platform/upstream/gstreamer.git
3 years agoMerging gst-plugins-ugly
Thibault Saunier [Fri, 24 Sep 2021 19:14:42 +0000 (16:14 -0300)]
Merging gst-plugins-ugly

3 years agoMove files from gst-plugins-bad into the "subprojects/gst-plugins-bad/" subdir
Thibault Saunier [Fri, 24 Sep 2021 19:14:36 +0000 (16:14 -0300)]
Move files from gst-plugins-bad into  the "subprojects/gst-plugins-bad/" subdir

3 years agoMerging gst-plugins-bad
Thibault Saunier [Fri, 24 Sep 2021 19:14:03 +0000 (16:14 -0300)]
Merging gst-plugins-bad

3 years agoMove files from gst-plugins-good into the "subprojects/gst-plugins-good/" subdir
Thibault Saunier [Fri, 24 Sep 2021 19:13:50 +0000 (16:13 -0300)]
Move files from gst-plugins-good into  the "subprojects/gst-plugins-good/" subdir

3 years agoMerging gst-plugins-good
Thibault Saunier [Fri, 24 Sep 2021 19:13:37 +0000 (16:13 -0300)]
Merging gst-plugins-good

3 years agoMove files from gst-plugins-base into the "subprojects/gst-plugins-base/" subdir
Thibault Saunier [Fri, 24 Sep 2021 19:13:26 +0000 (16:13 -0300)]
Move files from gst-plugins-base into  the "subprojects/gst-plugins-base/" subdir

3 years agoMerging gst-plugins-base
Thibault Saunier [Fri, 24 Sep 2021 19:13:17 +0000 (16:13 -0300)]
Merging gst-plugins-base

3 years agoMove files from gstreamer into the "subprojects/gstreamer/" subdir
Thibault Saunier [Fri, 24 Sep 2021 19:13:07 +0000 (16:13 -0300)]
Move files from gstreamer into  the "subprojects/gstreamer/" subdir

3 years agoRelease 1.19.2
Tim-Philipp Müller [Thu, 23 Sep 2021 00:34:47 +0000 (01:34 +0100)]
Release 1.19.2

3 years agoRelease 1.19.2
Tim-Philipp Müller [Thu, 23 Sep 2021 00:34:00 +0000 (01:34 +0100)]
Release 1.19.2

3 years agoRelease 1.19.2
Tim-Philipp Müller [Thu, 23 Sep 2021 00:33:39 +0000 (01:33 +0100)]
Release 1.19.2

3 years agoRelease 1.19.2
Tim-Philipp Müller [Thu, 23 Sep 2021 00:33:08 +0000 (01:33 +0100)]
Release 1.19.2

3 years agoRelease 1.19.2
Tim-Philipp Müller [Thu, 23 Sep 2021 00:32:32 +0000 (01:32 +0100)]
Release 1.19.2

3 years agoaudiobuffersplit: Remove unneeded buffer_clip wrapper
Nicolas Dufresne [Wed, 22 Sep 2021 18:17:35 +0000 (14:17 -0400)]
audiobuffersplit: Remove unneeded buffer_clip wrapper

This is just a small cleanup noticed while reading.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2544>

3 years agoopencv: cvtracker: add draw property
Vivek R [Fri, 10 Jul 2020 14:01:13 +0000 (19:31 +0530)]
opencv: cvtracker: add draw property

This property controls the drawing of rectangle around the tracked object.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2454>

3 years agoopencv: add cvtracker plugin
Vivek R [Sun, 24 May 2020 18:07:25 +0000 (23:37 +0530)]
opencv: add cvtracker plugin

This adds an object tracker plugin.
Tracker implementations from https://docs.opencv.org/3.4/d0/d0a/classcv_1_1Tracker.html

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2454>

3 years agoopencv: patch to ensure headers are detected
Vivek R [Mon, 25 May 2020 05:05:30 +0000 (10:35 +0530)]
opencv: patch to ensure headers are detected

This patch is used to ensure opencv headers are detected.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2454>

3 years agortph263pdepay: flag keyframes on output buffers
Tim-Philipp Müller [Wed, 22 Sep 2021 13:03:57 +0000 (14:03 +0100)]
rtph263pdepay: flag keyframes on output buffers

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1091>

3 years agoclocksync: Add some debug output to the clock waiting code
Sebastian Dröge [Wed, 23 Jun 2021 13:41:20 +0000 (16:41 +0300)]
clocksync: Add some debug output to the clock waiting code

