[PATCH 2/2] Make non-linear GPIO ranges accesible from gpiolib

[Stephen Warren]

On 06/25/2013 08:32 AM, Linus Walleij wrote:
> On Fri, Jun 21, 2013 at 11:17 PM, Stephen Warren <swarren at wwwdotorg.org> wrote:
>> On 06/20/2013 05:57 AM, Christian Ruppert wrote:
> 
>> The Linux pinctrl subsystem specifically doesn't provide mutual
>> exclusion between "mux function" and GPIO usage within a pin group,
>> although perhaps a driver could internally.
> 
> It used to block this at one point. But it doesn't make sense
> when the hardware looks like so:
> 
>>>                                                      +- SPI
>>>                  Physical pins --- GPIO --- pinctrl -+- I2C
>>>                                                      +- mmc
> 
> As in this case it is perfectly legal to enable the GPIO as
> input while the I2C bus is running and "spy" on the signals.
> 
> The driver should probably not allow the GPIO output to be
> driven while some peripheral is muxed in though, that could be
> disastrous...

Well, in the HW diagram above, GPIO output probably simply overrides
"mux function" output, so everything would work as requested. Whether it
makes sense to request such an override is a policy question that
pinctrl itself probably shouldn't decide. After all, what if there's a
pin group containing 4 pins, which are used by the board as 2 GPIO and 2
I2C. pinctrl shouldn't disallow selecting GPIO on any pin in that pin
group simply because I2C is selected on it; that fact doesn't
necessarily mean that the selected mux function actually uses every
single pin in the group.