Merge branch 'smp-hotplug-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull smp hotplug updates from Thomas Gleixner:
 "This is the final round of converting the notifier mess to the state
  machine. The removal of the notifiers and the related infrastructure
  will happen around rc1, as there are conversions outstanding in other
  trees.

  The whole exercise removed about 2000 lines of code in total and in
  course of the conversion several dozen bugs got fixed. The new
  mechanism allows to test almost every hotplug step standalone, so
  usage sites can exercise all transitions extensively.

  There is more room for improvement, like integrating all the
  pointlessly different architecture mechanisms of synchronizing,
  setting cpus online etc into the core code"

* 'smp-hotplug-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (60 commits)
  tracing/rb: Init the CPU mask on allocation
  soc/fsl/qbman: Convert to hotplug state machine
  soc/fsl/qbman: Convert to hotplug state machine
  zram: Convert to hotplug state machine
  KVM/PPC/Book3S HV: Convert to hotplug state machine
  arm64/cpuinfo: Convert to hotplug state machine
  arm64/cpuinfo: Make hotplug notifier symmetric
  mm/compaction: Convert to hotplug state machine
  iommu/vt-d: Convert to hotplug state machine
  mm/zswap: Convert pool to hotplug state machine
  mm/zswap: Convert dst-mem to hotplug state machine
  mm/zsmalloc: Convert to hotplug state machine
  mm/vmstat: Convert to hotplug state machine
  mm/vmstat: Avoid on each online CPU loops
  mm/vmstat: Drop get_online_cpus() from init_cpu_node_state/vmstat_cpu_dead()
  tracing/rb: Convert to hotplug state machine
  oprofile/nmi timer: Convert to hotplug state machine
  net/iucv: Use explicit clean up labels in iucv_init()
  x86/pci/amd-bus: Convert to hotplug state machine
  x86/oprofile/nmi: Convert to hotplug state machine
  ...

4 files changed, 86 insertions, 124 deletions

diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c
index f7a55e9ff2f7..577f2288d19f 100644
--- a/kernel/printk/printk.c
+++ b/kernel/printk/printk.c
@@ -2166,27 +2166,20 @@ void resume_console(void)
 
 /**
  * console_cpu_notify - print deferred console messages after CPU hotplug
- * @self: notifier struct
- * @action: CPU hotplug event
- * @hcpu: unused
+ * @cpu: unused
  *
  * If printk() is called from a CPU that is not online yet, the messages
  * will be spooled but will not show up on the console.  This function is
  * called when a new CPU comes online (or fails to come up), and ensures
  * that any such output gets printed.
  */
-static int console_cpu_notify(struct notifier_block *self,
-	unsigned long action, void *hcpu)
-{
-	switch (action) {
-	case CPU_ONLINE:
-	case CPU_DEAD:
-	case CPU_DOWN_FAILED:
-	case CPU_UP_CANCELED:
+static int console_cpu_notify(unsigned int cpu)
+{
+	if (!cpuhp_tasks_frozen) {
 		console_lock();
 		console_unlock();
 	}
-	return NOTIFY_OK;
+	return 0;
 }
 
 /**
@@ -2824,6 +2817,7 @@ EXPORT_SYMBOL(unregister_console);
 static int __init printk_late_init(void)
 {
 	struct console *con;
+	int ret;
 
 	for_each_console(con) {
 		if (!keep_bootcon && con->flags & CON_BOOT) {
@@ -2838,7 +2832,12 @@ static int __init printk_late_init(void)
 				unregister_console(con);
 		}
 	}
-	hotcpu_notifier(console_cpu_notify, 0);
+	ret = cpuhp_setup_state_nocalls(CPUHP_PRINTK_DEAD, "printk:dead", NULL,
+					console_cpu_notify);
+	WARN_ON(ret < 0);
+	ret = cpuhp_setup_state_nocalls(CPUHP_AP_ONLINE_DYN, "printk:online",
+					console_cpu_notify, NULL);
+	WARN_ON(ret < 0);
 	return 0;
 }
 late_initcall(printk_late_init);
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index 3bcb61b52f6c..71496a20e670 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -390,24 +390,16 @@ static int __init tick_nohz_full_setup(char *str)
 }
 __setup("nohz_full=", tick_nohz_full_setup);
 
