Nicolas Dufresne [Thu, 22 Oct 2020 16:38:11 +0000 (12:38 -0400)]
h265parse: Add missing const qualifier
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1750>
Seungha Yang [Mon, 2 Nov 2020 13:47:20 +0000 (22:47 +0900)]
Revert "d3d11decoder: Use D3D11/DXGI standard colorimetry"
This reverts commit
a52fc6deeda203add520cb59ae0026d109ecda95.
The change breaks H264/HEVC conformance bitstream tests
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1753>
Randy Li [Mon, 2 Nov 2020 08:46:25 +0000 (08:46 +0000)]
wlvideoformat: fix DMA format convertor
In the most of case, this typo would work. But for
ARGB8888 and XRGB8888, which shm format is not based on fourcc,
which would never appear in format enumeration.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1751>
Seungha Yang [Sat, 31 Oct 2020 18:58:30 +0000 (03:58 +0900)]
d3d11decoder: Use D3D11/DXGI standard colorimetry
D3D11/DXGI supports smaller set of colorimetry than all possible
combination. This restriction would make more streams convertible
by using ID3D11VideoProcessor
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1743>
Seungha Yang [Fri, 30 Oct 2020 18:28:55 +0000 (03:28 +0900)]
d3d11window: Use ID3D11VideoProcessor only if device supports corresponding conversion
... and drop support for ID3D11VideoProcessor if device doesn't
support ID3D11VideoContext1 interface and therefore we cannot
query conversion supportability.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1743>
Seungha Yang [Sun, 1 Nov 2020 11:52:11 +0000 (20:52 +0900)]
d3d11h{264,265}dec: Submit picture level parameters only once
Submit PICTURE_PARAMETERS and INVERSE_QUANTIZATION_MATRIX
buffers only once per picture. Multiple submission is redundant.
Also this modification would fix broken hevc decoding with
dependent slice.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1749>
Seungha Yang [Sat, 31 Oct 2020 11:36:13 +0000 (20:36 +0900)]
codecs: h265decocer: Rework for DPB management
* Move all DPB bumping process into GstH265Dpb internal
* Handle DPB add process in GstH265Dpb struct
* Make implementation to be 1:1 mappable with hevc specification
* Fix wrong DPB bumping implementation especially when no_output_of_prior_pics_flag
was specified.
With fixes from Nicolas Dufresne <nicolas.dufresne@collabora.com>
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1748>
Seungha Yang [Sat, 31 Oct 2020 11:31:51 +0000 (20:31 +0900)]
d3d11decoder: Get rid of framerate field from pad template
Framerate is optional value and we don't have any framerate
related restriction for those elements. This commit is to fix
negotiation failure when upstream doesn't set framerate on caps.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1747>
Jan Schmidt [Sat, 31 Oct 2020 10:46:16 +0000 (21:46 +1100)]
tests: Don't set dtlsenc state before linking.
Link the dtlsenc in the testsuite before setting it to paused, as it
starts a pad task that can generate a not-linked error otherwise.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1744>
Jan Schmidt [Fri, 30 Oct 2020 14:23:36 +0000 (01:23 +1100)]
dtls: Catch bus errors and fail instead of hanging.
If the DTLS elements fail, they post a bus error and don't signal any
key negotiation. Catch the bus error and fail the test early instead
of letting it hang and time out.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1741>
Jan Schmidt [Fri, 30 Oct 2020 11:52:18 +0000 (22:52 +1100)]
sctp: Do downward state change logic after chaining up.
Call the parent state_change function first when changing state
downward, to make sure that the element has stopped before cleaning
it up.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1741>
Jan Schmidt [Fri, 30 Oct 2020 11:49:22 +0000 (22:49 +1100)]
dtls: Avoid bio_buffer assertion on shutdown.
On shutdown, a previous iteration of dtsl_connection_process()
might be incomplete and leave a partial bio_buffer behind.
If the DTLS connection is already marked closed, drop out
of dtls_connection_process early without asserting.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1741>
Jan Schmidt [Fri, 30 Oct 2020 05:31:18 +0000 (16:31 +1100)]
webrtc: Fix a race on shutdown.
