Re: [PATCH v2 1/2] PM / OPP: add non-OF versions of dev_pm_opp_{cpumask_,}remove_table

[Sudeep Holla]

On 29/04/16 05:07, Viresh Kumar wrote:
> On 28-04-16, 18:07, Sudeep Holla wrote:
> > diff --git a/drivers/base/power/opp/core.c b/drivers/base/power/opp/core.c
> > index 433b60092972..e59b9e7c31ba 100644
> > --- a/drivers/base/power/opp/core.c
> > +++ b/drivers/base/power/opp/core.c

[...]

> >   {
> >   	struct opp_table *opp_table;
> >   	struct dev_pm_opp *opp, *tmp;
> > @@ -1884,7 +1885,7 @@ void dev_pm_opp_of_remove_table(struct device *dev)
> >   	if (list_is_singular(&opp_table->dev_list)) {
> >   		/* Free static OPPs */
> >   		list_for_each_entry_safe(opp, tmp, &opp_table->opp_list, node) {
> > -			if (!opp->dynamic)
> > +			if (!opp->dynamic || (opp->dynamic && remove_dyn))
>
> Well, that's a funny one :)
>
> The second conditional statement doesn't require opp->dynamic, as that is
> guaranteed to be true, as the first condition failed.
>
> So this should be:
>
> if (remove_all || !opp->dynamic)

Yes, initially I thought we may need to support a mix where both static
entries from DT and dynamic entries will be present and this function
must remove the latter only. But I soon realized that it's not required
and also will get ugly. However I forgot to change this part :(

> >   				_opp_remove(opp_table, opp, true);
> >   		}
> >   	} else {
> > @@ -1894,6 +1895,44 @@ void dev_pm_opp_of_remove_table(struct device *dev)
> >   unlock:
> >   	mutex_unlock(&opp_table_lock);
> >   }
> > +
> > +/**
> > + * dev_pm_opp_of_remove_table() - Free OPP table entries created from static DT
>
> No, this isn't the OF specific function.

Sorry, copy-paste disease :) esp. for documenting.

--
Regards,
Sudeep