Marek Vasut [Wed, 8 Sep 2021 03:28:22 +0000 (05:28 +0200)]
gsth264parser: Fix handling of NALs with emulation byte set
In case a set of NALs with emulation_prevention_three_byte is decoded using
hardware decoder like Hantro G1, wrong struct v4l2_ctrl_h264_decode_params
.dec_ref_pic_marking_bit_size is passed into the kernel, which results in
decoding artifacts. Subtract the number of emulation three bytes from the
.dec_ref_pic_m->bit_size to get the correct bit size and avoid having these
artifacts. Apply the exact same fix to slice->pic_order_cnt_bit_size as well.
The following NALs (7, 8, 6, 5) decode with artifacts,
.dec_ref_pic_marking_bit_size is set to 10 without this patch.
00000000 00 00 00 01 27 4d 00 20 89 8b 60 3c 04 bf 2e 02 |....'M. ..`<....|
00000010 d4 18 04 18 c0 c0 01 77 00 00 5d c1 7b df 05 00 |.......w..].{...|
00000020 00 00 01 28 ee 1f 20 00 00 01 06 05 10 b9 ed b9 |...(.. .........|
00000030 30 5d 21 4b 71 83 71 2c 10 a3 14 bb 29 80 00 00 |0]!Kq.q,....)...|
00000040 01 25 b8 00 05 00 00 03 03 7f fa 78 1e e7 fd fe |.%.........x....|
^^^^^^^^^^^^--- emulation 3 byte
00000050 b4 62 7a 31 ff 7d 81 fd 26 d8 62 b6 d6 25 46 ae |.bz1.}..&.b..%F.|
The following NALs (7, 8, 6, 5) decode fine,
.dec_ref_pic_marking_bit_size is set to 2 without this patch.
00000000 00 00 00 01 27 4d 00 20 89 8b 60 3c 04 bf 2e 02 |....'M. ..`<....|
00000010 d4 18 04 18 c0 c0 01 77 00 00 5d c1 7b df 05 00 |.......w..].{...|
00000020 00 00 01 28 ee 1f 20 00 00 01 06 05 10 b9 ed b9 |...(.. .........|
00000030 30 5d 21 4b 71 83 71 2c 10 a3 14 bb 29 80 00 00 |0]!Kq.q,....)...|
00000040 01 25 b8 00 04 c0 00 03 7f fa 78 1e e7 fd fe b4 |.%........x.....|
00000050 62 7a 31 ff 7d 81 fd 26 d8 62 b6 d6 25 46 ae ce |bz1.}..&.b..%F..|
Fixes:
d0d65fa875 ("codecparsers: h264: record dec_ref_pic_marking() size")
Fixes:
0cc7d6f093 ("codecparsers: h264: record pic_order_cnt elements size")
Signed-off-by: Marek Vasut <marex@denx.de>
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2517>
Aaron Boxer [Fri, 3 Sep 2021 18:57:09 +0000 (14:57 -0400)]
gsth264parser: reject memory management control op greater than 6
This prevents assertion from being thrown in
gst_h264_dpb_perform_memory_management_control_operation
if corrupt NAL has a control op greater than 6
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2508>
Ung, Teng En [Tue, 24 Aug 2021 09:59:59 +0000 (09:59 +0000)]
msdk: Adjust the plugin and factories description based on MFX_VERSION.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2485>
Mathieu Duponchelle [Wed, 8 Sep 2021 15:32:30 +0000 (17:32 +0200)]
vulkan: don't link to XOpenDisplay in documentation
hotdoc doesn't know about that symbol
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2518>
Jan Schmidt [Mon, 30 Aug 2021 13:26:39 +0000 (23:26 +1000)]
mpeg2enc: Only allow 1 pending frame for encoding
Having an unlimited input queue is very bad if the
encoder can't run at real-time. Eventually it will
consume all RAM. I don't really see any reason to
have more than 1 outstanding encoded frame, so
remove the queue and limit things to 1 pending frame.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2499>
Thibault Saunier [Wed, 1 Sep 2021 21:35:45 +0000 (17:35 -0400)]
wpe: Fix race condition on teardown
There was a race when going to PAUSED while pushing a buffer to the
pipeline process (where we weren't even cancelling anything).
