Sebastian Dröge [Mon, 21 Feb 2022 13:02:02 +0000 (15:02 +0200)]
gl: Add GST_GL_DEPRECATED
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1764>
Sebastian Dröge [Mon, 21 Feb 2022 12:17:11 +0000 (14:17 +0200)]
d3d11: Remove non-existing gst_d3d11_memory_get_type() API from header
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1764>
Dmitry Osipenko [Mon, 14 Feb 2022 20:30:45 +0000 (23:30 +0300)]
v4l2codecs: Support YUV420M format
YUV420M format of V4L essentially is I420 with separate memory planes.
It's used by NVIDIA Tegra V4L decoder. Support this format by v4l2codecs.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1727>
Matthew Waters [Mon, 21 Feb 2022 02:27:06 +0000 (13:27 +1100)]
ulpfecenc: slightly safer dispose impl
Technically dispose can be called more than once (even if gstelement is
not actually set up to do that) so need to protect against that.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1761>
Matthew Waters [Mon, 21 Feb 2022 02:24:07 +0000 (13:24 +1100)]
ulpfecenc: fix unmatched free() call
One must always match a g_slice_new with a g_slice_free and a g_new with
a g_free. This was not the case for the internal ctx struct.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1761>
Matthew Waters [Tue, 9 Nov 2021 06:37:24 +0000 (17:37 +1100)]
rtpulpfecenc: add some debug logging
Like, what configuration we are using or whether a fec packet is
generated.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1761>
Nirbheek Chauhan [Fri, 18 Feb 2022 09:53:13 +0000 (15:23 +0530)]
matroska-demux: Emit a warning when no codec data found
It is bad if an mkv file does not have codec data for the ProRes
variant, so we should emit a warning. ffmpeg does the same thing.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1739>
Sebastian Groß [Sun, 20 Feb 2022 13:05:05 +0000 (14:05 +0100)]
audiovisualizer: shader: Fix dframe out of bound write
shader_fade_and_move_horiz_out writes a complete stride at the end of its
dframe data
This led to SIGSEGV since the stride reached into sframe->map[0]->memory
which could not be umpapped later on.
This is due to `d` is increased twice. Once at the end of the upper loop
and at the start of the lower loop.
The corresponding dframe stride is therefore skipped.
Rewind `d` and start at the correct position.
Fixes https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/issues/1702
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1751>
Matthew Waters [Wed, 3 Nov 2021 06:05:07 +0000 (17:05 +1100)]
tests/clock: avoid a race cranking
Scenario:
- Source 1 requesting and waiting a clock id
- Source 2 requesting and waiting on a clock id
- Test attempting to crank both sources in the same GstHarness
gst_test_clock_crank() originally dropped locks between the retrieving
of the next clock id and advancing to the next clock id. This would
mean that both sources would race each other attempting to complete
their clock waits. Sometimes the operations would be performed in the
correct order, other times they would not and a FALSE return value would
be produced.
This would lead to an assertion in gst_harness_push_from_src() expecting
that all clock cranks to succeed.
