Mart Raudsepp [Thu, 26 Nov 2020 12:52:26 +0000 (14:52 +0200)]
gl/eagl: Fix automatic resize behaviour
https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/681
added a layoutSubViews, which never gets called, because it should have been
called layoutSubviews (non-capital "v"). However after fixing that, it still
doesn't work correctly, because window_width/height values are immediately
updated and then draw_cb will never trigger the resize path, because the
values are already up to date.
Update the values inside the resize path again instead, so the check for
entering the resize path is logically always correct.
This makes the layoutSubviews unnecessary, as it only updated the internal
size values prematurely, so it is deleted instead of method naming fixed.
These changes were originally done to avoid accessing UIKit objects on the
main thread, but no additional accesses are added here, only internal
private variable assignments under the same draw_lock, so there should be
no threading issues reintroduced.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/945>
Mart Raudsepp [Thu, 26 Nov 2020 12:45:05 +0000 (14:45 +0200)]
gl/eagl: Fix resize condition check in draw_cb to not get called unnecessarily
A CGSize contains CGFloat values (a typedef to double or float), which means
that the values aren't equal, despite it being equal after they are cast to
int by assigning them to window_height/width private members. This leads to
excessive gst_gl_window_resize calls on each frame, at least if the CGFloat
value has a .5 decimal value, e.g. 103.5.
Fix it by storing them as CGFloat instead of gint.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/945>
Matthew Waters [Fri, 20 Nov 2020 05:51:09 +0000 (16:51 +1100)]
video/converter: increase the number of cache lines for resampling
The exising hardcoded max default does not account for the possible
-1 offset when retrieving lines for resampling. As a result, when
another chain has the same number of cache lines (4), the resample
operation would be attempting to generate 5 lines with a cache size
of 4 and would overwrite the first cache line.
Fixes https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/issues/821
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/938>
Sebastian Dröge [Tue, 1 Dec 2020 11:13:40 +0000 (13:13 +0200)]
compositor: Don't crash in prepare_frame() if the pad was just removed
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/950>
Guillaume Desmottes [Tue, 1 Dec 2020 08:31:48 +0000 (09:31 +0100)]
videometa: gir annotate the size of plane array in new API
Fix #838
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/949>
Thibault Saunier [Thu, 19 Nov 2020 22:06:12 +0000 (19:06 -0300)]
encodebin: Enhance algorithm to check if profiles can be plugged to muxer
In particular muxing sinks have specific set of sinkpads and we need
to handle that.
Thibault Saunier [Thu, 19 Nov 2020 22:03:29 +0000 (19:03 -0300)]
encodebin: Ignore element classification when a muxer name specified
Instead of going through the list of known muxers go ahead and
instantiate the muxer specified as 'preset name' as this specifies
the exact element factory name to use.
Thibault Saunier [Fri, 13 Nov 2020 19:32:45 +0000 (16:32 -0300)]
pbutils: Add support for muxing sinks usage in encoding profiles
Thibault Saunier [Fri, 13 Nov 2020 19:32:45 +0000 (16:32 -0300)]
encoding: Implement encodebin2 with an ON_REQUEST src pad
Fixes https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/issues/304
Thibault Saunier [Fri, 13 Nov 2020 16:06:32 +0000 (13:06 -0300)]
encodebin: Split implementation into a base class
Create EncodeBaseBin as a base class for the existing
encodebin to allow other implementations.
Co-authored with Jan Schmidt <jan@centricular.com>
Seungha Yang [Fri, 27 Nov 2020 19:14:14 +0000 (04:14 +0900)]
glcontext: wgl: Implement check_feature vfunc
There are several WGL specific extenstions such as WGL_NV_DX_interop.
Currently we have no WGL specific extension support and
this commit is also only for debugging purpose.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/947>
Sanchayan Maity [Fri, 27 Nov 2020 06:11:36 +0000 (11:41 +0530)]
audiodecoder: Move max_errors out of GstAudioDecoderContext
Currently max-errors gets set during init to default or via property.
However, if a decoder element calls gst_audio_decoder_reset with 'full'
argument set to TRUE, it would result in all the fields of context being
zeroed with memset. This effectively results in max-errors getting a
value of 0 overriding the default or user requested value set during
init.
This would result in calls to GST_AUDIO_DECODER_ERROR which track error
counts and allow max-errors, to be ineffective.
