Tensor can be row or col major, but it's also possible that the order by we need
to read the tensor with more than two dimension need to be described. The
reserved field in GstTensorDim is there for this purpose. If we need this we
can add GST_TENSOR_DIM_ORDER_INDEXED, and follow an index defining order for
each dimension.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6000>
GstTensor contained two fields (data, dims) that were dynamicallay allocated. For
data it's for a GstBuffer and we have pool for efficient memory management. For
dims it's a small array to store the dimension of the tensor. The dims field
can be allocated inplace by moving it at the end of the structure. This will
allow a better memory management when GstTensor is stored in an analytics meta
which will take advantage of the _clear interface for re-use.
- New api to allocate and free GstTensor
To continue to support use-cases where GstTensor is not stored in an
analytics-meta we provide gst_tensor_alloc, gst_tensor_alloc_n and
gst_tensor_free that will facilitate memory management.
- Make GstTensor a boxed type
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6000>
Only in LTC mode we introduce additional latency that is depending on only on a
property and not on the framerate, so waiting for the framerate is not necessary.
In all other modes no latency is introduced at all and the latency query can
simply be proxied.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7831>
The gst_dep.get_variable('libexecdir') may fail in some scenarios
(e.g. building a module alone inside an uninstalled devenv) and
it shouldn't really be reached in the first place if docs are
disabled via options.
Also to avoid confusing meson messages when cross-compiling or
doing a static build.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7818>
The calculated position was off. I'm not sure of the exact cause;
possibly because we're in AU-aligned byte-stream mode, which means
`transform` is true.
Replacing the math that calculates the NALU positions with code more
similar to what is already in use for `idr_pos` seems to have fixed it.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7318>
Add missing pixel format constants, and mappings for
P010, packed variants of 420 and RGBA layouts to GStreamer
buffer formats. This fixes problems with android decoders
refusing to output raw video frames with decoders that
announce support for these common pixel formats and
only allowing the 'hardware surfaces output' path.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7743>
Some file format standards don't require mastering-display-info
and content-light-level values to be provided.
Decklink however requires the static HDR metdata for the PQ transfer
function which we may not have.
CTA-861-G mentions that in this case, 0 may provided as an 'unknown'
value which is what we use here.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7742>
* Consistently name parse functions according to their message type and
deprecate the misnamed ones,
* Add missing parse functions,
* Check for the correct message type when parsing
* Use correct field name for warning message details
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7754>
The diff between compared timestamps might be outside the gint range
resulting in wrong sorting results. This patch corrects that by
comparing the timestamps and then returning -1, 0 or 1 depending on the
result.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7726>
The wraparound handling code assumes that the PCR gets updated regularly for
being able to detect wraparounds. With ignore-pcr=true that was not the case and
it stayed initialized at 1h forever.
To avoid this problem, update the fake PCR whenever the PTS advanced by more
than 5s, and also detect wraparounds in these fake PCRs.
Problem can be reproduced with
$ gst-launch-1.0 videotestsrc pattern=black ! video/x-raw,framerate=1/5 ! \
x264enc speed-preset=ultrafast tune=zerolatency ! mpegtsmux ! \
tsdemux ignore-pcr=true ! fakesink
which restarts timestamps at 0 after around 26h30m.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7588>
When the stream resolution change it is needed to negotiate
a new pools and to update the caps.
Resolution change could occurs on a new sequence or a new
picture so move resolution change detection code in a common
function.
For memory allocation reasons, only allows resolution change
on non keyframe if the driver support remove buffer feature.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7684>
We must drain the pending output picture so that subclass can renegotiate
the caps. Not doing so while still renegotiating would mean that the
subclass would have to do an allocation query before pushing the caps.
Pushing the caps now without this would also not work since these caps
won't match the pending buffers format.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7684>
Add helpers function to call VIDIOC_REMOVE_BUFS ioctl.
If the driver support this feature buffers are removed from the queue when:
- the pool when is detached from the decoded.
- the pool is released.
- allocation failed.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7684>
Use VIDIOC_CREATE_BUFS ioctl to create buffers instead of VIDIOC_REQBUFS
because it allows to create buffers also while streaming.
