Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Fri, 4 Aug 2023 22:59:56 +0200
From:   Andi Shyti <andi.shyti@kernel.org>
To:     Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Cc:     Jarkko Nikula <jarkko.nikula@linux.intel.com>,
        Mario Limonciello <mario.limonciello@amd.com>,
        Wolfram Sang <wsa@kernel.org>, linux-i2c@vger.kernel.org,
        linux-kernel@vger.kernel.org,
        Mika Westerberg <mika.westerberg@linux.intel.com>,
        Jan Dabros <jsd@semihalf.com>
Subject: Re: [PATCH v1 4/9] i2c: designware: Propagate firmware node
Message-ID: <20230804205956.uuna4c76ww64z3zm@intel.intel>
References: <20230725143023.86325-1-andriy.shevchenko@linux.intel.com>
 <20230725143023.86325-5-andriy.shevchenko@linux.intel.com>
 <f0e0c31e-ad5e-c3ae-945e-c2af0d68beab@linux.intel.com>
 <ZMgU9Jm4u37D+70Y@smile.fi.intel.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <ZMgU9Jm4u37D+70Y@smile.fi.intel.com>
Precedence: bulk

Hi Andy,

On Mon, Jul 31, 2023 at 11:09:24PM +0300, Andy Shevchenko wrote:
> On Fri, Jul 28, 2023 at 03:25:58PM +0300, Jarkko Nikula wrote:
> > On 7/25/23 17:30, Andy Shevchenko wrote:
> > > Propagate firmware node by using a specific API call, i.e. device_set_node().
> 
> ...
> 
> > > +	device_set_node(&dev->adapter.dev, dev_fwnode(dev->dev));
> > 
> > Would this be better to put in the same place where ACPI_COMPANION_SET() is
> > removed like below? I'd keep this static inline function in the header file
> > as simple as possible. All extra code might invite adding even more.
> 
> We come again to the duplication and prone to deviation code, I wouldn't like
> to go this way. The idea of this call is to unify and avoid mistakes, like
> updating only in ACPI or DT (or any new one if happens in the future) case
> and leaving the second one unconsidered.

it's anyway an inline function becoming a bit too fat. Can't we
make it not inline?

> That said, I would rather drop this patch until i2c core will take this
> once for all (may be never in the reasonable future :-).

Which patch are you referring to that should be taken into i2c
core?

Andi