Re: [PATCH] sched/topology: Remove EM_MAX_COMPLEXITY limit

[Pierre Gondois]

Hi Ionela,

On 8/17/22 16:21, Ionela Voinescu wrote:
> Hi Pierre,
>
> On Friday 12 Aug 2022 at 12:16:19 (+0200), Pierre Gondois wrote:
> > From: Pierre Gondois <Pierre.Gondois@xxxxxxx>
> >
> > The Energy Aware Scheduler (EAS) estimates the energy consumption
> > of placing a task on different CPUs. The goal is to minimize this
> > energy consumption. Estimating the energy of different task placements
> > is increasingly complex with the size of the platform. To avoid having
> > a slow wake-up path, EAS is only enabled if this complexity is low
> > enough.
> >
> > The current complexity limit was set in:
> > commit b68a4c0dba3b1 ("sched/topology: Disable EAS on inappropriate
> > platforms").
> > base on the first implementation of EAS, which was re-computing
> > the power of the whole platform for each task placement scenario, cf:
> > commit 390031e4c309 ("sched/fair: Introduce an energy estimation helper
> > function").
> > but the complexity of EAS was reduced in:
> > commit eb92692b2544d ("sched/fair: Speed-up energy-aware wake-ups")
> > and find_energy_efficient_cpu() (feec) algorithm was updated in:
> > commit 3e8c6c9aac42 ("sched/fair: Remove task_util from effective
> > utilization in feec()")
> >
> > find_energy_efficient_cpu() (feec) is now doing:
> > feec()
> > \_ for_each_pd(pd) [0]
> >    // get max_spare_cap_cpu and compute_prev_delta
> >    \_ for_each_cpu(pd) [1]
> >
> >    \_ get_pd_busy_time(pd) [2]
> >      \_ for_each_cpu(pd)
> >
> >    // evaluate pd energy without the task
> >    \_ get_pd_max_util(pd, -1) [3.0]
> >      \_ for_each_cpu(pd)
> >    \_ compute_energy(pd, -1)
> >      \_ for_each_ps(pd)
> >
> >    // evaluate pd energy with the task on prev_cpu
> >    \_ get_pd_max_util(pd, prev_cpu) [3.1]
> >      \_ for_each_cpu(pd)
> >    \_ compute_energy(pd, prev_cpu)
> >      \_ for_each_ps(pd)
> >
> >    // evaluate pd energy with the task on max_spare_cap_cpu
> >    \_ get_pd_max_util(pd, max_spare_cap_cpu) [3.2]
> >      \_ for_each_cpu(pd)
> >    \_ compute_energy(pd, max_spare_cap_cpu)
> >      \_ for_each_ps(pd)
> >
> > [3.1] happens only once since prev_cpu is unique. To have an upper
> > bound of the complexity, [3.1] is taken into account for all pds.
> > So with the same definitions for nr_pd, nr_cpus and nr_ps,
> > the complexity is of:
> > nr_pd * (2 * [nr_cpus in pd] + 3 * ([nr_cpus in pd] + [nr_ps in pd]))
> >   [0]  * (     [1] + [2]      +       [3.0] + [3.1] + [3.2]          )
> > = 5 * nr_cpus + 3 * nr_ps
>
> I just want to draw your attention to [1] and the fact that the
> structure of the function changed. Your calculations largely remain the
> same - 3 calls to compute_energy() which in turn now calls
> eenv_pd_max_util() with operations for each cpu, plus some scattered
> calls to eenv_pd_busy_time(), all for each pd.

Yes indeed, there is:
s/get_pd_max_util/eenv_pd_max_util

and also as you spotted, the following pattern:
\_ eenv_pd_max_util(pd, dst_cpu)
  \_ for_each_cpu(pd)
\_ compute_energy(pd, dst_cpu)
  \_ for_each_ps(pd)

should actually be:
\_ compute_energy(pd, dst_cpu)
  \_ eenv_pd_max_util(pd, dst_cpu)
    \_ for_each_cpu(pd)
  \_ for_each_ps(pd)

Thanks,
Pierre

> [1]
> https://lore.kernel.org/lkml/20220621090414.433602-7-vdonnefort@xxxxxxxxxx/
>
> Thanks,
> Ionela.