Re: [PATCH v4] powernv, cpufreq: cpufreq driver for powernv platform

[Vaidyanathan Srinivasan <svaidy@linux.vnet.ibm.com>]

* Gautham R Shenoy <ego@linux.vnet.ibm.com> [2014-03-27 15:00:50]:

[snip]
 
> > > +       u32 len_ids, len_freqs;
> > > +
> > > +       power_mgt = of_find_node_by_path("/ibm,opal/power-mgt");
> > > +       if (!power_mgt) {
> > > +               pr_warn("power-mgt node not found\n");
> > > +               return -ENODEV;
> > > +       }
> > > +
> > > +       if (of_property_read_u32(power_mgt, "ibm,pstate-min", &pstate_min)) {
> > > +               pr_warn("ibm,pstate-min node not found\n");
> > > +               return -ENODEV;
> > > +       }
> > > +
> > > +       if (of_property_read_u32(power_mgt, "ibm,pstate-max", &pstate_max)) {
> > > +               pr_warn("ibm,pstate-max node not found\n");
> > > +               return -ENODEV;
> > > +       }
> > 
> > Why do you need to get these from DT? And not find that yourself here instead?
> >
> 
> Well, I think we can obtain a more accurate value from the DT which
> would have already computed it. But I'll let vaidy answer this.

DT provides the values that we should use.  Ideally these are the
upper and lower limits of the PState table,  however as per the
platform spec, we should use the PState IDs between these limits.  We
would like to parse it and keep it in the driver for debug purposes.

The platform exports this info in the DT to be more explicit and not
have a strong dependency on the PState numbering or PState table
itself.  In the driver we parse it for debug/validation purpose.

--Vaidy