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:   Mon, 27 Jun 2022 17:47:29 -0600
From:   Rob Herring <robh@kernel.org>
To:     Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Cc:     Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>,
        Chunfeng Yun <chunfeng.yun@mediatek.com>,
        Kishon Vijay Abraham I <kishon@ti.com>,
        Vinod Koul <vkoul@kernel.org>,
        Krzysztof Kozlowski <krzysztof.kozlowski+dt@linaro.org>,
        Anurag Kumar Vulisha <anurag.kumar.vulisha@xilinx.com>,
        Michal Simek <michal.simek@xilinx.com>,
        Matthias Brugger <matthias.bgg@gmail.com>,
        linux-arm-kernel@lists.infradead.org,
        linux-mediatek@lists.infradead.org, linux-phy@lists.infradead.org,
        devicetree@vger.kernel.org, linux-kernel@vger.kernel.org,
        Sean Anderson <sean.anderson@seco.com>
Subject: Re: [PATCH] dt-bindings: phy: make phy-cells description a text
Message-ID: <20220627234729.GA3169775-robh@kernel.org>
References: <20220619113325.21396-1-krzysztof.kozlowski@linaro.org>
 <Yq8LHN+WGVpXDwiM@pendragon.ideasonboard.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <Yq8LHN+WGVpXDwiM@pendragon.ideasonboard.com>
Precedence: bulk

On Sun, Jun 19, 2022 at 02:40:12PM +0300, Laurent Pinchart wrote:
> Hi Krzysztof,
> 
> Thank you for the patch.
> 
> On Sun, Jun 19, 2022 at 01:33:25PM +0200, Krzysztof Kozlowski wrote:
> > The description field is a string, so using YAML inside phy-cells
> > description is not actually helpful.
> 
> Does it hurt though ? 

Unfortunately, I see this a bit. It's convenient because the schema 
passes all the checks. Doh! And I usually stare at it wondering how it 
passed.

Though I probably did review this, so IDK...

> For xlnx,zynqmp-psgtr.yaml I wrote it that way to
> prepare for a future where it could be described using a YAML schema
> (but such future may never come).

There's 2 parts. There's the resolving the defines and then applying the 
schema to the cells. I actually think the latter would be easier. At 
least from a documenting standpoint, we just need to define our own 
keyword to stick the schema under. With the tools doing dtb based 
validation now, the tools already get the phandle node and get the cell 
size from the DT. It's just another step to extract the 
node's compatible, find it's schema, and get its cell format schema. Any 
volunteers?

Rob