CPU waits are more expensive and are only required if the CPU is ever going to
access the data. GPU waits perform inter-context synchronisation and are cheaper
as they don't require CPU intervention.
Adaptive demuxers need to start downloading from specific positions
(fragments) for every stream, this means that all streams can snap-seek
to a different position when requested. Snap seeking in this case will
be done in 2 steps:
1) do the snap seeking on the pad that received the seek event and
get the final position
2) use this position to do a regular seek on the other streams to
make sure they all start from the same position
More arguments were added to the stream_seek function, allowing better control
of how seeking is done. Knowing the flags and the playback direction allows
subclasses to handle snap-seeking.
And also adds a new return parameter to inform of the final
selected seeking position that is used to align the other streams.
https://bugzilla.gnome.org/show_bug.cgi?id=759158
FEC may only be used when PLC is enabled on the audio decoder,
as it relies on empty buffers to generate audio from the next
buffer. Hooking to the gap events doesn't work as the audio
decoder does not like more buffers output than it sends.
The length of data to generate using FEC from the next packet
is determined by rounding the gap duration to nearest. This
ensures that duration imprecision does not cause quantization
to 2.5 milliseconds less than available. Doing so causes the
Opus API to fail decoding. Such duration imprecision is common
in live cases.
The buffer to consider when determining the length of audio
to be decoded is the previous buffer when using FEC, and the
new buffer otherwise. In the FEC case, this means we determine
the amount of audio from the previous buffer, whether it was
missing or not (and get the data either from this buffer, or
the current one if the previous one was missing).
Was never ported and doesn't look like
we want it or need it in this form, can
do the same with the libgstvideo sample
conversion utility API now, but better
and in a more flexible way.
Just check whether LATM is defined which is only available
in 2.7 and later. Allows us to simplify the configure check
a little and we can get rid of some hackish workarounds for
problems with earlier version headers.
On some embedded systems, sqrt() is not supported in the shader,
use the actual value of sqrt(2) instead.
Signed-off-by: Haihua Hu <b55597@freescale.com>
Bugzilla: https://bugzilla.gnome.org/show_bug.cgi?id=761271
Allows the subclass to completely override the chosen src caps.
This is needed as videoaggregator generally has no idea exactly
what operation is being performed.
- Adds a fixate_caps vfunc for fixation
- Merges gst_video_aggregator_update_converters() into
gst_videoaggregator_update_src_caps() as we need some of its info
for proper caps handling.
- Pass the downstream caps to the update_caps vfunc
https://bugzilla.gnome.org/show_bug.cgi?id=756207
1. Otherwise rotating the video will clip and show black bars due to
gltransformation's implementation.
2. The other option of make gltransformation aspect-agnostic produces
incorrect output with perspective transformations.
Add a "mask" property that sets whether the edges by cvLaplace should be
used as a mask on the original input or not. The same way the original
image is copied to the edges in edgedetect.
Add a "mask" property that sets whether the detected derivative edges
should be used as a mask on the original input or not. The same way
the original image is added to the edges in edgedetect.
cvCvtPixToPlane() has been deprecated in OpenCV 3.0, and cvSplit() is the
suggested replacement. Since cvSplit() is available in OpenCV 2.4, it is
safe and cautious to update the function usage before it becomes an issue.
cvlaplace was also affected by the silent change in OpenCV API, same as
cvsobel. It hasn't been working for a while. It would return a plain black
image. This commit updates the usage of cvLaplace by using cvCvtColor to
create the grayscale intermediate image to process. This also means there
is no need anymore to use GstBaseTransform's transform_caps, since the pads
are RGB.
cvsobel han't been working for a while due to a silent change in OpenCV
API. It would return a plain black image. This commit updates the usage
of cvSobel by using cvCvtColor to create the grayscale image to process.
This also means there is no need to use GstBaseTransform's transform_caps
anymore, since the pads can be RGB.
In file included from effects/gstgleffectrgbtocurve.c:25:0:
effects/gstgleffectscurves.h:174:32: error: 'xray_curve' defined but not used
static const GstGLEffectsCurve xray_curve = {
...
For a live mpd, if availabilityStartTime is missing, adaptive demux asserts
with: Unexpected critical/warning: gst_date_time_to_g_date_time: assertion
'datetime != NULL' failed.
This patch improves the error message to:
Unexpected critical/warning: gst_mpd_client_seek_to_time: assertion
'client->mpd_node->availabilityStartTime != NULL' failed
https://bugzilla.gnome.org/show_bug.cgi?id=757602
Handling the ghostpad and its internal pad was causing more issues
than helping because of their coupled activation/deactivation
actions.
As we have to install custom chain,event and query functions it is
better to use a floating sink pad internally in the demuxer and just
use those pad functions to push through a standard pad in the demuxer
https://bugzilla.gnome.org/show_bug.cgi?id=757951
Reverses the transformation applied through the properties and forwards the
event.
