The cleanup routines in CreateDecodingContext() allowed for a deadlock to occur when the creation of surfaces failed early.

Update vaapi-xcb branch to 2.0.4 status.

Some codecs need less surfaces then 30. Relax the assert() to allow
other HW codecs too work too.

stream_Read() is designed to allow a NULL pointer as second argument.
Therefor this commit was wrong. I added a note to avoid making the same
mistake.

This reverts commit a7ef53d4.

(cherry picked from commit 621a5ee0)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

(cherry picked from commit fbe2c9bc)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

(cherry picked from commit 3e954c98)
(cherry picked from commit 0f82e032)
(cherry picked from commit f559e33c)
(cherry picked from commit 67b2f035)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

(cherry picked from commit c98e5d95)
(cherry picked from commit ba1a4be1)
(cherry picked from commit c3977541)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
(cherry picked from commit 64aa5fae)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

Bored as hell of those crashes...
(cherry picked from commit d84b8db9)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

(cherry picked from commit 8494d6cd)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

Manually backported from af480570Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

This stops rejecting 0-byte packets, adds checks everywhere we
access the first byte of a packet, updates the th_decode_packetin()
return value checking so that duplicate frames aren't thrown away,
and updates video pts generation so the resulting frames are
displayed correctly,

Fixes bug #3416.
Signed-off-by: Rafaël Carré <funman@videolan.org>
(cherry picked from commit 9c7198ea)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

This properly handles pre-skip, seeking pre-roll, and end-trim. It
uses the i_nb_samples field of block_t to signal to the decoder how
many samples should be returned after (possibly) skipping some at
the start of the block (for both pre-skip and pre-roll). In
addition, it abuses the i_length field of block_t to signal to the
decoder how many samples to trim from the end of the packet (for
end-trimming).

This patch does not compute correct timestamps for streams which
start at a non-zero offset (e.g., live streams joined in the
middle), nor does it correctly compute the length of a stream. But
neither of those things work for Vorbis, either. I'm leaving them
for a follow-up patch.
Signed-off-by: Rafaël Carré <funman@videolan.org>
(cherry picked from commit 675072a8)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

(cherry picked from commit 84dc54e0)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

(cherry picked from commit 52acf910)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

Fix #1859 #6983
(cherry picked from commit e80525ad)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

(cherry picked from commit fd94f73f)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

(cherry picked from commit 300b4d59)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

Only difference seems to be that in ES_OUT_SET_PCR we don't
have msg_Err showing up while playing ogg files
(cherry picked from commit 7ed15b71)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

Modified-by: Rafaël Carré <funman@videolan.org>
Signed-off-by: Rafaël Carré <funman@videolan.org>
(cherry picked from commit 7258c97a)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

(cherry picked from commit 3f17ad71)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

(cherry picked from commit b68f50f1)

(but not the other way around of course)

On OS X, fontconfig searches font only from /System/Library/Fonts, so this patch adds more search paths to its initialization.
Signed-off-by: Felix Paul Kühne <fkuehne@videolan.org>
(cherry picked from commit 6bad0899)

auhal: fixed audio device configuration callback and added an additional callback to watch the used device's health (close #4834)
(cherry picked from commit 8fbb1fd0)

Conflicts:
	modules/audio_output/auhal.c

This makes sure that the correct header files are checked and used.
That is to say the ncursesw headers as opposed to ncurses.
(cherry picked from commit 65d11fbb)

Conflicts:
	configure.ac

(cherry picked from commit d17ebaf1)

Top crash on Skins2 module on Win32
(cherry picked from commit 9e76fdbc)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

(cherry picked from commit 691b9c33)
(cherry picked from commit f19d35a1)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

(cherry picked from commit 6633ef04)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

(cherry picked from commit 0813e08f)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

(cherry picked from commit e7bfbd1d)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

(cherry picked from commit a1dbddfa)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

(cherry picked from commit 816e52c8)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

(cherry picked from commit ea87467e)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

extras/package: disable the macosx dialog provider in VLC.app release builds, since it should only be used by VLCKit
(cherry picked from commit 1c76df00)

Conflicts:
	extras/package/macosx/configure.sh

(cherry picked from commit dfb4fa14)

(cherry picked from commit 42dd1721)