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/841>

3 years agompegtsmux, mpegpsmux: remove GPL from choice of licenses and add SPDX license identifiers
Tim-Philipp Müller [Sun, 19 Sep 2021 00:18:00 +0000 (01:18 +0100)]
mpegtsmux, mpegpsmux: remove GPL from choice of licenses and add SPDX license identifiers

Some people need to avoid inclusion of GPL code for their use cases and thus
get easily spooked by GPL license headers. This code is actually licensed
under different licenses, only one of which is GPL, and it's already possible
to just upgrade from LGPL to GPL anyway so having the GPL listed explicitly
as one of the choices doesn't really add anything. So remove GPL from the list
and also add SPDX license identifiers while we're at it.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2539>

3 years agopbutils: codec-utils: fix g-ir-scanner warning
Tim-Philipp Müller [Tue, 21 Sep 2021 21:39:46 +0000 (22:39 +0100)]
pbutils: codec-utils: fix g-ir-scanner warning

Warning: GstPbutils: gst_codec_utils_h264_get_profile_flags_level:
unknown parameter 'codec_data' in documentation comment, should be 'codecs_data

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1279>

3 years agowasapideviceprovider: Add support for dynamic device add/remove
Seungha Yang [Mon, 23 Aug 2021 18:54:27 +0000 (03:54 +0900)]
wasapideviceprovider: Add support for dynamic device add/remove

Adding IMMDeviceEnumerator::RegisterEndpointNotificationCallback
in order to support device monitoring.
On OnDeviceAdded(), OnDeviceRemoved(), and OnDefaultDeviceChanged()
callback, wasapi device provider implementation will enumerate
devices again and will notify newly added and removed device
via GstDeviceProvider API.

As a bonus point, this IMMDeviceEnumerator abstraction object
will spawn a dedicated internal COM thread, so various COM thread
related issues of WASAPI plugin can be resolved by this commit.

Fixes: https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/issues/1649
Fixes: https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/issues/1110
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2484>

3 years agox264enc: Strip trailing whitespace from x264's log messages
Jan Alexander Steffens (heftig) [Tue, 21 Sep 2021 13:05:58 +0000 (15:05 +0200)]
x264enc: Strip trailing whitespace from x264's log messages

x264 has linebreaks at the end of its log messages that we should
ignore.

Add G_GNUC_PRINTF to the callback to make sure GCC lets us forward the
format string to another function marked as printf (g_strdup_vprintf)
without triggering -Wformat-nonliteral.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-ugly/-/merge_requests/85>

3 years agox264enc: Add the func/line info to the log callback
Jan Alexander Steffens (heftig) [Tue, 21 Sep 2021 12:57:29 +0000 (14:57 +0200)]
x264enc: Add the func/line info to the log callback

It looks awkward when it's missing. At least point us at the callback
function.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-ugly/-/merge_requests/85>

3 years agox264enc: Reduce log spam about bit depth support
Jan Alexander Steffens (heftig) [Tue, 21 Sep 2021 12:56:23 +0000 (14:56 +0200)]
x264enc: Reduce log spam about bit depth support

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-ugly/-/merge_requests/85>

3 years agoalsasink: Allow stop() function to happen during failing writes
Nicolas Dufresne [Tue, 24 Aug 2021 19:27:32 +0000 (15:27 -0400)]
alsasink: Allow stop() function to happen during failing writes

In ALSA, there is possible temporary failures that may require a retry,
though in certain situation, this may leak to the write() function
holding on a lock forever preventing the pipeline from going to pause
or stop. Fix this by shortly dropping the lock between retries.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1261>

3 years agoalsasink: Improve logging in write() function
Nicolas Dufresne [Tue, 24 Aug 2021 19:26:12 +0000 (15:26 -0400)]
alsasink: Improve logging in write() function

This moves the "written X frames" lower so that we don't trace
confusing negative values on errors and add the error code in the
"Write error" log.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1261>

3 years agoopenjpeg: add unit test
Stéphane Cerveau [Mon, 1 Feb 2021 15:21:59 +0000 (16:21 +0100)]
openjpeg: add unit test

Test various format supported with subframes.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/979>

