Commit graph

19 commits

Author SHA1 Message Date
Mengkejiergeli Ba
735b6af5b9 va: Fix some code defects
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3919>
2023-02-20 16:02:38 +00:00
Victor Manuel Jaquez Leal
171a626c6c vaav1dec: Remove double caps unref.
There was a duplicated caps unref raising a warning.

Also it reorgs the sorrounding code for simplicity.

Fixes: #1196
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3587>
2022-12-16 19:18:25 +00:00
Victor Manuel Jaquez Leal
4cbdf43e7d va: Handle input caps change.
Update output caps if it's notified by baseclass

See https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3328

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3480>
2022-12-01 18:54:14 +00:00
Victor Manuel Jaquez Leal
19b83bc156 vaav1dec: Use gst_va_base_dec_set_output_state().
And even that vaav1dec doesn't use vabasedec negotiate vmethod, it should align
with the new scheme of using base's width & height for surface size and
output_info structure for downstream display size negotiation.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3480>
2022-12-01 18:54:14 +00:00
Víctor Manuel Jáquez Leal
31c63a6e6c vaav1dec: Use gst_va_base_dec_prepare_output_frame().
And simplify a bit the code flow.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3480>
2022-12-01 18:54:14 +00:00
Víctor Manuel Jáquez Leal
f1a9728b30 va: Remove last_ret error handling in decoders.
It was used in the early development of the base classes. Now it
shouldn't be needed.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3480>
2022-12-01 18:54:14 +00:00
Víctor Manuel Jáquez Leal
de5b76a922 va: Add and use gst_va_base_dec_process_output().
This function will copy the frame, if it's needed, and will apply buffer flags.

The function is used by all the decoders.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3480>
2022-12-01 18:54:14 +00:00
He Junyan
765adf5325 va: Fix the caps memory leak by gst_va_pool_new_with_config().
The gst_va_pool_new_with_config() will ref the caps, and so we need
to unref after that.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3503>
2022-12-01 17:31:16 +00:00
Víctor Manuel Jáquez Leal
c8a0d706ca va: Add render node name in non-first devices.
There could be multi-GPU setups where the non-first has more
entrypoints than the first one, and the elements names are not
homogeneous, leading to pipeline building error.

This patch add the render node in the elements names when they belong
to the non-first device.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3491>
2022-12-01 16:00:55 +00:00
Victor Manuel Jaquez Leal
c60504bd9e va: Document device-path property.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3361>
2022-11-08 21:49:05 +00:00
He Junyan
9a991cfe9e va: av1dec: user internal buffer pool for non output layers.
The AV1 support multi spatial layers within one TU with different
resolutions, and only the highest spatial layer need to be output.
For example, there are two spatial layer, base level is 800x600
and higher level is 1920x1080. We need to decode both because the
higher level needs base layer as reference, but we only need to output
1920x1080 frames here.

The current manner always renegotiates the caps once we detect the
current picture resolution changes, so we renegotiate again and
again between different layers. That's a big waste and has very
low performance. We now only do the renegotiation for the highest
output layer. For other non output layers, we just keep a internal
buffer pool which is big enough to handle the surface allocation.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2382>
2022-09-16 09:45:00 +00:00
Víctor Manuel Jáquez Leal
c47255d148 va: basedec: Select format from template or negotiated caps.
Instead of using a hard-coded list of preferred formats according the
chroma type, now if now caps are pre-negotiated, from template caps
will choose the first format with the same chroma type. If
pre-negotiated, then it will choose the first format, with same chroma
type, from the first caps structure.

Also all the decoders will check if GST_VIDEO_FORMAT_UNKNOWN is
returned, failing the negotiation.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2351>
2022-05-09 08:50:58 +00:00
Seungha Yang
6a4c42c03f codecs: Signal required DPB size for AV1,MPEG2,VP8, and VP9 via new_sequence()
Make all codecs consistent so that subclass can know additional DPB
size requirement depending on render-delay configuration regardless
of codec. Note that render-delay feature is not implemented for AV1
yet but it's planned.

Also, consider new_sequence() is mandatory requirement, not optional

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2343>
2022-05-03 14:17:49 +00:00
Víctor Manuel Jáquez Leal
400faf7361 va: Set <gst/va/gstva.h> as library single point entry header.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2048>
2022-03-29 19:48:30 +00:00
He Junyan
990fbb3b52 va: Move allocators and pool objects into gstva library.
In order to other plugins use gstva objects, such as allocators and buffer
pools, this merge request move them from the va plugin to the gstva library.

This objects are not exposed in <gst/va/gstva.h> since they are not expected
to be used by users, only by plugin implementators.

Because of the surface copy design, which is used to implement allocator's
mem_copy() virtual function, depends on the vafilter, which is kept inside
the plugin, memory copy through VAPosproc is disabled and removed temporarly.

Also added some missing parameter validation.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2048>
2022-03-29 19:48:30 +00:00
Daniel Almeida
9a364464f0 av1decoder: Sync duplicate_picture with VP9 one
Pass the current frame to the duplicate_picture callback. This makes it easier
to set the frame's output_buffer if we already have one available. Also
documented that unlike VP9, it is not optional to implement this as the
picture will populate the DPB if it is a key-frame. To ensure this, remove the
default implementation.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1992>
2022-03-21 15:10:01 +00:00
Víctor Manuel Jáquez Leal
97e3f88c20 va: Move common variable need_negotiation to GstBaseDec.
This is a common variable to all decoders, so it's sound to move it to
the base helper.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1257>
2021-10-29 16:06:52 +00:00
Víctor Manuel Jáquez Leal
a2a3c81c85 va: Move back parent_object to each element.
Using GstBaseDec hack to access the parent_object of each element in
the element itself is a bit fragile. It would be better to keep its
own parent object as the usual global variable. It would make it
resistant to code changes.

The GstBaseDec macro to access the parent object now it's internal to
base decoder.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1257>
2021-10-29 16:06:52 +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/va/gstvaav1dec.c (Browse further)