Received-SPF: pass (google.com: domain of linux-kernel+bounces-46581-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:40f1:3f00::1 as permitted sender) client-ip=2604:1380:40f1:3f00::1;
Date: Wed, 31 Jan 2024 08:15:02 -0600
From: Rob Herring <robh@kernel.org>
To: Conor Dooley <conor@kernel.org>
Cc: Johan Jonker <jbx6244@gmail.com>, hjc@rock-chips.com, heiko@sntech.de,
	andy.yan@rock-chips.com, krzysztof.kozlowski+dt@linaro.org,
	conor+dt@kernel.org, airlied@gmail.com, daniel@ffwll.ch,
	maarten.lankhorst@linux.intel.com, tzimmermann@suse.de,
	mripard@kernel.org, markyao0591@gmail.com,
	dri-devel@lists.freedesktop.org, devicetree@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH v1 1/6] dt-bindings: display: rockchip: rockchip,dw-hdmi:
 deprecate port property
Message-ID: <20240131141502.GB966672-robh@kernel.org>
References: <6626d288-ecf1-436b-b0a3-3417308a8633@gmail.com>
 <20240130-denim-capacity-e1617595ac9f@spud>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20240130-denim-capacity-e1617595ac9f@spud>

On Tue, Jan 30, 2024 at 06:18:49PM +0000, Conor Dooley wrote:
> On Tue, Jan 30, 2024 at 03:55:43PM +0100, Johan Jonker wrote:
> > The hdmi-connector nodes are now functional and the new way to model
> > hdmi nodes with, so deprecate the port property and
> 
> This doesn't really explain what makes having hdmi-connector nodes
> replace the usecase for "port".
> 
> > make port@0 and
> > port@1 a requirement.
> 
> Why?

That means the deprecated way will always have warnings which makes 
documenting the deprecated stuff a bit pointless. Technically, new 
required properties are ABI break and something I'm working on making 
the tools check (by comparing 2 versions of schemas). That said, if all 
the upstream dts files are fixed already, then I don't care too much.


> > Also update example.
> 
> "Also do x" is a red flag when it comes to commit messages, as it
> immediately makes me think that this should be more than one commit.
> I'd probably write this as "Update the example to avoid use of the
> deprecated property" or something to avoid bad gut reactions.
> That's not worth resending for though obviously...
> 
> > 
> > Signed-off-by: Johan Jonker <jbx6244@gmail.com>
> > ---
> >  .../display/rockchip/rockchip,dw-hdmi.yaml    | 27 ++++++++++++++++---
> >  1 file changed, 23 insertions(+), 4 deletions(-)
> > 
> > diff --git a/Documentation/devicetree/bindings/display/rockchip/rockchip,dw-hdmi.yaml b/Documentation/devicetree/bindings/display/rockchip/rockchip,dw-hdmi.yaml
> > index 7e59dee15a5f..cd0a42f35f24 100644
> > --- a/Documentation/devicetree/bindings/display/rockchip/rockchip,dw-hdmi.yaml
> > +++ b/Documentation/devicetree/bindings/display/rockchip/rockchip,dw-hdmi.yaml
> > @@ -97,8 +97,11 @@ properties:
> >    ports:
> >      $ref: /schemas/graph.yaml#/properties/ports
> > 
> > -    patternProperties:
> > -      "^port(@0)?$":
> > +    properties:
> > +      port:
> > +        $ref: /schemas/graph.yaml#/properties/port
> > +        deprecated: true
> 
> This change makes the deprecated property's description incomplete,
> since it doesn't cover the endpoints any more. It also doesn't make
> port@0 and port mutually exclusive.

graph.yaml has a check that effectively does that.

Rob