Merge branches 'pm-core', 'pm-runtime' and 'pm-sleep'

Merge changes related to system sleep and runtime PM framework for
6.18-rc1:

 - Annotate loops walking device links in the power management core
   code as _srcu and add macros for walking device links to reduce the
   likelihood of coding mistakes related to them (Rafael Wysocki)

 - Document time units for *_time functions in the runtime PM API (Brian
   Norris)

 - Clear power.must_resume in noirq suspend error path to avoid resuming
   a dependant device under a suspended parent or supplier (Rafael
   Wysocki)

 - Fix GFP mask handling during hybrid suspend and make the amdgpu
   driver handle hybrid suspend correctly (Mario Limonciello, Rafael
   Wysocki)

 - Fix GFP mask handling after aborted hibernation in platform mode and
   combine exit paths in power_down() to avoid code duplication (Rafael
   Wysocki)

 - Use vmalloc_array() and vcalloc() in the hibernation core to avoid
   open-coded size computations (Qianfeng Rong)

 - Fix typo in hibernation core code comment (Li Jun)

 - Call pm_wakeup_clear() in the same place where other functions that do
   bookkeeping prior to suspend_prepare() are called (Samuel Wu)

* pm-core:
  PM: core: Add two macros for walking device links
  PM: core: Annotate loops walking device links as _srcu

* pm-runtime:
  PM: runtime: Documentation: ABI: Document time units for *_time

* pm-sleep:
  PM: hibernate: Combine return paths in power_down()
  PM: hibernate: Restrict GFP mask in power_down()
  PM: hibernate: Fix pm_hibernation_mode_is_suspend() build breakage
  drm/amd: Fix hybrid sleep
  PM: hibernate: Add pm_hibernation_mode_is_suspend()
  PM: hibernate: Fix hybrid-sleep
  PM: sleep: core: Clear power.must_resume in noirq suspend error path
  PM: sleep: Make pm_wakeup_clear() call more clear
  PM: hibernate: Fix typo in memory bitmaps description comment
  PM: hibernate: Use vmalloc_array() and vcalloc() to improve code

Author: rafaeljw

Documentation/ABI/testing/sysfs-devices-power

@@ -274,15 +274,15 @@ What:		/sys/devices/.../power/runtime_active_time
 Date:		Jul 2010
 Contact:	Arjan van de Ven <arjan@linux.intel.com>
 Description:
-		Reports the total time that the device has been active.
-		Used for runtime PM statistics.
+		Reports the total time that the device has been active, in
+		milliseconds. Used for runtime PM statistics.
 
 What:		/sys/devices/.../power/runtime_suspended_time
 Date:		Jul 2010
 Contact:	Arjan van de Ven <arjan@linux.intel.com>
 Description:
-		Reports total time that the device has been suspended.
-		Used for runtime PM statistics.
+		Reports total time that the device has been suspended, in
+		milliseconds. Used for runtime PM statistics.
 
 What:		/sys/devices/.../power/runtime_usage
 Date:		Apr 2010

drivers/base/base.h

@@ -251,6 +251,14 @@ void device_links_unbind_consumers(struct device *dev);
 void fw_devlink_drivers_done(void);
 void fw_devlink_probing_done(void);
 
+#define dev_for_each_link_to_supplier(__link, __dev)	\
+	list_for_each_entry_srcu(__link, &(__dev)->links.suppliers, c_node, \
+				 device_links_read_lock_held())
+
+#define dev_for_each_link_to_consumer(__link, __dev)	\
+	list_for_each_entry_srcu(__link, &(__dev)->links.consumers, s_node, \
+				 device_links_read_lock_held())
+
 /* device pm support */
 void device_pm_move_to_tail(struct device *dev);
 

drivers/base/power/main.c

@@ -40,10 +40,6 @@
 
 typedef int (*pm_callback_t)(struct device *);
 
-#define list_for_each_entry_rcu_locked(pos, head, member) \
-	list_for_each_entry_rcu(pos, head, member, \
-			device_links_read_lock_held())
-
 /*
  * The entries in the dpm_list list are in a depth first order, simply
  * because children are guaranteed to be discovered after parents, and
@@ -281,7 +277,7 @@ static void dpm_wait_for_suppliers(struct device *dev, bool async)
 	 * callbacks freeing the link objects for the links in the list we're
 	 * walking.
 	 */
-	list_for_each_entry_rcu_locked(link, &dev->links.suppliers, c_node)
+	dev_for_each_link_to_supplier(link, dev)
 		if (READ_ONCE(link->status) != DL_STATE_DORMANT)
 			dpm_wait(link->supplier, async);
 
