On 2013-05-21 15:36, Arun Raghavan wrote: > On Tue, 2013-05-21 at 16:06 +0300, Tanu Kaskinen wrote: >> On Tue, 2013-05-21 at 14:58 +0200, Christian Iversen wrote: >>> So my question to this list: Is module-suspend-on-idle really a >>> requirement? How difficult would this be to fix? >> >> Already fixed in the "next" branch :) > > And jfyi, since you expressed interest in fixing it, the commit is: > > http://cgit.freedesktop.org/pulseaudio/pulseaudio/commit/?h=next&id=cb1ef3c211a90e4ead4738c1059054d6e44d50d9 Ah, good to hear! It was interesting reading the patch. I think I understand the problem now. Without the suspend-on-idle module, there's no hook that fires pa_sink_suspend(s, FALSE, ...) after the rate update, because the module is not listening. Whereas in this patch, a new cause flag is implemented and used, so A) the module will not react to it, and B) it is taken care of outside of the suspend-on-idle module. Is that basically correct? -- Med venlig hilsen Christian Iversen
