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

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

The following variants are implemented:
lipshitz noise shaping dither
shibata noise shaping dither
low shibata noise shaping dither
high shibata noise shaping dither
f-weighted noise shaping dither
modified-e-weighted noise shaping dither
improved-e-weighted noise shaping dither

Data tables taken from SOX
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

* qatar/master:
  xan: Convert to bytestream2
  oggenc: add a page_duration option and deprecate the pagesize option
  x86: lavr: add SSE2/AVX dither_int_to_float()

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

* commit '1fb8f6a4':
  x86: lavr: add SSE2 quantize() for dithering
  doc/APIchanges: fill in missing dates and hashes.
  rtpdec_vp8: Request a keyframe if RTP packets are lost

Conflicts:
	doc/APIchanges
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '86d9181c':
  rtpdec: Support sending RTCP feedback packets

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

* commit '42805eda':
  rtpdec: Store the dynamic payload handler in the rtpdec context
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '9c80ed83':
  rtpdec_vp8: Avoid a warning about a possibly unused variable
  rtpdec_vp8: Make sure the previous packet is returned
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '92e354b6':
  rtpdec_vp8: Set the timestamp when returning a deferred packet
  hlsenc: Make the start_number option set the right variable

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

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

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

Regression since: ea6da80c
Fixes Ticket2123

I cannot reproduce any regressions by flipping the wrong condition
to how it should have been.

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

Sometimes the muxer modifies the packet, like for instance lavf/mp3enc
changing pkt->destruct in order to keep a copy. These changes must be
kept, even though the muxer behaviour is questionable. Regression since
0072116c.

Fixes #2124.

This uses page duration instead of byte size to determine when to buffer
the page. Also, it tries to avoid continued pages by buffering the current
page if there are already packets in the page and adding the next packet
would require it to be continued on a new page. This can improve seeking
performance.

The default page duration is 1 second, which is much saner than filling
all page segments by default.

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

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

This fixes a infinite loop with -loop and -vframes
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

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

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

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

This sends NACK for missed packets and PLI (picture loss indication)
if a depacketizer indicates that it needs a new keyframe, according
to RFC 4585.

This is only enabled if the SDP indicated that feedback is supported
(via the AVPF or SAVPF profile names).

The feedback packets are throttled to a certain maximum interval
(currently 250 ms) to make sure the feedback packets don't eat up
too much bandwidth (which might be counterproductive). The RFC
specifies a more elaborate feedback packet scheduling.

The feedback packets are currently sent independently from normal
RTCP RR packets, which is not totally spec compliant, but works
fine in the environments I've tested it in. (RFC 5506 allows this,
but requires a SDP attribute for enabling it.)
Signed-off-by: Martin Storsjö <martin@martin.st>

This allows calling other dynamic payload handler functions if
needed.
Signed-off-by: Martin Storsjö <martin@martin.st>

The warning is a false positive, but I prefer actually initializing
it over masking it with av_uninit, since the code is not performance
critical.
Signed-off-by: Martin Storsjö <martin@martin.st>

This is a bug from c7d4de3d - if the previous frame wasn't
returned yet (due to missing the final packets), but we have
enough data of it to return the first partition, we write that into
pkt and set returned_old_frame. That commit forgot returning 0 for
the case where this current packet didn't have the end_packet flag
set.
Signed-off-by: Martin Storsjö <martin@martin.st>

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

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

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

Found-by: durandal_1707
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Unlike the original ff_start_frame code, the incoming reference
may be freed before that point.

Fix CID966654.

* qatar/master:
  rtsp: Respect max_delay for the reordering queue when using custom IO
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '8729698d':
  rtsp: Recheck the reordering queue if getting a new packet
  lavr: log channel conversion description for any-to-any functions
  lavr: mix: reduce the mixing matrix when possible
  lavr: cosmetics: reindent
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '074a00d1':
  lavr: add a public function for setting a custom channel map
  lavr: typedef internal structs in internal.h
  doc: Extend commit message section

Conflicts:
	doc/APIchanges
	doc/developer.texi
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit 'ac2603be':
  doc: Mention memory allocation in the fuzz testing section
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit 'dc3e12d1':
  doc: Mention zzuf in the fuzz testing section
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit 'b5f9b9ac':
  doc: Merge disjointed bits about emailing patches

Conflicts:
	doc/developer.texi
Merged-by: Michael Niedermayer <michaelni@gmx.at>