linux-kernel - Re: [PATCH] arm64: dts: xilinx: Drop "label" property on dlg,slg7xl45106

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <0400eb44-4200-452e-9d1f-d1851f661cb8@kernel.org>
Date: Wed, 17 Dec 2025 13:44:02 +0100
From: Krzysztof Kozlowski <krzk@...nel.org>
To: Michal Simek <michal.simek@....com>, "Rob Herring (Arm)"
 <robh@...nel.org>, Krzysztof Kozlowski <krzk+dt@...nel.org>,
 Conor Dooley <conor+dt@...nel.org>
Cc: devicetree@...r.kernel.org, linux-arm-kernel@...ts.infradead.org,
 linux-kernel@...r.kernel.org
Subject: Re: [PATCH] arm64: dts: xilinx: Drop "label" property on
 dlg,slg7xl45106

On 17/12/2025 13:42, Krzysztof Kozlowski wrote:
> On 17/12/2025 13:38, Michal Simek wrote:
>>
>>
>> On 12/16/25 18:59, Rob Herring (Arm) wrote:
>>> The "label" property is not documented for the dlg,slg7xl45106. Nor is
>>> it common to use for GPIO controllers. So drop it.
>>>
>>
>> Correct it is not documented but it is at least used in Linux also by
>> drivers/gpio/gpio-mmio.c:810:	err = device_property_read_string(dev, "label", 
>> &label);
>>
>> which is also not documented in DT binding.
>>
>> I don't have a problem with it because I can't see it used anywhere.
>>
>> But there are other devices where label is the part of binding. And IMHO this 
>> should be more generic property.
> 
> No, please read the EXPLICIT comment:
> 
> "+	 * This property *must not* be used in device-tree sources, it's only"
> 
> It was added only for ACPI.

Uh, my bad, the comment is above "gpio-mmio,base", not "label", but I
have impression it was applicable to label as well. Maybe that's not the
case...

> 
> We have indeed the problem with documenting disallowed ABI for cases
> where drivers use fwnode/device_property(). So far I insisted on
> comments, but as we can see people can just ignore the comments and
> still use something explicitly disallowed.
> 
> Rob,
> 
> Would it be okay to add for such cases "label: false" in the binding to
> mark that this is not ABI for DT?
> 
> Best regards,
> Krzysztof


Best regards,
Krzysztof