On 20/04/17 09:23, Ulf Hansson wrote:
Viresh, Sudeep,
Sorry for jumping in late.
[...]
On the contrary(playing devil's advocate here), we can treat all existing regulators alone as OPP then if you strip the voltages and treat it as abstract number.
But then we are going to have lots of platform specific code which will program the actual hardware, etc. Which is all handled by the regulator framework. Also note that the regulator core selects the common voltage selected by all the children, while we want to select the highest performance point here.
If I understand correctly, Sudeep is not convinced that this is about PM domain regulator(s), right?
No, I am saying that it has to be modeled as regulators or some kind of advanced regulators. I am against modeling it as some new feature and using similar terminology that are quite close to OPP/CPPC in which case it's quite hard not to misunderstand the concepts and eventually use these bindings incorrectly.
To me there is no doubt, these regulators is exactly the definition of PM domain regulators.
+1
That said, long time ago we have decided PM domain regulator shall be modeled as exactly that. From DT point of view, this means the handle to the PM domain regulator belongs in the node of the PM domain controller - and not in each device's node of those belonging to the PM domain.
Isn't that what this discussion really boils down to? Or maybe I am not getting it.
I completely agree with you on all the above points. I am against the performance state terminology. Since the regulators and OPP are already defined in the bindings, all we need to explicitly state(if not already) is that there are hierarchical.