lavu/frame: put frame QP elements under a new version guard

These fields are still used, removal postponed until a replacement functionality is available.

Merge commit '8f12ef9860d0e164e4647fd5d5cebdb3cfb34a79'

* commit '8f12ef9860d0e164e4647fd5d5cebdb3cfb34a79':
  lavu: Drop deprecated duplicated AVFrame/AVCodecContext parameters

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>

lavu/cpu: remove old cmov cruft

Merge commit 'bf7114b6caad8cf94696b0299c13b0d26bf291be'

* commit 'bf7114b6caad8cf94696b0299c13b0d26bf291be':
  lavu: Drop deprecated AV_CPU_FLAG_MMX2 symbol

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>

Merge commit '2d40968dd3ff17b12f7c80dbfad409b14418e267'

* commit '2d40968dd3ff17b12f7c80dbfad409b14418e267':
  lavu: Drop deprecated audioconvert.h header

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>

Merge commit 'cdfe45ad371b7a8e6135b6c063b6b2a93152cb3a'

* commit 'cdfe45ad371b7a8e6135b6c063b6b2a93152cb3a':
  lavu: Drop deprecated av_reverse function

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>

Merge commit '2f8cbbc962dfc0dc1dd0a90b2cd6c21266380f51'

* commit '2f8cbbc962dfc0dc1dd0a90b2cd6c21266380f51':
  lavu: Drop deprecated external access to AVPixFmtDescriptor table

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>

Merge commit '183db02a51a422568084b113a7571c845ca68622'

* commit '183db02a51a422568084b113a7571c845ca68622':
  lavu: Drop deprecated old_pix_fmt.h and related code

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>

Merge commit 'cad40a3833ad81a352e7657ec6f7d637cea3b798'

* commit 'cad40a3833ad81a352e7657ec6f7d637cea3b798':
  lavc: Drop deprecated deinterlace module

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>

Merge commit '069713aa4b137781e270768d803b1f7456daa724'

* commit '069713aa4b137781e270768d803b1f7456daa724':
  lavc: Drop deprecated thread opaque and codec pkt

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>

Merge commit '9f90b24877016e7140b9b14e4b1acee663bb6d8a'

* commit '9f90b24877016e7140b9b14e4b1acee663bb6d8a':
  lavc: Drop deprecated get_buffer related functions

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>

Remove left-over FF_API_DESTRUCT_PACKET cruft

Merge commit '01bcc2d5c23fa757d163530abb396fd02f1be7c8'

* commit '01bcc2d5c23fa757d163530abb396fd02f1be7c8':
  lavc: Drop deprecated destruct_packet related functions

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>

Merge commit 'dc70c19476e76f1118df73b5d97cc76f0e5f6f6c'

* commit 'dc70c19476e76f1118df73b5d97cc76f0e5f6f6c':
  lavc: Drop deprecated request_channels related functions

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>

Remove left-over FF_API_AVFILTERBUFFER cruft

Merge commit 'f6974fe651d29ef6eb68d66d73f7b6c011062aa0'

* commit 'f6974fe651d29ef6eb68d66d73f7b6c011062aa0':
  lavfi: Drop deprecated AVFilterBuffer* code

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>

Merge commit 'e65e4cbbda03ca3c9087f069c9867d518415fca1'

* commit 'e65e4cbbda03ca3c9087f069c9867d518415fca1':
  lavfi: Drop deprecated *_count suffixed variables

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>

Merge commit '86e5056575f55f070609dd3926605302f7d2280e'

* commit '86e5056575f55f070609dd3926605302f7d2280e':
  lavfi: Drop deprecated public AVFilterPad struct

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>

Merge commit '11b2eed43e91b35b8295ed47115cae2e29bd687d'

* commit '11b2eed43e91b35b8295ed47115cae2e29bd687d':
  lavr: Drop deprecated context reinitialization if resampling was not enabled

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>

Merge commit '9d58639e270f7612874681e0ca30fa461e2667b7'

* commit '9d58639e270f7612874681e0ca30fa461e2667b7':
  sws: Drop deprecated SWS_CPU_CAPS defines

