Fixes 2 problems:
1) Number of unmapped memories does not always match number of mmaped ones in
dispatch_write().
2) When dispatch_write() is dispatched second time after an incomplete write,
already set offsets will not be taken into account, thus corrupt RTP data will
be sent.
This makes it unnecessary for callers to first merge together all
memories, and it allows API like GstRTSPConnection to write them out
without first copying all memories together or using writev()-style API
to write multiple memories out in one go.
Fixes https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/issues/370
g_source_remove() works only for a GSource which was attached
to default GMainContext, but the GSource might be attached to
custom context depending on how gst_discoverer_start() was called.
Whatever the attached context was, g_source_destroy() can clean it up.
Make consistent with what autotools puts into enabled_gl_apis
variable. Autotools puts 'gl' in there instead of 'opengl'.
This would cause problems when building -bad glmixers plugin
in meson against a -base that was built with autotools.
See https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/issues/871
Otherwise surface_width/surface_height stored in GstGLWindowPrivate
isn't changed, sometimes an unnecessary reconfigure event is sent on
sinkpad, then result in upstream reconfiguring.
Example pipeline:
gst-launch-1.0 videotestsrc ! msdkvpp ! glimagesink
This reverts commit 69c3c31608.
All devices have the same name, they are duplicated with pulseaudio one
and the provided does not respond to HW being plugged/unplugged. I think
it's not ready for 1.16.
The start_time and end_time in this context have already
been adjusted for the input's rate by converting them to running
time above. What is needed afterwards is to compare these
with the output's start/stop running time, which also takes
into account the rate, so we are comparing equal things.
Multiplying these with the output's rate here is only breaking
this logic. In most cases the input and output rate is the same,
so this multiplication effectively reverses the rate adjustment
that happened while converting to running time, which is why
we see the video playing with the original rate in tests.
Fixes#541
Binding the vertex array to 0 will unbind everything else already.
In the previous order older versions of the Intel GL driver caused
errors to be printed for every single call when disabling the vertex
attrib arrays after binding the vertex array to 0.
We use all those libraries internally and include headers from them in
the public headers.
And add the tag library to Requires.private as we use it internally and
it would be needed when doing static linking.
Fixes https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/issues/537
It might be the case that glgsinkbin would try to set a property to
its internal sink which doesn't exist in it, leading to a glib's
warning. For example, when playsink sets 'force-aspect-ratio' property
and glsinkbin has, as internal sink, appsink, which doesn't handle
that property.
The patch validates the incoming property to forward to internal sink
if it exists in the internal sink and both properties has the same
type.
We make an allocator for temporary lines and then use this for all
the steps in the conversion that can do in-place processing.
Keep track of the number of lines each step needs and use this to
allocate the right number of lines.
Previously we would not always allocate enough lines and we would
end up with conversion errors as lines would be reused prematurely.
Fixes#350
ISO 14496-3 defines that audioObjectType 5 is a special case that
indicates SBR is present and that an additional field has to be
parsed to find the true audioObjectType.
There are two ways of signaling SBR within an AAC stream - implicit
and explicit (see [1] section 4.2). When explicit signaling is used,
the presence of SBR data is signaled by means of the SBR
audioObjectType in the AudioSpecificConfig data.
Normally the sample rate is specified by an index into a
table of common sample rates. However index 0x0f is a special case
that indicates that the next 24 bits contain the real sample rate.
[1] https://www.telosalliance.com/support/A-closer-look-into-MPEG-4-High-Efficiency-AACFixes#39
(Initially discussed in
https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/issues/305)
The ticks waveform can be useful for audio synchronization diagnostics
and other cases where the time offset between waveforms is important.
However, in its current form, it is too limited, and has problems with
discontinuities, which result in severe artifacts when this waveform
is output by a DAC.
This patch fixes some discontinuities and considerably expand the ticks
waveform's flexibility. They also introduce the notion of a "marker tick";
every Nth tick can have a different amplitude (usually one that is larger
than the others). This is useful for combining frequent oscilloscope
triggering with large time offset detection. For example, without marker
ticks, the tick intervals must not be too small, otherwise the maximum time
offset that can be unambiguously detected is quite small (for example, if
the interval is 50ms, then no time offset larger than 25ms can be
unambiguously recognized). If the tick intervals are too far apart, then
no sudden changes can be clearly observed, since the oscilloscope is not
updated quickly enough. But with marker ticks, this is not an issue: If
there's for example a tick every 100 ms, then the oscilloscope can be
triggered every 100 ms. And, if every 20th tick is a marker tick, then
time offsets of up to 1 second can be discovered, even though the time
between ticks is 100 ms.
The patch also applies some minor cleanup to the audiotestsrc documentation.
Checking the address distance between given begin/end sequence
doesn't make sense. They are output params.
This is to fix weird failure of libs_rtp on Windows
Code in g_return_*() must not have side effects, as it
might be compiled out if -DG_DISABLE_CHECKS is used, in
which case we would read garbage off the stack.
Both versions are basically the same, but version 2.0 also allows
60000/1001 as framerate and allows to specify the field and line number
for each payload.
Put the major version into the caps so that elements can limit via caps
negotiation which versions they can support.
It breaks all the calculations. While it can make sense during
initialization, there's very little API that can be called with such
timecodes without ending up with wrong results.
