Re: [Patch v4 01/22] sched/cache: Introduce infrastructure for cache-aware load balancing

[Chen, Yu C]

On 4/15/2026 10:06 AM, Vern Hao wrote:
> Hi Tim,

[ ... ]

> > +
> > +static inline
> > +void account_mm_sched(struct rq *rq, struct task_struct *p, s64 
> > delta_exec)
> > +{
> > +    struct sched_cache_time *pcpu_sched;
> > +    struct mm_struct *mm = p->mm;
> > +    unsigned long epoch;
> > +
> > +    if (!sched_cache_enabled())
> > +        return;
> > +
> > +    if (p->sched_class != &fair_sched_class)
> > +        return;
> > +    /*
> > +     * init_task, kthreads and user thread created
> > +     * by user_mode_thread() don't have mm.
> > +     */
> > +    if (!mm || !mm->sc_stat.pcpu_sched)
> > +        return;
> > +
> > +    pcpu_sched = per_cpu_ptr(p->mm->sc_stat.pcpu_sched, cpu_of(rq));
>
> I do some test,  when i see many pids schedstat,  such as  cat proc/xxx/ 
> schedstat, there will cause a panic，maybe the 'p' is exit or others 
> reason,  sorry, i dont record these msg.  how about fix it like below.
>
> pcpu_sched = per_cpu_ptr(mm->sc_stat.pcpu_sched, cpu_of(rq));

Right, we have also received a report at
https://sashiko.dev/#/patchset/cover.1775065312.git.tim.c.chen%40linux.int.com
indicating that this issue needs to be fixed. We are working on it.
Thanks for the investigation.

thanks,
Chenyu