On Tue, 1 Jun 2010 12:43:21 +0200 (CEST) Thomas Gleixner <tglx@xxxxxxxxxxxxx> wrote: > On Tue, 1 Jun 2010, florian@xxxxxxxxxxx wrote: > > > The new extreme value is only depending on the old extreme value and > > the changed value. > > And how does that update to the next applicable constraint when the > current constraint is removed ? Your patch is creating a one way > decision. keeping the list sorted or using an rbtree is the answer then. if O(1) behaviour is needed maybe another constraint type could be implemented. Maybe using a bool or even a bitmap (to allow for 32 constraint-values). Is another constraint-type worthwile? > > > Signed-off-by: Florian Mickler <florian@xxxxxxxxxxx> > > --- > > This version actually compiles... :) > > That does not make it work. > > tglx indeed. Cheers, Flo _______________________________________________ linux-pm mailing list linux-pm@xxxxxxxxxxxxxxxxxxxxxxxxxx https://lists.linux-foundation.org/mailman/listinfo/linux-pm
