linux-kernel - Re: [PATCH v1 1/1] media: i2c: rdacm2x: Make use of device properties

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

Message-ID: <Z-rBQ8tsDHW9clYh@smile.fi.intel.com>
Date: Mon, 31 Mar 2025 19:22:27 +0300
From: Andy Shevchenko <andriy.shevchenko@...ux.intel.com>
To: Laurent Pinchart <laurent.pinchart@...asonboard.com>
Cc: Kieran Bingham <kieran.bingham+renesas@...asonboard.com>,
	linux-kernel@...r.kernel.org, linux-media@...r.kernel.org,
	acopo Mondi <jacopo+renesas@...ndi.org>,
	Mauro Carvalho Chehab <mchehab@...nel.org>
Subject: Re: [PATCH v1 1/1] media: i2c: rdacm2x: Make use of device properties

On Mon, Mar 31, 2025 at 06:34:35PM +0300, Laurent Pinchart wrote:
> On Mon, Mar 31, 2025 at 03:23:21PM +0300, Andy Shevchenko wrote:
> > On Mon, Mar 31, 2025 at 03:07:48PM +0300, Laurent Pinchart wrote:
> > > On Mon, Mar 31, 2025 at 09:16:36AM +0100, Kieran Bingham wrote:
> > > > Quoting Andy Shevchenko (2025-03-31 08:34:35)
> > > > > Convert the module to be property provider agnostic and allow
> > > > > it to be used on non-OF platforms.
> > > > 
> > > > Looks reasonable to me.
> > > 
> > > Is that going to work out of the box though ? The calls below read the
> > > "reg" property to get the device I2C addresses. AFAIK, ACPI handles I2C
> > > addresses using ACPI-specific methods.
> > > 
> > > Andy, have you tested this patch on an ACPI system ?
> > 
> > Only compile-tested. But you are right, this is something different here
> > between OF and ACPI.
> > 
> > I can rephrase the commit message to just point out that fwnode.h shouldn't
> > be in the drivers and either converting to device property in an assumption
> > that later it can be easier to support non-OF cases, or using of.h.
> 
> I wasn't aware that fwnode.h shouldn't be used in drivers, could you
> explain that ?

The fwnode.h provides the data types and definitions that are meant
to be used by the fwnode / device property API providers. The leaf drivers
shouldn't have any business with those definitions. Everything the drivers
need should be provided via property.h. property.h guarantees the necessary
data types to be visible to the users, when required (mostly think of
struct fwnode_reference_args). Yes, I am aware of v4l2-fwnode.h and it seems
it falls into the category of special device property API provider.

> If this patch is part of an effort to eliminate usage of some APIs from
> all drivers, I'm fine with it. Otherwise, I'm not sure it's worth
> modifying the driver.

These drivers basically include the wrong header.
If you insist, I can patch fwnode.h to add a comment summarizing the above.

-- 
With Best Regards,
Andy Shevchenko