Hi Patrick!
> On 29.11.2017, at 21:35, Patrick Menschel <menschel.p@xxxxxxxxx> wrote:
>
> Hello Martin,
>
>
> I didn't catch the whole discussion but you may want to check
>
> include/dt-bindings/gpio/gpio.h
>
>> #define GPIO_OPEN_DRAIN (GPIO_SINGLE_ENDED | GPIO_ACTIVE_LOW)
>
> This corresponds to
>
> Documentation/devicetree/bindings/gpio/gpio.txt
I understand, but the question still is: how to
present the information in a valid way.
To use gpio propperly it would require that the driver
implements a “sub-driver” pinctrl with all the extra
(boilerplate) code overhead.
Also this would mean mixing different types of
logical drivers into a single source - I doubt that
would be easy to get accepted...
Here again a summary of all the GPIOs that the mcp2517fd has:
* TXCAN: dedicated GPIO with single function,
individually conigurable as push/pull or open drain
* INT: main interrupt line - configurable as push/pull or
individually conigurable as push/pull or open drain
* GPIO0: general GPIO with in/out option, but 2 special “cases”:
tx-irq and TX-disable
group configurable as push/pull or open drain
* GPIO1: general GPIO with in/out potion, but 1 special “cases”:
rx-irq
group configurable as push/pull or open drain
* CLKO/SOF: clock output at (1/10th, 1/5th, 1/2, 1 of the
core frequency) or start of frame output
possibly group configurable as push/pull or open drain
(not explicitly specified in datasheet)
How would you try to present that HW-configuration in the
device tree instead?
How would it impact the driver design?
Thanks,
Martin
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