The main context can disappear in gst_webrtc_bin_enqueue_task()
between checking the is_closed flag and enqueueing a source on the
main context. Protect the main context with the object lock instead
of the PC lock, and hold a ref briefly to make sure it stays alive.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1741>
Olivier Crête [Wed, 8 Jul 2020 21:24:36 +0000 (17:24 -0400)]
webrtc: Set the DSCP markings based on the priority
This matches how the WebRTC javascript API works and the Chrome implementation.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1707>
Olivier Crête [Thu, 9 Jul 2020 17:39:03 +0000 (13:39 -0400)]
rtpsender: Add API to set the priority
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1707>
Olivier Crête [Thu, 9 Jul 2020 17:42:35 +0000 (13:42 -0400)]
rtptransceiver: Store the SSRC of the current stream
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1707>
Olivier Crête [Wed, 8 Jul 2020 23:13:33 +0000 (19:13 -0400)]
webrtc: Save the media kind in the transceiver
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1707>
Olivier Crête [Thu, 9 Jul 2020 17:45:20 +0000 (13:45 -0400)]
webrtcbin: Remove unused function
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1707>
Olivier Crête [Sat, 3 Oct 2020 01:38:00 +0000 (21:38 -0400)]
webrtc: Document more objects
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1707>
Seungha Yang [Fri, 30 Oct 2020 15:37:48 +0000 (00:37 +0900)]
d3d11decoder: Allow 10bits only profiles
HEVC_VLD_Main10 and VP9_VLD_10bit_Profile2 might not support
8bit format (i.e., NV12) depending on GPU vendor.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1742>
Sebastian Dröge [Sun, 25 Oct 2020 11:33:21 +0000 (13:33 +0200)]
decklink: Remove \n from debug output
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1721>
Sebastian Dröge [Sun, 25 Oct 2020 11:32:26 +0000 (13:32 +0200)]
decklink: Correctly indent everything
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1721>
Sebastian Dröge [Sun, 25 Oct 2020 11:30:55 +0000 (13:30 +0200)]
decklink: Add a default profile id
This causes no changes to the profile but keeps the existing settings.
The profile can also be changed from e.g. the card's configuration
application and in that case probably should be left alone.
The default is the new value as it keeps the profile setting as it is,
which is consistent with the previous behaviour in 1.18.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1721>
Sebastian Dröge [Sun, 25 Oct 2020 11:14:11 +0000 (13:14 +0200)]
decklink: Mark internal function as static
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1721>
Sebastian Dröge [Sun, 25 Oct 2020 11:13:37 +0000 (13:13 +0200)]
decklink: Remove some dead code
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1721>
Sebastian Dröge [Fri, 30 Oct 2020 08:02:32 +0000 (10:02 +0200)]
decklink: Update enum value bounds check in gst_decklink_get_mode()
The widescreen modes moved after GST_DECKLINK_MODE_2160p60 and using
them now would cause an assertion. This is a regression from
309f6187fef890c7ffa49305f38e89beac3b1423.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1737>
Chris Bass [Tue, 25 Aug 2020 13:56:50 +0000 (14:56 +0100)]
ttmlparse: Handle whitespace before XML declaration
When ttmlparse is in, e.g., an MPEG-DASH pipeline, there may be
whitespace between successive TTML documents in ttmlparse's accumulated
input. As libxml2 will fail to parse documents that have whitespace
before the opening XML declaration, ensure that any preceding whitespace
is not passed to libxml2.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1539>
Chris Bass [Tue, 25 Aug 2020 13:54:31 +0000 (14:54 +0100)]
ttmlparse: Ensure only single TTML doc parsed
The parser handles only one TTML file at a time, therefore if there are
multiple TTML documets in the input, parse only the first.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1539>
Xavier Claessens [Thu, 29 Oct 2020 17:43:16 +0000 (13:43 -0400)]
amc: Fix crash when encoding AVC
gstamcvideoenc.c calls gst_amc_avc_profile_to_string() with alternatives
set to NULL which causes a crash.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1732>
Guillaume Desmottes [Thu, 19 Mar 2020 14:07:47 +0000 (15:07 +0100)]
isac: add iSAC plugin
