Sebastian Dröge [Thu, 1 Jul 2021 09:41:11 +0000 (12:41 +0300)]
pbutils: Expose functions for getting a file extension for caps and flags for describing the format of the caps
This information was available internally already but not available from
the outside.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1221>
Sebastian Dröge [Wed, 4 Aug 2021 07:06:02 +0000 (10:06 +0300)]
playbin/uridecodebin: Emit source-setup signal early before doing the scheduling query
Some elements will require the source to be set up properly before the
scheduling query returns useful results, e.g. appsrc and giostreamsrc.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1241>
Devarsh Thakkar [Thu, 10 Jun 2021 07:55:23 +0000 (00:55 -0700)]
ext: alsa: Fix fallback paths for setting buffer and period times
Below fallback paths were introduced in
https://github.com/GStreamer/gst-plugins-base/commit/
9759810d8206b5f1aa199f98599caec3630a1813
if setting period time after buffer time failed :
1) Set period time and then buffer time if it doesn't work
2) Set only buffer time
3) Set only period time
These all were not functioning properly since they were using old
copy of snd_pcm_hw_params_t which already had some fields set
as per previous try and this was causing issues as driver was
referring to that old value while trying to set them again in
fallback paths.
So now we always use the initial copy of snd_pcm_hw_params_t
for every fallback and same is also being done at
https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/commit/
557c4295107dc7374c850b0bd5331dd35e8fdd0f
Also we change the sequence to set period time earlier than
buffer time since period bytes being the smaller unit, most of the times
if underlying alsa device has a dependency then it is of period bytes
to be a multiple of some value (as per underlying DMA constraint)
and rest of the parameters like buffer bytes need to be adjusted
as per period bytes.
The same sequence is also followed in alsa-utils at
https://github.com/alsa-project/alsa-utils/commit/
9b621eeac4d55c4e881f093be5b163ca07d01b63
Fix 2) and 3) scenarios by returning success if the exclusive setting is passed
and not doing any further setting for buffer time or period time.
Add new fallback path of not setting any buffer time and period time
if all above fallback paths fail. The same is also being
followed at aforementioned pulseaudio commit.
In case of alsasink, remove the retry goto label, since it is not
required anymore as fallback paths take care of setting default
values if driver is not accepting any of the fallback paths.
Use separate label for exit to free params structs and return err
code. This also fixes leak in no_rate goto path in alsasink
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1212>
Jakub Adam [Tue, 25 May 2021 19:16:48 +0000 (21:16 +0200)]
videoencoder: pass upstream HDR information through codec state
Don't copy HDR metadata from sink pad, because its caps may not have
been set yet if GstVideoEncoder::negotiate is called from
GstVideoEncoder::set_format, as e.g. vpx encoder does.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1175>
Jakub Adam [Tue, 25 May 2021 19:15:53 +0000 (21:15 +0200)]
videoutils: add HDR metadata fields to GstVideoCodecState
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1175>
Sebastian Dröge [Mon, 16 Aug 2021 07:19:07 +0000 (10:19 +0300)]
video-overlay-composition: Allow empty overlay compositions
Allowing to pass NULL to the constructor removes the need to
special-case the first rectangle in calling code and generally
simplifies application code.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1256>
Per Förlin [Tue, 1 Jun 2021 13:27:31 +0000 (15:27 +0200)]
gstrtspconnection: Add support to ignore x-server header reply
When connecting to an RTSP server in tunnled mode (HTTP) the server
usually replies with a x-server header. This contains the address
of the intended streaming server. However some servers return an
"invalid" address. Here follows two examples when it might happen.
1. A server use Apache combined with a separate RTSP process to handle
Https request on port 443. In this case Apache handle TLS and
connects to the local RTSP server, which results in a local
address 127.0.0.1 or ::1 in the x-server reply. This address is
returned to the actual RTSP client in the x-server header.
The client will receive this address and try to connect to it
and fail.
2. The client use a ipv6 link local address with a specified scope id
fe80::aaaa:bbbb:cccc:dddd%eth0 and connects via Http on port 80.
The RTSP server receives the connection and returns the address
in the x-server header. The client will receive this address and
try to connect to it "as is" without the scope id and fail.
In the case of streaming data from RTSP servers like 1. and 2. it's
useful to have the option to simply ignore the x-server header reply
and continue using the original address.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1192>
Nirbheek Chauhan [Fri, 13 Aug 2021 14:05:23 +0000 (19:35 +0530)]
sdp: Avoid using g_memdup() since it is deprecated
g_memdup() is deprecated since GLib 2.68 and we want to avoid
deprecation warnings with recent versions of GLib. Instead of using
g_memdup2(), we can simply use the new gst_buffer_new_memdup() added
in 1.19.x
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1254>
Matthew Waters [Fri, 4 Jun 2021 08:32:07 +0000 (18:32 +1000)]
glbuffer: support persistent buffer mappings
Requires OpenGL 4.4 or EXT_buffer_storage
Current mesa exposes GL_ARB_buffer_storage when retrieving the relevant
functions returns no-ops and causes failures.
