George Kiagiadakis [Wed, 12 Mar 2014 15:03:10 +0000 (17:03 +0200)]
waylandsink/wldisplay: bind to the latest available wl_compositor version
George Kiagiadakis [Wed, 12 Mar 2014 12:28:44 +0000 (14:28 +0200)]
waylandsink: create/destroy the display when entering/leaving the READY state instead of PAUSED
This is the only way to get the negotiation working with the dynamic
detection of formats from the display, because the pipeline needs
to know the supported formats in the READY state and the supported
formats can only be known if we open the display.
Unfortunately,in wayland we cannot have a separate connection to
the display from the rest of the application, so we need to ask for a
window handle when going to READY in order to get the display from it.
And since it's too early to create a top level window from the state
change to READY, create it in render() when there is no other window.
This also changes set_window_handle() to not support window handle
changes in PAUSED/PLAYING (because it's complex to handle and useless
in practice) and make sure that there is always a valid display pointer
around in the READY state.
George Kiagiadakis [Wed, 12 Mar 2014 12:13:49 +0000 (14:13 +0200)]
waylandsink: Support all video formats supported by the display
George Kiagiadakis [Wed, 12 Mar 2014 11:54:44 +0000 (13:54 +0200)]
waylandsink: fix crash in case there is no pool because of a caps negotiation error
George Kiagiadakis [Tue, 11 Mar 2014 17:47:52 +0000 (19:47 +0200)]
waylandsink: set an empty input region on the video surface
George Kiagiadakis [Tue, 11 Mar 2014 17:46:56 +0000 (19:46 +0200)]
waylandsink/wlwindow: reuse code between the two constructors
George Kiagiadakis [Tue, 11 Mar 2014 16:45:23 +0000 (18:45 +0200)]
waylandsink: increase debug messages
George Kiagiadakis [Tue, 11 Mar 2014 15:48:46 +0000 (17:48 +0200)]
waylandsink: Use a boolean in combination with render_cond to comply with GCond's usage documentation
George Kiagiadakis [Tue, 11 Mar 2014 15:45:05 +0000 (17:45 +0200)]
waylandsink: Implement expose() and handle resizing properly in non-PLAYING states
George Kiagiadakis [Tue, 11 Mar 2014 11:14:00 +0000 (13:14 +0200)]
waylandsink: Use wl_scaler/wl_viewport to scale the surface in the compositor/hardware
George Kiagiadakis [Tue, 11 Mar 2014 11:05:56 +0000 (13:05 +0200)]
waylandsink: Build bindings for the unstable wl_scaler spec
George Kiagiadakis [Mon, 10 Mar 2014 11:50:06 +0000 (13:50 +0200)]
waylandsink: Set external surfaces and their child objects to use our own event queue
This fixes weird freezes because of frame_redraw_callback() not being
called from the main thread when it should with weston's toy toolkit.
It's also safer to know that frame_redraw_callback() will always be
called from our display thread... Otherwise it could be called after
the sink has been destroyed for example.
George Kiagiadakis [Fri, 7 Mar 2014 15:25:00 +0000 (17:25 +0200)]
waylandsink: Wait for the frame_cb to redraw and drop frames meanwhile
We are not supposed to redraw until we receive a frame callback and this
is especially useful to avoid allocating too many buffers while the
window is not visible, because the compositor may not call wl_buffer.release
until the window becomes visible (ok, this is a wayland bug, but...).
George Kiagiadakis [Fri, 7 Mar 2014 14:16:30 +0000 (16:16 +0200)]
waylandsink: Handle wl_buffer::release and don't reuse buffers that are not released
This is achieved by adding an extra reference on the buffers, which does
not allow them to return to the pool. When they are released, this reference
is dropped.
The rest complexity of this patch (hash table, mutex, flag, explicit release calls)
merely exists to allow a safe, guaranteed and deadlock-free destruction sequence.
See the added comment on gstwaylandsink.c for details.
George Kiagiadakis [Thu, 6 Mar 2014 15:03:50 +0000 (17:03 +0200)]
waylandsink/waylandpool: remove useless munmap call
This data pointer is part of a bigger mmap'ed region,
it has not been returned from mmap itself.
George Kiagiadakis [Fri, 28 Feb 2014 11:37:30 +0000 (13:37 +0200)]
waylandsink/waylandpool: call the start/stop methods of the parent class
start() makes sure that the minimum ammount of buffers requested is allocated.
stop() makes sure that buffers are actually destroyed and prevents
filling the file system when resizing the surface a lot, because the
wayland-shm-* files will stay on the file system as long as the wl_buffers
created out of them are alive.