Wrapper on the iSAC reference encoder and decoder from webrtc,
see https://en.wikipedia.org/wiki/Internet_Speech_Audio_Codec
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1124>
Thibault Saunier [Wed, 28 Oct 2020 14:49:54 +0000 (11:49 -0300)]
transcodebin: Create the decodebin in _init
This way user can request pads right from the beginning
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1151>
Philippe Normand [Sun, 25 Oct 2020 18:04:05 +0000 (18:04 +0000)]
transcodebin: Accept more than one stream
Look-up the stream matching the given ID also after building the stream list
from the received collection. Without this change the transcoder would discard
the second incoming stream.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1151>
Thibault Saunier [Fri, 20 Mar 2020 12:27:48 +0000 (09:27 -0300)]
transcodebin: Port to decodebin3
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1151>
Thibault Saunier [Thu, 19 Mar 2020 12:35:57 +0000 (09:35 -0300)]
uritranscodebin: Move to using a urisourcebin for our source.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1151>
Thibault Saunier [Thu, 19 Mar 2020 12:34:54 +0000 (09:34 -0300)]
transcoder: Base sync transcoding variant on a GMainLoop
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1151>
Randy Li [Thu, 29 Oct 2020 06:13:05 +0000 (06:13 +0000)]
waylandsink: release frame callback when destroyed
We would use a frame callback from the surface to indicate
that last buffer is rendered, but when we destroy the surface
and that callback is not back yet, it may cause the wayland event
queue crash.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1729>
Seungha Yang [Wed, 28 Oct 2020 10:00:43 +0000 (19:00 +0900)]
rtmp2src: Set buffer timestamp on output buffer
This timestamp information would be useful for queue2 element
when calculating time level and also it makes buffering decision
more reliable.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1727>
Seungha Yang [Tue, 27 Oct 2020 15:47:49 +0000 (00:47 +0900)]
d3d11videoprocessor: Fix wrong input/output supportability check
The flag argument of ID3D11VideoProcessorEnumerator::CheckVideoProcessorFormat
method is output value, not input.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1726>
Seungha Yang [Sat, 24 Oct 2020 17:27:52 +0000 (02:27 +0900)]
nvcodec: nvsldec: Add support for CUDA memory
Add CUDA memory support. Note that zero copying is not supported yet
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1720>
Seungha Yang [Sun, 25 Oct 2020 20:15:33 +0000 (05:15 +0900)]
d3d11memory: Adjust log level for some spammy debug messages
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1723>
Seungha Yang [Sun, 25 Oct 2020 20:11:45 +0000 (05:11 +0900)]
d3d11convert: Use ID3D11VideoProcessor only if device supports colorspace
Check whether conversion with given combination of input/output
format and dxgi colorspace is supported or not by driver.
If not, we should use shader.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1723>
Seungha Yang [Sun, 25 Oct 2020 20:09:40 +0000 (05:09 +0900)]
d3d11videoprocessor: Add a method for device's conversion caps check
Add a wrapper method for
ID3D11VideoProcessorEnumerator1::CheckVideoProcessorFormatConversion.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1723>
Seungha Yang [Sun, 25 Oct 2020 20:04:55 +0000 (05:04 +0900)]
d3d11format: Map more colorimetry with dxgi colorspace
Map more logically identical set of GstVideoColorimetry formats
with dxgi color space.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1723>
He Junyan [Sun, 25 Oct 2020 15:13:46 +0000 (23:13 +0800)]
va: pool: Set the video_alignment after we get image info.
The set_format() of the allocator may change the stride of the
alloc_info. We should update the video_align.stride_align based
on it. Or, we get a warning in gst_video_meta_validate_alignment().
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1698>
He Junyan [Mon, 26 Oct 2020 03:50:59 +0000 (11:50 +0800)]
va: h264dec: Set the padding to VideoAlignment's right.
In our va implemenation, we just use image's info to map the buffer.
The padding info just plays a role as a place holder to expand the
allocation size in caps when decoding size is bigger than display
size. So the padding_right or padding_left does not change the result.
