- Oct 04, 2022
-
-
Tim-Philipp Müller authored
-
Tim-Philipp Müller authored
-
- Oct 03, 2022
-
-
Always hold a reference to the soft volume element provided by the playsinkaudioconvert bin helper, the same as when volume is provided by a sink element, or the soft volume element gets unreffed too soon. Part-of: <!3108>
-
If allocating an XvShmImage generates an X error, but still returns some allocated memory, make sure to free it so it doesn't leak. Part-of: <gstreamer/gstreamer!3100>
-
Clear the stack array before setting bits to pass to XISelectEvents(). Fixes spurious crash from an X error: X Error of failed request: BadValue (integer parameter out of range for operation) Part-of: <gstreamer/gstreamer!3100>
-
Use a stack variable instead of a temporarily malloced variable that wasn't being freed properly. Fixes a small leak of a GstXvTouchDevice Part-of: <gstreamer/gstreamer!3100>
-
Make sure we're operating on a private copy of an event when modifying it. Part-of: <gstreamer/gstreamer!3107>
-
Make sure we're operating on a private copy of an event before modifying it. Part-of: <gstreamer/gstreamer!3107>
-
Make sure not to give away the ref on the final EOS event for which the probe handler is returning GST_PAD_PROBE_REMOVE when pushing the event manually. Part-of: <gstreamer/gstreamer!3107>
-
As it's done elsewhere. Avoids a potential race of the field being modified in the meantime. Part-of: <gstreamer/gstreamer!3110>
-
This is a regression that was introduced in cca2f555 (yes, 9 years ago). The only place where a demuxer streaming thread should be stopped is when the sinkpad is deactivated from pull mode (i.e. PAUSED->READY). Attempting to stop the task in this function would cause this to happen when a FLUSH_STOP or STREAM_START event is received... which can cause deadlocks. Part-of: <gstreamer/gstreamer!3109>
-
- Oct 01, 2022
-
-
It would constantly want to renegotiate (and spam the debug log) even though the channel layout hasn't actually changed. We use the same fallback in gst_ffmpegauddec_negotiate() already. This happens with WMA files for example. Part-of: <gstreamer/gstreamer!3103>
-
If splitmuxsrc exposes multiple pads, but only one is linked, part pads will never see an EOS event. This shouldn't prevent the part from being eventually deactivated. Part-of: <gstreamer/gstreamer!3099>
-
- Sep 30, 2022
-
-
Part-of: <gstreamer/gstreamer!3094>
-
And add a little unit test. Part-of: <gstreamer/gstreamer!3094>
-
- Sep 29, 2022
-
-
We need to call this to register the MusixBrainz tags before we use them in an XMP schema. Fixes this critical when attempting to run jpegparse on a JPEG containing MusicBrainz XMP tags: GStreamer-CRITICAL **: 20:41:07.885: gst_tag_get_type: assertion 'info != NULL' failed Part-of: <gstreamer/gstreamer!3092>
-
Part-of: <gstreamer/gstreamer!3091>
-
Part-of: <gstreamer/gstreamer!3091>
-
Currently if the user is not able to access the devices under /dev/media*, either due to no media devices present on the system or simply no permission to access the device, v4l2codecs initialises with no features or debug messages. Since calling `GST_DEBUG="v4l2*:7" gst-inspect-1.0 v4l2codecs` is a typical way to diagnose why element(s) failed to enumerate, we should be more verbose here when the user is not able to access any /dev/media* device. So print a simple debug message in this case to aid debugging. Part-of: <!3088>
-
- Sep 28, 2022
-
-
The purpose of a deep buffer copy is to be able to release the source buffer and all its dependencies. Attaching the parent buffer meta to the newly created deep copy needlessly keeps holding a reference to the parent buffer. The issue this solves is the fact you need to allocate more buffers, as you have free buffers being held for no reason. In the good cases it will use more memory, in the bad case it will stall your pipeline (since codecs often need a minimum number of buffers to actually work). Fixes #283 Part-of: <!2928>
-
Since commit a79a756b we could change to ignore-pcr automatically at 500ms into a live stream when no PCR is seen by then. However the stream counting in program change detection was wrongly considering ignore-pcr programs to have a separate PCR PID, even though we are actually ignoring the PCR PID completely, resulting in an erroneous program switch getting triggered from the different stream count. This in turn would send an EOS and switch out the pads for what actually is still the same program, while we intended to simply apply a workaround for broken encoders. Part-of: <gstreamer/gstreamer!3060>