Fix by ensuring that the clock wait produced is dealt with before
processing the next by not dropping the relevant locks after retrieving
the next clock id.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1299>
Seungha Yang [Sun, 20 Feb 2022 14:59:46 +0000 (23:59 +0900)]
d3d11screencapturesrc: Use staging buffer pool for performance
By using staging pool/buffer, we can avoid per frame
staging texture -> system memory copy.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1756>
Nicolas Dufresne [Sun, 20 Feb 2022 20:20:07 +0000 (15:20 -0500)]
doc: Add NV12_16L32S into the cache
Autogenerated by CI
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1567>
Nicolas Dufresne [Fri, 18 Feb 2022 20:25:17 +0000 (15:25 -0500)]
tests: video: Add a unit test for stride extrapolation
This is a minimal unit test the show that the stride extrapolation can work
with all pixel format we support. This minimal verify that the extrapolation
match the stride we set into GstVideoInfo with 320x240 for all the pixel
format we support. The tiles formats are skipped, since their stride is
set as two 16bit integers, and we also skip over palette planes.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1567>
Nicolas Dufresne [Wed, 26 Jan 2022 18:39:40 +0000 (13:39 -0500)]
v4l2codecs: Enable MediaTek MM21 (NV12_16L32S)
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1567>
Nicolas Dufresne [Wed, 19 Jan 2022 20:53:05 +0000 (15:53 -0500)]
v4l2codecs: Add support for multi planar buffers
V4L2 have some pixel formats that produces 2 dmabuf, this add
support for these format.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1567>
Nicolas Dufresne [Wed, 26 Jan 2022 15:06:50 +0000 (10:06 -0500)]
v4l2transform: Handle caps changes
As this element is single threaded, we only need to stop the objects to
allow changing the format again. Fixes assertion notably on shutdown and
on some other situation where the format may be set twice without
actually activating the element.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1567>
Nicolas Dufresne [Wed, 26 Jan 2022 14:55:09 +0000 (09:55 -0500)]
v4l2object: Avoid crash on early failure
This happens while an external error lead to an early shutdown.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1567>
Nicolas Dufresne [Tue, 25 Jan 2022 19:34:32 +0000 (14:34 -0500)]
video4linux2: Add MM21 support
This enables mtk-vcodec and MDP driver from mainline Linux kernel.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1567>
Nicolas Dufresne [Tue, 25 Jan 2022 19:08:47 +0000 (14:08 -0500)]
Port plugins to gst_video_format_info_extrapolate_stride()
This reduces code duplication and simplify addition of new
pixel formats into related plugins.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1567>
Nicolas Dufresne [Tue, 25 Jan 2022 19:04:13 +0000 (14:04 -0500)]
video: Add an helper to extrapolate strides
Many of the legacy APIs, specifically in the Linux Kernel, have a
single stride for the pictures. In this context, it is common
to extrapolate the other strides based on the selected pixel
format. Such function have been copy pasted from video4linux2
plugin into wayland, kms and v4l2codecs plugins.
This patch implements a generalized from of that function and
make it available to everyone through the video library.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1567>
Nicolas Dufresne [Fri, 3 Dec 2021 02:46:59 +0000 (21:46 -0500)]
video: Add NV12_16L32S aka Mediatek MM21 support
Unlike other simple tiled formats, the Mediatek HW use different tile size
per-plane. The tile size is scaled according to the subsampling. Effectively,
using the name 16L32S to represent linearly layout tiles of size 16x32 bytes
in the Y plane, and 16x16 in the UV plane. In order to make this specificity
discoverable, a new SUBTILES flags have been added.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1567>
Rouven Czerwinski [Wed, 2 Feb 2022 11:49:29 +0000 (12:49 +0100)]
gstv4l2tuner: return NULL if no norm set
If the video4linux device supports norms but has no norm set, norm is
returned as an uninitialized variable after the ioctl call, leading to
gst_v4l2_tuner_get_norm_by_std_id() returning a random norm from the
supported norms. Catch this case and instead return NULL to indicate
that no norm is setup.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1625>
Seungha Yang [Sat, 19 Feb 2022 21:36:45 +0000 (06:36 +0900)]
winscreencap: Remove dxgiscreencapsrc element
We have the d3d11screencapturesrc element in d3d11 plugin
which is obviously better than this element in terms of performance
and design, so we don't need to make people be confused by two separate elements.
Let's pick the better implementation and remove unnecessary one.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1750>
Jan Alexander Steffens (heftig) [Tue, 8 Feb 2022 14:23:17 +0000 (15:23 +0100)]
wpe: Clean up build script
Use feature.require to check for gstgl and exit early if 'wpe' is
disabled (don't even check for wpe-webkit-1.1).
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1668>
Seungha Yang [Fri, 18 Feb 2022 15:37:34 +0000 (00:37 +0900)]
directshow: Fix build warnings
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1744>
Seungha Yang [Fri, 18 Feb 2022 15:14:38 +0000 (00:14 +0900)]
directshow: Merge plugins into single directshow plugin
... and lower rank of dshowvideosink and dshowdeviceprovider
to GST_RANK_MARGINAL since we don't prefer this plugin
by default
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1744>
Seungha Yang [Fri, 18 Feb 2022 15:12:55 +0000 (00:12 +0900)]
directshow: Remove white spaces
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1744>
Martin Reboredo [Tue, 19 Oct 2021 19:10:06 +0000 (16:10 -0300)]
vulkanshaderspv: SPIRV based filter
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1197>
Seungha Yang [Fri, 18 Feb 2022 18:26:00 +0000 (03:26 +0900)]
scenarios: Update for videoaggregator rounding behavior change
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1747>
Seungha Yang [Fri, 18 Feb 2022 16:58:20 +0000 (01:58 +0900)]
videoaggregator: Use floor() to calculate current position
... instead of round(). Depending on framerate, calculated position
may not be clearly represented by using uint64, 30000/1001 for example.
