On Fri, Jun 19, 2020 at 06:39:44PM +0100, Qais Yousef wrote:
> On 06/19/20 19:30, Peter Zijlstra wrote:
> > On Thu, Jun 18, 2020 at 08:55:24PM +0100, Qais Yousef wrote:
> >
> > > + for_each_clamp_id(clamp_id) {
> > > + memset(uc_rq[clamp_id].bucket,
> > > + 0,
> > > + sizeof(struct uclamp_bucket)*UCLAMP_BUCKETS);
> > > +
> > > + uc_rq[clamp_id].value = uclamp_none(clamp_id);
> >
> > I think you can replace all that with:
> >
> > *uc_rq = (struct uclamp_rq){
> > .value = uclamp_none(clamp_id),
> > };
> >
> > it's shorter and is free or weird line-breaks :-)
>
> Sure. I just sent v2 so that people will be encouraged to run tests hopefully.
> But will fix in v3.
>
> Do we actually need to 0 out anything here? Shouldn't the runqueues all be in
> BSS which gets initialized to 0 by default at boot?
>
> Maybe better stay explicit..
C99 named initializer (as used here) explicitly zero initializes all
unnamed members. Is that explicit enough? ;-)
