Hi! > > > > That it is same old hack it used to be. There's nothing magic about > > > > decrementer, and I do not think it is _neccessary_ to touch it just > > > > before cli. Just disable it from sysdev handler or something. > > > > > > No, you cannot disable it from sysdev handler because you have to be > > > able to take interrupts. Can we *please* move on to some more productive > > > discussion? > > > > So what happens if you disable it as a platform device? There has to > > be better solution than this. What is so special about decrementer? > > That you are dead since once it's disabled, kernel timers etc... stop > working. > > > We were able to handle MTRR registers on PC. Timer setup fit into the > > model too... > > I don't care what you do on x86. If I tell you I need to take control at > that point, then I need. Beside I really can't understand why you are so > dead set against an arch hook there. It's not like this will do anything > to you... it can't propagate to drivers, it's purely arch stuff, so > what ? There are about 7 places where we do stop_platform_devices() local_irq_disable() stop_sysdevs(). You call your hook from one of them and I suspect you need to call it from more than one. Plus, the hook needs to have documented semantics, so people know if they should do local_irq_disable() or if they should call your hook. Pavel -- (english) http://www.livejournal.com/~pavelmachek (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html _______________________________________________ linux-pm mailing list linux-pm@xxxxxxxxxxxxxxxxxxxxxxxxxx https://lists.linux-foundation.org/mailman/listinfo/linux-pm
