[PATCH -next 0/1] Restart rt period timer when rt runtime exceeded

[PATCH -next 0/1] Restart rt period timer when rt runtime exceeded

[Li Hua]

When rt_runtime is modified from -1 to a valid control value, it may cause
the task to be throttled all the time. Then the task cannot be killed.
E.g:
The FIFO task A execution while(1):
	#define _GNU_SOURCE
	#include <sched.h>
	#include <unistd.h>

	int main(int argc,char *argv[])
	{
		struct sched_param param;
		int priority = sched_get_priority_min(SCHED_FIFO);
		param.sched_priority = priority;
		sched_setscheduler(getpid(), SCHED_FIFO, &param);
		while(1) {}
		return 0;
	}

Following the steps:
lihua@octopus ~ # echo -1 > /proc/sys/kernel/sched_rt_runtime_us
lihua@octopus ~ # /bin/rt_throttled &
lihua@octopus ~ # echo 950000 > /proc/sys/kernel/sched_rt_runtime_us
lihua@octopus ~ # ps aux | grep rt_throttled
root 7026 65.1  0.0   1848   180 ttyAMA0  R    01:27   0:26 /bin/rt_throttled
root 7068  0.0  0.0   4072   620 ttyAMA0  S+   01:28   0:00 grep rt_throttled
lihua@octopus ~ # killall rt_throttled
lihua@octopus ~ # date
Thu Jan  1 01:28:30 UTC 1970
lihua@octopus ~ # date
Thu Jan  1 01:28:33 UTC 1970
lihua@octopus ~ # ps aux | grep rt_throttled
root 7026 41.0  0.0   1848   180 ttyAMA0  R    01:27   0:26 /bin/rt_throttled
root 7097  0.0  0.0   4072   576 ttyAMA0  S+   01:28   0:00 grep rt_throttled

So the patch try to wake up the timer when rt runtime exceeded.

*** BLURB HERE ***

Li Hua (1):
  sched/rt: Try to restart rt period timer when rt runtime exceeded

 kernel/sched/rt.c | 26 +++++++++++++++++++++++++-
 1 file changed, 25 insertions(+), 1 deletion(-)

-- 
2.17.1

[PATCH -next 1/1] sched/rt: Try to restart rt period timer when rt runtime exceeded

[Li Hua]

When rt_runtime is modified from -1 to a valid control value, it may
cause the task to be throttled all the time. Operations like the following
will trigger the bug. E.g:
1. echo -1 > /proc/sys/kernel/sched_rt_runtime_us
2. Run a FIFO task named A that executes while(1)
3. echo 950000 > /proc/sys/kernel/sched_rt_runtime_us

When rt_runtime is -1, The rt period timer will not be activated when task A
enqueued. And then the task will be throttled after setting rt_runtime to
950,000. The task will always be throttled because the rt period timer is not
activated.

Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Li Hua <hucool.lihua@huawei.com>
---
 kernel/sched/rt.c | 26 +++++++++++++++++++++++++-
 1 file changed, 25 insertions(+), 1 deletion(-)

diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c
index bb945f8faeca..630f2cbe37d0 100644
--- a/kernel/sched/rt.c
+++ b/kernel/sched/rt.c
@@ -947,6 +947,23 @@ static inline int rt_se_prio(struct sched_rt_entity *rt_se)
 	return rt_task_of(rt_se)->prio;
 }
 
+static inline void try_start_rt_bandwidth(struct rt_bandwidth *rt_b)
+{
+	raw_spin_lock(&rt_b->rt_runtime_lock);
+	if (!rt_bandwidth_enabled() || rt_b->rt_runtime == RUNTIME_INF) {
+		raw_spin_unlock(&rt_b->rt_runtime_lock);
+		return;
+	}
+
+	if (!rt_b->rt_period_active) {
+		rt_b->rt_period_active = 1;
+		hrtimer_forward_now(&rt_b->rt_period_timer, rt_b->rt_period);
+		hrtimer_start_expires(&rt_b->rt_period_timer,
+				      HRTIMER_MODE_ABS_PINNED_HARD);
+	}
+	raw_spin_unlock(&rt_b->rt_runtime_lock);
+}
+
 static int sched_rt_runtime_exceeded(struct rt_rq *rt_rq)
 {
 	u64 runtime = sched_rt_runtime(rt_rq);
@@ -1027,11 +1044,16 @@ static void update_curr_rt(struct rq *rq)
 		struct rt_rq *rt_rq = rt_rq_of_se(rt_se);
 
 		if (sched_rt_runtime(rt_rq) != RUNTIME_INF) {
+			int exceeded;
+
 			raw_spin_lock(&rt_rq->rt_runtime_lock);
 			rt_rq->rt_time += delta_exec;
-			if (sched_rt_runtime_exceeded(rt_rq))
+			exceeded = sched_rt_runtime_exceeded(rt_rq);
+			if (exceeded)
 				resched_curr(rq);
 			raw_spin_unlock(&rt_rq->rt_runtime_lock);
+			if (exceeded)
+				try_start_rt_bandwidth(sched_rt_bandwidth(rt_rq));
 		}
 	}
 }
@@ -2905,8 +2927,10 @@ static int sched_rt_global_validate(void)
 
