[PATCH v5 4/5] sched/core: Prevent race condition between cpuset and __sched_setscheduler()

[Juri Lelli]

From: Mathieu Poirier <mathieu.poirier@xxxxxxxxxx>

No synchronisation mechanism exists between the cpuset subsystem and calls
to function __sched_setscheduler(). As such, it is possible that new root
domains are created on the cpuset side while a deadline acceptance test
is carried out in __sched_setscheduler(), leading to a potential oversell
of CPU bandwidth.

Grab callback_lock from core scheduler, so to prevent situations such as
the one described above from happening.

Signed-off-by: Mathieu Poirier <mathieu.poirier@xxxxxxxxxx>
Signed-off-by: Juri Lelli <juri.lelli@xxxxxxxxxx>

---

v4->v5: grab callback_lock instead of cpuset_mutex, as callback_lock is
enough to get read-only access to cpusets [1] and it can be easily
converted to be a raw_spinlock (done in previous - new - patch).

[1] https://elixir.bootlin.com/linux/latest/source/kernel/cgroup/cpuset.c#L275
---
 include/linux/cpuset.h |  6 ++++++
 kernel/cgroup/cpuset.c | 18 ++++++++++++++++++
 kernel/sched/core.c    | 10 ++++++++++
 3 files changed, 34 insertions(+)

diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h
index 934633a05d20..8e5a8dd0622b 100644
--- a/include/linux/cpuset.h
+++ b/include/linux/cpuset.h
@@ -55,6 +55,8 @@ extern void cpuset_init_smp(void);
 extern void cpuset_force_rebuild(void);
 extern void cpuset_update_active_cpus(void);
 extern void cpuset_wait_for_hotplug(void);
+extern void cpuset_read_only_lock(void);
+extern void cpuset_read_only_unlock(void);
 extern void cpuset_cpus_allowed(struct task_struct *p, struct cpumask *mask);
 extern void cpuset_cpus_allowed_fallback(struct task_struct *p);
 extern nodemask_t cpuset_mems_allowed(struct task_struct *p);
@@ -176,6 +178,10 @@ static inline void cpuset_update_active_cpus(void)
 
 static inline void cpuset_wait_for_hotplug(void) { }
 
+static inline void cpuset_read_only_lock(void) { }
+
+static inline void cpuset_read_only_unlock(void) { }
+
 static inline void cpuset_cpus_allowed(struct task_struct *p,
 				       struct cpumask *mask)
 {
diff --git a/kernel/cgroup/cpuset.c b/kernel/cgroup/cpuset.c
index 5b43f482fa0f..8dc26005bb1e 100644
--- a/kernel/cgroup/cpuset.c
+++ b/kernel/cgroup/cpuset.c
@@ -2410,6 +2410,24 @@ void __init cpuset_init_smp(void)
 	BUG_ON(!cpuset_migrate_mm_wq);
 }
 
+/**
+ * cpuset_read_only_lock - Grab the callback_lock from another subsysytem
+ *
+ * Description: Gives the holder read-only access to cpusets.
+ */
+void cpuset_read_only_lock(void)
+{
+	raw_spin_lock(&callback_lock);
+}
+
+/**
+ * cpuset_read_only_unlock - Release the callback_lock from another subsysytem
+ */
+void cpuset_read_only_unlock(void)
+{
+	raw_spin_unlock(&callback_lock);
+}
+
 /**
  * cpuset_cpus_allowed - return cpus_allowed mask from a tasks cpuset.
  * @tsk: pointer to task_struct from which to obtain cpuset->cpus_allowed.
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 22f5622cba69..ac11ee599968 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -4228,6 +4228,13 @@ static int __sched_setscheduler(struct task_struct *p,
 	rq = task_rq_lock(p, &rf);
 	update_rq_clock(rq);
 
+	/*
+	 * Make sure we don't race with the cpuset subsystem where root
+	 * domains can be rebuilt or modified while operations like DL
+	 * admission checks are carried out.
+	 */
+	cpuset_read_only_lock();
+
 	/*
 	 * Changing the policy of the stop threads its a very bad idea:
 	 */
@@ -4289,6 +4296,7 @@ static int __sched_setscheduler(struct task_struct *p,
 	/* Re-check policy now with rq lock held: */
 	if (unlikely(oldpolicy != -1 && oldpolicy != p->policy)) {
 		policy = oldpolicy = -1;
+		cpuset_read_only_unlock();
 		task_rq_unlock(rq, p, &rf);
 		goto recheck;
 	}
@@ -4346,6 +4354,7 @@ static int __sched_setscheduler(struct task_struct *p,
 
 	/* Avoid rq from going away on us: */
 	preempt_disable();
+	cpuset_read_only_unlock();
 	task_rq_unlock(rq, p, &rf);
 
 	if (pi)
@@ -4358,6 +4367,7 @@ static int __sched_setscheduler(struct task_struct *p,
 	return 0;
 
 unlock:
+	cpuset_read_only_unlock();
 	task_rq_unlock(rq, p, &rf);
 	return retval;
 }
-- 
2.17.1