300)]
dashdemux: Smooth out skip distances in keyframe-only trick modes
This ensures smoother playback. It looks weird if we first do a big
jump, then play a couple of consecutive frames, just to again skip ahead
quite a bit because we ran late again.
Sebastian Dröge [Mon, 3 Apr 2017 12:51:43 +0000 (15:51 +0300)]
dashdemux: Take configure max bitrate/framerate into account for keyframe skipping
And by default only produce up to 10 fps.
Sebastian Dröge [Mon, 3 Apr 2017 12:48:13 +0000 (15:48 +0300)]
dashdemux: Try harder to not run into a loop over the same fragment over and over
While still making sure to not jump ahead one fragment further than
needed.
Sebastian Dröge [Wed, 15 Mar 2017 15:47:53 +0000 (17:47 +0200)]
dashdemux: Add some debug output to the target time selection
Sebastian Dröge [Tue, 28 Mar 2017 11:12:12 +0000 (14:12 +0300)]
dashdemux: Use the current clock running time in addition to the QoS earliest time
Sebastian Dröge [Mon, 27 Mar 2017 14:52:36 +0000 (17:52 +0300)]
dashdemux: Download keyframes from the current position if we're far enough ahead
Far enough here means more than 500ms or 4 times the average keyframe
download time. There is no need to jump ahead by one average keyframe
download time in this case.
This makes playback smooth if the network is fast enough.
Sebastian Dröge [Wed, 22 Mar 2017 10:05:20 +0000 (12:05 +0200)]
dashdemux: Fix sync sample selection based on target time in reverse playback mode
Sebastian Dröge [Wed, 22 Mar 2017 09:21:47 +0000 (11:21 +0200)]
dashdemux: Snap-seek for skipping ahead and use actual keyframe distance in the current fragment if applicable
Sebastian Dröge [Wed, 15 Mar 2017 14:47:59 +0000 (16:47 +0200)]
dashdemux: Don't adjust for fragment duration twice when seeking ahead in KEYUNITS mode
Sebastian Dröge [Wed, 15 Mar 2017 14:46:55 +0000 (16:46 +0200)]
dashdemux: Add current fragment duration in reverse playback mode to the position
We play from the end of the fragment to the beginning, not from the
beginning backwards.
Sebastian Dröge [Wed, 15 Mar 2017 14:45:33 +0000 (16:45 +0200)]
dashdemux: Never take more than the current fragment duration for estimations inside the fragment
Taking the average might give us from results.
Sebastian Dröge [Mon, 6 Mar 2017 12:27:20 +0000 (14:27 +0200)]
dashdemux: Don't increase current position if we just downloaded the moof in KEYUNIT mode
Edward Hervey [Thu, 12 Jan 2017 14:54:37 +0000 (15:54 +0100)]
dashdemux: Improve key-unit trick mode downloading
When dealing with key-unit trick mode downloads, the goal is to
provide the best "Quality of Experience". This is achieved by:
1) maximizing the number of frames displayed per second
2) avoiding "stalling" as much as possible (i.e. not downloading and
decoding frames fast enough)
This implementation achives this by:
1) Knowing very precisely the current keyframe being download (i.e
more accurate than at the fragment level which might contain more
than one keyfram). This is the new "actual_position" variable
introduced by this commit
2) Knowing the position of downstream (provided by QoS and stored
in the adaptivedemuxstream qos_earliest_time variable)
3) Knowing how long it takes to request and fully download a keyframe
(the average_download_time variable)
Taking those 3 variables into account, whenever a keyframe has been
pushed downstream we calculate a "target time" (target_time variable)
which is the ideal next keyframe time to request so that:
1) It will be requested/downloaded/demuxed/decoded in time to be
displayed without being too late
2) It will not be too far ahead that it would cause too few frames
per second to be displayed.
How far ahead we will request is inversily proportional to how close
the actual position (actual_position) is from the downstream
position (qos_earliest_time). The more is buffered between the source
and the sink, the "closer" the target time will be, and therefore
the more frames per seconds will be displayed (up to the limit
of keyframes_per_second * absolute_rate).