Improves throughput of uploads by roughly 30%-60% and download throughput by
roughly 10-30% across depending on the exact scenario and hardware.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1191>
Seungha Yang [Fri, 30 Jul 2021 14:57:20 +0000 (23:57 +0900)]
examples: win32-videooverlay: Add support for testing gst_video_overlay_set_render_rectangle
Add keyboard handler to test gst_video_overlay_set_render_rectangle()
API for Windows video elements
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1235>
Seungha Yang [Fri, 30 Jul 2021 14:04:57 +0000 (23:04 +0900)]
examples: win32-videooverlay: Use d3d11videosink by default
d3d11videosink was promoted to have primary rank and
it's recommended videosink element on Windows
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1235>
Seungha Yang [Thu, 8 Jul 2021 08:47:28 +0000 (17:47 +0900)]
tests: appsink: Add reverse stepping test case
To demonstrate reverse stepping issue of
https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/848
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1223>
Seungha Yang [Thu, 3 Jun 2021 10:15:22 +0000 (19:15 +0900)]
examples: win32-videooverlay: Add test option for threading scenario
Add an option to test the case where window thread and pipeline handling
thread are different. Mainly to test the HWND leak fixed by
https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2302
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1188>
Matthew Waters [Tue, 27 Jul 2021 07:44:02 +0000 (17:44 +1000)]
rtpbasedepayload: remove object locking an extension
Doing that is fraught with danger of deadlocks and is not conceptually
part of the API contract. The object lock is generally intended for
internal-object-use only.
If another lock is needed, that should be added separately.
This lock was erronously added as part of:
https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1118
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1233>
Matthew Waters [Tue, 27 Jul 2021 03:30:56 +0000 (13:30 +1000)]
gldownload: use the GstGLSyncMeta in all cases
fixes qmlglsrc ! gldownload ! videoconvert in some cases.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1232>
Nicolas Dufresne [Thu, 22 Jul 2021 21:11:26 +0000 (17:11 -0400)]
glcontext: egl: Stop comparing native surface pointer
This was noticed with wayland, sometimes the newly create native
handle can have the same pointer (even though its new). This lead
to unwanted errors or crash.
Fixes #927
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1231>
Nicolas Dufresne [Thu, 22 Jul 2021 21:02:51 +0000 (17:02 -0400)]
glwindow: Add "window-handle-changed" signal
This allow other objects to clear any wrapper object that depends
on the previous handle, and properly re-create the new wrappers without
having to resort into doing pointer comparison.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1231>
Nicolas Dufresne [Fri, 23 Jul 2021 17:39:34 +0000 (13:39 -0400)]
Revert "glwindow: wayland: Skip redoing surfaces if window haven't changed"
This reverts commit
aba6bd7822f4c0f572765bfaada76f454a594317.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1231>
Guillaume Desmottes [Mon, 22 Feb 2021 12:17:18 +0000 (13:17 +0100)]
appsrc: serialize custom events with buffers flow
Application may want to inject events to the pipeline and keep them
synchronized with the buffers flow.
Fix #247
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1046>
Guillaume Desmottes [Fri, 19 Feb 2021 12:32:48 +0000 (13:32 +0100)]
appsink: add API to catch events
There is currently no way for users to receive incoming events from
appsink while keeping them properly serialized with the buffers flow.
This can be especially useful when application is injecting custom
downstream events into the pipeline and needs to know when they reached
appsink.
Solving this by adding a new signal notifying about new incoming events
and a set of action signals and method to pull those events.
The API is actually pulling the samples and events all together as they
are actually fetched from the same queue.
Having a specific API to pull only events would have the side effect of
discarding samples (and pulling samples would discard events) making
this API not convenient for users.
Partially fix #247
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1046>
Guillaume Desmottes [Fri, 19 Feb 2021 13:45:08 +0000 (14:45 +0100)]
appsink: factor out dequeue_object()
No semantic change, will be used to implement new event API.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1046>
Nicolas Dufresne [Tue, 20 Jul 2021 13:37:58 +0000 (09:37 -0400)]
glwindow: wayland: Skip redoing surfaces if window haven't changed
The problem is that EGLNativeWindowSurface and wl_egl_surface are the
same object underneath, so we must recreate both together. As an
optimization, the EGLNativeWindowSurface wrapper is only re-created
if the window_handle changed.
