[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <2623685.EsipSBtvXi@phil>
Date: Tue, 15 Oct 2019 22:32:30 +0200
From: Heiko Stuebner <heiko@...ech.de>
To: Johan Jonker <jbx6244@...il.com>
Cc: robh+dt@...nel.org, mark.rutland@....com,
devicetree@...r.kernel.org, linux-arm-kernel@...ts.infradead.org,
linux-rockchip@...ts.infradead.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 2/2] include: dt-bindings: rockchip: remove RK_FUNC defines
Am Dienstag, 15. Oktober 2019, 22:26:14 CEST schrieb Johan Jonker:
> Hi Heiko,
>
> What's the plan for RK_FUNC_GPIO ? Change all to '0' or keep it?
RK_FUNC_GPIO I'd like to keep :-) .
Basic rationale is that mapping RK_FUNC_1 -> 1, RK_FUNC_2 -> 2, etc does
not provide any additional value, while telling it explicitly that we're
mapping to the gpio function does.
Heiko
>
> On 10/15/19 10:10 PM, Heiko Stuebner wrote:
> > Hi Johan,
> >
> > Am Dienstag, 15. Oktober 2019, 21:10:00 CEST schrieb Johan Jonker:
> >> The defines RK_FUNC_1, RK_FUNC_2, RK_FUNC_3 and RK_FUNC_4
> >> are no longer used, so remove them to prevent
> >> that someone start using them again.
> >
> > That won't work. Devicetree provides a slightly flexible promise of
> > backwards compatibilty. So a new kernel should still work old devicetrees.
> > (not exactly sure if this means dt-binaries and sources or only binaries)
> >
> > So while I think RK_FUNC_0-n should not be used anymore, we should
> > probably just mark them as "deprecated" in a first step.
> >
> >
> > Heiko
>
>
Powered by blists - more mailing lists