Tim-Philipp Müller [Thu, 8 Feb 2018 17:17:51 +0000 (17:17 +0000)]
meson: make version numbers ints and fix int/string comparison
WARNING: Trying to compare values of different types (str, int).
The result of this is undefined and will become a hard error
in a future Meson release.
Tim-Philipp Müller [Thu, 8 Feb 2018 16:57:53 +0000 (16:57 +0000)]
gl: g-i: expose platform-specific display API in bindings
https://bugzilla.gnome.org/show_bug.cgi?id=786391
Sebastian Cote [Fri, 2 Feb 2018 00:00:23 +0000 (00:00 +0000)]
rtspconnection: also add Content-Type to HTTP POST request when tunnelling
When the GstRTSPConnection class sends a RTSP over HTTP tunnelling
request, the HTTP Content-Type header is missing from the HTTP POST
request.
This isn't a problem with most servers, but there are servers that
rejects the request without there also being a Content-Type header.
RFC 1945:
Any HTTP/1.0 message containing an entity body should include a
Content-Type header field defining the media type of that body.
Apple Dispatch 28:
QuickTime Streaming uses the "application/x-rtsp-tunnelled" MIME
type in both the Content-Type and Accept headers. This reflects
the data type that is expected and delivered by the client and server.
https://bugzilla.gnome.org/show_bug.cgi?id=793110
Matthew Waters [Thu, 8 Feb 2018 10:14:56 +0000 (21:14 +1100)]
gl: remove more deprecated type aliases
Additions on top of
https://cgit.freedesktop.org/gstreamer/gst-plugins-base/commit/?id=
32a17f313494cbadaf8ec4e337d742e8d7e1b67b
https://cgit.freedesktop.org/gstreamer/gst-plugins-base/commit/?id=
c8b99139b1ef3f8891548b0f2607a135917c338e
Matthew Waters [Thu, 8 Feb 2018 05:13:23 +0000 (16:13 +1100)]
gl/display/egl: Fix precondition in display_egl_get_from_native
We don't really want type=NONE as input and it was already impossible
for that to occur with the other condtions.
CID #1427144
Nicolas Dufresne [Wed, 7 Feb 2018 19:48:00 +0000 (14:48 -0500)]
video: Fix GRAY10_LE32 depth information
Sebastian Dröge [Wed, 7 Feb 2018 16:40:49 +0000 (18:40 +0200)]
video: Change struct padding from void* to gpointer
gobject-introspection causes inconsistent type information for the
former and we use gpointer everywhere else.
Nicolas Dufresne [Tue, 6 Feb 2018 21:29:11 +0000 (16:29 -0500)]
meson: Enable libs_video unit test
It does not timeout anymore, even though it's a very slow test. For the
context, this test runs routines for a fixes amount of time and prints
the throughput. Which means the test takes more time everytime a pixel
format is added. If that becomes a problem again, we should disable the
benchmarks by default.
Nicolas Dufresne [Tue, 6 Feb 2018 21:16:15 +0000 (16:16 -0500)]
video-format: Fix 10LE32 formats packing function
The source offset (soff) was not incremented for each component and then
each group of 3 components were inverted. This was causing a staircase
effect combined with some noise.
https://bugzilla.gnome.org/show_bug.cgi?id=789876
Philippe Normand [Tue, 6 Feb 2018 16:13:07 +0000 (16:13 +0000)]
glcontext: remove deprecated type alias
Now for real without un-needed comments...
Philippe Normand [Tue, 6 Feb 2018 16:08:47 +0000 (16:08 +0000)]
glcontext: remove deprecated type alias
Tim-Philipp Müller [Sun, 4 Feb 2018 10:22:36 +0000 (11:22 +0100)]
autotools: use -fno-strict-aliasing where supported
https://bugzilla.gnome.org/show_bug.cgi?id=769183
Matthew Waters [Mon, 5 Feb 2018 03:58:06 +0000 (14:58 +1100)]
gl: some annotation fixes
Matthew Waters [Mon, 5 Feb 2018 03:56:07 +0000 (14:56 +1100)]
gl/build/meson: fix gl_api variable names
Tim-Philipp Müller [Wed, 22 Feb 2017 10:48:55 +0000 (10:48 +0000)]
tests: sdp: relicense code snippet from GPLv3 to LGPLv2.1+
Relicense with approval from Jose and Miguel. Code snippet
was supposed to be LGPL from the beginning.
