Reviewed-by: Thilo Borgmann

* commit '448c8cfe':
  movenc: Support setting fragment_index before the moov atom is written

Conflicts:
	libavformat/movenc.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '0c5e380c':
  movenc: Don't rely on the fragment index for vc1 info gathering
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit 'cf402d6f':
  rtpenc_mpegts: Set chain->rtp_ctx only after avformat_write_header succeeded
Reviewed-by: Andreas Cadhalpun <andreas.cadhalpun@googlemail.com>
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit 'c83dd2d2':
  rtpenc_mpegts: Free the right ->pb in the error path in the init function
Reviewed-by: Andreas Cadhalpun <andreas.cadhalpun@googlemail.com>
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '96a06dba':
  FATE: add support for testing hwaccels

Conflicts:
	tests/Makefile
	tests/fate-run.sh
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit 'dc7536ca':
  avconv: do not abort immediately if initializing hwaccel fails
Merged-by: Michael Niedermayer <michaelni@gmx.at>

This way, the caller doesn't need to coordinate setting the option
after the moov atom has been written. The downside is that it is
no longer possible to use the option for checking whether the moov
atom already has been written, but a caller is able to keep track
of that by other means anyway.
Signed-off-by: Martin Storsjö <martin@martin.st>

The previous use of the mov->fragments field, for determining whether
written packets were part of the first fragment or not, didn't
work as intended when using the empty_moov flag.
Signed-off-by: Martin Storsjö <martin@martin.st>

By making sure we at each time only have one pointer set, either a
local variable or one in the context, we avoid potential double frees
in the cleanup routines. If chain->rtp_ctx is set, it is closed by
calling avformat_write_trailer, but that shouldn't be called unless
avformat_write_header succeeded.

This issue was pointed out by Andreas Cadhalpun.
Signed-off-by: Martin Storsjö <martin@martin.st>

This fixes a typo from 8e32b1f0.
Signed-off-by: Martin Storsjö <martin@martin.st>

exit_program() will try to free the decoders, which is not a good idea
from within get_format().
Return an error instead.

This fixes artifacts when the dimensions are exact powers of 2
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Previous-version-reviewed-by: Tomas Härdin <tomas.hardin@codemill.se>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

So far it is only set in roq_encode_frame, but it is used in
roq_encode_end to free the coded_frame. This currently segfaults if
roq_encode_frame is not called between roq_encode_init and
roq_encode_end.
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

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

* commit '93f79481':
  libvpx: Fix mixed use of av_malloc() and av_reallocp()
Merged-by: Michael Niedermayer <michaelni@gmx.at>

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

* commit 'ad94c6ca':
  siff: Use the correct type for packet size variables

Conflicts:
	libavformat/siff.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '9f25a109':
  matroskaenc: Also validate chapter end time
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit 'bfeb83a8':
  rtpdec_hevc: Drop extra sanity check for size of input packet
Merged-by: Michael Niedermayer <michaelni@gmx.at>

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

allowing access to the size but not the extradata itself is not useful
and could lead to potential problems if writing happens through this field
Reviewed-by: Andreas Cadhalpun <andreas.cadhalpun@googlemail.com>
Reviewed-by: Lukasz Marek <lukasz.m.luki2@gmail.com>
Reviewed-by: Nicolas George <george@nsup.org>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

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

If resyncing leads to the same position as previously, it will again
lead to a resync attempt, resulting in an infinite loop.

Thus don't seek back beyond the last syncpoint.
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

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

This buffer is resized when vpx_codec_get_cx_data() returns a
VPX_CODEC_STATS_PKT packet.

CC: libav-stable@libav.org
Signed-off-by: Vittorio Giovara <vittorio.giovara@gmail.com>

The avio functions used here return an unsigned value.
Also reduce a variable scope.

CC: libav-stable@libav.org
Bug-Id: CID 1258461

This prevents it to be written as unsigned. Also add an error message.

CC: libav-stable@libav.org
Bug-Id: CID 1265717

In this case len is always at least 3, since it is checked against
RTP_HEVC_PAYLOAD_HEADER_SIZE + 1 before entering the switch block.

Bug-Id: CID 1238784

A negative time base can trigger assertions.
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

* commit '46d4d857':
  movenc: Avoid writing separate flags for the first sample if not necessary
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '00d751d4':
  movenc: Set tfhd default sample flags based on actual samples, if possible
Merged-by: Michael Niedermayer <michaelni@gmx.at>

The examples used to drop frames in most cases which
was probably not what users expected.

The mpegvideo decoder has set intra_dc_precision since forever.

Reported-by: Christoph Gerstbauer

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