Edward Hervey [Wed, 11 Jan 2017 16:11:27 +0000 (17:11 +0100)]
dashdemux: Store average download time
This will be used to bound the download rate when working in
keyframe-only trick mode
Edward Hervey [Wed, 11 Jan 2017 16:08:36 +0000 (17:08 +0100)]
adaptivedemux: Add various comments to the code
Edward Hervey [Tue, 15 Nov 2016 07:13:27 +0000 (08:13 +0100)]
adaptivedemux: Store QoS time
Allows subclasses to know where downstream is and make decisions
based upon that
Sebastian Dröge [Thu, 1 Sep 2016 14:47:11 +0000 (17:47 +0300)]
dashdemux: Collect keyunit distance and size statistics
Edward Hervey [Mon, 7 Nov 2016 11:22:09 +0000 (12:22 +0100)]
dashdemux: Disable bitrate switching in key-unit trick mode
This creates too much havoc for now
Edward Hervey [Mon, 31 Oct 2016 09:08:35 +0000 (10:08 +0100)]
dashdemux: Use new adaptivedemux trickmode macro
Reduces the lines of code, and makes it a bit more readable
Edward Hervey [Mon, 31 Oct 2016 09:08:04 +0000 (10:08 +0100)]
adaptivedemux: Add a macro to know if we are in trickmode-keyunit
Reduces the amount of lines of code in subclasses
Jan Schmidt [Thu, 18 May 2017 13:57:22 +0000 (15:57 +0200)]
uvch264src: Apply timestamps to outgoing aux buffers
When extracting an aux buffer from an MJPG carrier, at
*least* put the original timestamp on it, even if we
fail to apply any other timestamp (which we always do
at the moment, because the timestamp calculating code
was never finished). Apply a DTS using the camera
supplied delay value as well, assuming that there's
no re-ordering going on (there isn't in the C920,
which is really the only extant camera doing this
stuff) and a warning if that turns out not to be true.
Simon Himmelbauer [Thu, 18 May 2017 12:23:14 +0000 (15:23 +0300)]
qt: Use GST_GL_HAVE_PLATFORM_CGL instead of GST_GL_HAVE_PLATFORM_COCOA
The latter is not used/available anymore since years. Also fix a typo
in the include path for the Cocoa GL display header.
Dmitry Zhadinets [Sun, 7 May 2017 19:50:00 +0000 (19:50 +0000)]
motioncells: delay motionmaskcoords until caps arrive
motionmaskcoords is not applied on start because the information
about resolution isn't available until caps arrive.
https://bugzilla.gnome.org/show_bug.cgi?id=768666
Sebastian Dröge [Thu, 18 May 2017 11:34:04 +0000 (14:34 +0300)]
amcvideodec: Unref downstream caps after usage
https://bugzilla.gnome.org/show_bug.cgi?id=782771
Tim-Philipp Müller [Thu, 18 May 2017 09:58:20 +0000 (10:58 +0100)]
Fix up package name and origin in some plugins
Tim-Philipp Müller [Thu, 18 May 2017 09:53:48 +0000 (10:53 +0100)]
stereo: fix typo in plugin description
Sebastian Dröge [Thu, 18 May 2017 08:42:17 +0000 (11:42 +0300)]
h265parse: Fix calculation of codec_data buffer size
Haihua Hu [Wed, 17 May 2017 08:26:38 +0000 (16:26 +0800)]
glformat: Add missing GST_GL_RGB565 in some switch statement
https://bugzilla.gnome.org/show_bug.cgi?id=782736
Sebastian Dröge [Wed, 17 May 2017 14:38:01 +0000 (17:38 +0300)]
cocoa: Install gstgldisplay_cocoa.h
It's needed by e.g. qmlglsink.
Sebastian Dröge [Mon, 15 May 2017 17:31:31 +0000 (20:31 +0300)]
gst: Clear floating flag in constructor of all GstObject subclasses that are not owned by any parent
https://bugzilla.gnome.org/show_bug.cgi?id=743062
Sebastian Dröge [Mon, 15 May 2017 11:23:44 +0000 (14:23 +0300)]
decklink: Sink the clock reference in the constructor
This is now needed as GstClock does not do that internally anymore,
because that broke bindings.
https://bugzilla.gnome.org/show_bug.cgi?id=743062
Haihua Hu [Wed, 17 May 2017 02:58:05 +0000 (10:58 +0800)]
qml: Add EGL platform support for x11 backend
Add support for EGL platform when x11 is available. This can work
e.g. on imx6 platform.
https://bugzilla.gnome.org/show_bug.cgi?id=782718
Nicolas Dufresne [Tue, 16 May 2017 18:05:52 +0000 (14:05 -0400)]
Remove plugin specific static build option
Static and dynamic plugins now have the same interface. The standard
--enable-static/--enable-shared toggle are sufficient.
