1 files changed, 66 insertions, 46 deletions

diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index 8fdb710bfdd7..8dd2e66e8383 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -38,7 +38,6 @@
 #include <linux/hardirq.h>
 #include <linux/mempolicy.h>
 #include <linux/freezer.h>
-#include <linux/kallsyms.h>
 #include <linux/debug_locks.h>
 #include <linux/lockdep.h>
 #include <linux/idr.h>
@@ -48,6 +47,8 @@
 #include <linux/nodemask.h>
 #include <linux/moduleparam.h>
 #include <linux/uaccess.h>
+#include <linux/sched/isolation.h>
+#include <linux/nmi.h>
 
 #include "workqueue_internal.h"
 
@@ -1634,7 +1635,7 @@ static void worker_enter_idle(struct worker *worker)
 		mod_timer(&pool->idle_timer, jiffies + IDLE_WORKER_TIMEOUT);
 
 	/*
-	 * Sanity check nr_running.  Because wq_unbind_fn() releases
+	 * Sanity check nr_running.  Because unbind_workers() releases
 	 * pool->lock between setting %WORKER_UNBOUND and zapping
 	 * nr_running, the warning may trigger spuriously.  Check iff
 	 * unbind is not in progress.
@@ -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);
 
@@ -3939,6 +3940,37 @@ 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,
@@ -4001,29 +4033,8 @@ 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;
@@ -4463,6 +4474,12 @@ 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();
 		}
 	}
 
@@ -4490,6 +4507,12 @@ 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();
@@ -4510,9 +4533,8 @@ void show_workqueue_state(void)
  * cpu comes back online.
  */
 
-static void wq_unbind_fn(struct work_struct *work)
+static void unbind_workers(int cpu)
 {
-	int cpu = smp_processor_id();
 	struct worker_pool *pool;
 	struct worker *worker;
 
@@ -4589,16 +4611,6 @@ static void rebind_workers(struct worker_pool *pool)
 
 	spin_lock_irq(&pool->lock);
 
-	/*
-	 * XXX: CPU hotplug notifiers are weird and can call DOWN_FAILED
-	 * w/o preceding DOWN_PREPARE.  Work around it.  CPU hotplug is
-	 * being reworked and this can go away in time.
-	 */
-	if (!(pool->flags & POOL_DISASSOCIATED)) {
-		spin_unlock_irq(&pool->lock);
-		return;
-	}
-
 	pool->flags &= ~POOL_DISASSOCIATED;
 
 	for_each_pool_worker(worker, pool) {
@@ -4709,12 +4721,13 @@ int workqueue_online_cpu(unsigned int cpu)
 
 int workqueue_offline_cpu(unsigned int cpu)
 {
-	struct work_struct unbind_work;
 	struct workqueue_struct *wq;
 
 	/* unbinding per-cpu workers should happen on the local CPU */
-	INIT_WORK_ONSTACK(&unbind_work, wq_unbind_fn);
-	queue_work_on(cpu, system_highpri_wq, &unbind_work);
+	if (WARN_ON(cpu != smp_processor_id()))
+		return -1;
+
+	unbind_workers(cpu);
 
 	/* update NUMA affinity of unbound workqueues */
 	mutex_lock(&wq_pool_mutex);
@@ -4722,9 +4735,6 @@ int workqueue_offline_cpu(unsigned int cpu)
 		wq_update_unbound_numa(wq, cpu, false);
 	mutex_unlock(&wq_pool_mutex);
 
-	/* wait for per-cpu unbinding to finish */
-	flush_work(&unbind_work);
-	destroy_work_on_stack(&unbind_work);
 	return 0;
 }
 
@@ -4957,6 +4967,10 @@ int workqueue_set_unbound_cpumask(cpumask_var_t cpumask)
 	if (!zalloc_cpumask_var(&saved_cpumask, GFP_KERNEL))
 		return -ENOMEM;
 
+	/*
+	 * Not excluding isolated cpus on purpose.
+	 * If the user wishes to include them, we allow that.
+	 */
 	cpumask_and(cpumask, cpumask, cpu_possible_mask);
 	if (!cpumask_empty(cpumask)) {
 		apply_wqattrs_lock();
@@ -5555,7 +5569,7 @@ int __init workqueue_init_early(void)
 	WARN_ON(__alignof__(struct pool_workqueue) < __alignof__(long long));
 
 	BUG_ON(!alloc_cpumask_var(&wq_unbound_cpumask, GFP_KERNEL));
-	cpumask_copy(wq_unbound_cpumask, cpu_possible_mask);
+	cpumask_copy(wq_unbound_cpumask, housekeeping_cpumask(HK_FLAG_DOMAIN));
 
 	pwq_cache = KMEM_CACHE(pool_workqueue, SLAB_PANIC);
 
@@ -5638,6 +5652,8 @@ 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();
 
@@ -5649,8 +5665,12 @@ 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);