@@ -338,7 +334,7 @@ static void dpm_wait_for_consumers(struct device *dev, bool async)
 	 * continue instead of trying to continue in parallel with its
 	 * unregistration).
 	 */
-	list_for_each_entry_rcu_locked(link, &dev->links.consumers, s_node)
+	dev_for_each_link_to_consumer(link, dev)
 		if (READ_ONCE(link->status) != DL_STATE_DORMANT)
 			dpm_wait(link->consumer, async);
 
@@ -675,7 +671,7 @@ static void dpm_async_resume_subordinate(struct device *dev, async_func_t func)
 	idx = device_links_read_lock();
 
 	/* Start processing the device's "async" consumers. */
-	list_for_each_entry_rcu_locked(link, &dev->links.consumers, s_node)
+	dev_for_each_link_to_consumer(link, dev)
 		if (READ_ONCE(link->status) != DL_STATE_DORMANT)
 			dpm_async_with_cleanup(link->consumer, func);
 
@@ -724,8 +720,20 @@ static void device_resume_noirq(struct device *dev, pm_message_t state, bool asy
 	if (dev->power.syscore || dev->power.direct_complete)
 		goto Out;
 
-	if (!dev->power.is_noirq_suspended)
+	if (!dev->power.is_noirq_suspended) {
+		/*
+		 * This means that system suspend has been aborted in the noirq
+		 * phase before invoking the noirq suspend callback for the
+		 * device, so if device_suspend_late() has left it in suspend,
+		 * device_resume_early() should leave it in suspend either in
+		 * case the early resume of it depends on the noirq resume that
+		 * has not run.
+		 */
+		if (dev_pm_skip_suspend(dev))
+			dev->power.must_resume = false;
+
 		goto Out;
+	}
 
 	if (!dpm_wait_for_superior(dev, async))
 		goto Out;
@@ -1330,7 +1338,7 @@ static void dpm_async_suspend_superior(struct device *dev, async_func_t func)
 	idx = device_links_read_lock();
 
 	/* Start processing the device's "async" suppliers. */
-	list_for_each_entry_rcu_locked(link, &dev->links.suppliers, c_node)
+	dev_for_each_link_to_supplier(link, dev)
 		if (READ_ONCE(link->status) != DL_STATE_DORMANT)
 			dpm_async_with_cleanup(link->supplier, func);
 
@@ -1384,7 +1392,7 @@ static void dpm_superior_set_must_resume(struct device *dev)
 
 	idx = device_links_read_lock();
 
-	list_for_each_entry_rcu_locked(link, &dev->links.suppliers, c_node)
+	dev_for_each_link_to_supplier(link, dev)
 		link->supplier->power.must_resume = true;
 
 	device_links_read_unlock(idx);
@@ -1813,7 +1821,7 @@ static void dpm_clear_superiors_direct_complete(struct device *dev)
 
 	idx = device_links_read_lock();
 
-	list_for_each_entry_rcu_locked(link, &dev->links.suppliers, c_node) {
+	dev_for_each_link_to_supplier(link, dev) {
 		spin_lock_irq(&link->supplier->power.lock);
 		link->supplier->power.direct_complete = false;
 		spin_unlock_irq(&link->supplier->power.lock);
@@ -2065,7 +2073,7 @@ static bool device_prepare_smart_suspend(struct device *dev)
 
 	idx = device_links_read_lock();
 
-	list_for_each_entry_rcu_locked(link, &dev->links.suppliers, c_node) {
+	dev_for_each_link_to_supplier(link, dev) {
 		if (!device_link_test(link, DL_FLAG_PM_RUNTIME))
 			continue;
 

drivers/base/power/runtime.c

@@ -1903,8 +1903,7 @@ void pm_runtime_get_suppliers(struct device *dev)
 
 	idx = device_links_read_lock();
 
-	list_for_each_entry_rcu(link, &dev->links.suppliers, c_node,
-				device_links_read_lock_held())
+	dev_for_each_link_to_supplier(link, dev)
 		if (device_link_test(link, DL_FLAG_PM_RUNTIME)) {
 			link->supplier_preactivated = true;
 			pm_runtime_get_sync(link->supplier);

drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c

@@ -2665,7 +2665,7 @@ static int amdgpu_pmops_thaw(struct device *dev)
 	struct drm_device *drm_dev = dev_get_drvdata(dev);
 
