[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <Z7Rf9GPdO2atP89Z@bogus>
Date: Tue, 18 Feb 2025 10:24:52 +0000
From: Sudeep Holla <sudeep.holla@....com>
To: Peng Fan <peng.fan@....nxp.com>
Cc: Cristian Marussi <cristian.marussi@....com>,
Sudeep Holla <sudeep.holla@....com>,
Saravana Kannan <saravanak@...gle.com>,
Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
Linus Walleij <linus.walleij@...aro.org>,
Dong Aisheng <aisheng.dong@....com>,
Fabio Estevam <festevam@...il.com>, Shawn Guo <shawnguo@...nel.org>,
Jacky Bai <ping.bai@....com>,
Pengutronix Kernel Team <kernel@...gutronix.de>,
Sascha Hauer <s.hauer@...gutronix.de>, <arm-scmi@...r.kernel.org>,
<linux-arm-kernel@...ts.infradead.org>,
<linux-kernel@...r.kernel.org>, <linux-gpio@...r.kernel.org>,
<imx@...ts.linux.dev>, Peng Fan <peng.fan@....com>
Subject: Re: [PATCH 1/4] firmware: arm_scmi: bus: Bypass setting fwnode for
scmi cpufreq
On Tue, Feb 18, 2025 at 09:09:49AM +0800, Peng Fan wrote:
> A potential solution is not using reg in the protocol nodes. Define nodes
> as below:
> devperf {
> compatible ="arm,scmi-devperf";
> }
>
> cpuperf {
> compatible ="arm,scmi-cpuperf";
> }
>
> pinctrl {
> compatible ="arm,scmi-pinctrl";
> }
>
> The reg is coded in driver.
>
> But the upper requires restruction of scmi framework.
>
> Put the above away, could we first purse a simple way first to address
> the current bug in kernel? Just as I prototyped here:
> https://github.com/MrVan/linux/tree/b4/scmi-fwdevlink-v2
>
Good luck getting these bindings merged. I don't like it as it is pushing
software policy or issues into to the devicetree. What we have as SCMI
binding is more than required for a firmware interface IMO. So, you are
on your own to get these bindings approved as I am not on board with
these but if you convince DT maintainers, I will have a look at it then
to see if we can make that work really.
--
Regards,
Sudeep
Powered by blists - more mailing lists