https://bugzilla.gnome.org/show_bug.cgi?id=697808#c14
https://bugzilla.gnome.org/show_bug.cgi?id=697808#c15
Matthew Waters [Wed, 17 Jan 2018 04:24:20 +0000 (15:24 +1100)]
glmemory: fixup GL_RGB565 usage when performing texsubimage
An additional change for the texsubimage use case on top of
https://cgit.freedesktop.org/gstreamer/gst-plugins-bad/commit/?id=
c2b2c68beaddbea0ec4fe7b099507cc492f6dd7c
https://cgit.freedesktop.org/gstreamer/gst-plugins-base/commit/?id=
ea6bed111bab9b394d787f7760c6a3b953630d3a
https://bugzilla.gnome.org/show_bug.cgi?id=783066
https://bugzilla.gnome.org/show_bug.cgi?id=792584
Tim-Philipp Müller [Thu, 1 Feb 2018 11:21:19 +0000 (11:21 +0000)]
pkgconfig: only install gstreamer-gl-1.0.pc if OpenGL support was built
https://bugzilla.gnome.org/show_bug.cgi?id=793039
Edward Hervey [Thu, 1 Feb 2018 09:40:28 +0000 (10:40 +0100)]
oggdemux: Handle invalid-sized packets
On invalid packets there is the possibility we might end up wanting
to trim/offset more than what is available.
oss-fuzz issue #5866
Tim-Philipp Müller [Tue, 30 Jan 2018 20:33:36 +0000 (20:33 +0000)]
meson: use -fno-strict-aliasing if supported
https://bugzilla.gnome.org/show_bug.cgi?id=769183
Tim-Philipp Müller [Tue, 30 Jan 2018 20:32:44 +0000 (20:32 +0000)]
meson: use built-in pic kwarg when building static helper libs
instead of passing -fPIC manually.
Nicolas Dufresne [Mon, 29 Jan 2018 12:57:56 +0000 (12:57 +0000)]
glimagesink: Allow resetting render rectangle
As documented, passing -1 to x and/or y should reset the render
rectangle to the window/display size.
https://bugzilla.gnome.org/show_bug.cgi?id=792798
Nicolas Dufresne [Mon, 29 Jan 2018 12:45:06 +0000 (12:45 +0000)]
glimagesink: Add render-rectangle property
This allow controlling the render rectangle from gst-launch-1.0.
https://bugzilla.gnome.org/show_bug.cgi?id=792798
Nicolas Dufresne [Tue, 23 Jan 2018 20:04:21 +0000 (15:04 -0500)]
xvimagesink: Allow changing render-rectangle through property
This also enables setting the render rectangle before the window
is provided or created.
https://bugzilla.gnome.org/show_bug.cgi?id=792798
Nicolas Dufresne [Mon, 22 Jan 2018 20:40:32 +0000 (15:40 -0500)]
video-overlay: Add helpers for render-rectangle property
This is a set of helper that makes it easy to enable the render
rectangle to be controllable through a property.
https://bugzilla.gnome.org/show_bug.cgi?id=792798
Nicolas Dufresne [Fri, 3 Nov 2017 16:18:27 +0000 (12:18 -0400)]
video: Add NV16_10LE32 support
This adds a 10 bit variant for NV16 packed into 32 bits little endian
words. The MSB 2 bits are padding. This format is used on Xilinx SoC and
identified with the FOURCC XV20.
https://bugzilla.gnome.org/show_bug.cgi?id=789876
Nicolas Dufresne [Fri, 3 Nov 2017 16:11:58 +0000 (12:11 -0400)]
video: Add GRAY10_LE32 support
This add a 10bit variant of gray scale packed into 32bits little endian
words. The MSB 2 bits are padding and should be ignored. This format is
used on Xilinx SoC and is identified with the FOURCC XV10.
https://bugzilla.gnome.org/show_bug.cgi?id=789876
Nicolas Dufresne [Wed, 1 Nov 2017 21:26:31 +0000 (17:26 -0400)]
video: Add NV12_10LE32 support
This adds a 10bit variant for NV12 which packs 3 10bit components
into little endian 32bit words. The MSB 2 bits are padding and should be
ignored. This format is used on Xilinx SoC and is identified with there
with the FOURCC XV15
https://bugzilla.gnome.org/show_bug.cgi?id=789876
Tim-Philipp Müller [Thu, 25 Jan 2018 18:39:11 +0000 (18:39 +0000)]
subparse: fix pushing out of last chunk if last line has no newline
With playbin the last subtitle chunk would not get displayed
if the last chunk was missing a newline at the end. This is
because streamsynchronizer will hold back the EOS event until
the audio and video streams are finished too, so subparse
would never forcefully push out the last chunk until the very
end when it is too late.