Then the result of round() can be sliglhtly larger (1ns) than
buffer timestamp. And that will cause unnecessary frame delay.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1747>
Nicolas Dufresne [Fri, 18 Feb 2022 21:23:09 +0000 (16:23 -0500)]
v4l2codecs: vp9: Fix reset_frame_context parameter
It was assumed that the kernel parameters would match with the bitstream value
but instead the author when with another set of value. Surprisingly, this
makes no difference with the resulting fluster score.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1748>
Nicolas Dufresne [Fri, 18 Feb 2022 21:02:27 +0000 (16:02 -0500)]
v4l2codecs: vp9: Only fill compressed headers if needed
Fixes: 13944cf3ee871722 ("v4l2codecs: vp9: Make compressed hdr control optional")
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1748>
Tim Mooney [Sat, 15 Jan 2022 05:42:27 +0000 (23:42 -0600)]
v4l2: include <sys/ioccom.h> on Illumos
Needed for _IOR/_IORW
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1524>
Nicolas Dufresne [Tue, 2 Nov 2021 19:58:49 +0000 (15:58 -0400)]
doc: Fix doc comment for GstChildProxy
This removes warning like:
../subprojects/gstreamer/gst/gstchildproxy.h:57: Error: Gst: identifier not found on the first line:
* #GstChildProxyInterface::get_child_by_name:
^
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1297>
Nicolas Dufresne [Thu, 28 Oct 2021 21:09:34 +0000 (17:09 -0400)]
base: audioencoder: Keep serialize event behind buffers
If a serialized event arrives behind a buffer, it should not be send before
it. This fixes the pending event handling so that only early pending events,
the one that arrrived or was generated while the adapter was empty get send
before pushing buffer. All other events are not pushed after.
This issue lead the latency tracer to think our audio encoder did not have any
latency. This was testing with opusenc in a live pipeline.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1266>
Loic Le Page [Wed, 26 Jan 2022 20:06:45 +0000 (21:06 +0100)]
Add build files and fix build for gst-plugins-bad dshowsinkwrapper plugin
dshowsinkwrapper plugin is the old dshowvideosink plugin that has been
renamed to be compliant with the actuel plugin name in the source code.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1577>
Loic Le Page [Wed, 26 Jan 2022 20:03:10 +0000 (21:03 +0100)]
Add build files and fix build for gst-plugins-bad dshowsrcwrapper plugin
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1577>
Loic Le Page [Wed, 26 Jan 2022 19:58:12 +0000 (20:58 +0100)]
Add build files and fix build for gst-plugins-bad dshowdecwrapper plugin
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1577>
Loic Le Page [Wed, 26 Jan 2022 19:43:09 +0000 (20:43 +0100)]
Add directshow strmbase library source code from Microsoft repo
Original repo is here:
https://github.com/microsoft/Windows-classic-samples
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1577>
Loïc Le Page [Wed, 16 Feb 2022 13:53:12 +0000 (14:53 +0100)]
Move all directshow plugins to the same sub-folder
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1577>
Sebastian Wick [Thu, 17 Feb 2022 16:36:22 +0000 (17:36 +0100)]
matroska: default prores fourcc apcn
If there is no codec private data for prores it should default to Apple
ProRes 422 Standard Definition (apcn). Can be tested with
strobe_scientist.mkv from
https://developers.google.com/media/vp9/hdr-encoding
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1734>
Seungha Yang [Tue, 16 Nov 2021 08:35:25 +0000 (17:35 +0900)]
qtdemux: Do not send unnecessary GAP events
Each stream may have its own segment timeline
(i.g., different segment.start or segment.base)
depending on edit-list and composition-to-decode atom.
