Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1301366

[PATCH for v3.18.25 2/6] sched: Allow balance callbacks for check_class_changed()

Show all headers | View raw

From: Peter Zijlstra <peterz@infradead.org>

In order to remove dropping rq->lock from the
switched_{to,from}()/prio_changed() sched_class methods, run the
balance callbacks after it.

We need to remove dropping rq->lock because its buggy,
suppose using sched_setattr()/sched_setscheduler() to change a running
task from FIFO to OTHER.

By the time we get to switched_from_rt() the task is already enqueued
on the cfs runqueues. If switched_from_rt() does pull_rt_task() and
drops rq->lock, load-balancing can come in and move our task @p to
another rq.

The subsequent switched_to_fair() still assumes @p is on @rq and bad
things will happen.

By using balance callbacks we delay the load-balancing operations
{rt,dl}x{push,pull} until we've done all the important work and the
task is fully set up.

Furthermore, the balance callbacks do not know about @p, therefore
they cannot get confused like this.

Reported-by: Mike Galbraith <umgwanakikbuti@gmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: ktkhai@parallels.com
Cc: rostedt@goodmis.org
Cc: juri.lelli@gmail.com
Cc: pang.xunlei@linaro.org
Cc: oleg@redhat.com
Cc: wanpeng.li@linux.intel.com
Link: http://lkml.kernel.org/r/20150611124742.615343911@infradead.org
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Byungchul Park <byungchul.park@lge.com>

Conflicts:
	kernel/sched/core.c
---
 kernel/sched/core.c | 24 +++++++++++++++++++++++-
 1 file changed, 23 insertions(+), 1 deletion(-)

diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 3dff0ef..ca6dad6 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -999,6 +999,13 @@ inline int task_curr(const struct task_struct *p)
 	return cpu_curr(task_cpu(p)) == p;
 }
 
+/*
+ * switched_from, switched_to and prio_changed must _NOT_ drop rq->lock,
+ * use the balance_callback list if you want balancing.
+ *
+ * this means any call to check_class_changed() must be followed by a call to
+ * balance_callback().
+ */
 static inline void check_class_changed(struct rq *rq, struct task_struct *p,
 				       const struct sched_class *prev_class,
 				       int oldprio)
@@ -1485,8 +1492,12 @@ ttwu_do_wakeup(struct rq *rq, struct task_struct *p, int wake_flags)
 
 	p->state = TASK_RUNNING;
 #ifdef CONFIG_SMP
-	if (p->sched_class->task_woken)
+	if (p->sched_class->task_woken) {
+		/*
+		 * XXX can drop rq->lock; most likely ok.
+		 */
 		p->sched_class->task_woken(rq, p);
+	}
 
 	if (rq->idle_stamp) {
 		u64 delta = rq_clock(rq) - rq->idle_stamp;
@@ -3032,7 +3043,11 @@ void rt_mutex_setprio(struct task_struct *p, int prio)
 
 	check_class_changed(rq, p, prev_class, oldprio);
 out_unlock:
+	preempt_disable(); /* avoid rq from going away on us */
 	__task_rq_unlock(rq);
+
+	balance_callback(rq);
+	preempt_enable();
 }
 #endif
 
@@ -3559,10 +3574,17 @@ change:
 	}
 
 	check_class_changed(rq, p, prev_class, oldprio);
+	preempt_disable(); /* avoid rq from going away on us */
 	task_rq_unlock(rq, p, &flags);
 
 	rt_mutex_adjust_pi(p);
 
+	/*
+	 * Run balance callbacks after we've adjusted the PI chain.
+	 */
+	balance_callback(rq);
+	preempt_enable();
+
 	return 0;
 }
 
-- 
1.9.1

--
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/

Back to linux.kernel | Previous | Next — Previous in thread | Find similar | Unroll thread

Thread

[STABLE] kernel oops which can be fixed by peterz's patches Byungchul Park <byungchul.park@lge.com> - 2016-01-05 10:00 +0100
  [PATCH for v3.14.58 7/7] sched, dl: Convert switched_{from, to}_dl() / prio_changed_dl() to balance callbacks Byungchul Park <byungchul.park@lge.com> - 2016-01-05 10:20 +0100
  Re: [STABLE] kernel oops which can be fixed by peterz's patches Peter Zijlstra <peterz@infradead.org> - 2016-01-05 10:20 +0100
    Re: [STABLE] kernel oops which can be fixed by peterz's patches Byungchul Park <byungchul.park@lge.com> - 2016-01-12 09:50 +0100
    Re: [STABLE] kernel oops which can be fixed by peterz's patches Willy Tarreau <w@1wt.eu> - 2016-01-12 11:30 +0100
  [PATCH for v3.14.58 6/7] sched,dl: Remove return value from pull_dl_task() Byungchul Park <byungchul.park@lge.com> - 2016-01-05 10:20 +0100
  [PATCH for v3.14.58 1/7] sched: Clean up idle task SMP logic Byungchul Park <byungchul.park@lge.com> - 2016-01-05 10:20 +0100
    [PATCH for v3.14.58 4/7] sched,rt: Remove return value from pull_rt_task() Byungchul Park <byungchul.park@lge.com> - 2016-01-05 10:20 +0100
    [PATCH for v3.14.58 2/7] sched: Replace post_schedule with a balance callback list Byungchul Park <byungchul.park@lge.com> - 2016-01-05 10:20 +0100
    [PATCH for v3.14.58 3/7] sched: Allow balance callbacks for check_class_changed() Byungchul Park <byungchul.park@lge.com> - 2016-01-05 10:20 +0100
    [PATCH for v3.14.58 5/7] sched, rt: Convert switched_{from, to}_rt() / prio_changed_rt() to balance callbacks Byungchul Park <byungchul.park@lge.com> - 2016-01-05 10:20 +0100
  [PATCH for v3.18.25 1/6] sched: Replace post_schedule with a balance callback list Byungchul Park <byungchul.park@lge.com> - 2016-01-05 10:30 +0100
    [PATCH for v3.18.25 3/6] sched,rt: Remove return value from pull_rt_task() Byungchul Park <byungchul.park@lge.com> - 2016-01-05 10:30 +0100
    [PATCH for v3.18.25 4/6] sched, rt: Convert switched_{from, to}_rt() / prio_changed_rt() to balance callbacks Byungchul Park <byungchul.park@lge.com> - 2016-01-05 10:30 +0100
    [PATCH for v3.18.25 6/6] sched, dl: Convert switched_{from, to}_dl() / prio_changed_dl() to balance callbacks Byungchul Park <byungchul.park@lge.com> - 2016-01-05 10:30 +0100
    [PATCH for v3.18.25 5/6] sched,dl: Remove return value from pull_dl_task() Byungchul Park <byungchul.park@lge.com> - 2016-01-05 10:30 +0100
    [PATCH for v3.18.25 2/6] sched: Allow balance callbacks for check_class_changed() Byungchul Park <byungchul.park@lge.com> - 2016-01-05 10:30 +0100

csiph-web