3 years agojpeg2000parse + openjpeg: Switch striped mode to its own caps
Stéphane Cerveau [Fri, 22 Jan 2021 09:39:56 +0000 (10:39 +0100)]
jpeg2000parse + openjpeg: Switch striped mode to its own caps

It's not compatible with any other element that use the non-striped
mode. In addition, in this mode, we require that every frame have the
same number of stripes or that the MARKER bit be present, which is
different from the other
formats too.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/979>

3 years agoopenjpegdec: Fix crash with AYUV64 in subframe mode
Stéphane Cerveau [Wed, 23 Dec 2020 10:03:34 +0000 (11:03 +0100)]
openjpegdec: Fix crash with AYUV64 in subframe mode

Remove useless generic fill_frame methods to use
the packed one for AYUV and AYUV64.

Fix gst-launch-1.0 -v videotestsrc  !
video/x-raw,width=640,height=480,format=AYUV64  !  openjpegenc
num-stripes=8   !  openjpegdec max-threads=8  ! videoconvert !
autovideosink sync=false

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/979>

3 years agoopenjpegdec: support for a multithreaded decoding.
Stéphane Cerveau [Fri, 24 Apr 2020 14:15:42 +0000 (16:15 +0200)]
openjpegdec: support for a multithreaded decoding.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/979>

3 years agoopenjpegdec: enable sub frame mode
Aaron Boxer [Mon, 13 Jan 2020 19:02:39 +0000 (14:02 -0500)]
openjpegdec: enable sub frame mode

Rebuild output frame from multiple stripes input.

Keep the first frame and fill it with the following stripes to finish
a complete frame only once.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/979>

3 years agompegtsmux: Require frame alignment for JPEG 2000
Olivier Crête [Tue, 22 Dec 2020 23:19:40 +0000 (18:19 -0500)]
mpegtsmux: Require frame alignment for JPEG 2000

We have yet to implement stripe alignment with the required descriptor.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/979>

3 years agoopenjpegdec: Reject stripes for now
Olivier Crête [Tue, 22 Dec 2020 23:15:52 +0000 (18:15 -0500)]
openjpegdec: Reject stripes for now

They're not implemented.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/979>

3 years agoopenjpegenc: Only allow stripe with image/x-jpc format
Olivier Crête [Tue, 22 Dec 2020 23:20:35 +0000 (18:20 -0500)]
openjpegenc: Only allow stripe with image/x-jpc format

It's the only format that our MPEG-TS muxer allows and the carriage of
JPEG 2000 stripes is only defined for MPEG-TS as far as I know.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/979>

3 years agoopenjpegenc: support for a multithreaded encoding.
Stéphane Cerveau [Tue, 21 Apr 2020 18:56:03 +0000 (20:56 +0200)]
openjpegenc: support for a multithreaded encoding.

This commit introduces a multithreaded encoder allowing
to encode mulitple stripes or subframes in separated threads.
This feature aims to enhance the overall latency of a codec
pipeline.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/979>

3 years agowpe: context thread dispatch fixes
Philippe Normand [Sat, 18 Sep 2021 11:02:15 +0000 (12:02 +0100)]
wpe: context thread dispatch fixes

Use dedicated mutex/cond/flag for jobs being dispatched in the context thread.
The previous code was signalling the thread startup condition, which is wrong.
When WPEContextThread::dispatch() is invoked it means the thread has already
correctly been started up.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2533>

3 years agowpe: Properly wait on context thread startup condition
Philippe Normand [Sat, 18 Sep 2021 11:01:39 +0000 (12:01 +0100)]
wpe: Properly wait on context thread startup condition

Fixes #1661

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2533>

3 years agodoc: Update kmssink caps cache
Nicolas Dufresne [Mon, 20 Sep 2021 13:41:32 +0000 (09:41 -0400)]
doc: Update kmssink caps cache

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2530>

3 years agokmssink: Add RGB16/BGR16 support
Nicolas Dufresne [Fri, 17 Sep 2021 20:21:39 +0000 (16:21 -0400)]
kmssink: Add RGB16/BGR16 support

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2530>

3 years agokmssink: Sort format according to GST_VIDEO_FORMATS_ALL
Nicolas Dufresne [Fri, 17 Sep 2021 20:14:36 +0000 (16:14 -0400)]
kmssink: Sort format according to GST_VIDEO_FORMATS_ALL