Make sure whether time position of a stream has been actually
far behind than that of current target stream.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1352>
Tim Mooney [Sat, 15 Jan 2022 06:04:05 +0000 (00:04 -0600)]
meson: check for libsocket and libnsl
If present, add '-lsocket' and '-lnsl' to network_deps.
ext/curl/meson.build: add network_deps to dependencies
gst/festival/meson.build: same
sys/shm/meson.build: same
Fixes linking issues on Illumos distros.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1525>
Haihao Xiang [Sat, 20 Nov 2021 02:47:24 +0000 (18:47 -0800)]
vaapioverlay: inline sinkpad scaling support
Often, users will need to scale inputs (e.g.
with vaapipostproc) before they are submitted
to the vaapioverlay. However, this results in
multiple VPP passes/operations in the pipeline
which creates unnecessary process overhead.
This change allows for inputs to be submitted
at original scale to vaapioverlay with per-sinkpad
scale dimensions specified so they can be scaled
and blended/composited in a single VPP pass/operation
to avoid the unnecessary process overhead.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1380>
U. Artie Eoff [Fri, 19 Nov 2021 22:46:34 +0000 (14:46 -0800)]
vaapioverlay: use vpp blend state iif using alpha
Don't set VAAPI vpp blend flags if alpha == 1.0,
i.e. fully opaque. This can avoid extra processing
overhead on some drivers that apply blending
unconditionally when flags are present, even if the
end result is the same without blend flags (i.e. all
opaque alpha channels).
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1380>
Seungha Yang [Mon, 14 Feb 2022 13:59:01 +0000 (22:59 +0900)]
d3d11: Add support for AYUV, AYUV64, and RGBA64_LE formats
Note that AYUV and AYUV64 formats will be used to expand format
support, especially some packed YUV formats (e.g., Y410, YUY2)
are common DXGI formats used for hardware decoder/encoder on Windows
but those formats cannot be used as a render target. We need to handle
them differently without pixel shader help, using compute shader
for example.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1699>
Seungha Yang [Sun, 13 Feb 2022 11:32:48 +0000 (20:32 +0900)]
d3d11: Reorganize format mapping table
... and handle GST <-> D3D11 format mapping in a single place to
make adding format easier
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1699>
Seungha Yang [Sat, 12 Feb 2022 19:36:39 +0000 (04:36 +0900)]
d3d11: Store device format in struct
Holding pointer to struct was unsafe approach because
the pointer to d3d11 format will be invalidated once d3d11 device
object is released
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1699>
Seungha Yang [Sun, 6 Feb 2022 13:54:42 +0000 (22:54 +0900)]
gstinfo,ptpclock,libcheck: Use GetCurrentProcessId() instead of getpid() on Windows
getpid() shouldn't be used in case of UWP. Use GetCurrentProcessId()
instead which provides exactly the same functionality and can be
used with UWP as well.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1006>
Seungha Yang [Sun, 6 Feb 2022 17:49:15 +0000 (02:49 +0900)]
check: Disable some tests if pipe() is unavailable for UWP build
pipe() and _pipe() are not allowed on UWP
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1006>
Seungha Yang [Fri, 1 Oct 2021 11:27:28 +0000 (20:27 +0900)]
meson: Do hard build error for some MSVC warnings
Handle various MSVC warnings as errors for development version.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1006>
Seungha Yang [Tue, 8 Feb 2022 14:30:00 +0000 (23:30 +0900)]
Remove some trailing white spaces
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1006>
Dmitry Osipenko [Mon, 14 Feb 2022 13:18:54 +0000 (16:18 +0300)]
v4l2codecs: h264: Correct scaling matrix ABI check
Scaling matrix V4L UAPI control not presents on NVIDIA Tegra, the default
matrix should be used in this case. Mark scaling matrix presence optional.
Fixes: 47bfa71530c ("v4l2codecs: h264: Improve ABI check ")
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1726>
Mengkejiergeli Ba [Fri, 21 Jan 2022 07:32:09 +0000 (15:32 +0800)]
msdkdec: Enable SFC scaling for AVC and HEVC
Decoder SFC only supports down-scaling at most to 1/8.
