> >> That specifically was the intent of the change.  Please discuss with
> >> Ross Harvey.
> >
> > is he reading this list?
> >
> > i think that the change is broken and should be backed out at least.
> 
> I brought the message to his attention.

how long should i wait?

there are several choices.

0. keep it as-is.  (IMO broken)
1. remove the feedback completely.
2. do the feedback strictly.
   (keep how much p_estcpu is inherited from the parent.
   and decay it appropriately)
3. do the feedback in a compromised way.
   (eg. my proposed patch)
4. any other good idea?

YAMAMOTO Takashi