We get a STREAM_GROUP_DONE event from streamsynchronizer however,
so handle that like EOS and force out any remaining text then.
https://bugzilla.gnome.org/show_bug.cgi?id=771853
Vincent Penquerc'h [Fri, 7 Apr 2017 09:33:40 +0000 (10:33 +0100)]
gst-plugins-base.supp: add gl suppressions for one off init allocations
Add to valgrind suppressions file.
https://bugzilla.gnome.org/show_bug.cgi?id=781021
Branislav Katreniak [Fri, 20 Mar 2015 08:41:05 +0000 (09:41 +0100)]
alsasrc: lock calls to snd_pcm_delay() with mutex as in alsasink
Alsasrc introduced delay_lock in commit
519f85a43e73efb8f3fb2c7be45226e
because alsa-lib is not thread safe for the same handle.
Alsasrc uses the same threading pattern, it should be locked too.
https://bugzilla.gnome.org/show_bug.cgi?id=746015
Tim-Philipp Müller [Fri, 19 Jan 2018 20:43:57 +0000 (20:43 +0000)]
tag: id3v2: don't leak stack pointer outside of block where it's valid
https://bugzilla.gnome.org/show_bug.cgi?id=788548
Tim-Philipp Müller [Fri, 19 Jan 2018 18:50:07 +0000 (18:50 +0000)]
tests: audioconvert: fix up check for lost channel positions
The caps field is channel-mask these days, so that code path
was never active.
Jimmy Ohn [Mon, 5 Jun 2017 05:08:33 +0000 (14:08 +0900)]
tests: audioconvert: Fix memory leak in failure path
Don't set a bad example by leaking things, even if calling
g_error() will make the process abort.
https://bugzilla.gnome.org/show_bug.cgi?id=783418
Edward Hervey [Wed, 17 Jan 2018 13:35:11 +0000 (14:35 +0100)]
theoradec: Check for valid width/height
If width or height are zero ... there's no video :)
Ashish Kumar [Tue, 31 Oct 2017 09:34:47 +0000 (15:04 +0530)]
playback-utils: Fix caps leak on failure
https://bugzilla.gnome.org/show_bug.cgi?id=789358
Tim-Philipp Müller [Tue, 16 Jan 2018 20:13:45 +0000 (20:13 +0000)]
tests: fix build on Windows with MSVC
Tim-Philipp Müller [Tue, 16 Jan 2018 19:22:16 +0000 (19:22 +0000)]
meson: tests: fix build of tcp test on unix
Need to add gio-unix-2.0 dep to pipelines/tcp test otherwise it
won't find the gio/gunixfdmessage.h header which is not in the
same dir as the other gio headers. This issue was masked before
because we didn't include config.h so HAVE_GIO_UNIX_2_0
wasn't defined.
Tim-Philipp Müller [Tue, 16 Jan 2018 19:21:53 +0000 (19:21 +0000)]
meson: tests: skip allocators dmabuf test on non-Linux
Tim-Philipp Müller [Tue, 16 Jan 2018 18:14:59 +0000 (18:14 +0000)]
tests: include config.h and don't include unix headers
In many cases the unistd.h includes weren't actually needed.
Don't build tests that need it on windows with MSVC
(multifdsink, multisocketsink, pipelines/tcp).
Preparation for making tests work on Windows with MSVC.
