Re: [PATCH 1/4] regulator: core: If consumers don't call regulator_set_load() assume max

[Mark Brown]

On Tue, Aug 14, 2018 at 01:03:07PM -0700, Doug Anderson wrote:

> Specifically I will note that there are boatloads of drivers out there
> that use the regulator framework but don't have a call to
> regulator_set_load() in them.  Are these drivers all broken?  I don't
> think so.  IMO the regulator_set_load() API is an optional call for
> drivers that they can use to optimize power usage, not a required API.

Very few systems dynamically change modes in the first place, if we were
doing this as a matter of course you could claim the drivers were buggy
but really it's unusual for it to even be a useful thing to do - as it
is it's more an accomodation for a small subset of systems.  This does
mean that those systems have to pay attention to making sure that
everything works well which is unfortunate but it's not unreasonable.

Forcing the mode to the highest available would be a step backwards for
most systems.

Attachment: signature.asc
Description: PGP signature