From: Vincent Guittot <vincent.guittot@linaro.org> To: Sachin Sant <sachinp@linux.vnet.ibm.com> Cc: open list <linux-kernel@vger.kernel.org>, linuxppc-dev@lists.ozlabs.org, Peter Zijlstra <peterz@infradead.org>, Odin Ugedal <odin@uged.al> Subject: Re: [powerpc][5.13.0-rc7] Kernel warning (kernel/sched/fair.c:401) while running LTP tests Date: Mon, 21 Jun 2021 19:09:22 +0200 [thread overview] Message-ID: <CAKfTPtACzzoGhDFW0bTGgZRPB=3LR6kSwuUOrcKDFTAJ7BhTFQ@mail.gmail.com> (raw) In-Reply-To: <20210621162243.GA29874@vingu-book> Hi Sacha On Mon, 21 Jun 2021 at 18:22, Vincent Guittot <vincent.guittot@linaro.org> 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@uged.al/) > > 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@surriel.com/ > > 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 */
WARNING: multiple messages have this Message-ID (diff)
From: Vincent Guittot <vincent.guittot@linaro.org> To: Sachin Sant <sachinp@linux.vnet.ibm.com> Cc: Peter Zijlstra <peterz@infradead.org>, Odin Ugedal <odin@uged.al>, linuxppc-dev@lists.ozlabs.org, open list <linux-kernel@vger.kernel.org> Subject: Re: [powerpc][5.13.0-rc7] Kernel warning (kernel/sched/fair.c:401) while running LTP tests Date: Mon, 21 Jun 2021 19:09:22 +0200 [thread overview] Message-ID: <CAKfTPtACzzoGhDFW0bTGgZRPB=3LR6kSwuUOrcKDFTAJ7BhTFQ@mail.gmail.com> (raw) In-Reply-To: <20210621162243.GA29874@vingu-book> Hi Sacha On Mon, 21 Jun 2021 at 18:22, Vincent Guittot <vincent.guittot@linaro.org> 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@uged.al/) > > 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@surriel.com/ > > 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 */
next prev parent reply other threads:[~2021-06-21 17:09 UTC|newest] Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top 2021-06-21 6:32 [powerpc][5.13.0-rc7] Kernel warning (kernel/sched/fair.c:401) while running LTP tests Sachin Sant 2021-06-21 6:32 ` Sachin Sant 2021-06-21 9:39 ` Odin Ugedal 2021-06-21 9:39 ` Odin Ugedal 2021-06-21 9:50 ` Vincent Guittot 2021-06-21 9:50 ` Vincent Guittot 2021-06-21 9:54 ` Odin Ugedal 2021-06-21 9:54 ` Odin Ugedal 2021-06-21 10:57 ` Sachin Sant 2021-06-21 10:57 ` Sachin Sant 2021-06-21 11:04 ` Odin Ugedal 2021-06-21 11:04 ` Odin Ugedal 2021-06-21 12:42 ` Odin Ugedal 2021-06-21 12:42 ` Odin Ugedal 2021-06-21 16:22 ` Vincent Guittot 2021-06-21 16:22 ` Vincent Guittot 2021-06-21 16:45 ` Odin Ugedal 2021-06-21 16:45 ` Odin Ugedal 2021-06-21 17:07 ` Vincent Guittot 2021-06-21 17:07 ` Vincent Guittot 2021-06-21 17:09 ` Vincent Guittot [this message] 2021-06-21 17:09 ` Vincent Guittot 2021-06-21 17:31 ` Sachin Sant 2021-06-21 17:31 ` Sachin Sant 2021-06-21 17:44 ` Vincent Guittot 2021-06-21 17:44 ` Vincent Guittot
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to='CAKfTPtACzzoGhDFW0bTGgZRPB=3LR6kSwuUOrcKDFTAJ7BhTFQ@mail.gmail.com' \ --to=vincent.guittot@linaro.org \ --cc=linux-kernel@vger.kernel.org \ --cc=linuxppc-dev@lists.ozlabs.org \ --cc=odin@uged.al \ --cc=peterz@infradead.org \ --cc=sachinp@linux.vnet.ibm.com \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.