Seungha Yang [Tue, 16 May 2017 14:20:44 +0000 (23:20 +0900)]
dashdemux: Initialize sidx_position to GST_CLOCK_TIME_NONE
If a manifest has non-zero presentation time offset
(i.e., earliest presentation time specified by sidx box is not zero),
the initial sidx position shouldn't be zero. Since we cannot define
exact sidx position until parsing sidx box, set the value to unknown.
https://bugzilla.gnome.org/show_bug.cgi?id=782693
Edward Hervey [Mon, 15 May 2017 16:10:11 +0000 (18:10 +0200)]
adaptivedemux: Handle stop_type:SET, stop:NONE
Seek values of type GST_SEEK_TYPE_SET with values of GST_CLOCK_TIME_NONE
are perfectly valid (we essentially don't modify the existing position)
Sebastian Dröge [Mon, 15 May 2017 13:37:14 +0000 (16:37 +0300)]
decklinkvideosink: Copy min(gstreamer_stride, decklink_stride) per line
Instead of just bpp * width, which might be more than we can copy.
Sebastian Dröge [Mon, 15 May 2017 09:00:50 +0000 (12:00 +0300)]
glshader: Make reference counting of attach() consistent
https://bugzilla.gnome.org/show_bug.cgi?id=747990
https://bugzilla.gnome.org/show_bug.cgi?id=702960
Vincent Penquerc'h [Fri, 12 May 2017 09:00:56 +0000 (10:00 +0100)]
h265parse: fix caps leak in renegotiation
Vincent Penquerc'h [Fri, 12 May 2017 09:01:10 +0000 (10:01 +0100)]
h264parse: fix caps leak in renegotiation
Vincent Penquerc'h [Tue, 9 May 2017 09:32:05 +0000 (10:32 +0100)]
jpeg2000parse: fix negotiation with j2c and jpc both allowed upstream
If upstream supports both, but downstream supports only jpc, j2c
would have been selected as the first in the caps.
https://bugzilla.gnome.org/show_bug.cgi?id=782221
Sebastian Dröge [Wed, 19 Apr 2017 12:47:57 +0000 (13:47 +0100)]
hlssink2: New HLS sink element based on splitmuxsink
This embeds the muxer inside the sink and accepts elementary streams
while the old HLS sink required the muxer outside. Apart from that the
interface is the same as before.
Currently only mpegtsmux is supported, but support for other muxers is
just a matter of adding a property.
The advantage of the new sink is that it reduces complexity a lot and
properly handles pre-encoded streams with appropriately spaced
keyframes.
https://bugzilla.gnome.org/show_bug.cgi?id=781496
Sebastian Dröge [Sat, 25 Feb 2017 10:37:46 +0000 (12:37 +0200)]
decklinkvideo/audiosrc: Add GstReferenceTimestampMeta with the stream time to each buffer
This is basically a frame counter provided by the driver and it's
advancing at the speed of the HDMI/SDI input. Having this available on
each buffer allows to know what constant-framerate-based timestamp each
frame is corresponding to and can be used e.g. to write out files
accordingly without having the local pipeline clock timestamps used.
https://bugzilla.gnome.org/show_bug.cgi?id=779213
Sebastian Dröge [Fri, 12 May 2017 09:35:39 +0000 (11:35 +0200)]
decklinkaudiosrc: Consistently pass stream_time and stream_duration to audio/video src
It's the same value now, pass it consistently.
Nirbheek Chauhan [Tue, 9 May 2017 10:36:10 +0000 (16:06 +0530)]
directsoundsrc: Use a GstClockID to wait instead of Sleep()
The main advantage is that our sleeps can be interrupted in case of
an src_reset(). Earlier, we would need to wait for a read to complete
before we could do a reset, which could take a long time.
https://bugzilla.gnome.org/show_bug.cgi?id=781249
Sebastian Dröge [Thu, 11 May 2017 16:39:46 +0000 (18:39 +0200)]
audiobuffersplit: Allow changing the output-buffer-duration at any time
Previously this was only allowed before the CAPS event arrived.
