Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: James Almer <jamrial@gmail.com>

Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: James Almer <jamrial@gmail.com>

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

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

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

This was suggested by wm4 and stefano.
After this patch using align=1 the size used by various functions would not
contain padding, while the palette would be aligned at align>1

This patch makes it required to use align>=4 if the palette is to be accessed
as uint32

As a side-effect It fixes storing pal8 in nut with odd with&height
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

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

Fixes harmless integer overflow
Fixes Ticket5150

No speedloss measured, actually its slightly faster, but please benchmark & double check this
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

This avoid "libavformat/genh.c:43:14: warning: variable coef_splitted set but not used"
Fewer warnings makes it easier to see new and important warnings
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

It is not supported on 32bit.
Signed-off-by: Marton Balint <cus@passwd.hu>

Reviewed-by: Christophe Gisquet <christophe.gisquet@gmail.com>

The test is already slow.

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

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

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

1. Manipulate waiting count in pthread_cond_wait()
2. Use builtin atomic functions to manipulate waiting count
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

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

2.5x faster for 8-bit mode without autovectorization in GCC, 2x
slower with it on x86. However, since the platforms we enable GCC
autovectorization on most probably has support for SSE2
optimization (added in the subsequent commit), this commit should
in general do good.

Fixes bug 821.
Signed-off-by: Carl Eugen Hoyos <cehoyos@ag.or.at>

For xor mode, 1.8x faster on Haswell.

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

Signed-off-by: Alex Agranovsky <alex@sighthound.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

VLC uses hwaccel with frame threads and it works fine, but returning
an error here made it fail.

This regression was introduced in commit 31741aec.
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>

Warning message text by nevcairiel
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

Fixes 5023
Signed-off-by: Alex Agranovsky <alex@sighthound.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

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

Also increase precision to microsecs, and avoid writing creation_time as a
simple tag metadata item.
Signed-off-by: Marton Balint <cus@passwd.hu>

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

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

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

FYI this muxer bails out on parse error and not just warn the user.
Signed-off-by: Marton Balint <cus@passwd.hu>

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

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

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

Also parse timezone information previously ignored in ff_iso8601_to_unix_time.
Signed-off-by: Marton Balint <cus@passwd.hu>