This rework base all the cancellation around the GCancellable
"cancelled" signal trying to ensure that the streaming thread will not
block once a cancel operation happens.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2504>
Thibault Saunier [Wed, 1 Sep 2021 21:26:04 +0000 (17:26 -0400)]
wpe: Use the new element.get_current_running_time API
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2504>
Thibault Saunier [Wed, 1 Sep 2021 21:24:45 +0000 (17:24 -0400)]
wpe: Mark first buffer as starting at 0
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2504>
Seungha Yang [Thu, 2 Sep 2021 13:06:52 +0000 (22:06 +0900)]
videoparseutils: Fix for wrong CEA708 minimum size check
The minimum possible size of valid CEA708 data is 3 bytes, not 7 bytes
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2505>
Philippe Normand [Sun, 29 Aug 2021 10:04:17 +0000 (11:04 +0100)]
wpevideosrc: Uniformise default value for draw-background property
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2498>
Philippe Normand [Sun, 29 Aug 2021 09:30:53 +0000 (10:30 +0100)]
wpevideosrc: Implement basic heuristic for raw caps negotiation
Before this patch raw caps could be negotiated already with a capsfilter, but in
cases where wpesrc is being auto-plugged this approach can't be used.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2498>
Philippe Normand [Sun, 29 Aug 2021 09:28:57 +0000 (10:28 +0100)]
wpevideosrc: Ensure debug category is set
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2498>
Mathieu Duponchelle [Thu, 15 Jul 2021 19:10:14 +0000 (21:10 +0200)]
cccombiner: fix scheduling with interlaced video buffers
The initial code was written with the misunderstanding that
IS_TOP_FIELD indicated that an interlaced buffer contained
a top field, not that it contained only a top field
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2413>
Nicolas Dufresne [Fri, 27 Aug 2021 19:41:32 +0000 (15:41 -0400)]
Revert "kmssink: Fix fallback path for driver not able to scale scenario"
This reverts commit
d2a7b763bef3ca51f0c84cdac52eeed85b0db8fb.
After this change, non-scaled rendered were not centred as expected.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2496>
Mengkejiergeli Ba [Fri, 20 Aug 2021 05:28:51 +0000 (13:28 +0800)]
codecs: av1dec: Fix to output frame with highest spatial layer
During the output process, if there are multiple frames in a TU (i.e. multi-spatial
layers case), only one frame with the highest spatial layer id should be selected
according to av1 spec. The highest spatial layer id is obtained from idc value of
the operating point.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2475>
Alex Ashley [Tue, 24 Aug 2021 13:33:42 +0000 (14:33 +0100)]
dashdemux: copy ContentProtection element including xml namespaces
Commit
bc09d8cc changed gstmpdparser to put the entire
<ContentProtection> element in the "value" field, so that DRMs
other than PlayReady could make use of the data inside this
element.
However, the data in the "value" field does not include any
XML namespace declarations that are used within the element. This
causes problems for a namespace aware XML parser that wants to
make use of this data.
This commit modifies the way the XML is converted to a string
so that XML namespaces are preserved in the output.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2487>
Vivia Nikolaidou [Thu, 26 Aug 2021 18:26:01 +0000 (21:26 +0300)]
errorignore: Add ignore-eos mode
It's otherwise very complicated to ignore GST_FLOW_EOS without a
ghostpad's chain function to rewrite.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2492>
Brad Hards [Fri, 27 Aug 2021 07:25:04 +0000 (17:25 +1000)]
gsth264parser: fix typo in debug message
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2493>
Mathieu Duponchelle [Thu, 26 Aug 2021 02:12:07 +0000 (04:12 +0200)]
timecodestamper: add support for closedcaption input
Some closedcaption elements like sccenc except input buffers
to have timecode metas. The original use case is to serialize
closed captions extracted from a video stream, in that case
ccextractor copies the video time code metas to the closed
caption buffers, but no such mechanism exists when creating
a CC stream ex nihilo.
Remedy that by having timecodestamper accept closedcaption
input caps, as long as they have a framerate.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2490>
Aaron Boxer [Tue, 6 Jul 2021 16:31:42 +0000 (12:31 -0400)]
aes: add aes encryption and decryption elements
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1505>
Johan Sternerup [Mon, 10 May 2021 10:02:20 +0000 (12:02 +0200)]
webrtcbin: Return typed "sctp-transport"
With GstWebRTCSCTPTransport type exposed we can now define
"sctp-transport" property as being of this type.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2214>
Johan Sternerup [Fri, 7 May 2021 06:12:25 +0000 (08:12 +0200)]
webrtc: Split sctptransport into lib and implementation parts
GstWebRTCSCTPTransport is now made into into an abstract base class
that only contains property specifications matching the
RTCSctpTransport interface of the W3C WebRTC specification, see
https://w3c.github.io/webrtc-pc/#rtcsctptransport-interface. This
class is put into the WebRTC library to expose it for applications and
to allow for generation of bindings for non-dynamic languages using
GObject introspection.
The actual implementation is moved to the subclass WebRTCSCTPTransport
located in the WebRTC plugin.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2214>
Johan Sternerup [Mon, 3 May 2021 08:45:42 +0000 (10:45 +0200)]
webrtcbin: Expose SCTP Transport
Being able to access the SCTP Transport object from the application
means the application can access the associated DTLS Transport object
and its ICE Transport object. This means we can observe the ICE state
also for a data-channel-only session. The collated
ice-connection-state on webrtcbin only includes the ICE Transport
objects that resides on the RTP transceivers (which is exactly how it
is specified in
https://w3c.github.io/webrtc-pc/#rtciceconnectionstate-enum).
