Commit graph

29 commits

Author SHA1 Message Date
Nicolas Dufresne
e34f765a7e v4l2codecs: Don't unref allocation query caps
The caps obtained from parsing the allocation query is borrowed and
should not be unreffed. This fixes criticals assertion introduced in
1.24.1.

(gst-launch-1.0:242): GStreamer-CRITICAL **: 19:48:02.667:
    gst_mini_object_unref: assertion 'GST_MINI_OBJECT_REFCOUNT_VALUE (mini_object) > 0' failed

Fixes: 5189e8b956 ("v4l2codecs: decoders: Add DMA_DRM caps support")
Closes #3462

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6679>
2024-04-18 06:13:17 +00:00
Robert Mader
5189e8b956 v4l2codecs: decoders: Add DMA_DRM caps support
In order to simplify caps negotiations for clients and, notably, be more
compatible with va* decoders.
Crucially this allows clients to know ahead of time whether buffers will
actually be DMABufs.

Similar to GstVaBaseDec we only announce system memory caps if the peer
has ANY caps. Further more, and again like va decoders, we fail in
`decide_allocation()` if DMA_DRM caps are used without VideoMeta.
Apart from buggy peers this can happen e.g. when a peer with ANY caps
is used in combination with caps filters.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/5890>
2024-03-14 17:32:13 +00:00
Robert Mader
513d0d8cbb v4l2codecs: decoders: Introduce and use set_output_state helper class
Allowing us to avoid some code duplication. This will become more
important with upcoming changes to caps generation.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/5890>
2024-03-14 17:32:13 +00:00
Robert Mader
65896dab75 v4l2codecs: decoders: Use src template for negotiation filter
This ensures we don't create filter caps that are not supported by the
individual codec implementations, as well as that the resulting caps
have the required fields so they can be turned into a GstVideoFormat.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/5890>
2024-03-14 17:32:13 +00:00
Robert Mader
6b024b50ab v4l2codecs: h264: Fix a memory leak on renegotiation
We only use this anchor when streaming, in which case output_state is
set and needs to get unreffed.
This is in line with how it is handled for all other codecs.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6127>
2024-02-15 22:37:11 +01:00
Marek Vasut
3335af0efe v4l2codecs: Deduplicate picture frame number to timestamp in ns
Add macro which converts picture frame number to suitable timestamp in
nanoseconds for use in V4L2 VB2 buffer lookup. Since multiple codecs do
the same operation and almost all got it wrong, do it in one place so it
can be fixed in one place again, if needed.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/5791>
2023-12-20 18:47:39 +00:00
Marek Vasut
7725aa7d77 h264decoder: Align wraparound fix
Instead of casting GST_CODEC_PICTURE_FRAME_NUMBER (ref_pic) to u64,
use 1000ULL which is also u64 . This only aligns the behavior here
with '*decoder: Fix multiplication wraparound' commits.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/5791>
2023-12-20 18:47:39 +00:00
Seungha Yang
ea3dfadbed h264decoder: Port to GstCodecPicture struct
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/5285>
2023-09-08 11:51:23 +00:00
James Cowgill
70ff80a873 v4l2codecs: Always chain up to parent decide_allocation function
The `gst_video_decoder_negotiate_pool` function expects the
`decide_allocation` function to always provide a pool and will fail to
negotiate if the pool is missing. If we return immediately (even if we
don't need to do anything special) negotiation will fail if the
downstream element does not propose a pool.

Fix by chaining up to the default `decide_allocation` function which
adds a fallback pool if one was not already proposed.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/4630>
2023-06-15 17:02:28 +00:00
Sebastian Dröge
ccad9a7338 plugins: Fix various trivial clang compiler warnings
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/4210>
2023-03-18 16:16:55 +02:00
Nicolas Dufresne
d3c5fc815e v4l2codecs: Allow output caps to be updated
This change allow output caps to be updated even though we stay in
streaming state. This is needed so that any upstream updated to fields
like framerate, hdr data, etc. can result in a downstream caps event
being pushed.