To fix this move max-errors out of GstAudioDecoderContext, as changes to
context should not affect this. The error_count is anyways also in
GstAudioDecoderPrivate and not in context.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/946>
Stéphane Cerveau [Thu, 26 Nov 2020 16:37:01 +0000 (17:37 +0100)]
opus: disable docs option in meson fallback
As the docs are not necessary, no need to keep it
enabled.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/943>
Marijn Suijten [Wed, 25 Nov 2020 19:16:53 +0000 (20:16 +0100)]
video: Provide "deprecated in" version for gst_video_color_transfer fns
As requested in [1].
[1]: https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/940#note_706437
Fixes:
d0f36c7e1 ("video: Rename video_color_transfer to video_transfer_function")
Marijn Suijten [Tue, 24 Nov 2020 19:54:49 +0000 (20:54 +0100)]
audio: Move fill_silence into audio_format_info
With the function named gst_audio_format_fill_silence it would get
associated to the GstAudioFormat type in .gir which is incorrect and
confusing. See [1] for the discussion sparking this change.
https://gitlab.freedesktop.org/gstreamer/gstreamer-rs/-/merge_requests/630#note_694795
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/940>
Jan Schmidt [Wed, 25 Nov 2020 12:54:08 +0000 (23:54 +1100)]
xvimagesink: Add support for the XV_COLORSPACE attribute.
The XV_COLORSPACE attribute exists on some Xv adapters, with the same
semantics as the XV_ITURBT_709 attribute that was already supported.
A value of 0 is bt601, and 1 is for bt709 colorspace.
Fixes color shifting issues displaying bt709 content on some Xv
adapters.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/941>
Edward Hervey [Fri, 13 Nov 2020 16:45:12 +0000 (17:45 +0100)]
decodebin3: Take into account decoder latency for interleave size
Some decoders might introduce quite large latencies, which would result in
multiqueue draining out on some streams. In order to avoid that, check the
latency of decoders and adjust the minimum interleave time of multiqueue
accordingly.
Fixes https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/issues/800
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/931>
Edward Hervey [Thu, 19 Nov 2020 13:35:32 +0000 (14:35 +0100)]
decodebin3: Don't leak filter caps
Introduced by previous commit
See https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/933
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/934>
Edward Hervey [Sat, 14 Nov 2020 08:43:59 +0000 (09:43 +0100)]
decodebin3: Properly handle caps query with no filter
There's no guarantee that upstream elements will do queries with specified
filter caps. If that's the case, just return GST_CAPS_ANY
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/933>
Tim-Philipp Müller [Sat, 7 Nov 2020 17:30:19 +0000 (17:30 +0000)]
opus: add subproject fallback
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/924>
Mathieu Duponchelle [Sat, 7 Nov 2020 01:59:09 +0000 (02:59 +0100)]
compositor: expose zero-size-is-unscaled property
When that property is left to its default, the width and height
property considers frames from input pads with width or height <= 0
should be left unscaled in that dimension.
Setting this property to FALSE changes that behaviour to < 0, as when
animating these properties, 0 should be a valid end value (eg. shrinking
an input stream until it disappears).
The default value of the width and height properties is set to -1, so that
the default behaviour stays consistent whether that new property is set
or not.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/923>
Mathieu Duponchelle [Tue, 27 Oct 2020 23:01:16 +0000 (00:01 +0100)]
video-converter: switch to using a task pool ..
.. and make use of that API in videoaggregator.
When setting certain properties, such as cropping or the scaled
size of pads, a new converter is created by videoaggregator.
Before that patch, this implied spawning new threads, potentially
at each aggregate cycle when interpolating pad properties. This
is obviously wasteful, and re-using a task pool removes that
overhead.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/896>
David Keijser [Tue, 10 Nov 2020 22:26:39 +0000 (23:26 +0100)]
Fix segfault when using invalid encoding profile
Trying to use gst_encoding_profile_get_file_extension on a
GstEncodingProfile with a cap containing a typo would result in strcmp
being called with NULL. Instead use g_strcmp0 that handles this case.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/929>
Jan Alexander Steffens (heftig) [Tue, 3 Nov 2020 16:00:53 +0000 (17:00 +0100)]
videoaggregator: Fix locking around vagg->info
Take `GST_OBJECT_LOCK` when writing `vagg->info`, so that reading in
subclasses is protected against races, as documented in the struct.
