Benoit Parrot <bparrot@xxxxxx> wrote on Tue [2019-Nov-12 07:28:11 -0600]:
> Benoit Parrot <bparrot@xxxxxx> wrote on Wed [2019-Nov-06 14:58:39 -0600]:
> > Sakari Ailus <sakari.ailus@xxxxxx> wrote on Wed [2019-Nov-06 10:57:09 +0200]:
> > > Hi Benoit,
> > >
> > > On Mon, Nov 04, 2019 at 01:31:32PM -0600, Benoit Parrot wrote:
> > >
> > > ...
> > >
> > > > +static struct cal_data dra76x_cal_data = {
> > >
> > > const?
> >
> > Hmm, most likely.
>
> Well, it turns out they cannot be constified because we need to be able to
> update the register offset fir the regmap used for the syscon object.
> So I'll leave them as is.
Oh I guess you meant just the "static struct cal_data" ones?
I tried making the other one into const but obviously that didn't work.
But just constifying "static struct cal_data" appears to compiled just
fine.
>
> Benoit
>
> >
> > >
> > > > + .csi2_phy_core = dra76x_cal_csi_phy,
> > > > + .num_csi2_phy = ARRAY_SIZE(dra76x_cal_csi_phy),
> > > > +
> > >
> > > This newline seems extra.
> > >
> > > > + .flags = 0,
> > >
> > > And flags will be zero in any case, as one more struct members are assigned
> > > to.
> >
> > Is this guaranteed or compiler version dependent?
> >
> > >
> > > > +};
> > >
> > > --
> > > Regards,
> > >
> > > Sakari Ailus
