Re: [PATCH 1/2] dt-bindings: rtc: add properties to set battery-related functions

[Hugo Villeneuve]

On Tue, 8 Aug 2023 14:32:26 +0200
Alexandre Belloni <alexandre.belloni@xxxxxxxxxxx> wrote:

> On 08/08/2023 08:25:33-0400, Hugo Villeneuve wrote:
> > On Tue, 8 Aug 2023 12:21:24 +0100
> > Conor Dooley <conor@xxxxxxxxxx> wrote:
> > 
> > > Hey Hugo,
> > > 
> > > On Wed, Aug 02, 2023 at 03:11:52PM -0400, Hugo Villeneuve wrote:
> > > > From: Hugo Villeneuve <hvilleneuve@xxxxxxxxxxxx>
> > > > 
> > > > These properties can be defined in the board's device tree to set the
> > > > default power-on values for battery-related functions.
> > > > 
> > > > Signed-off-by: Hugo Villeneuve <hvilleneuve@xxxxxxxxxxxx>
> > > > ---
> > > >  .../devicetree/bindings/rtc/rtc.yaml          | 19 +++++++++++++++++++
> > > >  1 file changed, 19 insertions(+)
> > > > 
> > > > diff --git a/Documentation/devicetree/bindings/rtc/rtc.yaml b/Documentation/devicetree/bindings/rtc/rtc.yaml
> > > > index efb66df82782..0217d229e3fa 100644
> > > > --- a/Documentation/devicetree/bindings/rtc/rtc.yaml
> > > > +++ b/Documentation/devicetree/bindings/rtc/rtc.yaml
> > > > @@ -26,6 +26,25 @@ properties:
> > > >        0: not chargeable
> > > >        1: chargeable
> > > >  
> > > > +  battery-low-detect:
> > > > +    $ref: /schemas/types.yaml#/definitions/uint32
> > > > +    enum: [0, 1]
> > > > +    description: |
> > > > +      For RTC devices supporting a backup battery/supercap, this flag can be
> > > > +      used to configure the battery low detection reporting function:
> > > > +      0: disabled
> > > > +      1: enabled
> > > > +
> > > > +  battery-switch-over:
> > > > +    $ref: /schemas/types.yaml#/definitions/uint32
> > > > +    enum: [0, 1]
> > > > +    description: |
> > > > +      For RTC devices supporting a backup battery/supercap, this flag can be
> > > > +      used to configure the battery switch over when the main voltage source is
> > > > +      turned off:
> > > > +      0: disabled
> > > > +      1: enabled
> > > 
> > > Why are these implemented as enums? This seems to fall into the category
> > > of using DT to determine software policy - why's it not sufficient to
> > > have boolean properties that indicate hardware support and let the software
> > > decide what to do with them?
> > 
> > Hi Conor,
> > the reason is that I based the new properties on the existing property
> > "aux-voltage-chargeable":
> > 
> > -------------------
> >  aux-voltage-chargeable:
> >     $ref: /schemas/types.yaml#/definitions/uint32
> >     enum: [0, 1]
> >     description: |
> >       Tells whether the battery/supercap of the RTC (if any) is
> >       chargeable or not:
> >       0: not chargeable
> >       1: chargeable
> > -------------------
> > 
> > I agree with you that a boolean would be more appropriate. Should I
> > also submit a (separate) patch to fix the "aux-voltage-chargeable"
> > property to a boolean?
> > 
> 
> No, this is an enum on purpose.
> I will not take battery switch over related properties, this is not
> hardware description but software configuration. There is an ioctl for
> this.

Hi Alexandre,
can you suggest then how we can set default PWRMNG values for the
PCF2131 then?

I looked at Documentation/ABI/testing/rtc-cdev but couldn't find an
ioctl to activate the battery switch over function, nor one to activate
the battery-low detection...

Thank you,
Hugo.


> 
> > Hugo.
> > 
> > 
> > > Thanks,
> > > Conor.
> > > 
> > > > +
> > > >    quartz-load-femtofarads:
> > > >      description:
> > > >        The capacitive load of the quartz(x-tal), expressed in femto
> > > > -- 
> > > > 2.30.2
> > > > 
> 
> -- 
> Alexandre Belloni, co-owner and COO, Bootlin
> Embedded Linux and Kernel engineering
> https://bootlin.com