/*< public >*/
/* read-only, with OBJECT_LOCK */
GstVideoInfo info;
`gst_video_aggregator_default_negotiated_src_caps` should take the
`GST_VIDEO_AGGREGATOR_LOCK` to avoid racing with
`gst_video_aggregator_reset` called by
`gst_video_aggregator_release_pad` of the last sinkpad. Otherwise it can
happen that `latency = gst_util_uint64_scale (...` gets called with a
zero framerate.
There doesn't seem to be any reason not to use the local `info` instead
of `vagg->info`, so do that.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/915>
Guillaume Desmottes [Mon, 26 Oct 2020 07:59:34 +0000 (08:59 +0100)]
video: fix doc warning
@mode has been renamed to
gst_video_decoder_set_interlaced_output_state() but not in the header
file, raising a doc warning.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/894>
Jan Schmidt [Wed, 4 Nov 2020 17:43:47 +0000 (04:43 +1100)]
video-converter: Skip input lines where possible.
There is a case where there are no lines in the temp cache, and
it's possible to skip straight to the request line and not generate
intermediate ones. This is really only beneficial when doing
nearest-neighbour downscaling, as other methods generally require
all input lines sequentially to generate the output. In that case,
this change has no effect and all lines are generated and cached
as before.
As a side effect however, this fixes corruption when downscaling
using nearest-neighbour, as interactions with the pass_alloc flag
and reuse of temporary lines causes the unecessarily-generated
cache lines to overwrite the final output.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/919>
Jan Schmidt [Tue, 3 Nov 2020 15:11:30 +0000 (02:11 +1100)]
examples: Add a compositor mosaic example.
Add an example of configuring compositor to display 2 videos
side-by-side in a mosaic, and additionally configure some
video-converter properties to demonstrate how that works..
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/919>
He Junyan [Wed, 4 Nov 2020 15:05:27 +0000 (23:05 +0800)]
gluploadelement: Avoid race condition in propose_allocation().
The inside upload and context may have race condition in the function
of propose_allocation(). They may be destroyed while this function is
stilling using it.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/916>
He Junyan [Wed, 4 Nov 2020 09:02:13 +0000 (17:02 +0800)]
gluploadelement: Avoid race condition of base class' context.
The base class' context may change, we should use the common API
with lock to access it, rather than the just directly access the
struct field.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/916>
Nirbheek Chauhan [Wed, 4 Nov 2020 13:12:11 +0000 (18:42 +0530)]
meson: Enable some MSVC warnings for parity with GCC/Clang
This makes it easier to do development with MSVC by making it warn
on common issues that GCC/Clang error out for in our CI configuration.
Continuation from https://gitlab.freedesktop.org/gstreamer/gst-build/-/merge_requests/223
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/918>
He Junyan [Tue, 3 Nov 2020 12:19:16 +0000 (20:19 +0800)]
gluploadelement: Avoid race condition of inside upload creation.
The operations for the inside GstGLUploadElement->upload have race
condition. The _transform_caps() will creates this object if it does
not exist, while the _stop() and change_state() can destroy this object.
The _transform_caps() is called by the gst_base_transform_query(),
so it does not hold the stream lock. It may use the upload while the
_stop() and change_state() has already destroy that object, and then
crash.
Fix: #645
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/913>
Sebastian Dröge [Tue, 3 Nov 2020 14:40:38 +0000 (16:40 +0200)]
gl: Fix prototype of glGetSynciv()
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/914>
Thibault Saunier [Fri, 30 Oct 2020 16:56:16 +0000 (13:56 -0300)]
videoaggregator: Guarantee that the output format is supported
In the case `videoaggregator` is set as allowing format conversions,
and as we convert only on the sinkpads, we should ensure that the
chosen format is usable by the subclass. This in turns implies
that the format is usable on the srcpad.
When doing conversion *any* format can be used on the sinkpads, and this
is the only way that we can avoid race conditions during renegotiations
so we can not change that fact, we just need to ensure that the chosen
intermediary format is usable, which was not actually ensured before
that patch.