George Kiagiadakis [Fri, 28 Feb 2014 11:36:43 +0000 (13:36 +0200)]
waylandsink/waylandpool: unlink mmaped shm files so that they don't remain on the file system
George Kiagiadakis [Fri, 28 Feb 2014 09:48:30 +0000 (11:48 +0200)]
waylandsink: implement the GstVideoOverlay & GstWaylandVideo interfaces
This is the initial implementation, without the GstVideoOverlay.expose()
method. It only implements using an external (sub)surface and resizing
it with GstWaylandVideo.
George Kiagiadakis [Fri, 14 Feb 2014 15:08:56 +0000 (16:08 +0100)]
waylandsink: implement with stubs the GstWaylandVideo & GstVideoOverlay interfaces
George Kiagiadakis [Mon, 6 May 2013 10:16:02 +0000 (13:16 +0300)]
wayland: Add new gst-wayland library containing a new GstWaylandVideo interface
This interface is needed to be able to embed waylandsink into
other wayland surfaces. Due to the special nature of wayland,
GstVideoOverlay is not enough for this job.
George Kiagiadakis [Wed, 26 Feb 2014 16:35:29 +0000 (18:35 +0200)]
waylandsink: handle the list of supported formats properly
enum wl_shm_format is not a flags enum, as it may have been in the past,
so multiple formats cannot be stored in a bitfield. Use an array instead.
George Kiagiadakis [Wed, 26 Feb 2014 15:41:11 +0000 (17:41 +0200)]
waylandsink/wlvideoformat: add mappings for many common formats
George Kiagiadakis [Wed, 26 Feb 2014 14:20:41 +0000 (16:20 +0200)]
waylandsink: unref the buffer pool
George Kiagiadakis [Wed, 26 Feb 2014 14:11:29 +0000 (16:11 +0200)]
waylandsink/waylandpool: ref the display instead of the sink to avoid cyclic references
The reference to the sink is not really needed anyway in waylandpool,
what matters basically is that the display is active as long as the
pool is active, so we really want to reference the display object
instead of the sink.
George Kiagiadakis [Wed, 26 Feb 2014 12:56:21 +0000 (14:56 +0200)]
waylandsink: make the display property useful
Let the display property control the name of the display,
like in x(v)imagesink.
George Kiagiadakis [Fri, 14 Feb 2014 16:20:42 +0000 (17:20 +0100)]
waylandsink: access sink->pool in a more atomic fashion
George Kiagiadakis [Fri, 14 Feb 2014 15:33:10 +0000 (16:33 +0100)]
waylandsink: remove the useless wayland_lock
George Kiagiadakis [Thu, 13 Feb 2014 12:28:40 +0000 (13:28 +0100)]
waylandsink: apply the same debug category to all the subobjects
George Kiagiadakis [Thu, 13 Feb 2014 12:15:31 +0000 (13:15 +0100)]
waylandsink: cleanup header includes
George Kiagiadakis [Thu, 13 Feb 2014 10:59:45 +0000 (11:59 +0100)]
waylandsink: split window-related code out to a new GstWlWindow class
GstWlWindow also has API ready to support subsurfaces.
George Kiagiadakis [Thu, 13 Feb 2014 10:32:00 +0000 (11:32 +0100)]
waylandsink: remove callback and redraw_pending variables from the window structure
George Kiagiadakis [Thu, 13 Feb 2014 09:51:59 +0000 (10:51 +0100)]
waylandsink/waylandpool: improve debug message
George Kiagiadakis [Thu, 13 Feb 2014 09:37:01 +0000 (10:37 +0100)]
waylandsink/waylandpool: find the video format from the GstVideoInfo instead of accessing the sink
For the sake of isolation only. The format should be the same.