Roland Peffer [Tue, 16 Jan 2018 13:59:08 +0000 (13:59 +0000)]
meson: gl: fix build on windows
https://bugzilla.gnome.org/show_bug.cgi?id=791772
Tim-Philipp Müller [Tue, 16 Jan 2018 11:25:29 +0000 (11:25 +0000)]
meson: fix check whether both gles2 and opengl headers can be included
cc.compiles() doesn't support the prefix: kwarg currently, so it
never had any effect.
https://github.com/mesonbuild/meson/issues/2364
https://bugzilla.gnome.org/show_bug.cgi?id=787964
Sam Thursfield [Mon, 23 Jan 2017 14:30:20 +0000 (14:30 +0000)]
videotestsrc: Document the num-buffers property
It's not obvious from the existing docs that this option exists, nor
that it can be used to give a pipeline an exact length in video frames.
https://bugzilla.gnome.org/show_bug.cgi?id=777647
Tim-Philipp Müller [Mon, 8 Jan 2018 14:21:29 +0000 (14:21 +0000)]
examples: playback: don't use deprecated font button API
playback-test.c:2587:3: error: "gtk_font_button_get_font_name" is deprecated
Matthew Waters [Thu, 4 Jan 2018 04:33:33 +0000 (15:33 +1100)]
gl/wayland: move roundtrip on show to window thread
This makes it thread safe and fixes a possible deadlock.
Keeping the roundtrip off the window thread will result in two different
threads call wl_display_dispatch_queue() for the same queue which
violates the assumption for _dispatch_queue()'s thread-safety
guarantees.
https://bugzilla.gnome.org/show_bug.cgi?id=788754
https://bugzilla.gnome.org/show_bug.cgi?id=792156
https://bugzilla.gnome.org/show_bug.cgi?id=758984
Edward Hervey [Wed, 3 Jan 2018 14:31:04 +0000 (15:31 +0100)]
typefind: Fix mp3 typefinding with multiple different headers
(yes, this has never worked since it was introduced, don't worry)
If we want to actually detect layer/channels/samplerate changes,
it would be better to:
* not reset the various prev_* variables at every iteration.
* and actually store the values when they change
CID #206079
CID #206080
CID #206081
Matthew Waters [Wed, 3 Jan 2018 12:11:54 +0000 (23:11 +1100)]
libs/glheaders: move object creation to the gl thread where necessary
Some GL platforms (EGL, WGL) require deactivating the OpenGL context in
one thread before it can be used in another thread which this test
currently violates and would e.g. result in EGL_BAD_ACCESS errors from
gst_gl_context_activate().
Fix by moving the object creation into the GL thread instead and not
requiring additional gst_gl_context_activate() calls.
https://bugzilla.gnome.org/show_bug.cgi?id=792158
Matthew Waters [Wed, 3 Jan 2018 12:09:20 +0000 (23:09 +1100)]
tests/glheaders: use #if for platform selection
GST_GL_HAVE_* are always defined to 0 or 1 so an #ifdef will always
succeed which is not the intention here.
Tim-Philipp Müller [Fri, 22 Dec 2017 22:11:42 +0000 (23:11 +0100)]
glcolorconvert: re-enable -Wformat-nonliteral warning
We can pass string constants here to g_strdup_printf(),
so do so and re-enable the -Wformat-nonliteral warning
we had to disable when merging the opengl libs.
Tim-Philipp Müller [Wed, 3 Jan 2018 09:15:34 +0000 (09:15 +0000)]
glcolorconvert: fix missing printf arg for apple rectangular texture case
Found by -Wformat-nonliteral (to be re-enabled in a separate commit).
Edward Hervey [Wed, 3 Jan 2018 07:47:58 +0000 (08:47 +0100)]
configure: Fix build without gl plugins
The AM_CONDITIONAL always need to be evaluated, regardless of
whether we are building with or without gl plugins (the actual
checks are only called in AG_GST_GL_PLUGIN_CHECKS).
Philippe Normand [Sat, 30 Dec 2017 12:29:13 +0000 (12:29 +0000)]
gl: cocoa: fix formatting for function definition
This is a follow-up of
5d35e1fa15744b48c406fffa98695478fc8c9969
Philippe Normand [Tue, 26 Dec 2017 13:23:11 +0000 (13:23 +0000)]
gl: cocoa: Implement set_render_rectangle
Resize the internal NSView according to the provided render rectangle.
https://bugzilla.gnome.org/show_bug.cgi?id=791445
Tim-Philipp Müller [Tue, 26 Dec 2017 14:35:35 +0000 (15:35 +0100)]
gl: don't link to system gstallocators
We already link to the in-tree gstallocators, no need
to link to it again here, which also ends up being the
system gstallocators lib.
