Re: [PATCH 04/11] arm64: tegra: Enable XUSB host and device on Jetson AGX Orin

From: Thierry Reding <thierry.reding@gmail.com>
To: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Cc: Jon Hunter <jonathanh@nvidia.com>,
	Wayne Chang <waynec@nvidia.com>,
	gregkh@linuxfoundation.org, robh+dt@kernel.org,
	krzysztof.kozlowski+dt@linaro.org, treding@nvidia.com,
	heikki.krogerus@linux.intel.com, ajayg@nvidia.com, kishon@ti.com,
	vkoul@kernel.org, p.zabel@pengutronix.de, balbi@kernel.org,
	mathias.nyman@intel.com, jckuo@nvidia.com,
	linux-usb@vger.kernel.org, devicetree@vger.kernel.org,
	linux-kernel@vger.kernel.org, singhanc@nvidia.com,
	linux-i2c@vger.kernel.org, linux-phy@lists.infradead.org,
	linux-tegra@vger.kernel.org
Subject: Re: [PATCH 04/11] arm64: tegra: Enable XUSB host and device on Jetson AGX Orin
Date: Fri, 28 Oct 2022 14:38:43 +0200	[thread overview]
Message-ID: <Y1vNU1KeUH8LvG0r@orome> (raw)
In-Reply-To: <5676bcd2-14fc-4e1d-643e-89e575d190c3@linaro.org>

[-- Attachment #1: Type: text/plain, Size: 1427 bytes --]

On Fri, Oct 28, 2022 at 07:27:09AM -0400, Krzysztof Kozlowski wrote:
> On 28/10/2022 05:33, Jon Hunter wrote:
> >>> +			ucsi_ccg: ucsi_ccg@8 {
> >>
> >> No underscores in node names.
> >>
> >>> +				compatible = "cypress,cypd4226";
> >>> +				cypress,firmware-build = "gn";
> >>> +				interrupt-parent = <&gpio>;
> >>> +				interrupts = <TEGRA234_MAIN_GPIO(Y, 4) IRQ_TYPE_LEVEL_LOW>;
> >>> +				reg = <0x08>;
> >>> +				status = "okay";
> >>
> >> The pattern of redefining full path in Tegra is confusing - I have no
> >> clue which of these status=okay are correct which are redundant.
> >>
> >> Do you?
> > 
> > I understand you may not like this approach, however, this comment is 
> > not really relevant to just this patch, but a general comment. But yes 
> > we will ensure that this is correct.
> > 
> 
> Just to clarify - this status looks redundant, but I have no way to tell
> for sure...

But that's independent of whether we specify this using the full path or
reference the node by label, isn't it? The only way to make sure that a
status = "okay" is not redundant is by manual inspection. I don't know
of an automated way to do that. Perhaps it's something that could be
added as a check to DTC?

In this particular case I don't think the status is needed. As Jon
mentioned, this device is first defined here and status = "okay" is the
default, so this is redundant.

Thierry

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]