Re: [PATCH v8 3/3] input: pm8xxx-vibrator: add new SPMI vibrator support

[Fenglin Wu]

Hi Konrad,

On 4/11/2024 2:10 AM, Konrad Dybcio wrote:
> > +    if (regs->drv2_mask) {
> > +        if (on)
> > +            val = (vib->level << regs->drv2_shift) & regs->drv2_mask;
> > +        else
> > +            val = 0;
> > +        rc = regmap_write(vib->regmap, vib->drv2_addr, val);
>
> Are you purposefuly zeroing out the other bits?
>
> If yes, consider regmap_write_bits here
> If not, consider regmap_update_bits here
>
> > +        if (rc < 0)
> > +            return rc;
>
> Ignore regmap_r/w errors, these mean a complete failure of the API and
> we don't generally assume MMIO accesses can fail
>
> Unless SPMI is known to have issues here
Sorry, forgot to reply on this comment. Yes, SPMI transaction would fail 
(even with very low odds) on some boards if the layout of SPMI lines is 
not good enough. I'd like to keep the consistence since the whole driver 
also checks the regmap_r/w errors.