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>

avfilter/avf_showfreqs: Fix memleak of out frame

Fixes CID1322344

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

avfilter/avf_showfreqs: Fix "may be used uninitialized in this function" warning

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

avfilter/avf_showfreqs: Free fin

Fixes CID1322345

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

avfilter/af_amerge: avoid undefined shift (<<64) in outlayout setup

Fixes CID1322306

Reviewed-by: Ganesh Ajjanagadde <gajjanag@mit.edu>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avfilter/vf_stack: Fix memleak of out frame

Fixes CID1322347

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

avformat/hls: Check for av_opt_set_dict() failure

Fixes: CID1320426
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/rawenc: Use AVFrame parameters instead of AVCodecContext

This allows encoding raw frames with changing dimensions

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

avfilter/avfilter: Add a few more basic filters to the list which support frame size changes

Fixes assertion failures

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

lavf/aiffenc: Clarify an error message.

Only one audio stream is allowed in aiff.

fate: adjust AAC encoder TNS test target

The new commits improve the quality.

Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>

aacenc_tns: adjust coefficient calculation, add double filter support

This commit improves the TNS implementation to the point where it's
actually usable and very rarely results in nastyness (in all bitrates
except extremely low bitrates it's increasing the quality and prevents
some distortions from the coder being audiable).

Also adds a double filter support which is only used if the energy
difference between the top and bottom of the SFBs is above the
thresholds defined in the header file. Looking at the bitstream
that fdk_aac generates it sometimes used a double filter despite
the specs stating that a single filter should be enough for almost
all cases and purposes.

Unlike FAAC or fdk_aac we sometimes use a reverse filter in case
the energy difference isn't enought to use a double filter. This
actually works better.

Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>

lpc: increase error array size of ff_lpc_calc_ref_coefs_f by one

Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>

aacenc_is: add a flag to use pure coefficients instead

This commit adds a flag to use the pure coefficients instead
of the processed ones (sce->coeffs). This is needed because
IS will apply the changes to the coefficients immediately
before the adjust_common_prediction function and it doesn't
make sense to measure stereo channel coefficient difference
when one of the channels coefficients are all zero.

Therefore add a flag to use pure coefficients in that case.
TNS is the only thing touching the coefficients before IS
so common window prediction will not take that into account
but the effect of the TNS filter per coefficient can be small
(a few percent) so to some approximation it's fine to just
ignore that.

Also fixed a small error which doesn't alter the results
that much. pow(sqrt(number), 3.0/4.0) == pow(number, 3.0/8.0) !=
pow(number, 3.0/4.0).

Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>

avcodec/jpeg2000dec: Initialize ret to avoid warning and make the code more robust

"Fixes" CID1322361

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

avcodec/jpeg2000dec: Assert that step_x/y are valid

Fixes CID1322304, CID1322305

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

avcodec/hapdec: Check section_size for non negativity in parse_section_header()

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

lavf/mxfdec: Support more codecs in mxf_picture_essence_container_uls[].

Signed-off-by: Carl Eugen Hoyos <cehoyos@ag.or.at>

lavf/mxf: Support video essence container uls for vc1.

Signed-off-by: Carl Eugen Hoyos <cehoyos@ag.or.at>

lavf/riff: Support fourcc AVd1 for dvvideo.

Signed-off-by: Carl Eugen Hoyos <cehoyos@ag.or.at>

swscale: re-enable gamma

+added gamma conversion to refactored code

swscale: added proper error check for ff_init_filters

lavc/mxfdec: Support more mxf files with codec_ul.

Fixes decoding for the sample from ticket #4820.

lavc/j2kenc: Remove experimental flag.

The encoder produces valid and - if requested - lossless files.

avfilter/vf_vectorscope: change intensity to float

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

avfilter/vf_waveform: stop abusing s->size for calculating limit

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

avfilter/vf_waveform: change intensity to float

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

avfilter/vf_waveform: reduce number of operations a little

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

avfilter/vf_waveform: 9 and 10 bit depth support for lowpass & color filter

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

vp9_parse: fix parsing of pskip and profile 2/3.

The fate results change because we now correctly timestamp the pskip
frames, which means the results are now identical to -vsync 0.

libvpxdec: explicitly signal RGB.

vp9: don't erase values for {lf,ref,skip,q}_enabled if segmentation=0.

Instead, use segmentation.enabled before accessing each of these values.

vp9: fix segmentation map retention across keyframe boundaries.

vp9: fix RGB chroma subsampling.

vp9: read reserved bit in RGB header.

vp9: clip intermediates in dequant calculations.