-static int tick_nohz_cpu_down_callback(struct notifier_block *nfb,
-				       unsigned long action,
-				       void *hcpu)
+static int tick_nohz_cpu_down(unsigned int cpu)
 {
-	unsigned int cpu = (unsigned long)hcpu;
-
-	switch (action & ~CPU_TASKS_FROZEN) {
-	case CPU_DOWN_PREPARE:
-		/*
-		 * The boot CPU handles housekeeping duty (unbound timers,
-		 * workqueues, timekeeping, ...) on behalf of full dynticks
-		 * CPUs. It must remain online when nohz full is enabled.
-		 */
-		if (tick_nohz_full_running && tick_do_timer_cpu == cpu)
-			return NOTIFY_BAD;
-		break;
-	}
-	return NOTIFY_OK;
+	/*
+	 * The boot CPU handles housekeeping duty (unbound timers,
+	 * workqueues, timekeeping, ...) on behalf of full dynticks
+	 * CPUs. It must remain online when nohz full is enabled.
+	 */
+	if (tick_nohz_full_running && tick_do_timer_cpu == cpu)
+		return -EBUSY;
+	return 0;
 }
 
 static int tick_nohz_init_all(void)
@@ -428,7 +420,7 @@ static int tick_nohz_init_all(void)
 
 void __init tick_nohz_init(void)
 {
-	int cpu;
+	int cpu, ret;
 
 	if (!tick_nohz_full_running) {
 		if (tick_nohz_init_all() < 0)
@@ -469,7 +461,10 @@ void __init tick_nohz_init(void)
 	for_each_cpu(cpu, tick_nohz_full_mask)
 		context_tracking_cpu_set(cpu);
 
-	cpu_notifier(tick_nohz_cpu_down_callback, 0);
+	ret = cpuhp_setup_state_nocalls(CPUHP_AP_ONLINE_DYN,
+					"kernel/nohz:predown", NULL,
+					tick_nohz_cpu_down);
+	WARN_ON(ret < 0);
 	pr_info("NO_HZ: Full dynticks CPUs: %*pbl.\n",
 		cpumask_pr_args(tick_nohz_full_mask));
 
diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c
index 9c143739b8d7..89a2611a1635 100644
--- a/kernel/trace/ring_buffer.c
+++ b/kernel/trace/ring_buffer.c
@@ -479,9 +479,7 @@ struct ring_buffer {
 
 	struct ring_buffer_per_cpu	**buffers;
 
-#ifdef CONFIG_HOTPLUG_CPU
-	struct notifier_block		cpu_notify;
-#endif
+	struct hlist_node		node;
 	u64				(*clock)(void);
 
 	struct rb_irq_work		irq_work;
@@ -1274,11 +1272,6 @@ static void rb_free_cpu_buffer(struct ring_buffer_per_cpu *cpu_buffer)
 	kfree(cpu_buffer);
 }
 
-#ifdef CONFIG_HOTPLUG_CPU
-static int rb_cpu_notify(struct notifier_block *self,
-			 unsigned long action, void *hcpu);
-#endif
-
 /**
  * __ring_buffer_alloc - allocate a new ring_buffer
  * @size: the size in bytes per cpu that is needed.
@@ -1296,6 +1289,7 @@ struct ring_buffer *__ring_buffer_alloc(unsigned long size, unsigned flags,
 	long nr_pages;
 	int bsize;
 	int cpu;
+	int ret;
 
 	/* keep it in its own cache line */
 	buffer = kzalloc(ALIGN(sizeof(*buffer), cache_line_size()),
@@ -1303,7 +1297,7 @@ struct ring_buffer *__ring_buffer_alloc(unsigned long size, unsigned flags,
 	if (!buffer)
 		return NULL;
 
-	if (!alloc_cpumask_var(&buffer->cpumask, GFP_KERNEL))
+	if (!zalloc_cpumask_var(&buffer->cpumask, GFP_KERNEL))
 		goto fail_free_buffer;
 
 	nr_pages = DIV_ROUND_UP(size, BUF_PAGE_SIZE);
@@ -1318,17 +1312,6 @@ struct ring_buffer *__ring_buffer_alloc(unsigned long size, unsigned flags,
 	if (nr_pages < 2)
 		nr_pages = 2;
 
