Signed-off-by: Nicolas George <nicolas.george@normalesup.org>
Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
(cherry picked from commit 90441276)

Signed-off-by: Nicolas George <nicolas.george@normalesup.org>
Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
(cherry picked from commit fe174fc8)

Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
(cherry picked from commit f4a86bc9)

Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
(cherry picked from commit 5a6de4e7)

Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
(cherry picked from commit 74571e33)

wtv: do not use flag in stream_guid chunk to determine if stream is valid, as this method is unreliable

This fixes roundup issue 2556.
Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
(cherry picked from commit e4f85b84)

Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
(cherry picked from commit edbb0c07)

Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
(cherry picked from commit e8dcd730)

Using a preset file caused the address of a stack variable to be stored
in opt_names/values. This change causes the strings to be dup'd then
freed in uninit_opts.
Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
(cherry picked from commit 60ff9de6)

Add free to uninit_opts and relocate opt_names to same
Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
(cherry picked from commit 3a6a9cdf)

No behavior change; this makes DTS reliable with the next patch.
Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
(cherry picked from commit 6b474953)

This fixes roundup issue 2583.
Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
(cherry picked from commit 185a155e)

Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
(cherry picked from commit 5306bf41)

If required, the caller can do this itself. ff_write_chained rescales
timestamps as necessary, and all current callers of rtpenc_chain
use ff_write_chained, making this timebase copy unnecessary.
Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
(cherry picked from commit 397ffde1)

Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
(cherry picked from commit b22dbb29)

Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
(cherry picked from commit ff197489)

This function is useful for freeing data structures allocated by
muxers, which currently have to be freed manually by the caller.
Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
(cherry picked from commit f124b087)

This avoids having the chained AVStream->codec point to the same
AVCodecContext owned by the outer AVStream. The downside is that
changes to the AVCodecContext made after calling av_write_header
cannot be detected automatically within the chained muxer.

This avoids having to manually unlink the chained AVStream->codec
by setting it to null before freeing the chained muxer via generic
freeing functions.
Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
(cherry picked from commit 1338dc08)

(cherry picked from commit 8a2c99b4)

Convert hev_thresh logic to a LUT, simplify mbedge_lim calculation.
(cherry picked from commit 79dec154)

Signed-off-by: Mans Rullgard <mans@mansr.com>
(cherry picked from commit 290849e2)

Signed-off-by: Mans Rullgard <mans@mansr.com>
(cherry picked from commit 523d9407)

(cherry picked from commit a1b227bb)

(cherry picked from commit dd18c9a0)

Make avfilter_graph_free() free not only the internal structures, but
also the allocated graph, and set the graph pointer to NULL for
increased safety.

Simplify usage.
Signed-off-by: Mans Rullgard <mans@mansr.com>
(cherry picked from commit 4359288c)

Make avfilter_graph_parse() only release the internal structures
allocated during the parsing, and leave to free the graph itself to
the calling code.

This approach looks cleaner, as the graph is not allocated by the
function.
Signed-off-by: Mans Rullgard <mans@mansr.com>
(cherry picked from commit e8e5dde7)

Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
(cherry picked from commit e6fb5a4f)

Signed-off-by: Nicolas George <nicolas.george@normalesup.org>
Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
(cherry picked from commit 51b317d2)

Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
(cherry picked from commit f61dee2f)

Signed-off-by: Mans Rullgard <mans@mansr.com>
(cherry picked from commit a9d921cb)

Signed-off-by: Mans Rullgard <mans@mansr.com>
(cherry picked from commit 151595fe)

Tested to compile with lavc major bump.
Signed-off-by: Mans Rullgard <mans@mansr.com>
(cherry picked from commit b2ed95ec)

This fixes yet another memory leak, present since SVN rev 25418.
Signed-off-by: Mans Rullgard <mans@mansr.com>
(cherry picked from commit 1f56f5ed)

Now the first argument is URLContext *h. However, the function logs to
LOG_CONTEXT, which is #defined as 's' for new lavf major versions.

Therefore, rename h -> s.
Signed-off-by: Mans Rullgard <mans@mansr.com>
(cherry picked from commit 9ad4c65f)

Signed-off-by: Mans Rullgard <mans@mansr.com>
(cherry picked from commit e65b1934)

Signed-off-by: Mans Rullgard <mans@mansr.com>
(cherry picked from commit 75fd0668)

Signed-off-by: Mans Rullgard <mans@mansr.com>
(cherry picked from commit f5b82f45)

Signed-off-by: Mans Rullgard <mans@mansr.com>
(cherry picked from commit aa42cce5)

Signed-off-by: Mans Rullgard <mans@mansr.com>
(cherry picked from commit 437fb1c8)