For the consent freshness functionality (RFC 7675) to work the ICE
state must be accessible and consequently the SCTP transport must be
accessible for enabling consent freshness checking for a
data-channel-only session.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2214>
Tim-Philipp Müller [Tue, 20 Apr 2021 19:04:33 +0000 (20:04 +0100)]
openh264enc: fix broken header AU emission by base class
This encoder advertises alignment=au as output format, which means
each output frame should contain a full decodable access unit.
The video encoder base class is not aware of our output alignment
and will output spurious buffers with just the SPS/PPS inside when
we call gst_video_encoder_set_headers(), which is broken because
each buffer is supposed to contain a full decodable access unit
in our case.
Just don't tell the base class about our headers, they will be
sent at the beginning of each IDR frame anyway.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2178>
Tim-Philipp Müller [Tue, 20 Apr 2021 18:43:53 +0000 (19:43 +0100)]
openh264enc: fix caps and header buffer leak
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2178>
Tim-Philipp Müller [Tue, 20 Apr 2021 18:11:12 +0000 (19:11 +0100)]
openh264enc: fix broken sps/pps header generation
This was putting a truncated SPS into the initial header instead
of the PPS because it was always reading from the beginning of the
bitstream buffer (pBsBuf) and not from the offset where the current
NAL is at in the bitstream buffer (psBsBuf + nal_offset).
This was broken in commit
17113695.
Fixes https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/issues/1576
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2178>
Seungha Yang [Sat, 21 Aug 2021 15:33:58 +0000 (00:33 +0900)]
d3d11bufferpool: Hide buffer_size field from header
User can get the required buffer size by using buffer pool config.
Since d3d11 implementation is a candidate for public library in the future,
we need to hide everything from header as much as possible.
Note that the total size of allocated d3d11 texture memory by GPU is not
controllable factor. It depends on hardware specific alignment/padding
requirement. So, GstD3D11 implementation updates actual buffer size
by allocating D3D11 texture, since there's no way to get CPU accessible
memory size without allocating real D3D11 texture.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2482>
Seungha Yang [Fri, 20 Aug 2021 17:20:11 +0000 (02:20 +0900)]
nvcodec: Fix various typos
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2481>
Seungha Yang [Fri, 20 Aug 2021 17:10:37 +0000 (02:10 +0900)]
nvcodec: Get rid of G_GNUC_INTERNAL
Our default symbol visibility is hidden, so G_GNUC_INTERNAL
is pointless
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2481>
Nicolas Dufresne [Thu, 19 Aug 2021 20:45:18 +0000 (16:45 -0400)]
v4l2codecs: h264: Fix split field handling
Split fields ends up on multiple picture and requires accessing the
other_field to complete the information (POC).
This also cleanup the DPB from non-reference (was not useful) and skips
properly merge field instead of keeping them duplicated. This fixes most
of interlace decoding seen in fluster.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2474>
Nicolas Dufresne [Thu, 19 Aug 2021 15:40:22 +0000 (11:40 -0400)]
v4l2codec: h264: Implement support for split fields
When a frame is composed of two fields, the base class now split the
picture in two. In order to support this, we need to ensure that picture
buffer is held in VB2 queue so that the second field get decoded into
it. This also implements the new_field_picture() virtual and sets the
previous request on the new picture.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2474>
Nicolas Dufresne [Fri, 20 Aug 2021 15:23:57 +0000 (11:23 -0400)]
v4l2codecs: h264: Fix filling weight factors
This was a typo, the wrong index was used to set l1 weight (b-frames).
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2480>
Edward Hervey [Fri, 20 Aug 2021 12:34:53 +0000 (14:34 +0200)]
dashdemux: Properly initalize GError
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2476>
Seungha Yang [Thu, 19 Aug 2021 12:56:05 +0000 (21:56 +0900)]
mfvideosrc: Fix for negative MF stride
Negative stride value can be used in MediaFoundation to inform
whether memory layout is top-down or bottom-up manner. Note that
negative stride is allowed only for RGB, system memory.
See also
https://docs.microsoft.com/en-us/windows/win32/medfound/image-stride
Fixes: https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/issues/1646
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2473>
Nicolas Dufresne [Wed, 18 Aug 2021 15:14:37 +0000 (11:14 -0400)]
v4l2slh264dec: Fix slice header bit size calculation
The emulation bytes need to be removed as bytes, not bit. This fixes
decoding issues with files that have emulation bytes with the Cedrus
driver.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2471>
Víctor Manuel Jáquez Leal [Thu, 12 Aug 2021 12:08:19 +0000 (14:08 +0200)]
example: va: Add skin tone enhancement.
If camera is used as input stream and skin tone parameter is available
in vapostproc, and no random changes are enabled, the skin tone will
be enabled.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2470>
Víctor Manuel Jáquez Leal [Tue, 17 Aug 2021 12:04:41 +0000 (14:04 +0200)]
vapostproc: Use vapostproc as debug category name.