George Kiagiadakis [Thu, 13 Feb 2014 09:29:08 +0000 (10:29 +0100)]
waylandsink/waylandpool: refactor code
* make use of GstBufferPool::start/stop functions to allocate/deallocate memory
* get rid of struct shm_pool and do all operations cleanly inside WaylandBufferPool
* store a GstVideoInfo during configuration instead of the width & height
and use the stride from the video info instead of hardcoding its value
George Kiagiadakis [Thu, 13 Feb 2014 09:02:54 +0000 (10:02 +0100)]
waylandsink: split video format related functions out to a separate file
George Kiagiadakis [Wed, 12 Feb 2014 14:16:08 +0000 (15:16 +0100)]
waylandsink/waylandpool: move code around for better readability
George Kiagiadakis [Wed, 12 Feb 2014 13:41:52 +0000 (14:41 +0100)]
waylandsink: move struct shm_pool and its related functions to waylandpool.c
And also make the instance of this struct to be owned by the buffer
pool instead of the element, for the sake of isolation
George Kiagiadakis [Tue, 14 May 2013 16:35:33 +0000 (19:35 +0300)]
waylandsink: Use XDG_RUNTIME_DIR instead of /tmp for the shm file
George Kiagiadakis [Wed, 12 Feb 2014 13:15:52 +0000 (14:15 +0100)]
waylandsink: remove unused variables
George Kiagiadakis [Wed, 12 Feb 2014 10:28:40 +0000 (11:28 +0100)]
waylandsink: process display events in a separate thread
This also moves the display-related code into a new GstWlDisplay class,
which takes care of the new thread
George Kiagiadakis [Tue, 4 Feb 2014 15:32:31 +0000 (16:32 +0100)]
waylandsink: tidy up the header files
Sanjay NM [Thu, 12 Jun 2014 11:08:35 +0000 (16:38 +0530)]
pnmdec: use GstVideoDecoder Class
https://bugzilla.gnome.org/show_bug.cgi?id=731400
Edward Hervey [Mon, 16 Jun 2014 05:42:37 +0000 (07:42 +0200)]
tsdemux: Don't overwrite original seek value
In accurate mode, we'll be using start to fill in the seek segment,
therefore don't overwrite it with the shifted seek position.
https://bugzilla.gnome.org/show_bug.cgi?id=731698
Matthew Waters [Sun, 15 Jun 2014 05:18:46 +0000 (15:18 +1000)]
glvideomixer: silence incorrect number of arguments in format warning
Matthew Waters [Sun, 15 Jun 2014 05:17:07 +0000 (15:17 +1000)]
gldownload: plug a memory leak
Matthew Waters [Sun, 15 Jun 2014 03:59:07 +0000 (13:59 +1000)]
glvideomixer: wire up the alpha pad property
Matthew Waters [Sun, 15 Jun 2014 03:44:04 +0000 (13:44 +1000)]
glvideomixer: support input frame scaling
Matthew Waters [Sun, 15 Jun 2014 02:26:21 +0000 (12:26 +1000)]
glvideomixer: add positioning of input streams
https://bugzilla.gnome.org/show_bug.cgi?id=729798
Matthew Waters [Sun, 15 Jun 2014 02:24:38 +0000 (12:24 +1000)]
glmixer: allow the subclass to choose the sink pad type
Allows custom properties to be placed on the sink pads by subclasses
Edward Hervey [Fri, 13 Jun 2014 09:46:19 +0000 (11:46 +0200)]
mpegtspacketizer: live: handle pcr/dts discrepancies some more
When wrapover/reset occur, we end up with a small window of time where
the PTS/DTS will still be using the previous/next time-range.
In order not to return bogus values, return GST_CLOCK_TIME_NONE if the
PTS/DTS value to convert differs by more than 15s against the last seen
PCR
https://bugzilla.gnome.org/show_bug.cgi?id=674536
Edward Hervey [Fri, 13 Jun 2014 08:37:46 +0000 (10:37 +0200)]
mpegtspacketizer: Fix wraparound detection issue
Using 32bit unsigned values for corrected pcr/offset meant that we
potentially ended up in bogus values
Furthermore, refpcr - refpcroffset could end up being negative, which
PCRTIME_TO_GSTTIME() can't handle (and returned a massive positive value)
Matthew Waters [Fri, 13 Jun 2014 01:26:38 +0000 (11:26 +1000)]
docs: add gsteglimagememory
Matthew Waters [Thu, 12 Jun 2014 23:55:42 +0000 (09:55 +1000)]
docs: add glframebuffer object
Matthew Waters [Thu, 12 Jun 2014 23:52:01 +0000 (09:52 +1000)]
docs: misc adds for gl functions
Matthew Waters [Thu, 12 Jun 2014 23:41:51 +0000 (09:41 +1000)]
docs: add gluploadmeta object
Matthew Waters [Thu, 12 Jun 2014 23:32:56 +0000 (09:32 +1000)]
docs: update for some gl function name changes
Matthew Waters [Thu, 12 Jun 2014 13:14:45 +0000 (23:14 +1000)]
glfilterapp: actually emit the client-draw signal
Matthew Waters [Thu, 12 Jun 2014 13:13:11 +0000 (23:13 +1000)]
gl/examples: update for TEXTURE_2D change
Mathieu Duponchelle [Fri, 30 May 2014 14:52:09 +0000 (16:52 +0200)]
tsdemux: implement proper seeking for h264 streams.