Fixes https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/issues/834
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/909>
Seungha Yang [Fri, 30 Oct 2020 18:14:20 +0000 (03:14 +0900)]
videodecoder: Don't assume GstVideoChromaSite and GstVideoColorimetry
Even if given GstVideoChromaSite and/or GstVideoColorimetry has unknown
value(s), assumption for an unknown value should be done by subclass or
downstream element, not a role of video decoder. And subclass might
want to output unknown value as is.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/910>
Seungha Yang [Fri, 30 Oct 2020 18:16:00 +0000 (03:16 +0900)]
videodecoder: Remove trailing whitespace
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/910>
Jan Alexander Steffens (heftig) [Fri, 30 Oct 2020 12:52:29 +0000 (13:52 +0100)]
tcpserversink: Don't assume g_socket_get_remote_address succeeds
When the client disconnects immediately after connecting, the remote
address is no longer available.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/908>
Thibault Saunier [Thu, 29 Oct 2020 19:38:44 +0000 (16:38 -0300)]
video-aggregator: Fix renegotiation when using convert pads
Since
23189c60f4cff998c7880e1768cee2f6d1b719d0 we started using the
useless result of `modified_caps` which, was never used since it was
introduced 7 years ago (in videomixer2). The intersection is useless and
we should just avoid doing it at all (which was always the case before)
as it can end up failing renegotiation for bad reasons.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/905>
Seungha Yang [Wed, 27 May 2020 14:20:25 +0000 (23:20 +0900)]
gst-play: Add support for audio mute control
Assign 'm' keyboard input to control audio mute state
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/674>
Nicolas Dufresne [Wed, 28 Oct 2020 19:51:27 +0000 (15:51 -0400)]
tests: allocator: Fix FDMemory portability issue
This fixes few issues in the test but mainly some portability issue reported
on Ubutun. The test now uses a randomly name tempory file located into system
default tempory location and uses glib wrappers when available.
Fixes !895
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/901>
Dinesh Manajipet [Sat, 24 Oct 2020 18:15:19 +0000 (23:45 +0530)]
gl: optimize glcolorbalance by precomputing shader math
Precompute the rgb -> yuv conversion and color balance adjustment math
so that the shader does minimal work per pixel.
Merging these 15+ steps into 3 steps let us jump from choppy
360p video to smooth 720p video on our underpowered embedded system.
If we can remove the clamp() step inside the shader, or apply it after
rgba
conversion, there are more performance benefits to reap. But I am not
sure
what the side effects will be in that case.
<https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/893>
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/893>
Matthew Waters [Wed, 28 Oct 2020 12:26:27 +0000 (23:26 +1100)]
gl/build: use the brcm GL libraries on the rpi
Upstream RPi has moved to a completely separate GL library names now due
to conflicts.
See https://github.com/RPi-Distro/repo/issues/134
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/898>
Tobias Ronge [Tue, 27 Oct 2020 13:12:42 +0000 (14:12 +0100)]
fdmemory: Allow for change of protection mode
After a memory has been unmapped, protection mode can now be changed
when mapping it again.
See https://bugzilla.gnome.org/show_bug.cgi?id=789952.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/895>
Mathieu Duponchelle [Wed, 28 Oct 2020 00:19:37 +0000 (01:19 +0100)]
videoaggregator: document and fix locking in convert pad
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/897>
Tim-Philipp Müller [Wed, 28 Oct 2020 13:22:09 +0000 (13:22 +0000)]
gldeinterlace: fix doc blurb
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/899>
Xavier Claessens [Sun, 8 Apr 2018 00:52:12 +0000 (20:52 -0400)]
Meson: Use pkg-config generator
Seungha Yang [Thu, 22 Oct 2020 10:16:42 +0000 (19:16 +0900)]
urisourcebin: Fix crash caused by use after free
OutputSlotInfo doesn't hold ref of queue, so gst_bin_remove()
will free the queue memory.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/888>
Zebediah Figura [Wed, 21 Oct 2020 19:55:40 +0000 (14:55 -0500)]
theoradec: Set telemetry options only if they are nonzero
Setting telemetry options, even to zero, causes libtheora to enable an expensive code path. For large enough videos (e.g. 1920x1080) this can increase the time to decode each frame by 30-40 ms, which can be enough to cause noticeable stutter.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/887>
Michael Tretter [Wed, 21 Oct 2020 09:42:54 +0000 (11:42 +0200)]
glslstage: delete shader on finalize of stage
GLSLstage creates the glShader using glCreateShader, but never calls
glDeleteShader if the glShader is not used anymore. This forces the GL
library to keep the compiled shader around, because it might be used in
the future. Therefore, the glShader is leaked whenever a GLSLStage is
destroyed.