Otherwise is difficult to remember the different name.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2470>
Víctor Manuel Jáquez Leal [Thu, 12 Aug 2021 11:54:34 +0000 (13:54 +0200)]
examples: va: Add random cropping.
And remove unused caps filter.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2443>
Víctor Manuel Jáquez Leal [Wed, 28 Jul 2021 11:04:50 +0000 (13:04 +0200)]
vapostproc: Disable cropping in pass-through mode.
Originally, if a buffer arrives with crop meta but downstream doesn't
handle crop allocation meta, vapostproc tried to reconfigure itself to
non pass-through mode automatically. Sadly, this behavior was based on
the wrong assumption that propose_allocation() vmethod would bring
downstream allocation query, but it is not.
Now, if vapostproc is in pass-through mode, the cropping is passed to
downstream. Pass-through mode can be disabled via a parameter.
Finally, if pass-through mode isn't enabled, it's assumed the buffer
is going to be processed and, if cropping, downstream already
negotiated the cropped frame size, thus it's required to do the
cropping inside vapostproc to avoid artifacts because of the size of
downstream allocated buffers.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2443>
Víctor Manuel Jáquez Leal [Tue, 17 Aug 2021 12:54:21 +0000 (14:54 +0200)]
vapostproc: Update filters update_properties().
Right after instantiating the VA filter and changing the element
state, rebuild the image filters.
This will fix a regression from
f20b3b815, where properties in a
gst-launch pipeline are not applied.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2443>
Sebastian Dröge [Wed, 18 Aug 2021 06:13:45 +0000 (09:13 +0300)]
decklinkvideosrc: Fix PAL/NTSC widescreen autodetection when switching back to non-widescreen
Previously it would only switch to widescreen but never back.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2469>
Mengkejiergeli Ba [Tue, 20 Jul 2021 10:15:11 +0000 (18:15 +0800)]
msdkvpp: Fix frc from lower fps to higher fps
There are three framerate conversion algorithms described in
<https://github.com/Intel-Media-SDK/MediaSDK/blob/master/doc/mediasdk-man.md>,
interpolation is not implemented so far and thus distributed timestamp algorihtm
is considered to be more practical which evenly distributes output timestamps
according to output framerate. In this case, newly generated frames are inserted
between current frame and previous one, timestamp is calculated by msdk API.
This implementation first pushes newly generated buffers(outbuf_new) forward and
the current buffer(outbuf) is handled at last round by base transform automatically.
A flag "create_new_surface" is used to indicate if new surfaces have been generated
and then push new outbuf forward accordingly.
Considering the upstream element may not be the msdk element, it is necessary to
always set the input surface timestamp as same as input buffer's timestamp and
convert it to msdk timestamp.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2418>
Matthew Waters [Thu, 6 May 2021 12:22:12 +0000 (22:22 +1000)]
webrtc: improve matching on the correct jitterbuffer
The mapping between an RTP session and the SDP m= line is not always the
same, especially when BUNDLEing is used.
This causes a failure in a specific case where if when bundling,
if mline 0 is a data channel, and mline 1 an audio/video section,
then retrieving the transceiver at mline 0 (rtp session used) will fail
and cause an assertion.
This fix is actually potentially a regression for cases where the remote
part does not provide the a=ssrc: media level SDP attributes as is now
becoming common, especially when simulcast is involved.
The correct fix actually requires reading out header extensions as used
with bundle for signalling in the actual data, what media and therefore
transceiver is being used.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2467>
He Junyan [Mon, 16 Aug 2021 05:45:39 +0000 (13:45 +0800)]
va: Use GST_CAPS_FEATURE_MEMORY_VA to replace "memory:VAMemory".
"memory:VAMemory" is a commonly used string which notates our VA-kind
memory type. We now used a definition in va lib to replace the simply
string usage.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2466>
He Junyan [Mon, 16 Aug 2021 05:32:51 +0000 (13:32 +0800)]
va: Use MEMORY_DMABUF definition to replace "memory:DMABuf" strings.
GST_CAPS_FEATURE_MEMORY_DMABUF is already a common definition, we should
just use it rather than use the "memory:DMABuf" strings by ourselves.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2466>
Thibault Saunier [Mon, 9 Aug 2021 23:02:56 +0000 (19:02 -0400)]
fdkaacdec: Add Converter class to hint gst-validate
fdkaacdec have minimal conversion capability, adding the Converter class allow
gst-validate to behave properly and not spit an error when it notice that the
number of channels or rate miss-match in and out.
Same logic as with opusdec, see: https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/1142>
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2462>
Seungha Yang [Wed, 9 Jun 2021 14:29:43 +0000 (23:29 +0900)]
wasapi2: Increase rank to primary + 1
wasapi2 plugin should be preferred than old wasapi plugin if available because:
* wasapi2 supports automatic stream routing, and it's highly recommended
feature for application by MS. See also
https://docs.microsoft.com/en-us/windows/win32/coreaudio/automatic-stream-routing
* This implementation must be various COM threading issue free by design
since wasapi2 plugin spawns a new dedicated COM thread and all COM objects'
life-cycles are managed correctly.
