On Wed, Nov 5, 2014 at 8:08 PM, Pantelis Antoniou <pantelis.antoniou@xxxxxxxxxxxx> wrote: > Hi Grant, > >> On Nov 5, 2014, at 22:01 , Grant Likely <grant.likely@xxxxxxxxxxxx> wrote: >> >> On Tue, 28 Oct 2014 22:33:52 +0200 >> , Pantelis Antoniou <pantelis.antoniou@xxxxxxxxxxxx> >> wrote: >>> The notifier now includes the old_prop argument when updating >>> properties, propagate this API to changeset internals while >>> also retaining the old behaviour of retrieving the old_property >>> when NULL is passed. >>> >>> Signed-off-by: Pantelis Antoniou <pantelis.antoniou@xxxxxxxxxxxx> >> >> I'm still fuzzy on the need for this patch. Is this just an optimization >> so that the property doesn't get looked up twice? Or is there a reason >> when the oldprop really needs to be passed in explicitly? >> > > In the case of overlay's applying a property change the old property has > been already retrieved. Passing it as an argument saves us a traversal of the > property list. > > On the original series were removal was supported, the old property was required > since you can't find the old property anymore on the node (it was on another list). > > When we go back to revisit the removal case, that API is useful. Since we're not doing removal anymore, lets drop this patch. g. -- To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
