On Wed, 2014-12-03 at 09:38 -0800, S?ren Brinkmann wrote:
> On Wed, 2014-12-03 at 03:03PM +0200, Ivan T. Ivanov wrote:
> > On Thu, 2014-11-27 at 17:26 -0800, Soren Brinkmann wrote:
> > > Instead of the driver caring about implementation details like device
> > > tree, just provide information about driver specific pinconf parameters
> > > to pinconf-generic which takes care of parsing the DT.
> > >
> > > Signed-off-by: Soren Brinkmann brinkmann at xilinx.com>
> > > ---
> > > This is compile tested only. So, it's likely that it needs more tweaking
> > > to make it actually work on HW. But it illustrates the potential
> > > benefits of the pinconf-generic changes in this series.
> > > ---
> > > drivers/pinctrl/qcom/pinctrl-spmi-gpio.c | 125 +++----------------------------
> > > 1 file changed, 11 insertions(+), 114 deletions(-)
> > >
> > > diff --git a/drivers/pinctrl/qcom/pinctrl-spmi-gpio.c b/drivers/pinctrl/qcom/pinctrl-spmi-
> > > gpio.c
> > > index b863b5080890..2db85e53ef73 100644
> > > --- a/drivers/pinctrl/qcom/pinctrl-spmi-gpio.c
> > > +++ b/drivers/pinctrl/qcom/pinctrl-spmi-gpio.c
> > > @@ -131,14 +131,14 @@ struct pmic_gpio_state {
> > > struct gpio_chip chip;
> > > };
> > >
> > > -struct pmic_gpio_bindings {
> > > - const char*property;
> > > - unsigned param;
> > > +static const struct pinconf_generic_dt_params pmic_gpio_bindings[] = {
> > > + {"qcom,pull-up-strength",PMIC_GPIO_CONF_PULL_UP,0},
> > > + {"qcom,drive-strength",PMIC_GPIO_CONF_STRENGTH,0},
> > > };
> > >
> > > -static struct pmic_gpio_bindings pmic_gpio_bindings[] = {
> > > - {"qcom,pull-up-strength",PMIC_GPIO_CONF_PULL_UP},
> > > - {"qcom,drive-strength",PMIC_GPIO_CONF_STRENGTH},
> > > +static const struct pin_config_item pmic_conf_items[] = {
> > > + PCONFDUMP(PMIC_GPIO_CONF_PULL_UP, "pull-up-strength", NULL, true),
> >
> > s/pull-up-strength/qcom,pull-up-strength/
>
> This is used for the debugfs output. I don't think you want the qcom
> prefix here, do you?
Then at lest lets make it "pull up strength".
>
> > > + PCONFDUMP(PMIC_GPIO_CONF_STRENGTH, "drive-strength", NULL, true),
> > > };
> > >
> >
> >
> >
> > I am not happy that we have to define the same think two times.
>
> It's not really the same thing. One is the actual DT binding the other
> how the property is displayed in debugfs. Two different things. It's how
> the core does it. That's not new.
This didn't make me more happy :-). Could both be merged?
Regards,
Ivan