There are unsolved COM issues around old wasapi plugin. Such issues are
very tricky to be solved unless old wasapi plugin's threading model
is re-designed.
Note that, in case of UWP, wasapi2 plugin's rank is primary + 1 already
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2314>
Mathieu Duponchelle [Thu, 12 Aug 2021 18:39:24 +0000 (20:39 +0200)]
ccconverter: fix overflow when not doing framerate conversion
When converting from one framerate to another, counters are
reset periodically, however when not converting they never are
and can_genearte_output ends up making overflow-prone calculations
with large values for input_frames and output_frames.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2465>
Sebastian Dröge [Thu, 12 Aug 2021 12:26:27 +0000 (15:26 +0300)]
webrtcbin: Don't assume that non-audio medias are video medias when creating transceivers
And print the unknown media kind in the logs.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2464>
Sebastian Dröge [Thu, 12 Aug 2021 12:25:50 +0000 (15:25 +0300)]
webrtcbin: Use the correct media for deciding the media kind when creating the transceiver from the SDP
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2464>
He Junyan [Thu, 29 Jul 2021 13:30:32 +0000 (21:30 +0800)]
codecs: h264dec: Output the picture directly if already a frame.
We forget one case that is the frame and field pictures may be mixed
together. For this case, the dpb is interlaced while the last picture
may be a complete frame. We do not need to cache that complete picture
and should output it directly.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2448>
He Junyan [Fri, 6 Aug 2021 09:11:55 +0000 (17:11 +0800)]
va: caps: Make the template raw video caps classified by features.
The current output of raw video caps is not good. When we have multi
profiles and each profile support different formats, the output of
gst-inspect may like:
SRC template: 'src'
Availability: Always
Capabilities:
video/x-raw(memory:VAMemory)
width: [ 1, 16384 ]
height: [ 1, 16384 ]
format: NV12
video/x-raw
width: [ 1, 16384 ]
height: [ 1, 16384 ]
format: NV12
video/x-raw(memory:VAMemory)
width: [ 1, 16384 ]
height: [ 1, 16384 ]
format: P010_10LE
video/x-raw
width: [ 1, 16384 ]
height: [ 1, 16384 ]
format: P010_10LE
video/x-raw(memory:VAMemory)
width: [ 1, 16384 ]
height: [ 1, 16384 ]
format: P012_LE
video/x-raw
width: [ 1, 16384 ]
height: [ 1, 16384 ]
format: P012_LE
The gst_caps_simplify does not classify the caps by same features, but
just leave them interweaved. We need to handle them manually here, the
result should be:
SRC template: 'src'
Availability: Always
Capabilities:
video/x-raw
width: [ 1, 16384 ]
height: [ 1, 16384 ]
format: { (string)P010_10LE, (string)P012_LE, (string)NV12 }
video/x-raw(memory:VAMemory)
width: [ 1, 16384 ]
height: [ 1, 16384 ]
format: { (string)P010_10LE, (string)P012_LE, (string)NV12 }
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2456>
Víctor Manuel Jáquez Leal [Tue, 27 Jul 2021 11:22:02 +0000 (13:22 +0200)]
vapostproc: Inherit from GstVaBaseTransform.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2442>
Víctor Manuel Jáquez Leal [Wed, 17 Feb 2021 16:15:22 +0000 (17:15 +0100)]
va: Add base transform class.
This base transform class is a derivable class for VA-based filters,
for example vapostproc right now, but it will be used also for
future elements such as vadeinterlace.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2442>
Víctor Manuel Jáquez Leal [Tue, 27 Jul 2021 11:03:37 +0000 (13:03 +0200)]
va: pool: Add gst_va_pool_new_with_config().
It is a function helper.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2442>
Seungha Yang [Mon, 9 Aug 2021 17:48:01 +0000 (02:48 +0900)]
d3d11window: Misc code cleanup
* Remove unnecessary upcasting. We are now dealing with C++ class objects
and don't need explicit C-style casting in C++ world
* Use helper macro IID_PPV_ARGS() everywhere. It will make code
a little short.
* Use ComPtr smart pointer instead of calling manual IUnknown::Release()
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2461>
Seungha Yang [Mon, 9 Aug 2021 17:48:45 +0000 (02:48 +0900)]
d3d11compositor: Fix indent
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2461>
Thibault Saunier [Fri, 28 May 2021 21:36:15 +0000 (17:36 -0400)]
openh264: Respect level set downstream
We were not specifying the requested level to openh264 meaning that
it was choosing anything and was not respecting what was specified\
downstream
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2289>
He Junyan [Wed, 4 Aug 2021 07:02:01 +0000 (15:02 +0800)]
x265: Fix a deadlock when failing to create the x265enc.
