On 24/07/2023 13:44, Alexander Stein wrote:
>>>>> Documentation/devicetree/bindings/gpio/st,stmpe-gpio.yaml | 4 ++++
>>>>> 1 file changed, 4 insertions(+)
>>>>>
>>>>> diff --git a/Documentation/devicetree/bindings/gpio/st,stmpe-gpio.yaml
>>>>> b/Documentation/devicetree/bindings/gpio/st,stmpe-gpio.yaml index
>>>>> 22c0cae73425..4555f1644a4d 100644
>>>>> --- a/Documentation/devicetree/bindings/gpio/st,stmpe-gpio.yaml
>>>>> +++ b/Documentation/devicetree/bindings/gpio/st,stmpe-gpio.yaml
>>>>>
>>>>> @@ -28,6 +28,10 @@ properties:
>>>>> gpio-controller: true
>>>>>
>>>>> + gpio-line-names:
>>>>> + minItems: 1
>>>>> + maxItems: 24
>>>>> +
>>>>
>>>> I am sure there is no variant with one GPIO.
>>>
>>> That's true. But if only one GPIO is actually connected there is no use
>>> enforcing to add empty entries.
>>> AFAIK it is also allowed to provide an incomplete array.
>>
>> Did you test it? Linux prints warning. Warning means "not allowed".
>
> I did and I don't see any warnings.
> I using the following (modified) GPIO subnode on an stmpe811 chip:
> gpio {
> compatible = "st,stmpe-gpio";
> gpio-controller;
> #gpio-cells = <2>;
> interrupt-controller;
> #interrupt-cells = <2>;
> /* GPIO 5-7 used for touch */
> st,norequest-mask = <0xf0>;
> gpio-line-names = "GPIO_ADC_I2C1_1";
> };
>
> I only see the warning "stmpe-gpio stmpe-gpio: DMA mask not set", but that's a
> different matter.
> Only GPIO 0-3 are connected. Original property is
> gpio-line-names = "GPIO_ADC_I2C1_1",
> "GPIO_ADC_I2C1_2",
> "GPIO_ADC_I2C1_3",
> "GPIO_ADC_I2C1_4";
>
> Also gpioinfo shows that the gpiochip has 8 pins while only the first one is
> named.
>
> gpiochip7 - 8 lines:
> line 0: "GPIO_ADC_I2C1_1" unused input active-high
> line 1: unnamed unused input active-high
> line 2: unnamed unused input active-high
> line 3: unnamed unused input active-high
> line 4: unnamed unused input active-high
> line 5: unnamed unused input active-high
> line 6: unnamed unused input active-high
> line 7: unnamed unused input active-high
>
>
> If 4 GPIOs are named the output shows names for lines 1-3 accordingly, but no
> errors/warnings as well.
>
>> The expectation is that the rest are named "NC".
>
> Where does this expectation come from? Is there any pending change I am not
> aware of?
>
Hm, indeed, I mistaken the case for multiple gpio chips in one device
(so having an offset). Bindings are actually explicitly allowing array
for only some GPIOs.
Best regards,
Krzysztof
