avcodec/mediacodec: add av_mediacodec_render_buffer_at_time()

The existing av_mediacodec_release_buffer allows the user to render
or discard the Surface-backed frame. This new method allows the user
to control exactly when the frame will be rendered to its SurfaceView.

Available since Android API 21.

Signed-off-by: Aman Gupta <aman@tmm1.net>

libavcodec/libaomenc.c: Added code for computing PSNR/SSIM

Signed-off-by: James Almer <jamrial@gmail.com>

avcodec/ffv1dec_template: Check for the end of input after 1024 pixels

Fixes: Timeout
Fixes: 10385/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_FFV1_fuzzer-5689206987292672
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/aacdec_template: Allow duplicated elements

Such streams are invalid according to
4.5.2.1 Top level payloads for the audio object types AAC main, AAC SSR, AAC LC and AAC LTP
4.5.2.1.1 Definitions
...cIn the raw_data_block(), several instances of the
same syntactic element may occur, but must have a different 4 bit
element_instance_tag, except for data_stream_element()'s and
fill_element()'s.

Fixes: Ticket7477
Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avformat/dashenc: only write video streams into HLS master playlist

Tool mediastreamvalidator reports error "Variant media_[N].m3u8 is
missing audio group" for audio streams in HLS master playlist. As audio
streams are already listed in audio group, skip them as variant media
streams in master playlist.

lavu/qsv: make a copy as libmfx alignment requirement for uploading

Libmfx requires 16 bytes aligned input/output for uploading.
Currently only output is 16 byte aligned and assigning same width/height to
input with smaller buffer size actually, thus definitely will cause segment fault.

Can reproduce with any 1080p nv12 rawvideo input:
ffmpeg -init_hw_device qsv=qsv:hw -hwaccel qsv -filter_hw_device qsv -f rawvideo -pix_fmt nv12 -s:v 1920x1080
-i 1080p_nv12.yuv -vf 'format=nv12,hwupload=extra_hw_frames=16,hwdownload,format=nv12' -an -y out_nv12.yuv

It can fix #7418

Signed-off-by: Zhong Li <zhong.li@intel.com>

lavu/hwcontext_qsv: Add support for AV_PIX_FMT_BGRA.

RGB32(AV_PIX_FMT_BGRA on intel platforms) format may be used as overlay with alpha blending.
So add AV_PIX_FMT_BGRA format support.

One example of alpha blending overlay: ffmpeg -hwaccel qsv -c:v h264_qsv -i BA1_Sony_D.jsv
-filter_complex 'movie=lena-rgba.png,hwupload=extra_hw_frames=16[a];[0:v][a]overlay_qsv=x=10:y=10'
-c:v h264_qsv -y out.mp4

Rename RGB32 to be BGRA to make it clearer as Mark Thompson's suggestion.

V2: Add P010 format support else will introduce HEVC 10bit encoding regression.
    Thanks for LinJie's discovery.

Signed-off-by: Zhong Li <zhong.li@intel.com>
Verified-by: Fu, Linjie <linjie.fu@intel.com>

avfilter/af_afir: remove again option, merge it with gtype

libavfilter: bump micro version to 101

libavfilter/ebur128: add scale parameter

This allows switching between absolute (LUFS) and relativ (LU) display
in the status line.

Signed-off-by: Daniel Molkentin <daniel@molkentin.de>
Signed-off-by: Conrad Zelck <c.zelck@imail.de>

libavfilter/ebur128: introduce target range

This eases meeting the target level during live mixing.

Signed-off-by: Daniel Molkentin <daniel@molkentin.de>
Signed-off-by: Conrad Zelck <c.zelck@imail.de>

libavfilter/ebur128: add gauge option

Allow to show short-term instead of momentary in gauge. Useful for monitoring
whilst live mixing.

Signed-off-by: Daniel Molkentin <daniel@molkentin.de>
Signed-off-by: Conrad Zelck <c.zelck@imail.de>

libavfilter/ebur128: add target value to statistics line

Signed-off-by: Daniel Molkentin <daniel@molkentin.de>
Signed-off-by: Conrad Zelck <c.zelck@imail.de>

libavfilter/ebur128: add target level option for EBUR128 visualization filter

Signed-off-by: Daniel Molkentin <daniel@molkentin.de>
Signed-off-by: Conrad Zelck <c.zelck@imail.de>

lavf/ftp: Remove an unneeded forward declaration.

avfilter/af_asetnsamples: do not leak frame on ENOMEM

Fixes Coverity CID 1416352.

