Old Intel GPUs expect the reference frame index to the actual surface,
instead of the index into RefFrameList as specified by the spec.

This workaround should be set when using one of the "ClearVideo" decoder
devices.
Signed-off-by: Anton Khirnov <anton@khirnov.net>

The latest H.264 DXVA specification states that the index in this
structure should refer to a valid entry in the RefFrameList of the picture
parameter structure, and not to the actual surface index.

Fixes H.264 DXVA2 decoding on recent Intel GPUs (tested on Sandy and Ivy)
Signed-off-by: Anton Khirnov <anton@khirnov.net>

If framerate estimation is enabled, it makes sense to always apply it,
instead of limiting it to a few specific cases.

This caused mpv (and possibly others) to fallback to software decoding after
seeking a VC1 stream.

Bug-Id: 667
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>

Bug-Id: 664
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>

Signed-off-by: Luca Barbato <lu_zero@gentoo.org>

Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>

Optimized for the default filter length 16.

30% faster opus silk decoding.

Required for arch optimized resampling.

Some applications really mean to send negative pts.

No point in sending the message multiple time.

Messages should start with a capital letter and possibly end with a "."
if they are statements.

Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>

The equations can't overflow or underflow anyway.
Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>

Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>

Fix compilation
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>

This fixes usage of AV_TIME_BASE_Q in C++ applications, which
cannot use compound literals directly in their code.
Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>

Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>

The matrices are not zigzagged.

Fixes artefacting.
Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>

From the ARMv7 NEON version. 16 times faster as the C version, overall
more than 12% faster vorbis decoding on Apple's A7.

30%/25% (fixed/float) faster mp3 decoding on Apple's A7. The floating
point decoder is approximately 7% faster.

Approximately as fast as the ARM NEON version on Apple's A7.

Approximately as fast as the ARM NEON version on Apple's A7.

The parser has no way of knowing video duration, and therefore no video
parsers set it.

Currently, if a decoder sets AVFrame.pts, we'd send the same timestamp
to it twice, which is wrong.

It is not supposed to be changed from outside of lavc.
Set the stream timebase and average framerate instead.

The picture slot can be recycled by select_input_picture and
only current_picture is populated with the valid pts.

Unbreak timestamps when in cbr mode.
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>

Fixes part of Ticket3466
Found-by: Andrey_Karpov / PVS-Studio
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>

Fixes artifacts
Fixes use of freed memory
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>

fix integer overflows
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>

Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>

Also set the RGBA pixel format correctly as the native endian format,
which is what it returns.

This fixes the tests on big endian.
Signed-off-by: Martin Storsjö <martin@martin.st>

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

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