https://bugzilla.gnome.org/show_bug.cgi?id=791956
Tim-Philipp Müller [Tue, 26 Dec 2017 12:49:25 +0000 (13:49 +0100)]
meson: skip translations if gettext is not available
Tim-Philipp Müller [Fri, 22 Dec 2017 21:28:20 +0000 (22:28 +0100)]
meson: gl: define HAVE_PNG etc. in config.h instead of c_args
So that the tests have access to them as well (gl-launch-lines.c)
Miguel Paris [Mon, 5 Jun 2017 16:11:42 +0000 (18:11 +0200)]
rtp: fix gst_rtp_buffer_ext_timestamp taking into account backwards
If timestamp goes forwards more than allowed, we consider that the
timestamp belongs to the previous counting, so the extended timestamp
is unwrapped.
https://bugzilla.gnome.org/show_bug.cgi?id=783443
Mathieu Duponchelle [Thu, 21 Dec 2017 18:32:03 +0000 (19:32 +0100)]
tests: add missing comma
Mathieu Duponchelle [Thu, 21 Dec 2017 18:28:00 +0000 (19:28 +0100)]
tests: conditionally define the glimagesink test as well
Forgotten in previous commit
Mathieu Duponchelle [Thu, 21 Dec 2017 18:24:20 +0000 (19:24 +0100)]
tests: fix build when not building gstgl
gstgl_dep is only defined when build_gstgl is true
Tim-Philipp Müller [Wed, 20 Dec 2017 23:47:33 +0000 (23:47 +0000)]
meson: docs: link gtk-doc scanner with gstgl
Tim-Philipp Müller [Wed, 20 Dec 2017 16:01:38 +0000 (16:01 +0000)]
gl: fix build if libjpeg is not available
If libjpeg is not available, leave HAVE_JPEG undefined in
config.h, instead of defining it to 0. Fixes mismatch between
autotools conditional and ifdefs in the code.
Tim-Philipp Müller [Tue, 19 Dec 2017 16:51:24 +0000 (16:51 +0000)]
gl: fix up a few more #if HAVE_FOO
error: "HAVE_PNG" is not defined, evaluates to 0 [-Werror=undef]
Tim-Philipp Müller [Tue, 19 Dec 2017 15:48:19 +0000 (15:48 +0000)]
autotools: gl: don't check for gstreamer-allocators via pkg-config
It's in -base too after all, so this doesn't make sense and is no
longer needed.
Tim-Philipp Müller [Tue, 19 Dec 2017 12:44:35 +0000 (12:44 +0000)]
allocators: fix meson build after physmem move from -bad
Josep Torra [Tue, 12 Dec 2017 07:31:47 +0000 (08:31 +0100)]
gl: cocoa: fix a warning when building in MacOS 10.12
gstglwindow_cocoa.m:186:60: error: incompatible pointer types sending 'GstGLContextCocoa *'
(aka 'struct _GstGLContextCocoa *') to parameter of type 'GstGLContext *' (aka 'struct _GstGLContext *')
Víctor Manuel Jáquez Leal [Fri, 8 Dec 2017 13:46:35 +0000 (14:46 +0100)]
gl: display: demote error level at display creation error
https://bugzilla.gnome.org/show_bug.cgi?id=791391
Tim-Philipp Müller [Sun, 17 Dec 2017 20:48:11 +0000 (20:48 +0000)]
meson: set info variables in gstreamer-gl .pc file
Tim-Philipp Müller [Sun, 17 Dec 2017 15:21:19 +0000 (15:21 +0000)]
docs: plugins: add opengl plugin to docs
Tim-Philipp Müller [Sun, 17 Dec 2017 15:18:49 +0000 (15:18 +0000)]
docs: update for git master
Tim-Philipp Müller [Fri, 15 Dec 2017 00:40:25 +0000 (00:40 +0000)]
docs: add moved gl lib to documentation
Tim-Philipp Müller [Mon, 11 Dec 2017 11:59:57 +0000 (11:59 +0000)]
gl: hook up tests
Also move over simple gl launch line test.
Tim-Philipp Müller [Sun, 10 Dec 2017 20:02:26 +0000 (20:02 +0000)]
gl: hook up to build system
Tests and documentation will follow separately.
