Re: [PATCH v2] sched/cache: Avoid publishing partial LLC utilization

[Tim Chen]

On Tue, 2026-10-13 at 16:43 +0800, Jemmy Wong wrote:
> update_sg_lb_stats() sums utilization over the CPUs in both
> sched_group_span(group) and env->cpus, but takes capacity from
> group->sgc->capacity for the whole group. record_sg_llc_stats()
> publishes both.
> 
> Rejecting an all-affinity-pinned source runqueue removes that CPU from
> env->cpus and retries group selection. The retry can publish the
> partial utilization with the full capacity.
> 
> can_migrate_llc() then treats a busy preferred LLC as having spare
> capacity. More tasks can aggregate into it, migrations out can be
> blocked, and two busy LLCs may not fall back to generic balancing.
> This increases CPU contention and task latency. Found by inspection.
> 
> Skip publication on affinity retries using LBF_ALL_PINNED, retaining
> the first-pass snapshots and avoiding repeated cpumask checks. On the
> first pass, publish either shared LLC statistic only when env->cpus
> covers the whole group. Load balancing still uses the affinity-filtered
> statistics.
> 
> The same window exists during CPU hotplug. env->cpus drops a CPU at
> set_cpu_active(false), but the group span and sgc->capacity still
> include it until cpuset_cpu_inactive() rebuilds the domains. Keeping
> the previous snapshot avoids the mismatched publish. After the rebuild,
> the next periodic balance records both statistics again.

Thanks.

Reviewed-by: Tim Chen <tim.c.chen@xxxxxxxxxxxxxxx>

Tim

> 
> Suggested-by: K Prateek Nayak <kprateek.nayak@xxxxxxx>
> Suggested-by: Tim Chen <tim.c.chen@xxxxxxxxxxxxxxx>
> Fixes: f025ef275388 ("sched/cache: Record per LLC utilization to guide cache aware scheduling decisions")
> Signed-off-by: Jemmy Wong <jemmywong512@xxxxxxxxx>
> ---
> 
> v1: https://lore.kernel.org/r/20261010113425.34829-1-jemmywong512@xxxxxxxxx/
> 
> Changes since v1:
> - Skip publication on affinity retries using LBF_ALL_PINNED, retaining
>   first-pass snapshots and avoiding repeated cpumask checks, as suggested
>   by Prateek and Tim. Keep cpumask_subset() for partial group coverage
>   during CPU hotplug on the first pass.
> - Update the comment and commit message to explain the retry behavior.
> - Add Fixes and Suggested-by tags.
> 
>  kernel/sched/fair.c | 8 ++++++++
>  1 file changed, 8 insertions(+)
> 
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index 57360f5cdde4..dc01fdc0526d 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -12111,6 +12111,14 @@ static void record_sg_llc_stats(struct lb_env *env,
>  	if (env->sd->child != rcu_dereference_all(per_cpu(sd_llc, env->dst_cpu)))
>  		return;
> 
> +	/*
> +	 * A redo repeats stats published on the first pass. Hotplug can
> +	 * still leave env->cpus partial, so util would not match capacity.
> +	 */
> +	if ((env->flags & LBF_ALL_PINNED) ||
> +	    !cpumask_subset(sched_group_span(group), env->cpus))
> +		return;
> +
>  	/*
>  	 * At this point we know this group spans a LLC domain.
>  	 * Record the statistic of this group in its corresponding
> 
> --
> 2.54.0 (Apple Git-157)