On Fri, 26 Oct 2018, Pascal PAILLET-LME wrote:

> Hello Lee,
>
> Le 10/26/2018 08:46 AM, Lee Jones a écrit :
> > On Thu, 25 Oct 2018, Rob Herring wrote:
> >
> >> On Thu, Oct 25, 2018 at 4:44 AM Lee Jones <[email protected]> wrote:
> >>> Rob: please grep your name for some feedback.
> > [Moved to the top]
> >
> >>> I'm surprised Rob allowed you to add register bits in a DT property?
> >> Yeah, they are certainly questionable and not something we encourage,
> >> but the alternative would be dozens of properties which only apply for
> >> this one device and just get translated back to register values.
> > I don't see how this device is any different to the 100's of devices
> > which are already supported in DT.
> >
> >>> On Thu, 18 Oct 2018, Pascal PAILLET-LME wrote:
> >>>
> >>>> From: pascal paillet <[email protected]>
> >>>> +Optional parent device properties:
> >>>> +- st,main-control-register:
> >>>> + -bit 1: Power cycling will be performed on turn OFF condition
> >>>> + -bit 2: PWRCTRL is functional
> >>>> + -bit 3: PWRCTRL active high
> >>>> +- st,pads-pull-register:
> >>>> + -bit 1: WAKEUP pull down is not active
> >>>> + -bit 2: PWRCTRL pull up is active
> >>>> + -bit 3: PWRCTRL pull down is active
> >>>> + -bit 4: WAKEUP detector is disabled
> > This should be part of the Pinctrl configuration/driver.
> >
> > And Pinctrl probably already has properties for this?
> I will remove this in the next version.
> >>>> +- st,vin-control-register:
> >>>> + -bit 0: VINLOW monitoring is enabled
> >>>> + -bit [1...3]: VINLOW rising threshold
> >>>> + 000 VINOK_f + 50mV
> >>>> + 001 VINOK_f + 100mV
> >>>> + 010 VINOK_f + 150mV
> >>>> + 011 VINOK_f + 200mV
> >>>> + 100 VINOK_f + 250mV
> >>>> + 101 VINOK_f + 300mV
> >>>> + 110 VINOK_f + 350mV
> >>>> + 111 VINOK_f + 400mV
> >>>> + -bit [4...5]: VINLOW hyst
> >>>> + 00 100mV
> >>>> + 01 200mV
> >>>> + 10 300mV
> >>>> + 11 400mV
> >>>> + -bit 6: SW_OUT detector is disabled
> >>>> + -bit 7: SW_IN detector is enabled.
> > This should be part of the Regulator configuration/driver?
> I will also remove this in the next version. I think this could be
> handled by an hardware monitor future driver.
> >>>> +- st,usb-control-register:
> >>>> + -bit 3: SW_OUT current limit
> >>>> + 0: 600mA
> >>>> + 1: 1.1A
> >>>> + -bit 4: VBUS_OTG discharge is enabled
> >>>> + -bit 5: SW_OUT discharge is enabled
> >>>> + -bit 6: VBUS_OTG detection is enabled
> >>>> + -bit 7: BOOST_OVP is disabled
> > This should be part of the USB configuration/driver?
> >
> I will also remove this in the next version. And I'm going to add active
> discharge ops in the regulator driver.

And unless this device is in any way special, these subsystems should
already have generic bindings to set these hardware configuration
options.

I would say that passing raw register values is not the way to go.

--
Lee Jones [李琼斯]
Linaro Services Technical Lead
Linaro.org │ Open source software for ARM SoCs
Follow Linaro: Facebook | Twitter | Blog

2018-10-26 20:35:48

by Mark Brown

[permalink] [raw]

Subject: Re: [PATCH v4 4/8] regulator: stpmic1: add stpmic1 regulator driver

On Thu, Oct 25, 2018 at 01:23:08PM +0000, Pascal PAILLET-LME wrote:

> I have reworked the code so that we don't touch any more to the init_data.
> the new loop to register the regulators is below:
>
> for (i = 0; i < ARRAY_SIZE(stpmic1_regulator_cfgs); i++) {
> ret = stpmic1_regulator_register(pdev, i, &stpmic1_matches[i],
> regul);
> if (ret < 0)
> return ret;
> regul++;
> }
>
> Each regulator is registered, even is it is not described in the
> device-tree.

Looks good.

Attachments:

(No filename) (573.00 B)
signature.asc (499.00 B)
Download all attachments