sched: Distangle worker accounting from rqlock

The worker accounting for cpu bound workers is plugged into the core
scheduler code and the wakeup code. This is not a hard requirement and
can be avoided by keeping track of the state in the workqueue code
itself.
    
Keep track of the sleeping state in the worker itself and call the
notifier before entering the core scheduler. There might be false
positives when the task is woken between that call and actually
scheduling, but that's not really different from scheduling and being
woken immediately after switching away. There is also no harm from
updating nr_running when the task returns from scheduling instead of
accounting it in the wakeup code.
    
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: http://lkml.kernel.org/r/20110622174919.135236139@linutronix.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

3 files changed, 41 insertions, 99 deletions

diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 878db4d97d1d91..21cba30df6f693 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -1735,10 +1735,6 @@ static inline void ttwu_activate(struct rq *rq, struct task_struct *p, int en_fl
 {
 	activate_task(rq, p, en_flags);
 	p->on_rq = TASK_ON_RQ_QUEUED;
-
-	/* if a worker is waking up, notify workqueue */
-	if (p->flags & PF_WQ_WORKER)
-		wq_worker_waking_up(p, cpu_of(rq));
 }
 
 /*
@@ -2055,52 +2051,6 @@ out:
 }
 
 /**
- * try_to_wake_up_local - try to wake up a local task with rq lock held
- * @p: the thread to be awakened
- *
- * Put @p on the run-queue if it's not already there. The caller must
- * ensure that this_rq() is locked, @p is bound to this_rq() and not
- * the current task.
- */
-static void try_to_wake_up_local(struct task_struct *p)
-{
-	struct rq *rq = task_rq(p);
-
-	if (WARN_ON_ONCE(rq != this_rq()) ||
-	    WARN_ON_ONCE(p == current))
-		return;
-
-	lockdep_assert_held(&rq->lock);
-
-	if (!raw_spin_trylock(&p->pi_lock)) {
-		/*
-		 * This is OK, because current is on_cpu, which avoids it being
-		 * picked for load-balance and preemption/IRQs are still
-		 * disabled avoiding further scheduler activity on it and we've
-		 * not yet picked a replacement task.
-		 */
-		lockdep_unpin_lock(&rq->lock);
-		raw_spin_unlock(&rq->lock);
-		raw_spin_lock(&p->pi_lock);
-		raw_spin_lock(&rq->lock);
-		lockdep_pin_lock(&rq->lock);
-	}
-
-	if (!(p->state & TASK_NORMAL))
-		goto out;
-
-	trace_sched_waking(p);
-
-	if (!task_on_rq_queued(p))
-		ttwu_activate(rq, p, ENQUEUE_WAKEUP);
-
-	ttwu_do_wakeup(rq, p, 0);
-	ttwu_stat(p, smp_processor_id(), 0);
-out:
-	raw_spin_unlock(&p->pi_lock);
-}
-
-/**
  * wake_up_process - Wake up a specific process
  * @p: The process to be woken up.
  *
@@ -3281,21 +3231,6 @@ static void __sched notrace __schedule(bool preempt)
 		} else {
 			deactivate_task(rq, prev, DEQUEUE_SLEEP);
 			prev->on_rq = 0;
-
-			/*
-			 * If a worker went to sleep, notify and ask workqueue
-			 * whether it wants to wake up a task to maintain
-			 * concurrency.
-			 * Only call wake up if prev isn't blocked on a sleeping
-			 * spin lock.
-			 */
-			if (prev->flags & PF_WQ_WORKER && !prev->saved_state) {
-				struct task_struct *to_wakeup;
-
-				to_wakeup = wq_worker_sleeping(prev, cpu);
-				if (to_wakeup)
-					try_to_wake_up_local(to_wakeup);
-			}
 		}
 		switch_count = &prev->nvcsw;
 	}
@@ -3328,6 +3263,14 @@ static inline void sched_submit_work(struct task_struct *tsk)
 {
 	if (!tsk->state || tsk_is_pi_blocked(tsk))
 		return;
+
+	/*
+	 * If a worker went to sleep, notify and ask workqueue whether
+	 * it wants to wake up a task to maintain concurrency.
+	 */
+	if (tsk->flags & PF_WQ_WORKER)
+		wq_worker_sleeping(tsk);
+
 	/*
 	 * If we are going to sleep and we have plugged IO queued,
 	 * make sure to submit it to avoid deadlocks.
@@ -3336,6 +3279,12 @@ static inline void sched_submit_work(struct task_struct *tsk)
 		blk_schedule_flush_plug(tsk);
 }
 