The mixer elements in the opengl plugin need to stay
in -bad for now since they use GstVideoAggregator.
https://bugzilla.gnome.org/show_bug.cgi?id=754094
Tim-Philipp Müller [Mon, 11 Dec 2017 00:26:36 +0000 (00:26 +0000)]
gl: examples: fix c++ code for -Wreturn-aggregate
.. and use #ifdef instead of #if
Tim-Philipp Müller [Sun, 10 Dec 2017 22:27:31 +0000 (22:27 +0000)]
gl: use #ifdef HAVE_* instead of #if HAVE_FOO
Tim-Philipp Müller [Sat, 9 Dec 2017 20:16:48 +0000 (20:16 +0000)]
gl: remove GST_USE_UNSTABLE_API bits
Tim-Philipp Müller [Sun, 10 Dec 2017 21:08:28 +0000 (21:08 +0000)]
opus: remove unused build variable and define
Sebastian Dröge [Fri, 17 Feb 2017 17:26:41 +0000 (19:26 +0200)]
allocators: move GstPhysMemoryAllocator abstraction from -bad to -base
This can be used in a generic way as common interface by all platforms
that, in one way or another, pass around physical memory addresses.
This is used by the gl lib and seems useful enough, so might just as
well move it next to the other allocators.
https://bugzilla.gnome.org/show_bug.cgi?id=779067
Tim-Philipp Müller [Tue, 19 Dec 2017 12:00:43 +0000 (12:00 +0000)]
Move OpenGL library and plugin from -bad
Merge branch 'opengl-move'
https://bugzilla.gnome.org/show_bug.cgi?id=754094
Nicolas Dufresne [Sun, 10 Dec 2017 19:59:54 +0000 (14:59 -0500)]
videoconvert: Filter-out crop meta
To passthrough crop-meta, the converter would need to allocate and
convert buffers of the size of the originating buffer. This is currently
made difficult by GstBaseTransform since we cannot alter the caps passed
though the allocation query. We would also need to wait for the first
input buffer to be received in order to make the decision around that
size.
So the short and safe solution is just to stop pretending we can
passthrought that meta.
https://bugzilla.gnome.org/show_bug.cgi?id=791412
Dongil Park [Fri, 15 Dec 2017 01:50:44 +0000 (10:50 +0900)]
playbin3: Fix accessing invalid index in GstStream when received select-stream event
If select-stream event was send to playbin3 as missing any GstStream of ES type
(V or A or TEX) of collection then, playbin will access to invalid address of
GstStream due to invalid index limit. This caused SIGSEGV.
https://bugzilla.gnome.org/show_bug.cgi?id=791638
Matthew Waters [Thu, 14 Dec 2017 03:48:54 +0000 (14:48 +1100)]
Automatic update of common submodule
From e8c7a71 to 3fa2c9e
Jun Xie [Mon, 11 Dec 2017 03:45:37 +0000 (11:45 +0800)]
xvimagesink: fix inaccurate error message
It's about not being able to calculate the display size, not the display
ratio.
https://bugzilla.gnome.org/show_bug.cgi?id=791463
Víctor Manuel Jáquez Leal [Tue, 5 Dec 2017 09:37:40 +0000 (10:37 +0100)]
gldownload: fix wrong enum
When compiling with clang, an enum conversion error is triggered
since GstVideoFrameFlags are not GstVideoFlags.
This patch sets GST_VIDEO_FRAME_FLAG_NONE to the added video meta.
https://bugzilla.gnome.org/show_bug.cgi?id=791251
Nicolas Dufresne [Sun, 3 Dec 2017 00:28:30 +0000 (19:28 -0500)]
gldownload: Add missing ifdef for dmabuf and egl
This fixes the build for platforms that don't support one or the other.
Nicolas Dufresne [Sat, 2 Dec 2017 20:25:38 +0000 (15:25 -0500)]
gldisplay: Prefer wayland over X11
As most Wayland compositors supports XWayland, X11 backend get
selected. This also realign better GStreamer decision to what
happens with GTK and other stack out there.
