On Thursday 12 April 2007 8:03 am, Rafael J. Wysocki wrote: > On Thursday, 12 April 2007 13:23, Benjamin Herrenschmidt wrote: > > Why not give this added flexibility ? Archs who don't care don't need to > > bother and it will make us happy... it's not like we are about to -add- > > burden to other architectures. > > Well, I think it would be reasonable to add the quiesce()/activate() hooks for > all of the above reasons. Makes sense to me. Though I'd rather see pm_set_ops() patch the default hooks than expect all platforms to change ... that would shrink the size of the patch adding these, as well as the potential cost of removing them. > However, once we've done that, it'll be quite > difficult to remove them, so we should better be sure they are really really > needed and there's no other way to implement what you need (or all of the > alternative ways are far worse). In general I think too much of the way PM "works" now is a bit more due to convenient side effects than by clear intent. So while I agree that adding hooks should be done with care, this one certain seems to be a step in the right direction. - Dave _______________________________________________ linux-pm mailing list linux-pm@xxxxxxxxxxxxxxxxxxxxxxxxxx https://lists.linux-foundation.org/mailman/listinfo/linux-pm