Fix the leak by deleting the glShader when finishing the GLSLStage.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/886>
Sebastian Dröge [Tue, 20 Oct 2020 09:44:21 +0000 (12:44 +0300)]
audio/videodecoder: Initialize max_errors in instance_init()
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/882>
Sebastian Dröge [Tue, 20 Oct 2020 08:51:08 +0000 (11:51 +0300)]
audio/videodecoder: Don't reset max-errors in reset()
Otherwise setting the property on the elements has no effect at all
because it's immediately reset during startup.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/881>
Seungha Yang [Mon, 12 Oct 2020 18:34:07 +0000 (03:34 +0900)]
examples: Add an uridecodebin3 example
Demonstrate a way to select multiple streams
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/880>
Seungha Yang [Mon, 19 Oct 2020 11:03:40 +0000 (20:03 +0900)]
decodebin3: Store stream-start event on output pad before exposing it
It's required for users to be able to figure out associated GstStream
object with a pad on pad-added callback.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/880>
Sebastian Dröge [Sat, 17 Oct 2020 09:35:16 +0000 (12:35 +0300)]
Add some missing nullable annotations
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/876>
Sebastian Dröge [Sat, 17 Oct 2020 07:42:49 +0000 (10:42 +0300)]
typefind/xdgmime: Validate mimetypes to be valid GstStructure names before using them
On macOS, for example, "text/*" can be returned as mimetype for
plaintext files but we don't allow '*' in structure names and this would
cause critical warnings.
It's a valid mimetype but not a valid structure name.
Fixes https://gitlab.freedesktop.org/gstreamer/gstreamer/-/issues/616
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/875>
Stéphane Cerveau [Thu, 15 Oct 2020 16:20:37 +0000 (18:20 +0200)]
meson: update glib minimum version to 2.56
In order to support the symbol g_enum_to_string in various
project using GStreamer ( gst-validate etc.), the glib minimum
version should be 2.56.0.
Remove compat code as glib requirement
is now > 2.56
Version used by Ubuntu 18.04 LTS
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/874>
Seungha Yang [Sun, 20 Sep 2020 10:47:08 +0000 (19:47 +0900)]
videoconvert: Enhance debug message to be more human readable
Numerical representation of GstVideoFormat is not debugging
friendly
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/822>
Seungha Yang [Tue, 13 Oct 2020 08:57:53 +0000 (17:57 +0900)]
uridecodebin3: Forward upstream events to decodebin3 directly
Otherwise default handler will try to forward it to sink element
or srcpad which might not be available at that moment.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/867>
Seungha Yang [Mon, 14 Sep 2020 11:01:18 +0000 (20:01 +0900)]
tests: appsrc: Fix unstable test case
Wait all buffers to be consumed before sending flush seek event,
so that checking timestamp and segment as expected.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/816>
Jan Alexander Steffens (heftig) [Tue, 13 Oct 2020 13:31:41 +0000 (15:31 +0200)]
videoaggregator: Limit accepted caps by template caps
This seems like an obvious mistake, since `modified_caps` was created
but not used.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/868>
Jan Alexander Steffens (heftig) [Tue, 13 Oct 2020 13:28:54 +0000 (15:28 +0200)]
videoaggregator: Log the format name instead of index
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/868>
Will Miller [Mon, 12 Oct 2020 11:21:17 +0000 (12:21 +0100)]
gstrtpbuffer: fix header extension length validation
We validate the header extensions length of an RTP buffer by comparing
it against the block size. Since we multiply the length in words by 4 to
get the length in bytes, a suitably large length could cause a wrapround
of the uint16, giving a lower length which erroneously passes the check
and allows the buffer to be mapped.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/864>
Marijn Suijten [Fri, 11 Sep 2020 18:28:55 +0000 (20:28 +0200)]
gl: glslstage: Sync new_default_{fragment,vertex} documentation name
Documentation for the parameter and return value don't show up when
this header mismatches the function it tries to describe.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/814>
François Laignel [Sat, 10 Oct 2020 15:26:42 +0000 (17:26 +0200)]
decodebin3: only force streams-selected seqnum after a select-streams
The initial streams-selected message might not match any select-streams,
in which case it is currently overriden with GST_SEQNUM_INVALID.