This makes values consistent with libvpx.

vp9: fix segmentation map retention if segmentation is turned off.

vp9: use resetctx in the same way as libvpx.

vp9: fix indentation.

vp9: deal with the case where update_map=0 but we're key/intraonly.

vp9: reset segmentation information on context reset.

avcodec/g2meet: Also clear tile dimensions on header_fail

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

avcodec/g2meet: Fix potential overflow in tile dimensions check

Fixes CID1322351

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

avcodec/eatgq: Check init_get_bits8() for failure

Fixes CID1322315

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

avcodec/adpcm: Check init_get_bits8() for failure

Fixes CID1322317

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

ffmpeg: Drop redundant ist check

stream copy always has a input stream, it cannot use complex video/audio filters with unambigous input
Fixes CID1322348

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

doc/examples/http_multiclient: Fix occured typo

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

avcodec/libfdk-aacdec: Remove unused variable

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

avcodec/hevc_parser: Check init_get_bits8() for failure

Fixes: CID1322322
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/ira288: Check init_get_bits8() for failure

Fixes: CID1322321
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/smacker: Check init_get_bits8() for failure

Fixes: CID1322314
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/sonic: Check init_get_bits8() for failure

Fixes: CID1322310
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/svq1dec: Check init_get_bits8() for failure

Fixes: CID1322313
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/tta: Check init_get_bits8() for failure

Fixes: CID1322319
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/vp3: Check init_get_bits8() for failure

Fixes CID1322316

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

avfilter/af_channelmap: Reorder operations to avoid memleak

Fixes CID1322346
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avfilter/af_sidechaincompress: Also assert that i < 2

This should help static analyzers (CID1322339)

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

avfilter/vf_atadenoise: Check for ff_get_video_buffer() failure

Fixes CID1322338

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

avfilter/vf_histogram: Fix order of operations with flags

Fixes CID1322325

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

avcodec/gsmdec_template: avoid undefined negative left shifts

Fixes: unknown_unknown_338_824_cov_1045285351_sample-gsm-8000.mov
Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avformat/mov: Change the type of the r/g/b variables

Fixes integer overflow
Fixes: unknown_unknown_31b_795_cov_1818643045_raybauduc.mov
Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

lavf/http: Remove superfluous parenthesis.

Signed-off-by: Stephan Holljes <klaxa1337@googlemail.com>
Reviewed-by: Ganesh Ajjanagadde <gajjanag@mit.edu>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avfilter/avf_showwaves: Check max_samples

Fixes potential division by zero
Fixes: CID1292295
Reviewed-by: Ganesh Ajjanagadde <gajjanag@mit.edu>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

lavf/http: Fix incorrectly placed parenthesis.

The assignment had incorrectly placed parentheses which resulted in ret
always being > 0.

Reviewed-by: wm4 <nfxjfg@googlemail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avfilter/af_dynaudnorm: Fix typo in assert

Fixes: CID1322303
Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avfilter/avf_showfreqs: Use floating point division in WFUNC_BHANN

Fixes: CID1322365
Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avformat/concatdec: Check file variable before dereferencing

Fixes CID1322328

Reviewed-by: Ganesh Ajjanagadde <gajjanag@mit.edu>
Reviewed-by: Nicolas George <george@nsup.org>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

AAC: MIPS: Add missing codebooks in quantize funcs

Add entries on the quantize function tables for the missing
codebooks (which all behave like ZERO)

AAC: Increase fuzziness of fate-aac tests

Needed to make them pass in mips

doc: Explain how to use the fps and the fieldmatch filter together.

Fixes ticket #3968.

avformat/file: Fix copy and paste error

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

avfilter/vf_sab: use the name 's' for the pointer to the private context

Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>

avcodec/fraps: use init_get_bits8()

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

avformat/hlsenc: Fix memleak of path

Fixes: CID1322343
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avformat/hlsenc: Initialize vtt_oc to help static analyzers

This is also more robust in case it ever is used

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

avformat/latmenc: Add assert to avoid coverity warning

Fixes CID1322323

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

avfilter/vf_drawgraph: add rscroll slide mode

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

avcodec/vorbisdec: use init_get_bits8()

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

avfilter/vf_vectorscope: constify more variables

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

avfilter/vf_vectorscope: support more formats for color4 mode in common case

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

avfilter/vf_elbg: make it possible to output to pal8 pixel format

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