On Mesa, this would cause crash, which will be fixed by:
https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11979
And will lead to proper errors in the future or on other GL stack. This
issue was encounter using a permanent GstGLDisplay after cycling one of
multiple independent pipelines through NULL state.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1230>
Nicolas Dufresne [Tue, 20 Jul 2021 13:36:22 +0000 (09:36 -0400)]
glwindow: wayland: Remove redundant create_surfaces call
The surfaces will be created in _roundtrip_async, so no need to call
this early. This should cause no functional difference.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1230>
Michael Olbrich [Fri, 11 Jun 2021 07:02:29 +0000 (09:02 +0200)]
decodebin3: improve decoder selection
Currently the decoder selection is very naive: The type with the highest
rank that matches the current caps is used. This works well for software
decoders. The exact supported caps are always known and the static caps are
defined accordingly.
With hardware decoders, e.g. vaapi, the situation is different. The decoder
may reject the caps later during a caps query. At that point, a new decoder
is created. However, the same type is chosen an after several tries,
decodebin fails.
To avoid this, do the caps query while adding the decoder and try again
with other decoder types if the query fails:
1. create the decoder from the next matching type
2. add and link the decoder
3. change the decoder state to READY
4. do the caps query
if it fails then remove the decoder again and go back to 1.
5. expose the source pad
6. sync the decoder state with the parent.
This way, the decoder is already part of the pipeline when the state change
to READY happens. So context handling should work as before.
Exposing the source pad after the query was successful is important:
Otherwise the thread from the decoder source pad may block in a blocked pad
downstream in the playsink waiting for other pads to be ready.
The thread now blocks trying to set the state back to NULL while holding
the SELECTION_LOCK. Other streams may block on the SELECTION_LOCK and the
playsink never unblocks the pad. The result is a deadlock.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1201>
Nicolas Dufresne [Wed, 14 Jul 2021 20:09:41 +0000 (16:09 -0400)]
gl: x11: Issue XSync to close our top level window
This is similar action as when the window handle is modified, we now issue
XSync whenever we destroy our internal window. This ensure that the window is
properly closed before the connecgtion is dropped.
Fixes #815
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1226>
Nicolas Dufresne [Wed, 14 Jul 2021 15:43:10 +0000 (11:43 -0400)]
gl: wayland: Fix hinding the window on close()
When the window is called, we properly destroy all surfaces, which effectively
will unmap that surface and should make it disapear on screen, but we also
destroy the wl_source, a GSource that is resposibble of dispatching and executing
messages to/from the Wayland server.
As a side effect, the server never gets the message and the surfaces are
"leaked" on the server. We fix this using wl_display_flush() before destroying
the wl_source.
Fixes #815
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1226>
Nicolas Dufresne [Thu, 15 Jul 2021 15:09:35 +0000 (11:09 -0400)]
tests: example: Add missing glx_dep when building sdlshare
Might be realted to some recent Mesa cleanup, but GLX is not longer visible
through libOpenGL, so add the missing deps now.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1227>
Matthew Waters [Mon, 17 May 2021 02:04:50 +0000 (12:04 +1000)]
examples/qt/textureshare: add explicit dep on glx_dep
Fixes linking:
/usr/bin/ld: subprojects/gst-plugins-base/tests/examples/gl/qt/qglwtextureshare/qglwtextureshare.p/qglrenderer.cpp.o: undefined reference to symbol 'glXGetCurrentContext'
/usr/bin/ld: /usr/lib64/libGLX.so.0: error adding symbols: DSO missing from command line
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1143>
Haelwenn (lanodan) Monnier [Fri, 14 May 2021 12:10:55 +0000 (14:10 +0200)]
gl: Try GLVND 'opengl' and 'glx' first
This fixes targetting desktop OpenGL without libGL.so
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1143>
Sebastian Dröge [Mon, 12 Jul 2021 06:37:24 +0000 (09:37 +0300)]
audioaggregator: Only post QoS messages if the property is enabled
Previously one of the branches did not check for the property value. To
avoid this in the future, check inside the QoS calculation function
instead.
As a side effect this now always prints the debug messages into the logs
when samples are dropped, which is useful information even without the
QoS messages.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1224>
Sebastian Dröge [Fri, 9 Jul 2021 06:49:15 +0000 (09:49 +0300)]
audioaggregator: Resync on the next buffer when dropping a buffer on discont resyncing
If a buffer is dropped during resyncing on a discont because either its
end offset is already before the current output offset of the
aggregator or because it fully overlaps with the part of the current
output buffer that was already filled, then don't just assume that the
next buffer is going to start at exactly the expected offset. It might
still require some more dropping of samples.
This caused the input to be mixed with an offset to its actual position
in the output stream, causing additional latency and wrong
synchronization between the different input streams.
