workqueue: Convert to state machine callbacks

Get rid of the prio ordering of the separate notifiers and use a proper state
callback pair.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Anna-Maria Gleixner <anna-maria@linutronix.de>
Reviewed-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Acked-by: Tejun Heo <tj@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Lai Jiangshan <jiangshanlai@gmail.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Nicolas Iooss <nicolas.iooss_linux@m4x.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: rt@linutronix.de
Link: http://lkml.kernel.org/r/20160713153335.197083890@linutronix.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>

include/linux/cpu.h

@@ -55,15 +55,6 @@ extern ssize_t arch_cpu_release(const char *, size_t);
 #endif
 struct notifier_block;
 
-/*
- * CPU notifier priorities.
- */
-enum {
-	/* bring up workqueues before normal notifiers and down after */
-	CPU_PRI_WORKQUEUE_UP	= 5,
-	CPU_PRI_WORKQUEUE_DOWN	= -5,
-};
-
 #define CPU_ONLINE		0x0002 /* CPU (unsigned)v is up */
 #define CPU_UP_PREPARE		0x0003 /* CPU (unsigned)v coming up */
 #define CPU_UP_CANCELED		0x0004 /* CPU (unsigned)v NOT coming up */

include/linux/cpuhotplug.h

@@ -12,6 +12,7 @@ enum cpuhp_state {
 	CPUHP_PERF_BFIN,
 	CPUHP_PERF_POWER,
 	CPUHP_PERF_SUPERH,
+	CPUHP_WORKQUEUE_PREP,
 	CPUHP_NOTIFY_PREPARE,
 	CPUHP_BRINGUP_CPU,
 	CPUHP_AP_IDLE_DEAD,
@@ -49,6 +50,7 @@ enum cpuhp_state {
 	CPUHP_AP_PERF_S390_SF_ONLINE,
 	CPUHP_AP_PERF_ARM_CCI_ONLINE,
 	CPUHP_AP_PERF_ARM_CCN_ONLINE,
+	CPUHP_AP_WORKQUEUE_ONLINE,
 	CPUHP_AP_NOTIFY_ONLINE,
 	CPUHP_AP_ONLINE_DYN,
 	CPUHP_AP_ONLINE_DYN_END		= CPUHP_AP_ONLINE_DYN + 30,

include/linux/workqueue.h

@@ -625,4 +625,10 @@ void wq_watchdog_touch(int cpu);
 static inline void wq_watchdog_touch(int cpu) { }
 #endif	/* CONFIG_WQ_WATCHDOG */
 
+#ifdef CONFIG_SMP
+int workqueue_prepare_cpu(unsigned int cpu);
+int workqueue_online_cpu(unsigned int cpu);
+int workqueue_offline_cpu(unsigned int cpu);
+#endif
+
 #endif

kernel/cpu.c

@@ -1185,6 +1185,11 @@ static struct cpuhp_step cpuhp_bp_states[] = {
 		.startup = perf_event_init_cpu,
 		.teardown = perf_event_exit_cpu,
 	},
+	[CPUHP_WORKQUEUE_PREP] = {
+		.name = "workqueue prepare",
+		.startup = workqueue_prepare_cpu,
+		.teardown = NULL,
+	},
 	/*
 	 * Preparatory and dead notifiers. Will be replaced once the notifiers
 	 * are converted to states.
@@ -1267,6 +1272,11 @@ static struct cpuhp_step cpuhp_ap_states[] = {
 		.startup = perf_event_init_cpu,
 		.teardown = perf_event_exit_cpu,
 	},
+	[CPUHP_AP_WORKQUEUE_ONLINE] = {
+		.name = "workqueue online",
+		.startup = workqueue_online_cpu,
+		.teardown = workqueue_offline_cpu,
+	},
 
 	/*
 	 * Online/down_prepare notifiers. Will be removed once the notifiers

kernel/workqueue.c

@@ -4611,84 +4611,65 @@ static void restore_unbound_workers_cpumask(struct worker_pool *pool, int cpu)
 						  pool->attrs->cpumask) < 0);
 }
 
