[Bug 701015] Using a g_thread for gnonlin pipeline update / forward eos on the streaming thread.

GStreamer (bugzilla.gnome.org) bugzilla at gnome.org
Sun May 26 12:06:33 PDT 2013


https://bugzilla.gnome.org/show_bug.cgi?id=701015
  GStreamer | gnonlin | git

--- Comment #6 from Sebastian Dröge <slomo at circular-chaos.org> 2013-05-26 19:06:26 UTC ---
(In reply to comment #5)
> (In reply to comment #3)
> > The alternative would be to only start the thread when it is needed and shut it
> > down afterwards. Starting threads is nowadays very fast, so shouldn't cause
> > much overhead. Starting/stopping the thread in NULL<->READY sounds ok, but
> > actually READY<->PAUSED should be sufficient here too (there's no dataflow in
> > READY, and this thread is only used when dataflow is happening).
> 
> It's not a problem to put the thread creation / joining in READY<->PAUSED, we
> were wondering about the standard way to do so.

I have no strong opinion on this, doesn't really matter much :)

> > Using a GstTask here would probably simplify code a bit, as it has functions to
> > start/stop/pause the thread which you otherwise have to implement yourself. Not
> > exactly rocket science, but still nice to not worry about that ;)
> 
> If you look at the code, the thread starting / stopping is actually extra light
> (one boolean). Don't know if it's really needed to switch to GstTask ?

Well, and the GCond and all around that :) It's not *needed*, just seems nice
to make things even simpler.

-- 
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.


More information about the gstreamer-bugs mailing list