The previous code ended in multiple different infinite
loops. See stl_ten_1_big.sfd as example with and without zzuf
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Signed-off-by: Justin Ruggles <justin.ruggles@gmail.com>

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Signed-off-by: Justin Ruggles <justin.ruggles@gmail.com>

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

The reference encoder does not generate any streams using this, but the
reference decoder can handle it, so we should as well.

This isn't used by the reference encoder, but it is supported by the decoder.

Now that there is official documentation from Apple about this, we don't have
to guess anymore.

Add it to doc/APIchanges and update and correct other entries.

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

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

This means that avformat_find_stream_info() now works properly with
multiple threads, so there's no need to force thread count to 1.

Signed-off-by: Janne Grunau <janne-libav@jannau.net>

Signed-off-by: Janne Grunau <janne-libav@jannau.net>

Adds a new member to MpegEncContext to hold the number of used slice
contexts. Fixes segfaults with '-threads 17 -thread_type slice' and
fate-vsynth{1,2}-mpeg{2,4}thread{,_ilace} with --disable-pthreads.

A rebuild in debian revealed that this broke compilation of a number packages:
http://bugs.debian.org/cgi-bin/pkgreport.cgi?users=siretart@tauware.de;tag=ftbfs-libav0.8

Both the Indeo 4 decoder and the SMJPEG demuxer were included in the
0.8_beta1 release.

Codec timebase is not framerate and the situation when it's different
from AVStream.r_frame_rate is quite common, nothing to warn about.

Signed-off-by: Justin Ruggles <justin.ruggles@gmail.com>

Also fix calculation of new position when switching buffers.
This fixes "overread" error messages when seeking.

The fate-h264-bsf-mp4toannexb failures were caused by an integer
overflow of the unneeded multiplication.

Inspired by patch by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Anton Khirnov <anton@khirnov.net>

Signed-off-by: Anton Khirnov <anton@khirnov.net>

None of the raw audio files have timestamps, thus setting
start_time to 0 should be reasonable.
Signed-off-by: Anton Khirnov <anton@khirnov.net>

Signed-off-by: Anton Khirnov <anton@khirnov.net>

Signed-off-by: Anton Khirnov <anton@khirnov.net>

Signed-off-by: Anton Khirnov <anton@khirnov.net>

The extra thread added in {frame_}*thread_init was not taken into
account. Explicitly sets thread_count to 1 if only one CPU core was
detected. Also fixes two typos in comments.

Signed-off-by: Anton Khirnov <anton@khirnov.net>

Palette is as supposed in native endianness. Converting the pal8 output
to rgb24 is thus necessary for identical CRCs on big and little endian
systems.