But we find if using padding_left, it is hard to meet the requirement
of gst_video_meta_validate_alignment(), when the video meta's stride
is different from the allocation width.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1698>
He Junyan [Sun, 25 Oct 2020 17:22:12 +0000 (01:22 +0800)]
va: basedec: No need to call base class' decide_allocation().
We have already done the jobs in gst_va_base_dec_decide_allocation()
and no need to call base class' decide_allocation() again. The base
class' decide_allocation() will set_format() again and let use do the
image/surface testing again, which is low performance and no needed.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1698>
He Junyan [Tue, 20 Oct 2020 06:31:22 +0000 (14:31 +0800)]
va: allocator: use _update_image_info() to set allocator parameters.
Use this standalone function to update the allocator info and make
all ensure_image() and mem_alloc() API clean.
We also change the default way of using image. We now set the non
derive manner as the default manner, and if it fails, then fallback
to the derived image manner.
On a lot of platforms, the derived image does not have caches, so the
read and write operations have very low performance.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1698>
He Junyan [Tue, 20 Oct 2020 06:09:35 +0000 (14:09 +0800)]
va: allocator: Add a helper function to update the image info.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1698>
He Junyan [Mon, 19 Oct 2020 15:32:44 +0000 (23:32 +0800)]
va: allocator: Decide the allocator's parameters when set_format().
Moving the parameters testing and setting from the allocator_alloc_full()
to the allocator_try(). The allocator_alloc_full() will be called every
time when we need to allocate a new memory. But all these parameters such
as the surface and the image format, rt_format, etc, are unchanged during
the whole allocator lifetime. Just setting them in set_format() is enough.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1698>
Nicola Murino [Sun, 18 Oct 2020 15:59:44 +0000 (17:59 +0200)]
opencv: allow compilation against 4.5.x
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1709>
Aaron Boxer [Tue, 24 Mar 2020 13:18:28 +0000 (09:18 -0400)]
jpeg2000parse: sub-sampling parse should take component into account
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1653>
Stéphane Cerveau [Tue, 21 Apr 2020 12:16:45 +0000 (14:16 +0200)]
jpeg2000parse: no pts interpolation with subframe.
The jpeg2000parser must not interpolate PTS with subframes.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1653>
Aaron Boxer [Mon, 13 Jan 2020 19:01:19 +0000 (14:01 -0500)]
jpeg2000parse: support frame and stripe alignment in caps
forward alignment and num-stripes caps properties
Use caps height when setting caps for subframe
We want downstream to use full frame height, not subframe height
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1653>
Sebastian Dröge [Sun, 25 Oct 2020 09:46:29 +0000 (11:46 +0200)]
decklink: Reset skip counters when starting the sources
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/378>
Vivia Nikolaidou [Thu, 10 May 2018 11:05:12 +0000 (14:05 +0300)]
decklink*src: Post a warning message on the bus about dropped frames
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/378>
Georg Lippitsch [Tue, 28 Nov 2017 12:44:18 +0000 (13:44 +0100)]
decklink*src: Aggregate dropped frame/packet logging
decklink*src currently prints a log entry for every dropped frame and
audio packet. That completely spams the logs.
This change aggregates information about dropped packets and only prints
a message once when dropping starts, and a summary when dropping ends.
Fixes https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/issues/705
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/378>
Seungha Yang [Sat, 24 Oct 2020 11:59:55 +0000 (20:59 +0900)]
d3d11memory: Protect view object with lock
Make resource allocation more thread-safe
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1718>
Seungha Yang [Fri, 23 Oct 2020 17:47:22 +0000 (02:47 +0900)]
d3d11convert: Add support for conversion using ID3D11VideoProcessor
Output texture of d3d11 decoder cannot have the bind flag
D3D11_BIND_SHADER_RESOURCE (meaning that it cannot be used for shader
input resource). So d3d11convert (and it's subclasses) was copying
texture into another internal texture to use d3d11 shader.
It's obviously overhead and we can avoid texture copy for
colorspace conversion or resizing via ID3D11VideoProcessor
as it supports decoder output texture.
