This is notably the case with Nikon-Camera generated movies

The issue here is that android-x86 doesn't guarantee an aligned stack, which means some of the inline asm functions will waste an extra register on realigning the stack.

libavcodec does however have stuff in place in all entrypoints that realign the stack on all calls into the library. If you build with --extra-cflags='-mincoming-stack-boundary=4', gcc assumes that the stack actually is aligned already, which the entry points ensure, so this inline asm builds fine.
Pointed-out-by: Martin Storsjö <martin@martin.st>
Signed-off-by: Rafaël Carré <funman@videolan.org>

This package needs exceptions.
Signed-off-by: Rafaël Carré <funman@videolan.org>

This package demands exceptions, but if it is enabled globally, it conflicts with other packages.
Signed-off-by: Rafaël Carré <funman@videolan.org>

Signed-off-by: Rafaël Carré <funman@videolan.org>

That plugin was written for Maemo 4. It barely worked with Maemo 5.
Maemo 6 did not have Hildon. Hildon is unmaintained and unused.

especially:
- don't show outdated items in sub menus
- disable unusable sub menus

close #6998

This avoids listing all video filters (not splitters!). However, the
video splitters are still wrongly included in the --video-filter list.

That way, only sources show up in the list of sources and filters in
the list of filters.

Unfortunately, some callers cannot be trusted not to leak their COM
context, causing COM initialization to fail.

Memory associated with a FcPattern is freed when the pattern is
destroyed, so copying the font filename needs to be done before
the call to FcPatternDestroy.
Signed-off-by: Thierry Reding <thierry.reding@avionic-design.de>
Modified-by: Rafaël Carré <funman@videolan.org> (simplify return)
Signed-off-by: Rafaël Carré <funman@videolan.org>

This reverts commit 5fc0ccda.

Signed-off-by: Rafaël Carré <funman@videolan.org>

Headers normally present under normal Linux are missing in bionic.
Signed-off-by: Rafaël Carré <funman@videolan.org>

Signed-off-by: Rafaël Carré <funman@videolan.org>

The library introduces noise in unclear circumstances. There does not
seem to be any problem on VLC side though. I do not have time and
expertise to troubleshoot this.

The time report callback could only be called from pf_play() so this is
cleaner and less prone to coding mistake.

Also pass the relative drift instead of the absolute play time stamp.
The drift is less subject to scheduling noise, i.e. drift evolves much
slower than current time advances.