Signed-off-by: Marton Balint <cus@passwd.hu>

ffmpeg: check return value of avcodec_parameters_from_context

Fixes Coverity CID 1427273.

Signed-off-by: Marton Balint <cus@passwd.hu>

avfilter/f_cue: always check the return value of ff_inlink_consume_frame

Fixes Coverity CID 1439936.

Signed-off-by: Marton Balint <cus@passwd.hu>

h264_slice: Copy the value of x264_build before calling h264_slice_header_init during thread init

If we don't copy this value first, it is seen as 0 by h264_slice_header_init,
due to zero-allocation of the new context, triggering an old hack that
multiplied the denominator by 2 for files produced by old x264 versions, but
only if more than one thread was used.

Fixes #7475 and #7083.

Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>

avfilter/f_reverse: add packed sample format support to areverse

avfilter/af_silenceremove: fix max for duration options

ffplay: add options to specify window position

Signed-off-by: Marton Balint <cus@passwd.hu>

avutil/tests/parseutils: add some big duration tests

Signed-off-by: Marton Balint <cus@passwd.hu>

avutil/parseutils: fix some overflows in duration calculations

Also properly return AVERROR(ERANGE) in case of actual overflows.

Signed-off-by: Marton Balint <cus@passwd.hu>

avutil/opt: check for minimum and maximum values when setting AV_OPT_TYPE_DURATION fields

Signed-off-by: Marton Balint <cus@passwd.hu>

avformat/segment: fix segment_time_delta option min/max value

Signed-off-by: Marton Balint <cus@passwd.hu>

avfilter/af_afade: fix crossfade duration maximum value

Signed-off-by: Marton Balint <cus@passwd.hu>

avfilter/af_afade: add logistic sigmoid curve

avfilter/avf_showspectrum: add legend support to showspectrum filter

avformat/utils: Fix potential integer overflow in extract_extradata()

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avformat/utils: Do not use "i" as a context pointer, "i" is normally the integer counter in loops

This avoids surprising developers. Its bad to surprise developers with
such unexpected things.

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avformat/utils: Do not ignore failure in extract_extradata_init()

We check for the documented explanation of the "Ignore code" in extract_extradata_check() already

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/speedhq: Make speedhq_vlc const

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avfilter/vf_tonemap_opencl: Fix seperate typo

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/vc1_loopfilter: Fix colums typos

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

libavcodec/v4l2_buffers.h: fix cant typos

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avfilter/af_biquads: Remove l from %lf in av_log environment

The l modifier does nothing in C99 and it was undefined in C89 for %f

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avfilter/af_aiir: Remove l from %lf in av_log environment

The l modifier does nothing in C99 and it was undefined in C89 for %f

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avformat/movenc: Remove unneeded variable from mov_find_codec_tag()

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

libavformat/ty: use decimal constant for shift

This was the only case in the source that uses a hexadecimal shift value.
The change removed a special case in respect to greping

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/unary: Improve get_unary() docs

Found-by: kierank
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avfilter/avf_showspectrum: switch to activate and add fps option

Merge commit '21733b39d0af5211d7b9f168ff3667ea86362e2b'

* commit '21733b39d0af5211d7b9f168ff3667ea86362e2b':
  lavu/qsv: fix a random hwupload failure regression

Fixes CID #1439585.

Merged-by: Mark Thompson <sw@jkqxz.net>

lavfi/deshake: fix deshake crash issue.

