GitLab

The priority of the thread that executes audioringbuffer_thread_func
is incremented on Windows by the usage of the AvSetMmThreadCharacteristics
API. This change has to be restored, as described on the documentation
of the API (https://docs.microsoft.com/en-us/windows/win32/api/avrt/nf-avrt-avsetmmthreadcharacteristicsw#remarks),
with a call to the AvRevertMmThreadCharacteristics. If this is not done,
a handle will be leaked.

Part-of: <!760>

All APIs in avrt.h are desktop only.

Part-of: <!763>

We don't install any cocoa/win32 specific headers.

Part-of: <!759>

Fixes -Wmissing-field-initializers in Clang.

Part-of: <!757>

Part-of: <!757>

Part-of: <gstreamer/gst-plugins-base!754>

This adds linear 32x32 NV12 based tiles. This format is notably used by
Allwinner VCU and exposed in V4L2 as being "SUNXI Tiled" format. In this
patch we generalize the plane info calculation so we can share this part
with the 4L4 variant.

Part-of: <gstreamer/gst-plugins-base!754>

Part-of: <!753>

When using tile format, the stride has a different meaning. It used
the MSB and LSB 16bits to encode respectively the width and height in
number of tiles.

This issue was introduce with commit e5b70d38 which was fixing
missing size recalculation when strides and offset is updated.

Part-of: <!753>

This format is produced by Verisillicon VC8000D VPU decoder, it is a simple 4x4
tiling layout in a linear way.

Part-of: <!753>

Part-of: <gstreamer/gst-plugins-base!707>

-Waggregate-return: used by some Qt clases extensively and not super
useful for this example. Supress it.

warning: "GL_GLEXT_VERSION" redefined: Perform the same workaround as
qmlglsink by defining the old gl/GL.h header guard if the new GL/gl.h
guard exists.

Part-of: <gstreamer/gst-plugins-base!752>

Part-of: <gstreamer/gst-plugins-base!476>

Fixes #684

Part-of: <!749>

All UI elements will follow Single-Threaded Apartments (STA) model.
As a result, we should access them from dedicated UI thread.
Due to the nature of the threading model, ANGLE will wait the UI
thread while closing internal window/swapchain objects.

A problem here is that when destroying GstGLWindow from the UI thread,
it will wait GstGLContext's internal thread. Meanwhile, the GstGLContext's
internal thread will be blocked because ANGLE wants to access the UI thread.
That will cause a deadlock or exceptions.

In short, application should not try to call
gst_element_set_state(pipeline, GST_STATE_NULL) from a UI thread.
That's a limitation of current implementation.

Part-of: <!745>

GstGLWindow implmentaion should be able to report native window size
and also it need to handle resize event for glimagesink.

Note that GstD3D11Window implementation was referenced for this change.

Part-of: <!745>

For native window size query, c++ API is much convenient than c API

Part-of: <!745>

Don't put double ':' there

Part-of: <!663>

Part-of: <!663>

Add property "handle-segment-change" for user to allow pushing
custom segment event. For now, this property can work only for
time format GstSegment.
This property can be useful in case application controls timeline
of stream such as there is timestamp discontinuity but playback is
expected to be continuous. Multi-period scenario of MPEG-DASH is an
example of this use case.

Part-of: <!663>

We have the notion of presence, and when the user want to be in control
it is totally legitimate for him to have several occurrences of a
similar profile

Part-of: <!746>

Since we are using VideoMeta, the converter (similarly to the video_frame_copy
utility) should have no issue dealing with frames that are slightly larger.
This situation occure as some element will use padded width/height for
allocation, which results in a VideoMeta width/height being larger then the
display width/height found in the negotiated caps.

Fixes #790

Part-of: <!747>

get_allowed_caps() will return NULL, which is not a problem in itself.
Just take the template caps for negotiation in that case instead of
erroring out.

Part-of: <!744>

And fix up DOAP file XML. Parser would complain about
unknown entity ! here.

Part-of: <!743>

Don't put gl into the libraries list if the gst-gl library isn't
being built, and also don't include it in the list of linker flags
then.

Fixes gstreamer/gst-plugins-base#613

Part-of: <gstreamer/gst-plugins-base!742>

Calling gst_gl_window_send_{key,mouse}_event() from GstGLContext
thread might cause a deadlock. Instead, use the dedicated event handling
thread in GstGLDisplay.

Part-of: <!721>

Add test option "--fullscreen" for testing fullscreen mode switch.
When enabled, user can switch fullscreen mode via SPACE key or
right mouse button click.

Part-of: <!721>

We added GL_CFLAGS in autotools because we unconditionally included OpenGL
headers. We do not do that anymore for a standard #include <gst/gl/gl.h>.
OpenGL headers are only included for the platform specific files
e.g. #include <gst/gl/x11/gstgldisplay_x11.h> which must always be hidden
behind a #if GST_GL_HAVE_WINDOW_X11, or #include <gst/gl/gstglfuncs.h> to
gain access to the contents of the GstGLFuncs object.

Users of our headers must include the necessary include paths for the
platforms they are intending to support.

We might introduce extra .pc files for what GstGL was built against though.

Fixes #735

Part-of: <!741>

The header file might not be available on non-*nix platform.

Part-of: <!740>

meson.build:4: WARNING: Trying to compare values of different types (ExecutableHolder, bool) using ==.

Part-of: <!739>

Part-of: <!738>

Forcing software-decoder has no effect on playbin3 now.

Part-of: <!738>

This reverts commit f4bcf829.

Part-of: <!738>

Whenever a new collection is calculated, the internal `select_streams_seqnum`
variable is reset. This ensures that we reliably know whether a select-streams
event has been received for that new collection.

Use that to decide whether we should add previously un-selected streams or new
streams in the current selection

Fixes #784

Part-of: <!733>

Part-of: <!727>

Part-of: <!734>

Add target to update backup orc -dist.[ch] files.

Part-of: <!734>

For example, BT709, BT601, and BT2020_10 all have theoretically
different transfer functions, but the same function in practice. In
these cases, we should use the fast path for negotiating. Also,
BT2020_12 is essentially the same as the other three, just with one more
decimal point, so it gives the same result for fewer bits. This is now
also aliased to the former three.

Also make videoconvert do passthrough if the caps have equivalent
transfer functions but are otherwise matching.

As of the previous commit, we write the correct transfer function for
BT601, instead of the (functionally identical but different ISO code)
transfer function for BT709. Files created using GStreamer prior to that
commit write the wrong transfer function for BT601 and are, strictly
speaking, 2:4:5:4 instead. However, this commit takes care of
negotiation, so that conversions from/to the same transfer function are
done using the fast path.

Fixes #783

Part-of: <!724>

Functionally the same as 709 but technically has a different value, and
external software (e.g. ffmpeg) finds "wrong" values produced by
GStreamer.

Part-of: <!724>