3.4.22-rt34-rc1 stable review

Dear RT Folks,

This is the RT stable review cycle of patch 3.4.22-rt34-rc1.

Please scream at me if I messed something up. Please test the patches too.

The -rc release will be uploaded to kernel.org and will be deleted when the final release is out. This is just a review release (or release candidate).

The pre-releases will not be pushed to the git repository, only the final release is.

If all goes well, this patch will be converted to the next main release on 12/14/2012, let's hope it's not an unexpected journey!

Enjoy,

-- Steve

To build 3.4.22-rt34-rc1 directly, the following patches should be applied:

http://www.kernel.org/pub/linux/kernel/v3.x/linux-3.4.tar.xz

http://www.kernel.org/pub/linux/kernel/v3.x/patch-3.4.22.xz

http://www.kernel.org/pub/linux/kernel/projects/rt/3.4/patch-3.4.22-rt34-rc1.patch.xz

You can also build from 3.4.22-rt33 by applying the incremental patch:

http://www.kernel.org/pub/linux/kernel/projects/rt/3.4/incr/patch-3.4.22-rt33-rt34-rc1.patch.xz

Changes from 3.4.22-rt33:

---

Steven Rostedt (1):

      Linux 3.4.22-rt34-rc1

Thomas Gleixner (1):
sched: Queue RT tasks to head when prio drops

----

 kernel/sched/core.c |   16 ++++++++++++----
 localversion-rt     |    2 +-
 2 files changed, 13 insertions(+), 5 deletions(-)

From: Thomas Gleixner <[email protected]>

The following scenario does not work correctly:

Runqueue of CPU1 contains two runnable and pinned tasks:

	 T1: SCHED_FIFO, prio 80
	 T2: SCHED_FIFO, prio 80

T1 is on the cpu and executes the following syscalls (classic priority ceiling scenario):

sys_sched_setscheduler(pid(T1), SCHED_FIFO, .prio = 90);
...
sys_sched_setscheduler(pid(T1), SCHED_FIFO, .prio = 80);
...

Now T1 gets preempted by T3 (SCHED_FIFO, prio 95). After T3 goes back to sleep the scheduler picks T2. Surprise!

The same happens w/o actual preemption when T1 is forced into the scheduler due to a sporadic NEED_RESCHED event. The scheduler invokes pick_next_task() which returns T2. So T1 gets preempted and scheduled out.

This happens because sched_setscheduler() dequeues T1 from the prio 90 list and then enqueues it on the tail of the prio 80 list behind T2. This violates the POSIX spec and surprises user space which relies on the guarantee that SCHED_FIFO tasks are not scheduled out unless they give the CPU up voluntarily or are preempted by a higher priority task. In the latter case the preempted task must get back on the CPU after the preempting task schedules out again.

We fixed a similar issue already in commit 60db48c(sched: Queue a deboosted task to the head of the RT prio queue). The same treatment is necessary for sched_setscheduler().

While analyzing the problem I noticed that the fix in rt_mutex_setprio() is one off. The head queueing depends on old priority greater than new priority (user space view), but in fact it needs to have the same treatment for equal priority. Instead of blindly changing the condition to <= it's better to avoid the whole dequeue/requeue business for the equal priority case completely.

---
 kernel/sched/core.c |   16 ++++++++++++----
 1 file changed, 12 insertions(+), 4 deletions(-)

index 1f9d6f5..054e669 100644
@@ -4117,6 +4117,8 @@ void rt_mutex_setprio(struct task_struct *p, int prio)
 
 	trace_sched_pi_setprio(p, prio);
 	oldprio = p->prio;
+	if (oldprio == prio)
+		goto out_unlock;
 	prev_class = p->sched_class;
 	on_rq = p->on_rq;
 	running = task_current(rq, p);

@@ -4472,6 +4474,13 @@ recheck:
 		task_rq_unlock(rq, p, &flags);
 		goto recheck;
 	}
+
+	p->sched_reset_on_fork = reset_on_fork;
+
+	oldprio = p->prio;
+	if (oldprio == param->sched_priority)
+		goto out;
+
 	on_rq = p->on_rq;
 	running = task_current(rq, p);
 	if (on_rq)

@@ -4479,18 +4488,17 @@ recheck:
 	if (running)
 		p->sched_class->put_prev_task(rq, p);
 
-	p->sched_reset_on_fork = reset_on_fork;
-
-	oldprio = p->prio;
 	prev_class = p->sched_class;
 	__setscheduler(rq, p, policy, param->sched_priority);
 
 	if (running)
 		p->sched_class->set_curr_task(rq);
 	if (on_rq)
-		enqueue_task(rq, p, 0);
+		enqueue_task(rq, p, oldprio < param->sched_priority ?
+			     ENQUEUE_HEAD : 0);
 
 	check_class_changed(rq, p, prev_class, oldprio);
+out:
 	task_rq_unlock(rq, p, &flags);
 
 	rt_mutex_adjust_pi(p);

1.7.10.4

Hello,

we tested this patch successfully with 3.2.35-rt53-rc1. The failure could not be reproduced in a 16 hours test run, so we are pretty sure that the failure is gone.