Instead consider each buffer after a discont as a discont until the
aggregator actually resynced and starts mixing samples from the input
again.
Also update the start output offset of a new input buffer if samples
have to be dropped at the beginning. Otherwise it might be mixed too
early into the output and overwrite part of the output buffer that
already took samples from this input into account.
Fixes https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/issues/912
which is a regression introduced by https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1180/
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1224>
Olivier Crête [Wed, 26 May 2021 22:20:02 +0000 (18:20 -0400)]
audiomixer: Add test for QoS message posting
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1209>
Olivier Crête [Wed, 26 May 2021 14:38:18 +0000 (10:38 -0400)]
audio aggregator: Post QoS message when dropping audio
Post a QoS message every time some audio samples are dropped.
Also print log messages to make it easier to debug
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1209>
Olivier Crête [Tue, 25 May 2021 22:05:05 +0000 (18:05 -0400)]
audio aggregator: Count samples that are dropped or processed
Keep a count of samples that are dropped or processed as statistics
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1209>
Olivier Crête [Fri, 21 May 2021 20:16:50 +0000 (16:16 -0400)]
audio aggregator: Add QoS property to pad
Add a property to emit a QoS message whenever any data is dropped.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1209>
Olivier Crête [Fri, 21 May 2021 20:10:06 +0000 (16:10 -0400)]
audio aggregator: Rename property enum to match class name
Add "CONVERT" into the property enum as we're going to add an
enum specifically for the base pad.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1209>
Stéphane Cerveau [Tue, 24 Sep 2019 15:14:10 +0000 (17:14 +0200)]
videodecoder: add API to receive subframes
A video decoder can now receive subframes and start decoding
instead of waiting for the full frame to be complete.
Subframe support will reduce latency as described in the
video encoder base class.
A unit test illustrating this API is available in
tests/check/libs/videodecoder.c.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/454>
Seungha Yang [Sat, 3 Jul 2021 10:36:06 +0000 (19:36 +0900)]
gl/context/wgl: Add missing NULL init
The value of uninitialized local variable is varying depending
on compiler and not guaranteed to be NULL initialized.
That results in pointing random address instead of expected function pointer.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1222>
Francisco Javier Velázquez-García [Fri, 12 Mar 2021 12:55:38 +0000 (13:55 +0100)]
videotestsrc: Add SMPTE75 RP-219 color bars conformant
Implement 8-bit values of SMPTE RP 2019-1:2014. The bar widths and
heights are the result of fractions as integers. The remainders of
widths are distributed in a way that they match the values in Table
C.1 (a) in the specification.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1063>
Jan Alexander Steffens (heftig) [Fri, 12 Mar 2021 19:58:40 +0000 (20:58 +0100)]
videotestsrc: Add a start parameter to _blend_line
Makes it easier to paint part of a line.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1063>
Jan Alexander Steffens (heftig) [Fri, 12 Mar 2021 19:57:13 +0000 (20:57 +0100)]
videotestsrc: Keep tmpline unchanged in_convert_tmpline
This will allow us to repeatedly
call it to render subsequent lines.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1063>
Michael de Gans [Tue, 15 Jun 2021 20:22:55 +0000 (13:22 -0700)]
appsink: fix incorrect return nullability
This commit fixes the annoations for return nullability on several
GstAppSink functions. This was causing bindings to be generated
incorrectly.
Fixes #914
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1203>
Seungha Yang [Wed, 19 May 2021 07:22:46 +0000 (16:22 +0900)]
compositor: Add scaling policy to support PAR-aware scaling
Adding "sizing-policy" property for user to be able to specify
scaling policy (aspect-ratio for example).
At the moment, supported mode is only keep-aspect-ratio, but we might
be able to add more policies such as cropping, etc.
Fixes: https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/issues/696
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1156>
Seungha Yang [Wed, 19 May 2021 11:11:15 +0000 (20:11 +0900)]
video: Deprecate gst_video_sink_center_rect()
... and add gst_video_center_rect() method as a replacement.
The method is useful for outside of videosink subclasses as well
but the old naming might be able to mislead people.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1156>
Jakub Adam [Mon, 24 May 2021 17:11:51 +0000 (19:11 +0200)]
rtpbasepayload: don't write empty extension header
When some header extensions are present but none decides to write any
data to the currently processed RTP buffer, remove the extension data
section.
Resulting RTP buffer wasn't formatted correctly.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1173>
Jakub Adam [Mon, 24 May 2021 17:02:42 +0000 (19:02 +0200)]
rtpbuffer: Add gst_rtp_buffer_remove_extension_data()
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1173>
Jakub Adam [Mon, 24 May 2021 17:01:24 +0000 (19:01 +0200)]
rtpbasepayload: map RTP buffer READWRITE when setting headers
GstRTPHeaderExtension::write can map the RTP buffer for reading. If that
happens on a buffer that is already mapped WRITE-only by the payloader,
the payloader's mapping gets invalidated (GstRTPBuffer::map will point
to a different instance of GstMemory).
