Commit graph

285 commits

Author SHA1 Message Date
Víctor Manuel Jáquez Leal
52bd1931b8 vkencoder-private: usage structure is provided by caller
As all the profile structure, it's not intended to be filled in
gst_vulkan_encoder_start() function, but by the caller.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7974>
2024-11-27 20:13:12 +00:00
Nirbheek Chauhan
23a006c64f meson: Don't unconditionally invoke the libsoup subproject
fallback: kwarg will invoke the specified subproject even if required:
false, which is not what we want here.

Reported at https://gitlab.freedesktop.org/gstreamer/gstreamer/-/issues/4045#note_2674340

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7965>
2024-11-27 12:12:36 +00:00
Zhong Hongcheng
821754e2d5 tests: Add the VVC(H266) parser test cases
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/5003>
2024-11-24 16:49:25 +00:00
Víctor Manuel Jáquez Leal
d9aa8a78ea h264bitwriter: implement gst_h264_bit_writer_filler()
This is required for vulkan encoder since it can only write slides after aligned
offsets.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7785>
2024-11-11 16:05:38 +00:00
Diego Nieto
0d85cdafd5 exiftag: handle GST_TAG_CAPTURING_LIGHT_SOURCE tag
This exif tag allows to specify the different light conditions
when taking a picture. This tag is defined in:
https://exiftool.org/TagNames/EXIF.html#LightSource

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/5571>
2024-11-10 12:57:36 +00:00
Sebastian Dröge
b5e119bbcc ccconverter: Don't override in_fps_entry when trying to take output
This allows to handle CDP streams where the framerate is not provided by the
caps and generally gives preference to the framerate inside the CDP packets over
the one in the caps.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7532>
2024-11-10 08:37:36 +00:00
Xavier Claessens
468dcbe9b7 Revert "unixfd: disable flaky test_unixfd_segment for now"
This reverts commit 06cd4e2457.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6765>
2024-10-29 12:12:26 +00:00
Xavier Claessens
9b946098df unixfd: Fix racy unit test by adding wait-for-connection property
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6765>
2024-10-29 12:12:26 +00:00
Jordan Petridis
22ec1d8e4e ci: add suppressions for OpenSSL false positives
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7455>
2024-10-25 13:55:20 +00:00
Jordan Petridis
bc666db5fe gst-plugins-bad.supp: Remvoe gssdp leaks that have been fixed
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7455>
2024-10-25 13:55:20 +00:00
Peter Stensson
06d4629521 codectimestamper: Fix gint wraparound in pts_compare_func
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>
2024-10-25 01:49:10 +00:00
Jochen Henneberg
2ae846a77a examples: va: Added VP8 encoder to dynamic reconfigure
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6959>
2024-10-22 11:48:49 +00:00
Daniel Morin
9fc017667f test: Adding a test for segmentation analytics-meta
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6026>
2024-10-17 18:13:03 +00:00
Jordan Petridis
4b8d43446e lc3: tests: Zero out the buffer we allocate for the tests
Otherwise liblc3 will try to access the uninitialized memory
and it makes valgrind very sad.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7657>
2024-10-16 14:16:43 +00:00
Víctor Manuel Jáquez Leal
809ab829d0 tests: va: fix vapostproc test for DMABuf
Now it picks the first format in the template srcpad list and do
the convertion. Also the format size is reduced because not all
drives support 4K as DMABuf (radeonsi).

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7636>
2024-10-10 16:34:04 -04:00
Jordan Petridis
ab54e45f67 tests/lc3: Allocate the same size for the buffer and the data
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7631>
2024-10-09 22:16:13 +00:00
Sebastian Dröge
b7b24573ce common: Use more efficient versions of GstCapsFeatures API where possible
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7432>
2024-09-26 19:26:18 +03:00
Sebastian Dröge
6233eb0ff3 common: Stop using GQuark-based GstStructure field name API
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7432>
2024-09-26 19:21:29 +03:00
Edward Hervey
6ce1910c5b mpegts: Handle ISO 13818 / ITU H.222.0 base extension descriptor
Previously this was hardcoded to the DVB extension descriptors (0x7f), but it
should also be applied for the base specification extension descriptors (0x3f)

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7172>
2024-09-16 18:31:01 +01:00
Seungha Yang
25f696c7c0 examples: Add application CUDA memory pool example
An example to show application managed CUDA memory pool usage

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7427>
2024-09-11 17:15:14 +00:00
He Junyan
9327458cfb tests: Add the jpeg bit code writer test case
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6022>
2024-09-05 09:56:02 +00:00
Edward Hervey
81e7bde67c check: Disable failing test
Test hasn't been properly fixed for several years with modern libsoup, and it
only for the legacy adaptive demuxer.