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2530>

3 years agokmssink: Remove big endian format inversion
Nicolas Dufresne [Fri, 17 Sep 2021 19:42:25 +0000 (15:42 -0400)]
kmssink: Remove big endian format inversion

This has been a bad interpretation of the DRM docuemntation. The formats are
fixed regardless the CPU, but for some formats, they expressed in the opposite
order as GStreamer. Same change was done in waylandsink 2 years ago.

https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/commit/b393b650ab9bfb9654fc116163ab331907216d74

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2530>

3 years agokmssink: Add NV61 support
Nicolas Dufresne [Fri, 17 Sep 2021 19:41:41 +0000 (15:41 -0400)]
kmssink: Add NV61 support

This identically handled to NV16, so no reason not to inclue it.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2530>

3 years agokmssink: Add NV24 support
Nicolas Dufresne [Fri, 17 Sep 2021 19:39:54 +0000 (15:39 -0400)]
kmssink: Add NV24 support

This was tested on RK3566 platform, using vendor DRM driver.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2530>

3 years agocodecs: av1decoder: Use GstFlowReturn everywhere
Seungha Yang [Fri, 17 Sep 2021 15:33:12 +0000 (00:33 +0900)]
codecs: av1decoder: Use GstFlowReturn everywhere

The same modification as that of VP8 decoder

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2528>

3 years agocodecs: vp9decoder: Use GstFlowReturn everywhere
Seungha Yang [Fri, 17 Sep 2021 15:09:24 +0000 (00:09 +0900)]
codecs: vp9decoder: Use GstFlowReturn everywhere

The same modification as that of VP8 decoder

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2528>

3 years agocodecs: vp8decoder: Use GstFlowReturn everywhere
Seungha Yang [Fri, 17 Sep 2021 14:23:06 +0000 (23:23 +0900)]
codecs: vp8decoder: Use GstFlowReturn everywhere

boolean return value is not sufficient for representing the reason
of error in most cases. For instance, any errors around new_sequence()
would mean negotiation error, not just *ERROR*.
And some subclasses will allocate buffer/memory/surface on new_picture()
but it could be failed because of expected error, likely flushing

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2528>

3 years agogst: Initialize optional event/message fields when parsing
Sebastian Dröge [Mon, 20 Sep 2021 10:12:12 +0000 (13:12 +0300)]
gst: Initialize optional event/message fields when parsing

These might not exist inside the structure and then we would potentially
keep around uninitialized memory from the caller in the out parameter.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/887>

3 years agovideodecoder: Add properties to automatically request sync points and vfunc to allow...
Sebastian Dröge [Fri, 10 Sep 2021 12:10:46 +0000 (15:10 +0300)]
videodecoder: Add properties to automatically request sync points and vfunc to allow subclasses to handle packet loss / missing data

Subclasses could use the new vfunc to activate packet loss concealment,
for example.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1274>

3 years agompegtspacketizer: memcmp potentially seen_before data
Vivia Nikolaidou [Mon, 16 Aug 2021 15:15:42 +0000 (18:15 +0300)]
mpegtspacketizer: memcmp potentially seen_before data

Theoretically the version number is incremented every time there's a new
section, but in a world of streaming we can't easily make that
assumption.

An example of a broken use case is when we're cat-ing two mpeg-ts files
together, which is equivalent of capturing a DVB stream while switching
channels. A set-top box would know that we switched the channels and
reset the demuxer, but in practice this might not happen.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2468>

3 years agoplayer: Fix/add various annotations
Sebastian Dröge [Mon, 20 Sep 2021 08:35:51 +0000 (11:35 +0300)]
player: Fix/add various annotations

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2538>

3 years agomeson: va: Make AV1 support always optional
Seungha Yang [Sat, 18 Sep 2021 09:07:43 +0000 (18:07 +0900)]
meson: va: Make AV1 support always optional

Otherwise meson configure with -Dva=enabled will be failed
when installed libva version is < 1.8

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2532>

3 years agova: caps: Don't use image formats for decoded frames.
Víctor Manuel Jáquez Leal [Sat, 18 Sep 2021 09:03:16 +0000 (11:03 +0200)]
va: caps: Don't use image formats for decoded frames.

Initially we tried to use the internal color conversion used in i965
and Gallium drivers when decoding. But this approach has showed
limitations and problems.

This patch removes completely the possible color conversion at
decoding, since it show problems with deinterlacing, for example:

gst-launch-1.0 filesrc location=interlaced.mpg2 ! parsebin ! vampeg2dec ! vadeinterlace ! xvimagesink

Allowing only the surface formats when decoding is more stable.

