Re: [PATCH 3/3 v2] sched/fair: reduce cases for active balance

From: Vincent Guittot
Date: Thu Jan 07 2021 - 07:21:25 EST


On Thu, 7 Jan 2021 at 12:26, Valentin Schneider
<valentin.schneider@xxxxxxx> wrote:
>
> On 07/01/21 11:33, Vincent Guittot wrote:
> > Active balance is triggered for a number of voluntary cases like misfit
> > or pinned tasks cases but also after that a number of load balance
> > attempts failed to migrate a task. There is no need to use active load
> > balance when the group is overloaded because an overloaded state means
> > that there is at least one waiting task. Nevertheless, the waiting task
> > is not selected and detached until the threshold becomes higher than its
> > load. This threshold increases with the number of failed lb (see the
> > condition if ((load >> env->sd->nr_balance_failed) > env->imbalance) in
> > detach_tasks()) and the waiting task will end up to be selected after a
> > number of attempts.
> >
> > Signed-off-by: Vincent Guittot <vincent.guittot@xxxxxxxxxx>
> > ---
> > kernel/sched/fair.c | 45 +++++++++++++++++++++++----------------------
> > 1 file changed, 23 insertions(+), 22 deletions(-)
> >
> > diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> > index a3515dea1afc..00ec5b901188 100644
> > --- a/kernel/sched/fair.c
> > +++ b/kernel/sched/fair.c
> > @@ -9499,13 +9499,32 @@ asym_active_balance(struct lb_env *env)
> > }
> >
> > static inline bool
> > -voluntary_active_balance(struct lb_env *env)
> > +imbalanced_active_balance(struct lb_env *env)
> > +{
> > + struct sched_domain *sd = env->sd;
> > +
> > + /*
> > + * The imbalanced case includes the case of pinned tasks preventing a fair
> > + * distribution of the load on the system but also the even distribution of the
> > + * threads on a system with spare capacity
> > + */
>
> Do you mean s/imbalanced/migrate_task/? This part here will affect
> group_imbalanced, group_asym_packing, and some others.

I really mean the imbalanced case which refers to the function name
and includes:
- the pinned tasks case aka group_imbalanced and which is the primary
target of this function ( which explains its name)
- but also the case where we want to evenly spread tasks on system
with spare capacity and removed this imbalance

>
> > + if ((env->migration_type == migrate_task) &&
> > + (sd->nr_balance_failed > sd->cache_nice_tries+2))
> > + return 1;
> > +
> > + return 0;
> > +}
> > +