On Mon, May 11, 2020 at 12:58:00PM +0200, Peter Zijlstra wrote:
> On Wed, Apr 29, 2020 at 08:31:39PM -0500, Scott Wood wrote:
> > > If you were to do a queue_balance_callback() from somewhere in the
> > > pick_next_task() machinery, then the balance_callback() at the end of
> > > __schedule() would run it, and it'd be gone. How would
> > > rt_mutex_setprio() / __sched_setscheduler() be affected?
> >
> > The rq lock is dropped between queue_balance_callback() and the
> > balance_callback() at the end of __schedule(). What stops
> > setprio/setscheduler on another cpu from doing the callback at that
> > point?
>
> Hurmm.. fair point, and that might explain some issues I had a while
> back. Let me poke a little at that.
How's this?
---
kernel/sched/core.c | 109 ++++++++++++++++++++++++++++++---------------------
kernel/sched/sched.h | 2 +
2 files changed, 67 insertions(+), 44 deletions(-)
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index dfb8ab61cbdd..610e9da557ed 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -3078,6 +3078,59 @@ static inline void finish_task(struct task_struct *prev)
#endif
}
+#ifdef CONFIG_SMP
+
+/* rq->lock is NOT held, but preemption is disabled */
+static void do_balance_callbacks(struct callback_head *head)
+{
+ void (*func)(struct rq *rq);
+ struct callback_head *next;
+
+ while (head) {
+ func = (void (*)(struct rq *))head->func;
+ next = head->next;
+ head->next = NULL;
+ head = next;
+
+ func(rq);
+ }
+}
+
+static inline struct callback_head *splice_balance_callbacks(struct rq *rq)
+{
+ struct callback_head *head = rq->balance_callback;
+ if (head)
+ rq->balance_callback = NULL;
+}
+
+static void __balance_callbacks(struct rq *rq)
+{
+ do_balance_callbacks(splice_balance_callbacks(rq));
+}
+
+static inline void balance_callbacks(struct rq *rq, struct callback_head *head)
+{
+ unsigned long flags;
+
+ if (unlikely(head)) {
+ raw_spin_lock_irqsave(&rq->lock, flags);
+ do_balance_callbacks(head);
+ raw_spin_unlock_irqrestore(&rq->lock, flags);
+ }
+}
+
+#else
+
+static inline void __balance_callbacks(struct rq *rq)
+{
+}
+
+static inline void balance_callbacks(struct rq *rq, struct callback_head *head)
+{
+}
+
+#endif
+
static inline void
prepare_lock_switch(struct rq *rq, struct task_struct *next, struct rq_flags *rf)
{
@@ -3103,6 +3156,7 @@ static inline void finish_lock_switch(struct rq *rq)
* prev into current:
*/
spin_acquire(&rq->lock.dep_map, 0, 0, _THIS_IP_);
+ __balance_callbacks(rq));
raw_spin_unlock_irq(&rq->lock);
}
@@ -3244,43 +3298,6 @@ static struct rq *finish_task_switch(struct task_struct *prev)
return rq;
}
-#ifdef CONFIG_SMP
-
-/* rq->lock is NOT held, but preemption is disabled */
-static void __balance_callback(struct rq *rq)
-{
- struct callback_head *head, *next;
- void (*func)(struct rq *rq);
- unsigned long flags;
-
- raw_spin_lock_irqsave(&rq->lock, flags);
- head = rq->balance_callback;
- rq->balance_callback = NULL;
- while (head) {
- func = (void (*)(struct rq *))head->func;
- next = head->next;
- head->next = NULL;
- head = next;
-
- func(rq);
- }
- raw_spin_unlock_irqrestore(&rq->lock, flags);
-}
-
-static inline void balance_callback(struct rq *rq)
-{
- if (unlikely(rq->balance_callback))
- __balance_callback(rq);
-}
-
-#else
-
-static inline void balance_callback(struct rq *rq)
-{
-}
-
-#endif
-
/**
* schedule_tail - first thing a freshly forked thread must call.
* @prev: the thread we just switched away from.
@@ -3300,7 +3317,6 @@ asmlinkage __visible void schedule_tail(struct task_struct *prev)
*/
rq = finish_task_switch(prev);
- balance_callback(rq);
preempt_enable();
if (current->set_child_tid)
@@ -4090,10 +4106,11 @@ static void __sched notrace __schedule(bool preempt)
rq = context_switch(rq, prev, next, &rf);
} else {
rq->clock_update_flags &= ~(RQCF_ACT_SKIP|RQCF_REQ_SKIP);
- rq_unlock_irq(rq, &rf);
- }
- balance_callback(rq);
+ rq_unpin_lock(rq, &rf);
+ __balance_callbacks(rq);
+ raw_spin_unlock_irq(&rq->lock);
+ }
}
void __noreturn do_task_dead(void)
@@ -4499,9 +4516,11 @@ void rt_mutex_setprio(struct task_struct *p, struct task_struct *pi_task)
out_unlock:
/* Avoid rq from going away on us: */
preempt_disable();
- __task_rq_unlock(rq, &rf);
- balance_callback(rq);
+ rq_unpin_lock(rq, &rf);
+ __balance_callbacks(rq);
+ raw_spin_unlock(&rq->lock);
+
preempt_enable();
}
#else
@@ -4775,6 +4794,7 @@ static int __sched_setscheduler(struct task_struct *p,
int retval, oldprio, oldpolicy = -1, queued, running;
int new_effective_prio, policy = attr->sched_policy;
const struct sched_class *prev_class;
+ struct callback_head *head;
struct rq_flags rf;
int reset_on_fork;
int queue_flags = DEQUEUE_SAVE | DEQUEUE_MOVE | DEQUEUE_NOCLOCK;
@@ -5013,6 +5033,7 @@ static int __sched_setscheduler(struct task_struct *p,
/* Avoid rq from going away on us: */
preempt_disable();
+ head = splice_balance_callbacks(rq);
task_rq_unlock(rq, p, &rf);
if (pi) {
@@ -5021,7 +5042,7 @@ static int __sched_setscheduler(struct task_struct *p,
}
/* Run balance callbacks after we've adjusted the PI chain: */
- balance_callback(rq);
+ balance_callbacks(head);
preempt_enable();
return 0;
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index d7fc4caf0dfd..3855d354760d 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -1189,6 +1189,8 @@ static inline void rq_pin_lock(struct rq *rq, struct rq_flags *rf)
#ifdef CONFIG_SCHED_DEBUG
rq->clock_update_flags &= (RQCF_REQ_SKIP|RQCF_ACT_SKIP);
rf->clock_update_flags = 0;
+
+ SCHED_WARN_ON(rq->balance_callback);
#endif
}