To prepare the introduction of VIDIOC_REMOVE_BUFFERS create
the buffers one per one instead of a range of them. This way
it can, in the futur, fill the holes.
gst_v4l2_decoder_request_buffers() is stil used to remove all
the buffers of the queue.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7684>
When a datachannel within a session is removed after proper close,
reference to the error_ignore_bin elements of the datachannel
appsrc/appsink were left in webrtcbin.
This caused the bin-objects to be left and not freed until the whole
webrtc session was terminated. Among other things that includes a thread
from the appsrc.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7675>
- Add mtd_meta_clear to allow specific analytics-meta to handle their clear
operation specific to their type.
- Clear mtd's attached when analytic-meta is freed. When the buffer where
analytics-meta is attached is not from a buffer pool
gst_analytics_relation_meta_clear will not be called unless we explicitly call
it in _free. This important otherwise _mtd_clear are not called and lead to
leak if embedded mtd's allocated memory
- Un-ref in transform if it's a copy
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6026>
In my tests with the new GCC 14 compiler for Cerbero, I got the
following error:
> In file included from include/directxmath/DirectXMath.h:2275,
> from ../gst-libs/gst/d3d11/gstd3d11converter.cpp:46:
> include/directxmath/DirectXMathMatrix.inl: In function 'bool
> DirectX::XMMatrixDecompose(XMVECTOR*, XMVECTOR*, XMVECTOR*, FXMMATRIX)':
> include/directxmath/DirectXMathMatrix.inl:1161:16:
> error: variable 'aa' set but not used [-Werror=unused-but-set-variable]
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7658>
Check and generate remote reception statistics from the info stored on
internal sources, as they are stored there when running against newer rtpbin
since MR !7424
This fixes cases where statistics are incomplete when
peers send RR reports from a single remote ssrc, which GStreamer does
when bundling is enabled and other RTP stacks may too.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7425>
In some cases, decodebin3 will send us incomplete caps (not containing
codec_data), and then a GAP event, which will force a negotiation.
This segfaults due to a null pointer deref because self->input_state
is NULL.
The only possible fix is to avoid negotiating when we get incomplete
caps (to avoid re-negotiationg immediately afterwards, which isn't
supported by some muxers), but also set as much input state as
possible so that a renegotiation triggered by a GAP event can complete
successfully.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7634>
This new LCEVC encoder plugin is meant to implement all LCEVC encoder elements.
For now, it only implements the LCEVC H264 encoder (lcevch264enc) element. This
element essentially encodes raw video frames using a specific EIL plugin, and
outputs H264 frames with LCEVC data. Depending on the encoder properties, the
LCEVC data can be either part of the video stream as SEI NAL Units, or attached
to buffers as GstMeta.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7330>
This new element wraps both the base H264 decoder and lcevcdec elements into a
bin so that LCEVC decoding works with auto-plugging elements such as decodebin.
By default, the H264 decoder element with higher rank is used as base decoder,
but any particular H264 decoder can be used by manually setting the base-decoder
property.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7330>
This new LCEVC decoder plugin is meant to implement all LCEVC decoder elements.
For now, it only implements the LCEVC enhancement decoder (lcevcdec) element.
This element essentially enhances raw video frames using the LCEVC metadata
attached to input buffers into a higher resolution frame. The element is only
meant to be used after any base decoder (eg avdec_h264).
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7330>
There was an override to fake an IDR as soon as a SPS/PPS
is encountered, but that's not valid, at least an i-slice is needed.
Amend the visl result, as the output is slightly more correct, not
duplicating frame_num.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7330>
This improves the h264parse element to attach LCEVC enhancement data to buffers
using the new GstLcevcMeta API. This metadata will eventually be used downstream
by LCEVC decoders to enhance the RAW video frame.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7330>
This new metadata API allows elements to attach LCEVC enhancement data to video
buffers. Usually, the video parser elements are charged to parse the LCEVC
enhancement data from SEI Nal units (Supplemental enhancement Information).
However, other elements such as demuxers can also use this API if the LCEVC
enhancement data of the video is stored in a separate stream in the container.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7330>
D3D12_HEAP_FLAG_CREATE_NOT_ZEROED flag was introduced as of
Windows 10 May 2020 Update, and older versions don't understand
the heap flag. Checks the feature support and enables the
D3D12_HEAP_FLAG_CREATE_NOT_ZEROED only if it's supported by OS
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7573>
The uvcsink was limited to only transfer YUY2 and MJPEG. For the
uncompressed formats there is no technical reason not to support them.