Nicolas Dufresne [Sat, 2 Dec 2017 20:25:16 +0000 (15:25 -0500)]
gldisplay: Add missing GL API to the doc
Matt Fischer [Wed, 11 Jan 2017 01:23:58 +0000 (19:23 -0600)]
gldownload: Add dmabuf exporting
This patch adds code to gldownload to export the image as a
dmabuf if requested. The element now exposes memory:DMABuf as
a cap feature, and if it is selected, the element exports the
texture to an EGL image and then a dmabuf. It also implements a
fallback to system memory download in case the exportation failed.
https://bugzilla.gnome.org/show_bug.cgi?id=776927
Tim-Philipp Müller [Sat, 2 Dec 2017 16:01:25 +0000 (16:01 +0000)]
Remove GstAggregator from -bad, moved to core
https://bugzilla.gnome.org/show_bug.cgi?id=739010
Matthew Waters [Mon, 27 Nov 2017 03:44:58 +0000 (14:44 +1100)]
gl/caopengllayer: use public GstGLContext instead of Cocoa-specific one
Allows keeping the GstGLCAOpenGLLayer public but not the winsys-specific
context/display/window.
Matthew Waters [Mon, 27 Nov 2017 03:44:24 +0000 (14:44 +1100)]
Revert "gl: cocoa: sprinkle some GST_EXPORT"
This reverts commit
94d798c3337013bd902c0fd0a66f7f6842243eeb.
cocoac headers don't need to be public as all the functionality can be
provided by the base class
Tim-Philipp Müller [Sun, 26 Nov 2017 22:36:35 +0000 (22:36 +0000)]
gl: cocoa: sprinkle some GST_EXPORT
Undefined symbols for architecture x86_64:
"_gst_gl_context_cocoa_get_type", referenced from:
__create_layer in libgstopengl_la-caopengllayersink.o
Might need some more in other headers, but first need to
clarify what exactly should be exported, there are some
inconsistencies (installed header files vs. funcs in docs).
Nicolas Dufresne [Fri, 24 Nov 2017 22:06:22 +0000 (17:06 -0500)]
sdlshare: Destroy GL backend buffer before the GL Context
This was otherwise leading to a deadlock in the GL library.
Nicolas Dufresne [Fri, 24 Nov 2017 21:58:21 +0000 (16:58 -0500)]
sdlshare2: Fix race conditions
The client-draw callback is running on the GL Thread, which will
be required to map the buffer. Map early, and pass the mapped
frame instead. On top of that, make sure to signal any pending
draw before trying to push EOS, as some pad locks might be taken.
This is the cost of using the same thread to control GStreamer and
to render GL.
Nicolas Dufresne [Fri, 24 Nov 2017 21:58:01 +0000 (16:58 -0500)]
meson: Build gl/sdl examples
Jan Schmidt [Thu, 23 Nov 2017 12:01:52 +0000 (23:01 +1100)]
viv-fb: Don't destroy the native FB display
It causes crashes in applications because the result of
fbGetDisplay() might be in use elsewhere in the application
and Vivante doesn't seem to do any refcounting
Jan Schmidt [Thu, 23 Nov 2017 11:58:40 +0000 (22:58 +1100)]
Revert "gl: Use GstGLDisplayEGL directly instead of creating a GstGLDisplayVIVFb subclass"
This reverts commit
47fd4d391e775c11f529705bb0f457a9d25ba5e7.
This patch is incorrect. It doesn't actually compile, and causes a crash
because the viv-fb window implementation needs a native EGL handle
to pass to fbCreateWindow, but the GstGLDisplayEGL handleis actually
an EGLDisplay now (and gets cast to the wrong type)
Jan Schmidt [Thu, 23 Nov 2017 14:37:48 +0000 (01:37 +1100)]
3dvideo example: Correct video overlay for HiDPI
Use the GTK scaling factor to scale the video allocation
so video displays correctly on hi-dpi screens
Tim-Philipp Müller [Thu, 23 Nov 2017 10:15:49 +0000 (11:15 +0100)]
meson: remove outdated comment in build file
Mathieu Duponchelle [Mon, 6 Nov 2017 20:07:51 +0000 (21:07 +0100)]
aggregator: Remove klass->sinkpads_type
This posed problems for the python bindings (and possibly others).
Instead, subclasses now use add_pad_template_with_gtype.
https://bugzilla.gnome.org/show_bug.cgi?id=789986
Edward Hervey [Mon, 20 Nov 2017 16:06:07 +0000 (17:06 +0100)]
all: Fix left-shift undefined behaviour
Cast to the target type before shifting (or use macro if available)