by Soren Brinkmann

[permalink] [raw]

Subject: Re: [PATCH RFC v2 8/8] ARM: zynq: DT: Add pinctrl information

On Fri, 2014-10-31 at 10:40AM -0700, Sören Brinkmann wrote:
> On Fri, 2014-10-31 at 06:36PM +0100, Linus Walleij wrote:
> > On Fri, Oct 31, 2014 at 5:57 PM, Sören Brinkmann
> > <[email protected]> wrote:
> > > On Fri, 2014-10-31 at 09:17AM +0100, Linus Walleij wrote:
> >
> > >> Again it seems to be a sequencing problem. And device tree is
> > >> not good at sequences, therefore all states should be self-contained.
> > >
> > > I agree, but how would I define a pin with pull-up enabled and
> > > tri-state disabled - assume the pin is currently in a random state that
> > > can have those things set/not set arbitrarily.
> >
> > I was more thinking as everything you don't enable explicitly
> > in a state is per definition disabled.
> >
> > So if you are in state A and tri-state is enabled there and you
> > move to state B where pull-up is enabled, then tri-state should
> > be disabled, since it is not explicitly enabled.
> >
> > > I can't put bias-disable in DT since it would potentially disable both
> > > and the pull-up enable would have only a transient effect.
> >
> > Well look at the callback from the core:
> >
> > int (*pin_config_set) (struct pinctrl_dev *pctldev,
> > unsigned pin,
> > unsigned long *configs,
> > unsigned num_configs);
> >
> > You get all configs in an array. The driver can walk over the list and
> > make informed decisions on what to do *BEFORE* poking any registers.
> >
> > Avoiding transients as you describe is part of why the callback
> > looks as it does. This is why every driver has its own for-loop.
>
> Okay, I guess that is possible. I find usage of the arguments more
> elegant since it is more explicit and reduces code in the driver, but I
> suspect it should work.

It does work with some limitation though.
This was how I originally described a state in DT:
pinctrl_uart1_default: pinctrl-uart1-default {
common {
groups = "uart1_10_grp";
function = "uart1";
bias-pull-up = <0>;
slew-rate = <0>;
io-standard = <1>;
};

rx {
pins = "MIO49";
bias-high-impedance = <1>;
};

tx {
pins = "MIO48";
bias-high-impedance = <0>;
};
};

Now, I removed the arguments for tri-state and pull-up. The problem,
this state is handled per-sub-node. I.e. one call to the cfg-set
callback per node. I.e. I cannot split things in common, rx and tx, but
I need to duplicate the pinconf props in rx and tx, resulting in:
pinctrl_uart1_default: pinctrl-uart1-default {
common {
groups = "uart1_10_grp";
function = "uart1";
};

rx {
pins = "MIO49";
slew-rate = <0>;
io-standard = <1>;
bias-high-impedance;
};

tx {
pins = "MIO48";
slew-rate = <0>;
io-standard = <1>;
};
};

In a nutshell, yes, it's possible to work without the arguments for
pull-up or tri-state. But it adds complexity in the driver and the DT
description.

Sören

2014-11-03 14:40:24

by Linus Walleij

[permalink] [raw]

Subject: Re: [PATCH RFC v2 8/8] ARM: zynq: DT: Add pinctrl information

On Sun, Nov 2, 2014 at 9:20 PM, Sören Brinkmann
<[email protected]> wrote:
(...)
> common {
> groups = "uart1_10_grp";
> function = "uart1";
> bias-pull-up = <0>;
> slew-rate = <0>;
> io-standard = <1>;
> };
>
> rx {
> pins = "MIO49";
> bias-high-impedance = <1>;
> };
>
> tx {
> pins = "MIO48";
> bias-high-impedance = <0>;
> };
(...)
> common {
> groups = "uart1_10_grp";
> function = "uart1";
> };
>
> rx {
> pins = "MIO49";
> slew-rate = <0>;
> io-standard = <1>;
> bias-high-impedance;
> };
>
> tx {
> pins = "MIO48";
> slew-rate = <0>;
> io-standard = <1>;
> };
> };
>
> In a nutshell, yes, it's possible to work without the arguments for
> pull-up or tri-state. But it adds complexity in the driver and the DT
> description.

But isn't it obvious that the latter is easier for a human to read
and understand (and not misunderstand) and remember readability
of config is one of the design goals of DT...

Yours,
Linus Walleij