Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Mon, 30 May 2022 14:37:38 +0530
From:   Viresh Kumar <viresh.kumar@linaro.org>
To:     Pierre Gondois <pierre.gondois@arm.com>
Cc:     linux-kernel@vger.kernel.org, Ionela.Voinescu@arm.com,
        Dietmar.Eggemann@arm.com,
        Shaokun Zhang <zhangshaokun@hisilicon.com>,
        "Rafael J. Wysocki" <rafael@kernel.org>, linux-pm@vger.kernel.org
Subject: Re: [PATCH v1] cpufreq: CPPC: Fix unused-function warning
Message-ID: <20220530090738.7ycfr7g52b54jzj3@vireshk-i7>
References: <20220530081236.40728-1-pierre.gondois@arm.com>
 <20220530082025.vqzk37dvyzxiq7dv@vireshk-i7>
 <3a26ce90-5d2d-0164-3799-85a9dc1abee6@arm.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <3a26ce90-5d2d-0164-3799-85a9dc1abee6@arm.com>
Precedence: bulk

On 30-05-22, 10:44, Pierre Gondois wrote:
> 
> 
> On 5/30/22 10:20, Viresh Kumar wrote:
> > On 30-05-22, 10:12, Pierre Gondois wrote:
> > > Building the cppc_cpufreq driver with for arm64 with
> > > CONFIG_ENERGY_MODEL=n triggers the following warnings:
> > >   drivers/cpufreq/cppc_cpufreq.c:550:12: error: ‘cppc_get_cpu_cost’ defined but not used
> > > [-Werror=unused-function]
> > >     550 | static int cppc_get_cpu_cost(struct device *cpu_dev, unsigned long KHz,
> > >         |            ^~~~~~~~~~~~~~~~~
> > >   drivers/cpufreq/cppc_cpufreq.c:481:12: error: ‘cppc_get_cpu_power’ defined but not used
> > > [-Werror=unused-function]
> > >     481 | static int cppc_get_cpu_power(struct device *cpu_dev,
> > >         |            ^~~~~~~~~~~~~~~~~~
> > > 
> > > Fixes: 740fcdc2c20e ("cpufreq: CPPC: Register EM based on efficiency class information")
> > > Reported-by: Shaokun Zhang <zhangshaokun@hisilicon.com>
> > > Signed-off-by: Pierre Gondois <pierre.gondois@arm.com>
> > > ---
> > >   drivers/cpufreq/cppc_cpufreq.c | 6 +++---
> > >   1 file changed, 3 insertions(+), 3 deletions(-)
> > > 
> > > diff --git a/drivers/cpufreq/cppc_cpufreq.c b/drivers/cpufreq/cppc_cpufreq.c
> > > index d092c9bb4ba3..ecd0d3ee48c5 100644
> > > --- a/drivers/cpufreq/cppc_cpufreq.c
> > > +++ b/drivers/cpufreq/cppc_cpufreq.c
> > > @@ -478,7 +478,7 @@ static inline unsigned long compute_cost(int cpu, int step)
> > >   			step * CPPC_EM_COST_STEP;
> > >   }
> > > -static int cppc_get_cpu_power(struct device *cpu_dev,
> > > +static __maybe_unused int cppc_get_cpu_power(struct device *cpu_dev,
> > >   		unsigned long *power, unsigned long *KHz)
> > >   {
> > >   	unsigned long perf_step, perf_prev, perf, perf_check;
> > > @@ -547,8 +547,8 @@ static int cppc_get_cpu_power(struct device *cpu_dev,
> > >   	return 0;
> > >   }
> > > -static int cppc_get_cpu_cost(struct device *cpu_dev, unsigned long KHz,
> > > -		unsigned long *cost)
> > > +static __maybe_unused int cppc_get_cpu_cost(struct device *cpu_dev,
> > > +		unsigned long KHz, unsigned long *cost)
> > >   {
> > >   	unsigned long perf_step, perf_prev;
> > >   	struct cppc_perf_caps *perf_caps;
> > 
> > Should we actually run cppc_cpufreq_register_em() for
> > !CONFIG_ENERGY_MODEL ? Why?
> > 
> 
> Hello Viresh,
> It seems that when CONFIG_ENERGY_MODEL=n, the compiler is already
> considering cppc_cpufreq_register_em() as an empty function.
> 
> Indeed, CONFIG_ENERGY_MODEL=n makes em_dev_register_perf_domain()
> an empty function, so cppc_cpufreq_register_em() is only made of
> variable definitions. This compiler optimization also explains
> why cppc_get_cpu_power() and cppc_get_cpu_cost() trigger the
> -Wunused-function warning.
> 
> Putting cppc_cpufreq_register_em() inside an
> #ifdef CONFIG_ENERGY_MODEL
> guard seems also valid to me. To avoid too many empty definitions
> of cppc_cpufreq_register_em(), I guess it should be inside an
> #if defined(CONFIG_ARM64) && defined(CONFIG_ENERGY_MODEL)
> guard instead.
> Please let me know what you prefer.

In that case we shouldn't do:

cppc_cpufreq_driver.register_em = cppc_cpufreq_register_em;

as well, as that is extra work for the cpufreq core, which won't be
used at all.

So instead of __maybe_unused, lets put all dependent stuff within
CONFIG_ENERGY_MODEL ?

-- 
viresh