Hi,

Peter Chen <peter.chen@nxp.com> writes:
> Introduce runtime PM and wakeup interrupt handler for cdns3,
> the runtime PM is default off since other cdns3 has not implemented
> glue layer support for runtime PM.

it should be left disabled by default even after every glue implements
it. Enabling runtime PM is a policy decision better left to userspace,
IMO.

> When the controller is in low power mode, the lpm flag will be set.
> The interrupt triggered later than lpm flag is set considers as
> wakeup interrupt and handled at cdns_drd_irq. Once the wakeup

This sentence doesn't parse very well. Do you want to rephrase it?

> occurs, it first disables interrupt to avoid later interrupt
> occurrence since the controller is in low power mode at that
> time, and access registers may be invalid at that time. At wakeup
> handler, it will call pm_runtime_get to wakeup controller, and
> at runtime resume handler, it will enable interrupt again.

Then what? Are you going to, somehow, process pending interrupts? At
what point?

> The API platform_suspend is introduced for glue layer to implement
> platform specific PM sequence.

I really think that's unnecessary. Why can't we rely on parent->child
relationship for this?

-- 
balbi