linux-kernel - Re: [PATCH 1/3] sched/fair: Prepare variables for increased precision of EAS estimated energy

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAJZ5v0iOzp5FKo4NsNE-m+sEXZUvv1TbkAO_9+jSidx9c0iq8A@mail.gmail.com>
Date:   Wed, 30 Jun 2021 19:01:03 +0200
From:   "Rafael J. Wysocki" <rafael@...nel.org>
To:     Lukasz Luba <lukasz.luba@....com>
Cc:     Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        Chris Redpath <Chris.Redpath@....com>,
        Dietmar Eggemann <dietmar.eggemann@....com>,
        Morten Rasmussen <morten.rasmussen@....com>,
        Quentin Perret <qperret@...gle.com>,
        Linux PM <linux-pm@...r.kernel.org>,
        Peter Zijlstra <peterz@...radead.org>,
        "Rafael J. Wysocki" <rjw@...ysocki.net>,
        Viresh Kumar <viresh.kumar@...aro.org>,
        Vincent Guittot <vincent.guittot@...aro.org>,
        Ingo Molnar <mingo@...hat.com>,
        Juri Lelli <juri.lelli@...hat.com>,
        Steven Rostedt <rostedt@...dmis.org>, segall@...gle.com,
        Mel Gorman <mgorman@...e.de>,
        Daniel Bristot de Oliveira <bristot@...hat.com>,
        CCj.Yeh@...iatek.com
Subject: Re: [PATCH 1/3] sched/fair: Prepare variables for increased precision
 of EAS estimated energy

On Fri, Jun 25, 2021 at 5:26 PM Lukasz Luba <lukasz.luba@....com> wrote:
>
> The Energy Aware Scheduler (EAS) tries to find best CPU for a waking up
> task. It probes many possibilities and compares the estimated energy values
> for different scenarios. For calculating those energy values it relies on
> Energy Model (EM) data and em_cpu_energy(). The precision which is used in
> EM data is in milli-Watts (or abstract scale), which sometimes is not
> sufficient. In some cases it might happen that two CPUs from different
> Performance Domains (PDs) get the same calculated value for a given task
> placement, but in more precised scale, they might differ. This rounding
> error has to be addressed. This patch prepares EAS code for better
> precision in the coming EM improvements.
>
> Signed-off-by: Lukasz Luba <lukasz.luba@....com>

If you want me to pick up this series, this patch requires an ACK from
the scheduler maintainers.

> ---
>  kernel/sched/fair.c | 13 +++++++------
>  1 file changed, 7 insertions(+), 6 deletions(-)
>
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index 7b8990fd4896..b517c9e79768 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -6582,7 +6582,7 @@ static unsigned long cpu_util_next(int cpu, struct task_struct *p, int dst_cpu)
>   * to compute what would be the energy if we decided to actually migrate that
>   * task.
>   */
> -static long
> +static u64
>  compute_energy(struct task_struct *p, int dst_cpu, struct perf_domain *pd)
>  {
>         struct cpumask *pd_mask = perf_domain_span(pd);
> @@ -6689,12 +6689,13 @@ compute_energy(struct task_struct *p, int dst_cpu, struct perf_domain *pd)
>   */
>  static int find_energy_efficient_cpu(struct task_struct *p, int prev_cpu)
>  {
> -       unsigned long prev_delta = ULONG_MAX, best_delta = ULONG_MAX;
>         struct root_domain *rd = cpu_rq(smp_processor_id())->rd;
> +       u64 prev_delta = ULLONG_MAX, best_delta = ULLONG_MAX;
>         int cpu, best_energy_cpu = prev_cpu, target = -1;
> -       unsigned long cpu_cap, util, base_energy = 0;
> +       unsigned long cpu_cap, util;
>         struct sched_domain *sd;
>         struct perf_domain *pd;
> +       u64 base_energy = 0;
>
>         rcu_read_lock();
>         pd = rcu_dereference(rd->pd);
> @@ -6718,9 +6719,9 @@ static int find_energy_efficient_cpu(struct task_struct *p, int prev_cpu)
>                 goto unlock;
>
>         for (; pd; pd = pd->next) {
> -               unsigned long cur_delta, spare_cap, max_spare_cap = 0;
> +               unsigned long spare_cap, max_spare_cap = 0;
>                 bool compute_prev_delta = false;
> -               unsigned long base_energy_pd;
> +               u64 base_energy_pd, cur_delta;
>                 int max_spare_cap_cpu = -1;
>
>                 for_each_cpu_and(cpu, perf_domain_span(pd), sched_domain_span(sd)) {
> @@ -6790,7 +6791,7 @@ static int find_energy_efficient_cpu(struct task_struct *p, int prev_cpu)
>          * Pick the best CPU if prev_cpu cannot be used, or if it saves at
>          * least 6% of the energy used by prev_cpu.
>          */
> -       if ((prev_delta == ULONG_MAX) ||
> +       if ((prev_delta == ULLONG_MAX) ||
>             (prev_delta - best_delta) > ((prev_delta + base_energy) >> 4))
>                 target = best_energy_cpu;
>
> --
> 2.17.1
>