From: Ionela Voinescu <ionela.voinescu@arm.com>
To: Lukasz Luba <lukasz.luba@arm.com>
Cc: linux-kernel@vger.kernel.org, dietmar.eggemann@arm.com,
Pierre.Gondois@arm.com, viresh.kumar@linaro.org,
rafael@kernel.org, daniel.lezcano@linaro.org,
linux-pm@vger.kernel.org, mka@chromium.org, nm@ti.com,
sboyd@kernel.org, linux-arm-kernel@lists.infradead.org,
linux-mediatek@lists.infradead.org, cristian.marussi@arm.com,
sudeep.holla@arm.com, matthias.bgg@gmail.com
Subject: Re: [RESEND][PATCH 1/8] PM: EM: Add .get_cost() callback
Date: Mon, 4 Apr 2022 17:00:47 +0100 [thread overview]
Message-ID: <YksWL2cWQDXPCuKv@arm.com> (raw)
In-Reply-To: <20220321095729.20655-2-lukasz.luba@arm.com>
On Monday 21 Mar 2022 at 09:57:22 (+0000), Lukasz Luba wrote:
> The Energy Model (EM) supports devices which report abstract power scale,
> not only real Watts. The primary goal for EM is to enable the Energy Aware
> Scheduler (EAS) for a given platform. Some of the platforms might not be
> able to deliver proper power values. The only information that they might
> have is the relative efficiency between CPU types.
>
> Thus, it makes sense to remove some restrictions in the EM framework and
> introduce a mechanism which would support those platforms. What is crucial
> for EAS to operate is the 'cost' field in the EM. The 'cost' is calculated
> internally in EM framework based on knowledge from 'power' values.
> The 'cost' values must be strictly increasing. The existing API with its
> 'power' value size restrictions cannot guarantee that the 'cost' will meet
> this requirement.
>
> Since the platform is missing this detailed information, but has only
> efficiency details, introduce a new custom callback in the EM framework.
> The new callback would allow to provide the 'cost' values which reflect
> efficiency of the CPUs. This would allow to provide EAS information which
> has different relation than what would be forced by the EM internal
> formulas calculating 'cost' values. Thanks to this new callback it is
> possible to create a system view for EAS which has no overlapping
> performance states across many Performance Domains.
>
> Signed-off-by: Lukasz Luba <lukasz.luba@arm.com>
> ---
> include/linux/energy_model.h | 24 +++++++++++++++++++++++-
> 1 file changed, 23 insertions(+), 1 deletion(-)
>
> diff --git a/include/linux/energy_model.h b/include/linux/energy_model.h
> index 9f3c400bc52d..0a3a5663177b 100644
> --- a/include/linux/energy_model.h
> +++ b/include/linux/energy_model.h
> @@ -114,9 +114,30 @@ struct em_data_callback {
> */
> int (*active_power)(unsigned long *power, unsigned long *freq,
> struct device *dev);
> +
> + /**
> + * get_cost() - Provide the cost at the given performance state of
> + * a device
> + * @dev : Device for which we do this operation (can be a CPU)
> + * @freq : Frequency at the performance state in kHz
> + * @cost : The cost value for the performance state
> + * (modified)
> + *
> + * In case of CPUs, the cost is the one of a single CPU in the domain.
> + * It is expected to fit in the [0, EM_MAX_POWER] range due to internal
> + * usage in EAS calculation.
> + *
> + * Return 0 on success, or appropriate error value in case of failure.
> + */
> + int (*get_cost)(struct device *dev, unsigned long freq,
> + unsigned long *cost);
> };
> -#define EM_DATA_CB(_active_power_cb) { .active_power = &_active_power_cb }
> #define EM_SET_ACTIVE_POWER_CB(em_cb, cb) ((em_cb).active_power = cb)
> +#define EM_ADV_DATA_CB(_active_power_cb, _cost_cb) \
> + { .active_power = _active_power_cb, \
> + .get_cost = _cost_cb }
> +#define EM_DATA_CB(_active_power_cb) \
> + EM_ADV_DATA_CB(_active_power_cb, NULL)
>
> struct em_perf_domain *em_cpu_get(int cpu);
> struct em_perf_domain *em_pd_get(struct device *dev);
> @@ -264,6 +285,7 @@ static inline int em_pd_nr_perf_states(struct em_perf_domain *pd)
>
> #else
> struct em_data_callback {};
> +#define EM_ADV_DATA_CB(_active_power_cb, _cost_cb) { }
> #define EM_DATA_CB(_active_power_cb) { }
> #define EM_SET_ACTIVE_POWER_CB(em_cb, cb) do { } while (0)
>
Reviewed-by: Ionela Voinescu <ionela.voinescu@arm.com>
Thanks,
Ionela.
> --
> 2.17.1
>
next prev parent reply other threads:[~2022-04-04 16:01 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-03-21 9:57 [RESEND][PATCH 0/8] Introduce support for artificial Energy Model Lukasz Luba
2022-03-21 9:57 ` [RESEND][PATCH 1/8] PM: EM: Add .get_cost() callback Lukasz Luba
2022-04-04 16:00 ` Ionela Voinescu [this message]
2022-03-21 9:57 ` [RESEND][PATCH 2/8] PM: EM: Add artificial EM flag Lukasz Luba
2022-04-04 16:01 ` Ionela Voinescu
2022-03-21 9:57 ` [RESEND][PATCH 3/8] PM: EM: Use the new .get_cost() callback while registering EM Lukasz Luba
2022-04-04 16:01 ` Ionela Voinescu
2022-03-21 9:57 ` [RESEND][PATCH 4/8] PM: EM: Change the order of arguments in the .active_power() callback Lukasz Luba
2022-04-04 16:02 ` Ionela Voinescu
2022-03-21 9:57 ` [RESEND][PATCH 5/8] PM: EM: Remove old debugfs files and print all 'flags' Lukasz Luba
2022-04-04 16:02 ` Ionela Voinescu
2022-04-04 16:31 ` Lukasz Luba
2022-03-21 9:57 ` [RESEND][PATCH 6/8] Documentation: EM: Add artificial EM registration description Lukasz Luba
2022-04-04 16:02 ` Ionela Voinescu
2022-03-21 9:57 ` [RESEND][PATCH 7/8] thermal: cooling: Check Energy Model type in cpufreq_cooling and devfreq_cooling Lukasz Luba
2022-03-28 3:27 ` Viresh Kumar
2022-03-29 6:37 ` Lukasz Luba
2022-04-04 16:02 ` Ionela Voinescu
2022-03-21 9:57 ` [RESEND][PATCH 8/8] powercap: DTPM: Check for Energy Model type Lukasz Luba
2022-04-04 16:03 ` Ionela Voinescu
2022-04-04 12:35 ` [RESEND][PATCH 0/8] Introduce support for artificial Energy Model Lukasz Luba
2022-04-12 6:53 ` Lukasz Luba
2022-04-13 14:29 ` Rafael J. Wysocki
2022-04-13 14:47 ` Lukasz Luba
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=YksWL2cWQDXPCuKv@arm.com \
--to=ionela.voinescu@arm.com \
--cc=Pierre.Gondois@arm.com \
--cc=cristian.marussi@arm.com \
--cc=daniel.lezcano@linaro.org \
--cc=dietmar.eggemann@arm.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mediatek@lists.infradead.org \
--cc=linux-pm@vger.kernel.org \
--cc=lukasz.luba@arm.com \
--cc=matthias.bgg@gmail.com \
--cc=mka@chromium.org \
--cc=nm@ti.com \
--cc=rafael@kernel.org \
--cc=sboyd@kernel.org \
--cc=sudeep.holla@arm.com \
--cc=viresh.kumar@linaro.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).