This commit would be a visible optimization for d3d11 decoder with
d3d11compositor use case because we can avoid texture copy per frame.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1718>
Seungha Yang [Fri, 23 Oct 2020 17:33:29 +0000 (02:33 +0900)]
d3d11memory: Store ID3D11VideoProcessorOutputView object
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1718>
He Junyan [Fri, 23 Oct 2020 14:29:57 +0000 (22:29 +0800)]
codecs: h265decoder: Fix picture leaks because of reference set.
The last frame's reference set has no one to cleanup. We need to
clean all pictures in the stop() func.
We also add a helper function to cleanup all the pictures in the
reference picture set.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1717>
He Junyan [Fri, 23 Oct 2020 13:21:05 +0000 (21:21 +0800)]
codecs: h265decoder: Fix 3 ref array leaks in finalize.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1717>
Seungha Yang [Fri, 23 Oct 2020 07:59:00 +0000 (16:59 +0900)]
d3d11window: Reuse ID3D11VideoProcessorInputView if possible
GstMemory object could be disposed if GstBuffer is not allocated
by GstD3D11BufferPool such as via gst_buffer_copy() and/or
gst_buffer_make_writable(). So attaching qdata on GstMemory
object would cause unnecessary view alloc/free.
By using view pool which is implemented in GstD3D11Allocator,
we can avoid redundant view alloc/free.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1716>
Seungha Yang [Wed, 21 Oct 2020 07:28:11 +0000 (16:28 +0900)]
d3d11memory: Implement ID3D11VideoProcessorInputView pool
Similar to ID3D11VideoDecoderOutputView pool implementation
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1716>
Xavier Claessens [Sat, 7 Apr 2018 20:33:47 +0000 (16:33 -0400)]
Meson: Use pkg-config generator
Nicolas Dufresne [Fri, 31 Jul 2020 00:23:37 +0000 (20:23 -0400)]
h265decoder: Add support for l0/l1
Add support for reference list needed for VA-API and some V4L2 decoders.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1713>
Nicolas Dufresne [Tue, 28 Jul 2020 22:37:38 +0000 (18:37 -0400)]
h265decoder: Sync with the H264 implementation
This ensures that we get the last reference to picture being outputed,
avoiding GstBuffer structure copies and simplifying the buffer management.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1713>
Seungha Yang [Tue, 20 Oct 2020 08:31:17 +0000 (17:31 +0900)]
d3d11decoder: Directly access ID3D11VideoDecoderOutputView for decoding
Decoder output view is stored in GstD3D11Memory object instead of
wrapper struct now. So qdata is no more required.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1712>
Seungha Yang [Mon, 19 Oct 2020 16:59:35 +0000 (01:59 +0900)]
d3d11memory: Implement ID3D11VideoDecoderOutputView pool
Similar to texture-array pool, we can reuse decoder output view
since the life time of output view is identical to that of texture-array.
In this way, we can avoid frequent output view alloc/free.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1712>
Seungha Yang [Sun, 4 Oct 2020 14:39:05 +0000 (23:39 +0900)]
d3d11memory: Move to GArray to store texture-array status
The size D3D11_REQ_TEXTURE2D_ARRAY_AXIS_DIMENSION is 2048
which is too large in practice especially for a texture
of dpb
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1712>
Víctor Manuel Jáquez Leal [Mon, 12 Oct 2020 17:20:10 +0000 (19:20 +0200)]
va: Add VP9 decoder
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1702>
Philippe Normand [Fri, 16 Oct 2020 14:46:20 +0000 (15:46 +0100)]
wpe: Convert launch lines to markdown and move since tag
Seems like the examples don't appear in the generated docs because the Since tag
was badly positioned in the doc blurb.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1706>
Víctor Manuel Jáquez Leal [Fri, 16 Oct 2020 08:35:36 +0000 (10:35 +0200)]
codecs: vp9decoder: Add segmentation to picture.
VA-API needs AC and DC quant scales for both luma and chroma, and the loop
filter level for current frame, but these values are not available outside
the private GstVp9Parser structure. And these values may change from frame
to frame, so they are picture specific.
This patch add GstVp9Segmentation structure array to GstVp9Picture to expose
it to derived classes. This approach is safer than passing the parser at
picture handling flow.
