lrintf is anyway used, suggesting we only care up to floating precision.
Rurthermore, there is a compat hack in avutil/libm for this function,
and it is used in avcodec/aacps_tablegen.h.

This yields a non-negligible speedup. Sample benchmark:
x86-64, Haswell, GNU/Linux:

old (draw_mandelbrot):
274635709 decicycles in draw_mandelbrot,     256 runs,      0 skips
300287046 decicycles in draw_mandelbrot,     512 runs,      0 skips
371819935 decicycles in draw_mandelbrot,    1024 runs,      0 skips
336663765 decicycles in draw_mandelbrot,    2048 runs,      0 skips
581851016 decicycles in draw_mandelbrot,    4096 runs,      0 skips

new (draw_mandelbrot):
269882717 decicycles in draw_mandelbrot,     256 runs,      0 skips
296359285 decicycles in draw_mandelbrot,     512 runs,      0 skips
370076599 decicycles in draw_mandelbrot,    1024 runs,      0 skips
331478354 decicycles in draw_mandelbrot,    2048 runs,      0 skips
571904318 decicycles in draw_mandelbrot,    4096 runs,      0 skips
Reviewed-by: NMichael Niedermayer <michael@niedermayer.cc>
Signed-off-by: NGanesh Ajjanagadde <gajjanagadde@gmail.com>

This rewrites into a similar expression avoiding sqrt. Similarity is
assured since sqrt(x^2 + y^2)/(x+y) lies in [1/sqrt(2), 1] for x, y > 0.

Tested on x86-64, Haswell, GNU/Linux.
Command:
ffmpeg -f lavfi -i mandelbrot -f null -

old (draw_mandelbrot):
277625266 decicycles in draw_mandelbrot,     256 runs,      0 skips
304527322 decicycles in draw_mandelbrot,     512 runs,      0 skips
377593582 decicycles in draw_mandelbrot,    1024 runs,      0 skips
338539499 decicycles in draw_mandelbrot,    2048 runs,      0 skips
583630357 decicycles in draw_mandelbrot,    4096 runs,      0 skips

new (draw_mandelbrot):
274635709 decicycles in draw_mandelbrot,     256 runs,      0 skips
300287046 decicycles in draw_mandelbrot,     512 runs,      0 skips
371819935 decicycles in draw_mandelbrot,    1024 runs,      0 skips
336663765 decicycles in draw_mandelbrot,    2048 runs,      0 skips
581851016 decicycles in draw_mandelbrot,    4096 runs,      0 skips
Reviewed-by: NMichael Niedermayer <michael@niedermayer.cc>
Signed-off-by: NGanesh Ajjanagadde <gajjanagadde@gmail.com>

Reviewed-by: NRostislav Pehlivanov <atomnuker@gmail.com>
Signed-off-by: NGanesh Ajjanagadde <gajjanagadde@gmail.com>

See e.g https://stackoverflow.com/questions/1666353/are-typedef-and-define-the-same-in-c
for rationale.
Signed-off-by: NGanesh Ajjanagadde <gajjanagadde@gmail.com>

See e.g https://stackoverflow.com/questions/1666353/are-typedef-and-define-the-same-in-c
for rationale.
Signed-off-by: NGanesh Ajjanagadde <gajjanagadde@gmail.com>

This option can be used to select useful frames from an ffconcat file which is
using inpoints and outpoints but where the source files are not intra frame
only.
Reviewed-by: NStefano Sabatini <stefasab@gmail.com>
Signed-off-by: NMarton Balint <cus@passwd.hu>

Reviewed-by: NNicolas George <george@nsup.org>
Signed-off-by: NMarton Balint <cus@passwd.hu>

If duration is still AV_NOPTS_VALUE when opening the next file, we can assume
that outpoint is not set.
Reviewed-by: NNicolas George <george@nsup.org>
Signed-off-by: NMarton Balint <cus@passwd.hu>

Reviewed-by: NNicolas George <george@nsup.org>
Signed-off-by: NMarton Balint <cus@passwd.hu>

Fixes out of array access
Fixes: 1430e9c43fae47a24c179c7c54f94918/signal_sigsegv_421427_2049_f2192b6829ab6e0eefcb035329c03c60.264

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

* commit 'fb8753ad':
  qsvenc: factor out common options
Merged-by: NDerek Buitenhuis <derek.buitenhuis@gmail.com>

Fixes a floating point exception when width and height are not supplied
(and therefore are zero).