Thanks for the quick inclusion of this pacth in stable-rt series!

Best Regards, Gerhard Engleder

-----Ursprüngliche Nachricht-----
Von: [email protected] [mailto:[email protected]] Im Auftrag von Steven Rostedt
Gesendet: Mittwoch, 12. Dezember 2012 01:46
An: [email protected]; linux-rt-users
Cc: Thomas Gleixner; Carsten Emde; John Kacur; [email protected]; [email protected]
Betreff: [PATCH RT 1/2] sched: Queue RT tasks to head when prio drops

From: Thomas Gleixner <[email protected]>

The following scenario does not work correctly:

Runqueue of CPU1 contains two runnable and pinned tasks:

	 T1: SCHED_FIFO, prio 80
	 T2: SCHED_FIFO, prio 80

T1 is on the cpu and executes the following syscalls (classic priority ceiling scenario):

sys_sched_setscheduler(pid(T1), SCHED_FIFO, .prio = 90); ... sys_sched_setscheduler(pid(T1), SCHED_FIFO, .prio = 80); ...

Now T1 gets preempted by T3 (SCHED_FIFO, prio 95). After T3 goes back to sleep the scheduler picks T2. Surprise!

The same happens w/o actual preemption when T1 is forced into the scheduler due to a sporadic NEED_RESCHED event. The scheduler invokes pick_next_task() which returns T2. So T1 gets preempted and scheduled out.

This happens because sched_setscheduler() dequeues T1 from the prio 90 list and then enqueues it on the tail of the prio 80 list behind T2.
This violates the POSIX spec and surprises user space which relies on the guarantee that SCHED_FIFO tasks are not scheduled out unless they give the CPU up voluntarily or are preempted by a higher priority task. In the latter case the preempted task must get back on the CPU after the preempting task schedules out again.

We fixed a similar issue already in commit 60db48c(sched: Queue a deboosted task to the head of the RT prio queue). The same treatment is necessary for sched_setscheduler().

While analyzing the problem I noticed that the fix in
rt_mutex_setprio() is one off. The head queueing depends on old priority greater than new priority (user space view), but in fact it needs to have the same treatment for equal priority. Instead of blindly changing the condition to <= it's better to avoid the whole dequeue/requeue business for the equal priority case completely.

---
 kernel/sched/core.c |   16 ++++++++++++----
 1 file changed, 12 insertions(+), 4 deletions(-)

@@ -4117,6 +4117,8 @@ void rt_mutex_setprio(struct task_struct *p, int prio)
 
 	trace_sched_pi_setprio(p, prio);
 	oldprio = p->prio;
+	if (oldprio == prio)
+		goto out_unlock;
 	prev_class = p->sched_class;
 	on_rq = p->on_rq;
 	running = task_current(rq, p);

@@ -4472,6 +4474,13 @@ recheck:
 		task_rq_unlock(rq, p, &flags);
 		goto recheck;
 	}
+
+	p->sched_reset_on_fork = reset_on_fork;
+
+	oldprio = p->prio;
+	if (oldprio == param->sched_priority)
+		goto out;
+
 	on_rq = p->on_rq;
 	running = task_current(rq, p);
 	if (on_rq)

@@ -4479,18 +4488,17 @@ recheck:
 	if (running)
 		p->sched_class->put_prev_task(rq, p);
 
-	p->sched_reset_on_fork = reset_on_fork;
-
-	oldprio = p->prio;
 	prev_class = p->sched_class;
 	__setscheduler(rq, p, policy, param->sched_priority);
 
 	if (running)
 		p->sched_class->set_curr_task(rq);
 	if (on_rq)
-		enqueue_task(rq, p, 0);
+		enqueue_task(rq, p, oldprio < param->sched_priority ?
+			     ENQUEUE_HEAD : 0);
 
 	check_class_changed(rq, p, prev_class, oldprio);
+out:
 	task_rq_unlock(rq, p, &flags);
 
 	rt_mutex_adjust_pi(p);

--
1.7.10.4

--
To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
--
To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Unfortunately there seems to be a bug in the patch, and it needs to be fixed in Thomas's tree before I take it.

	if (oldprio == (MAX_RT_PRIO - 1) - param->sched_priority)

From: Steven Rostedt <[email protected]>

---
 localversion-rt |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

index e1d8362..c2c1097 100644
@@ -1 +1 @@
--rt33
+-rt34-rc1

1.7.10.4

Sure enough I was taken on an unexpected journey today. Unfortunately it wasn't in the theaters.

I guess there's a small bug with one of the patches, with a fix sent to me by Mike. This needs to be in Thomas's latest before it goes into stable.

        oldprio = p->prio;
-       if (oldprio == param->sched_priority)
+       if (oldprio == (MAX_RT_PRIO - 1) - param->sched_priority)
                goto out;

-- Steve