On 29/04/2019 15:09:18+0000, Howey, Dylan wrote:
> Thanks for the quick response. This is actually the first time I've
> submitted changes to this project.
>
> The 04/27/2019 15:00, Alexandre Belloni wrote:
> > > -static int pcf2123_write_reg(struct device *dev, u8 reg, u8 val)
> > > -{
> > > - u8 txbuf[2];
> > > +static const struct regmap_range pcf2123_ranges[] = {
> > > + {
> > > + .range_min = PCF2123_REG_CTRL1,
> > > + .range_max = PCF2123_REG_CTDWN_TMR,
> > > + },
> > > +};
> > >
> > > - txbuf[0] = reg;
> > > - txbuf[1] = val;
> > > - return pcf2123_write(dev, txbuf, sizeof(txbuf));
> > > -}
> > > +static const struct regmap_access_table pcf2123_access_table = {
> > > + .yes_ranges = pcf2123_ranges,
> > > + .n_yes_ranges = ARRAY_SIZE(pcf2123_ranges),
> > > +};
> >
> > This covers all the registers, I don't think this is necessary.
> >
> This would cover the same registers that are exposed by the sysfs
> interface. I can take out the timer registers since this driver does not
> support the periodic timer.
>
What I meant is that you probably don't need the access table as the
whole range is already between 0 and max_register so there is no need to
double check.
> I agree with the rest of your feedback. When I send a V2 patch should I
> respond to this thread?
>
This should be sent as a separate thred.
--
Alexandre Belloni, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com
