Re: [PATCH v4 2/2] gpio: mxc: Always set GPIOs used as interrupt source to INPUT mode

[Marek Vasut <marex@denx.de>]

On 7/26/22 10:15, Linus Walleij wrote:
> On Mon, Jul 25, 2022 at 12:50 AM Marek Vasut <marex@denx.de> wrote:
> 
>> Always configure GPIO pins which are used as interrupt source as INPUTs.
>> In case the default pin configuration is OUTPUT, or the prior stage does
>> configure the pins as OUTPUT, then Linux will not reconfigure the pin as
>> INPUT and no interrupts are received.
>>
>> Always configure interrupt source GPIO pin as input to fix the above case.
>>
>> Fixes: 07bd1a6cc7cbb ("MXC arch: Add gpio support for the whole platform")
>> Signed-off-by: Marek Vasut <marex@denx.de>
>> Cc: Bartosz Golaszewski <bgolaszewski@baylibre.com>
>> Cc: Linus Walleij <linus.walleij@linaro.org>
>> Cc: Loic Poulain <loic.poulain@linaro.org>
>> Cc: Marc Zyngier <maz@kernel.org>
>> Cc: NXP Linux Team <linux-imx@nxp.com>
>> Cc: Peng Fan <peng.fan@nxp.com>
>> Cc: Shawn Guo <shawnguo@kernel.org>
>> ---
>> V2: Actually update and clear bits in GDIR register
>> V3: Rebase on top of new patch 1/2, expand CC list, add Fixes tag
>> V4: No change
> 
> I understand what you are trying to achieve, and it makes sense.
> 
> There's is just this one generic GPIO-based driver that makes me
> a little bit nervous here.
> 
> Consider:
> drivers/media/cec/platform/cec-gpio/cec-gpio.c
> Look what the driver is doing with the gpiod_* operations on it's
> cec->cec_gpio.
> 
> A certain GPIO pin is switched back and forth between input and
> output and in input mode, it is used to generate interrupts as well.
> 
> Will this still work fine with the MXC driver after this change?
> At least it will be set to input mode twice, but I suppose that is
> fine, it's not your fault that the frameworks are orthogonal.

Ugh. I don't see why it shouldn't work, esp. if the CEC driver controls 
the direction and the default is input, but I wonder what other corner 
cases there are.