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1173>
Olivier Crête [Thu, 24 Jun 2021 18:56:11 +0000 (14:56 -0400)]
rtphdrext: Make all fields private
The presence of a method and a field with the same name confuses the C#
binding generator. As there are accessor functions for all the fields,
let's just make them private.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1218>
Olivier Crête [Sat, 26 Jun 2021 16:50:58 +0000 (12:50 -0400)]
gst: don't use volatile to mean atomic
volatile is not sufficient to provide atomic guarantees and real atomics
should be used instead. GCC 11 has started warning about using volatile
with atomic operations.
https://gitlab.gnome.org/GNOME/glib/-/merge_requests/1719
Discovered in gst-plugins-good#868
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1219>
Jan Schmidt [Fri, 25 Jun 2021 13:42:34 +0000 (23:42 +1000)]
video-converter: Set up matrix tables only once.
When configuring a multi-thread converter, only allocate the
shared colour conversion matrices once for the first thread,
to avoid allocating multiple times and leaking memory.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1216>
Jan Alexander Steffens (heftig) [Thu, 6 May 2021 17:01:41 +0000 (19:01 +0200)]
video-converter: Set up gamma tables only once
When the video converter is using multiple threads, the gamma tables
were created multiple times, leaking the tables set up for the previous
thread.
Only calculate the tables once.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1140>
Jan Alexander Steffens (heftig) [Thu, 6 May 2021 16:22:45 +0000 (18:22 +0200)]
audio-converter: Free config when gst_audio_converter_new fails
The config got leaked when parameter validation fails.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1140>
Seungha Yang [Fri, 25 Jun 2021 06:24:21 +0000 (15:24 +0900)]
glprototypes: Add GST_GL_API_OPENGL to available version of sync
Make sync APIs usable if supported, even when GST_GL_API_OPENGL3 is
not selected
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1215>
Per Förlin [Fri, 23 Apr 2021 16:03:20 +0000 (18:03 +0200)]
gstrtspconnection: Add IPv6 support for tunneled mode
An IPv6 address must be specified within [] brackets.
Add brackets for IPv6 address used for tunneled mode,
for non-tunneled this is already supported.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1145>
Nicolas Dufresne [Wed, 16 Jun 2021 18:49:14 +0000 (14:49 -0400)]
videodecoder: Call drain() rather then finish() on segment-done
The finish() virtual function documentation state that "Sub-classes can refuse
to decode new data after." Though, it is very common to issue a non-flushing
seek after that event in gapless playback uses case. This fixes potential
stalls with code using segment seeks, by using drain() virtual funciton
instead.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1206>
Matthew Waters [Tue, 8 Jun 2021 04:55:36 +0000 (14:55 +1000)]
oggdemux: fix a race in push mode when performing the duration seek
There may be two or more threads involved here however the important
interaction is the use of ogg->seeK_event_drop_till value that was only
set in the push-mode seek-event thread and could race with upstream
sending e.g. and EOS (or data).
Scenario is this:
1. oggdemux performs a seek to near the end of the file to try and find
the duration. ogg->push_state is set to PUSH_DURATION.
2. Seek is picked up by the dedicated seek event thread and sets
ogg->seek_event_drop_till to the seek event's seqnum.
3. Most operations are blocked or dropped waiting on the duration to
be determined and processing continues until a duration is found.
4. Two branching options for how this ultimately plays out
4a. The source is too fast and we receive an EOS event which is dropped
because ogg->push_state == PUSH_DURATION. In this case everything
works.
4b. We hit our 'almost at the end' check in
gst_ogg_pad_handle_push_mode_state() and attempt to seek back to the
beginning (or to a user-provided seek). This seek is marshalled to
the seek event thread without setting ogg->seek_event_drop_till but
with change ogg->push_state = PUSH_PLAYING. If an EOS event or
e.g. buffers arrive from upstream before the seek event thread has
picked up the seek event, then the EOS/data is processed as if it
came as a result of the seek event. This is the case that fails.
The fix is two-fold:
1. Preemptively set ogg->seek_event_drop_till when setting the seek
event so that data and other events can be dropped correctly.