Fixes #3783

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7454>
2024-09-05 10:09:58 +02:00
Philippe Normand
89f335f173 webrtcbin: Prevent crash when attempting to set answer on invalid SDP
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>
2024-09-02 04:00:57 +00:00
Edward Hervey
087cb87d27 bad: Add suppression for libsrt issues
This is not code we control

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7415>
2024-08-28 06:54:02 +00:00
Edward Hervey
38271fc9e4 check: Fix leak in lc3 test
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7415>
2024-08-28 06:54:02 +00:00
Carlos Bentzen
77faf0a163 webrtcbin: fix regression with missing RTP header extensions in Answer SDP
webrtcsrc first creates recvonly transceivers with codec-preferences
and expects that after applying a remote description, the
previously created transceivers are used rather than having new
transceivers created.

When pairing webrtcsink + webrtcsrc, the offer sdp from webrtcsink has a media
section with sendonly direction. In !7156, which was implemented following
RFC9429 Section 5.10, we only reuse a unassociated transceiver when applying a
remote description if the media is sendrecv or recvonly, and that caused creation
of new transceivers when applying a remote offer in webrtcsrc, thus losing
information from codec preferences like the RTP extension headers in the
previously created transceivers.

Since the change in !7156 broke existing code from webrtcsrc, relax the condition
for reusing unassociated transceivers and add a test to document this behavior which
wasn't covered by any tests before.

Fixes #3753.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7417>
2024-08-27 23:56:00 +00:00
Thibault Saunier
87c69e5174 ci: Fail tests if we forget to checkout expectation files
And add missing expectation files

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7400>
2024-08-21 17:53:38 +00:00
Seungha Yang
1b5f026119 examples: Add CUDA based in-place transform element example
Adding a CUDA example element for plugin developers

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7004>
2024-08-20 23:48:24 +00:00
Jan Schmidt
96c4bd8d9f webrtc: Fix racy unit test
Don't reuse the same stats state structure across multiple
get-stats calls. Make each callback take a copy of the
non-changing fields it needs and use a local working copy
to avoid crashing.

Fixes problems with the unit test crashing sometimes for the
unit test introduced in MR !7338

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7387>
2024-08-20 12:07:02 +00:00
Jan Schmidt
97845475c5 webrtcbin: Fix uint64 -> uint confusion for ice-candidate priority
ICE candidate priority is a 32-bit field and reported as such in the
webrtcbin statistics, but the documentation was incorrect, and the
unit test was looking for a uint64.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7338>
2024-08-19 21:07:52 +10:00
Jan Schmidt
7da5d03b29 webrtcbin: Fixes for bundled statistics generation
When multiple streams are bundled on the same transport,
the statistics would end up incorrectly generated,
as each pad would regenerate stats for every ssrc on the
transport, overwriting previous iterations and assigning
bogus media kind and other values to the wrong ssrc.

Fix by making sure each pad only loops and generates
statistics for the one ssrc that pad is receiving / sending.

Add a unit test that the codec kind field in RTP statistics
are now generated correctly.

Fixes: https://gitlab.freedesktop.org/gstreamer/gstreamer/-/issues/2555
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7338>
2024-08-19 21:07:51 +10:00
Seungha Yang
4bb3854772 d3d12: Add d3d12swapchainsink element
Adding a new videosink element for Windows composition API based
applications. Unlike d3d12videosink, this element will create only
DXGI swapchain by using IDXGIFactory2::CreateSwapChainForComposition()
without actual window handle, so that video scene can be composed
via Windows native composition API, such as DirectComposition.
Note that this videosink does not support GstVideoOverlay interface
because of the design.