For color conversion is better to do it explicitly with vapostproc.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2531>

3 years agotest: bitwriter: Add a test for reset_and_get_data when not byte unaligned.
He Junyan [Sun, 19 Sep 2021 13:01:21 +0000 (21:01 +0800)]
test: bitwriter: Add a test for reset_and_get_data when not byte unaligned.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/886>

3 years agobitwriter: Fix a memory leak in reset_and_get_buffer.
He Junyan [Sun, 19 Sep 2021 14:39:09 +0000 (22:39 +0800)]
bitwriter: Fix a memory leak in reset_and_get_buffer.

We should record the ownership of the data before we reset the bitwriter.
Or we will always dup the buffer data and leak the memory.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/886>

3 years agobitwriter: Fix the trailing bits lost when getting its data.
He Junyan [Sat, 18 Sep 2021 16:19:43 +0000 (00:19 +0800)]
bitwriter: Fix the trailing bits lost when getting its data.

In reset_and_get_data and reset_and_get_buffer, it fails to include
the trailing bits less than 8. So, when the bit_size is not byte
aligned, the trailing bits are lost in the return buffer.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/886>

3 years agoplayer: Add missing nullable annotations
Marijn Suijten [Tue, 27 Apr 2021 09:59:15 +0000 (11:59 +0200)]
player: Add missing nullable annotations

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2197>

3 years agoplay: Add missing nullable annotations
Marijn Suijten [Tue, 27 Apr 2021 09:58:58 +0000 (11:58 +0200)]
play: Add missing nullable annotations

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2197>

3 years agosys: shm: Define shm_enable and shm_deps before escape meson subdir
Fabian Orccon [Sat, 18 Sep 2021 12:29:25 +0000 (14:29 +0200)]
sys: shm: Define shm_enable and shm_deps before escape meson subdir

Fixes meson configure in tests if the shm plugin is disabled

Fixes #1664

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2534>

3 years agod3d11videosink: Display title of content if possible
Seungha Yang [Fri, 17 Sep 2021 17:27:51 +0000 (02:27 +0900)]
d3d11videosink: Display title of content if possible

Update title text of window (currently it's always "Direct3D11 renderer")
when we are rendering on internal HWND, not external HWND.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2529>

3 years agod3d11videosink: Remove unused enum value
Seungha Yang [Fri, 17 Sep 2021 16:32:11 +0000 (01:32 +0900)]
d3d11videosink: Remove unused enum value

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2529>

3 years agomsdkenc: Pass color properties to MediaSDK for encoding
Mengkejiergeli Ba [Wed, 15 Sep 2021 05:59:17 +0000 (13:59 +0800)]
msdkenc: Pass color properties to MediaSDK for encoding

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2523>

3 years agomsdkh265enc: Add profile main10 still picture for hevc
Mengkejiergeli Ba [Wed, 15 Sep 2021 08:32:02 +0000 (16:32 +0800)]
msdkh265enc: Add profile main10 still picture for hevc

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2524>

3 years agowaylandsink: Fix double render check
Nicolas Dufresne [Thu, 16 Sep 2021 21:12:58 +0000 (17:12 -0400)]
waylandsink: Fix double render check

Our code does not support rendering twice the same wl_buffer in a row, so it
tries to skip that case, but for this it relied on the GstBuffer pointer,
while the cache actually works at the GstMemory level now. To avoid this
compare the GstWlBuffer instead.

This fixes crash when use in zero-copy with videorate element.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2526>

3 years agocodecs: mpeg2decoder: Use tsg framerate for latency.
Víctor Manuel Jáquez Leal [Tue, 7 Sep 2021 07:45:54 +0000 (09:45 +0200)]
codecs: mpeg2decoder: Use tsg framerate for latency.

Latency setting relies on src pad caps, but they aren't set when the
function is called, and latency is never updated.

In order to fix it, this patch uses TSG framerate first, and if it's
not set yet, sinkpad caps are used to get the framerate.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2514>

3 years agod3d11decoder: Refactor for more unified decoding flow
Seungha Yang [Wed, 15 Sep 2021 15:59:37 +0000 (00:59 +0900)]
d3d11decoder: Refactor for more unified decoding flow

... and various code cleanup.

