Re: [PATCH v7 1/2] dt-bindings: cpufreq: Introduce QCOM CPUFREQ Firmware bindings

From: Sudeep Holla <sudeep.holla@arm.com>
To: Stephen Boyd <sboyd@kernel.org>
Cc: Taniya Das <tdas@codeaurora.org>,
	skannan@codeaurora.org, "Rafael J. Wysocki" <rjw@rjwysocki.net>,
	Viresh Kumar <viresh.kumar@linaro.org>,
	linux-kernel@vger.kernel.org, linux-pm@vger.kernel.org,
	Rajendra Nayak <rnayak@codeaurora.org>,
	Amit Nischal <anischal@codeaurora.org>,
	devicetree@vger.kernel.org, robh@kernel.org,
	amit.kucheria@linaro.org, evgreen@google.com,
	Sudeep Holla <sudeep.holla@arm.com>
Subject: Re: [PATCH v7 1/2] dt-bindings: cpufreq: Introduce QCOM CPUFREQ Firmware bindings
Date: Wed, 8 Aug 2018 09:37:38 +0100	[thread overview]
Message-ID: <20180808083738.GA25416@e107155-lin> (raw)
In-Reply-To: <153370890232.220756.13782706852981763402@swboyd.mtv.corp.google.com>

On Tue, Aug 07, 2018 at 11:15:02PM -0700, Stephen Boyd wrote:
> Quoting Taniya Das (2018-08-07 19:46:01)
[...]

> > Sudeep,
> > 
> > Earlier the design was the freq_domain would take the CPU phandles
> > 
> > freq_domain:
> >    cpus = <&cpu0 &cpu1....>;
> > 
> 
> I believe Sudeep is recommending something I recommended earlier. It
> would look like:
> 
>    cpu7 {
> 	   qcom,freq-domain = <&cpufreq_hw 1>;
>    }
> 
> to indicate that cpu7 is in cpufreq_hw's frequency domain #1. That
> should probably be called clk domain BTW.
>

Thanks Stephen, that's exactly what I meant. You have explained all the
details saving me time :)

> If that was done with a phandle and a single cell, then we should have
> something similar on the cpufreq_hw node side indicating how to parse
> the cells in qcom,freq-domain. A property like #qcom,freq-domain-cells =
> <1> to indicate that one u32 follows the phandle.
>

As Saravana mentions in the other email I can't believe it's just made up
of logical CPU numbers. If that's the case is it software configurable.

--
Regards,
Sudeep