-	/*
-	 * In case of non-hotplug cpu, if the ring-buffer is allocated
-	 * in early initcall, it will not be notified of secondary cpus.
-	 * In that off case, we need to allocate for all possible cpus.
-	 */
-#ifdef CONFIG_HOTPLUG_CPU
-	cpu_notifier_register_begin();
-	cpumask_copy(buffer->cpumask, cpu_online_mask);
-#else
-	cpumask_copy(buffer->cpumask, cpu_possible_mask);
-#endif
 	buffer->cpus = nr_cpu_ids;
 
 	bsize = sizeof(void *) * nr_cpu_ids;
@@ -1337,19 +1320,15 @@ struct ring_buffer *__ring_buffer_alloc(unsigned long size, unsigned flags,
 	if (!buffer->buffers)
 		goto fail_free_cpumask;
 
-	for_each_buffer_cpu(buffer, cpu) {
-		buffer->buffers[cpu] =
-			rb_allocate_cpu_buffer(buffer, nr_pages, cpu);
-		if (!buffer->buffers[cpu])
-			goto fail_free_buffers;
-	}
+	cpu = raw_smp_processor_id();
+	cpumask_set_cpu(cpu, buffer->cpumask);
+	buffer->buffers[cpu] = rb_allocate_cpu_buffer(buffer, nr_pages, cpu);
+	if (!buffer->buffers[cpu])
+		goto fail_free_buffers;
 
-#ifdef CONFIG_HOTPLUG_CPU
-	buffer->cpu_notify.notifier_call = rb_cpu_notify;
-	buffer->cpu_notify.priority = 0;
-	__register_cpu_notifier(&buffer->cpu_notify);
-	cpu_notifier_register_done();
-#endif
+	ret = cpuhp_state_add_instance(CPUHP_TRACE_RB_PREPARE, &buffer->node);
+	if (ret < 0)
+		goto fail_free_buffers;
 
 	mutex_init(&buffer->mutex);
 
@@ -1364,9 +1343,6 @@ struct ring_buffer *__ring_buffer_alloc(unsigned long size, unsigned flags,
 
  fail_free_cpumask:
 	free_cpumask_var(buffer->cpumask);
-#ifdef CONFIG_HOTPLUG_CPU
-	cpu_notifier_register_done();
-#endif
 
  fail_free_buffer:
 	kfree(buffer);
@@ -1383,18 +1359,11 @@ ring_buffer_free(struct ring_buffer *buffer)
 {
 	int cpu;
 
-#ifdef CONFIG_HOTPLUG_CPU
-	cpu_notifier_register_begin();
-	__unregister_cpu_notifier(&buffer->cpu_notify);
-#endif
+	cpuhp_state_remove_instance(CPUHP_TRACE_RB_PREPARE, &buffer->node);
 
 	for_each_buffer_cpu(buffer, cpu)
 		rb_free_cpu_buffer(buffer->buffers[cpu]);
 
-#ifdef CONFIG_HOTPLUG_CPU
-	cpu_notifier_register_done();
-#endif
-
 	kfree(buffer->buffers);
 	free_cpumask_var(buffer->cpumask);
 
@@ -4633,62 +4602,48 @@ int ring_buffer_read_page(struct ring_buffer *buffer,
 }
 EXPORT_SYMBOL_GPL(ring_buffer_read_page);
 
-#ifdef CONFIG_HOTPLUG_CPU
-static int rb_cpu_notify(struct notifier_block *self,
-			 unsigned long action, void *hcpu)
+/*
+ * We only allocate new buffers, never free them if the CPU goes down.
+ * If we were to free the buffer, then the user would lose any trace that was in
+ * the buffer.
+ */
+int trace_rb_cpu_prepare(unsigned int cpu, struct hlist_node *node)
 {
-	struct ring_buffer *buffer =
-		container_of(self, struct ring_buffer, cpu_notify);
-	long cpu = (long)hcpu;
+	struct ring_buffer *buffer;
 	long nr_pages_same;
 	int cpu_i;
 	unsigned long nr_pages;
 
