1 files changed, 54 insertions, 25 deletions

diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index d1d460bb9b02..017044c26233 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -48,6 +48,7 @@
 #include <linux/moduleparam.h>
 #include <linux/uaccess.h>
 #include <linux/sched/isolation.h>
+#include <linux/nmi.h>
 
 #include "workqueue_internal.h"
 
@@ -2135,7 +2136,7 @@ __acquires(&pool->lock)
 	 * stop_machine. At the same time, report a quiescent RCU state so
 	 * the same condition doesn't freeze RCU.
 	 */
-	cond_resched_rcu_qs();
+	cond_resched();
 
 	spin_lock_irq(&pool->lock);
 
@@ -3940,6 +3941,37 @@ 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,
@@ -4002,29 +4034,8 @@ 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;
@@ -4464,6 +4475,12 @@ void show_workqueue_state(void)
 			if (pwq->nr_active || !list_empty(&pwq->delayed_works))
 				show_pwq(pwq);
 			spin_unlock_irqrestore(&pwq->pool->lock, flags);
+			/*
+			 * We could be printing a lot from atomic context, e.g.
+			 * sysrq-t -> show_workqueue_state(). Avoid triggering
+			 * hard lockup.
+			 */
+			touch_nmi_watchdog();
 		}
 	}
 
@@ -4491,6 +4508,12 @@ void show_workqueue_state(void)
 		pr_cont("\n");
 	next_pool:
 		spin_unlock_irqrestore(&pool->lock, flags);
+		/*
+		 * We could be printing a lot from atomic context, e.g.
+		 * sysrq-t -> show_workqueue_state(). Avoid triggering
+		 * hard lockup.
+		 */
+		touch_nmi_watchdog();
 	}
 
 	rcu_read_unlock_sched();
@@ -5630,6 +5653,8 @@ int __init workqueue_init(void)
 	 * archs such as power and arm64.  As per-cpu pools created
 	 * previously could be missing node hint and unbound pools NUMA
 	 * affinity, fix them up.
+	 *
+	 * Also, while iterating workqueues, create rescuers if requested.
 	 */
 	wq_numa_init();
 
@@ -5641,8 +5666,12 @@ int __init workqueue_init(void)
 		}
 	}
 
-	list_for_each_entry(wq, &workqueues, list)
+	list_for_each_entry(wq, &workqueues, list) {
 		wq_update_unbound_numa(wq, smp_processor_id(), true);
+		WARN(init_rescuer(wq),
+		     "workqueue: failed to create early rescuer for %s",
+		     wq->name);
+	}
 
 	mutex_unlock(&wq_pool_mutex);