[Bug 733169] decodebin: improve deadend pads handling

[GStreamer (bugzilla.gnome.org)]

https://bugzilla.gnome.org/show_bug.cgi?id=733169
  GStreamer | gst-plugins-base | unspecified

--- Comment #4 from Thiago Sousa Santos <thiagossantos at gmail.com> 2014-07-17 23:00:22 UTC ---
(In reply to comment #3)
> (From update of attachment 280664 [details])
> Not 100% sure this makes sense. A deadend pad could still receive buffers and
> then at some point an EOS. Right?

The drained is being only set on the EOS handling, so definitely there was an
EOS on the chain. Can it happen that the pad isn't the last one?

> 
> So this would be an optimization that allows deadend pads to be considered
> drained earlier already if they're the last ones?

I haven't reinforced a check that they are the last ones, but in my tests this
seems to happen only with them. If they are not marked as drained then the next
group switch will never happen.

> 
> Or what am I missing here?

I might also be missing a scenario where this EOS handler is called for a non
final pad. Is this possible?

-- 
Configure bugmail: https://bugzilla.gnome.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the QA contact for the bug.
You are the assignee for the bug.