 	/* do not resume device if it's normal hibernation */
-	if (!pm_hibernate_is_recovering())
+	if (!pm_hibernate_is_recovering() && !pm_hibernation_mode_is_suspend())
 		return 0;
 
 	return amdgpu_device_resume(drm_dev, true);

include/linux/suspend.h

@@ -418,6 +418,12 @@ static inline int hibernate_quiet_exec(int (*func)(void *data), void *data) {
 }
 #endif /* CONFIG_HIBERNATION */
 
+#if defined(CONFIG_HIBERNATION) && defined(CONFIG_SUSPEND)
+bool pm_hibernation_mode_is_suspend(void);
+#else
+static inline bool pm_hibernation_mode_is_suspend(void) { return false; }
+#endif
+
 int arch_resume_nosmt(void);
 
 #ifdef CONFIG_HIBERNATION_SNAPSHOT_DEV

kernel/power/hibernate.c

@@ -80,6 +80,17 @@ static const struct platform_hibernation_ops *hibernation_ops;
 
 static atomic_t hibernate_atomic = ATOMIC_INIT(1);
 
+#ifdef CONFIG_SUSPEND
+/**
+ * pm_hibernation_mode_is_suspend - Check if hibernation has been set to suspend
+ */
+bool pm_hibernation_mode_is_suspend(void)
+{
+	return hibernation_mode == HIBERNATION_SUSPEND;
+}
+EXPORT_SYMBOL_GPL(pm_hibernation_mode_is_suspend);
+#endif
+
 bool hibernate_acquire(void)
 {
 	return atomic_add_unless(&hibernate_atomic, -1, 0);
@@ -695,19 +706,13 @@ static void power_down(void)
 
 #ifdef CONFIG_SUSPEND
 	if (hibernation_mode == HIBERNATION_SUSPEND) {
+		pm_restore_gfp_mask();
 		error = suspend_devices_and_enter(mem_sleep_current);
-		if (error) {
-			hibernation_mode = hibernation_ops ?
-						HIBERNATION_PLATFORM :
-						HIBERNATION_SHUTDOWN;
-		} else {
-			/* Restore swap signature. */
-			error = swsusp_unmark();
-			if (error)
-				pr_err("Swap will be unusable! Try swapon -a.\n");
+		if (!error)
+			goto exit;
 
-			return;
-		}
+		hibernation_mode = hibernation_ops ? HIBERNATION_PLATFORM :
+						     HIBERNATION_SHUTDOWN;
 	}
 #endif
 
@@ -718,10 +723,9 @@ static void power_down(void)
 	case HIBERNATION_PLATFORM:
 		error = hibernation_platform_enter();
 		if (error == -EAGAIN || error == -EBUSY) {
-			swsusp_unmark();
 			events_check_enabled = false;
 			pr_info("Wakeup event detected during hibernation, rolling back.\n");
-			return;
+			goto exit;
 		}
 		fallthrough;
 	case HIBERNATION_SHUTDOWN:
@@ -740,6 +744,15 @@ static void power_down(void)
 	pr_crit("Power down manually\n");
 	while (1)
 		cpu_relax();
+
+exit:
+	/* Match the pm_restore_gfp_mask() call in hibernate(). */
+	pm_restrict_gfp_mask();
+
+	/* Restore swap signature. */
+	error = swsusp_unmark();
+	if (error)
+		pr_err("Swap will be unusable! Try swapon -a.\n");
 }
 
 static int load_image_and_restore(void)

kernel/power/process.c

@@ -132,7 +132,6 @@ int freeze_processes(void)
 	if (!pm_freezing)
 		static_branch_inc(&freezer_active);
 
-	pm_wakeup_clear(0);
 	pm_freezing = true;
 	error = try_to_freeze_tasks(true);
 	if (!error)

kernel/power/snapshot.c

@@ -363,7 +363,7 @@ static void *chain_alloc(struct chain_allocator *ca, unsigned int size)
  *
  * One radix tree is represented by one struct mem_zone_bm_rtree. There are
  * two linked lists for the nodes of the tree, one for the inner nodes and
- * one for the leave nodes. The linked leave nodes are used for fast linear
+ * one for the leaf nodes. The linked leaf nodes are used for fast linear
  * access of the memory bitmap.
  *
  * The struct rtree_node represents one node of the radix tree.

kernel/power/suspend.c

@@ -595,6 +595,7 @@ static int enter_state(suspend_state_t state)
 	}
 
 	pm_pr_dbg("Preparing system for sleep (%s)\n", mem_sleep_labels[state]);
+	pm_wakeup_clear(0);
 	pm_suspend_clear_flags();
 	error = suspend_prepare(state);
 	if (error)