Commit graph

26404 commits

Author SHA1 Message Date
Seungha Yang
f62ecc1625 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>
2020-10-16 15:56:49 +00:00
Seungha Yang
6e1c1178c2 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>
2020-10-16 15:56:49 +00:00
Seungha Yang
fe83a12b10 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>
2020-10-16 15:56:49 +00:00
Seungha Yang
592a8d5400 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>
2020-10-16 15:56:49 +00:00
Seungha Yang
4cc73ff9d6 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>
2020-10-16 15:56:49 +00:00
Seungha Yang
dd995bd9d4 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>
2020-10-16 15:56:49 +00:00
Seungha Yang
b1ac8bafc9 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>
2020-10-16 15:56:49 +00:00
Seungha Yang
d16a9237c3 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>
2020-10-16 15:56:49 +00:00
Seungha Yang
a8e9d616d6 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>
2020-10-16 15:56:49 +00:00
Seungha Yang
8635d12929 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>
2020-10-16 15:56:49 +00:00
Seungha Yang
cf5ef5635f 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>
2020-10-16 15:56:49 +00:00
Andrew Wesie
11353b3f6e codecparsers: h264parser: guard against ref_pic_markings overflow
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1703>
2020-10-16 15:11:26 +00:00
Tim-Philipp Müller
3f8d33abed hlssink2: fix and flesh out docs
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1699>
2020-10-16 13:37:18 +00:00
Stéphane Cerveau
cbd61e28b2 meson: update glib minimum version to 2.56
In order to support the symbol g_enum_to_string in various
project using GStreamer ( gst-validate etc.), the glib minimum
version should be 2.56.0.

Remove compat code as glib requirement
is now > 2.56

Version used by Ubuntu 18.04 LTS

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1695>
2020-10-16 09:16:34 +00:00
He Junyan
edc6e0fa1d d3d11: vp8dec: No need to check show_frame flag when output_picture.
The VP8 base class has already handled it for us.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1670>
2020-10-15 19:01:54 +00:00
He Junyan
6b3ff669ee codecs: vp8decoder: handle the show_frame check in base class.
Move the show_frame check from sub class to vp8 decoder's base class.
Calling the sub class' output_picture() function only when the frame
is displayed and marking the other automatically as decode only.

This is done to avoid logic and code repetition in subclasses.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1670>
2020-10-15 19:01:54 +00:00
Seungha Yang
a31a6608fe codecs: vp9decoder: handle the show_frame check in base class
Same as vp8 decoder update https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1670

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1697>
2020-10-15 18:36:54 +00:00
Nicolas Dufresne
0f5fc9646e v4l2codecs: vp8: Allow baseclass from skipping frames
In preparation for !1670, this will allow the base class from skipping frames
that should not be displayed. Previously it would complain about unordered
decoding taking place in the driver.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1696>
2020-10-15 13:15:02 -04:00
Nicolas Dufresne
a3e6d9fc24 v4l2codecs: decoder: Unmark previously pending request
requests are executed in order, so while dequeuing sink buffers for previous
request, also mark these request as no longer pending. This will allow reusing
the request later.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1696>
2020-10-15 13:15:02 -04:00
Nicolas Dufresne
2e3bc60422 v4l2codecs: decoder: Properly remove pending requests
Pass the pointer instead of NULL in order to find and remove properly any
pending request from the queue. This coding error was leading to use after
free in error and early exit cases.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1696>
2020-10-15 13:14:57 -04:00
He Junyan
4af33eda76 va: basedec: Create the other pool anyway.
Fix a bug in _create_other_pool(). The old way of checking the
base->other_pool make that other_pool never be changed until the
gst_va_base_dec_stop() to stop the current decoding context.
But in some stream, the resolution may change during the decoding
process, and we need to re-negotiate the buffer pool. Then, the
old other_pool can not be clean correctly and the new correct one
can not be created.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1692>
2020-10-15 13:22:33 +00:00
He Junyan
ee739003e8 va: basedec: Should unmap src frame when dst frame map failing.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1692>
2020-10-15 13:22:33 +00:00
He Junyan
21f4c31a52 va: bufferpool: use release_buffer to clean the mem.
The current bufferpool wastes all pre-allocate buffers when the
buffer pool is actived.
The pool->priv->size is 0 for va buffer pool. And every time, the
reset_buffer() will clean all mem and make the buffer size 0, that
can cache the gst_buffer in the buffer pool.
But when the buffer pool is activing, the default_start() just
allocate the buffer and release_buffer() immediately, all the pre
allocated buffers and surfaces are destroyed because of
gst_buffer_get_size (buffer) != pool->priv->size.
We need to use release_buffer() to do the clean job at the pool
start time.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1686>
2020-10-15 12:02:29 +08:00
Seungha Yang
634eb1fc38 h265parse: Don't enable passthrough by default
SEI messages contain various information which wouldn't be conveyed
by using upstream CAPS (HDR, timecode for example).