 static void sched_rt_do_global(void)
 {
+	raw_spin_lock(&def_rt_bandwidth.rt_runtime_lock);
 	def_rt_bandwidth.rt_runtime = global_rt_runtime();
 	def_rt_bandwidth.rt_period = ns_to_ktime(global_rt_period());
+	raw_spin_unlock(&def_rt_bandwidth.rt_runtime_lock);
 }
 
 int sched_rt_handler(struct ctl_table *table, int write, void *buffer,
-- 
2.17.1

Re: [PATCH -next 1/1] sched/rt: Try to restart rt period timer when rt runtime exceeded

[Peter Zijlstra]

On Mon, Nov 15, 2021 at 01:46:28AM +0000, Li Hua wrote:
> When rt_runtime is modified from -1 to a valid control value, it may
> cause the task to be throttled all the time. Operations like the following
> will trigger the bug. E.g:
> 1. echo -1 > /proc/sys/kernel/sched_rt_runtime_us
> 2. Run a FIFO task named A that executes while(1)
> 3. echo 950000 > /proc/sys/kernel/sched_rt_runtime_us
> 
> When rt_runtime is -1, The rt period timer will not be activated when task A
> enqueued. And then the task will be throttled after setting rt_runtime to
> 950,000. The task will always be throttled because the rt period timer is not
> activated.
> 
> Reported-by: Hulk Robot <hulkci@huawei.com>
> Signed-off-by: Li Hua <hucool.lihua@huawei.com>
> ---
>  kernel/sched/rt.c | 26 +++++++++++++++++++++++++-
>  1 file changed, 25 insertions(+), 1 deletion(-)
> 
> diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c
> index bb945f8faeca..630f2cbe37d0 100644
> --- a/kernel/sched/rt.c
> +++ b/kernel/sched/rt.c
> @@ -947,6 +947,23 @@ static inline int rt_se_prio(struct sched_rt_entity *rt_se)
>  	return rt_task_of(rt_se)->prio;
>  }
>  
> +static inline void try_start_rt_bandwidth(struct rt_bandwidth *rt_b)
> +{
> +	raw_spin_lock(&rt_b->rt_runtime_lock);
> +	if (!rt_bandwidth_enabled() || rt_b->rt_runtime == RUNTIME_INF) {
> +		raw_spin_unlock(&rt_b->rt_runtime_lock);
> +		return;
> +	}
> +
> +	if (!rt_b->rt_period_active) {
> +		rt_b->rt_period_active = 1;
> +		hrtimer_forward_now(&rt_b->rt_period_timer, rt_b->rt_period);
> +		hrtimer_start_expires(&rt_b->rt_period_timer,
> +				      HRTIMER_MODE_ABS_PINNED_HARD);
> +	}
> +	raw_spin_unlock(&rt_b->rt_runtime_lock);
> +}

This is almost a verbatim copy of start_rt_bandwidth() surely we can do
better.

> +
>  static int sched_rt_runtime_exceeded(struct rt_rq *rt_rq)
>  {
>  	u64 runtime = sched_rt_runtime(rt_rq);
> @@ -1027,11 +1044,16 @@ static void update_curr_rt(struct rq *rq)
>  		struct rt_rq *rt_rq = rt_rq_of_se(rt_se);
>  
>  		if (sched_rt_runtime(rt_rq) != RUNTIME_INF) {
> +			int exceeded;
> +
>  			raw_spin_lock(&rt_rq->rt_runtime_lock);
>  			rt_rq->rt_time += delta_exec;
> -			if (sched_rt_runtime_exceeded(rt_rq))
> +			exceeded = sched_rt_runtime_exceeded(rt_rq);
> +			if (exceeded)
>  				resched_curr(rq);
>  			raw_spin_unlock(&rt_rq->rt_runtime_lock);
> +			if (exceeded)
> +				try_start_rt_bandwidth(sched_rt_bandwidth(rt_rq));
>  		}
>  	}
>  }
> @@ -2905,8 +2927,10 @@ static int sched_rt_global_validate(void)
>  
>  static void sched_rt_do_global(void)
>  {
> +	raw_spin_lock(&def_rt_bandwidth.rt_runtime_lock);
>  	def_rt_bandwidth.rt_runtime = global_rt_runtime();
>  	def_rt_bandwidth.rt_period = ns_to_ktime(global_rt_period());
> +	raw_spin_unlock(&def_rt_bandwidth.rt_runtime_lock);

And that's just wrong I think; did you test this with lockdep enabled?
IIRC this lock is irq-safe, it has to be if you're using it form a timer
context.