[Bug 744106] Maximum latency handling broken everywhere

[bugzilla at gnome.org]

https://bugzilla.gnome.org/show_bug.cgi?id=744106

--- Comment #17 from Sebastian Dröge (slomo) <slomo at coaxion.net> ---
<wtay> slomo, yeah, I saw it. popcorn time
<wtay> slomo, I think the only problem is that the default for basesrc should
be 0,0 instead of 0,-1
<slomo> wtay: that might be one change, but how should overall be the expected
behaviour? -1 meaning infinite maximum latency, meaning that usually elements
just pass that through... unless they themselves have leaky queueing behaviour
like audioringbuffer, in which case they would reset it to their queue size?
also do you agree that all our queues should do something with the latency
query somehow? :)
<wtay> slomo, yes, -1 in infinity and stays that way unless you have a leaky
queue that actually throws away things
<wtay> slomo, queues should do something with the query. I think queue does
(but only when time limits are set, bytes and buffer limits can't be taken into
account)
<slomo> they could be taken into account if upstream properly implements the
convert query i guess :)
<wtay> probably
<slomo> ok, i think then it all makes sense now and needs some clarification in
part-latency.txt :) i'll do that later and fix all the code that is currently
doing weird things
<slomo> thanks

-- 
You are receiving this mail because:
You are the QA Contact for the bug.
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/gstreamer-bugs/attachments/20150210/bedd2644/attachment-0001.html>