-
Part-of: <gstreamer/gstreamer!3074>
-
- Sep 27, 2022
-
-
See: gstreamer/gstreamer#1136 Part-of: <gstreamer/gstreamer!2166>
-
We no longer create any symlinks. Part-of: <gstreamer/gstreamer!2166>
-
Part-of: <gstreamer/gstreamer!3079>
-
If the SETUP request returns an IPv6 server address in the Transport field, we would generate an incorrect URI, and multiudpsink would fail to initialize: ``` rtspsrc gstrtspsrc.c:9780:dump_key_value:<source> key: 'Transport', value: 'RTP/AVP;unicast;source=fe80::dc27:25ff:fe5e:bd13:8080;client_port=62696-62697;server_port=4000-4001' ... rtspsrc gstrtspsrc.c:4595:gst_rtspsrc_stream_configure_udp_sinks:<source> configure RTP UDP sink for fe80::dc27:25ff:fe5e:bd13:8080:4000 ... multiudpsink gstmultiudpsink.c:1229:gst_multiudpsink_configure_client:<udpsink0> error: Invalid address family (got 23) ``` We can't look at stream->is_ipv6 because we can't rely on the server returning the right value there. In the issue reported about this, server reported itself as `KuP RTSP Server/0.1`, and the SDP was: ``` c=IN IP4 m=video 54608 RTP/AVP 96 a=rtpmap:96 H264/90000 ``` So we need to parse the string value and figure out the family ourselves. Fixes gstreamer/gstreamer#1058 Part-of: <gstreamer/gstreamer!1819>
-
Fixes warning with meson 0.62: gst-plugins-bad| subprojects/gst-plugins-bad/meson.build:546: WARNING: Project targets '>= 0.62' but uses feature deprecated since '0.62.0': pkgconfig.generate variable for builtin directories. They will be automatically included when referenced and more. Part-of: <gstreamer/gstreamer!3086>
-
This had a couple of issues: * The backing strings (from GstStream) could disappear * The actual list wasn't properly reset/freed when decodebin3 was re-used Part-of: <gstreamer/gstreamer!3083>
-
Could possibly cause an endless loop. Fixes #1400. Part-of: <gstreamer/gstreamer!3084>
-
Update unit test for some mpd cases that were reporting timestamps including the period start time, while dashdemux2 expects that it needs to add the period start time itself. Fix the tests to not expect the period start time to be included. Part-of: <gstreamer/gstreamer!3025>
-
These values will be referred to as timestamp relative to period start so need to subtract period start time from the values. Fixes a problem with determining the start position when playing Live content with SegmentTimeline, presentationTimeOffset and a non-0 period start time. Part-of: <gstreamer/gstreamer!3025>
-
- Sep 26, 2022
-
-
Sebastian Dröge authored
Otherwise code like gst_uri_set_host(uri, gst_uri_get_host(uri)) would first free the string, then create a copy of the freed string and then assigned that. Part-of: <gstreamer/gstreamer!3076>
-
- Sep 23, 2022
-
-
Xavier Claessens authored
Update all wraps available in WrapDB. This is simply running the command "meson wrap update --force" with that Meson PR: https://github.com/mesonbuild/meson/pull/10357. Part-of: <gstreamer/gstreamer!3042>
-
Jakub Adam authored
Detect at runtime if libvpx is compiled with --enable-vp9-highbitdepth and enable 10bit video formats in element caps accordingly. Part-of: <gstreamer/gstreamer!3067>
-
- Sep 22, 2022
-
-
Previously we were unconditionally cloning the orc subproject because we want the developer environment to use the latest orc. However, some people want to use the system orc instead, or want to auto-detect which to use, with the system orc getting preference. This requires adding a new option to select that. See discussion at: gstreamer/gstreamer!2556 Fixes gstreamer/gstreamer#1282 Part-of: <gstreamer/gstreamer!3062>
-
Nirbheek Chauhan authored
gstreamer/gstreamer!3065 (comment 1563706) Part-of: <gstreamer/gstreamer!3068>
-
macOS runners can get overworked and cause cerbero pipelines to take longer than 3h to complete, because each cross-macos-universal job takes about 2h and so it can take that long to get a runner allocated. Part-of: <gstreamer/gstreamer!3065>
-
Part-of: <gstreamer/gstreamer!3065>
-
This will allow gst-plugins-rs documentation CI to include this template via a remote include and doesn't need to manually update the image. This causes cascading failures across all repos, for example gstreamer/gst-plugins-rs!874 is required for making gst-plugins-rs CI pass in Cerbero, which is also run on the gstreamer monorepo. Part-of: <gstreamer/gstreamer!3065>
-
If QoS is disabled, skip the whole computation (avoids calculating values which won't be needed) Part-of: <gstreamer/gstreamer!3034>
-