If --syslog is specified on the commandline then vlc loads the logger in file mode first, because --file-logging is the default. This is counter intuitive for users, they expect *only* syslog mode to be active. Note that the logger module can only have one mode active at the sametime, either html, text or syslog.
(cherry picked from commit 5e2f7b2f90dbab0488d62557ba59159925a88e3a)

(cherry picked from commit a45ba22e)

(cherry picked from commit 918910e2)

(cherry picked from commit d3ab3afe)

(cherry picked from commit 6094cb67)

We don't want to leak descriptors to child PROCESSES (!= threads) such
as the pseudo-CGI interface. The socket wrappers ahve already been doing
that for a while. Unfortunately, there are still many "leaks", through
underlying libraries, opendir(), fopen(), etc. And then, this is not
thread-safe (we leak if another thread calls fork() between open() and
fcntl()). This is a well-known limitation in POSIX anyway.
(cherry picked from commit 90ba04f4)

(cherry picked from commit 471fa3a8)
Signed-off-by: Derk-Jan Hartman <hartman@videolan.org>

(cherry picked from commit a5ad91e7)
Signed-off-by: Derk-Jan Hartman <hartman@videolan.org>

(cherry picked from commit 29a15bf0)
Signed-off-by: Derk-Jan Hartman <hartman@videolan.org>

This is a feature, but almost fully copied from existing avcodec code.

Cherry-Picked from:
[d47c4459]
[74095abc]
[2b737862]

(cherry picked from commit 64632922)
Signed-off-by: Derk-Jan Hartman <hartman@videolan.org>

(cherry picked from commit 356fafa5)
Signed-off-by: Derk-Jan Hartman <hartman@videolan.org>

(cherry picked from commit eeaa54cf)
Signed-off-by: Derk-Jan Hartman <hartman@videolan.org>

(cherry picked from commit 3f7a8904)
Signed-off-by: Derk-Jan Hartman <hartman@videolan.org>

(cherry picked from commit 74d28b3e)
Signed-off-by: Derk-Jan Hartman <hartman@videolan.org>

(cherry picked from commit 9e211e0c)
Signed-off-by: Derk-Jan Hartman <hartman@videolan.org>

Signed-off-by: Felix Paul Kühne <fkuehne@videolan.org>

The code was never enabled.

It should fix a potential segfault when the vout is closed.

(cherry picked from commit 8a9bd2d8)
Signed-off-by: Derk-Jan Hartman <hartman@videolan.org>

(cherry picked from commit 264444a1)
Signed-off-by: Derk-Jan Hartman <hartman@videolan.org>

(cherry picked from commit 2624f9af)
Signed-off-by: Derk-Jan Hartman <hartman@videolan.org>

(cherry picked from commit aab61e71)
Signed-off-by: Derk-Jan Hartman <hartman@videolan.org>

(cherry picked from commit cb89f6d5)
Signed-off-by: Derk-Jan Hartman <hartman@videolan.org>

(cherry picked from commit fc4b7b3c)
Signed-off-by: Derk-Jan Hartman <hartman@videolan.org>

(cherry picked from commit 9dd13ebf)

(cherry picked from commit 1f0135f4)
Signed-off-by: Derk-Jan Hartman <hartman@videolan.org>

It has little benefits in general, (esp. in our usecase) and huge disadvantages for files on network mounts.
"Basically, IOKit should be so fast and efficient, that there are few usecases left for mmap() on Darwin."
(cherry picked from commit 400b6736)
Signed-off-by: Derk-Jan Hartman <hartman@videolan.org>