netdev - Re: [PATCH] dt-bindings: net: rfkill-gpio: add reset-gpio property

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]

Date: Tue, 19 Mar 2024 11:27:28 +0000
From: Josua Mayer <josua@...id-run.com>
To: Krzysztof Kozlowski <krzysztof.kozlowski@...aro.org>, "David S. Miller"
	<davem@...emloft.net>, Eric Dumazet <edumazet@...gle.com>, Jakub Kicinski
	<kuba@...nel.org>, Paolo Abeni <pabeni@...hat.com>, Rob Herring
	<robh+dt@...nel.org>, Krzysztof Kozlowski
	<krzysztof.kozlowski+dt@...aro.org>, Conor Dooley <conor+dt@...nel.org>,
	Johannes Berg <johannes@...solutions.net>, Philipp Zabel
	<p.zabel@...gutronix.de>
CC: "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
	"devicetree@...r.kernel.org" <devicetree@...r.kernel.org>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] dt-bindings: net: rfkill-gpio: add reset-gpio property

Am 19.03.24 um 11:25 schrieb Krzysztof Kozlowski:
> On 19/03/2024 11:13, Josua Mayer wrote:
>> rfkill-gpio driver supports management of two gpios: reset, shutdown.
>> Reset seems to have been missed when bindings were added.
> No, you do not add properties just because driver supports them.
> Bindings are for hardware, not for drivers.
>
> You now revert all the comments from v1, without actually addressing them.
"Reset of rfkill? It seems entire binding is a workaround of missing
reset in your device. I don't think this is suitable for binding."
> NAK.
Ack. I provide rationale below, but ultimately agree with your NAK.
> Or provide rationale why my previous comments, from adding the
> binding, should be reversed/ignored/overruled.
As mentioned by Philipp, reset signal might be used to radio-silence
a device in case more appropriate signals are not available or wired.

M.2 connector has two interesting signals that fit surprisingly well
to the rfkill-gpio driver's signal names:

FULL_CARD_POWER_OFF# and RESET#.

If for some reason both signals are under software control,
and a specific radio card is known to be connected,
then it would seem correct to describe both reset and shutdown gpios.

However your remark
("It seems entire binding is a workaround of missing reset in your device")
still stands:

When we have a standard m.2 connector, users might connect any
compatible card, and we know cards are often composite devices
utilizing multiple buses.

After thinking this through I agree that combining reset- and shutdown-gpio
in rfkill is a workaround to a different problem:
We are unable to describe control signals shared by multiple components
from different buses, i.e. usb, i2c, pci-e, sata.

Perhaps an m.2 connector should be its own entity, with various
gpios and buses linked to it.
Such a structural change is currently beyond my capabilities.

sincerely
Josua Mayer