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:   Sun, 4 Sep 2022 03:02:56 +0200
From:   Andrew Lunn <andrew@lunn.ch>
To:     Pali =?iso-8859-1?Q?Roh=E1r?= <pali@kernel.org>
Cc:     Gregory Clement <gregory.clement@bootlin.com>,
        Sebastian Hesselbarth <sebastian.hesselbarth@gmail.com>,
        Rob Herring <robh+dt@kernel.org>,
        Krzysztof Kozlowski <krzysztof.kozlowski+dt@linaro.org>,
        Marek Behun <marek.behun@nic.cz>,
        linux-arm-kernel@lists.infradead.org, devicetree@vger.kernel.org,
        linux-kernel@vger.kernel.org
Subject: Re: [PATCH] ARM: dts: turris-omnia: Add mcu node
Message-ID: <YxP5QD6v4HmBUVGp@lunn.ch>
References: <20220819131152.6513-1-pali@kernel.org>
 <20220831142809.lcmnv3l4rnulo522@pali>
 <YxE2JqJutZ0ilghH@lunn.ch>
 <20220903152735.2ga3iwrivh6zhibf@pali>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20220903152735.2ga3iwrivh6zhibf@pali>
Precedence: bulk

> > I'm also not sure what the DT people will say about the node name mcu.
> > I don't see any examples of that in the binding documentation. They
> > might request you rename it to gpio-controller, unless it does more
> > than GPIO? And if it does do more than GPIO we are then into mfd
> > territory, and the binding then becomes much more interesting. Then we
> > start the questions, are you defining a ABI now, before there is even
> > a driver for it?
> > 
> >        Andrew
> 
> Yes, there is already driver. See my previous email, I mentioned it and
> also I wrote link for this driver. Moreover now driver is merged in
> upstream u-boot.

I'm not comfortable accepting a DT binding for a driver which does not
exist in Linux. As i said, there are interesting ABI issues here, and
it could be the MFD, GPIO or reset Maintainers don't accept a binding
until a Linux driver exists.

At minimum, you need an Acked-by from the GPIO Maintainer, of the
binding before this DT change is merged via MVEBU.

	Andrew