[PATCH 1/1] arm64: dts: imx8mp: Update pin function file according to Rev.D RM

[Aisheng Dong]

> From: Ahmad Fatoum <a.fatoum at pengutronix.de>
> Sent: Monday, November 24, 2025 9:54 PM
> 
> Hi,
> 
> On 11/24/25 2:39 PM, Aisheng Dong wrote:
> >> From: Ahmad Fatoum <a.fatoum at pengutronix.de>
> >> Sent: Monday, November 24, 2025 8:07 PM
> >> Subject: Re: [PATCH 1/1] arm64: dts: imx8mp: Update pin function file
> >> according to Rev.D RM
> >>
> >> Hi,
> >>
> >> On 11/24/25 10:54 AM, Dong Aisheng wrote:
> >>> From: Anson Huang <Anson.Huang at nxp.com>
> >>>
> >>> Update i.MX8MP imx8mp-pinfunc.h file according to reference manual
> >> Rev.D.
> >>>
> >>> Signed-off-by: Anson Huang <Anson.Huang at nxp.com>
> >>> Signed-off-by: Dong Aisheng <aisheng.dong at nxp.com>
> >>
> >> At the very least, we should keep around the old defines.
> >>
> >
> > Could you help elaborate a bit more why need keep the old defines as I
> > saw the previous update patch also didn't keep them?
> 
> Which previous update patch do you refer to?
> 

I mean this patch:

commit bcf7206fe9c35e048e1dc90cf62216b0f5eaf091
Author: Anson Huang <Anson.Huang at nxp.com>
Date:   Fri Aug 14 17:27:19 2020 +0800

    arm64: dts: imx8mp: Update pinfunc header file

    Update some pins' name and adjust pin options to i.MX8MP pinfunc
    header file according to latest reference manual.

    Signed-off-by: Anson Huang <Anson.Huang at nxp.com>
    Signed-off-by: Shawn Guo <shawnguo at kernel.org>

> Generally, If the defines are wrong or misleading, I am all for renaming them.
> 
> In this case, NXP changed their mind and renamed the function in an
> (unreleased)) reference manual.

This is not accurate. The RM with updated names has been released.

> 
> The tradeoff here is between:
> 
> - amount of confusion avoided when we rename USB_OTG to USB
> - amount of overhead introduced to adapt device trees
> 
> I think the benefit of the rename is marginal at best and not worth the
> unnecessary breakage it would impose on countless downstream users with
> out-of-tree board device trees.

I agree the benefit of USB renaming may be arguable.
But how about the remain changes (drop invalid defines and adding new ones)?
Are they still need to be fixed?

Regards
Aisheng