For example, given a 4K bistream, SFC can scale it to 1080p via:
"msdkh265dec ! video/x-raw,width=1920,height=1080 ! glimagesink"
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1444>
Mengkejiergeli Ba [Wed, 15 Dec 2021 03:49:40 +0000 (11:49 +0800)]
msdkdec: Enable SFC csc for AVC and HEVC
Decoder SFC will be triggered when default output format is not accept at
downstream. One use case below can work without using msdkvpp:
"! msdkh265dec ! "video/x-raw,format=BGRA" ! glimagesink",
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1444>
Mengkejiergeli Ba [Wed, 17 Nov 2021 09:05:09 +0000 (17:05 +0800)]
msdkdec: Add SFC codes
SFC refers to Scaler Format Converter, more details here:
https://01.org/sites/default/files/documentation/intel-gfx-prm-osrc-kbl-vol15-sfc.pdf
By attaching mfxExtDecVideoProcessing buffer to mfxVideoParam, decoder
can do SFC for csc and scaling.
MSDK will check if surface bitdepth equals to input when no SFC is
triggered, which means in the case of SFC, there is no need to update
surface bitdepth via gst_msdk_update_mfx_frame_info_from_mfx_video_param.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1444>
Sebastian Dröge [Wed, 16 Feb 2022 07:54:07 +0000 (09:54 +0200)]
subprojects: Switch gst-plugins-rs.wrap to the main branch
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1725>
Seungha Yang [Tue, 15 Feb 2022 17:23:58 +0000 (02:23 +0900)]
d3d11converter: Fix for missing GRAY conversion
Add missing Y410 -> GRAY and GRAY -> semi-planar conversion
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1724>
Seungha Yang [Tue, 15 Feb 2022 17:11:53 +0000 (02:11 +0900)]
d3d11converter: Don't use FIXME_OBJECT for non-GstObject
... and print ERROR messages for unexpected input/output formats
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1724>
Sebastian Dröge [Thu, 8 Apr 2021 09:18:09 +0000 (12:18 +0300)]
qtmux: Don't post an error message if pushing a sample failed with FLUSHING
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1711>
He Junyan [Wed, 10 Nov 2021 07:18:24 +0000 (15:18 +0800)]
va: Delete the display lock/unlock when we call VA-API.
According to the VA-API's description, they are MT-safe, we do not
need to care about that in our middleware layer. The lock/unlock
operations have overhead and make the performance lower.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1333>
Jan Alexander Steffens (heftig) [Tue, 8 Feb 2022 14:22:39 +0000 (15:22 +0100)]
openaptx: Support libfreeaptx
[libfreeaptx][1] is a fork of libopenapt 0.2.0, used by pipewire.
[1]: https://github.com/iamthehorker/libfreeaptx
Fixes: https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/issues/1642
Closes: https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/issues/1589
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1667>
Nirbheek Chauhan [Mon, 14 Feb 2022 23:28:13 +0000 (04:58 +0530)]
ci: Increase cerbero trigger timeout to 3h
All cerbero jobs explicitly override the default project timeout from
1h to 3h, so this should also be 3h. It is quite common for cerbero
trigger jobs to timeout due to slow download speeds from the artifact
server to FDO gitlab.
https://gitlab.freedesktop.org/heirecka/gstreamer/-/jobs/
18803342
https://gitlab.freedesktop.org/gstreamer/cerbero/-/jobs/
18803360
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1686>
Heiko Becker [Fri, 11 Feb 2022 20:35:54 +0000 (21:35 +0100)]
meson: Don't build lame plugin with -Dlame=disabled
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1686>
Seungha Yang [Mon, 14 Feb 2022 17:26:46 +0000 (02:26 +0900)]
d3d11converter: Fix RGB to GRAY conversion
Fix typo in shader code
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1697>
Stéphane Cerveau [Mon, 14 Feb 2022 11:57:44 +0000 (12:57 +0100)]
tsdemux: unlock mutex on -1 start_offfset
Closing #1013
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1694>
Seungha Yang [Sun, 13 Feb 2022 13:39:46 +0000 (22:39 +0900)]
d3d11download: Fix decide_allocation
Caps should be parsed in any cases.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1691>
Víctor Manuel Jáquez Leal [Wed, 9 Feb 2022 05:01:34 +0000 (06:01 +0100)]
vaapi: Use meson's features for option selection.
