rcu: Create RCU-specific workqueues with rescuers

RCU's expedited grace periods can participate in out-of-memory deadlocks
due to all available system_wq kthreads being blocked and there not being
memory available to create more.  This commit prevents such deadlocks
by allocating an RCU-specific workqueue_struct at early boot time, and
providing it with a rescuer to ensure forward progress.  This uses the
shiny new init_rescuer() function provided by Tejun (but indirectly).

This commit also causes SRCU to use this new RCU-specific
workqueue_struct.  Note that SRCU's use of workqueues never blocks them
waiting for readers, so this should be safe from a forward-progress
viewpoint.  Note that this moves SRCU from system_power_efficient_wq
to a normal workqueue.  In the unlikely event that this results in
measurable degradation, a separate power-efficient workqueue will be
creates for SRCU.

Change-Id: I2988819b553b769ccfdeabd62394c3aad63d6668
Reported-by: Prateek Sood <prsood@codeaurora.org>
Reported-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Acked-by: Tejun Heo <tj@kernel.org>
Git-commit: ad7c946b35ad455417fdd4bc0e17deda4011841b
Git-Repo: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
Signed-off-by: Prateek Sood <prsood@codeaurora.org>
Signed-off-by: mydongistiny <jaysonedson@gmail.com>
Signed-off-by: DennySPB <dennyspb@gmail.com>
Signed-off-by: Twisted <36546624+TwistedPrime@users.noreply.github.com>
Signed-off-by: Sorayukii <sorayukii69@gmail.com>

kernel/rcu/rcu.h

@@ -562,6 +562,7 @@ void show_rcu_gp_kthreads(void);
 void rcu_force_quiescent_state(void);
 void rcu_bh_force_quiescent_state(void);
 void rcu_sched_force_quiescent_state(void);
+extern struct workqueue_struct *rcu_gp_wq;
 #endif /* #else #ifdef CONFIG_TINY_RCU */
 
 #ifdef CONFIG_RCU_NOCB_CPU

kernel/rcu/srcutree.c

@@ -465,8 +465,7 @@ static bool srcu_queue_delayed_work_on(int cpu, struct workqueue_struct *wq,
  */
 static void srcu_schedule_cbs_sdp(struct srcu_data *sdp, unsigned long delay)
 {
-	srcu_queue_delayed_work_on(sdp->cpu, system_power_efficient_wq,
-				   &sdp->work, delay);
+	srcu_queue_delayed_work_on(sdp->cpu, rcu_gp_wq, &sdp->work, delay);
 }
 
 /*
@@ -664,8 +663,7 @@ static void srcu_funnel_gp_start(struct srcu_struct *sp, struct srcu_data *sdp,
 	    rcu_seq_state(sp->srcu_gp_seq) == SRCU_STATE_IDLE) {
 		WARN_ON_ONCE(ULONG_CMP_GE(sp->srcu_gp_seq, sp->srcu_gp_seq_needed));
 		srcu_gp_start(sp);
-		queue_delayed_work(system_power_efficient_wq, &sp->work,
-				   srcu_get_delay(sp));
+		queue_delayed_work(rcu_gp_wq, &sp->work, srcu_get_delay(sp));
 	}
 	raw_spin_unlock_irqrestore_rcu_node(sp, flags);
 }
@@ -1189,7 +1187,7 @@ static void srcu_reschedule(struct srcu_struct *sp, unsigned long delay)
 	raw_spin_unlock_irq_rcu_node(sp);
 
 	if (pushgp)
-		queue_delayed_work(system_power_efficient_wq, &sp->work, delay);
+		queue_delayed_work(rcu_gp_wq, &sp->work, delay);
 }
 
 /*

kernel/rcu/tree.c

@@ -4197,6 +4197,8 @@ static void __init rcu_dump_rcu_node_tree(struct rcu_state *rsp)
 	pr_cont("\n");
 }
 
+struct workqueue_struct *rcu_gp_wq;
+
 void __init rcu_init(void)
 {
 	int cpu;
@@ -4224,6 +4226,10 @@ void __init rcu_init(void)
 		if (IS_ENABLED(CONFIG_TREE_SRCU))
 			srcu_online_cpu(cpu);
 	}
+
+	/* Create workqueue for expedited GPs and for Tree SRCU. */
+	rcu_gp_wq = alloc_workqueue("rcu_gp", WQ_MEM_RECLAIM, 0);
+	WARN_ON(!rcu_gp_wq);
 }
 
 #include "tree_exp.h"

kernel/rcu/tree_exp.h

@@ -606,7 +606,7 @@ static void _synchronize_rcu_expedited(struct rcu_state *rsp,
 		rew.rew_rsp = rsp;
 		rew.rew_s = s;
 		INIT_WORK_ONSTACK(&rew.rew_work, wait_rcu_exp_gp);
-		schedule_work(&rew.rew_work);
+		queue_work(rcu_gp_wq, &rew.rew_work);
 	}
 
 	/* Wait for expedited grace period to complete. */

kernel/workqueue.c

@@ -3971,6 +3971,37 @@ static int wq_clamp_max_active(int max_active, unsigned int flags,
 	return clamp_val(max_active, 1, lim);
 }
 
+/*
+ * Workqueues which may be used during memory reclaim should have a rescuer
+ * to guarantee forward progress.
+ */
+static int init_rescuer(struct workqueue_struct *wq)
+{
+	struct worker *rescuer;
+	int ret;
+
+	if (!(wq->flags & WQ_MEM_RECLAIM))
+		return 0;
+
+	rescuer = alloc_worker(NUMA_NO_NODE);
+	if (!rescuer)
+		return -ENOMEM;
+
+	rescuer->rescue_wq = wq;
+	rescuer->task = kthread_create(rescuer_thread, rescuer, "%s", wq->name);
+	ret = PTR_ERR_OR_ZERO(rescuer->task);
+	if (ret) {
+		kfree(rescuer);
+		return ret;
+	}
+
+	wq->rescuer = rescuer;
+	kthread_bind_mask(rescuer->task, cpu_possible_mask);
+	wake_up_process(rescuer->task);
+
+	return 0;
+}
+
 struct workqueue_struct *__alloc_workqueue_key(const char *fmt,
 					       unsigned int flags,
 					       int max_active,
@@ -4033,29 +4064,8 @@ struct workqueue_struct *__alloc_workqueue_key(const char *fmt,
 	if (alloc_and_link_pwqs(wq) < 0)
 		goto err_free_wq;
 
-	/*
-	 * Workqueues which may be used during memory reclaim should
-	 * have a rescuer to guarantee forward progress.
-	 */
-	if (flags & WQ_MEM_RECLAIM) {
-		struct worker *rescuer;
-
-		rescuer = alloc_worker(NUMA_NO_NODE);
-		if (!rescuer)
-			goto err_destroy;
-
-		rescuer->rescue_wq = wq;
-		rescuer->task = kthread_create(rescuer_thread, rescuer, "%s",
-					       wq->name);
-		if (IS_ERR(rescuer->task)) {
-			kfree(rescuer);
-			goto err_destroy;
-		}
-
-		wq->rescuer = rescuer;
-		kthread_bind_mask(rescuer->task, cpu_possible_mask);
-		wake_up_process(rescuer->task);
-	}
+	if (wq_online && init_rescuer(wq) < 0)
+		goto err_destroy;
 
 	if ((wq->flags & WQ_SYSFS) && workqueue_sysfs_register(wq))
 		goto err_destroy;