Also, this patch, in order to solve Documentation CI, mark as private the
GstVp9Picture structure.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1700>
Víctor Manuel Jáquez Leal [Mon, 12 Oct 2020 09:07:47 +0000 (11:07 +0200)]
codecs: vp9decoder: Pass parser as new_sequence() parameter.
In order to know the chroma format, besides profile, subsampling_x and
subsampling_y are needed (Spec 7.2.2 Color config semantics). These values are
in GstVp9Parser but not in GstVp9Framehdr.
Also, bit_depth is available in parser but not frame header. Evenmore, those
values are copied to picture structure later.
In case of VA-API, to configure the pipeline, it is require to know the chroma
format and depth.
It is possible to know chroma and depth through caps coming from vp9parser, but
it requires string parsing. It would be less error prone to get these values
through the parser structure at new_sequence() virtual method.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1700>
Nicolas Dufresne [Wed, 23 Sep 2020 20:43:30 +0000 (16:43 -0400)]
rtpsrc: Cleanup on BYE, timeout or when pad is reused
In this patch, we enabled 'autoremove' feature of rtpbin and also call
'clear-ssrc' on the rtpssrcdemux element when a pad is being reused. This
ensure that the jitterbuffer is removed and no threads accumulates.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1575>
George Kiagiadakis [Fri, 4 Sep 2020 11:18:13 +0000 (14:18 +0300)]
rtpsrc: re-use the same src pad for streams that have the same payload type
Also use payload type when naming pads, this will make it easier to identify
pads and simplify the code.
Fixes #1395
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1575>
Seungha Yang [Tue, 2 Jun 2020 16:26:12 +0000 (01:26 +0900)]
d3d11: Introduce d3d11compositor element
Add new video composition element which is equivalent to compositor
and glvideomixer elements. When d3d11 decoder elements are used,
d3d11compositor can do efficient graphics memory handling
(zero copying or at least copying memory on GPU memory space).
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1323>
Seungha Yang [Wed, 5 Aug 2020 08:27:30 +0000 (17:27 +0900)]
d3d11shader: Allow drawing without shader resource view
... for the case that we are rendering on target without input texture.
For example, we might want to draw arbitrary shape on render target view
without shader resource view.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1323>
Seungha Yang [Sun, 2 Aug 2020 18:19:34 +0000 (03:19 +0900)]
d3d11convert: Add new subclasses for only color convert or resize
New d3d11colorconvert and d3d11scale elements will perform only
colorspace conversion and rescale, respectively. Those new elements
would be useful when only colorspace conversion or rescale is required
and the other part should be done by another elements.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1323>
Seungha Yang [Sat, 1 Aug 2020 11:16:52 +0000 (20:16 +0900)]
d3d11colorconverter: Allow setting alpha value to use
... used for reordering case for now. In other words, non-alpha formats
such as NV12 is not supported case yet.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1323>
Seungha Yang [Tue, 2 Jun 2020 16:20:41 +0000 (01:20 +0900)]
d3d11colorconverter: Add support conversion with blending
This is pre-work for d3d11compositor support
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1323>
Seungha Yang [Tue, 2 Jun 2020 15:59:15 +0000 (00:59 +0900)]
d3d11colorconverter: Add method to support updating destination rect
It's equivalent to GST_VIDEO_CONVERTER_OPT_DEST_* options of GstVideoConverter
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1323>
Seungha Yang [Tue, 2 Jun 2020 15:46:13 +0000 (00:46 +0900)]
d3d11: Clarify target rect to be updated
Rename internal methods to clarify which rect (i.e., input or output)
should be updated
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1323>
Julian Bouzas [Fri, 2 Oct 2020 14:02:38 +0000 (10:02 -0400)]
nvcodec: Report latency in decoder based on max-display-delay
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1633>
Julian Bouzas [Fri, 2 Oct 2020 13:22:34 +0000 (09:22 -0400)]
nvcodec: Add max-display-delay decoder property
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1633>
Julian Bouzas [Thu, 24 Sep 2020 15:25:33 +0000 (11:25 -0400)]
