* Correctly handle FMA instructions with memory operands.
 * Print a warning if FMA instructions are used without the correct cpuflag.
 * Simplify the instantiation code.
 * Clarify documentation.

Only the last operand in FMA3 instructions can be a memory operand. When
converting FMA4 instructions to FMA3 instructions we can utilize the fact
that multiply is a commutative operation and reorder operands if necessary
to ensure that a memory operand is used only as the last operand.

It seems to miscompile them

Should fix fate-ra-288 and fate-twinvq
Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: James Almer <jamrial@gmail.com>

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

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

The version structure in the main decoder context was not (and
apparently has never been) populated since it was added.
Still, having VC-2 break the existing Dirac Low Delay mode was odd and
easily avoidable had the specifications authors noticed/cared.
Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>

The specs define some more overlapping and some rarely used parse codes
which were missing.
Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>

Reviewed-by: Rostislav Pehlivanov <atomnuker@gmail.com>
Signed-off-by: James Almer <jamrial@gmail.com>

Forgotten from previous commit.
Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>

The parser scans for "BBCD" to appear in the bitstream which indicate a
parse info header and once that happens, checks if the parse offsets are
sane. Since random BBCD strings might appear in the bitstream the parser
will emit a pointless warning if that happens.

This commit improves parsing by checking for a valid parse code as well
as keeping the original checks for valid parse offsets. The warnings
were removed as they serve no real purpose.
Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>

Fixes assertion failure
Fixes: 6568d187979ce17878b6fe5fbbb89142/signal_sigabrt_7ffff6ae7cb7_7176_564bbc6741bdcf907f5c4e685c9a77a2.mpg

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

The DSP lacked a function needed to convert signed to unsigned. This was
ignored when originally adding support and templating for bit depths
greater than 8. The 10 bit function was used for 12 bit pictures and
resulted in an improper conversion.
Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>

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

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

Should prevent some leaks if header writing fails
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

Can happen in cases where's there's zero autocorrelation (pulses),
and it also implies NaN/inf coeffs

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

Reported by Bob, no6b no6b com

Reviewed-by: Andreas Cadhalpun <andreas.cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

Fixes null pointer dereference
Fixes: c4faf8280ba366bf00a79d425f2910a8/signal_sigsegv_1f96477_5177_1448ba7e4125faceb966f44ceb69abfa.qcp
Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

Cleaner solution is welcome!
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

This was somehow forgotten in a8bb81a0.

Signed-off-by: Vittorio Gambaletta <ffmpeg-dev@vittgam.net>
Signed-off-by: Marton Balint <cus@passwd.hu>

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

This is needed as near infinite values on the input side result in only some
output to be non finite.
Also it may still be insufficient if subsequent computations overflow

Fixes null pointer dereference
Fixes: ae66c0f6c12ac1cd5c2c237031240f57/signal_sigsegv_2618c99_9516_6007026f2185a26d7afea895fbed6e38.ogg

Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Reviewed-by: Claudio Freire <klaussfreire@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

This feature is not know much or used much AFAIK, and it might be helpfull in
exploits.
No specific case is known where it can be used in an exploit though
subsequent commits depend on this commit though
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: James Almer <jamrial@gmail.com>

Leftover from the experimental flag removal.
Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>

Has been marked for removal for over a month and has not been improved
or touched at all since it was implemented.
Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>

Too many crashes observed. Can't be helped until the autocorrelation
function is massively checked for sanity.
Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>

Documentation of eval, param0 and param1 parameters
Signed-off-by: Bela Bodecs <bodecsb@vivanet.hu>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

With this, user applications which use custom IO and have set a IO context will not have
their already opened IO context ignored and glob/seq being interpreted

Comments and tests from maintainers of user apps are welcome!
Liked-by: wm4 <nfxjfg@googlemail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

Updates libkvazaar to pass the exact frame rate to Kvazaar by setting
the numerator and denominator separately instead of a single floating
point number. The exact frame rate is needed for writing timing info to
the bitstream.

Requires Kvazaar version 0.8.1.
Signed-off-by: Arttu Ylä-Outinen <arttu.yla-outinen@tut.fi>

Fixes: 55d71971da50365d542ed14b65565fe1/signal_sigsegv_4765a4_8499_f146af090a94f591d6254515c7700ef5.mkv

Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Reviewed-by: Andreas Cadhalpun <andreas.cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

This feature is mostly only used by NLE software, and is
both of dubious value being enabled by default, and a
possible security risk.
Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

vf_scale: Detecting changes of incoming frame properties and dinamically evaluate width and height expressions

Currently scale filter accepts expressions in its width and height
parameters but evaluates them only once at init and replaces them with
their actual values. Later on, if any parameter of incoming frames
changes - ie those were used in the original size expressions -  then
they new values will not have any affect for width and heigth values.
They remain the same. This patch makes possible that width and height
expressions be evaluated frame-by-frame basis if width/height/sar/format
properties of incoming frame would change. To retain the current
behaviour and not to break any earlier app, a new config parameter has
been introduced. Its name is "eval" and it has two distinct values:
"init" and "frame". The default value is "init".
This feature is very usefull in case of DVBT mpeg-ts streams where SAR
may change time-by-time from 4/3 to 16/9 and vica-versa and the size
remains the same and you want to create a variable sized output with 1/1
SAR.
Signed-off-by: Bela Bodecs <bodecsb@vivanet.hu>
Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

vf_overlay video filter accepts expressions in its parameters. In
'frame-by-frame' evaluation mode it recalculates them regularly, but
incoming video frame size changes did not reflect in their values. So if
you used width or height of any source videos in expressions as
parameters, they stayed on their initial values. This patch corrects
this bug.
Signed-off-by: Bela Bodecs <bodecsb@vivanet.hu>
Reviewed-by: Paul B Mahol <onemda@gmail.com
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>