[PATCH 04/23] Subject: SCHED - RT overloaded runqueues accounting

!MAILaRCHIVE_VOTE_RePLACE
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
To: <mingo@...>
Cc: <rostedt@...>, <ghaskins@...>, <linux-rt-users@...>, <linux-kernel@...>
Date: Tuesday, December 4, 2007 - 4:44 pm

From: Steven Rostedt <srostedt@redhat.com>

This patch adds an RT overload accounting system. When a runqueue has
more than one RT task queued, it is marked as overloaded. That is that it
is a candidate to have RT tasks pulled from it.

Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Gregory Haskins <ghaskins@novell.com>
---

 kernel/sched_rt.c |   36 ++++++++++++++++++++++++++++++++++++
 1 files changed, 36 insertions(+), 0 deletions(-)

diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c
index b5ef4b8..b8c758a 100644
--- a/kernel/sched_rt.c
+++ b/kernel/sched_rt.c
@@ -3,6 +3,38 @@
  * policies)
  */
 
+#ifdef CONFIG_SMP
+static cpumask_t rt_overload_mask;
+static atomic_t rto_count;
+static inline int rt_overloaded(void)
+{
+	return atomic_read(&rto_count);
+}
+static inline cpumask_t *rt_overload(void)
+{
+	return &rt_overload_mask;
+}
+static inline void rt_set_overload(struct rq *rq)
+{
+	cpu_set(rq->cpu, rt_overload_mask);
+	/*
+	 * Make sure the mask is visible before we set
+	 * the overload count. That is checked to determine
+	 * if we should look at the mask. It would be a shame
+	 * if we looked at the mask, but the mask was not
+	 * updated yet.
+	 */
+	wmb();
+	atomic_inc(&rto_count);
+}
+static inline void rt_clear_overload(struct rq *rq)
+{
+	/* the order here really doesn't matter */
+	atomic_dec(&rto_count);
+	cpu_clear(rq->cpu, rt_overload_mask);
+}
+#endif /* CONFIG_SMP */
+
 /*
  * Update the current task's runtime statistics. Skip current tasks that
  * are not in our scheduling class.
@@ -33,6 +65,8 @@ static inline void inc_rt_tasks(struct task_struct *p, struct rq *rq)
 #ifdef CONFIG_SMP
 	if (p->prio < rq->rt.highest_prio)
 		rq->rt.highest_prio = p->prio;
+	if (rq->rt.rt_nr_running > 1)
+		rt_set_overload(rq);
 #endif /* CONFIG_SMP */
 }
 
@@ -54,6 +88,8 @@ static inline void dec_rt_tasks(struct task_struct *p, struct rq *rq)
 		} /* otherwise leave rq->highest prio alone */
 	} else
 		rq->rt.highest_prio = MAX_RT_PRIO;
+	if (rq->rt.rt_nr_running < 2)
+		rt_clear_overload(rq);
 #endif /* CONFIG_SMP */
 }
 

--
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

Messages in current thread:
[PATCH 00/23] RT balance v7, Gregory Haskins, (Tue Dec 4, 4:44 pm)
Re: [PATCH 00/23] RT balance v7, Ingo Molnar, (Tue Dec 4, 5:27 pm)
[PATCH 0/3] RT balance v7a, Gregory Haskins, (Tue Dec 4, 10:55 pm)
[PATCH 1/3] Subject: SCHED - Add sched-domain roots, Gregory Haskins, (Tue Dec 4, 10:55 pm)
Re: [PATCH 00/23] RT balance v7, Gregory Haskins, (Tue Dec 4, 5:35 pm)
[PATCH 21/23] Subject: SCHED - Add sched-domain roots, Gregory Haskins, (Tue Dec 4, 4:46 pm)
[PATCH 16/23] Subject: SCHED - Avoid overload, Gregory Haskins, (Tue Dec 4, 4:45 pm)
[PATCH 19/23] Subject: SCHED - Optimize out cpu_clears, Gregory Haskins, (Tue Dec 4, 4:46 pm)
[PATCH 15/23] Subject: SCHED - Optimize rebalancing, Gregory Haskins, (Tue Dec 4, 4:45 pm)
[PATCH 03/23] Subject: SCHED - push RT tasks, Gregory Haskins, (Tue Dec 4, 4:44 pm)
[PATCH 06/23] Subject: SCHED - wake up balance RT, Gregory Haskins, (Tue Dec 4, 4:44 pm)
[PATCH 05/23] Subject: SCHED - pull RT tasks, Gregory Haskins, (Tue Dec 4, 4:44 pm)
[PATCH 04/23] Subject: SCHED - RT overloaded runqueues accou..., Gregory Haskins, (Tue Dec 4, 4:44 pm)