[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <d899a49e-8934-496b-b390-263aae70122c@kernel.org>
Date: Mon, 27 Nov 2023 10:29:01 +0100
From: Krzysztof Kozlowski <krzk@...nel.org>
To: Conor Dooley <conor.dooley@...rochip.com>
Cc: Sia Jee Heng <jeeheng.sia@...rfivetech.com>,
paul.walmsley@...ive.com, palmer@...belt.com,
aou@...s.berkeley.edu, conor@...nel.org, kernel@...il.dk,
robh+dt@...nel.org, emil.renner.berthing@...onical.com,
linux-riscv@...ts.infradead.org, linux-kernel@...r.kernel.org,
leyfoon.tan@...rfivetech.com
Subject: Re: [PATCH 7/7] riscv: dts: starfive: Add initial StarFive JH8100
device tree
On 27/11/2023 10:20, Conor Dooley wrote:
> On Mon, Nov 27, 2023 at 09:50:35AM +0100, Krzysztof Kozlowski wrote:
>> On 27/11/2023 02:36, Sia Jee Heng wrote:
>
>>> + chosen {
>>> + stdout-path = "serial0:115200n8";
>>> + };
>>> +
>>> + cpus {
>>> + timebase-frequency = <4000000>;
>>
>> This was fixed recently on the mailing lists.
>
> Actually, I only fixed this for the microchip stuff, as I don't know
> whether or not having this in $board.dts is valid for other cases.
>
> As far as I understand it, this works just like regular clock nodes, so
> if the clock is provided by an off-chip crystal it is okay to have it in
> $board.dts, but otherwise it should go in $soc.dts.
>
> The same applies to the jh7100/jh7110, so if you could clarify it for
> that platform too, that'd be great. Doing it in response to
> https://lore.kernel.org/all/20231126-attractor-swampland-b1b95bd1322b@spud/
> would be great.
OK, then it should be added via label syntax, not by full node path.
Best regards,
Krzysztof
Powered by blists - more mailing lists