Re: [PATCH v6 linux 1/2] sched/cache: Reduce the overhead of task_cache_work by only scan the visisted cpus

[Chen, Yu C]

On 7/17/2026 12:12 PM, Luo Gengkun wrote:

>   struct sched_cache_stat {
> @@ -2398,6 +2399,7 @@ struct sched_cache_stat {
>   	unsigned long next_scan;
>   	unsigned long footprint;
>   	int cpu;
> +	cpumask_var_t visited_cpus;

need zalloc_cpumask_var()/free_cpumask_var() in this patch.

>   } ____cacheline_aligned_in_smp;
>   
>   #else
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index d78467ec6ee1..ab9010b2ec49 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -1585,6 +1585,7 @@ void mm_init_sched(struct mm_struct *mm,
>   		pcpu_sched->runtime = 0;
>   		/* a slightly stale cpu epoch is acceptible */
>   		pcpu_sched->epoch = rq->cpu_epoch;
> +		pcpu_sched->epoch_last_visit = rq->cpu_epoch;
>   		epoch = rq->cpu_epoch;
>   	}
>   
> @@ -1635,13 +1636,23 @@ static inline void __update_mm_sched(struct rq *rq,
>   	}
>   }
>   
> -static unsigned long fraction_mm_sched(struct rq *rq,
> -				       struct sched_cache_time *pcpu_sched)
> +static unsigned long fraction_mm_sched(int cpu,
> +				       struct mm_struct *mm)
>   {
> +	struct sched_cache_time *pcpu_sched =
> +		per_cpu_ptr(mm->sc_stat.pcpu_sched, cpu);
> +	struct rq *rq = cpu_rq(cpu);
> +
>   	guard(raw_spinlock_irqsave)(&rq->cpu_epoch_lock);
>   
>   	__update_mm_sched(rq, pcpu_sched);
>   
> +	/* Skip the rq that has not been hit for a long time */
> +	if ((rq->cpu_epoch - pcpu_sched->epoch_last_visit) > llc_epoch_affinity_timeout) {
> +		cpumask_clear_cpu(cpu, mm->sc_stat.visited_cpus);

It may be worth adding a comment in account_mm_sched() or task_cache_work()
to clarify the race condition. As Tim noted,  it happens between setting 
and reading
visited_cpus. That said, it's a trade-off between accuracy and efficiency
- locking would fix it but at the cost of extra overhead IMO - and the 
update-to-date
visited_cpus could be read properly in the next invoke of task_cache_work().

thanks,
Chenyu