2. In addition to dropping and EOS events while ogg->push_state ==
PUSH_DURATION, also drop any EOS events that are received before the
seek event has been processed by also tracking the seqnum of the seek.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1196>
Sergei Kovalev [Mon, 21 Jun 2021 14:06:14 +0000 (14:06 +0000)]
audiobasesink: Fix of double lock release
Add missing "return;" which prevents double lock release.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1208>
Corentin Damman [Mon, 21 Jun 2021 08:34:07 +0000 (08:34 +0000)]
Update COPYING
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1207>
Matthew Waters [Wed, 16 Jun 2021 05:55:49 +0000 (15:55 +1000)]
gl/context: fill a GError on platform-specific fill_info() error
Fixes bindings assuming that GError is always set on error:
https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/809#note_957493
https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/809#note_957494
https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/809#note_957498
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1204>
François Laignel [Wed, 16 Jun 2021 10:13:21 +0000 (12:13 +0200)]
Check mandatory ClockTime arguments
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1205>
Scott Moreau [Fri, 11 Jun 2021 15:27:46 +0000 (09:27 -0600)]
gl/wayland: Use consistent wl_display when creating work queue for proxy wrapper
Without this, glimagesink since wayland
727c7903 fails with
gst-launch-1.0: ../src/wayland-client.c:2181: wl_proxy_set_queue:
Assertion 'proxy->display == queue->display' failed.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1200>
Haihua Hu [Mon, 7 Jun 2021 09:54:46 +0000 (17:54 +0800)]
gl/viv-fb: fix gl plugin hang when run with viv-fb backend
below commit change the window resize thread and cause viv-fb backend
hang, need move resize code after window->open is called. Otherwise,
the resize message will send to a thread that not start running and
window resize call will waiting forever.
Commit:
b887db1efe816c0c28b60a6842fa9005a26c1502
glwindow: fix racy resize updates
Take locks around resize handling and marshall all resizes to the
windowing thread by default.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1195>
Nicolas Dufresne [Thu, 10 Jun 2021 16:41:31 +0000 (12:41 -0400)]
eglimage: Add AV12 DMABuf import support
The per plane format mapping for AV12 was missing, which would force
raw upload.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1199>
Nicolas Dufresne [Thu, 10 Jun 2021 16:39:34 +0000 (12:39 -0400)]
eglimage: Fix wrong stride when importing DMABuf
When the code was ported to use component index instead of plane
index, the call to GST_VIDEO_INFO_PLANE_STRIDE() was accidently ported
to use component index, but this macro takes a plane index.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1199>
Sebastian Dröge [Tue, 8 Jun 2021 14:27:55 +0000 (17:27 +0300)]
appsrc: When dropping buffers before handling the initial segment use the latest input segment
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1198>
Sebastian Dröge [Tue, 8 Jun 2021 10:05:46 +0000 (13:05 +0300)]
appsrc: Correctly check if this is the first buffer that was queued
By checking the queue length this would also count events and caps. We
already keep track of the number of buffers separately so just use that.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1197>
Mathieu Duponchelle [Sat, 5 Jun 2021 00:35:30 +0000 (02:35 +0200)]
appsrc: signal enough-data even when leaking
this is convenient for application that wish to monitor whether
the appsrc is leaking.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1193>
Marijn Suijten [Thu, 3 Jun 2021 19:14:42 +0000 (21:14 +0200)]
rtp/header: Add missing `array length` annotation to read/write methods
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1190>
Sebastian Dröge [Thu, 3 Jun 2021 10:40:33 +0000 (13:40 +0300)]
video: Sort video formats correctly
AV12 should be right after A420 because it is the same format with just
one plane less, instead of being next to I420/NV12 which don't have an
alpha channel.
RGBP should be before GBR because it's the same format except for the
more canonical component order.
See https://gitlab.freedesktop.org/gstreamer/gstreamer-rs/-/merge_requests/790
which actually checks on the CI if the algorithm defined in
video-format.h is implemented correctly.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1189>
Matthew Waters [Thu, 3 Jun 2021 05:57:39 +0000 (15:57 +1000)]
glvideomixer: hold extra pad ref while calling parent
Our subsequent cleanup needs a ref on the pad and calling the parent may
release the last reference and could cause a use-after-free.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1187>
Matthew Waters [Thu, 3 Jun 2021 05:57:17 +0000 (15:57 +1000)]
gl/stereo: fix a coupld of caps leaks
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1187>
Matthew Waters [Sat, 22 May 2021 05:42:17 +0000 (15:42 +1000)]
gl/display: remove choosing egl fallback from GST_GL_PLATFORM
If GST_GL_WINDOW is unset but GST_GL_PLATFORM=egl, then we were choosing
to create an GstGLDisplayEGL directly instead of going through the any
more specific windowing system implementation (X11, Wayland).
The 'create an GstGLDisplayEGL when GST_GL_PLATFORM=egl' was a fallback
as we did not have entries for all EGL-using window systems previously.
