[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20230705094638.mosz6wk76ztd5xf5@CAB-WSD-L081021>
Date: Wed, 5 Jul 2023 12:46:38 +0300
From: Dmitry Rokosov <ddrokosov@...rdevices.ru>
To: Conor Dooley <conor@...nel.org>
CC: <gregkh@...uxfoundation.org>, <robh+dt@...nel.org>,
<krzysztof.kozlowski+dt@...aro.org>, <conor+dt@...nel.org>,
<neil.armstrong@...aro.org>, <jbrunet@...libre.com>,
<jirislaby@...nel.org>, <khilman@...libre.com>,
<martin.blumenstingl@...glemail.com>, <kelvin.zhang@...ogic.com>,
<xianwei.zhao@...ogic.com>, <kernel@...rdevices.ru>,
<rockosov@...il.com>, <linux-amlogic@...ts.infradead.org>,
<linux-serial@...r.kernel.org>, <devicetree@...r.kernel.org>,
<linux-kernel@...r.kernel.org>,
<linux-arm-kernel@...ts.infradead.org>
Subject: Re: [PATCH v1 5/5] arm64: dts: meson: a1: change uart compatible
string
Conor,
On Tue, Jul 04, 2023 at 08:19:00PM +0300, Dmitry Rokosov wrote:
> On Tue, Jul 04, 2023 at 06:10:46PM +0100, Conor Dooley wrote:
> > On Tue, Jul 04, 2023 at 08:08:43PM +0300, Dmitry Rokosov wrote:
> > > On Tue, Jul 04, 2023 at 06:02:58PM +0100, Conor Dooley wrote:
> > > > On Tue, Jul 04, 2023 at 04:59:36PM +0300, Dmitry Rokosov wrote:
> > > > > In the current implementation, the meson-a1 configuration incorporates a
> > > > > unique compatibility tag "amlogic,meson-a1-uart' within the meson-uart
> > > > > driver due to its usage of the new console device name "ttyS".
> > > > > Consequently, the previous compatibility tag designated for the
> > > > > 'amlogic,meson-gx-uart' configuration has become obsolete and is no
> > > > > longer relevant to the current setup.
> > > >
> > > > I don't really see why you would remove the gx-uart to be honest, and
> > > > not use it as a fallback. Neil's platform though, so his call :)
> > > >
> > >
> > > Because of amlogic,meson-gx-uart has legacy devname, we do not want to
> > > use it in the A1.
> >
> > Which I did read in your commit message, fallback being the operative
> > word here.
>
> Although it is difficult for me to envision a situation where we would
> require this fallback, but gx-uart fallback will function from a kernel
> perspective (without taking into account bootloader setup or userspace
> daemon script). I don't have any objections to stay gx-uart as a
> fallback, will do it in the v2.
Unfortunately, it's not possible based on schema rules. During dtbs check
I've got the such error:
arch/arm64/boot/dts/amlogic/meson-a1-ad401.dtb: serial@...0: compatible: 'oneOf' conditional failed, one must be fixed:
['amlogic,meson-a1-uart', 'amlogic,meson-ao-uart', 'amlogic,meson-gx-uart'] is too long
'amlogic,meson-g12a-uart' was expected
'amlogic,meson-gx-uart' was expected
'amlogic,meson-ao-uart' was expected
From schema: Documentation/devicetree/bindings/serial/amlogic,meson-uart.yaml
Of course, we can change dt bindings schema, but is it really needed? As
I said before, it's difficult to imagine the situation, when gx-uart
fallback will be helpful.
--
Thank you,
Dmitry
Powered by blists - more mailing lists