The GST_ELEMENT_ERROR will call the gst_object_get_path_string and
use gst_object_get_parent to get the full object path name, which
needs to lock the object. But we are already in a locked context and
so this will cause a deadlock, the pipeline can not exit normally.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2451>
R S Nikhil Krishna [Thu, 22 Jul 2021 20:58:02 +0000 (20:58 +0000)]
rtmpsrc: mention setting librtmp flags in docs
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2424>
Mathieu Duponchelle [Thu, 5 Aug 2021 21:11:26 +0000 (23:11 +0200)]
mpeg2enc: fix interlace-mode detection
Previously, the code was always assuming progressive input,
fix this by looking at the caps.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2455>
Tim-Philipp Müller [Sun, 23 May 2021 18:15:25 +0000 (19:15 +0100)]
Use g_memdup2() where available and add fallback for older GLib versions
g_memdup() is deprecated since GLib 2.68 and we want to avoid
deprecation warnings with recent versions of GLib.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2280>
Sebastian Dröge [Thu, 5 Aug 2021 10:02:00 +0000 (13:02 +0300)]
timecodestamper: Fix latency calculation
The LTC extra latency is in ms already and not in frames, so multiplying
with the framerate will end up with a wrong number.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2453>
Haihao Xiang [Sat, 17 Jul 2021 16:51:04 +0000 (00:51 +0800)]
msdk: make sure child context is destroyed first
The parent context shares some resources with child context, so the
child context should be destroyed first, otherwise the command below
will trigger a segmentation fault
$> gst-launch-1.0 videotestsrc num-buffers=100 ! msdkh264enc ! \
msdkh264dec ! fakesink videotestsrc num-buffers=50 ! \
msdkh264enc ! msdkh264dec ! fakesink
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2435>
Seungha Yang [Mon, 2 Aug 2021 07:22:06 +0000 (16:22 +0900)]
d3d11videosink: Fix warning around GstVideoOverlay::expose()
When expose() is called, d3d11videosink needs to redraw using
cached buffer, so gst_d3d11_window_render() should allow null buffer.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2450>
Seungha Yang [Fri, 30 Jul 2021 16:05:47 +0000 (01:05 +0900)]
d3d11videosink: Forward navigation event without modification
Current implementation for translating native coordinate and
video coordinate is very wrong because d3d11videosink doesn't
understand native HWND's coordinate. That should be handled
by GstD3D11Window implementation as an enhancement.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2450>
Seungha Yang [Fri, 30 Jul 2021 15:59:14 +0000 (00:59 +0900)]
d3d11videosink: Add support for GstVideoOverlay::set_render_rectangle
Inspired by an MR https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2382
The idea is that we can make use of MoveWindow() in WIN32 d3d11window
implementation safely because WIN32 d3d11window implementation creates
internal HWND even when external HWND is set and then subclassing is used to
draw on internal HWND in any case. So the coordinates passed to MoveWindow()
will be relative to parent HWND, and it meets well to the concept of
set_render_rectangle().
On MoveWindow() event, WM_SIZE event will be generated by OS and then
GstD3D11WindowWin32 implementation will update render area including swapchain
correspondingly, as if it's normal window move/resize case.
But in case of UWP (CoreWindow or SwapChainPanel), we need more research to
meet expected behavior of set_render_rectangle()
Fixes: https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/issues/1416
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2450>
He Junyan [Thu, 29 Jul 2021 10:05:35 +0000 (18:05 +0800)]
va: vp8: fix the overflow in _fill_quant_matrix().
The gint8 of qi and qi_base may overflow when calculation the matrix
parameters and change the decoding result.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2447>
Seungha Yang [Sun, 20 Jun 2021 15:19:17 +0000 (00:19 +0900)]
d3d11: Disable packed and subsampled YUV formats
Direct3D11 sampler doesn't support them very well, and conversion
outputs usually result in poor visual quality with our shader code.
Should disable support for such formats for now
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2344>
Jan Alexander Steffens (heftig) [Mon, 26 Jul 2021 14:43:47 +0000 (16:43 +0200)]
tsdemux: Notify when ignore_pcr is set
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2437>
Mathieu Duponchelle [Tue, 27 Jul 2021 21:53:06 +0000 (23:53 +0200)]
cccombiner: fix CDP padding detection
While a cc_data_pkt with cc_valid 0 should be considered padding,
it might be followed up by valid DTVCC packets, and should not
cause the whole CDP packet to get discarded.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2440>
He Junyan [Tue, 27 Jul 2021 04:51:08 +0000 (12:51 +0800)]
codecs: h264dec: Improve the find_first_field_picture().
We need to consider the first field of the last picture when the
last picture can not enter the DPB.
Another change is, when prev field's frame_num is not equal to the
current field's frame_num, we should also return FASLE because it
is also a case of losing some field.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2430>
He Junyan [Tue, 27 Jul 2021 04:16:13 +0000 (12:16 +0800)]
codecs: h264dec: consider the last field when add picture to DPB.