* Move spreaded decoding API calls into one method
Previously, decoding flow of most codecs are
- Call DecoderBeginFrame() on start_picture()
- Call {Get,Release}DecoderBuffer() on decode_slice()
- Call SubmitDecoderBuffers() and DecoderEndFrame() on end_picture()
Such spreaded API calls make it hard to keep track of status
of decoding. Now it will be done at once in a new method.

* Drop a code for non-zero wBadSliceChopping
When bitstream buffer provided by driver is not sufficient
to write compressed bitstream data, host decoder needs to make use
of wBadSliceChopping so that driver can understand there are
multiple bitstream buffer. But it's a bit unrealistic and
not tested. Since FFMpeg's DXVA implemetaion doesn't support it,
we might be able to ignore the case for now.

* Make code more portable
Consider common logic of GstCodecs -> DXVA translation for all D3D APIs
(i,e., D3D9, D3D11, and D3D12).

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2525>

3 years agod3d11decoder: Remove duplicated class_init and property related code
Seungha Yang [Wed, 15 Sep 2021 14:41:39 +0000 (23:41 +0900)]
d3d11decoder: Remove duplicated class_init and property related code

Move them into the decoder helper code to remove duplication

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2525>

3 years agod3d11: Get rid of "extern "C"" wrapping for GST_DEBUG_CATEGORY_EXTERN
Seungha Yang [Fri, 10 Sep 2021 15:43:26 +0000 (00:43 +0900)]
d3d11: Get rid of "extern "C"" wrapping for GST_DEBUG_CATEGORY_EXTERN

Instead, change the file defining debug category to cpp

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2525>

3 years agovideodecoder: Fix min-force-key-unit-interval logic and logging
Havard Graff [Fri, 10 Sep 2021 14:12:51 +0000 (16:12 +0200)]
videodecoder: Fix min-force-key-unit-interval logic and logging

The new keyframe is needed when the deadline of the buffer has exeeded
the waiting time, not while it is within it.

Also, since we look at the deadline of the frame, log that instead of PTS.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1278>

3 years agotests: skip cc tests if plugin is disabled
U. Artie Eoff [Tue, 30 Jun 2020 18:15:43 +0000 (11:15 -0700)]
tests: skip cc tests if plugin is disabled

Skip the closedcaption element tests if the
closedcaption option is disabled at compile
time (i.e. -Dclosedcaption=disabled).

v2: rename pangocairo_dep to avoid conflict
with later definition in ext/ttml/meson.build
as suggested by @tpm.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1388>

3 years agortphdrhext-twcc: Return failure on map failure
Olivier Crête [Wed, 18 Aug 2021 23:47:40 +0000 (19:47 -0400)]
rtphdrhext-twcc: Return failure on map failure

This feels like exactly like a case that should fail.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1059>

3 years agortphdrext: Update write() API to return a signed value
Olivier Crête [Wed, 18 Aug 2021 23:46:25 +0000 (19:46 -0400)]
rtphdrext: Update write() API to return a signed value

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1059>

3 years agortphdrext: Make write function return a signed value
Olivier Crête [Wed, 18 Aug 2021 23:40:55 +0000 (19:40 -0400)]
rtphdrext: Make write function return a signed value

Since the return value is documented to possibly be smaller than 0,
then it needs to be signed.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1258>

3 years agovideorate: Add unit test for closing a segment and opening a separate one
Olivier Crête [Wed, 16 Jun 2021 19:07:13 +0000 (15:07 -0400)]
videorate: Add unit test for closing a segment and opening a separate one

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/767>

3 years agovideorate: Drop incoming buffers that are outside of the segment
Olivier Crête [Wed, 16 Jun 2021 19:06:57 +0000 (15:06 -0400)]
videorate: Drop incoming buffers that are outside of the segment

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/767>

3 years agovideorate: Only "close" the segment if it is discontinous
Olivier Crête [Fri, 24 Jul 2020 21:41:57 +0000 (17:41 -0400)]
videorate: Only "close" the segment if it is discontinous

Otherwise, it will drop valid buffers on a simple segment update

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/767>

3 years agovideorate: Add test for segment update
Olivier Crête [Fri, 24 Jul 2020 21:38:58 +0000 (17:38 -0400)]
videorate: Add test for segment update

Continue as-is on segment update.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/767>

3 years agovideorate: Update the base time on segment updates
Olivier Crête [Fri, 24 Jul 2020 20:35:04 +0000 (16:35 -0400)]
videorate: Update the base time on segment updates

Dropping it to 0 makes videorate push buffers from timestamp 0 again.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/767>

3 years agocodecs: h264dec: Check bumping again after inserting current picture.
He Junyan [Tue, 31 Aug 2021 09:16:05 +0000 (17:16 +0800)]
codecs: h264dec: Check bumping again after inserting current picture.

