On 08/12/2013 12:18 PM, Tanu Kaskinen wrote: > On Mon, 2013-08-12 at 09:15 +0200, David Henningsson wrote: >> On 08/09/2013 08:57 AM, Tanu Kaskinen wrote: >>> Monitor sources don't have the update_rate() callback set, so their rate was >>> not being changed when changing the sink rate. >> >> For better understanding (I was a little confused first), one could add >> a sentence to the commit comment saying e g "This patch fixes this by >> changing the rate correctly, even if the update_rate callback is not set". > > Ok. > >> Can this also happen to sinks, that there might be types of sinks that >> do not have update_rate set, and might fall into the same bug? > > I don't think there are such sinks. Monitor sources are special, because > they are automatically created by the core. All other sources and sinks > are supposed to set update_rate() if they support rate switching. > So, with this patch, it looks like sources not supporting rate switching, will not set update_rate initially, and then crash in assert(monitor_of) ? -- David Henningsson, Canonical Ltd. https://launchpad.net/~diwic
