Re: [PATCH] pinctrl: qcom: sm8250: correct sdc2_clk and ngpios

[Bjorn Andersson <bjorn.andersson@xxxxxxxxxx>]

On Sun 13 Sep 18:52 CDT 2020, Dmitry Baryshkov wrote:

> Hello,
> 
> On Mon, 14 Sep 2020 at 02:44, Bjorn Andersson
> <bjorn.andersson@xxxxxxxxxx> wrote:
> >
> > On Sun 13 Sep 17:50 CDT 2020, Dmitry Baryshkov wrote:
> >
> > > Correct sdc2_clk pin definition (register offset) and ngpios (SM8250 has
> > > 180 GPIO pins).
> > >
> > > Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@xxxxxxxxxx>
> > > Fixes: 4e3ec9e407ad5058003309072b37111f7b8c900a
> > > ---
> > >  drivers/pinctrl/qcom/pinctrl-sm8250.c | 4 ++--
> > >  1 file changed, 2 insertions(+), 2 deletions(-)
> > >
> > > diff --git a/drivers/pinctrl/qcom/pinctrl-sm8250.c b/drivers/pinctrl/qcom/pinctrl-sm8250.c
> > > index a660f1274b66..4f7dae61b089 100644
> > > --- a/drivers/pinctrl/qcom/pinctrl-sm8250.c
> > > +++ b/drivers/pinctrl/qcom/pinctrl-sm8250.c
> > > @@ -1308,7 +1308,7 @@ static const struct msm_pingroup sm8250_groups[] = {
> > >       [178] = PINGROUP(178, WEST, _, _, _, _, _, _, _, _, _),
> > >       [179] = PINGROUP(179, WEST, _, _, _, _, _, _, _, _, _),
> > >       [180] = UFS_RESET(ufs_reset, 0xb8000),
> > > -     [181] = SDC_PINGROUP(sdc2_clk, 0x7000, 14, 6),
> > > +     [181] = SDC_PINGROUP(sdc2_clk, 0xb7000, 14, 6),
> >
> > You're right.
> >
> > >       [182] = SDC_PINGROUP(sdc2_cmd, 0xb7000, 11, 3),
> > >       [183] = SDC_PINGROUP(sdc2_data, 0xb7000, 9, 0),
> > >  };
> > > @@ -1320,7 +1320,7 @@ static const struct msm_pinctrl_soc_data sm8250_pinctrl = {
> > >       .nfunctions = ARRAY_SIZE(sm8250_functions),
> > >       .groups = sm8250_groups,
> > >       .ngroups = ARRAY_SIZE(sm8250_groups),
> > > -     .ngpios = 181,
> > > +     .ngpios = 180,
> >
> > But, while it is correct that SM8250 only has 180 GPIOs, we're exposing
> > UFS_RESET as a gpio as well, so this should therefor be 181. Sorry for
> > the obfuscation here...
> 
> True. Could you please drop the second part of the patch on your own?
> Or would you prefer for me to send v2?
> 

This goes through Linus' tree, so it would be up to him. If you want to
avoid waiting for his response go ahead and post a v2 - and I'll give
you a r-b on it.

Regards,
Bjorn