Related to:
- https://gitlab.freedesktop.org/gstreamer/gstreamer-rs/-/merge_requests/601
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/860>
Jan Alexander Steffens (heftig) [Fri, 9 Oct 2020 15:15:37 +0000 (17:15 +0200)]
examples/gl/gtk: Add missing dependency on gstgl
Occasionally, a clean build would fail like this:
In file included from ../subprojects/gst-plugins-base/tests/examples/gl/gtk/gstgtk.c:24:
../subprojects/gst-plugins-base/gst-libs/gst/gl/gl.h:25:10: fatal error: gst/gl/gl-enumtypes.h: No such file or directory
25 | #include <gst/gl/gl-enumtypes.h>
| ^~~~~~~~~~~~~~~~~~~~~~~
Add the missing dependency so that the headers are generated beforehand.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/855>
Mathieu Duponchelle [Fri, 4 Sep 2020 19:09:17 +0000 (21:09 +0200)]
rtsptransport: support single port for client_port and server_port
This is explicitly specified as valid in the RFC, where the
syntax for both parameters is:
";" "client_port" "=" port [ "-" port ]
";" "server_port" "=" port [ "-" port ]
This is useful for applications where RTCP is either not necessary
or not possible.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/846>
Guillaume Desmottes [Fri, 9 Oct 2020 10:02:14 +0000 (12:02 +0200)]
video-info: clarify GST_VIDEO_INTERLACE_MODE_ALTERNATE doc
Fields really should alternate in this mode.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/852>
Guillaume Desmottes [Mon, 5 Oct 2020 10:07:22 +0000 (12:07 +0200)]
videodecoder: fix output state interlace-mode
When user is passing the actual interlace-mode when calling
gst_video_decoder_set_interlaced_output_state() it should not be
overidden by the input interlace-mode.
Needed to fix #825 as we want to keep interlace-mode=interleaved from
parsers and have the OMX decoder producing interlace-mode=alternate.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/852>
Guillaume Desmottes [Mon, 5 Oct 2020 09:38:09 +0000 (11:38 +0200)]
videodecoder: add interlace-mode to debug output when setting output state
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/852>
Matthew Waters [Thu, 16 Jul 2020 08:10:45 +0000 (18:10 +1000)]
tests: add gl structs to abi check
Tested on x86, x86_64, armv7l, aarch64.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/854>
Matthew Waters [Fri, 9 Oct 2020 01:25:21 +0000 (12:25 +1100)]
build: use cpu_family for arch checks
e.g. on 32-bit arm, we may have armv6, armv7l, armv7hf, etc which all
generally have the same layouts. cpu_family() groups all of these into
just 'arm' that the ABI check table is expecting.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/854>
Jan Alexander Steffens (heftig) [Thu, 1 Oct 2020 09:22:51 +0000 (11:22 +0200)]
multihandlesink: Don't pass NULL caps to gst_caps_is_equal
Apparently the sinkpad caps can get cleared when shutting down the
pipeline while the sink is working, provoking a critical warning.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/842>
Sebastian Dröge [Thu, 8 Oct 2020 14:10:30 +0000 (17:10 +0300)]
timeoverlay: Add elapsed-running-time mode
This takes the first running time and the first after each flush-stop as
an offset for the running time that is rendered.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/737>
Jan Alexander Steffens (heftig) [Thu, 8 Oct 2020 11:52:33 +0000 (13:52 +0200)]
audio: video: Fix in/outbuf confusion of transform_meta
There are three instances where in- and outbuf have been swapped. This
didn't affect the correctness of the libs *filter code, but the
videoscale implementation swapped the arguments of meta->transform_func.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/853>
Xavier Claessens [Thu, 1 Oct 2020 18:45:24 +0000 (14:45 -0400)]
pango: Add fallback dependency
When we'll depend on Meson >=0.55 this can be replaced by [provide]
section in gst-build's pango.wrap file. The dependency variable name can
also be dropped when pango is updated to use override_dependency:
https://gitlab.gnome.org/GNOME/pango/-/merge_requests/250
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/844>
Seungha Yang [Tue, 15 Sep 2020 19:30:40 +0000 (04:30 +0900)]
meson: Always wrap "prefix" option with join_paths() to make Windows happy
Both "\" and "/" separators are allowed on Windows but the problem
is that "\" is not converted to "\\" in this case.
