https://bugzilla.gnome.org/show_bug.cgi?id=797363

We were reading more bytes than we allocated.

https://bugzilla.gnome.org/show_bug.cgi?id=797363

There is new code that ensures that we renegotiate after an
uploader transition if the negotiated caps have changed.

The problem is that the raw uploader will not really try and
fixate the input caps, but instead of return a subset with the
only the supported target texture.

This had two effect, raw uploads was always done renegotiated
once and the raw upload unit test was now failing as it didn't
expect a renegotiation.

As it's a valid check, simply relax the gst_caps_is_equal() check
and use a gst_caps_is_subset() instead.

https://bugzilla.gnome.org/show_bug.cgi?id=783521

The direct dmabuf upload does color conversion, so when it transforms
the caps, it replaces the format with all formats found through the
format query. When this uploader can't be used, it makes the upstream
source pick a unsupported format.

To fix this, we only append the caps with a list of format. So the
source will only pick one of these formats if the downstream preferred
format is not supported. A negotiation failure after this would be
normal.

This fixes pipelines without a glcolorconvert element.

https://bugzilla.gnome.org/show_bug.cgi?id=783521

The unit test makes mixed usage of ret value. Sometimes its does
stores an enum and at other moment a boolean. Also fix test
using boolean instead of the correct enum value.

https://bugzilla.gnome.org/show_bug.cgi?id=783521

https://bugzilla.gnome.org/show_bug.cgi?id=797134

This new property controls the synchronisation offset between the text and video
streams. Positive values make the text ahead of the video and negative values
make the text go behind the video.

https://bugzilla.gnome.org/show_bug.cgi?id=797134

This new property controls the synchronisation offset between the text and video
streams. Positive values make the text ahead of the video and negative values
make the text go behind the video.

https://bugzilla.gnome.org/show_bug.cgi?id=797134

When the playsink contains a text chain this property controls the
synchronisation of the subtitles and video by controlling the underlying
subtitleoverlay::subtitle-ts-offset property.

https://bugzilla.gnome.org/show_bug.cgi?id=797134

This property controls the synchronisation offset between text and video in
nanoseconds, by updating the parser src pad offset.

https://bugzilla.gnome.org/show_bug.cgi?id=797134

wangfei authored Oct 09, 2018 and Nicolas Dufresne committed Oct 31, 2018

This pixel format is packed format with 4:2:2 sample and 10
available bits of each channel.

https://bugzilla.gnome.org/show_bug.cgi?id=797267

Philipp Zabel authored Oct 17, 2018 and Michael Olbrich committed Oct 30, 2018

Report in the DEBUG log if the driver does not support importing a given
format with linear modifiers non-externally.

https://bugzilla.gnome.org/show_bug.cgi?id=783521

Philipp Zabel authored Oct 17, 2018 and Michael Olbrich committed Oct 30, 2018

The EXT_image_dma_buf_import_modifiers extension [1] states regarding
eglQueryDmaBufModifiersEXT:

    The format must be one of those returned by the
    eglQueryDmaBufFormatsEXT command.

To comply with this requirement eglQueryDmaBufFormatsEXT must be called
before eglQueryDmaBufModifiersEXT.

[1] https://www.khronos.org/registry/EGL/extensions/EXT/EGL_EXT_image_dma_buf_import_modifiers.txt

https://bugzilla.gnome.org/show_bug.cgi?id=783521

Philipp Zabel authored Oct 17, 2018 and Michael Olbrich committed Oct 30, 2018

https://bugzilla.gnome.org/show_bug.cgi?id=783521

Philipp Zabel authored Oct 17, 2018 and Michael Olbrich committed Oct 30, 2018

Calculate DRM fourcc and report to the DEBUG log about it only once
instead of three times in gst_egl_image_from_dmabuf_direct().

https://bugzilla.gnome.org/show_bug.cgi?id=783521

The idea is that some GPUs (like the Vivante series) can actually
perform the YUV->RGB conversion internally, so no custom conversion
shaders are needed. To make use of this feature, we need an additional
uploader that can import DMABUF FDs and also directly pass the pixel
format, relying on the GPU to do the conversion.