Co-Authored by: Thibault Saunier <tsaunier@gnome.org>
From a high level perspective, the new process for seeking h264
streams is as follows:
1) Rewind the stream until we find the first I-slice of a frame,
and mark its offset in the stream.
2) Rewind the stream until we find SPS and PPS informations,
to make sure the subsequent parser is up to date.
3) Accumulate optionnal SEI NAL units on the way.
4) Push the SPS, PPS and SEI units before the new keyframe.
https://bugzilla.gnome.org/show_bug.cgi?id=675132
Vincent Penquerc'h [Thu, 12 Jun 2014 09:14:27 +0000 (10:14 +0100)]
dvbsrc: do not try to close negative file descriptors
Interestingly, Coverity implies that close takes an unsigned
argument, while my close(2) man page shows it taking a signed
argument. I guess it may be platforms specific.
Coverity 1214602
Tim-Philipp Müller [Thu, 12 Jun 2014 09:17:10 +0000 (10:17 +0100)]
hls: fix build with GLib 2.32
Provide internal copy of g_list_copy_deep() until we
bump the GLib requirement.
https://bugzilla.gnome.org/show_bug.cgi?id=731555
Vincent Penquerc'h [Thu, 12 Jun 2014 09:00:55 +0000 (10:00 +0100)]
aiffparse: bail out early when bytes per sample is not set
This should always be set for valid files when we get there,
and checking this avoids having ad hoc checks further down
in several places.
Coverity 1139698
Gwenole Beauchesne [Wed, 11 Jun 2014 13:11:40 +0000 (15:11 +0200)]
codecparsers: h264: fix quantization matrix conversion routine names.
Fix routine names for zigzag/raster scan order conversion routines for
quantization matrices. This ought to use the gst_h264_quant_matrix_*()
naming convention instead of gst_h264_video_quant_matrix_*(), which
derived from the MPEG-2 function names.
https://bugzilla.gnome.org/show_bug.cgi?id=731524
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
Gwenole Beauchesne [Mon, 26 May 2014 16:29:21 +0000 (18:29 +0200)]
codecparsers: add C++ guards for MPEG-4 and VP8 parsers.
Fix MPEG-4 and VP8 APIs to export their external symbols as pure C
symbols, i.e. un-mangled for C++.
https://bugzilla.gnome.org/show_bug.cgi?id=731522
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
Matthew Waters [Thu, 12 Jun 2014 04:58:47 +0000 (14:58 +1000)]
glmixer: send the stream-start event
and the caps event after.
https://bugzilla.gnome.org/show_bug.cgi?id=730944
Matthew Waters [Thu, 12 Jun 2014 03:37:51 +0000 (13:37 +1000)]
gl/examples: add GST_PLUGINS_BASE_CFLAGS
Matthew Waters [Thu, 12 Jun 2014 03:15:10 +0000 (13:15 +1000)]
gl/examples: remove spurious include for doublecube example
Matthew Waters [Thu, 12 Jun 2014 03:06:31 +0000 (13:06 +1000)]
gl/examples: add the srcdir and builddir includes
And remove references to $(GST_PLUGINS_GL_*)
Matthew Waters [Thu, 12 Jun 2014 02:49:42 +0000 (12:49 +1000)]
glfilterapp: remove the reshape/draw properties
The reshape property was never used.
Replace the draw property with a signal.
Based on patch by Mathieu Duponchelle <mathieu.duponchelle@epitech.eu>
https://bugzilla.gnome.org/show_bug.cgi?id=704507
Matthew Waters [Thu, 12 Jun 2014 02:14:35 +0000 (12:14 +1000)]
glmixer: lock the size of mix->frames to the number of pads
Fixes a segfault with decodebin ! glmixer where the request pads on
both sides were being requested after the state change to PAUSED.
Also fixes dynamically adding and removing pads while glmixer is
in a state >= PAUSED.