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1639>
2020-10-15 03:25:17 +09:00
Vivia Nikolaidou
94e1623434 cameracalibrate: Improve gst-inspect documentation
Thanks to @kazz_naka on Twitter

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1691>
2020-10-13 17:21:59 +03:00
Matthew Waters
2f29a4cde6 wpesrc: add some debug logging around WPEView creation/destruction
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1663>
2020-10-13 08:48:05 +00:00
Matthew Waters
da18a8d93d wpesrc: fix a memory leak of the bytes
free the previous GBytes if load-bytes is called multiple times
before view creation.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1663>
2020-10-13 08:48:05 +00:00
Matthew Waters
356fee4dd6 wpesrc: only create webview if not already created
e.g. _decide_allocation() can be called multiple times throughout the
element's lifetime and we only want to create the view once rather than
overwriting.

Fixes a leak of the WPEView under certain circumstances.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1663>
2020-10-13 08:48:05 +00:00
Matthew Waters
b84c8821de wpe: free a previous pending image/shm buffer
Don't blindly overwrite a possibly previously set buffer.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1663>
2020-10-13 08:48:05 +00:00
Jan Alexander Steffens (heftig)
4eeff95f92 srtsrc: Prevent delay from being negative
`delay` should be a GstClockTimeDiff since SRT time is int64_t.

All values are in local time so we should never see a srctime that's in
the future. If we do, clamp the delay to 0 and warn about it.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1674>
2020-10-12 12:58:22 +00:00
Jan Alexander Steffens (heftig)
ec11ad9d55 srtsrc: Don't calculate a delay if the srctime is 0
A zero srctime is a missing srctime. Apparently this can happen when
["the connection is not between SRT peers or if Timestamp-Based Packet
Delivery mode (TSBPDMODE) is not enabled"][1] so it may not apply to us,
but it's best to be defensive.

[1]: https://github.com/Haivision/srt/blob/v1.4.2/docs/API.md#sending-and-receiving

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1674>
2020-10-12 12:58:22 +00:00
Jan Alexander Steffens (heftig)
6b2fcb52e5 srtsrc: Defend against missing clock
If we don't have a clock, stop the source instead of asserting in
gst_clock_get_time. This can happen when the element is removed from the
pipeline while it's playing.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1674>
2020-10-12 12:58:22 +00:00
Marc Leeman
0be59181d7 rtpmanagerbad: remove duplicate parent declaration
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1689>
2020-10-12 13:56:50 +02:00
Tim-Philipp Müller
1ed969d276 rtmp2sink: fix since marker on new "stop-commands" property
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1687>
2020-10-12 11:55:46 +01:00
Víctor Manuel Jáquez Leal
18444fd993 va: basedec: copy frames logic to decide_allocation()
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1673>
2020-10-10 17:18:04 +00:00
Víctor Manuel Jáquez Leal
5658c4182a va: basedec: refactor the other video pool instantiation
Just a code clean up

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1673>
2020-10-10 17:18:04 +00:00
Víctor Manuel Jáquez Leal
a6398eca17 va: basedec: add gstvabasedec helper
This is a helper for all decoders.