nvcodec: Fix compiler error if OpenGL is not enabled
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1633>
Julian Bouzas [Thu, 24 Sep 2020 14:33:58 +0000 (10:33 -0400)]
nvcodec: Add missing CUDAMemory src caps in h264 decoder
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1633>
Julian Bouzas [Wed, 23 Sep 2020 17:49:43 +0000 (13:49 -0400)]
nvcodec: Add missing CUDAMemory sink caps in h264 and h265 encoders
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1633>
Julian Bouzas [Tue, 22 Sep 2020 17:07:19 +0000 (13:07 -0400)]
nvcodec: Fix description of cudadownload element
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1633>
Seungha Yang [Sun, 11 Aug 2019 06:02:04 +0000 (15:02 +0900)]
tests: Add CUDA filter unit tests
Adding a test for buffer meta and colorspace conversion
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1633>
Seungha Yang [Wed, 16 Oct 2019 13:43:09 +0000 (22:43 +0900)]
nvcodec: Add CUDA video scale element
Add new element for video resizing using CUDA
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1633>
Seungha Yang [Wed, 16 Oct 2019 13:42:55 +0000 (22:42 +0900)]
nvcodec: Add CUDA video convert element
Add new element for colorspace conversion using CUDA.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1633>
Seungha Yang [Wed, 16 Oct 2019 13:42:39 +0000 (22:42 +0900)]
nvcodec: Add generic CUDA video convert object
Introducing generic video convert object similar to video-converter
but using CUDA.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1633>
Seungha Yang [Wed, 16 Oct 2019 13:42:24 +0000 (22:42 +0900)]
nvcodec: Add support runtime CUDA kernel source compilation
Add util functions for runtime CUDA kernel source compilation
using NVRTC library. Like other nvcodec dependent libraries,
NVRTC library will be loaded via g_module_open.
Note that the NVRTC library naming is not g_module_open friendly
on Windows.
(i.e., nvrtc64_{CUDA major version}{CUDA minor version}.dll).
So users can specify the dll name using GST_NVCODEC_NVRTC_LIBNAME
environment.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1633>
Seungha Yang [Wed, 16 Oct 2019 13:42:06 +0000 (22:42 +0900)]
nvcodec: Add CUDA upload/download elements with base class for CUDA filters
Similar to glupload/gldownload elements but for CUDA memory.
It will help transfer memory between system and nvidia GPU
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1633>
Seungha Yang [Wed, 16 Oct 2019 13:21:05 +0000 (22:21 +0900)]
nvcodec: Peer direct access support
If support direct access each other, use device to device memory copy
without staging host memory
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1633>
Seungha Yang [Fri, 30 Aug 2019 08:19:44 +0000 (17:19 +0900)]
cudacontext: Enable direct CUDA memory access over multiple GPUs
If each device context can access each other, enable peer access
for better interoperability.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1633>
Seungha Yang [Fri, 30 Aug 2019 04:57:15 +0000 (13:57 +0900)]
nvenc: Support CUDA buffer pool
When upstream support CUDA memory (only nvdec for now), we will create
CUDA buffer pool.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1633>
Seungha Yang [Fri, 30 Aug 2019 04:55:25 +0000 (13:55 +0900)]
nvdec: Support CUDA buffer pool
If downstream can accept CUDA memory caps feature (currently nvenc only),
always CUDA memory is preferred.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1633>
Seungha Yang [Mon, 19 Aug 2019 09:02:56 +0000 (18:02 +0900)]
nvcodec: Add CUDA specific memory and bufferpool
Introducing CUDA buffer pool with generic CUDA memory support.
Likewise GL memory, any elements which are able to access CUDA device
memory directly can map this CUDA memory without upload/download
overhead via the "GST_MAP_CUDA" map flag.
Also usual GstMemory map/unmap is also possible with internal staging memory.
For staging, CUDA Host allocated memory is used (see CuMemAllocHost API).
The memory is allowing system access but has lower overhead
during GPU upload/download than normal system memory.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1633>
Andrew Wesie [Fri, 16 Oct 2020 11:29:02 +0000 (12:29 +0100)]
codecparsers: h264parser: guard against ref_pic_markings overflow
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1703>