RT keeps track of the utilization on a per-rq basis with the structure
avg_rt. This utilization is updated during task_tick_rt(),
put_prev_task_rt() and set_next_task_rt(). However, when the current
running task changes its policy, set_next_task_rt() which would usually
take care of updating the utilization when the rq starts running RT tasks,
will not see a such change, leaving the avg_rt structure outdated. When
that very same task will be dequeued later, put_prev_task_rt() will then
update the utilization, based on a wrong last_update_time, leading a huge
spike in the RT utilization signal.
The signal would eventually recover from this issue after few ms. Even if
no RT tasks are run, avg_rt is also updated in __update_blocked_others().
But as the CPU capacity depends partly on the avg_rt, this issue has
nonetheless a significant impact on the scheduler.
Fix this issue by ensuring a load update when a running task changes
its policy to RT.
Fixes: 371bf427 ("sched/rt: Add rt_rq utilization tracking")
Signed-off-by: Vincent Donnefort <[email protected]>
diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c
index a525447..5ac0f31 100644
--- a/kernel/sched/rt.c
+++ b/kernel/sched/rt.c
@@ -2341,10 +2341,8 @@ void __init init_sched_rt_class(void)
static void switched_to_rt(struct rq *rq, struct task_struct *p)
{
/*
- * If we are already running, then there's nothing
- * that needs to be done. But if we are not running
- * we may need to preempt the current running task.
- * If that current running task is also an RT task
+ * If we are not running we may need to preempt the current
+ * running task. If that current running task is also an RT task
* then see if we can move to another run queue.
*/
if (task_on_rq_queued(p) && rq->curr != p) {
@@ -2355,6 +2353,13 @@ static void switched_to_rt(struct rq *rq, struct task_struct *p)
if (p->prio < rq->curr->prio && cpu_online(cpu_of(rq)))
resched_curr(rq);
}
+
+ /*
+ * If we are running, update the avg_rt tracking, as the running time
+ * will now on be accounted into the latter.
+ */
+ if (task_current(rq, p))
+ update_rt_rq_load_avg(rq_clock_pelt(rq), rq, 0);
}
/*
--
2.7.4
DL keeps track of the utilization on a per-rq basis with the structure
avg_dl. This utilization is updated during task_tick_dl(),
put_prev_task_dl() and set_next_task_dl(). However, when the current
running task changes its policy, set_next_task_dl() which would usually
take care of updating the utilization when the rq starts running DL
tasks, will not see a such change, leaving the avg_dl structure outdated.
When that very same task will be dequeued later, put_prev_task_dl() will
then update the utilization, based on a wrong last_update_time, leading a
huge spike in the DL utilization signal.
The signal would eventually recover from this issue after few ms. Even
if no DL tasks are run, avg_dl is also updated in
__update_blocked_others(). But as the CPU capacity depends partly on the
avg_dl, this issue has nonetheless a significant impact on the scheduler.
Fix this issue by ensuring a load update when a running task changes
its policy to DL.
Fixes: 3727e0e ("sched/dl: Add dl_rq utilization tracking")
Signed-off-by: Vincent Donnefort <[email protected]>
diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c
index 3829c5a..915227a 100644
--- a/kernel/sched/deadline.c
+++ b/kernel/sched/deadline.c
@@ -2497,6 +2497,8 @@ static void switched_to_dl(struct rq *rq, struct task_struct *p)
check_preempt_curr_dl(rq, p, 0);
else
resched_curr(rq);
+ } else {
+ update_dl_rq_load_avg(rq_clock_pelt(rq), rq, 0);
}
}
--
2.7.4
On Fri, 18 Jun 2021 at 15:32, Vincent Donnefort
<[email protected]> wrote:
>
> RT keeps track of the utilization on a per-rq basis with the structure
> avg_rt. This utilization is updated during task_tick_rt(),
> put_prev_task_rt() and set_next_task_rt(). However, when the current
> running task changes its policy, set_next_task_rt() which would usually
> take care of updating the utilization when the rq starts running RT tasks,
> will not see a such change, leaving the avg_rt structure outdated. When
> that very same task will be dequeued later, put_prev_task_rt() will then
> update the utilization, based on a wrong last_update_time, leading a huge
> spike in the RT utilization signal.
>
> The signal would eventually recover from this issue after few ms. Even if
> no RT tasks are run, avg_rt is also updated in __update_blocked_others().
> But as the CPU capacity depends partly on the avg_rt, this issue has
> nonetheless a significant impact on the scheduler.
>
> Fix this issue by ensuring a load update when a running task changes
> its policy to RT.
>
> Fixes: 371bf427 ("sched/rt: Add rt_rq utilization tracking")
> Signed-off-by: Vincent Donnefort <[email protected]>
>
> diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c
> index a525447..5ac0f31 100644
> --- a/kernel/sched/rt.c
> +++ b/kernel/sched/rt.c
> @@ -2341,10 +2341,8 @@ void __init init_sched_rt_class(void)
> static void switched_to_rt(struct rq *rq, struct task_struct *p)
> {
> /*
> - * If we are already running, then there's nothing
> - * that needs to be done. But if we are not running
> - * we may need to preempt the current running task.
> - * If that current running task is also an RT task
> + * If we are not running we may need to preempt the current
> + * running task. If that current running task is also an RT task
> * then see if we can move to another run queue.