Modernize option selection, so if a required dependency is missing,
produce a meaningful error message.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1676>
He Junyan [Sun, 13 Feb 2022 02:49:31 +0000 (10:49 +0800)]
meson: va: Add va option check into plugin's build.
Because the liblibgstva is built unconditionally, we now move the
va option check into va plugin's build.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1685>
He Junyan [Sun, 13 Feb 2022 02:39:37 +0000 (10:39 +0800)]
meson: va: Remove va option check from gst-libs
Building libgstva-1.0 unconditionally for the va implementation can be
used by non-va plugins such as MSDK even when va plugin is disabled
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1685>
Philippe Normand [Sat, 12 Feb 2022 10:05:11 +0000 (10:05 +0000)]
dtlstransport: Notify ICE transport property changes
The application might track the underlying ICE transport, so not notifying
changes might lead to use-after-free issues.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1688>
Marek Vasut [Fri, 11 Feb 2022 22:55:57 +0000 (23:55 +0100)]
jpegdec: Pull row_stride from GST_VIDEO_FRAME_PLANE_STRIDE()
The libjpeg-turbo internal state might not be correctly initialized for
the first frame in a stream, pull the frame stride from gstreamer frame
metadata instead, which is correct even for the first frame, and which
makes this code consistent with the surrounding lines.
Fixes: e6d83d8f96 ("jpegdec: Support libjpeg-turbo colorspace conversion")
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1687>
Marek Vasut [Fri, 11 Feb 2022 22:44:24 +0000 (23:44 +0100)]
jpegdec: Call gst_jpeg_turbo_parse_ext_fmt_convert() before jpeg_start_decompress()
It is imperative that the libjpeg-turbo state is properly initialized
before jpeg_start_decompress() is called. Make sure cinfo.out_color_space
and cinfo.raw_data_out are set to their final values matching their peer
caps before calling jpeg_start_decompress().
Fixes: e6d83d8f96 ("jpegdec: Support libjpeg-turbo colorspace conversion")
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1687>
Marek Vasut [Fri, 11 Feb 2022 22:44:20 +0000 (23:44 +0100)]
jpegdec: Factor out gst_jpeg_turbo_parse_ext_fmt_convert()
Pull out peer caps checking code into gst_jpeg_turbo_parse_ext_fmt_convert().
This code is used by libjpeg-turbo extras to determine whether peer is capable
of handling buffers into which libjpeg-turbo can directly decode data. This
kind of check must be performed before jpeg_start_decompress() is called in
gst_jpeg_dec_prepare_decode() as well as in gst_jpeg_dec_negotiate(), hence
the common code.
This commit does modify the code a little to make it easier to call from both
call sites without much duplication, hence the extra `if (*clrspc)` test.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1687>
Marek Vasut [Fri, 11 Feb 2022 22:29:27 +0000 (23:29 +0100)]
Revert "jpegdec: only allow conversions from RGB"
This reverts commit
2aa2477208c029b0e1b8232d69f4f99a3bf1d473.
The commit is completely wrong, libjpeg-turbo is perfectly capable
of decoding I420 (YUV) to RGB. The test case provided alongside the
aforementioned commit passes without this revert because it decodes
image of JCS_YCrCb color space, so the new `if (clrspc == JCS_RGB)`
condition is false on that image, and the libjpeg-turbo decoding
does not get used. The real bug is hidden by that commit.
The real problem is in the call order of gst_jpeg_dec_prepare_decode()
and gst_jpeg_dec_negotiate(). The gst_jpeg_dec_prepare_decode() calls
jpeg_start_decompress() which sets up internal state of the libjpeg,
however, neither cinfo.out_color_space nor cinfo.raw_data_out are
set correctly yet. Those two are set up in gst_jpeg_dec_negotiate()
which is called a bit later. Therefore, the real fix is the set up
cinfo.out_color_space and cinfo.raw_data_out before calling
jpeg_start_decompress(). This is however a separate patch.
