Re: [powerpc][5.13.0-rc7] Kernel warning (kernel/sched/fair.c:401) while running LTP tests

[Vincent Guittot]

Hi Sacha

On Mon, 21 Jun 2021 at 18:22, Vincent Guittot
<vincent.guittot@xxxxxxxxxx> wrote:
>
> Le lundi 21 juin 2021 à 14:42:23 (+0200), Odin Ugedal a écrit :
> > Hi,
> >
> > Did some more research, and it looks like this is what happens:
> >
> > $ tree /sys/fs/cgroup/ltp/ -d --charset=ascii
> > /sys/fs/cgroup/ltp/
> > |-- drain
> > `-- test-6851
> >     `-- level2
> >         |-- level3a
> >         |   |-- worker1
> >         |   `-- worker2
> >         `-- level3b
> >             `-- worker3
> >
> > Timeline (ish):
> > - worker3 gets throttled
> > - level3b is decayed, since it has no more load
> > - level2 get throttled
> > - worker3 get unthrottled
> > - level2 get unthrottled
> >   - worker3 is added to list
> >   - level3b is not added to list, since nr_running==0 and is decayed
> >
> >
> > The attached diff (based on
> > https://lore.kernel.org/lkml/20210518125202.78658-3-odin@xxxxxxx/)
> > fixes the issue for me. Not the most elegant solution, but the
> > simplest one as of now, and to show what is wrong.
> >
> > Any thoughts Vincent?
>
>
> I would prefer that we use the reason of adding the cfs in the list instead.
>
> Something like the below should also fixed the problem. It is based on a
> proposal I made to Rik sometimes ago when he tried to flatten the rq:
> https://lore.kernel.org/lkml/20190906191237.27006-6-riel@xxxxxxxxxxx/
>
> This will ensure that a cfs is added in the list whenever one of its  child
> is still in the list.

Could you confirm that this patch fixes the problem for you too ?

