Received-SPF: pass (google.com: domain of linux-kernel+bounces-17186-linux.lists.archive=gmail.com@vger.kernel.org designates 139.178.88.99 as permitted sender) client-ip=139.178.88.99;
Precedence: bulk
MIME-Version: 1.0
References: <20240104171553.2080674-1-lukasz.luba@arm.com> <20240104171553.2080674-13-lukasz.luba@arm.com>
In-Reply-To: <20240104171553.2080674-13-lukasz.luba@arm.com>
From: "Rafael J. Wysocki" <rafael@kernel.org>
Date: Thu, 4 Jan 2024 20:55:54 +0100
Message-ID: <CAJZ5v0g9nEis2Bcvygn70vAT=iifHisZ_7_T4PcmQSU_=_Ymgg@mail.gmail.com>
Subject: Re: [PATCH v6 12/23] PM: EM: Add helpers to read under RCU lock the
 EM table
To: Lukasz Luba <lukasz.luba@arm.com>
Cc: linux-kernel@vger.kernel.org, linux-pm@vger.kernel.org, rafael@kernel.org, 
	dietmar.eggemann@arm.com, rui.zhang@intel.com, amit.kucheria@verdurent.com, 
	amit.kachhap@gmail.com, daniel.lezcano@linaro.org, viresh.kumar@linaro.org, 
	len.brown@intel.com, pavel@ucw.cz, mhiramat@kernel.org, qyousef@layalina.io, 
	wvw@google.com
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Thu, Jan 4, 2024 at 6:15=E2=80=AFPM Lukasz Luba <lukasz.luba@arm.com> wr=
ote:
>
> To use the runtime modifiable EM table there is a need to use RCU
> read locking properly. Add helper functions for the device drivers and
> frameworks to make sure it's done properly.
>
> Signed-off-by: Lukasz Luba <lukasz.luba@arm.com>
> ---
>  include/linux/energy_model.h | 19 +++++++++++++++++++
>  1 file changed, 19 insertions(+)
>
> diff --git a/include/linux/energy_model.h b/include/linux/energy_model.h
> index f33257ed83fd..cfaf5d8b1aad 100644
> --- a/include/linux/energy_model.h
> +++ b/include/linux/energy_model.h
> @@ -338,6 +338,20 @@ static inline int em_pd_nr_perf_states(struct em_per=
f_domain *pd)
>         return pd->nr_perf_states;
>  }
>
> +static inline struct em_perf_state *em_get_table(struct em_perf_domain *=
pd)
> +{
> +       struct em_perf_table __rcu *table;
> +
> +       rcu_read_lock();
> +       table =3D rcu_dereference(pd->em_table);
> +       return table->state;
> +}
> +
> +static inline void em_put_table(void)
> +{
> +       rcu_read_unlock();
> +}

The lack of symmetry between em_get_table() and em_put_table() is kind
of confusing.

I don't really like these wrappers.

IMO it would be better to use rcu_read_lock()/rcu_read_unlock()
directly everywhere they are needed and there can be a wrapper around
rcu_dereference(pd->em_table), something like

static inline struct em_perf_state *em_perf_state_from_pd(struct
em_perf_domain *pd)
{
        return rcu_dereference(pd->em_table)->state;
}

> +
>  #else
>  struct em_data_callback {};
>  #define EM_ADV_DATA_CB(_active_power_cb, _cost_cb) { }
> @@ -384,6 +398,11 @@ int em_dev_update_perf_domain(struct device *dev,
>  {
>         return -EINVAL;
>  }
> +static inline struct em_perf_state *em_get_table(struct em_perf_domain *=
pd)
> +{
> +       return NULL;
> +}
> +static inline void em_put_table(void) {}
>  #endif
>
>  #endif
> --