Re: [PATCH v11 02/13] dt-bindings: usb: Add bindings for multiport properties on DWC3 controller

From: Johan Hovold <johan@kernel.org>
To: Krishna Kurapati PSSNV <quic_kriskura@quicinc.com>
Cc: Thinh Nguyen <Thinh.Nguyen@synopsys.com>,
	Krzysztof Kozlowski <krzysztof.kozlowski+dt@linaro.org>,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	Philipp Zabel <p.zabel@pengutronix.de>,
	Andy Gross <agross@kernel.org>,
	Bjorn Andersson <andersson@kernel.org>,
	Konrad Dybcio <konrad.dybcio@linaro.org>,
	Rob Herring <robh+dt@kernel.org>, Felipe Balbi <balbi@kernel.org>,
	Wesley Cheng <quic_wcheng@quicinc.com>,
	Mathias Nyman <mathias.nyman@intel.com>,
	linux-usb@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-arm-msm@vger.kernel.org, devicetree@vger.kernel.org,
	quic_pkondeti@quicinc.com, quic_ppratap@quicinc.com,
	quic_jackp@quicinc.com, ahalaney@redhat.com,
	quic_shazhuss@quicinc.com,
	Bjorn Andersson <quic_bjorande@quicinc.com>,
	Rob Herring <robh@kernel.org>
Subject: Re: [PATCH v11 02/13] dt-bindings: usb: Add bindings for multiport properties on DWC3 controller
Date: Sat, 11 Nov 2023 11:55:17 +0100	[thread overview]
Message-ID: <ZU9dlfM8tdfdqKkx@hovoldconsulting.com> (raw)
In-Reply-To: <c8b0af46-6513-43fc-994d-4d04f6c65b5e@quicinc.com>

On Sat, Nov 11, 2023 at 03:17:40PM +0530, Krishna Kurapati PSSNV wrote:
> On 11/11/2023 2:00 PM, Krishna Kurapati PSSNV wrote:
> > On 11/10/2023 6:58 PM, Johan Hovold wrote:

> >>>     phy-names:
> >>>       minItems: 1
> >>> -    maxItems: 2
> >>> -    items:
> >>> -      enum:
> >>> -        - usb2-phy
> >>> -        - usb3-phy
> >>> +    maxItems: 8
> >>> +    oneOf:
> >>> +      - items:
> >>> +          enum: [ usb2-phy, usb3-phy ]
> >>> +      - items:
> >>> +          pattern: "^usb[23]-port[0-3]$"
> >>
> >> Shouldn't this just be
> >>
> >>     pattern: "^usb[23]-[0-3]$"
> >>
> >> so that it matches the names that are used by the nvidia bindings?
> >>
> >> We already have some inconsistency in that Amlogic uses a variant based
> >> on the legacy names that needlessly includes "phy" in the names:
> >>
> >>     const: usb2-phy0
> >>     const: usb2-phy1
> >>     const: usb3-phy0
> >>     ...
> >>
> >> I don't think we should be introducing a third naming scheme here so I
> >> suggest just following the nvidia bindings.

> >> > In that case, why don't we use  "^usb[23]-phy[0-3]$". I think its close
> > to what we have on dwc3 core already today (usb2-phy/usb3-phy).
>
> I mean, it isn't needless. It is a phy and shouldn't the binding suggest 
> that and include "-phy" in the name ?

No, adding a '-phy' suffix to each name is unnecessary since the
property is called 'phy-names'.

This is also documented:

	For names used in {clock,dma,interrupt,reset}-names, do not add
	any suffix, e.g.: "tx" instead of "txirq" (for interrupt).

	https://docs.kernel.org/devicetree/bindings/writing-bindings.html

and we've already discussed this when I asked you to drop the likewise
redundant '_irq' suffix from the interrupt names.

Johan