It will result to unknown/unrecognized character escape sequence.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/818>
Xavier Claessens [Fri, 25 Sep 2020 18:05:09 +0000 (14:05 -0400)]
meson: Add GL summary
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/836>
Matthew Waters [Mon, 21 Sep 2020 13:50:03 +0000 (23:50 +1000)]
compositor: fix YUY2/UYVY/VYUY blending
Don't overead memory.
Fixup for https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/755
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/823>
Nicolas Dufresne [Tue, 15 Sep 2020 18:15:36 +0000 (14:15 -0400)]
meson: Add a qt5 feature
This allow opting out items of the builds the depends on QT5 library.
Auto-detection of QT5 in cross-build requires host tools to match with the
sysroot, and detection of mis-match is not fully reliable.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/817>
sohwan.park [Mon, 14 Sep 2020 08:56:07 +0000 (17:56 +0900)]
meson: add abi configuration for meson build
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/815>
Matthew Waters [Wed, 15 Jul 2020 05:46:56 +0000 (15:46 +1000)]
compositor: blend with multiple threads
Increases the throughput of compositing by using more CPU cycles across
multiple threads. Simple cases (the output contains one pixel from at
most one input) can have up to a 70% increase in throughput. Not so
simple cases are limited by the region with the most number of
composite operations.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/755>
Marijn Suijten [Fri, 11 Sep 2020 18:15:14 +0000 (20:15 +0200)]
video: Rename video_color_transfer to video_transfer_function
Rename remaining `gst_video_color_transfer_{encode,decode}` functions on
the `GstVideoTransferFunction` enumeration to
`gst_video_transfer_function_{encode,decode}` permitting
gobject-introspection to turn these into associated functions and place
them under the respective `<enumeration>` block in gir XML files.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/805>
Matthew Waters [Fri, 11 Sep 2020 11:40:26 +0000 (21:40 +1000)]
videoaggregator: ensure peek_next_sample() uses the correct caps
gst_pad_get_current_caps() may be wrong when there is a renegotiation in
progress for the pad and we have not yet received or selected the buffer
with different caps yet.
Fix by storing the caps through in a similar way to the existing code
for buffer/video-info selection machinery.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/813>
Stian Selnes [Fri, 18 Aug 2017 15:01:02 +0000 (17:01 +0200)]
videodecoder: Bump log level of marking a sync point
Sync points are often important when debugging and deserves DEBUG
level.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/730>
Sebastian Dröge [Wed, 1 Jul 2020 15:19:09 +0000 (18:19 +0300)]
videodecoder: Add API for subclasses to request a new sync point
This allows subclasses that notice missing reference frames to request a
new sync point to allow seamless decoding again. While doing so the
subclass can also signal whether it wants a) all following input frames
until the sync point to be discarded or b) all output frames until the
sync point to be marked as corrupt.
Sending of force-keyunit events for this can be throttled by the
application via the "min-force-keyunit-interval" property.
This replaces custom behaviour for the same in various decoders, for
example openh264dec.
Based on patches by Haakon Sporsheim <haakon@pexip.com> and
Stian Selnes <stian@pexip.com>.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/730>
Sebastian Dröge [Wed, 1 Jul 2020 15:14:40 +0000 (18:14 +0300)]
videodecoder: Add API for subclasses to signal that they need the stream to start with a sync point
If the first frame(s) at the very beginning or after a flush are not a
sync point then the base class would discard them before passing them to
the subclass.
This also fixes the previously broken distance_from_sync handling: it
was never reset at sync points.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/730>
Sebastian Dröge [Wed, 1 Jul 2020 15:11:25 +0000 (18:11 +0300)]
videodecoder: Add "discard-corrupted-frames" property
This can be used by applications to configure decoders so that corrupted
frames are directly discarded instead of being forwarded inside the
pipeline. It is a replacement for the "output-corrupt" property of the
ffmpeg decoders.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/730>
Sebastian Dröge [Wed, 1 Jul 2020 15:09:22 +0000 (18:09 +0300)]
videodecoder: Add API for marking output frames as corrupted
This can be used by subclasses to mark output frames as known to be
corrupted, for example if reference frames were missing. ffmpeg's
decoders can signal this.