The swapchain created by this element can be used with
* DirectComposition's IDCompositionVisual in Win32 app
* WinRT and WinUI3's UI.Composition in Win32/UWP app
* UWP and WinUI3 XAML's SwapChainPanel

See also examples in this commit which show usage of the videosink

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7287>
2024-08-19 11:07:17 +09:00
Jan Schmidt
d266995323 tests/webrtcbin: Add a lock around the stats test
Prevent any race if both webrtcbin end up generating their
statistics simultaneously, however unlikely.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7365>
2024-08-16 14:18:35 +00:00
Jan Schmidt
460f5dcb33 tests/webrtcbin: Fix racy rollback test
Prevent the default webrtc test machinery from attempting to
create and set an answer when we're just testing rollback
of the offers. Add some locking / waiting to ensure the test
is complete before exiting.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7365>
2024-08-16 14:18:35 +00:00
Jan Schmidt
490c21a72e tests/webrtcbin: Use fail_unless_matches_string()
Use pattern matching against expected error strings that
might include internal element names, where the names
are default assigned with incrementing integers. When running
with CK_FORK=no, there may have been previous tests that
ran in the same process and incremented the counters more
than when running in the default fork-per-test mode.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7365>
2024-08-16 14:18:35 +00:00
Qian Hu (胡骞)
104dcc90f1 h26xparse: bypass check for length_size_minus_one
fix playback fail, when some file with length_size_minus_one == 2

According to the spec 2 cannot be a valid value, so that stream has a
bad config record. but breaking the decoding because of that, perhaps is too much.
and ffmpeg seem not check this

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7213>
2024-08-14 08:31:15 +00:00
Jan Schmidt
0f8fc27892 webrtcbin: Fix renegotiation checks
When checking for renegotiation against a local offer,
reverse the remote direction in the corresponding answer
to fix falsely not triggering on-negotiation needed when
switching (for example) from local sendrecv -> recvonly
against a peer that answered 'recvonly'.

In the other direction, when the local was the answerer,
renegotiation might trigger when it didn't need to -
whenever the local transceiver direction differs from
the intersected direction we chose. Instead what we want
is to check if the intersected direction we would now
choose differs from what was previously chosen.

This makes the behaviour in both cases match the
behaviour described in
https://www.w3.org/TR/webrtc/#dfn-check-if-negotiation-is-needed

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7303>
2024-08-11 21:45:10 +00:00
Jan Schmidt
4b775228bf webrtcbin: Make basic rollbacks work
Fixes for basic rollback (from have-local-offer or have-remote-offer to
stable). Allow having no SDP attached to the webrtc session description
in that case, and avoid all the transceiver and ICE update logic
normally applied when entering the stable signalling state

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7304>
2024-08-07 21:10:43 +10:00
Carlos Bentzen
efa0a3ec6a webrtcbin: connect output stream on recv transceivers
With MR 7156, transceivers and transports are created earlier,
but for sendrecv media we could get `not-linked` errors due to
transportreceivebin not being connected to rtpbin yet when incoming
data arrives.

This condition wasn't being tested in elements_webrtcbin, but could be
reproduced in the webrtcbidirectional example. This commit now also
adds a test for this, so that this doesn't regress anymore.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7294>
2024-08-05 08:25:04 +00:00
Carlos Bentzen
48ae40f477 webrtcbin: create and associate transceivers earlier in negotation
According to https://w3c.github.io/webrtc-pc/#set-the-session-description
(steps in 4.6.10.), we should be creating and associating transceivers when
setting session descriptions.

Before this commit, webrtcbin deviated from the spec:
1. Transceivers from sink pads where created when the sink pad was
   requested, but not associated after setting local description, only
   when signaling is STABLE.
2. Transceivers from remote offers were not created after applying the
   the remote description, only when the answer is created, and were then
   only associated once signaling is STABLE.

This commit makes webrtcbin follow the spec more closely with regards to
timing of transceivers creation and association.

A unit test is added, checking that the transceivers are created and
associated after every session description is set.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7156>
2024-08-01 07:38:46 +00:00
Víctor Manuel Jáquez Leal
2990cc5f71 vulkan: add source pipeline stage to _operation_add_frame_barrier()
Instead of dragging the last destination pipeline stage as current barrier
source pipeline stage (which isn't a valid semantic) this patch adds a parameter
to gst_vulkan_operation_add_frame_barrier() to set the source pipeline stage to
define the barrier.

The previous logic brought problems particularly with queue transfers, when the
new queue doesn't support the stage set during a previous operation in a
different queue.

