[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <942d6dc7-26ad-405a-bb6a-270e2261a329@kernel.org>
Date: Fri, 6 Feb 2026 17:49:08 +0100
From: Krzysztof Kozlowski <krzk@...nel.org>
To: Victor.Duicu@...rochip.com, linux@...ck-us.net
Cc: corbet@....net, linux-hwmon@...r.kernel.org, devicetree@...r.kernel.org,
robh@...nel.org, linux-kernel@...r.kernel.org, krzk+dt@...nel.org,
linux-doc@...r.kernel.org, Marius.Cristea@...rochip.com, conor+dt@...nel.org
Subject: Re: [PATCH v3 1/2] dt-bindings: hwmon: add support for MCP998X
On 06/02/2026 15:17, Victor.Duicu@...rochip.com wrote:
>>> +
>>> + interrupts:
>>> + items:
>>> + - description: Signal coming from ALERT/THERM pin.
>>> + - description: Signal coming from THERM/ADDR pin.
>>> + - description: Signal coming from SYS_SHDN pin.
>>> +
>>> + interrupt-names:
>>> + items:
>>> + - const: alert-therm
>>> + - const: therm-addr
>>> + - const: sys-shutdown
>>
>> The top-level definition of interrupt-names specifies exactly 3
>> items.
>> How does this interact with variants that only have 2 interrupts?
>>
>
> The chips with "D" in the family have the sys-shutdown and alert-therm
> interrupt pins. The rest have alert-therm and therm-addr interrupt
> pins. The conditional assigns the interrupt names depending on the
> chip.
No, the top level says you have three interrupts. Do not create bindings
which contradict themselves.
More important I am 100% sure this fails tests if you wrote proper, so a
complete example. It passes only because you made a limited example,
without properties.
No, drop review, fix and request re-review.
Best regards,
Krzysztof
Powered by blists - more mailing lists