Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Thu, 28 Oct 2021 16:12:27 -0700
From:   Bjorn Andersson <bjorn.andersson@linaro.org>
To:     Lukasz Luba <lukasz.luba@arm.com>
Cc:     linux-kernel@vger.kernel.org, linux-pm@vger.kernel.org,
        linux-arm-kernel@lists.infradead.org,
        linux-arm-msm@vger.kernel.org, sudeep.holla@arm.com,
        will@kernel.org, catalin.marinas@arm.com, linux@armlinux.org.uk,
        gregkh@linuxfoundation.org, rafael@kernel.org,
        viresh.kumar@linaro.org, amitk@kernel.org,
        daniel.lezcano@linaro.org, amit.kachhap@gmail.com,
        thara.gopinath@linaro.org, agross@kernel.org
Subject: Re: [PATCH v2 1/5] arch_topology: Introduce thermal pressure update
 function
Message-ID: <YXsuW+wOWoJzgMAA@ripper>
References: <20211015144550.23719-1-lukasz.luba@arm.com>
 <20211015144550.23719-2-lukasz.luba@arm.com>
 <YXmdyeOmNS2x3K0W@ripper>
 <a75b08d9-cf0f-b132-b07e-878b10d133a2@arm.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <a75b08d9-cf0f-b132-b07e-878b10d133a2@arm.com>
Precedence: bulk

On Thu 28 Oct 00:16 PDT 2021, Lukasz Luba wrote:

> 
> 
> On 10/27/21 7:43 PM, Bjorn Andersson wrote:
> > On Fri 15 Oct 07:45 PDT 2021, Lukasz Luba wrote:
> > > diff --git a/arch/arm64/include/asm/topology.h b/arch/arm64/include/asm/topology.h
> > [..]
> > > +/**
> > > + * topology_thermal_pressure_update() - Update thermal pressure for CPUs
> > > + * @cpus	: The related CPUs for which capacity has been reduced
> > > + * @capped_freq	: The maximum allowed frequency that CPUs can run at
> > 
> > I know this matches what I see in e.g. the Qualcomm cpufreq hw driver,
> > but in what cases will @capped_freq differ from
> > cpufreq_get_hw_max_freq(cpumask_first(cpus))?
> 
> The @capped_freq is the maximum allowed frequency value due to
> thermal reasons, which will always be lower or equal to the value
> returned by cpufreq_get_hw_max_freq()
> (effectively: 'policy->cpuinfo.max_freq').
> 

Read patch 3 and 4 again and now this makes sense to me.

Thanks,
Bjorn

> We limit the frequency (and voltage) of CPU to reduce power (and heat)
> in the passive cooling system. That information is important to us,
> because scheduler needs to know how fast the CPU can go. It cannot
> assume that the speed is always 'policy->cpuinfo.max_freq'. Often
> it's less then that at heavy load or GPU heavy load (the same SoC).
> 
> Regards,
> Lukasz