Matthew Waters [Thu, 12 Jun 2014 02:09:56 +0000 (12:09 +1000)]
gl/examples: Port to using signals instead of properties
Based on patch by Mathieu Duponchelle <mathieu.duponchelle@epitech.eu>
https://bugzilla.gnome.org/show_bug.cgi?id=704507
Matthew Waters [Sun, 1 Jun 2014 05:02:52 +0000 (15:02 +1000)]
glimagesink: replace pointer properties with signals
Based on patch by Mathieu Duponchelle <mathieu.duponchelle@epitech.eu>
https://bugzilla.gnome.org/show_bug.cgi?id=704507
Edward Hervey [Wed, 11 Jun 2014 07:38:00 +0000 (09:38 +0200)]
mpegtspacketizer: Use first group for handling reference pcr
If _set_current_pcr_offset gets called after a flushing seek, we ended
up using the current group for delta calculation ... whereas we should
be using the first group to calculate shifts.
Also add an early exit if there are no changes to apply
Edward Hervey [Tue, 10 Jun 2014 12:50:10 +0000 (14:50 +0200)]
tsdemux: Remove unused variable
The seeksegment was no longer used since the switch to calculating segments
when we see data.
Edward Hervey [Tue, 10 Jun 2014 12:27:53 +0000 (14:27 +0200)]
mpegtspacketizer: Fix debug statement
Edward Hervey [Tue, 10 Jun 2014 12:27:01 +0000 (14:27 +0200)]
mpegtspacketizer: Fix duration evaluation in push mode
When working in push mode, we need to be able to evaluate the duration
based on a single group of observations.
To do that we use the current group values
Edward Hervey [Mon, 2 Jun 2014 09:07:01 +0000 (11:07 +0200)]
mpegtspacketizer: Handle early PTS conversion
When handling the PTS/DTS conversion in new groups, there's a possibility
that the PTS might be smaller than the first PCR value observed, due to
re-ordering.
When using the current group, only apply the wraparound correction when we
are certain it is one (i.e. differs by more than a second) and not when it's
just a small difference (like out-of-order PTS).
https://bugzilla.gnome.org/show_bug.cgi?id=731088
Tim-Philipp Müller [Mon, 9 Jun 2014 09:10:01 +0000 (10:10 +0100)]
gdp: don't print already-freed string
CID 1221987
Edward Hervey [Mon, 9 Jun 2014 08:56:48 +0000 (10:56 +0200)]
speed: Fix position query some more
Forgot to remove code in
16fd917632c0992349ca28cf9108e9562c493ad7
CID #1139677
Edward Hervey [Sat, 7 Jun 2014 05:13:59 +0000 (07:13 +0200)]
mpegtsdemux: Fix sticky event handling
When we receive sticky events from upstream, always return TRUE.
Fixes the issue where we receive custom sticky events (such as "uri")
and no pads are created yet.
Damian Ziobro [Wed, 4 Jun 2014 17:34:44 +0000 (18:34 +0100)]
hlsdemux: Improve parsing quoted key URIs and apply it for I-frame-based stream URI
https://bugzilla.gnome.org/show_bug.cgi?id=730830
Sebastian Dröge [Fri, 6 Jun 2014 10:08:04 +0000 (13:08 +0300)]
hlsdemux: Don't store the current key in the playlist
It's per fragment and applying to all following fragments
until the next key is specified.
Sebastian Dröge [Fri, 6 Jun 2014 10:04:04 +0000 (13:04 +0300)]
hlsdemux: Fix compiler warnings
Thomas Bluemel [Fri, 30 May 2014 22:34:18 +0000 (16:34 -0600)]
hlsdemux: Reload the variant playlist if refreshing a playlist or downloading a fragment fails
This can happen if the playlists have moved due to the variant playlist
now being redirected to another target. This currently only works as long
as the referenced playlists don't change in relation to the variant
playlist, and the new location is purely due to a new path triggered by a
new redirection target of the variant playlist, or a new redirection
target of the playlist itself.
https://bugzilla.gnome.org/show_bug.cgi?id=731164
Olivier Crête [Thu, 5 Jun 2014 20:45:12 +0000 (16:45 -0400)]
schroenc: Use an arbitrary framerate if none if given
The schro encoder always requires a framerate, but some source material,
such as RTP doesn't have a fixed one, so just fake it.
https://bugzilla.gnome.org/show_bug.cgi?id=729054
Reynaldo H. Verdejo Pinochet [Wed, 4 Jun 2014 18:35:51 +0000 (14:35 -0400)]
dvbsrc: drop redundant assignment
Reynaldo H. Verdejo Pinochet [Wed, 4 Jun 2014 05:02:20 +0000 (01:02 -0400)]
dvbsrc: improve timeout handing at locking loop
New approach attempts to be more accurate by measuring
the elapsed time by iteration. Also:
* Use a 10 seconds default timeout and a half a second
polling step. New values should better match the tuning
process on real-life scenarios.