Calculate packet size only once, and propagate errors earlier in the chain.

Also remove use of the deprecated av_image_get_buffer_size().

More readable and less breakable.

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

Requested-by: Mike Brown, brown at mrvideo vidiot com

avcodec/mpegvideo_enc: Remove slice structured mode from H.263 as well as the code automatically enabing it

There is no such thing as a slice structured mode in the original version 1  H.263,
that mode was added in H.263+ in 1998. Also the headers for slice structured mode
are not part of the older version 1 and this would result in unplayable files

An alternative to this patch would be to merge the H263 and H263P AVCodecs and use
other means to distinguish the older and newer versions.
Signed-off-by: NMichael Niedermayer <michael@niedermayer.cc>

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

Signed-off-by: NMatthieu Bouron <matthieu.bouron@stupeflix.com>
Signed-off-by: NClément Bœsch <clement@stupeflix.com>

Signed-off-by: NMatt Oliver <protogonoi@gmail.com>

FATE test changes because of the switch from shift to division.

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

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

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

This reverts commit 82c5f317.
Signed-off-by: NGanesh Ajjanagadde <gajjanagadde@gmail.com>

Found-by: NClément Bœsch <u@pkh.me>
Signed-off-by: NGanesh Ajjanagadde <gajjanagadde@gmail.com>

This actually fixes an incorrect float literal. It is believed by
examining the precision that the literals were all pre-computed as
floats, resulting in this needless loss of precision. There is no
benefit to keeping such reduced precision:
1. These constants are used for static array computation, hence
compile-time.
2. They will be treated as doubles anyway, since f specifier was not
present.
Signed-off-by: NGanesh Ajjanagadde <gajjanagadde@gmail.com>

Fixed point value remains unchanged.
Signed-off-by: NGanesh Ajjanagadde <gajjanagadde@gmail.com>

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

This uses M_SQRT1_2, M_SQRT2 instead of the actual literals. This yields
greater precision in some places in avcodec/ac3, while fixed point
values remain unchanged.
Reviewed-by: NClément Bœsch <u@pkh.me>
Signed-off-by: NGanesh Ajjanagadde <gajjanagadde@gmail.com>

This uses M_SQRT1_2, M_SQRT2 instead of the actual literals. Fixed point
values remain unchanged.
Patch tested with FATE on x86.
Signed-off-by: NGanesh Ajjanagadde <gajjanagadde@gmail.com>

This simplifies the code.
Reviewed-by: NMichael Niedermayer <michael@niedermayer.cc>
Signed-off-by: NGanesh Ajjanagadde <gajjanagadde@gmail.com>

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

* commit '3edac01f':
  qsvenc: fix setting maxrate for VBR
Merged-by: NDerek Buitenhuis <derek.buitenhuis@gmail.com>

* commit '4d8f536b':
  qsvenc: print the actual video parameters used by MSDK
Merged-by: NDerek Buitenhuis <derek.buitenhuis@gmail.com>

* commit 'f6c94457':
  mpegvideo_enc: enable rtp_mode when multiple slices are used
Merged-by: NDerek Buitenhuis <derek.buitenhuis@gmail.com>

* commit 'd80f0a4a':
  mpevideo_enc: disallow multiple slices for h261 and flv
Merged-by: NDerek Buitenhuis <derek.buitenhuis@gmail.com>

* commit 'ca489564':
  movenc: add fallback audio track tref support
Merged-by: NDerek Buitenhuis <derek.buitenhuis@gmail.com>

* commit '79ae1e63':
  avcodec: Define side data type for fallback track
Merged-by: NDerek Buitenhuis <derek.buitenhuis@gmail.com>