Now that we do, the fallback can be removed. An EGLDisplay can still
be created by setting GST_GL_WINDOW=egl or as one option.
Fixup of https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1154
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1169>
Tim-Philipp Müller [Sun, 23 May 2021 22:51:27 +0000 (23:51 +0100)]
Use g_memdup2() where available and add fallback for older GLib versions
g_memdup() is deprecated since GLib 2.68 and we want to avoid
deprecation warnings with recent versions of GLib.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1171>
Tim-Philipp Müller [Tue, 1 Jun 2021 14:28:24 +0000 (15:28 +0100)]
Back to development
Tim-Philipp Müller [Mon, 31 May 2021 23:09:54 +0000 (00:09 +0100)]
Release 1.19.1
Nicolas Dufresne [Wed, 26 May 2021 20:57:28 +0000 (16:57 -0400)]
glcontext: Ask for an alpha channel and fallback
While this was already possible through the GLContext machinary, this simply
request an alpha channel by default and fallback if this is not possible. This
obsolete some RPi Dispmanx hack, since this is near equivalent will allow see
through when playgin WebM Alpha or other transparent files.
Application are still free to pass their own EGLContext attribute, this is
specially for the case the application let GStreamer chose (e.g. gst-launch).
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1176>
Daniel Knobe [Mon, 31 May 2021 07:14:53 +0000 (09:14 +0200)]
rawbaseparse: check destination format correctly
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1181>
Olivier Crête [Thu, 27 May 2021 02:50:22 +0000 (22:50 -0400)]
audiomixer: Add test for discont going backwards
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1180>
Olivier Crête [Thu, 27 May 2021 20:05:04 +0000 (16:05 -0400)]
audioaggregator: Don't overwrite already written samples
On re-sync, don't forget what has already been written. Instead, just
drop any samples that overlap with parts that were already filled.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1180>
Seungha Yang [Thu, 27 May 2021 07:20:09 +0000 (16:20 +0900)]
audiobasesrc: Fix divide by zero assertion
GstAudioRingBufferSpec can be cleared from other thread, then
rate value will be zero
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1179>
Marijn Suijten [Wed, 26 May 2021 21:22:45 +0000 (23:22 +0200)]
audio,video-format: Make generate_raw_formats idempotent for assertions
When compiling without assertions `g_assert` and its contents disappear
resulting in no list being deserialized at all and the
`gst_{audio,video}_formats_raw` functions to return an empty collection.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1177>
Tim-Philipp Müller [Sun, 23 May 2021 22:48:39 +0000 (23:48 +0100)]
Use new gst_buffer_new_memdup()
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1170>
Sebastian Dröge [Mon, 24 May 2021 11:25:55 +0000 (14:25 +0300)]
compositor: Consider the converter-config when deciding whether one pad obscures another
If the converter configuration is set to not fill any borders, or if the
border fill color is not full opaque, then the pad has to be handled
as potentially transparent and can't be considered to obscure another
one.
This prevents pads from being wrongly skipped and doing alpha-blending
with uninitialized memory.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1172>
Nicolas Dufresne [Fri, 21 May 2021 19:04:16 +0000 (15:04 -0400)]
doc: Update cache for AV12 pixel format
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1152>
Nicolas Dufresne [Fri, 21 May 2021 17:28:32 +0000 (13:28 -0400)]
video: Update video-orc-dist
This adds the new symbols needed for AV12 support.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1152>
Daniel Almeida [Tue, 18 May 2021 19:20:36 +0000 (16:20 -0300)]
gl: add support for AV12
AV12 is an internally conceived format that is actually the
combination of NV12 and an alpha plane.
This format is to add to gstreamer's webM transparency support for
vp8 and vp9. To this end, two I420 streams are independently decoded
simultaneously for the actual content and the alpha plane respectively
and these are then combined into A420.
This patch adds GL conversion support so that it is possible to convert
from AV12 to RGBA for the purposes of rendering it on a display.
The reverse conversion is also supplied.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1152>
Daniel Almeida [Fri, 21 May 2021 15:24:37 +0000 (12:24 -0300)]
video: add support for AV12
AV12 is an internally conceived format that is actually
the combination of NV12 and an alpha plane.
This format is to add to gstreamer's webM
transparency support for vp8 and vp9. To this end, two
I420 streams are independently decoded simultaneously for
the actual content and the alpha plane respectively
and these are then combined into A420.