* Correct elapsed_time computation.
* Add _retry_ioctl() to avoid bailing out on temporary
ioctl EINTR failures (no need to check for EAGAIN cause
we are opening the frontend on blocking mode)
* Small corrections to fail condition handling
Matthew Waters [Thu, 5 Jun 2014 08:43:30 +0000 (18:43 +1000)]
gl/wayland: create a dummy display instead of an EGLDisplay
Currently, GstGLWindowWaylandEGL holds the wayland display connection
If we create the EGLDisplay at the GstDisplay creation time, then
libEGL will internally open another connection to the wayland server.
These two display connections are unable to communicate resulting in
no window output/display and hangs inside libEGL.
Eventually we will move the wl_display from GstGLWindow to GstGLDisplay.
Edward Hervey [Thu, 5 Jun 2014 07:40:38 +0000 (09:40 +0200)]
rtmp: proxy logging from librtmp
Helps with debugging various librtmp issues
Edward Hervey [Thu, 5 Jun 2014 06:29:50 +0000 (08:29 +0200)]
rtmpsrc: Fix position querying
It's the position we're querying, not the duration :)
Jan Schmidt [Wed, 4 Jun 2014 11:39:48 +0000 (21:39 +1000)]
tsdemux: Remove extraneous extra parameter from check_pending_buffers function
Jan Schmidt [Wed, 4 Jun 2014 11:31:21 +0000 (21:31 +1000)]
mpegtspacketiser: Always initialise packet AFC flags and PCR fields
Also fix a slight typo in a comment about the smoothing algorithm
Jan Schmidt [Fri, 23 May 2014 05:07:19 +0000 (15:07 +1000)]
tsdemux: Reset pending_ts on a discont, to re-measure initial timestamps
Since all the other timestamp tracking now gets reset on a discont,
it makes sense to wait for a PCR and timestamp buffers like when
playback first starts
Aleix Conchillo Flaqué [Sat, 22 Mar 2014 05:16:41 +0000 (22:16 -0700)]
srtp: add support for rollover counters and replay protection window size
We add a new signal, get-rollover-counter, to the SRTP encoder. Given a
ssrc the signal will return the currently internal SRTP rollover counter
for the given stream.
For the SRTP decoder we have a new SRTP caps parameter "roc" that needs
to be set when a new SRTP stream is created for a given SSRC.
https://bugzilla.gnome.org/show_bug.cgi?id=726861
Michael Olbrich [Mon, 14 Oct 2013 14:14:42 +0000 (16:14 +0200)]
liveadder: handle gap buffer flag
Remove GAP flag unless both sides have gap data, ignore
incoming data if it is gap data.
https://bugzilla.gnome.org/show_bug.cgi?id=722397
Matthieu Bouron [Sun, 25 May 2014 20:43:22 +0000 (21:43 +0100)]
glimagesink: improve state change debug message
Matthew Waters [Tue, 3 Jun 2014 03:59:51 +0000 (13:59 +1000)]
glupload: always release the previous buffer
Fixes the case where _perform_with_buffer() is called without
intervening calls to _release_buffer() as is the case on start up
with glimagesink.
Also release the buffer when reseting the upload.
https://bugzilla.gnome.org/show_bug.cgi?id=731107
Olivier Crête [Tue, 3 Jun 2014 01:51:38 +0000 (21:51 -0400)]
tests: Take account of memory alignment in shm test
Olivier Crête [Tue, 3 Jun 2014 01:43:56 +0000 (21:43 -0400)]
gdp: Fail gracefully if event can't be parsed
https://bugzilla.gnome.org/show_bug.cgi?id=731093
Olivier Crête [Tue, 3 Jun 2014 01:43:34 +0000 (21:43 -0400)]
shmsink: Allocate enough memory to do alignment
https://bugzilla.gnome.org/show_bug.cgi?id=731093