Re: [PATCH v4 3/3] arm64: dts: imx93: Add phyBOARD-Segin-i.MX93 support

From: Primoz Fiser <primoz.fiser@norik.com>
To: Wadim Egorov <w.egorov@phytec.de>,
	Stefan Wahren <wahrenst@gmx.net>,
	Mathieu Othacehe <othacehe@gnu.org>
Cc: Rob Herring <robh+dt@kernel.org>,
	Krzysztof Kozlowski <krzysztof.kozlowski+dt@linaro.org>,
	Conor Dooley <conor+dt@kernel.org>,
	Shawn Guo <shawnguo@kernel.org>,
	Sascha Hauer <s.hauer@pengutronix.de>,
	Pengutronix Kernel Team <kernel@pengutronix.de>,
	Fabio Estevam <festevam@gmail.com>,
	NXP Linux Team <linux-imx@nxp.com>, Li Yang <leoyang.li@nxp.com>,
	Christoph Stoidner <c.stoidner@phytec.de>,
	devicetree@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org, upstream@lists.phytec.de
Subject: Re: [PATCH v4 3/3] arm64: dts: imx93: Add phyBOARD-Segin-i.MX93 support
Date: Thu, 25 Jan 2024 07:12:13 +0100	[thread overview]
Message-ID: <26efc165-7319-434a-82b3-b557949f671d@norik.com> (raw)
In-Reply-To: <13f6edd6-f3ae-4e32-88e9-03355d78153c@phytec.de>

Hi,

On 25. 01. 24 07:05, Wadim Egorov wrote:
> 
> Am 24.01.24 um 18:28 schrieb Stefan Wahren:
>> Hello Mathieu,
>>
>> Am 24.01.24 um 14:48 schrieb Mathieu Othacehe:
>>> Hello Stefan,
>>>
>>>>> Defining line names should be fine. But I would still prefer to have
>>>>> the muxing in an overlay bound to a specific use case.
>>>> I'm fine with this. Unfortunately Mathieu dropped the line names in V5
>>>> today :-(
>>>>
>>>> AFAIR reviewers should have 2 weeks time maximum. This was just 2 days.
>>> I am sorry but it is not easy for me to deal with contradictory input. I
>>> chose to remove the gpio-line-names even though it also seemed like a
>>> nice addition to me. The idea was to not interfere with Phytec plans in
>>> the future.
>> tbh sending v5 before the discussion between Wadim and me was finished
>> made it more complicated. Please keep in mind that some reviewers do
>> this in their spare time, so a response could take some time.
>>
>> In this particular case Wadim and me agreed on a solution, so no action
>> from your side was necessary except a little bit patience.
>>
>> The reason why i suggested the gpio-line-names in the first place is
>> that users doesn't need to care about different versions of the DT files
>> (except the downstream one). Changing the line names afterwards leads to
>> confusion.
>>
>> So before we discuss on a v6, just a question: are on the X16 connector
>> just 2 pins muxable as GPIO? This is hard to believe.
> 
> In theory you can use more of the Pins as GPIOs. But at this point I
> should mention that the Segin board became slightly more complicated
> since it started to support more SoMs with different SoCs. We have
> routings for various pins to help with the compatibility. So the naming
> in the schematics is not really trivial. And IMO the dt should follow
> the naming of the schematics.
> 
> I would prefer to go with v5 without having any namings for now.
> 
> Regards,
> Wadim

This would also be my preference.

Thanks,

BR,
Primoz