On Tue, Apr 26, 2022 at 11:04:30PM PDT, Krzysztof Kozlowski wrote:
>On 27/04/2022 03:01, Zev Weiss wrote:
>> These Super I/O chips have an i2c interface that some systems expose
>> to a BMC; the BMC's device tree can now describe that via this
>> binding.
>>
>> Signed-off-by: Zev Weiss <[email protected]>
>
>I already reviewed it so I guess you did not include the tag because of
>significant changes?
>
Yeah, the nuvoton,tsi-channel-mask property is new this round, so I
dropped the previous R-B -- and since it looks like I missed some stuff,
thanks for taking another look (though perhaps some of it could have
been avoided if I'd remembered to run 'make dt_binding_check').
>> ---
>> .../bindings/hwmon/nuvoton,nct6775.yaml | 56 +++++++++++++++++++
>> 1 file changed, 56 insertions(+)
>> create mode 100644 Documentation/devicetree/bindings/hwmon/nuvoton,nct6775.yaml
>>
>> diff --git a/Documentation/devicetree/bindings/hwmon/nuvoton,nct6775.yaml b/Documentation/devicetree/bindings/hwmon/nuvoton,nct6775.yaml
>> new file mode 100644
>> index 000000000000..418477374fdb
>> --- /dev/null
>> +++ b/Documentation/devicetree/bindings/hwmon/nuvoton,nct6775.yaml
>> @@ -0,0 +1,56 @@
>> +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
>> +%YAML 1.2
>> +---
>> +
>> +$id: http://devicetree.org/schemas/hwmon/nuvoton,nct6775.yaml#
>> +$schema: http://devicetree.org/meta-schemas/core.yaml#
>> +
>> +title: Nuvoton NCT6775 and compatible Super I/O chips
>> +
>> +maintainers:
>> + - Zev Weiss <[email protected]>
>> +
>> +properties:
>> + compatible:
>> + enum:
>> + - nuvoton,nct6106
>> + - nuvoton,nct6116
>> + - nuvoton,nct6775
>> + - nuvoton,nct6776
>> + - nuvoton,nct6779
>> + - nuvoton,nct6791
>> + - nuvoton,nct6792
>> + - nuvoton,nct6793
>> + - nuvoton,nct6795
>> + - nuvoton,nct6796
>> + - nuvoton,nct6797
>> + - nuvoton,nct6798
>> +
>> + reg:
>> + maxItems: 1
>> +
>> + nuvoton,tsi-channel-mask:
>> + description:
>> + Bitmask indicating which TSI temperature sensor channels are
>> + active. LSB is TSI0, bit 1 is TSI1, etc.
>
>Need a type/ref.
>
Ack, thanks.
>> + maximum: 0xff
>> + default: 0
>
>Since by default it is disabled, doesn't it make a required property?
>IOW, if you add a node without this mask, will the device operate
>properly and usefully?
>
Yeah, zero active TSI channels is a totally legitimate way for these
devices to operate. TSI is just an optional source of additional
temperature readings that's used on some (AMD) systems; all the basic
Super I/O functionality works fine without it.
Thanks,
Zev
On 27/04/2022 08:44, Zev Weiss wrote:
>
>>> + maximum: 0xff
>>> + default: 0
>>
>> Since by default it is disabled, doesn't it make a required property?
>> IOW, if you add a node without this mask, will the device operate
>> properly and usefully?
>>
>
> Yeah, zero active TSI channels is a totally legitimate way for these
> devices to operate. TSI is just an optional source of additional
> temperature readings that's used on some (AMD) systems; all the basic
> Super I/O functionality works fine without it.
OK, thanks.
Best regards,
Krzysztof
On 27/04/2022 08:44, Zev Weiss wrote:
>>> + reg:
>>> + maxItems: 1
>>> +
>>> + nuvoton,tsi-channel-mask:
>>> + description:
>>> + Bitmask indicating which TSI temperature sensor channels are
>>> + active. LSB is TSI0, bit 1 is TSI1, etc.
>>
>> Need a type/ref.
>>
>
> Ack, thanks.
Did you test the bindings after the changes? Using reviewers time
instead of testing by yourself with an automated tool is quite a waste.
Best regards,
Krzysztof
On Wed, Apr 27, 2022 at 09:37:20AM PDT, Krzysztof Kozlowski wrote:
>On 27/04/2022 08:44, Zev Weiss wrote:
>>>> + reg:
>>>> + maxItems: 1
>>>> +
>>>> + nuvoton,tsi-channel-mask:
>>>> + description:
>>>> + Bitmask indicating which TSI temperature sensor channels are
>>>> + active. LSB is TSI0, bit 1 is TSI1, etc.
>>>
>>> Need a type/ref.
>>>
>>
>> Ack, thanks.
>
>Did you test the bindings after the changes? Using reviewers time
>instead of testing by yourself with an automated tool is quite a waste.
>
Yeah, sorry about that -- with uint32 $ref added it passes dt_binding_check;
I'll re-send with that change.
Zev