In order to get the lowest latency, we can add another bumping check after
inserting the current picture into the DPB immediately. That can avoid
waiting for another decoding circle of the next frame and so the latency
is lower.

Fix: #1628
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2501>

3 years agocodecs: h264: Add protection to to_insert picture in bump check.
He Junyan [Tue, 31 Aug 2021 09:37:11 +0000 (17:37 +0800)]
codecs: h264: Add protection to to_insert picture in bump check.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2501>

3 years agocodecs: h264dec: Improve the policy to infer max_num_reorder_frames.
He Junyan [Tue, 31 Aug 2021 08:39:06 +0000 (16:39 +0800)]
codecs: h264dec: Improve the policy to infer max_num_reorder_frames.

The max_num_reorder_frames number can change the way we bumping the
pictures in the DPB. The smaller it is, the lower latency we will
get. So it is important for live mode streams, but it is not given
in VUI parameters sometimes. We now improve the policy to infer it:
1. Never guess it in the "strict" compliance.
2. For baseline and constrained baseline profiles, which do not have
   B frames, set it to 0.
3. For -intra only profiles, set it to 0.
4. Otherwise, not guess it.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2501>

3 years agotests: skip aes test if elements not built
U. Artie Eoff [Wed, 15 Sep 2021 03:57:30 +0000 (20:57 -0700)]
tests: skip aes test if elements not built

In ext/aes/meson.build, the aes_dep will return
not-found if -Daes=disabled, regardless of whether
openssl is found or not.  Thus, we don't need a
separate check for the option.  This will also
ensure that aes_dep is always defined and we can
use it in the tests/check/meson.build unit.

Fixes #1660

v2: handle -Daes=disabled, too.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2522>

3 years agoqtdemux: Try to build AAC codec-data whenever it's possible
Seungha Yang [Tue, 14 Sep 2021 08:26:27 +0000 (17:26 +0900)]
qtdemux: Try to build AAC codec-data whenever it's possible

AAC codec_data is a just collection of AAC profile, samplerate, and
channels. We can know samplerate and channels from parsed
SampleEntry data. Although the AAC profile is unknown there,
let's assume it as AAC-LC like we've been doing for the version 1
atom.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1082>

3 years agowpe: Add support for web:// URIs
Philippe Normand [Tue, 31 Aug 2021 16:33:02 +0000 (17:33 +0100)]
wpe: Add support for web:// URIs

The CEF source already supports this. No good reason for wpesrc not too ;)

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2503>

3 years agomultiqueue: fix obsolete comment re initial flow status
Mathieu Duponchelle [Thu, 9 Sep 2021 23:43:18 +0000 (01:43 +0200)]
multiqueue: fix obsolete comment re initial flow status

The initial single queue srcresult is OK, it hasn't been
NOT_LINKED since 2007.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/885>

3 years agomultiqueue: never consider a queue that is not waiting
Mathieu Duponchelle [Thu, 9 Sep 2021 18:25:25 +0000 (20:25 +0200)]
multiqueue: never consider a queue that is not waiting

.. when computing the high id.

After a flush for instance, sq->srcresult is reset to OK,
yet it doesn't make sense to pick a non-existing position
id as the high id when a queue doesn't contain any items
in that situation either.

It is in any case completely OK to let the not-linked stream
get consumed without throttling at this stage, as any
first packet arriving on other single queues will get assigned
a higher position id.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/885>

3 years agova: h264dec: Try to use ConstrainedBaseline or Main to decode BaseLine.
He Junyan [Fri, 23 Jul 2021 15:38:22 +0000 (23:38 +0800)]
va: h264dec: Try to use ConstrainedBaseline or Main to decode BaseLine.

