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

One step in untangling the mpegvideo code and fixing some problems in
the order that initialization is being done in h263dec and h261dec.

Intrinsics only used on aarch64 since the existing ARMv7 NEON asm
is slightly faster (Cortex-A9, gcc-4.8, micro-benchmarks and full
decoding time).

Signed-off-by: James Yu <james.yu@linaro.org>
Signed-off-by: Janne Grunau <janne-libav@jannau.net>

It is not used outside of the file.

Since error resilience uses AVFrame pointers instead of references it
has to copy NULL pointers too. After a codec flush the last/next frame
pointers in MpegEncContext are NULL and the old pointers remaining in
ERContext are invalid. Fixes a crash in vlc for android thumbnailer.
Reported and debugged by Adrien Maglo <magsoft@videolan.org>.

This describes more accurately what this field is for.

The most interesting parts are initialization in ff_MPV_common_init() and
uninitialization in ff_MPV_common_end().

ff_mpeg_unref_picture and ff_thread_release_buffer have additional NULL
checks for Picture.f, because these functions can be called on
uninitialized or partially initialized Pictures.

NULL pointer checks are added to ff_thread_release_buffer() stub function.

Signed-off-by: Vittorio Giovara <vittorio.giovara@gmail.com>

Drop the mpegvideo dependency for svq3 in configure.

This is done to disentangle ER from mpegvideo. In order to use a
classic Picture, callers can use ff_mpeg_set_erpic() or use a custom function
to set the fields. Please note that buffers need to be allocated before
calling ff_er_frame_end().

Added libavutil/timer.h include to all files with {START,STOP}_TIMER.

Allocating edges will be removed from generic code, so the encoder must
handle this explicitly.

Several decoders disable those anyway and they are not measurably faster
on x86. They might be somewhat faster on other platforms due to missing
emu edge SIMD, but the gain is not large enough (and those decoders
relevant enough) to justify the added complexity.

It has been disabled as 'broken' over 3 years ago in b716a792

This introduces some code duplication. However, much of it should go
away once the decoders stop using MpegEncContext.

Otherwise the generic code will unref them, which can then result in
last_picture_ptr == current_picture_ptr, which causes deadlocks at least
in rv40.

Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC:libav-stable@libav.org

This is a temporary workaround to allow deprecating
avcodec_get_frame_defaults(). The proper solution will be using a
properly allocated AVFrame in Picture.

ff_mpeg_unref_picture() already resets the frame.

It is only called from one place in mpegvideo.c now. Also remove the
remove_current parameter, which is always 1.

It is not called from outside of mpegvideo.c anymore.

It is used for encoding only.

It is encoding-only, frame threading is not used for encoding.

The field remains in MpegEncContext for use in the encoder.

data[0] may be NULL for valid frames with hwaccel pixel formats.

XvMC has long ago been superseded by newer acceleration APIs, such as
VDPAU, and few downstreams still support it. Furthermore XvMC is not
implemented within the hwaccel framework, but requires its own specific
code in the MPEG-1/2 decoder, which is a maintenance burden.

Alpha has been end-of-lifed and no more test machines are available.

It has not been actually used since
37045e422903695e610cca6ecb753df643ab9380, when the broken vismv code was
removed.

That check is easily reachable by faulty input.

CC:libav-stable@libav.org
Reported-by: Torsten Sadowski <tsadowski@gmx.net>

This fixes breakage in a few fate tests on certain setups
(that for some reason didn't break on OS X) after the previous
commit (8812a8057). Currently, some video streams are initialized
in ff_MPV_common_init with width/height set at 0 and only changed
to a proper video size with ff_MPV_common_frame_size_change later.

The breakage was diagnosed by Anton Khirnov.

Signed-off-by: Martin Storsjö <martin@martin.st>