> */
> if (task_on_rq_queued(p) && rq->curr != p) {
> @@ -2355,6 +2353,13 @@ static void switched_to_rt(struct rq *rq, struct task_struct *p)
> if (p->prio < rq->curr->prio && cpu_online(cpu_of(rq)))
> resched_curr(rq);
> }
> +
> + /*
> + * If we are running, update the avg_rt tracking, as the running time
> + * will now on be accounted into the latter.
> + */
> + if (task_current(rq, p))
task_current(rq,p) equals the "rq->curr != p" above. Having 2
different ways to get the same value in the function is a bit
confusing.
And call it once
Otherwise, the fix looks good to me
> + update_rt_rq_load_avg(rq_clock_pelt(rq), rq, 0);
> }
>
> /*
> --
> 2.7.4
>
On Fri, 18 Jun 2021 at 15:32, Vincent Donnefort
<[email protected]> wrote:
>
> DL keeps track of the utilization on a per-rq basis with the structure
> avg_dl. This utilization is updated during task_tick_dl(),
> put_prev_task_dl() and set_next_task_dl(). However, when the current
> running task changes its policy, set_next_task_dl() which would usually
> take care of updating the utilization when the rq starts running DL
> tasks, will not see a such change, leaving the avg_dl structure outdated.
> When that very same task will be dequeued later, put_prev_task_dl() will
> then update the utilization, based on a wrong last_update_time, leading a
> huge spike in the DL utilization signal.
>
> The signal would eventually recover from this issue after few ms. Even
> if no DL tasks are run, avg_dl is also updated in
> __update_blocked_others(). But as the CPU capacity depends partly on the
> avg_dl, this issue has nonetheless a significant impact on the scheduler.
>
> Fix this issue by ensuring a load update when a running task changes
> its policy to DL.
>
> Fixes: 3727e0e ("sched/dl: Add dl_rq utilization tracking")
> Signed-off-by: Vincent Donnefort <[email protected]>
Reviewed-by: Vincent Guittot <[email protected]>
>
> diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c
> index 3829c5a..915227a 100644
> --- a/kernel/sched/deadline.c
> +++ b/kernel/sched/deadline.c
> @@ -2497,6 +2497,8 @@ static void switched_to_dl(struct rq *rq, struct task_struct *p)
> check_preempt_curr_dl(rq, p, 0);
> else
> resched_curr(rq);
> + } else {
> + update_dl_rq_load_avg(rq_clock_pelt(rq), rq, 0);
> }
> }
>
> --
> 2.7.4
>
[...]
> > /*
> > - * If we are already running, then there's nothing
> > - * that needs to be done. But if we are not running
> > - * we may need to preempt the current running task.
> > - * If that current running task is also an RT task
> > + * If we are not running we may need to preempt the current
> > + * running task. If that current running task is also an RT task
> > * then see if we can move to another run queue.
> > */
> > if (task_on_rq_queued(p) && rq->curr != p) {
> > @@ -2355,6 +2353,13 @@ static void switched_to_rt(struct rq *rq, struct task_struct *p)
> > if (p->prio < rq->curr->prio && cpu_online(cpu_of(rq)))
> > resched_curr(rq);
> > }
> > +
> > + /*
> > + * If we are running, update the avg_rt tracking, as the running time
> > + * will now on be accounted into the latter.
> > + */
> > + if (task_current(rq, p))
>
> task_current(rq,p) equals the "rq->curr != p" above. Having 2
> different ways to get the same value in the function is a bit
> confusing.
>
> And call it once
>
> Otherwise, the fix looks good to me
I can change it to something like:
if (task_current(rq, p)) {
update_rt_rq_load_avg()
return;
}
if (task_on_rq_queued(p) {
...
if (p->prio ...)
resched_curr(rq);
}
?
>
> > + update_rt_rq_load_avg(rq_clock_pelt(rq), rq, 0);
> > }
> >
> > /*
> > --
> > 2.7.4
> >
On Mon, 21 Jun 2021 at 11:03, Vincent Donnefort
<[email protected]> wrote:
>
> [...]
>
> > > /*
> > > - * If we are already running, then there's nothing
> > > - * that needs to be done. But if we are not running
> > > - * we may need to preempt the current running task.
> > > - * If that current running task is also an RT task
> > > + * If we are not running we may need to preempt the current
> > > + * running task. If that current running task is also an RT task
> > > * then see if we can move to another run queue.
> > > */
> > > if (task_on_rq_queued(p) && rq->curr != p) {
> > > @@ -2355,6 +2353,13 @@ static void switched_to_rt(struct rq *rq, struct task_struct *p)
> > > if (p->prio < rq->curr->prio && cpu_online(cpu_of(rq)))
> > > resched_curr(rq);
> > > }
> > > +
> > > + /*
> > > + * If we are running, update the avg_rt tracking, as the running time
> > > + * will now on be accounted into the latter.
> > > + */
> > > + if (task_current(rq, p))
> >
> > task_current(rq,p) equals the "rq->curr != p" above. Having 2
> > different ways to get the same value in the function is a bit
> > confusing.
> >
> > And call it once
> >
> > Otherwise, the fix looks good to me
>
> I can change it to something like:
>
> if (task_current(rq, p)) {
> update_rt_rq_load_avg()
> return;
> }
>
> if (task_on_rq_queued(p) {
> ...
> if (p->prio ...)
> resched_curr(rq);
> }
>
> ?
yes, looks good to me
>
> >
> > > + update_rt_rq_load_avg(rq_clock_pelt(rq), rq, 0);
> > > }
> > >
> > > /*
> > > --
> > > 2.7.4
> > >