[pulseaudio-tickets] [Bug 65703] High latency in HTTP streaming

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=65703

--- Comment #58 from Wang Xingchao <wangxingchao2011 at gmail.com> ---
(In reply to comment #56)
> I very much doubt that adding a rewind callback is the correct fix. The
> reason is that nothing in the code base currently implements the source
> output process_rewind() callback, so it doesn't seem to be important (I
> don't fully understand how source rewinding works - it's quite different
> from sink rewinding).
> 
> Also, source outputs can be rewound only when connected to a monitor source.
> Are you using a monitor source for testing?

yes, the source outputs connected to the monitor source "upnp" in the test
case. please look at comment #47.
And it's easy to reproduce the issue with comment#47, you can check from the
log the callback was called, and the latency reduced effeciently.i.e. without
the patch, when stop the first stream which rendering MP3 to monitor source, i
still hear musck playing for almost ~2 seconds. With the patch applied, it
would shtop playback immediately.


> If not, then the callback is
> never called. How can the patch then have any effect on the latency? It
> seems that pa_source_output_push() behaves differently if the
> process_rewind() callback is set (even if it's never called):
> 
>     limit = o->process_rewind ? 0 : o->source->thread_info.max_rewind;

-- 
You are receiving this mail because:
You are the QA Contact for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/pulseaudio-bugs/attachments/20130709/397c9249/attachment.html>