-	switch (action) {
-	case CPU_UP_PREPARE:
-	case CPU_UP_PREPARE_FROZEN:
-		if (cpumask_test_cpu(cpu, buffer->cpumask))
-			return NOTIFY_OK;
-
-		nr_pages = 0;
-		nr_pages_same = 1;
-		/* check if all cpu sizes are same */
-		for_each_buffer_cpu(buffer, cpu_i) {
-			/* fill in the size from first enabled cpu */
-			if (nr_pages == 0)
-				nr_pages = buffer->buffers[cpu_i]->nr_pages;
-			if (nr_pages != buffer->buffers[cpu_i]->nr_pages) {
-				nr_pages_same = 0;
-				break;
-			}
-		}
-		/* allocate minimum pages, user can later expand it */
-		if (!nr_pages_same)
-			nr_pages = 2;
-		buffer->buffers[cpu] =
-			rb_allocate_cpu_buffer(buffer, nr_pages, cpu);
-		if (!buffer->buffers[cpu]) {
-			WARN(1, "failed to allocate ring buffer on CPU %ld\n",
-			     cpu);
-			return NOTIFY_OK;
+	buffer = container_of(node, struct ring_buffer, node);
+	if (cpumask_test_cpu(cpu, buffer->cpumask))
+		return 0;
+
+	nr_pages = 0;
+	nr_pages_same = 1;
+	/* check if all cpu sizes are same */
+	for_each_buffer_cpu(buffer, cpu_i) {
+		/* fill in the size from first enabled cpu */
+		if (nr_pages == 0)
+			nr_pages = buffer->buffers[cpu_i]->nr_pages;
+		if (nr_pages != buffer->buffers[cpu_i]->nr_pages) {
+			nr_pages_same = 0;
+			break;
 		}
-		smp_wmb();
-		cpumask_set_cpu(cpu, buffer->cpumask);
-		break;
-	case CPU_DOWN_PREPARE:
-	case CPU_DOWN_PREPARE_FROZEN:
-		/*
-		 * Do nothing.
-		 *  If we were to free the buffer, then the user would
-		 *  lose any trace that was in the buffer.
-		 */
-		break;
-	default:
-		break;
 	}
-	return NOTIFY_OK;
+	/* allocate minimum pages, user can later expand it */
+	if (!nr_pages_same)
+		nr_pages = 2;
+	buffer->buffers[cpu] =
+		rb_allocate_cpu_buffer(buffer, nr_pages, cpu);
+	if (!buffer->buffers[cpu]) {
+		WARN(1, "failed to allocate ring buffer on CPU %u\n",
+		     cpu);
+		return -ENOMEM;
+	}
+	smp_wmb();
+	cpumask_set_cpu(cpu, buffer->cpumask);
+	return 0;
 }
-#endif
 
 #ifdef CONFIG_RING_BUFFER_STARTUP_TEST
 /*
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index 8696ce6bf2f6..465d56febc5b 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -7659,10 +7659,21 @@ __init static int tracer_alloc_buffers(void)
 
 	raw_spin_lock_init(&global_trace.start_lock);
 
+	/*
+	 * The prepare callbacks allocates some memory for the ring buffer. We
+	 * don't free the buffer if the if the CPU goes down. If we were to free
+	 * the buffer, then the user would lose any trace that was in the
+	 * buffer. The memory will be removed once the "instance" is removed.
+	 */
+	ret = cpuhp_setup_state_multi(CPUHP_TRACE_RB_PREPARE,
+				      "trace/RB:preapre", trace_rb_cpu_prepare,
+				      NULL);
+	if (ret < 0)
+		goto out_free_cpumask;
 	/* Used for event triggers */
 	temp_buffer = ring_buffer_alloc(PAGE_SIZE, RB_FL_OVERWRITE);
 	if (!temp_buffer)
-		goto out_free_cpumask;
+		goto out_rm_hp_state;
 
 	if (trace_create_savedcmd() < 0)
 		goto out_free_temp_buffer;
@@ -7723,6 +7734,8 @@ out_free_savedcmd:
 	free_saved_cmdlines_buffer(savedcmd);
 out_free_temp_buffer:
 	ring_buffer_free(temp_buffer);
+out_rm_hp_state:
+	cpuhp_remove_multi_state(CPUHP_TRACE_RB_PREPARE);
 out_free_cpumask:
 	free_cpumask_var(global_trace.tracing_cpumask);
 out_free_buffer_mask: