Re: [PATCH] pinctrl: sh-pfc: r8a7795: Add DRIF support

[Geert Uytterhoeven <geert@xxxxxxxxxxxxxx>]

Hi Ramesh,

CC linux-gpio, LinusW, Laurent

On Fri, Jun 17, 2016 at 4:16 PM, Ramesh Shanmugasundaram
<ramesh.shanmugasundaram@xxxxxxxxxxxxxx> wrote:
>> >  drivers/pinctrl/sh-pfc/pfc-r8a7795.c | 121
>> > +++++++++++++++++++++++++++++++++++
>> >  1 file changed, 121 insertions(+)
>> >
>> > diff --git a/drivers/pinctrl/sh-pfc/pfc-r8a7795.c
>> > b/drivers/pinctrl/sh-pfc/pfc-r8a7795.c
>> > index 33be5d56..6f246ec 100644
>> > --- a/drivers/pinctrl/sh-pfc/pfc-r8a7795.c
>> > +++ b/drivers/pinctrl/sh-pfc/pfc-r8a7795.c
>> > @@ -1658,6 +1658,91 @@ static const unsigned int canfd1_data_mux[] = {
>> >         CANFD1_TX_MARK,         CANFD1_RX_MARK,
>> >  };
>> >
>> > +/* - DRIF
>> > +--------------------------------------------------------------- */
>> static const unsigned int drif0_data_a_pins[] = {
>> > +       /* CLK, SYNC, D0, D1 */
>> > +       RCAR_GP_PIN(6, 8), RCAR_GP_PIN(6, 9), RCAR_GP_PIN(6, 10),
>> > +       RCAR_GP_PIN(6, 7),
>> > +};
>> > +static const unsigned int drif0_data_a_mux[] = {
>> > +       RIF0_CLK_A_MARK, RIF0_SYNC_A_MARK, RIF0_D0_A_MARK,
>> > +RIF0_D1_A_MARK, };
>>
>> According to my information, each DRIF module consists of two sub-modules
>> (that's why there are 8 module clocks for 4 DRIF modules), each handling a
>> data pin, but sharing the CLK and SYNC signals.
>> Hence it's possible to receive using only one data pin. Is that correct?
>
> Yes, that is correct.
>
>> Shouldn't the pinctrl data reflect that? The second unused data pin could
>> be used for something else.
>
> Is that possible? For e.g. when MOD_SEL0(bit8 -> sel_drif2) is set to 0, all the 4 pins are owned by DRIF IP(RIF2_XXX_A set) even though one of the RIF2_D0_A or RIF2_D1_A may be unused depending on the master it interfaces with. Am I missing something?

You're right, the whole block is switched.

The unused data pin can still be configured as a GPIO, as the GPSR register
(to select between GPIO and IP function) takes precedence.

Linus/Laurent: I assume the GPIO also takes precedence in the Linux pinctrl
subsystem, or would it return an error, as the pin is claimed by the IP group?

Gr{oetje,eeting}s,

                        Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@xxxxxxxxxxxxxx

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds
--
To unsubscribe from this list: send the line "unsubscribe linux-gpio" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html