Nicolas Dufresne [Tue, 9 May 2017 19:13:10 +0000 (15:13 -0400)]
opencv/meson: Allow 3.2.0
This was already added and tested in autoconf.
Matthew Waters [Thu, 11 May 2017 10:05:24 +0000 (20:05 +1000)]
build/gl/meson: check for GLES3/gl3ext.h existence
791e7522ebcb75beb31fcace271dee1342d3505d for meson
https://bugzilla.gnome.org/show_bug.cgi?id=781885
Jens Georg [Thu, 11 May 2017 08:29:58 +0000 (10:29 +0200)]
build: Check for GLES3/gl3ext.h existence
Some OpenGL drivers do not ship this but use GLES2/gl2ext.h instead.
This is also in line with Khronos's recommendations
https://bugzilla.gnome.org/show_bug.cgi?id=781885
Scott D Phillips [Thu, 11 May 2017 05:29:10 +0000 (22:29 -0700)]
msdk: Declare libmfx.a as a C++ static library
This lets meson know that the overall plugin needs built with C++
linking.
https://bugzilla.gnome.org/show_bug.cgi?id=781561
Sebastian Dröge [Wed, 10 May 2017 13:59:42 +0000 (15:59 +0200)]
decklink: Always use the video stream time for audio too
The audio packet times can be completely unrelated to the video stream
time, depending on the card. While this looks like a bug in the driver,
just always using the video stream time (which is correct) works as a
workaround for now.
Nicolas Dufresne [Tue, 9 May 2017 19:06:52 +0000 (15:06 -0400)]
opencv/meson: Ensure variable opencv_found is set
If the required version is not satisfied, we need to make sure this
variable is set, otherwise build will fail.
Nicolas Dufresne [Tue, 9 May 2017 17:16:50 +0000 (13:16 -0400)]
Bump and update for meson 0.40.1
This patch bumps the required meson to 0.40.1 as gstreamer core just
did, and cleanup some code to use a feature from 0.37 that allow
specifying version range when checking dependency.
https://bugzilla.gnome.org/show_bug.cgi?id=780654
Vincent Penquerc'h [Fri, 5 May 2017 10:05:40 +0000 (11:05 +0100)]
opusparse: do not drop preskip and gain from OpusHead header
https://bugzilla.gnome.org/show_bug.cgi?id=753275
Nicolas Dufresne [Tue, 9 May 2017 13:47:10 +0000 (09:47 -0400)]
qmlgl: Make the plugin name match the pugin file name
Nicolas Dufresne [Tue, 9 May 2017 13:43:01 +0000 (09:43 -0400)]
gme: Make the plugin name match the plugin library name
Nicolas Dufresne [Tue, 9 May 2017 13:41:17 +0000 (09:41 -0400)]
neon: Make the plugin name match the plugin library
Sebastian Dröge [Tue, 9 May 2017 09:25:20 +0000 (11:25 +0200)]
glmemory: Only use glDrawBuffer if available
Otherwise fall back to glDrawBuffers. Also check if glReadBuffer exists
before using it.
glDrawBuffer does not exist for GLES, only glDrawBuffers does.
https://bugzilla.gnome.org/show_bug.cgi?id=782376
Edward Hervey [Mon, 8 May 2017 13:34:50 +0000 (15:34 +0200)]
adaptivedemux: Handle more live seeking use-cases
This commit fixes the following assumptions with live seeking:
1) start was always valid and of type GST_SEEK_TYPE_SET
2) direction was always forward
3) stop should be offsetted when handling non-accurate seeks before
the range start position.
In order to handle more live seeking use-cases (including reverse playback),
only do non-accurate start/stop value clamping for GST_SEEK_TYPE_SET values.
Also add a bit more debugging lines for issues
https://bugzilla.gnome.org/show_bug.cgi?id=782330
Edward Hervey [Fri, 5 May 2017 16:25:43 +0000 (18:25 +0200)]
adaptivedemux: Implement GST_SEEK_TYPE_END usage for live
When dealing with live streams, we can't rely on GstSegment calculation
since it uses the segment duration to calculate the absolute values.
