On Tue, 2014-04-15 at 08:05 +0200, Peter Meerwald wrote: > Hello Tanu, > > some information that describes the bug would be good, such as how to > trigger, what the observed effect is, what goes wrong > > the fix is not obvious (to me at least) Sorry for being lazy, you're right, I should explain the change. The bug is that "i->save_muted = i->save_muted || mute" just makes no sense at all. I looked up the history of this line, and it was introduced when reverting the volume ramping code. The reverting apparently involved some manual intervention, and this bug slipped in at that point. > the negated condition, i->muted == mute, looks a bit weird Yes, I presume the negation used because when we still had pa_bool_t, booleans could in theory be defined as int. The negation is unnecessary with the bool type. I'll remove the negation in v2. -- Tanu
