This can happen when seeking via http.

This is interlocked with the control queue as it most probably should.

Also make sure that no new controls are pushed after input_Stop().

Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

And also with dxgidebug.h to show leaks
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

The input is already stopped if it has reached end or error.

After each and every MainLoop() call is a call to End(), which
inconditionally sets the input state to END_S.

```--
  CC       access_output/file.lo
access_output/file.c: In function 'Open':
access_output/file.c:288:9: error: implicit declaration of function 'setmode' [-Werror=implicit-function-declaration]
         setmode (STDOUT_FILENO, O_BINARY);
         ^
cc1.exe: some warnings being treated as errors
make.exe[4]: *** [access_output/file.lo] Error 1
```

--
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

Use signed variables for calculating the buffers to be sent to the component
as negative values may occur.
Signed-off-by: Julian Scheel <julian@jusst.de>
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

Save some memory by allocating fewer buffers. This is still not optimal, but
close to where we can get with the current implementation.
Signed-off-by: Julian Scheel <julian@jusst.de>
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

As we use packetized input we normally don't need more than a handful of input
buffers. Using the default of mmal, which is 20, seems absolutely sufficient.
Signed-off-by: Julian Scheel <julian@jusst.de>
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

To be able to handle seeking properly we need to do a full decoder flush on a
discontinuity. As flushing the in- and output ports alone would not cause the
codecs internal state to be reset we have to disable the ports, reconfigure
and reenable. To ensure the flushing is synchronous the input buffers in
transit are now tracked as well.
Signed-off-by: Julian Scheel <julian@jusst.de>
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

Signed-off-by: Julian Scheel <julian@jusst.de>
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

When running in opaque mode the mmal codec, deinterlace and vout must be used
together, so it is known that all pictures in use are allocated by the mmal
vout and contain an associated mmal_buffer_header. This allows us to enable
the zerocopy feature of mmal, which allocates shared memory between GPU and
CPU for holding the buffer payloads. Albeit the payloads are just small opaque
handles that reference the GPU side pictures it saves a little bit of
performance, because less VCHIQ transfers are required.
Signed-off-by: Julian Scheel <julian@jusst.de>
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

Using mmal_port_pool_allocate/destroy adds some extra safety on maintaining
the pool lifecycle in relation to the component lifecycle. Furthermore it is a
prerequisite to use zerocopy mode for ports.
Signed-off-by: Julian Scheel <julian@jusst.de>
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

If the input port returns a buffer which has no picture attached to it it
would be leaked. While this in fact should not happen it seems sane to at
least release the buffer and spit a warning in case it does happen.
Signed-off-by: Julian Scheel <julian@jusst.de>
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>