linux-kernel - Re: [PATCH] iio: accel: bma220: move set_wdt() out of bma220

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: <aPp2BeNqFJABsAkg@smile.fi.intel.com>
Date: Thu, 23 Oct 2025 21:37:57 +0300
From: Andy Shevchenko <andriy.shevchenko@...ux.intel.com>
To: Jonathan Cameron <jonathan.cameron@...wei.com>
Cc: Petre Rodan <petre.rodan@...dimension.ro>,
	Jonathan Cameron <jic23@...nel.org>,
	David Lechner <dlechner@...libre.com>,
	Nuno Sá <nuno.sa@...log.com>,
	Andy Shevchenko <andy@...nel.org>, linux-iio@...r.kernel.org,
	linux-kernel@...r.kernel.org, kernel test robot <lkp@...el.com>
Subject: Re: [PATCH] iio: accel: bma220: move set_wdt() out of bma220_core

On Thu, Oct 23, 2025 at 06:23:18PM +0100, Jonathan Cameron wrote:
> On Wed, 22 Oct 2025 14:50:18 +0300
> Andy Shevchenko <andriy.shevchenko@...ux.intel.com> wrote:
> > On Tue, Oct 21, 2025 at 01:31:49PM +0300, Petre Rodan wrote:
> > > Move bma220_set_wdt() into bma220_i2c.c instead of using a conditional
> > > based on i2c_verify_client() in bma220_core.c that would make core
> > > always depend on the i2c module.  
> > 
> > Reviewed-by: Andy Shevchenko <andriy.shevchenko@...ux.intel.com>
> > 
> > But Kconfig for this driver is a bit strange. Usually we do other way around,
> > i.e. make user visible selection of the glue drivers, while core is selected if
> > at least one of the leaf driver selected by the user.
> > 
> This comes up from time to time.  There kind of isn't a right answer
> to my mind in the trade off between complexity of configuration 
> and desire for minimum useful set of Kconfig symbols and people wanting
> to build only exactly what they want.  So we've ended up with a mix.
> 
> I don't mind setting a policy on this for new code going forwards, but
> that means we need to decide which approach we prefer and document
> it somewhere.

One immediate thing from top of my mind why current approach worse than
proposed by me.

- when you have an SPI or I²C enabled the glue driver will be silently build.
As a used on the embedded system with smaller resources (flash/ROM) I prefer
the opposite behaviour, i.e. when _I_ control what SPI/I²C component driver is
needed even if the subsystem is present.

But let's give a time to gather other opinions and justifications.

-- 
With Best Regards,
Andy Shevchenko