The process for finding the coordinates on the video are as follows:
1. Convert the given pointer_x and pointer_y to model space at the near and far planes
2. Get the equation of the video plane
3. Find where the ray in 1 intersects the plane
4. Profit!
It performs the exact same operation as videobalance but with opengl shaders and
was tested with glvideomixer by comparing frames from videobalance and
glcolorbalance.
Allows more blending options than just A over B
e.g. frame comparisons are now possible.
glvideomixer name=m
sink_0::zorder=0
sink_1::zorder=1
sink_1::blend-equation-rgb={subtract,reverse-subtract}
sink_1::blend-function-src-rgb=src-color
sink_1::blend-function-dst-rgb=dst-color
! glimagesinkelement
videotestsrc pattern=checkers-4 ! m.sink_0
videotestsrc pattern=checkers-8 ! m.sink_1
SBC frame length calculation wasn't being rounded up to the nearest byte
(as specified in the A2DP 1.0 specification, section 12.9). This could
cause 'stereo' and 'joint stereo' mode SBC streams to have incorrectly
calculated frame lengths.
https://bugzilla.gnome.org/show_bug.cgi?id=742446
After commit 64080e632, configure checks for all the header files that
should be available in OpenCV 2.3 and later. If any of these files isn't
there the OpenCV elements won't be part of the build.
No need to recheck for opencv2/legacy/legacy.hpp again in
gstpyramidsegment.h. Minimum supported OpenCV version must have this header
and configure already checks for it. Removing check.
Run the transform function of pyramidsegment in place, reusing the image
data as both source and destination in cvPyrSegmentation. This avoids
copying the image back and forth and the extra memory.
Properly handle snap flags during reverse seeking. In this case
the before/after are also reversed, so handle those as such.
For example: with a sequence of 1s fragments:
|-- 0 --|-- 1 --|-- 2 --|-- 3 --|
If you seek to 1.5s it is inside fragment 1. With reverse and
snap-before: should play from the end of fragment 1
snap-after: should play from the end of fragment 0
Avoids an unlikely crash.
Arguably, if allocation fails we have no chance of
recovering but nonetheless, RTMP_Alloc can fail and
librtmp's RTMP_init() (called next) assumes a non-NULL
pointer is passed without checking.
Additionally, unify exit path on error.
Avoids an unlikely crash.
Arguably, if allocation fails we have no chance of
recovering but nonetheless, RTMP_Alloc can fail and
librtmp's RTMP_init() (called next) assumes a non-NULL
pointer is passed without checking.
Additionally, unify exit path on error.
All the gleffects shaders can be run against a gles2 or a legacy opengl glsl
compiler but weren't being advertised as such.
Fixes gleffects under desktop opengl < 3.2.
The property location has been changed in favor of vertex/fragment
string properties; the doc had not been updated and was still referring
to the previous property; also, now the #version header has become mandatory
https://bugzilla.gnome.org/show_bug.cgi?id=759902
No need to attempt splitting the RGB string in 255 tokens
if we only expect 3.
Left max_tokens at 4 to preserve the current logic (which
allows for extra stuff at the end) and added a warning on
parsing failure instead of silently discarding the value.
The URI attribute from the EXT-X-KEY tag and the URI attribute from the
EXT-X-I-FRAMES-ONLY tag are both quoted-string attibutes that have their
quotation marks removed during parsing. The CODECS attribute of the
EXT-X-STREAM-INF is also a quoted-string attribute, but this attribute
was not being un-quoted.
This commit changes the parser to always unquote all quoted-string
attributes and adjusts the unit tests to this new bevahiour for the
CODECS attribute.
An additional test is added to check that parsing of all of the fields
in the EXT-X-STREAM tag is correct, including those that contain comma
characters.
https://bugzilla.gnome.org/show_bug.cgi?id=758384
Clear error as soon as we determine that the download failed,
otherwise there are code paths where we might return without
clearing it ever, which would leak the GError then. Also, we
can pass a NULL GError pointer to _fetch_uri(), so just do that
instead of passing one that we're going to just free again
right away anyway.
Adds more meaningful error than
"Failed to convert multiview video buffer", which is always used
when prepare_next_buffer() fails in gst_glimage_sink_prepare().
https://bugzilla.gnome.org/show_bug.cgi?id=743345
Update opencvtextoverlay to inherit from GstOpencvVideoFilter instead of
from GstElement. This means less code and more uniformity with other OpenCV
elements. The chain/transform function is now a third of the size than
before.
Update pyramidsegment to inherit from GstOpencvVideoFilter instead of from
GstElement. This means less code and more uniformity with other OpenCV
elements.
Update pyramidsegment to inherit from GstOpencvVideoFilter instead of from
GstElement. This means less code and more uniformity with other OpenCV
elements.
Update motioncells to inherit from GstOpencvVideoFilter instead of from
GstElement. This means less code and more uniformity with other OpenCV
elements.
Add gst_gl_memory_allocator_get_default to get the default allocator based on
the opengl version. Allows us to stop hardcoding the PBO allocator which isn't
supported on gles2.
Fixes GL upload on iOS9 among other things.