It is not an abstract subclass, just merely a helper that avoids code
duplication among the decoders.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1673>
2020-10-10 17:18:04 +00:00
Víctor Manuel Jáquez Leal
a5dcb35ad6 va: vp8dec: add element documentation
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1673>
2020-10-10 17:18:04 +00:00
Víctor Manuel Jáquez Leal
a0ba72ed50 va: h264dec: set video alignment definition earlier
This patch renames need_cropping variable to need_videoalign which is clearer
with its function. And now GstVideoAlignment is part of GstVaH264Dec structure,
so it can be set earlier.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1673>
2020-10-10 17:18:04 +00:00
Olivier Crête
8a0d1d85cf dtlsconnection: Ignore OpenSSL system call errors
OpenSSL shouldn't be making real system calls, so we can safely
ignore syscall errors. System interactions should happen through
our BIO. So especially don't look at the system's errno, as it
should be meaningless.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1656>
2020-10-10 15:34:21 +00:00
Jan Alexander Steffens (heftig)
3ea6387f96 tests: svthevcenc: Fix test_encode_simple
Pick the same I420 format the other test use. Without this, the source
picks AYUV64, which fails.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1573>
2020-10-10 04:34:56 +00:00
Víctor Manuel Jáquez Leal
fbbf9c629a va: allocator: add _set_format() and _get_format()
Since allocators keep an available memory queue to reuse, video format and usage
hint are now persistant while allocator's memories are around.

This patch adds _set_format() and _get_format() for both VA allocators.

_set_format() validates if given format can be used or reused. If no allocated
surface previously it creates a dummy one to fetch its offsets and
strides. Updated info is returned to callee.

GstVaPool uses _set_format() at config to verify the allocator capacity and to
get the surfaces offsets and strides, which are going to be used by the video
meta.

Allocator extracted caps are compared with caps from config and if they have
different strides or offsets, force_videometa is set.

A new bufferpool method gst_va_pool_requires_video_meta() is added return the
value of force_videometa. This value is checked in order to know if decoders
need to copy the surface if downstream doesn't announce video meta support.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1667>
2020-10-09 16:34:57 +02:00
Víctor Manuel Jáquez Leal
482e93b4d8 va: caps: added gst_caps_is_raw()
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1667>
2020-10-09 16:34:57 +02:00
Víctor Manuel Jáquez Leal
6a8fa67f42 va: pool: call parent's start() method
Without preallocating buffers and memories a deadlock in pool allocator is
highly probably since it might hit the case were buffer is returned to the pool
but their memories are still hold by a copy downstream, without other
preallocated buffers available.

This kind of a hack, where buffer_reset() follow the normal path if it's called
from start().

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1667>
2020-10-09 16:34:57 +02:00
Víctor Manuel Jáquez Leal
8f936baffe va: pool: fix log's object
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1667>
2020-10-09 16:34:57 +02:00
Víctor Manuel Jáquez Leal
26ee5b1e4e va: allocator: remove noisy log message
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1667>
2020-10-09 16:34:57 +02:00
Víctor Manuel Jáquez Leal
9254045b3e va: allocator: add a todo for dmabuf_memories_setup()
It would be nice to add a surface pool for this type of surface allocation in
order to have a better control of them.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1667>
2020-10-09 16:34:56 +02:00
Víctor Manuel Jáquez Leal
a299dff9d4 va: allocator: add a surface counter
Every time a new surface is created the counter increases by one, and when it is
destroyed (or will be destroyed in case of GstVaAllocator), the counter is
decreased. Then, at allocator dispose, it is warning if the counter is not zero.

This counter will be also used to check if the allocator can change its
configuration if the counter is zero or can not.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1667>
2020-10-09 16:34:56 +02:00
Víctor Manuel Jáquez Leal
30281af83f va: allocator: remove GstVideoInfo from GstVaBufferSurface
Don't store it them anymore since it is related with the negotiated stream and
not the concrete buffer.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1667>
2020-10-09 16:34:56 +02:00
Víctor Manuel Jáquez Leal
67eb0a9440 va: remove GstVideoInfo parameter from _get_surface() functions
There shouldn't be need to retrieve GstVideoInfo per buffer or memory since it
is the same for all the negotiated stream.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1667>
2020-10-09 16:34:56 +02:00