Fixes ticket #7441.

for block contrast calculate, the block is like this:

|<---------------- stride-----------------------|
+----------------------------------------------->  X
|
|            w = 16
|    (cx,cy)+------+
|           |      |
|h=blocksize|      |
|           |      |
|           +------+
V

Y

so we calc the block contrast use:
   (cy + y) * stride + (cx + x)

Signed-off-by: Jun Zhao <mypopydev@gmail.com>

av1_metadata: Fix constraint on setting chroma_sample_position

It can't be set for 4:2:2 streams.

avfilter/avf_showspectrum: increase padding size for low sample rates

avfilter/avf_showspectrum: add zoom mode to showspectrumpic

avfilter/avf_showspectrum: add green color map

lavfi/atempo: fix tempo range limit inconsistency

yae_set_tempo was overlooked when max tempo limit was raised to 100.

tested with:
./ffmpeg_g -i Delerium/SemanticSpaces/Gateway.mp3 \
-af asendcmd=f=asendcmd.cfg,atempo=1.0 -y /tmp/asendcmd-atempo.wav

where asendcmd.cfg was:
15.0-45.0 [enter] atempo tempo 2.0,
          [leave] atempo tempo 0.5;

60.0-300.0 [enter] atempo tempo 4.0,
           [leave] atempo tempo 1.0;

avfilter/avf_showspectrum: fix scaling in zoom mode

avfilter/avf_showspectrum: implement zoom mode

avilter/af_afir: add irfmt option

avcodec/libaomdec: remove bogus comment

It's a remnant from libvpx that's not valid for libaom.

Signed-off-by: James Almer <jamrial@gmail.com>

avfilter/af_afir: use internal lavfi queue

avfilter/af_headphone: use lavfi internal queue instead

Signed-off-by: Paul B Mahol <onemda@gmail.com>

avfilter/af_asetnsamples: use lavfi internal queue

avfilter: add ff_inlink_queued_samples()

avcodec: add an AV1 parser

Simple parser to set keyframes, frame type, structure, width, height, and pixel
format, plus stream profile and level.

Reviewed-by: Mark Thompson <sw@jkqxz.net>
Signed-off-by: James Almer <jamrial@gmail.com>

avcodec/av1_metadata: add an option to insert and remove Temporal Delimiter OBUs

Reviewed-by: Mark Thompson <sw@jkqxz.net>
Signed-off-by: James Almer <jamrial@gmail.com>

avfilter/f_cue: use inlink fifo for queueing frames

Signed-off-by: Marton Balint <cus@passwd.hu>

avfilter/filters: add ff_inlink_peek_frame and ff_inlink_queued_frames to access frames in the inlink fifo

Signed-off-by: Marton Balint <cus@passwd.hu>

doc/protocols: simplify and clarify UDP localaddr option

Signed-off-by: Marton Balint <cus@passwd.hu>

avformat/udp: always use IP_ADD_SOURCE_MEMBERSHIP for subscribing to an UDP multicast source group in IPv4

That alone supports specifying the interface based on its address. Getting the
interface index from the local address seems quite a bit of work in a platform
independent way...

Obviously for IPv6 we still always use MCAST_JOIN_SOURCE_GROUP.

As a side effect this also fixes ticket #7459.

Signed-off-by: Marton Balint <cus@passwd.hu>

avformat/udp: specify the local address for some source filtered multicast joins

We already use localaddr for the multicast joins without source filters, so we
should use them for source filters as well. This patch only fixes the
IP_ADD_SOURCE_MEMBERSHIP and the IP_BLOCK_SOURCE case.

Unless we do this, the kernel automatically selects an interface based on the
source address, and that interface might be different from the one set in
localaddr. For blocked sources this even casues EINVAL because we joined the
multicast group on a different interface.

Signed-off-by: Marton Balint <cus@passwd.hu>

avformat/udp: add support for generic source filtering

This allows getting data only from a specific source IP. This is useful not
only for unicast but for multicast as well because multicast source
subscriptions do not act as source filters for the incoming packets.

Signed-off-by: Marton Balint <cus@passwd.hu>

avformat/rtpproto: use factorized ip functions

Signed-off-by: Marton Balint <cus@passwd.hu>

avformat/udp: use factorized ip functions

Signed-off-by: Marton Balint <cus@passwd.hu>

avformat/ip: factorize some IP filtering and resolving functions to a new file

These are based on the very similar UDP and RTP protocol functions.

Signed-off-by: Marton Balint <cus@passwd.hu>

avfilter/af_afir: make IR gain control more flexible

For this reason introduce two more options.

avfilter/avfilter: fix typos in comments

lavu/qsv: fix a random hwupload failure regression

Variable 'ret' hasn't been initialized，thus introducing a random
hwupload failure regression due to qsv session uninitialized.