Based on patches from Nicolas Dufresne <nicolas.dufresne@collabora.com> and
Carlos Rafael Giani <dv@pseudoterminal.org>.

https://bugzilla.gnome.org/show_bug.cgi?id=783521

Without this, a buffer is dropped if glupload indicates that it is
necessary to reconfigure.
Avoid this by explicitly reconfiguring immediately and uploading the buffer
again.

https://bugzilla.gnome.org/show_bug.cgi?id=783521

If a upload method is selected then use it exclusively in transform_caps().
Also, reconfigure if the current caps don't match the current upload
method.

https://bugzilla.gnome.org/show_bug.cgi?id=783521

This should not be necessary, but currently not all plugins that provide
dmabuf memory announce this with caps features, e.g. v4l2.
The static caps already contain the system memory. It didn't break before
because other upload methods provide the necessary transformation.

https://bugzilla.gnome.org/show_bug.cgi?id=783521

Reconfigure will trigger a set_caps which clears the upload method.
Remember the method in this case and start with it.
Wrap around once to try all methods if necessary.

https://bugzilla.gnome.org/show_bug.cgi?id=783521

The colorspace conversion happens during the upload so the necessary hints
must be provided to ensure that the conversion works correctly.

At least the Mesa Intel driver will create a texture without error but
produces an incorrect result. Use eglQueryDmaBufModifiersEXT() to check if
non-external upload is supported for the given format.

Based on a patch from Carlos Rafael Giani <dv@pseudoterminal.org>.

https://bugzilla.gnome.org/show_bug.cgi?id=783521

Nicolas Dufresne authored Jun 07, 2017 and Michael Olbrich committed Oct 30, 2018

gst_gl_memory_setup_buffer() was not properly using the number
of pointers to wrapped. This also fixes the validation, as we
only support 1 wrapper per view, or num_planes * views wrapper.

https://bugzilla.gnome.org/show_bug.cgi?id=783521

Johan Bjäreholt authored Oct 29, 2018 and Sebastian Dröge committed Oct 29, 2018

https://bugzilla.gnome.org/show_bug.cgi?id=797353

overlaycomposition: New element that allows applications to draw GstVideoOverlayComposition on a stream

https://bugzilla.gnome.org/show_bug.cgi?id=797234

Before freeing pending pads it is required to hold the mutex, that's
what is protecting the list of pending pads in other places.

https://bugzilla.gnome.org/show_bug.cgi?id=796935

Edward Hervey authored Oct 27, 2018 and Edward Hervey committed Oct 27, 2018

The previous failure was a timeout which was due to the sending pipeline
pushing test buffer *before* the remote client was accepted. We would
therefore never get the buffer on the other side.

While the client socket would indeed appear as "connected", this doesn't
mean that the remote server side did "accept" it (which is where we then
add it to the list of remote parties to which data will be sent).

The problem isn't with the element implementation, but to the nature of
TCP 3-way handshake.

In order to make the test reliable, wait for the sink to have accepted
the remote client (by checking the number of handles) before sending out
test buffers.

rlandj authored Jul 24, 2018 and Tim-Philipp Müller committed Oct 18, 2018

https://bugzilla.gnome.org/show_bug.cgi?id=796860

Seungha Yang authored Sep 27, 2018 and Tim-Philipp Müller committed Oct 17, 2018

Use 'copy' keyword to avoid meson warning message.
Note that 'copy' keyword in configure_file() is available
since meson 0.47.0

https://bugzilla.gnome.org/show_bug.cgi?id=797298

Otherwise it's very easy to miss them when gst_video_frame_map() fails

rtsp_connection_send takes care of adding those already,
and some reverse proxies such as nginx will reject the request
altogether if the Authorization header is present twice,
even with the same value.

https://bugzilla.gnome.org/show_bug.cgi?id=797272

Stian Selnes authored Oct 10, 2018 and Olivier Crête committed Oct 10, 2018

Add a source-info property that will read/write meta to the buffers
about RTP source information. The GstRTPSourceMeta can be used to
transport information about the origin of a buffer, e.g. the sources
that is included in a mixed audio buffer.

A new function gst_rtp_base_payload_allocate_output_buffer() is added
for payloaders to use to allocate the output RTP buffer with the correct
number of CSRCs according to the meta and fill it.

RTPSourceMeta does not make sense on RTP buffers since the information
is in the RTP header. So the payloader will strip the meta from the
output buffer.

https://bugzilla.gnome.org/show_bug.cgi?id=761947

https://github.com/mesonbuild/meson/pull/4169

https://bugzilla.gnome.org/show_bug.cgi?id=797173

https://bugzilla.gnome.org/show_bug.cgi?id=797225

Matej authored Sep 29, 2018 and George Kiagiadakis committed Oct 05, 2018

otherwise caps intersection always fails when converting non-interleaved to interleaved audio

https://bugzilla.gnome.org/show_bug.cgi?id=797225

This reverts commit 452d90a8.

This was actually correct. sorry for the noise!

Fixes the internal viewconvert to not scale buffers for output with the
following pipeline:

gltestsrc ! glimagesink

It also fixes overlay composition with a resized output with an OpenGL
upstream:

gltestsrc ! timeoverlay ! glimagesink