Re: sched: hang in migrate_swap

[Michael wang <wangyun@linux.vnet.ibm.com>]

On 02/24/2014 08:12 PM, Peter Zijlstra wrote:
[snip]
>>
>> ...what about move idle_balance() back to it's old position?
> 
> I've always hated that, idle_balance() is very much a fair policy thing
> and shouldn't live in the core code.
> 
>> pull_rt_task() logical could be after idle_balance() if still no FAIR
>> and DL, then go into the pick loop, that may could make things more
>> clean & clear, should we have a try?
> 
> So the reason pull_{rt,dl}_task() is before idle_balance() is that we
> don't want to add the execution latency of idle_balance() to the rt/dl
> task pulling.

Yeah, that make sense, just wondering... since RT also has balance
stuff, may be we can use a new call back for each class in the old position?

The new idle_balance could like:

	void idle_balance() {
		for_each_class(class)
			if class->idle_balance()
				break
	}

> 
> Anyway, the below seems to work; it avoids playing tricks with the idle
> thread and instead uses a magic constant.
> 
> The comparison should be faster too; seeing how we avoid dereferencing
> p->sched_class.

Great, it once appeared in my mind but you achieved this without new
parameter, now let's ignore my wondering above :)

Regards,
Michael Wang

> 
> ---
> Subject: sched: Guarantee task priority in pick_next_task()
> From: Peter Zijlstra <peterz@infradead.org>
> Date: Fri Feb 14 12:25:08 CET 2014
> 
> Michael spotted that the idle_balance() push down created a task
> priority problem.
> 
> Previously, when we called idle_balance() before pick_next_task() it
> wasn't a problem when -- because of the rq->lock droppage -- an rt/dl
> task slipped in.
> 
> Similarly for pre_schedule(), rt pre-schedule could have a dl task
> slip in.
> 
> But by pulling it into the pick_next_task() loop, we'll not try a
> higher task priority again.
> 
> Cure this by creating a re-start condition in pick_next_task(); and
> triggering this from pick_next_task_{rt,fair}().
> 
> Fixes: 38033c37faab ("sched: Push down pre_schedule() and idle_balance()")
> Cc: Juri Lelli <juri.lelli@gmail.com>
> Cc: Ingo Molnar <mingo@kernel.org>
> Cc: Steven Rostedt <rostedt@goodmis.org>
> Reported-by: Michael Wang <wangyun@linux.vnet.ibm.com>
> Signed-off-by: Peter Zijlstra <peterz@infradead.org>
> ---
>  kernel/sched/core.c  |   12 ++++++++----
>  kernel/sched/fair.c  |   13 ++++++++++++-
>  kernel/sched/rt.c    |   10 +++++++++-
>  kernel/sched/sched.h |    5 +++++
>  4 files changed, 34 insertions(+), 6 deletions(-)
> 
> --- a/kernel/sched/core.c
> +++ b/kernel/sched/core.c
> @@ -2586,24 +2586,28 @@ static inline void schedule_debug(struct
>  static inline struct task_struct *
>  pick_next_task(struct rq *rq, struct task_struct *prev)
>  {
> -	const struct sched_class *class;
> +	const struct sched_class *class = &fair_sched_class;
>  	struct task_struct *p;
> 
>  	/*
>  	 * Optimization: we know that if all tasks are in
>  	 * the fair class we can call that function directly:
>  	 */
> -	if (likely(prev->sched_class == &fair_sched_class &&
> +	if (likely(prev->sched_class == class &&
>  		   rq->nr_running == rq->cfs.h_nr_running)) {
>  		p = fair_sched_class.pick_next_task(rq, prev);
> -		if (likely(p))
> +		if (likely(p && p != RETRY_TASK))
>  			return p;
>  	}
> 
> +again:
>  	for_each_class(class) {
>  		p = class->pick_next_task(rq, prev);
> -		if (p)
> +		if (p) {
> +			if (unlikely(p == RETRY_TASK))
> +				goto again;
>  			return p;
> +		}
>  	}
> 
>  	BUG(); /* the idle class will always have a runnable task */
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -4687,6 +4687,7 @@ pick_next_task_fair(struct rq *rq, struc
>  	struct cfs_rq *cfs_rq = &rq->cfs;
>  	struct sched_entity *se;
>  	struct task_struct *p;
> +	int new_tasks;
> 
>  again:
>  #ifdef CONFIG_FAIR_GROUP_SCHED
> @@ -4785,7 +4786,17 @@ pick_next_task_fair(struct rq *rq, struc
>  	return p;
> 
>  idle:
> -	if (idle_balance(rq)) /* drops rq->lock */
> +	/*
> +	 * Because idle_balance() releases (and re-acquires) rq->lock, it is
> +	 * possible for any higher priority task to appear. In that case we
> +	 * must re-start the pick_next_entity() loop.
> +	 */
> +	new_tasks = idle_balance(rq);
> +
> +	if (rq->nr_running != rq->cfs.h_nr_running)
> +		return RETRY_TASK;
> +
> +	if (new_tasks)
>  		goto again;
> 
>  	return NULL;
> --- a/kernel/sched/rt.c
> +++ b/kernel/sched/rt.c
> @@ -1360,8 +1360,16 @@ pick_next_task_rt(struct rq *rq, struct
>  	struct task_struct *p;
>  	struct rt_rq *rt_rq = &rq->rt;
> 
> -	if (need_pull_rt_task(rq, prev))
> +	if (need_pull_rt_task(rq, prev)) {
>  		pull_rt_task(rq);
> +		/*
> +		 * pull_rt_task() can drop (and re-acquire) rq->lock; this
> +		 * means a dl task can slip in, in which case we need to
> +		 * re-start task selection.
> +		 */
> +		if (unlikely(rq->dl.dl_nr_running))
> +			return RETRY_TASK;
> +	}
> 
>  	if (!rt_rq->rt_nr_running)
>  		return NULL;
> --- a/kernel/sched/sched.h
> +++ b/kernel/sched/sched.h
> @@ -1090,6 +1090,8 @@ static const u32 prio_to_wmult[40] = {
> 
>  #define DEQUEUE_SLEEP		1
> 
> +#define RETRY_TASK		((void *)-1UL)
> +
>  struct sched_class {
>  	const struct sched_class *next;
> 
> @@ -1104,6 +1106,9 @@ struct sched_class {
>  	 * It is the responsibility of the pick_next_task() method that will
>  	 * return the next task to call put_prev_task() on the @prev task or
>  	 * something equivalent.
> +	 *
> +	 * May return RETRY_TASK when it finds a higher prio class has runnable
> +	 * tasks.
>  	 */
>  	struct task_struct * (*pick_next_task) (struct rq *rq,
>  						struct task_struct *prev);
> --
> 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/
>