Since most hardware decoders output NV12, this patch adds
NV12+A to make the same workflow possible.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1152>
Daniel Almeida [Fri, 21 May 2021 15:24:00 +0000 (12:24 -0300)]
video: video-orc: remove trailing spaces
Remove unnecessary trailing spaces at the end of some orc functions.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1152>
Thibault Saunier [Thu, 20 May 2021 15:16:30 +0000 (11:16 -0400)]
tests: Update expectation files with sorted structure fields
Thibault Saunier [Thu, 20 May 2021 15:24:15 +0000 (11:24 -0400)]
meson: Fix the name of the `sdp_deps` variable
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1167>
Matthew Waters [Thu, 20 May 2021 12:52:56 +0000 (22:52 +1000)]
gl/context/wgl: implement a better get_proc_address()
Look in opengl32.dll first, then wglGetProcAddress(), and only then
possibly from any linked in libraries.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1165>
Matthew Waters [Thu, 20 May 2021 12:50:23 +0000 (22:50 +1000)]
gl/context: add opengl32.dll as a library to dlopen() on windows
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1165>
Tim-Philipp Müller [Thu, 20 May 2021 22:27:21 +0000 (23:27 +0100)]
overlaycomposition: fix cut-off example code in docs
Include everything to the end.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1164>
Daniel Almeida [Thu, 6 May 2021 20:26:50 +0000 (17:26 -0300)]
gl: add support for A420 conversion
A420 is a four planar format similar to I420 but with an extra buffer
for alpha values.
A common use of the gl stack is for GPU format conversions using
shaders, in which case one can use glupload, glcolorconvert and
gldownload elements to upload their buffer to the GPU context, perform
the conversion on the GPU itself and then retrieve the data to CPU
context.
A420 was not supported. This patch adds said support mainly by adding
the corresponding shader to perform the conversion and updating the
supported caps.
Both A420->RGBA and RGBA->A420 conversions are supported.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1153>
Enrique Ocaña González [Wed, 19 May 2021 17:44:29 +0000 (19:44 +0200)]
glcolorbalance: Error out on unsupported texture target types
The issue can be reproduced on a computer with a Radeon graphics card
when trying to force GStreamer Editing Services to use GL for video
mixing in GESSmartMixer, instead of the GstCompositor that smart mixer
would normally use. This change causes the resulting video stream to
have "video/x-raw(memory:GLMemory) ... texture-target: 2D" caps (instead
of "video/x-raw ..." caps). At the PlaySink stage of the pipeline, a
GstGLImageSinkBin is plugged, with a GstGLColorBalance on it. For some
reason that is still to be debugged (and out of the scope of this
patch), gst_gl_filter_set_caps() is never called on that color balance
element, leaving filter->in_texture_target set to its default
GST_GL_TEXTURE_TARGET_NONE value. The incomplete _create_shader() logic
does the rest and silently generates a shader code that doesn't build.
This is the command I use to reproduce the issue (I'm not sure if I
would be able to isolate the issue in a simple pipeline, though):
GST_PLUGIN_FEATURE_RANK=vaapih265enc:NONE,vaapih264enc:NONE,vaapisink:NONE,vaapidecodebin:NONE,vaapipostproc:NONE,vaapih265dec:NONE,vaapivc1dec:NONE,vaapih264dec:NONE,vaapimpeg2dec:NONE,vaapijpegdec:NONE,glvideomixer:260
ges-launch-1.0 +clip /tmp/video.mp4
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1159>
Nicolas Dufresne [Wed, 19 May 2021 20:35:01 +0000 (16:35 -0400)]
compositor: Fix NV12 blend operation
The full src_height/width was being used instead of the remaining
width/height for the current band. As a side effect, that value would
get erroneously reset and would cause overrun.
Fixes #887
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1160>
Nicolas Dufresne [Tue, 18 May 2021 19:36:36 +0000 (15:36 -0400)]
video: Pass component index not plane index
While so far it worked, we are about to introduce a format that break this
assuming. We have a format which consist of NV12 with alpha, and this format
does not have a direct mapping of the component against their plane indexes.
Fix this by using gst_video_format_info_component() introduced in 1.18 for
this purpose.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1151>
Matthew Waters [Tue, 18 May 2021 10:00:01 +0000 (20:00 +1000)]
gl/context: move egl creation lower in priority on _new()
e.g. if running a dual wgl/egl built library, then egl will always
succeed in creating the GstGLContext because almost anything could
support egl, as long as eglGetDisplay() works.
wgl, however has a check for the correct display type so should move
earlier in the tried list.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1154>
Matthew Waters [Tue, 18 May 2021 09:56:13 +0000 (19:56 +1000)]
gl/display: provide a gst_gl_display_new_with_type()
Allows more fine-grained control over the exact display type that is
created.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1154>
Seungha Yang [Wed, 19 May 2021 07:23:43 +0000 (16:23 +0900)]
videoaggregator: Don't try to return void
warning C4098: 'gst_video_aggregator_parallel_convert_pad_finalize':
'void' function returning a value
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1155>