Previously, any of these changes was being ignored and the downstream
caps would not reflect it.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3328>
2022-11-16 13:12:38 +00:00
Nicolas Dufresne
316b182a86 v4l2codecs: h264: Reset the number of slices on picture start
This way, if we aborted a frame in the middle, we don't endup thinking
there is multiple slices to send.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1079>
2022-07-22 10:21:03 -04:00
Nicolas Dufresne
c2144ec586 v4l2codecs: h264: Fix initialization of the slice array
The function g_array_sized_new() leaves the len to 0, but the slice
implementation assumes it would be set to 4. Sending multiple slices is
not yet support for H.264 as no driver needed it yet, but if that code
was to be used it would have overflowed as the array would never grow as
multiple 0 by 2 always results in 0.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1079>
2022-07-22 10:21:03 -04:00
Nicolas Dufresne
a707d984ae v4l2codecs: h264: Rename count variable to num_controls
This is the counter for the used control slot. The term num_controls was
introduced in the HEVC decoder and felt like a better choice.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1079>
2022-07-22 10:21:02 -04:00
Dmitry Osipenko
fd694d840a v4l2codecs: h264: Set frame type flags
V4L spec now requires decode_params flags to be set in accordance to the
frame's type. In particular this is required by H.264 decoder of NVIDIA
Tegra SoC to operate properly. Set the flags based on type of parsed
slices.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1757>
2022-05-06 13:35:59 +00:00
Nicolas Dufresne
efa360ce82 v4l2codecs: h264: Set other field poc if available
Whenever the other field POC is available, pass it back to the
driver. This should not be strictly required, but it makes things
easier to debug and more forgiven.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2009>
2022-03-25 21:40:28 +00:00
Nicolas Dufresne
dcd62c35f0 v4l2codecs: Trace request by it FD
This helps when crossing the debug logs with the kernel logs.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2009>
2022-03-25 21:40:28 +00:00
Nicolas Dufresne
d24dc74170 v4l2codecs: h264: Fix dpb entry flag FLAG_FIELD
The logic to guess back this value was broken. Use the value
now saved into the picture instead. This was tested using
LibreELEC patched 5.15 kernel, and fixed the interlaced decode
issues.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2009>
2022-03-25 21:40:28 +00:00
Dmitry Osipenko
b757d12cff 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: 47bfa71530 ("v4l2codecs: h264: Improve ABI check ")
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1726>
2022-02-16 13:16:17 +00:00
Seungha Yang
6434d69f8c codecs: Stop claiming constness for refcounted object
It's almost pointless and makes little sense as subclass might
want to modify refcount of the object or so. And all subclasses
are already casting them to non-const version as well.
In a general sense, we need to avoid passing refcounted object
with const qualifier.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1238>
2022-02-06 17:17:22 +00:00
Nicolas Dufresne
47bfa71530 v4l2codecs: h264: Improve ABI check
This moves the ABI check to the registration, so we don't expose
decoders with the wrong ABI or that are just broken somehow. It
also makes few enhancement:

- Handle missing, but required controls
- Prints the controls macro name instead of id

This should fix RK3399 support with a currently release minor
regression in the Hantro driver that cause errors.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1599>
2022-01-29 15:33:49 +00:00
Nicolas Dufresne
d7363bd838 v4l2codecs: Fix debug assertion in register functions
As now, we warn if the decoder have no support src pixel format, but that
warning is called before the type (hence the debug category) is initialized.
Fix this by moving the debug category init out of the type initialization,
into the register funcitons.

This will fix an assertion that occures in the register function and allow
relevant log to be seen by the users.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1588>
2022-01-27 18:53:41 +00:00
Nicolas Dufresne
107088a658 v4l2codecs: Unify the src template caps format
Notably NV12_4L4 ended up being applied to only VP9 decoder. This fixes the
situation by using a central define for all static src pad templated formats.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/965>
2022-01-21 19:32:30 +00:00
Nicolas Dufresne
a6eb1e7782 v4l2codecs: Fix renegotiation
If we hold the last reference to the allocator, leaving the device
streaming will cause an EBUSY error when trying to free the allocate
buffers.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1404>
2021-12-02 15:45:26 -05:00
Nicolas Dufresne
4ca7f1e014 v4l2codecs: h264: Fix typo in comment
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1387>
2021-11-30 00:28:34 +00:00
Nicolas Dufresne
83933b7772 v4l2codecs: h264: Properly set pic_num/frame_num
The V4L2 uAPI uses pic_num for both PicNum and ShortTermPicNum. It also
doe the same for both FrameNum and LongTermFrameIdx. This change does
not change the fluster score, but fixed a visual corruption noticed
with some third party streams.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1387>
2021-11-30 00:28:34 +00:00
James Cowgill
8b932c105f v4l2codecs: Handle allocator creation failure
If `VIDIOC_REQBUFS` doesn't return enough buffers the allocator creation
function can fail and return `NULL`. Handle this by generating an error
and returning instead of segfaulting.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1220>
2021-11-02 15:25:48 +00:00
Seungha Yang
e322745763 codecs: h264decoder: Use GstFlowReturn everywhere
boolean return value is not sufficient for representing the reason
of error in most cases. For instance, any errors around new_sequence()
would mean negotiation error, not just *ERROR*.
And some subclasses will allocate buffer/memory/surface on new_picture()
but it could be failed because of expected error, likely flushing

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1019>
2021-10-04 20:56:46 +00:00
Thibault Saunier
019971a3c7 Move files from gst-plugins-bad into the "subprojects/gst-plugins-bad/" subdir 2021-09-24 16:14:36 -03:00
Renamed from sys/v4l2codecs/gstv4l2codech264dec.c (Browse further)