Re: [Patch v2 1/6] sched/fair: Determine active load balance for SMT sched groups

[Tim Chen]

On Mon, 2023-06-12 at 13:13 +0200, Peter Zijlstra wrote:
> 
> 
> 
> > @@ -9537,6 +9581,18 @@ static bool update_sd_pick_busiest(struct lb_env *env,
> >  		break;
> >  
> >  	case group_has_spare:
> > +		/*
> > +		 * Do not pick sg with SMT CPUs over sg with pure CPUs,
> > +		 * as we do not want to pull task off half empty SMT core
> > +		 * and make the core idle.
> > +		 */
> > +		if (smt_vs_nonsmt_groups(sds->busiest, sg)) {
> > +			if (sg->flags & SD_SHARE_CPUCAPACITY)
> > +				return false;
> > +			else
> > +				return true;
> > +		}
> 
> However, here I'm not at all sure. Consider SMT-4 with 2 active CPUs, we
> still very much would like to pull one task off if we have an idle core
> somewhere, no?
> 

How about making this modification to take care of SMT-4 case?

Tim

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 33246dce10db..e2261c24e536 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -9642,11 +9642,11 @@ static bool update_sd_pick_busiest(struct lb_env *env,
        case group_has_spare:
                /*
                 * Do not pick sg with SMT CPUs over sg with pure CPUs,
-                * as we do not want to pull task off half empty SMT core
+                * as we do not want to pull task off SMT core with one task
                 * and make the core idle.
                 */
                if (smt_vs_nonsmt_groups(sds->busiest, sg)) {
-                       if (sg->flags & SD_SHARE_CPUCAPACITY)
+                       if (sg->flags & SD_SHARE_CPUCAPACITY && sgs->sum_h_nr_running <= 1)
                                return false;
                        else
                                return true;