Re: [PATCH v3 1/1] gpio: mpfs: add polarfire soc gpio support

[Lewis.Hanly]

On 16/07/2022 11:44, Marc Zyngier wrote:
> EXTERNAL EMAIL: Do not click links or open attachments unless you know the content is safe
> 
> On Sat, 16 Jul 2022 08:11:13 +0100,
> <lewis.hanly@xxxxxxxxxxxxx> wrote:
>>
>> From: Lewis Hanly <lewis.hanly@xxxxxxxxxxxxx>
>>
>> Add a driver to support the Polarfire SoC gpio controller.
>>
>> Signed-off-by: Lewis Hanly <lewis.hanly@xxxxxxxxxxxxx>
>> ---
>>   drivers/gpio/Kconfig     |   9 +
>>   drivers/gpio/Makefile    |   1 +
>>   drivers/gpio/gpio-mpfs.c | 361 +++++++++++++++++++++++++++++++++++++++
>>   3 files changed, 371 insertions(+)
>>   create mode 100644 drivers/gpio/gpio-mpfs.c
> 
> A couple of other nits:
> 
>> +static const struct of_device_id mpfs_of_ids[] = {
>> +     { .compatible = "microchip,mpfs-gpio", },
> 
> Where is the DT binding for this?
Already upstream, 
Documentation/devicetree/bindings/gpio/microchip%2Cmpfs-gpio.yaml
> 
>> +     { /* end of list */ }
>> +};
>> +
>> +static struct platform_driver mpfs_gpio_driver = {
>> +     .probe = mpfs_gpio_probe,
>> +     .driver = {
>> +             .name = "microchip,mpfs-gpio",
>> +             .of_match_table = mpfs_of_ids,
>> +     },
>> +     .remove = mpfs_gpio_remove,
> 
> No, please. You cannot enforce that there are no interrupts being used
> (and nothing checks for this), and you're pretty much guaranteed that
> the system will catch fire on the first interrupt being delivered.
> Moreover, your "remove" callback only turns the clock off, which is
> yet another nail on that coffin.
Will remove.
> 
>          M.
> 
> --
> Without deviation from the norm, progress is not possible.