-/*
- * Workqueues should be brought up before normal priority CPU notifiers.
- * This will be registered high priority CPU notifier.
- */
-static int workqueue_cpu_up_callback(struct notifier_block *nfb,
-					       unsigned long action,
-					       void *hcpu)
+int workqueue_prepare_cpu(unsigned int cpu)
+{
+	struct worker_pool *pool;
+
+	for_each_cpu_worker_pool(pool, cpu) {
+		if (pool->nr_workers)
+			continue;
+		if (!create_worker(pool))
+			return -ENOMEM;
+	}
+	return 0;
+}
+
+int workqueue_online_cpu(unsigned int cpu)
 {
-	int cpu = (unsigned long)hcpu;
 	struct worker_pool *pool;
 	struct workqueue_struct *wq;
 	int pi;
 
-	switch (action & ~CPU_TASKS_FROZEN) {
-	case CPU_UP_PREPARE:
-		for_each_cpu_worker_pool(pool, cpu) {
-			if (pool->nr_workers)
-				continue;
-			if (!create_worker(pool))
-				return NOTIFY_BAD;
-		}
-		break;
-
-	case CPU_DOWN_FAILED:
-	case CPU_ONLINE:
-		mutex_lock(&wq_pool_mutex);
+	mutex_lock(&wq_pool_mutex);
 
-		for_each_pool(pool, pi) {
-			mutex_lock(&pool->attach_mutex);
+	for_each_pool(pool, pi) {
+		mutex_lock(&pool->attach_mutex);
 
-			if (pool->cpu == cpu)
-				rebind_workers(pool);
-			else if (pool->cpu < 0)
-				restore_unbound_workers_cpumask(pool, cpu);
+		if (pool->cpu == cpu)
+			rebind_workers(pool);
+		else if (pool->cpu < 0)
+			restore_unbound_workers_cpumask(pool, cpu);
 
-			mutex_unlock(&pool->attach_mutex);
-		}
+		mutex_unlock(&pool->attach_mutex);
+	}
 
-		/* update NUMA affinity of unbound workqueues */
-		list_for_each_entry(wq, &workqueues, list)
-			wq_update_unbound_numa(wq, cpu, true);
+	/* update NUMA affinity of unbound workqueues */
+	list_for_each_entry(wq, &workqueues, list)
+		wq_update_unbound_numa(wq, cpu, true);
 
-		mutex_unlock(&wq_pool_mutex);
-		break;
-	}
-	return NOTIFY_OK;
+	mutex_unlock(&wq_pool_mutex);
+	return 0;
 }
 
-/*
- * Workqueues should be brought down after normal priority CPU notifiers.
- * This will be registered as low priority CPU notifier.
- */
-static int workqueue_cpu_down_callback(struct notifier_block *nfb,
-						 unsigned long action,
-						 void *hcpu)
+int workqueue_offline_cpu(unsigned int cpu)
 {
-	int cpu = (unsigned long)hcpu;
 	struct work_struct unbind_work;
 	struct workqueue_struct *wq;
 
-	switch (action & ~CPU_TASKS_FROZEN) {
-	case CPU_DOWN_PREPARE:
-		/* 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);
-
-		/* update NUMA affinity of unbound workqueues */
-		mutex_lock(&wq_pool_mutex);
-		list_for_each_entry(wq, &workqueues, list)
-			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);
-		break;
-	}
-	return NOTIFY_OK;
+	/* 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);
+
+	/* update NUMA affinity of unbound workqueues */
+	mutex_lock(&wq_pool_mutex);
+	list_for_each_entry(wq, &workqueues, list)
+		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;
 }
 
 #ifdef CONFIG_SMP
@@ -5490,9 +5471,6 @@ static int __init init_workqueues(void)
 
 	pwq_cache = KMEM_CACHE(pool_workqueue, SLAB_PANIC);
 
-	cpu_notifier(workqueue_cpu_up_callback, CPU_PRI_WORKQUEUE_UP);
-	hotcpu_notifier(workqueue_cpu_down_callback, CPU_PRI_WORKQUEUE_DOWN);
-
 	wq_numa_init();
 
 	/* initialize CPU pools */