In the h264, the Baseline profile is widely misused. A lot of streams declare
that they are the Baseline, but in fact they just conform to ConstrainedBaseline.
The features such as FMO and ASO are not used at all.
If the decoder does not strictly conforms to the SPEC, we can just use Baseline
or Main profile to decode it to avoid lots of streams failure.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2428>

3 years agoflv: fix seqnum handling for seeks
Vivienne Watermeier [Tue, 7 Sep 2021 20:23:01 +0000 (22:23 +0200)]
flv: fix seqnum handling for seeks

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1078>

3 years agoisomp4: also allow muxing different h264/5 profiles/levels/etc
Matthew Waters [Mon, 18 Jan 2021 05:06:27 +0000 (16:06 +1100)]
isomp4: also allow muxing different h264/5 profiles/levels/etc

All of that is advertised through the codec_data itself so can change
just fine within isomp4.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1071>

3 years agomatroska: Add support for muxing/demuxing ffv1
Sebastian Dröge [Sat, 11 Sep 2021 06:24:35 +0000 (09:24 +0300)]
matroska: Add support for muxing/demuxing ffv1

Previously only demuxing when stored via the RIFF/AVI mapping was
supported.

See https://github.com/FFmpeg/FFV1/blob/master/ffv1.md#matroska-file-format

Fixes https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/issues/923

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1080>

3 years agocodecs: h264dec: Improve the fast bump for the live mode.
He Junyan [Wed, 28 Jul 2021 15:19:15 +0000 (23:19 +0800)]
codecs: h264dec: Improve the fast bump for the live mode.

We control the policy of fast bump by the profile and the compliance
property. For baseline and constrained baseline profiles, we can use
more radical bump policy. User can also change the bump policy by
setting the compliance property in run time.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2432>

3 years agocodecs: h264: Change the low_latency to an enum for dpb_needs_bump().
He Junyan [Wed, 28 Jul 2021 14:48:21 +0000 (22:48 +0800)]
codecs: h264: Change the low_latency to an enum for dpb_needs_bump().

The bool parameter of low_latency is not enough. We have multi policies for
low latency bumping, from the safest to something radical. So we need an enum
to represent the proper latency requirement.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2432>

3 years agocodecs: h264dec: Add a compliance property to control behavior.
He Junyan [Mon, 26 Jul 2021 08:09:19 +0000 (16:09 +0800)]
codecs: h264dec: Add a compliance property to control behavior.

Some features such as the low-latency DPB bumping and mapping the
baseline profile as the constrained-baseline profile do not conform
to the H264 offical spec. But in practice, they are very useful and
are widely needed. We add this compliance property to control the
behavior of the decoder, make it fit more requirement.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2432>

3 years agodocs: Update cache
Philippe Normand [Sun, 12 Sep 2021 11:23:36 +0000 (12:23 +0100)]
docs: Update cache

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2521>

3 years agodocs: Update cache
Philippe Normand [Sun, 12 Sep 2021 11:18:32 +0000 (12:18 +0100)]
docs: Update cache

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1081>

3 years agodiscoverer: Prevent stream tags from leaking in global tags
Philippe Normand [Sun, 12 Sep 2021 09:07:49 +0000 (10:07 +0100)]
discoverer: Prevent stream tags from leaking in global tags

The PrivateStream should keep track of stream tags only. Likewise, the
GstDiscovererInfo should keep track of global tags only.

This patch fixes the issue where the discoverer would report duplicated tag
titles, especially for Matroska media files. The Matroska demuxer emits
correctly-scoped tags, but downstream was making no distinction of them.

Fixes #598, #836, https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/issues/827

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1275>

3 years agova: Update vapostproc documentation.
Víctor Manuel Jáquez Leal [Tue, 7 Sep 2021 08:55:10 +0000 (10:55 +0200)]
va: Update vapostproc documentation.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2495>

3 years agova: Update todo lists, removing deinterlacing.
Víctor Manuel Jáquez Leal [Tue, 7 Sep 2021 08:16:05 +0000 (10:16 +0200)]
va: Update todo lists, removing deinterlacing.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2495>

3 years agoAdd vadeinterlace element.
Víctor Manuel Jáquez Leal [Mon, 23 Aug 2021 09:24:40 +0000 (11:24 +0200)]
Add vadeinterlace element.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2495>

3 years agova: filter: Add past and future frames in GstVaSample.
Víctor Manuel Jáquez Leal [Tue, 24 Aug 2021 11:53:12 +0000 (13:53 +0200)]
va: filter: Add past and future frames in GstVaSample.

And add them in the pipeline structure if they are provided.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2495>