https://bugs.kde.org/show_bug.cgi?id=421662
--- Comment #8 from Gabriel Marcano <gabemarc...@yahoo.com> --- Here's the PowerDevil commit that introduced the delay: https://github.com/KDE/powerdevil/commit/eca79138c15575f6f523a8680919b407f84da2e2 It's a conscious decision by the powerdevil developers. As far as I can tell, even though powerdevil takes 5 seconds to post an inhibit request, they DO track if they receive an uninhibit request before those 5 seconds are up (https://github.com/KDE/powerdevil/blob/master/daemon/powerdevilpolicyagent.cpp#L599). As far as I can tell powerdevil returns a cookie immediately, so elisa should receive one before the inhibit is officially posted. In other words, for some reason elisa isn't sending a request when it should to uninhibit. I "fixed" this problem by doing synchronous dbus commands on the destructor (not ideal, because of possible exceptions? Maybe? Not sure, haven't looked to see how dangerous calling dbus related classes in the destructor is), making sure that if we have a cookie, to send a message to uninhibit. I could do a bit more digging to make sure https://github.com/KDE/elisa/blob/master/src/audiowrapper_qtmultimedia.cpp#L194 is being fired when we're exiting, and that elisa is actually requesting for the power manager to be uninhibited. -- You are receiving this mail because: You are watching all bug changes.