But since we are dealing with live *and* we know the ranges, we can
compute the absolute seeking values using the range stop (i.e. "now")
as the END position.
Allows seeking back to "live" by using start_type:GST_SEEK_TYPE_END
and start:0
https://bugzilla.gnome.org/show_bug.cgi?id=782228
Scott D Phillips [Mon, 8 May 2017 18:01:39 +0000 (11:01 -0700)]
meson: Ignore msvc warnings in C++ as well
We were only ignoring the listed msvc warnings for C language
files and not C++. This was working by the coincidence that we did
not have any instances of these warnings in C++ files. Lately the
build of decklink has been fixed on windows, and it has an
instance of one of these warnings in a C++ file.
https://bugzilla.gnome.org/show_bug.cgi?id=782345
Nirbheek Chauhan [Wed, 3 May 2017 17:20:27 +0000 (22:50 +0530)]
directsoundsrc: Use latency-time and buffer-time settings
Earlier, the plugin was ignoring those settings and blindly setting
buffer-time to 2 seconds and latency-time to 200ms, which forced all
pipelines to have a minimum latency of 200ms + sink latency.
The values of segsize and segtotal were also not derived correctly.
Now we obey these values, and you can get close to the previous
behaviour by setting buffer-time and latency-time manually. Note that
they are set in microseconds.
As a consequence, when we haven't received enough data from the
device, we now sleep for a time proportional to the data remaining.
However, Directsound is a deprecated API so it maintains its own
software ringbuffer which updates at arbitrary intervals. Hence we
might have to wait a full segsize to get the last 10% of data. To
avoid tight loops, we clamp our sleep floor at 10ms.
In my testing, this keeps the wakeups not-too-high (proportional to
the latency-time set on the source). Further improvements should be
made by fixing the WASAPI audio source plugin instead of this.
Directsound is deprecated and as the comments explain, it is
impossible to get low latency, decent quality, or good performance
from it.
Based on a patch by Sebastian Dröge <sebastian@centricular.com>
https://bugzilla.gnome.org/show_bug.cgi?id=781249
Tim-Philipp Müller [Sun, 7 May 2017 10:47:40 +0000 (11:47 +0100)]
glvideomixer: fix whole example launch line actually
Tim-Philipp Müller [Sun, 7 May 2017 10:41:06 +0000 (11:41 +0100)]
glvideomixer: remove extraneous \ from example launch line in docs
Edward Hervey [Fri, 5 May 2017 16:22:18 +0000 (18:22 +0200)]
adaptivedemux: Allow live seeking range to go back to "now"
The allowed live seek ranges returned by subclasses are "inclusive", that is
to say that the "range_stop" value they return is the highest acceptable position
one can seek to (i.e. "now").
Allow seeking to exactly that value
Nirbheek Chauhan [Fri, 5 May 2017 13:22:24 +0000 (18:52 +0530)]
Revert "decklink: Fix debug logging warnings on Windows"
This reverts commit
845832263ba6b9e135b23e9a29c9d109cb2ee9c4.
The commit broke cross-mingw CI:
https://ci.gstreamer.net/job/GStreamer-master/8659/console
It seems that cross-mingw on Autotools and native-mingw on Meson
disagree about the size of HRESULT. Revert for now till I can
investigate the Meson side of things some more.
Nirbheek Chauhan [Thu, 4 May 2017 23:00:59 +0000 (04:30 +0530)]
decklink: Fix linking on MinGW
MinGW does not provide comsupp.lib, so there's no implementation of
_com_util::ConvertBSTRToString. Use a fallback implementation that
uses wcstombs() instead.
On MinGW we also truncate the name to 100 chars which should be fine.
Nirbheek Chauhan [Thu, 4 May 2017 22:32:29 +0000 (04:02 +0530)]
decklink: Fix building on Windows
BSTR is already a pointer to a string
Nirbheek Chauhan [Thu, 4 May 2017 22:29:45 +0000 (03:59 +0530)]
decklink: Fix debug logging warnings on Windows
HRESULT is unsigned long int, not unsigned int