Signed-off-by: Zhong Li <zhong.li@intel.com>
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>

avcodec/rasc: unref both frames on reinit

Fixes: integer overflow
Fixes: inconsistent frame dimensions
Fixes: 10454/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_RASC_fuzzer-5656301162463232
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/rasc: Fix off by 1 error in vertical coordinate

Fixes: out of array read
Fixes: 10311/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_RASC_fuzzer-4856330905452544
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/tiff: check remaining packet size for strips

Fixes: Timeout
Fixes: 10280/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_TIFF_fuzzer-5676217211027456
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avformat/matroskaenc: reserve free space for metadata on request

Signed-off-by: James Almer <jamrial@gmail.com>

avcodec/scpr: Skip frames which change nothing

Fixes: Timeout
Fixes: 10292/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_SCPR_fuzzer-5687943864254464
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/wcmv: Fix integer overflows

Fixes: signed integer overflow: 262140 * 65535 cannot be represented in type 'int'
Fixes: 10090/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_WCMV_fuzzer-5691269368512512
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/gdv: Remove unused variable

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

ffmpeg: break loop when dts_heuristic check done

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avdevice/decklink_dec: add option to align capture start time

This option is useful for maintaining input synchronization across N
different hardware devices deployed for 'N-way' redundancy.
The system time of different hardware devices should be synchronized
with protocols such as NTP or PTP, before using this option.

Signed-off-by: Marton Balint <cus@passwd.hu>

avcodec/hevc_parser: very minor simplification in hevc_parse()

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

lavc/hevc: Don't parse NAL unit for a dummy buffer

hevc parser mistakenly reports the following message if a dummy buffer
is padded for EOF

   [hevc @ 0x559b63848610] missing picture in access unit

Signed-off-by: Haihao Xiang <haihao.xiang@intel.com>
Reviewed-by: Steven Liu <lq@chinaffmpeg.org>
Reviewed-by: "Li, Zhong" <zhong.li@intel.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/libzvbi-teletextdec: add formatted ass output

Inspired by the VideoLAN text decoder and its port to FFmpeg made by Aman
Gupta.

Signed-off-by: Marton Balint <cus@passwd.hu>

avcodec/diracdec: check that GetBitContext has not ended in codeblock()

Fixes: Timeout (part 2 of 2)
Fixes: 9774/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_DIRAC_fuzzer-5748957085958144
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/diracdec: Propagate errors from codeblock()

Consider a component to be damaged if more than 50% of its subbands are damaged

Fixes: Timeout (part 1 of 2)
Fixes: 9774/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_DIRAC_fuzzer-5748957085958144
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/bintext: Use ff_get_buffer instead of ff_reget_buffer and simplify

reget seems unneeded and it is slower

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/bintext: Check input size before allocating the input image

Fixes: Timeout
Fixes: 9795/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_XBIN_fuzzer-5768631928487936
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/gdv: Error out if we had insufficent input for the output frame in decompress_2()

Fixes: Timeout (with the previous commits to gdv.c)
Fixes: 9422/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_GDV_fuzzer-5143490001371136
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/gdv: Optimize 2x scaling loop a little in gdv_decode_frame()

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/gdv: Replace loop in gdv_decode_frame() by memcpy()

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/gdv: Eliminate 50% of the reads in the first inner loop in rescale()

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/gdv: Simplify first inner loop in rescale()

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/gdv: Combine first and 3rd case in rescale()

Simplifies code

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/gdv: Factor terms out of inner loops in rescale()

This also replaces some loops by memcpy()

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/gdv: Replace divisions by shifts in rescale()

Divisions tend to be slower than shifts unless the compiler optimizes them out.
And some of these are in inner loops.

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

lavf/mpeg: Support demuxing HEVC in mpeg-ps.

lavc: Add AV1 metadata bitstream filter

Can adjust colour and timing information.

lavc: Add coded bitstream read/write support for AV1

avfilter/af_afftdn: add missing break

avformat/matroskaenc: reindent after the previous commit

Signed-off-by: James Almer <jamrial@gmail.com>