There are cases that the first field of the last picture is not a
ref but the second field is a ref. We need to add both of them
because the bumping always needs a complete frame in the DPB.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2430>
He Junyan [Tue, 27 Jul 2021 02:51:03 +0000 (10:51 +0800)]
codecs: h264dec: Consider the field case when directly output.
For interlaced streams, it is also possible that the last frame is
not able to be inserted into DPB when the DPB is full and the last
frame is a non ref. For this case, we need to hold a extra ref for
the first field of the last frame and wait for the complete frame
with both top and bottom fields. For the progressive stream, the
behaviour is unchanged.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2430>
He Junyan [Sun, 25 Jul 2021 17:16:34 +0000 (01:16 +0800)]
codecs: h264dec: Fix a error print of dpb_add.
When the dpb is interlaced, the max size should be 2*dpb->max_num_frames,
correcting the error print info for that.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2430>
Seungha Yang [Wed, 28 Jul 2021 07:11:36 +0000 (16:11 +0900)]
decklink: Don't print error for dlopen failure
This is not a fatal error on systems without decklink
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2441>
Imanol Fernandez [Tue, 27 Jul 2021 10:21:41 +0000 (12:21 +0200)]
player: Add static keyword to _config_quark_table
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2439>
Matthew Waters [Tue, 27 Jul 2021 04:52:38 +0000 (14:52 +1000)]
applemedia: silence a couple of MoltenVK warnings
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2433>
Nicolas Dufresne [Tue, 27 Jul 2021 15:49:47 +0000 (11:49 -0400)]
debugutils: Only proxy the properties once
The needed once call was removed accidently during porting. This was catch by
the CI as memory leaks.
Related to !2426
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2438>
Edward Hervey [Tue, 27 Jul 2021 10:13:43 +0000 (12:13 +0200)]
mxf: Handle D10 "picture only" variant
Fixes https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/issues/80
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2436>
Víctor Manuel Jáquez Leal [Sat, 24 Jul 2021 11:19:39 +0000 (13:19 +0200)]
va: filter: refactor convert_surface() to process()
The idea of this change is to add, in the future,
process_with_generator(), when multiple input surfaces are processed,
for blending.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2431>
Víctor Manuel Jáquez Leal [Sun, 18 Jul 2021 10:46:08 +0000 (12:46 +0200)]
va: filter: Refactor set_formats() to set_video_info().
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2431>
Víctor Manuel Jáquez Leal [Sun, 18 Jul 2021 15:13:16 +0000 (17:13 +0200)]
vapostproc: Don't add video alignment option in buffer pool.
vapostproc will not call gst_buffer_pool_config_set_video_alignment(),
thus this option is not required.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2431>
Edward Hervey [Tue, 27 Jul 2021 07:37:49 +0000 (09:37 +0200)]
mxfvc3: Also accept clip-wrapped vc-3
We can now handle this fine
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2434>
Edward Hervey [Tue, 27 Jul 2021 05:59:52 +0000 (07:59 +0200)]
mxfdemux: Handle EOS with non-frame wrapping
When reaching the end of non-frame wrapping track in pull mode, we want to force
the switch to the next non-eos pad. This is similar to when we exceed the
maximum drift.
Fixes issues on EOS where not everything would be drained out and stray errors
would pop out.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2434>
Edward Hervey [Sun, 25 Jul 2021 05:52:06 +0000 (07:52 +0200)]
mxfdemux: More granular interleaved content handling
An interleave of 500ms can be way too big for some downstream queueing
elements. Instead use a smaller 100ms interleave and silence the various
warnings about resyncing (it's normal)
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2434>
Philippe Normand [Fri, 23 Jul 2021 08:36:10 +0000 (09:36 +0100)]
debugutils: De-duplicate proxy_properties function to a new utils module
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2426>
Seungha Yang [Mon, 7 Jun 2021 16:40:34 +0000 (01:40 +0900)]
Introduce Steinberg ASIO (Audio Streaming Input/Output) plugin
Adds a new plugin for ASIO devices.
Although there is a standard low-level audio API, WASAPI, on Windows,
ASIO is still being broadly used for audio devices which are aiming to
professional use case. In case of such devices, ASIO API might be able
to show better quality and latency performance depending on manufacturer's
driver implementation.
In order to build this plugin, user should provide path to
ASIO SDK as a build option, "asio-sdk-path".
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2309>
Seungha Yang [Sun, 6 Jun 2021 13:32:08 +0000 (22:32 +0900)]
audiolatency: Expose samplesperbuffer property
... for user to be able to set the number of required samples.
For instance, our default value is 240 samples
(about 5ms latency in case that sample rate is 48000), which might
be larger than actual buffer size of audio capture device.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2307>
He Junyan [Fri, 23 Jul 2021 14:02:05 +0000 (22:02 +0800)]
codecs: h264decoder: let print_ref_pic_list_b print the correct list name.
