mpeg4_unpack_bframes_bsf bitstream filters constructs
resulting packet using av_packet_from_data() function.
This function however modifies only buffer (data) and leaves
other fields untouched, so the content of other fields
of the output packet is undefined.
It is working with old BSF API, since old API filters
just data and the packet fields are copied in
av_apply_bitstream_filters from input packet.
This change fixes the behaviour for the new BSF API.
Signed-off-by: Jan Sebechlebsky <sebechlebskyjan@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
VPx (VP8/VP9) alpha encoding has been part of FFmpeg. Now, add the
ability to decode such files with alpha channel.
Signed-off-by: Vignesh Venkatasubramanian <vigneshv@google.com>
The code conflicts with moving the h264_init_ps() call point
Without this, ff_h264_parse_ref_count() fills ref and list count and
h264_init_ps() subsequently wipes them out on a "success" path.
Subsequently things crash as the wiped fields are used.
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
This ensures that captions are written towards the bottom of the screen
when tuning into mid-stream. The row will be reset on the receipt of the
next PAC command. Row 10 was chosen as it corresponds to the value of
"0" in a PAC (see row_map in handle_pac()).
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
... and attempt to preserve compatibility with clang that was
introduced in 311a953c76 (untested)
Reviewed-by: "Ronald S. Bultje" <rsbultje@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
* commit 'd78fd2fa21cde28465e40dd0be4446b1387d22a6':
Add MagicYUV decoder
Changes observed from Libav:
- many cosmetics (function renames/move, spacing, line breaks)
- MagicYUVContext.slices_size is now unsigned
- use of pixdesc (include fixed in FFmpeg)
- mention of "Lossless" in the long name dropped (also removed from
general.texi in FFmpeg)
- addition of the FF_CODEC_CAP_INIT_THREADSAFE caps
- use of qsort() instead of AV_QSORT() (NOT MERGED)
- use of AVCodecContext.{width,height} instead of AVCodecContext.coded_{width,height} (NOT MERGED)
See also 77f9c4b7aa
Merged-by: Clément Bœsch <u@pkh.me>
Adding a check for bits == 0 would still make Coverity misdetect this,
so just revert to the normal way of setting the residue to 0.
Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>
Can also be used in future cleanups since 99% of the time the leftover
appending will just append to an already empty residue.
Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>
Shifting by more than 63 bits is undefined behavior, athough any
compiler not returning 0 after shifting by any amount would be insane.
Found by Coverity, fixes CID1363959 and CID1363960
Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>
Still much left to optimize, but it provides a significant performance
improvement - 10% for 300Mbps (1080p30), 25% for 1.5Gbps (4k 60fps) in
comparison with the default implementation.
Signed-off-by: Rostislav Pehlivanov <rpehlivanov@obe.tv>
Now coefficients are written to a buffer and are then dequantized by the
new SIMD dequantization functions. For the lower bands without enough
coefficients to fill a register (and hence they overwrite) the C version
of the dequantization function is used.
The buffer is per-thread and will be realloc'd if anything changes.
This prevents regressions and having to limit slice size.
Signed-off-by: Rostislav Pehlivanov <rpehlivanov@obe.tv>
Prevents having to have random magic values in the decoder and a
separate macro in the encoder.
Signed-off-by: Rostislav Pehlivanov <rpehlivanov@obe.tv>
In preparation for the following commits, this commit simplifies the
coefficient parsing and dequantization function. It was needlessly
inlined without much performance gain.
Signed-off-by: Rostislav Pehlivanov <rpehlivanov@obe.tv>
