Hi Linus,

Le 07/01/2013 00:46, Linus Walleij a ?crit :

On Wed, Dec 19, 2012 at 9:18 PM, Maxime Ripard
[off-list ref] wrote:

quoted

The Allwinner SoCs have an IP module that handle both the muxing and the
GPIOs.

Sorry for very slow review :-( :-(

That's fine, at that time of the year, I wasn't expecting a fast review
anyway :)

quoted

 include/linux/pinctrl/pinconf-generic.h            |    1 +

Can you break this into a separate patch and rebase it?

It does not apply anymore after v3.8...

Ah, yes, of course... Will do.

quoted

diff --git a/include/linux/pinctrl/pinconf-generic.h b/include/linux/pinctrl/pinconf-generic.h
index 4f0abb9..5f5968d 100644
--- a/include/linux/pinctrl/pinconf-generic.h
+++ b/include/linux/pinctrl/pinconf-generic.h

@@ -74,6 +74,7 @@ enum pin_config_param {
        PIN_CONFIG_DRIVE_PUSH_PULL,
        PIN_CONFIG_DRIVE_OPEN_DRAIN,
        PIN_CONFIG_DRIVE_OPEN_SOURCE,
+       PIN_CONFIG_DRIVE_CURRENT,
        PIN_CONFIG_INPUT_SCHMITT,
        PIN_CONFIG_INPUT_DEBOUNCE,
        PIN_CONFIG_POWER_SOURCE,

Above the definitions there is some kerneldoc and that is where this has to be
defined.

You also have to defines what the argument to this parameter is.

I think it should be renamed PIN_CONFIG_DRIVE_STRENGTH and the
argument should be the number of drivers stages.

These things are constructed with totem-pole-like outputs and the number
of totempoles define the drive strength, usually it will be something like

1 = 2mA
2 = 4mA
3 = 6mA
4 = 8mA

or similar, as you see 2 mA for each added driver stage. The driver can
convert to any internal representation...

Ok, I will do it.

Do you have other comments on the driver that I should wait for before
sending a v3?

Thanks,
Maxime

-- 
Maxime Ripard, Free Electrons
Kernel, drivers, real-time and embedded Linux
development, consulting, training and support.
http://free-electrons.com