In addition this flag is propagated downstream if the input frame had it
set.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/730>
Sebastian Dröge [Tue, 23 Jun 2020 15:40:07 +0000 (18:40 +0300)]
videodecoder: Remove unused reorder_depth variable
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/730>
Mikhail Fludkov [Wed, 1 Jul 2020 16:18:36 +0000 (19:18 +0300)]
rtpbasedepayload: Mark GAP events sent because of packet loss as such
This allows downstream to distinguish packet loss from normal GAP events
that are sent simply because of gaps in the timeline.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/731>
Nirbheek Chauhan [Tue, 8 Sep 2020 10:11:55 +0000 (15:41 +0530)]
meson: Actually build gstgl without implicit include dirs
Fixes a typo from https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/800
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/810>
Sebastian Dröge [Thu, 10 Sep 2020 11:19:04 +0000 (14:19 +0300)]
glmixer: Fix unit test to actually work reliably
Don't run the harness in live mode, or otherwise it would output frames
already in the very beginning before a buffer was provided to it due to
timeout.
Also send EOS/a second buffer before pulling a buffer as videoaggregator
has one frame of latency.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/812>
Haihao Xiang [Thu, 13 Aug 2020 04:49:19 +0000 (12:49 +0800)]
gl: add support for GBR and GBRA
example pipeline:
gst-launch-1.0 videotestsrc ! video/x-raw,format=GBR ! glimagesink
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/789>
Sebastian Dröge [Thu, 20 Aug 2020 06:06:04 +0000 (09:06 +0300)]
videoaggregator: Don't require any pads to be configured for negotiating source pad caps
This is not actually required (anymore?). Source pad caps can be
negotiated at any time regardless of any configured (or existing) sink
pads and videoaggregator comes up with some fixated caps based on the
downstream caps.
Subclasses can override this behaviour as needed by overriding
update_src_caps().
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/793>
Marijn Suijten [Thu, 3 Sep 2020 19:19:29 +0000 (21:19 +0200)]
gst-libs: gl: Fix documentation typo and clarify gl_memory_texsubimage
`gst_gl_memory_read_pixels` reads pixels from `GLMemory` into the
pointer, effectively writing to it. This is opposite from
`gst_gl_memory_texsubimage` which reads texture data from `read_pointer`
into `GLMemory`.
Both cases are clarified by changing `read_pointer` to `write_pointer`,
and explaining what `gst_gl_memory_texsubimage` does in addition to
referring back to `gst_gl_memory_read_pixels`.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/806>
Sebastian Dröge [Thu, 20 Aug 2020 13:31:38 +0000 (16:31 +0300)]
audioaggregator: Reset offset if the output rate is renegotiated
On next aggregation the new offset will be calculated based on the
segment position.
Without this a rate change would cause a jump forwards or backwards in
the output timeline.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/794>
Sebastian Dröge [Tue, 18 Aug 2020 07:42:09 +0000 (10:42 +0300)]
video-anc: Implement transform functions for AFD/Bar metas
If the transformation is just a copy, copy over the metadata. For AFD
also copy over the metadata if the aspect ratio stays the same in
scaling transformations. In all other cases fail the transformation.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/791>
Sebastian Dröge [Mon, 24 Aug 2020 16:37:15 +0000 (19:37 +0300)]
appsrc: Wake up the create() function on caps changes
This allows the new caps to be sent downstream as soon as possible
instead of having to wait for the next buffer/buffer list.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/798>
Thibault Saunier [Tue, 8 Sep 2020 23:45:53 +0000 (20:45 -0300)]
encodebin: Be less restrictive on muxer caps when preset name is specified
This way the user can simply define its caps by only providing the
structure name which is what we do in the encoding profile
deserialization function for example.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/811>
Thibault Saunier [Thu, 30 Jul 2020 22:49:52 +0000 (18:49 -0400)]
encodebin: Add vp8 and vp9 smart encoding support
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/770>
Thibault Saunier [Thu, 30 Jul 2020 22:49:18 +0000 (18:49 -0400)]
encodebin: Fix refcount mishandling
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/770>