Fixes: 2aa2477208 ("jpegdec: only allow conversions from RGB")
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1687>
He Junyan [Sat, 12 Feb 2022 06:51:51 +0000 (14:51 +0800)]
vavpp: Fix the caps leak in the transform_caps() function.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1689>
Seungha Yang [Fri, 11 Feb 2022 15:51:56 +0000 (00:51 +0900)]
meson: d3d11: Remove d3d11 option check from gst-libs
Build gstd3d11-1.0 unconditionally for d3d11 implementation can be used
by non-d3d11 plugins even when d3d11 plugin is disabled
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1684>
Seungha Yang [Thu, 10 Feb 2022 13:31:06 +0000 (22:31 +0900)]
d3d11: Update build-time dependency
Remove all the d3d11 and dxgi header version dependent ifdef
and bump the minimum requirement to d3d11_4.h and dxgi1_6.h.
We are already failing support old Visual Studio (Windows SDK actually)
such as Visual Studio 2015. Note that our MinGW toolchain satisfies
the requirement.
From runtime point of view, this change should be fine since
we are checking OS version with IUnknown::QueryInterface()
everywhere in order to check API availability
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1684>
Seungha Yang [Fri, 7 Jan 2022 11:02:46 +0000 (20:02 +0900)]
videoaggregator: Fix for unhandled negative rate
Nagative rates have been considered only in
gst_video_aggregator_advance_on_timeout(). Update other places
to fix broken reverse playback.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1500>
Seungha Yang [Wed, 25 Aug 2021 11:18:20 +0000 (20:18 +0900)]
typefindfunctions: Fix WebVTT format detection
If WebVTT file consists of "WebVTT" header without body,
the file size can be smaller than 10 bytes.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1359>
Vivienne Watermeier [Mon, 31 Jan 2022 12:01:10 +0000 (13:01 +0100)]
navigation: add more constructors for navigation events
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1610>
He Junyan [Sat, 8 May 2021 09:54:34 +0000 (17:54 +0800)]
MSDK: Import VA surface as encoder's input.
We make all MSDK encoders declare "memory:VAMemory" feature. Then
the pipeline such as:
gst-launch-1.0 -vf filesrc location=xxx.h264 ! h264parse ! \
vah264dec ! msdkh265enc ! fakesink
will choose VA memory caps between the VA decoder and MSDK encoder.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1087>
He Junyan [Sat, 8 May 2021 09:26:27 +0000 (17:26 +0800)]
MSDK: Add helper function to peek VA surface of VA kind gstbuffer.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1087>
He Junyan [Sat, 8 May 2021 07:51:11 +0000 (15:51 +0800)]
MSDK: Handle context query into the encoder's query function.
The MSDK encoder's query function is not set and it just forwards
all query to its base class. We now need to answer the context
query correctly. Other VA plugins need to query the VA display.
By the way, the current query of "gst.msdk.Context" is also missing.
The other MSDK elements must depend on the bin's context message(
sent in context_propagate()) to set their MsdkContext correctly.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1087>
He Junyan [Tue, 17 Aug 2021 14:36:59 +0000 (22:36 +0800)]
MSDK: Handle context query into the VPP's query function.
The MSDK VPP's query function is not set and it just forwards
all query to its base class. We now need to answer the context
query correctly. Other VA plugins need to query the VA display.
By the way, the current query of "gst.msdk.Context" is also missing.
The other MSDK elements must depend on the bin's context message(
sent in context_propagate()) to set their MsdkContext correctly.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1087>
He Junyan [Tue, 17 Aug 2021 14:34:43 +0000 (22:34 +0800)]
MSDK: Handle context query into the decoder's query function.
The MSDK decoder's query function is not set and it just forwards
all query to its base class. We now need to answer the context
query correctly. Other VA plugins need to query the VA display.
By the way, the current query of "gst.msdk.Context" is also missing.
The other MSDK elements must depend on the bin's context message(
sent in context_propagate()) to set their MsdkContext correctly.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1087>
He Junyan [Tue, 17 Aug 2021 14:33:18 +0000 (22:33 +0800)]
MSDK: Add a handle_context_query() helper function.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1087>
He Junyan [Tue, 17 Aug 2021 13:51:58 +0000 (21:51 +0800)]
MSDK: Improve the GstContext setting logic.