Now the operation API is closer to Vulkan semantics.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7165>
2024-07-19 14:45:39 +02:00
Víctor Manuel Jáquez Leal
dd4027388e vulkan: fix wrong stages or access in barriers
While working on !7165 we found out that some parameters for barriers were wrong
or the destination pipeline stage was too coarse.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7200>
2024-07-19 12:44:51 +02:00
Taruntej Kanakamalla
f54320b161 lc3: remove bitstream comparison in the tests
since the encoded output is changing based on version
it does not make sense to check the output bitstream with a fixed
bytearray since the version in the target might vary. So sticking
to checking the number of output buffers and encoded frame size
similar to the other tests

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7141>
2024-07-05 18:27:20 +05:30
Seungha Yang
37e1847464 d3d12videosink: Add support for window handle update
A large refactoring commit for adding features and improve performance

* Reuse internal converter and overlay compositor:
  Converter can be reused as long as input and display formats are not
  changed. Also overlay compositor reconstruction is required only if
  display format is changed

* Don't wait for full GPU flush on resize or close:
  D3D12 swapchain requires GPU idle in order to resize backbuffer.
  Thus CPU side waiting is required for swapchain related commands
  to be finished. However, don't need to wait for full GPU flushing.

* Support multiple sink on a single external window
  Keep installed subclass window procedure even if there's no associated
  our internal HWND. This will make window procedure hooking less racy.
  Then parent HWND's message will be transferred to our internal HWNDs
  if needed.

* Adding support for window handle update
  Application can change target HWND even when videosink is playing or
  paused state. So, users can call gst_video_overlay_set_window_handle()
  against d3d12videosink anytime. The videosink will be able to update
  internal state and setup resource upon requested.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7013>
2024-06-17 16:05:00 +00:00
Seungha Yang
6fb0c7b928 examples: cuda: Add CUDA memory synchronization example
Add an example code for external CUDA context sharing and
gst_cuda_memory_sync()

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6864>
2024-06-06 11:27:26 +00:00
Stéphane Cerveau
df33ae2da6 gst-plugins-bad: tests: rename vkvideoencode tests
Rename vulkan encode tests to be able to use the namespace
libs_vkvideoencode*.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6992>
2024-06-05 17:50:27 +00:00
Stéphane Cerveau
21ee264d65 vulkan: remove GST_VULKAN_HAVE_VIDEO_ENCODERS
Use 2.3.275 as first supported SDK version

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6992>
2024-06-05 17:50:27 +00:00
Philippe Normand
299a000917 webrtcbin: Allow session level setup attribute in SDP
An SDP answer can declare its setup attribute at the session level or at the
media level. Until this patch we were validating only the latter case and an
assert was raised in the former case.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6930>
2024-05-27 14:21:15 +00:00
Seungha Yang
84aecab150 d3d12videosink: Add overlay signal to support d3d12/d3d11/d2d overlay
Conceptually identical to the present signal of d3d11videosink.
This signal will be emitted with current render target
(i.e., swapchain backbuffer) and command queue. Signal handler
can record GPU commands for an overlay image or to blend
an image to the render target.

In addition to d3d12 resources, videosink will send
d3d11 and d2d resources depending on "overlay-mode"
property, so that signal handler can render by using
preferred/required DirectX API.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6838>
2024-05-24 15:55:17 +00:00
Víctor Manuel Jáquez Leal
e913b4870a vkformat: try UNORM format first and decouple them from colorimetry
From the spec (chapter 34, v1.3.283):

````
UNORM: the components are unsigned normalized values in the range [0, 1]

SRGB: the R, G and B components are unsigned normalized value that represent
      values using sRGB nonlinear encoding, while the A component (if one
      exists) is a regular unsigned normalized value
```

The difference is the storage encoding, the first one is aimed for image
transfers, while the second is for shaders, mostly in the swapchain stage in the
pipeline, and it's done automatically if needed [1].

As far as I have checked, other frameworks (FFmpeg, GTK+), when import or export
images from/to Vulkan, use exclusively UNORM formats, while SRGB formats are
ignored.

My conclusion is that Vulkan formats are related on how bits are stored in
memory rather their transfer functions (colorimetry).

This patch does two interrelated changes:

1. It swaps certain color format maps to try first, in both
gst_vulkan_format_from_video_info() and gst_vulkan_format_from_video_info_2(),
the UNORM formats, when comparing its usage, and later check for SRGB.

2. It removes the code that check for colorimetry in
gst_vulkan_format_from_video_info_2(), since it not storage related.

1. https://community.khronos.org/t/noob-difference-between-unorm-and-srgb/106132/7

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6797>
2024-05-23 17:41:30 +00:00