Re: [PATCH v4] sched/deadline: support dl task migration during cpu hotplug

[Kirill Tkhai <ktkhai@parallels.com>]

В Вт, 11/11/2014 в 21:07 +0800, Wanpeng Li пишет:
> Hi Kirill,
> On 11/11/14, 7:10 PM, Kirill Tkhai wrote:
> > В Вт, 11/11/2014 в 10:30 +0800, Wanpeng Li пишет:
> >> I observe that dl task can't be migrated to other cpus during cpu hotplug, in
> >> addition, task may/may not be running again if cpu is added back. The root cause
> >> which I found is that dl task will be throtted and removed from dl rq after
> >> comsuming all budget, which leads to stop task can't pick it up from dl rq and
> >> migrate to other cpus during hotplug.
> >>
> >> The method to reproduce:
> >> schedtool -E -t 50000:100000 -e ./test
> >> Actually test is just a simple for loop. Then observe which cpu the test
> >> task is on.
> >> echo 0 > /sys/devices/system/cpu/cpuN/online
> >>
> >> This patch fix it by push the task to another cpu in dl_task_timer() if
> >> rq is offline.
> >>
> >> Signed-off-by: Wanpeng Li <wanpeng.li@linux.intel.com>
> > I'm still thinking we don't have to guarantee any "deadlines" during cpu hotplug...
> > But, if speaking about this way:
> >
> >> ---
> >> v3 -> v4:
> >>   * use tsk_cpus_allowed wrapper
> >>   * fix compile error
> >> v2 -> v3:
> >>   * don't get_task_struct
> >>   * if cannot preempt any rq, fallback to pick any online cpus
> >>   * use cpu_active_mask as original later_mask if cpu is offline
> >> v1 -> v2:
> >>   * push the task to another cpu in dl_task_timer() if rq is offline.
> >>
> >>
> >>   kernel/sched/deadline.c | 51 ++++++++++++++++++++++++++++++++++++++++++++++---
> >>   1 file changed, 48 insertions(+), 3 deletions(-)
> >>
> >> diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c
> >> index 00324af..e0fbba4 100644
> >> --- a/kernel/sched/deadline.c
> >> +++ b/kernel/sched/deadline.c
> >> @@ -487,6 +487,7 @@ static int start_dl_timer(struct sched_dl_entity *dl_se, bool boosted)
> >>   	return hrtimer_active(&dl_se->dl_timer);
> >>   }
> >>   
> >> +static struct rq *find_lock_later_rq(struct task_struct *task, struct rq *rq);
> >>   /*
> >>    * This is the bandwidth enforcement timer callback. If here, we know
> >>    * a task is not on its dl_rq, since the fact that the timer was running
> >> @@ -538,6 +539,46 @@ again:
> >>   	update_rq_clock(rq);
> >>   	dl_se->dl_throttled = 0;
> >>   	dl_se->dl_yielded = 0;
> >> +
> >> +	/*
> >> +	 * So if we find that the rq the task was on is no longer
> >> +	 * available, we need to select a new rq.
> >> +	 */
> >> +	if (!rq->online) {
> >> +		struct rq *later_rq = NULL;
> >> +
> >> +		raw_spin_unlock(&rq->lock);
> >> +
> >> +		later_rq = find_lock_later_rq(p, rq);
> > find_lock_later_rq() expects that rq is locked.
> >
> > The comment near its head confuses a reader. It locks newly found rq.
> 
> Sorry for my bad, what's you think should be changed?

raw_spin_unlock(&rq->lock) is wrong here. It's not need.

> 
> >
> >> +
> >> +		if (!later_rq) {
> >> +			int cpu;
> >> +
> >> +			/*
> >> +			 * If cannot preempt any rq, fallback to pick any
> >> +			 * online cpu.
> >> +			 */
> >> +			for_each_cpu(cpu, tsk_cpus_allowed(p))
> >> +				if (cpu_online(cpu))
> >> +					later_rq = cpu_rq(cpu);
> >> +			if (!later_rq) {
> >> +				pr_warn("fail to find any online and task "
> >> +				    "will never come back to us\n");
> >> +				goto out;
> >> +			}
> >> +		}
> >> +
> >> +		deactivate_task(rq, p, 0);
> >> +		set_task_cpu(p, later_rq->cpu);
> >> +		activate_task(later_rq, p, 0);
> >> +
> >> +		resched_curr(later_rq);
> >> +
> >> +		double_unlock_balance(rq, later_rq);
> > double_unlock_balance() unlocks later_rq only.
> >
> >> +
> >> +		goto out;
> >> +	}
> >> +
> >>   	if (task_on_rq_queued(p)) {
> >>   		enqueue_task_dl(rq, p, ENQUEUE_REPLENISH);
> >>   		if (dl_task(rq->curr))
> >> @@ -555,7 +596,7 @@ again:
> >>   	}
> >>   unlock:
> >>   	raw_spin_unlock(&rq->lock);
> >> -
> >> +out:
> >>   	return HRTIMER_NORESTART;
> >>   }
> >>   
> >> @@ -1185,8 +1226,12 @@ static int find_later_rq(struct task_struct *task)
> >>   	 * We have to consider system topology and task affinity
> >>   	 * first, then we can look for a suitable cpu.
> >>   	 */
> >> -	cpumask_copy(later_mask, task_rq(task)->rd->span);
> >> -	cpumask_and(later_mask, later_mask, cpu_active_mask);
> >> +	if (likely(task_rq(task)->online)) {
> >> +		cpumask_copy(later_mask, task_rq(task)->rd->span);
> >> +		cpumask_and(later_mask, later_mask, cpu_active_mask);
> >> +	} else
> >> +		/* for offline cpus we have a singleton rd */
> >> +		cpumask_copy(later_mask, cpu_active_mask);
> >>   	cpumask_and(later_mask, later_mask, &task->cpus_allowed);
> >>   	best_cpu = cpudl_find(&task_rq(task)->rd->cpudl,
> >>   			task, later_mask);
> >
> > --
> > To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> > the body of a message to majordomo@vger.kernel.org
> > More majordomo info at  http://vger.kernel.org/majordomo-info.html
> > Please read the FAQ at  http://www.tux.org/lkml/
>