Performing any GL function marshalling off the GL thread with glimagesink's
render lock is prone to deadlocks between the GL thread and the non-GL thread.
What can happen is this:
1. non-GL thread attempts to function marshal to the GL thread.
2. while 1 is happening, the winsys gives an event (say resize)
3. This calls back into glimagesink which taks the render lock.
4. As the GL function marshalling is attempting to run on the GL
and already has glimagesink's render lock locked. This deadlocks
as the threads are waiting for each other.
Update edgedetect to inherit from GstOpencvVideoFilter instead of from
GstElement. This means less code and more uniformity with other OpenCV
elements.
Adding the support for the two other OpenCV linear filters to smooth
images. The new API does support spatial sigma in the bilateral filter,
hence bringing that property back.
Adding reference to new documentation.
The OpenCV cvSmooth function is deprecated [0] and the documentation
recommends to use GaussianBlur (). This makes the spatial property go
unused. Marking it as deprecated, making it non-functional and will remove
in the next cycle.
[0] http://docs.opencv.org/2.4/modules/imgproc/doc/filtering.html
gst_mpdparser_parse_utctiming_node does not validate the parsed values completely. The following scenarios are incorrectly accepted:
- elements with no schemeIdUri property should be rejected
- elements with unrecognized UTCTiming scheme should be rejected
- elements with empty values should be rejected
The last one triggers a division by 0 in gst_dash_demux_poll_clock_drift:
clock_drift->selected_url = clock_drift->selected_url % g_strv_length (urls);
because it urls is a valid pointer to an empty array.
https://bugzilla.gnome.org/show_bug.cgi?id=759547
We don't directly link to GL in the element, though we use GL headers.
For this reason we need to include the proper GL headers path. This
prevent this element from using a different GL header then libgstgl.
rename gst-launch --> gst-launch-1.0
replace old elements with new elements(ffmpegcolorspace -> videoconvert, ffenc_** -> avenc_**)
fix caps in examples
https://bugzilla.gnome.org/show_bug.cgi?id=759432
The base class is useful for having multiple backing memory types other
than the default. e.g. IOSurface, EGLImage, dmabuf?
The PBO transfer logic is now inside GstGLMemoryPBO which uses GstGLBuffer
to manage the PBO memory.
This also moves the format utility functions into their own file.
The opencv element includes were full of duplicates and uneeded headers.
For example a few elements that stopped using gstcvopencvutils still
included that header file.
Since commit 45ca8876b2 nobody is using
gst_opencv_get_ipl_depth_and_channels() or
gst_opencv_parse_iplimage_params_from_structure(). Remove this dead
code.
Setting the seek flags to GST_SEEK_FLAG_SNAP_* will change the seek
target time to a segment boundary.
Based on original work by Ben Willers <bwillers@digisoft.tv>.
https://bugzilla.gnome.org/show_bug.cgi?id=759108
Dashdemux has set the width and height information from MPD manifest.
Some embedded devices which are not insufficient H/W resources need more information such as framerate
to assign H/W resources. So I suggested that dashdemux also needs to set the framerate information from MDP manifest.
https://bugzilla.gnome.org/show_bug.cgi?id=758515
As HLS does not provide any way of knowing the server's clock, and we do
buffering of "live" streams, at some point we will fall behind the server in
many cases and would have to advance to a fragment that is not in the playlist
anymore.
Previously we would've just resynced to the next oldest fragment that is still
there, but this causes problems as from this point onwards we would always
fall off the playlist again all the time.
Instead we now resync and move to the 3rd newest fragment like we would do
when starting playback of a live stream.
https://bugzilla.gnome.org/show_bug.cgi?id=758987
If connection speed is set, playlist according
to connection speed is selected as current playlist.
Problem is that the current variant of main playlist still
points to previously set variant.
If previously set variant doesn't correspond to current
playlist, then it causes unnecessary change of playlist
to the same playlist after first fragment is downloaded,
because of not updated current variant.
To fix this, we need to make sure that current variant
of main playlist corresponds to the current playlist
https://bugzilla.gnome.org/show_bug.cgi?id=758946
Don't jump backward to 3 files from the end of the playlist
when switching variants - it just means we downloaded
fragments fast and caught up to the end of the playlist.
Disable that by treating a variant switch as a playlist
update, not a restart due to a seek or so.
If the stream is discont, we must provide a timestamp in any case. Elements
like tsdemux are not going to output anything if we give a NONE timestamp
after a discont.
Also marking a stream as discont if a playlist change was not successful would
lead to the above situation, but in that case we are not required at all to
mark the stream discont as we're still at the old playlist.
commit da5c41930c removed the two uses of the
new value of data:
channels = opus_packet_get_nb_channels (data);
bandwidth = opus_packet_get_bandwidth (data);
Since then, data isn't being used between incrementing it by packet_offset
and going out of scope. Removing this uneeded statement.
The scene graph can be initialized when the we receive window handle change
notification and so we will not receive a scenegraph initialization
notification. Initialize ourself in this case.
https://bugzilla.gnome.org/show_bug.cgi?id=758337
