Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

about 1% better compression
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

* qatar/master:
  configure: Separate MinGW32 and MinGW64 libc handling
Merged-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Include the freetype header, in-directly through a macro, like it
is done in the drawtext filter. Do not break if the header is moved.

Unfortunately the drawtext filter included the file where the include
macros are defined in a wrong way. This is not needed and breaks the
build. Remove that #include line too.
Signed-off-by: Alexander Strasser <eclipse7@gmx.net>

Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

* qatar/master:
  libvpx: do not mark VP9 as experimental when using libvpx >= 1.3.0

Conflicts:
	libavcodec/libvpxdec.c
	libavcodec/libvpxenc.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: James Almer <jamrial@gmail.com>
Reviewed-by: James Zern <jzern@google.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Diego Biurrun <diego@biurrun.de>

Signed-off-by: James Almer <jamrial@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

* rbultje/vp9-simd:
  vp9: implement top/left half (4x4) sub-8x8-IDCT.
  vp9: split pre-load of 11585x2 out of 1d idct macro.
  vp9: minor refactorings in idct ssse3 assembly.
  vp9: split x86 assembly in two files.
Merged-by: Michael Niedermayer <michaelni@gmx.at>

For that specific case (eob>3&&eob<=12), runtime of idct8x8 goes from
668 to 477 cycles. For all idct8x8, runtime goes from 521 to 490 cycles.

This allows us to load it only once, instead of twice, in this function.

Make register usage in macros explicit; change mulsub_2w_4x to use 2
instead of 3 temp registers.

(And in future, loopfilter or intra pred could be put in their own
respective files also.)

* cus/stable:
  ffplay: set default window size before starting audio
  ffplay: factor out function setting default window size
  ffplay: remove no longer necessary codec flush
  ffplay: do not wait for flushing the picture queue on flush packet
Merged-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

* qatar/master:
  dcadec: Decode LFE to avoid adding random data when downmixing with LFE

See: 6e7de114Merged-by: Michael Niedermayer <michaelni@gmx.at>

This avoids collecting statistics for rfps values that very likely
will get rejected later.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

it makes no sense for other streams
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

makes the first call to sws_getContext() 1ms faster
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

this safes about 0.1ms
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Tim Walker <tdskywalker@gmail.com>
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>

These occur when no context is set for example, thus they are common
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

about 1/3 or so faster
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Should fix issues compiling lavr's Windows resource file.
Signed-off-by: James Almer <jamrial@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

* lukaszmluki/master:
  lavd/fbdev_enc: add default device
Merged-by: Michael Niedermayer <michaelni@gmx.at>

Fixes ticket #2381.
Signed-off-by: Marton Balint <cus@passwd.hu>

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

It was introduced in c2e8691c, but since we no
longer no longer provide a custom get_buffer callback, the original cause of
the issue is gone.
Signed-off-by: Marton Balint <cus@passwd.hu>

It is no longer necessary. Also do frame timer and video current pos reset in
the main thread because with the wait removed, the timing would not be optimal
in the read thread.
Signed-off-by: Marton Balint <cus@passwd.hu>

Since we don't write lavf's string when bitexact is requested, this will
prevent the tag from being copied from the source stream.
Signed-off-by: James Almer <jamrial@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Use device specified by environment variable FRAMEBUFFER or /dev/fb0 as default.
Signed-off-by: Lukasz Marek <lukasz.m.luki@gmail.com>