Since gst_video_format_to_string is already supporting more fourcc than
only YUY2 we use the default path in gst_v4l2uvc_fourcc_to_bare_struct
to create structures for more formats and bail out if the returned
format is not from the uncompressed type.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6037>
In fact, the va decoder is just a internal helper class and its access
is under the control of all dec elements. So far, there is no parallel
operation on it now.
At the other side, some code scan tools report race condition issues.
For example, the "context" field is just protected with lock at _open()
but is not protected at _add_param_buffer().
So we just delete all its lock usage.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7547>
Be smarter when allocating sink and source memory pools to reduce the
memory footprint. Use gst_v4l2_decoder_get_render_delay() to know the
need number of buffers for downstream element.
Handle errors in case of memory allocation failures.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7544>
By setting the earliest time to timestamp + 2 * diff there would be a difference
of 1 * diff between the current clock time and the earliest time the element
would let through in the future. If e.g. a frame is arriving 30s late at the
sink, then not just all frames up to that point would be dropped but also 30s of
frames after the current clock time.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7459>
This makes sure that if upstream has different latencies that we're still
outputting buffers with increasining timestamps across the different streams
unless buffers are arriving after the latency deadline.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7500>
Adding prefer-stream-ordered-alloc property to GstCudaContext.
If stream ordered allocation buffer pool option is not configured
and this property is enabled, buffer pool will enable the stream
ordered allocation. Otherwise it will follow default behavior.
If GST_CUDA_ENABLE_STREAM_ORDERED_ALLOC env is set,
default behavior is enabling the stream ordered allocation.
Otherwise sync alloc/free method will be used.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7427>
Default CUDA memory allocation will cause implicit global
synchronization. This stream ordered allocation can avoid it
since memory allocation and free operations are asynchronous
and executed in the associated cuda stream context
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7427>
While transforming the internals of waylandsink into a library, the
context type name was accidentally changed, causing an ABI break. Change
it back to its original (as used by the libgstgl), and add support for
the misnamed version as a backward compatibility measure.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7482>
If the UVC gadget announces multiple formats in the descriptors the uvcsink
doesn't select the actual format but let's the UVC hosts select the format.
If the GStreamer pipeline is started before a UVC host selected the format,
upstream decides on a format until the UVC host has decided. In this case, the
current format needs to be set based on the caps from the caps event to be able
to detect if the format selection by the UVC host requires a format change on
the GStreamer pipeline.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7473>
The uvcsink may be put into the READY state to start listening for UVC requests.
Therefore, the UVC host may set a streaming format before the GStreamer pipeline
is started and the uvcsink received a caps event. In this case, prev_caps will
be NULL.
If the EVENT_CAPS has not been received, skip the check if the format needs to
be changed, since the sink will be started with the format selected by the UVC
host, anyway.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7473>
Adding a property to control the number of in-flight GPU commands
(default is unlimited). Note that actual maximum number is defined
in d3d12device's direct command queue object which is 32 now,
thus total number of scheduled GPU commands cannot exceed 32.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7444>
Sometimes under certain loads, VT can error out with kVTVideoEncoderMalfunctionErr or kVTVideoEncoderNotAvailableNowErr.
These have been reported to happen more often than usual if CopyProperty/SetProperty() is used close to the encode call.
Both can be worked around by restarting the encoding session.
These errors can be returned either directly from VTCompressionSessionEncodeFrame() or later in the encoding callback.
This patch handles both scenarios the same way - a session restart is be attempted on the next encode_frame() call.
If the error is returned immediately by the encode call, it's possible that some correct frames will still be given to
the output callback, but for simplicity (+ because I wasn't able to verify this scenario) let's just discard those.
In addition, this commit also simplifies the beach/drop logic in enqueue_buffer.
Related bug reports in other projects:
http://www.openradar.me/45889262https://github.com/aws/amazon-chime-sdk-ios/issues/170#issuecomment-741908622
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7173>
If state is changing from playing to paused, and rate is reset to 1
which causes seek position is valid, current code will do seek for
streams that are not seekable. So need to check whether stream is
seekable before seeking.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7441>
If the pending remote description has an invalid BUNDLE group _parse_bundle()
triggers early return from _create_answer_task(), before ret has been
initialized, so it needs to be checked before attempting to call
gst_sdp_message_copy().
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7423>
