linux-kernel - Re: (subset) [PATCH v1 0/3] spidev: introduce trivial abb sensor device

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 for Android: free password hash cracker in your pocket

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

Message-ID: <20250722112013.0000597e@huawei.com>
Date: Tue, 22 Jul 2025 11:20:13 +0100
From: Jonathan Cameron <Jonathan.Cameron@...wei.com>
To: Guenter Roeck <linux@...ck-us.net>
CC: <hs@...x.de>, Krzysztof Kozlowski <krzk@...nel.org>, Mark Brown
	<broonie@...nel.org>, <linux-spi@...r.kernel.org>,
	<linux-kernel@...r.kernel.org>, Andrei Lalaev <andrey.lalaev@...il.com>,
	Chanh Nguyen <chanh@...amperecomputing.com>, Conor Dooley
	<conor+dt@...nel.org>, Fabio Estevam <festevam@...il.com>, Grant Peltier
	<grantpeltier93@...il.com>, Krzysztof Kozlowski <krzk+dt@...nel.org>, Laurent
 Pinchart <laurent.pinchart@...asonboard.com>, Michal Simek
	<michal.simek@....com>, Naresh Solanki <naresh.solanki@...ements.com>,
	Pengutronix Kernel Team <kernel@...gutronix.de>, Rob Herring
	<robh@...nel.org>, Rodrigo Gobbi <rodrigo.gobbi.7@...il.com>, Sascha Hauer
	<s.hauer@...gutronix.de>, Shawn Guo <shawnguo@...nel.org>,
	<devicetree@...r.kernel.org>, <imx@...ts.linux.dev>,
	<linux-arm-kernel@...ts.infradead.org>
Subject: Re: (subset) [PATCH v1 0/3] spidev: introduce trivial abb sensor
 device

On Mon, 21 Jul 2025 21:58:10 -0700
Guenter Roeck <linux@...ck-us.net> wrote:

> On 7/21/25 21:05, Heiko Schocher wrote:
> > Hello Krzysztof,
> > 
> > On 21.07.25 18:24, Krzysztof Kozlowski wrote:  
> >> On 21/07/2025 17:56, Mark Brown wrote:  
> >>> On Sat, 19 Jul 2025 08:33:51 +0200, Heiko Schocher wrote:  
> >>>> This series introduces the changes needed for trivial spi
> >>>> based sensors from ABB, currently operated from userspace.
> >>>>
> >>>> The last patch adds the spidevices to the DTS files, already
> >>>> in mainline.
> >>>>
> >>>> make dtbs_check showed no errors/warnings for the dts files
> >>>>
> >>>> [...]  
> >>>
> >>> Applied to
> >>>
> >>>     https://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi.git for-next
> >>>
> >>> Thanks!
> >>>
> >>> [1/3] dt-bindings: trivial-devices: Document ABB sensors
> >>>        commit: aad2f87cbcab56b322109d26d7b11842a09df91f
> >>> [2/3] spi: spidev: Add an entry for the ABB spi sensors
> >>>        commit: d60f7cab7c04944a79af16caa43c141e780a59c6
> >>>  
> >>
> >>
> >> That's unexpected, Mark. Patches received two objections/comments and I
> >> don't think discussion was resolved.
> >>
> >> ABB is huge company, probably making hundreds or more of sensors. The
> >> patchset basically claims that all of them work with spidev. It does not
> >> providing any model names or details, so it seems really incomplete to
> >> call them trivial devices.  
> > 
> > I do not know how many different sensors they have, nor if that department can
> > speak for the whole company...
> > 
> > What I have as information is:
> > https://lore.kernel.org/linux-spi/2477dc64-92a0-9dc9-d168-56646d0d796e@denx.de/
> > 
> > and I get no more information about them currently. May I should
> > add some sort of trivial into compatible name? Something like
> > 
> > "abb,spi-trivial-sensor"
> > or
> > "abb,spidev-trivial-sensor"
> > 
> > which makes it clearer, that only ABB trivial sensor, controlled through spidev
> > driver, is connected here?
> >   
> 
> FWIW, I always thought that devicetree is not supposed to contain such generic
> information. Is it even appropriate to list something like this in devicetree
> in the first place ?
> 
> If so, what prevents anyone from submitting hundreds of
> "<company>,spidev-trivial-<device-type>" entries, using the same line of argument ?

Agreed.  These should have separate compatibles based on what any OS etc
might want to bind to them.  Just because their model in Linux is spidev etc
that shouldn't mean a generic ID is appropriate.

Can we at least have some examples to motivate the discussion?

Jonathan

> 
> Guenter
>