>
> ---
>  kernel/sched/fair.c | 28 ++++++++++++++++++++++++++++
>  1 file changed, 28 insertions(+)
>
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index ea7de54cb022..e751061a9449 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -3272,6 +3272,31 @@ static inline void cfs_rq_util_change(struct cfs_rq *cfs_rq, int flags)
>
>  #ifdef CONFIG_SMP
>  #ifdef CONFIG_FAIR_GROUP_SCHED
> +/*
> + * Because list_add_leaf_cfs_rq always places a child cfs_rq on the list
> + * immediately before a parent cfs_rq, and cfs_rqs are removed from the list
> + * bottom-up, we only have to test whether the cfs_rq before us on the list
> + * is our child.
> + * If cfs_rq is not on the list, test wether a child needs its to be added to
> + * connect a branch to the tree  * (see list_add_leaf_cfs_rq() for details).
> + */
> +static inline bool child_cfs_rq_on_list(struct cfs_rq *cfs_rq)
> +{
> +       struct cfs_rq *prev_cfs_rq;
> +       struct list_head *prev;
> +
> +       if (cfs_rq->on_list) {
> +               prev = cfs_rq->leaf_cfs_rq_list.prev;
> +       } else {
> +               struct rq *rq = rq_of(cfs_rq);
> +
> +               prev = rq->tmp_alone_branch;
> +       }
> +
> +       prev_cfs_rq = container_of(prev, struct cfs_rq, leaf_cfs_rq_list);
> +
> +       return (prev_cfs_rq->tg->parent == cfs_rq->tg);
> +}
>
>  static inline bool cfs_rq_is_decayed(struct cfs_rq *cfs_rq)
>  {
> @@ -3287,6 +3312,9 @@ static inline bool cfs_rq_is_decayed(struct cfs_rq *cfs_rq)
>         if (cfs_rq->avg.runnable_sum)
>                 return false;
>
> +       if (child_cfs_rq_on_list(cfs_rq))
> +               return false;
> +
>         return true;
>  }
>
> --
> 2.17.1
>
>
>
> >
> > Thanks
> > Odin
> >
> >
> > diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> > index bfaa6e1f6067..aa32e9c29efd 100644
> > --- a/kernel/sched/fair.c
> > +++ b/kernel/sched/fair.c
> > @@ -376,7 +376,8 @@ static inline bool list_add_leaf_cfs_rq(struct
> > cfs_rq *cfs_rq)
> >         return false;
> >  }
> >
> > -static inline void list_del_leaf_cfs_rq(struct cfs_rq *cfs_rq)
> > +/* Returns 1 if cfs_rq was present in the list and removed */
> > +static inline bool list_del_leaf_cfs_rq(struct cfs_rq *cfs_rq)
> >  {
> >         if (cfs_rq->on_list) {
> >                 struct rq *rq = rq_of(cfs_rq);
> > @@ -393,7 +394,9 @@ static inline void list_del_leaf_cfs_rq(struct
> > cfs_rq *cfs_rq)
> >
> >                 list_del_rcu(&cfs_rq->leaf_cfs_rq_list);
> >                 cfs_rq->on_list = 0;
> > +               return 1;
> >         }
> > +       return 0;
> >  }
> >
> >  static inline void assert_list_leaf_cfs_rq(struct rq *rq)
> > @@ -3298,24 +3301,6 @@ static inline void cfs_rq_util_change(struct
> > cfs_rq *cfs_rq, int flags)
> >
> >  #ifdef CONFIG_SMP
> >  #ifdef CONFIG_FAIR_GROUP_SCHED
> > -
> > -static inline bool cfs_rq_is_decayed(struct cfs_rq *cfs_rq)
> > -{
> > -       if (cfs_rq->load.weight)
> > -               return false;
> > -
> > -       if (cfs_rq->avg.load_sum)
> > -               return false;
> > -
> > -       if (cfs_rq->avg.util_sum)
> > -               return false;
> > -
> > -       if (cfs_rq->avg.runnable_sum)
> > -               return false;
> > -
> > -       return true;
> > -}
> > -
> >  /**
> >   * update_tg_load_avg - update the tg's load avg
> >   * @cfs_rq: the cfs_rq whose avg changed
> > @@ -4109,11 +4094,6 @@ static inline void update_misfit_status(struct
> > task_struct *p, struct rq *rq)
> >
> >  #else /* CONFIG_SMP */
> >
> > -static inline bool cfs_rq_is_decayed(struct cfs_rq *cfs_rq)
> > -{
> > -       return true;
> > -}
> > -
> >  #define UPDATE_TG      0x0
> >  #define SKIP_AGE_LOAD  0x0
> >  #define DO_ATTACH      0x0
> > @@ -4771,10 +4751,11 @@ static int tg_unthrottle_up(struct task_group
> > *tg, void *data)
> >         if (!cfs_rq->throttle_count) {
> >                 cfs_rq->throttled_clock_task_time += rq_clock_task(rq) -
> >                                              cfs_rq->throttled_clock_task;
> > -
> > -               /* Add cfs_rq with load or one or more already running
> > entities to the list */
> > -               if (!cfs_rq_is_decayed(cfs_rq) || cfs_rq->nr_running)
> > +               if (cfs_rq->insert_on_unthrottle) {
> >                         list_add_leaf_cfs_rq(cfs_rq);
> > +                       if (tg->parent)
> > +
> > tg->parent->cfs_rq[cpu_of(rq)]->insert_on_unthrottle = true;
> > +                       }
> >         }
> >
> >         return 0;
> > @@ -4788,7 +4769,7 @@ static int tg_throttle_down(struct task_group
> > *tg, void *data)
> >         /* group is entering throttled state, stop time */
> >         if (!cfs_rq->throttle_count) {
> >                 cfs_rq->throttled_clock_task = rq_clock_task(rq);
> > -               list_del_leaf_cfs_rq(cfs_rq);
> > +               cfs_rq->insert_on_unthrottle = list_del_leaf_cfs_rq(cfs_rq);
> >         }
> >         cfs_rq->throttle_count++;
> >
> > @@ -8019,6 +8000,23 @@ static bool __update_blocked_others(struct rq
> > *rq, bool *done)
> >
> >  #ifdef CONFIG_FAIR_GROUP_SCHED
> >
> > +static inline bool cfs_rq_is_decayed(struct cfs_rq *cfs_rq)
> > +{
> > +       if (cfs_rq->load.weight)
> > +               return false;
> > +
> > +       if (cfs_rq->avg.load_sum)
> > +               return false;
> > +
> > +       if (cfs_rq->avg.util_sum)
> > +               return false;
> > +
> > +       if (cfs_rq->avg.runnable_sum)
> > +               return false;
> > +
> > +       return true;
> > +}
> > +
> >  static bool __update_blocked_fair(struct rq *rq, bool *done)
> >  {
> >         struct cfs_rq *cfs_rq, *pos;
> > diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
> > index a189bec13729..12a707d99ee6 100644
> > --- a/kernel/sched/sched.h
> > +++ b/kernel/sched/sched.h
> > @@ -602,6 +602,7 @@ struct cfs_rq {
> >         u64                     throttled_clock_task_time;
> >         int                     throttled;
> >         int                     throttle_count;
> > +       int                     insert_on_unthrottle;
> >         struct list_head        throttled_list;
> >  #endif /* CONFIG_CFS_BANDWIDTH */
> >  #endif /* CONFIG_FAIR_GROUP_SCHED */