This reverts commit 9413bdc381.
That commit broke the fate HEVC tests - unfortunately I only
tested checkasm for that patch, and that function is still
lacking checkasm coverage.
Signed-off-by: Martin Storsjö <martin@martin.st>
pts may not be set on input packets, which could result in the entire stream
being discarded.
This reverts commit 8fc2dedfe6, reintroducing the
behavior it replaced but now allowing the caller to manually drop the preroll
samples by looking at the skip_samples side data at the start while ignoring it
on seek, by setting the skip_manual avctx flag.
Fixes ticket #10251.
Suggested-by: Anton Khirnov <anton@khirnov.net>
Signed-off-by: James Almer <jamrial@gmail.com>
These fields are supposed to store information about the packet the
frame was decoded from, specifically the byte offset it was stored at
and its size.
However,
- the fields are highly ad-hoc - there is no strong reason why
specifically those (and not any other) packet properties should have a
dedicated field in AVFrame; unlike e.g. the timestamps, there is no
fundamental link between coded packet offset/size and decoded frames
- they only make sense for frames produced by decoding demuxed packets,
and even then it is not always the case that the encoded data was
stored in the file as a contiguous sequence of bytes (in order for pos
to be well-defined)
- pkt_pos was added without much explanation, apparently to allow
passthrough of this information through lavfi in order to handle byte
seeking in ffplay. That is now implemented using arbitrary user data
passthrough in AVFrame.opaque_ref.
- several filters use pkt_pos as a variable available to user-supplied
expressions, but there seems to be no established motivation for using them.
- pkt_size was added for use in ffprobe, but that too is now handled
without using this field. Additonally, the values of this field
produced by libavcodec are flawed, as described in the previous
ffprobe conversion commit.
In summary - these fields are ill-defined and insufficiently motivated,
so deprecate them.
The AVPacket given to an encoder's encode callback
is unreferenced generically on error.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
The AVPacket given to an encoder's encode callback
is unreferenced generically on error.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
The AVPacket given to an encoder's encode callback
is unreferenced generically on error.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
FFmpeg's assembly code currently does not abide by the
plattform-specific ABIs wrt its handling of the X86 MMX flag:
Resetting the MMX state is deferred to avoid doing it multiple times
instead of ensuring that the CPU is in floating point state
upon return from any function.
Furthermore, resetting said state is sometimes done generically,
namely for all the decoders using the ordinary decode callback;
yet this is not done for the decoders using the receive_frame API.
This led to problems when MJPEG (and the MJPEG-based decoders)
were switched to the receive_frame API in commit
e9a2a87773, because ff_mjpeg_decode_sos()
only resets the MMX state on success, not on failure.
Such issues are probably still possible with SMVJPEG, which still
uses the receive_frame API. See issue #10210.
This commit therefore also resets the MMX state for
the receive_frame API to avoid any more surprises of this sort.
Reviewed-by: Anton Khirnov <anton@khirnov.net>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
The base_enable_flag is parallel to three_Spline_enable_flag. The
typesetting of the specification is very misleading.
Signed-off-by: Zhao Zhili <zhilizhao@tencent.com>
It conflicts the comments. The operation based on Delta_enable_mode
can be applied by user during tone mapping.
Signed-off-by: Zhao Zhili <zhilizhao@tencent.com>
This also fixes the number of planes for the NV formats
(this seems to not have caused any problems).
Reviewed-by: Anton Khirnov <anton@khirnov.net>
Reviewed-by: Jan Ekström <jeebjp@gmail.com>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
Reviewed-by: Anton Khirnov <anton@khirnov.net>
Reviewed-by: Jan Ekström <jeebjp@gmail.com>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
Reviewed-by: Anton Khirnov <anton@khirnov.net>
Reviewed-by: Jan Ekström <jeebjp@gmail.com>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
Generally if maxrate is set, the calculation should be maxrate over
bufsize. This additionally enables CRF + maxrate & bufsize usage.
In order to keep negative values from enabling zero to be treated
as larger and causing a division by zero, check that one of the
variables is larger than zero.
It is currently abused to store packet size, which breaks
AV_CODEC_FLAG_COPY_OPAQUE.
Use stream_index instead, which is unused in libavcodec and has the
same type as size.
Found-by: Martin Storsjö
This includes Mastering Display, Content light level, and some ITU-T T35
metadata like closed captions and HDR10+.
Signed-off-by: James Almer <jamrial@gmail.com>
If the packets returned by libvpx and our internal frame properties
queue get desynchronized for some reason (should not happen, but it is
not clear libvpx API guarantees this), we will keep adding to the queue
indefinitely and never remove anything.
Change the code to drain the queue even if timestamps do not match.
This AVFifo is used to propagate HDR metadata from input frames to
output packets, since libvpx does not allow passing through arbitrary
user data.
It will be extended to pass through other kinds of data in future
commits, so give it a more generic name.
It is not used, except to check whether the packet is valid before
writing HDR metadata to the packet in storeframe(). However, that check
serves no purpose, as the encoded packet is already treated as valid
higher up in this function.
