On Tue, May 14, 2024 at 11:34:29AM +0100, Mark Brown wrote: > On Mon, May 13, 2024 at 05:22:54PM +0100, Conor Dooley wrote: > > On Fri, May 10, 2024 at 08:06:25PM +0800, Alina Yu wrote: > > > > + richtek,fixed-microvolt = <1200000>; > > > regulator-min-microvolt = <1200000>; > > > regulator-max-microvolt = <1200000>; > > > I'm dumb and this example seemed odd to me. Can you explain to me why > > it is not sufficient to set min-microvolt == max-microvolt to achieve > > the same thing? > > This is for a special mode where the voltage being configured is out of > the range usually supported by the regulator, requiring a hardware > design change to achieve. The separate property is because otherwise we > can't distinguish the case where the mode is in use from the case where > the constraints are nonsense, and we need to handle setting a fixed > voltage on a configurable regulator differently to there being a > hardware fixed voltage on a normally configurable regulator. Cool, I think an improved comment message and description would be helpful then to describe the desired behaviour that you mention here. The commit message in particular isn't great: | Since there is no way to check is ldo is adjustable or not. | As discussing in v2 series, 'richtek,fixed-microvolt' is added for that. | user is supposed to know whether vout of ldo is adjustable. It also doesn't seem like this sort of behaviour would be limited to Richtek either, should this actually be a common property in regulator.yaml w/o the vendor prefix? Cheers, Conor.
Attachment:
signature.asc
Description: PGP signature