The print_ref_pic_list_b now not only needs to trace the ref_pic_list_b0/1,
but also need to trace the ref_frame_list_0_short_term. We need to pass the
name directly to it rather than an index to refer to ref_pic_list_b0/1.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2425>
He Junyan [Fri, 23 Jul 2021 04:31:17 +0000 (12:31 +0800)]
codecs: h264dec: Fix a typo in construct_ref_field_pic_lists_b.
The array sort of ref_frame_list_0_short_term has some typo. The
typo makes this list not in the POC ascend order and generate wrong
decoding result for interlaced streams.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2425>
Devarsh Thakkar [Thu, 15 Jul 2021 12:12:37 +0000 (05:12 -0700)]
kmssink: Fix fallback path for driver not able to scale scenario
When driver return error on update plane request, kmssink
disables the scaling and retries plane update.
While doing so kmssink was matching the source rectangle dimensions
to the target rectangle dimensions which were calculated
as per scaling but this is incorrect, instead what we want here is
that target rectangle dimensions should match the source rectangle
dimensions as scaling is disabled now and so we match result
rectangle dimensions with source rectangle dimensions.
While at it, also match the result rectangle coordinates for
horizontal and vertical offsets with source rectange coordinates,
as since there is no scaling being done so no recentering is
required.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2415>
He Junyan [Fri, 23 Jul 2021 08:49:49 +0000 (16:49 +0800)]
videoparsers: vp9: Need to process the first frame even not key.
Some cut VP9 streams begin with a non key frame. The current code
just bail out the parse_process_frame() if not a key frame. Because
of this, we do not set the valid caps before we push the data of the
first frame(even this first frame will be discarded by the downstream
decoder because it is not a key frame).
The pipeline such as:
gst-launch-1.0 filesrc location=some.ivf ! ivfparse ! vp9parse !
vavp9dec ! fakesink
will get a negotiation error and the pipeline can not continue. The
correct behaviour should be: the decoder discard the first frame and
continue to decode later frames successfully.
So, when the parse does not have valid stream info(should be the first
frame case), we should continue and report caps.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2427>
Nirbheek Chauhan [Wed, 21 Jul 2021 14:10:17 +0000 (19:40 +0530)]
audiolatency: Handle audio buffers with invalid duration
pipewiresrc outputs audio buffers without a valid duration, so we need
to calculate it manually in that case.
Upstream issue: https://gitlab.freedesktop.org/pipewire/pipewire/-/issues/1438
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2419>
He Junyan [Thu, 22 Jul 2021 14:00:38 +0000 (22:00 +0800)]
va: h265dec: Do not assign the frame->output_buffer until output_picture.
We may need to drop the slices such as RASL pictures with the NoRaslOutputFlag, so
the current picture of h265decoder may be freed. We should not assign the frame->
output_buffer too early until we really output it. Or, the later coming slices will
allocate another picture and trigger the assert of:
gst_video_decoder_allocate_output_frame_with_params:
assertion 'frame->output_buffer == NULL' failed
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2421>
Edward Hervey [Thu, 22 Jul 2021 13:14:26 +0000 (15:14 +0200)]
tsdemux: Handle PCR-less streams
Some programs specify a PCR PID but don't actually store any PCR values, or are
way too far apart.
In order to gracefully handle those situations, we will queue up to a certain
amount of pending buffers before deciding to give up on that PCR PID and not use
any (i.e. using DTS/PTS values as-is)
Fixes https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/issues/1629
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2422>
He Junyan [Thu, 22 Jul 2021 02:44:27 +0000 (10:44 +0800)]
va: H265: Add odd bit depth and chroma depth in get_rtformat.
In H265, the stream may have odd bit depth such as 9 or 11. And
the bit depth of luma and chroma may differ. For example, the
stream with luma depth of 8 and chroma depth of 9 should use the
10 bit rtformat as the decoded picture format.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2420>
He Junyan [Tue, 20 Jul 2021 16:04:18 +0000 (00:04 +0800)]
codecs: h264dec: Improve the algorithm for low latency mode.
In low_latency mode, try to bump the picture as soon as possible
without the frames disorder:
1. We can directly output the continuous non-reference frame.
2. Consider max_num_reorder_frames, which is special useful for
I-P mode.
3. Consider the leading pictures with negative POC.
4 Output small POC pictures when non-reference frame comes.
4. Output the POC increment<=2 pictures. This is not 100% safe,
but in practice this condition can be used.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2373>
He Junyan [Tue, 20 Jul 2021 15:49:12 +0000 (23:49 +0800)]
codecs: h264dec: Add help function of dpb_set_max_num_reorder_frames.
The max_num_reorder_frames can be useful for bump check. We store it
in the DPB and no need for the decoder now.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2373>
He Junyan [Tue, 20 Jul 2021 15:36:38 +0000 (23:36 +0800)]
codecs: h264dec: Add a flag to record whether picture is reference.
The picture->ref field will change from time to time according to decoder's
state and reference sliding window. We need another flag to record whether
the picture is a reference picture when it is created, and this can help
the bumping check.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2373>