We now can use the gst va lib's display to create our MSDK context,
and use its helper functions to simplify our code. The improved logic
is like this:
1. Every MSDK element should use gst_msdk_context_find() to find a MSDK
context from neighbour. If valid, reuse it.
2. Use gst_msdk_ensure_new_context(). It will first query neighbours
about the GstVaDisplay, if found(e.g. some VA element is connected),
use gst_msdk_context_from_external_display() to create a MSDK context.
3. Then, creating the MSDK context from scratch. It creates both the
display and MSDK context.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1087>
He Junyan [Sat, 8 May 2021 05:49:23 +0000 (13:49 +0800)]
MSDK: Add new_with_va_display() helper function to create context.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1087>
He Junyan [Sat, 8 May 2021 03:44:57 +0000 (11:44 +0800)]
MSDK: Use the VA display object to replace the raw display handle.
The VA display object from VA lib is a common defined object. which
contain the whole display things. It is easier to use, and more important,
we can share it with the other VA plugins and keep all the VA related
plugins working on the same GPU device.
We also delete the useless gst_msdk_context_get_fd() API.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1087>
He Junyan [Fri, 11 Feb 2022 03:11:03 +0000 (11:11 +0800)]
MSDK: Add the gst va lib into MSDK's dependency.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1087>
He Junyan [Thu, 10 Feb 2022 04:52:30 +0000 (12:52 +0800)]
av1parse: Add temporal unit check when TD is absent.
The current manner for deciding the new temporal unit is based on
temporal delimiter(TD) OBU. We only start a new temporal unit when
the TD comes.
But some streams do not have TD at all, which makes the output "TU"
alignment fail to work. We now add check based on the relationship
between the different layers and it can successfully judge the TU edge.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1634>
He Junyan [Fri, 4 Feb 2022 09:12:15 +0000 (17:12 +0800)]
av1parse: let the parse continue when MISSING_OBU_REFERENCE error.
Some streams may have verbose OBUs before a valid sequence header. We
should let the parse continue rather than return a error.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1634>
He Junyan [Fri, 4 Feb 2022 03:40:18 +0000 (11:40 +0800)]
av1parse: Continue when we fail to detect the alignment.
Some streams may have problematic OBUs at the beginning, which causes
the parse fail to detect the alignment and return error. For example,
there may be verbose OBUs before a valid sequence, which should be
discarded until we meet a valid sequence. We should let the parse
continue when we meet such cases, rather than just return error.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1634>
Bastien Nocera [Tue, 1 Feb 2022 13:28:24 +0000 (14:28 +0100)]
gtk: Fix rotation not being applied when paused
The video wouldn't be redrawn immediately when a rotation was applied
but the pipeline was paused, as no new buffers were scheduled to be
displayed.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1618>
Bastien Nocera [Tue, 1 Feb 2022 13:26:02 +0000 (14:26 +0100)]
gtk: Add a way to queue redrawing the base GTK widget
This will be used to request a redraw of the GTK widget should the
display be changed using properties not directly handled by the base GTK
widget, but by one of its descendants.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1618>
Seungha Yang [Tue, 30 Mar 2021 10:23:12 +0000 (19:23 +0900)]
ivfparse: Don't set zero resolution on caps
It could be zero if the information is not available at ivfparse
side, or not implemented. In that case, simply don't set
width/height on caps, otherwise downstream would be confused
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1649>
Seungha Yang [Fri, 28 Jan 2022 16:52:26 +0000 (01:52 +0900)]
mediafoundation: Add support for MP3 decoding
See also
https://docs.microsoft.com/en-us/windows/win32/medfound/windows-media-mp3-decoder
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1596>
Seungha Yang [Wed, 26 Jan 2022 17:20:37 +0000 (02:20 +0900)]
mediafoundation: Add support for AAC decoding
See also
https://docs.microsoft.com/en-us/windows/win32/medfound/aac-decoder
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1596>
Seungha Yang [Thu, 27 Jan 2022 16:45:30 +0000 (01:45 +0900)]
mediafoundation: Cosmetic changes
Rename baseclass to be consistent with other Windows plugins
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1596>
Seungha Yang [Wed, 26 Jan 2022 17:26:02 +0000 (02:26 +0900)]
mediafoundation: Use & operator instead of GetAddressOf()
Use shorter expression
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1596>