+static void sched_update_worker(struct task_struct *tsk)
+{
+	if (tsk->flags & PF_WQ_WORKER)
+		wq_worker_running(tsk);
+}
+
 asmlinkage __visible void __sched schedule(void)
 {
 	struct task_struct *tsk = current;
@@ -3346,6 +3295,7 @@ asmlinkage __visible void __sched schedule(void)
 		__schedule(false);
 		sched_preempt_enable_no_resched();
 	} while (need_resched());
+	sched_update_worker(tsk);
 }
 EXPORT_SYMBOL(schedule);
 
diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index 31a2cea908e879..1ffc1696c19846 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -811,44 +811,31 @@ static void wake_up_worker(struct worker_pool *pool)
 }
 
 /**
- * wq_worker_waking_up - a worker is waking up
- * @task: task waking up
- * @cpu: CPU @task is waking up to
+ * wq_worker_running - a worker is running again
+ * @task: task returning from sleep
  *
- * This function is called during try_to_wake_up() when a worker is
- * being awoken.
- *
- * CONTEXT:
- * spin_lock_irq(rq->lock)
+ * This function is called when a worker returns from schedule()
  */
-void wq_worker_waking_up(struct task_struct *task, int cpu)
+void wq_worker_running(struct task_struct *task)
 {
 	struct worker *worker = kthread_data(task);
 
-	if (!(worker->flags & WORKER_NOT_RUNNING)) {
-		WARN_ON_ONCE(worker->pool->cpu != cpu);
+	if (!worker->sleeping)
+		return;
+	if (!(worker->flags & WORKER_NOT_RUNNING))
 		atomic_inc(&worker->pool->nr_running);
-	}
+	worker->sleeping = 0;
 }
 
 /**
  * wq_worker_sleeping - a worker is going to sleep
  * @task: task going to sleep
- * @cpu: CPU in question, must be the current CPU number
- *
- * This function is called during schedule() when a busy worker is
- * going to sleep.  Worker on the same cpu can be woken up by
- * returning pointer to its task.
- *
- * CONTEXT:
- * spin_lock_irq(rq->lock)
- *
- * Return:
- * Worker task on @cpu to wake up, %NULL if none.
+ * This function is called from schedule() when a busy worker is
+ * going to sleep.
  */
-struct task_struct *wq_worker_sleeping(struct task_struct *task, int cpu)
+void wq_worker_sleeping(struct task_struct *task)
 {
-	struct worker *worker = kthread_data(task), *to_wakeup = NULL;
+	struct worker *next, *worker = kthread_data(task);
 	struct worker_pool *pool;
 
 	/*
@@ -857,14 +844,15 @@ struct task_struct *wq_worker_sleeping(struct task_struct *task, int cpu)
 	 * checking NOT_RUNNING.
 	 */
 	if (worker->flags & WORKER_NOT_RUNNING)
-		return NULL;
+		return;
 
 	pool = worker->pool;
 
-	/* this can only happen on the local cpu */
-	if (WARN_ON_ONCE(cpu != raw_smp_processor_id() || pool->cpu != cpu))
-		return NULL;
+	if (WARN_ON_ONCE(worker->sleeping))
+		return;
 
+	worker->sleeping = 1;
+	spin_lock_irq(&pool->lock);
 	/*
 	 * The counterpart of the following dec_and_test, implied mb,
 	 * worklist not empty test sequence is in insert_work().
@@ -877,9 +865,12 @@ struct task_struct *wq_worker_sleeping(struct task_struct *task, int cpu)
 	 * lock is safe.
 	 */
 	if (atomic_dec_and_test(&pool->nr_running) &&
-	    !list_empty(&pool->worklist))
-		to_wakeup = first_idle_worker(pool);
-	return to_wakeup ? to_wakeup->task : NULL;
+	    !list_empty(&pool->worklist)) {
+		next = first_idle_worker(pool);
+		if (next)
+			wake_up_process(next->task);
+	}
+	spin_unlock_irq(&pool->lock);
 }
 
 /**
diff --git a/kernel/workqueue_internal.h b/kernel/workqueue_internal.h
index 45215870ac6cc2..f000c4d6917e92 100644
--- a/kernel/workqueue_internal.h
+++ b/kernel/workqueue_internal.h
@@ -43,6 +43,7 @@ struct worker {
 	unsigned long		last_active;	/* L: last active timestamp */
 	unsigned int		flags;		/* X: flags */
 	int			id;		/* I: worker id */
+	int			sleeping;	/* None */
 
 	/*
 	 * Opaque string set with work_set_desc().  Printed out with task
@@ -68,7 +69,7 @@ static inline struct worker *current_wq_worker(void)
  * Scheduler hooks for concurrency managed workqueue.  Only to be used from
  * sched/core.c and workqueue.c.
  */
-void wq_worker_waking_up(struct task_struct *task, int cpu);
-struct task_struct *wq_worker_sleeping(struct task_struct *task, int cpu);
+void wq_worker_running(struct task_struct *task);
+void wq_worker_sleeping(struct task_struct *task);
 
 #endif /* _KERNEL_WORKQUEUE_INTERNAL_H */