Merge branches 'pm-sleep' and 'pm-tools'

* pm-sleep:
  PM / hibernate: Introduce test_resume mode for hibernation
  x86 / hibernate: Use hlt_play_dead() when resuming from hibernation
  PM / hibernate: Image data protection during restoration
  PM / hibernate: Add missing braces in __register_nosave_region()
  PM / hibernate: Clean up comments in snapshot.c
  PM / hibernate: Clean up function headers in snapshot.c
  PM / hibernate: Add missing braces in hibernate_setup()
  PM / hibernate: Recycle safe pages after image restoration
  PM / hibernate: Simplify mark_unsafe_pages()
  PM / hibernate: Do not free preallocated safe pages during image restore
  PM / suspend: show workqueue state in suspend flow
  PM / sleep: make PM notifiers called symmetrically
  PM / sleep: Make pm_prepare_console() return void
  PM / Hibernate: Don't let kasan instrument snapshot.c

* pm-tools:
  PM / tools: scripts: AnalyzeSuspend v4.2
  tools/turbostat: allow user to alter DESTDIR and PREFIX

Documentation/kernel-parameters.txt

@@ -3594,6 +3594,9 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
 				present during boot.
 		nocompress	Don't compress/decompress hibernation images.
 		no		Disable hibernation and resume.
+		protect_image	Turn on image protection during restoration
+				(that will set all pages holding image data
+				during restoration read-only).
 
 	retain_initrd	[RAM] Keep initrd memory after extraction
 

arch/x86/include/asm/smp.h

@@ -135,6 +135,7 @@ int native_cpu_up(unsigned int cpunum, struct task_struct *tidle);
 int native_cpu_disable(void);
 int common_cpu_die(unsigned int cpu);
 void native_cpu_die(unsigned int cpu);
+void hlt_play_dead(void);
 void native_play_dead(void);
 void play_dead_common(void);
 void wbinvd_on_cpu(int cpu);

arch/x86/kernel/smpboot.c

@@ -1622,7 +1622,7 @@ static inline void mwait_play_dead(void)
 	}
 }
 
-static inline void hlt_play_dead(void)
+void hlt_play_dead(void)
 {
 	if (__this_cpu_read(cpu_info.x86) >= 4)
 		wbinvd();

arch/x86/power/cpu.c

@@ -12,6 +12,7 @@
 #include <linux/export.h>
 #include <linux/smp.h>
 #include <linux/perf_event.h>
+#include <linux/tboot.h>
 
 #include <asm/pgtable.h>
 #include <asm/proto.h>
@@ -266,6 +267,35 @@ void notrace restore_processor_state(void)
 EXPORT_SYMBOL(restore_processor_state);
 #endif
 
+#if defined(CONFIG_HIBERNATION) && defined(CONFIG_HOTPLUG_CPU)
+static void resume_play_dead(void)
+{
+	play_dead_common();
+	tboot_shutdown(TB_SHUTDOWN_WFS);
+	hlt_play_dead();
+}
+
+int hibernate_resume_nonboot_cpu_disable(void)
+{
+	void (*play_dead)(void) = smp_ops.play_dead;
+	int ret;
+
+	/*
+	 * Ensure that MONITOR/MWAIT will not be used in the "play dead" loop
+	 * during hibernate image restoration, because it is likely that the
+	 * monitored address will be actually written to at that time and then
+	 * the "dead" CPU will attempt to execute instructions again, but the
+	 * address in its instruction pointer may not be possible to resolve
+	 * any more at that point (the page tables used by it previously may
+	 * have been overwritten by hibernate image data).
+	 */
+	smp_ops.play_dead = resume_play_dead;
+	ret = disable_nonboot_cpus();
+	smp_ops.play_dead = play_dead;
+	return ret;
+}
+#endif
+
 /*
  * When bsp_check() is called in hibernate and suspend, cpu hotplug
  * is disabled already. So it's unnessary to handle race condition between

include/linux/suspend.h

@@ -18,12 +18,11 @@ static inline void pm_set_vt_switch(int do_switch)
 #endif
 
 #ifdef CONFIG_VT_CONSOLE_SLEEP
-extern int pm_prepare_console(void);
+extern void pm_prepare_console(void);
 extern void pm_restore_console(void);
 #else
-static inline int pm_prepare_console(void)
+static inline void pm_prepare_console(void)
 {
-	return 0;
 }
 
 static inline void pm_restore_console(void)

kernel/power/Makefile

@@ -1,6 +1,8 @@
 
 ccflags-$(CONFIG_PM_DEBUG)	:= -DDEBUG
 
+KASAN_SANITIZE_snapshot.o	:= n
+
 obj-y				+= qos.o
 obj-$(CONFIG_PM)		+= main.o
 obj-$(CONFIG_VT_CONSOLE_SLEEP)	+= console.o

kernel/power/console.c

@@ -126,17 +126,17 @@ out:
 	return ret;
 }
 
-int pm_prepare_console(void)
+void pm_prepare_console(void)
 {
 	if (!pm_vt_switch())
-		return 0;
+		return;
 
 	orig_fgconsole = vt_move_to_console(SUSPEND_CONSOLE, 1);
 	if (orig_fgconsole < 0)
-		return 1;
+		return;
 
 	orig_kmsg = vt_kmsg_redirect(SUSPEND_CONSOLE);
-	return 0;
+	return;
 }
 
 void pm_restore_console(void)

kernel/power/hibernate.c

@@ -52,6 +52,7 @@ enum {
 #ifdef CONFIG_SUSPEND
 	HIBERNATION_SUSPEND,
 #endif
+	HIBERNATION_TEST_RESUME,
 	/* keep last */
 	__HIBERNATION_AFTER_LAST
 };
@@ -409,6 +410,11 @@ int hibernation_snapshot(int platform_mode)
 	goto Close;
 }
 
+int __weak hibernate_resume_nonboot_cpu_disable(void)
+{
+	return disable_nonboot_cpus();
+}
+
 /**
  * resume_target_kernel - Restore system state from a hibernation image.
  * @platform_mode: Whether or not to use the platform driver.
@@ -433,7 +439,7 @@ static int resume_target_kernel(bool platform_mode)
 	if (error)
 		goto Cleanup;
 
-	error = disable_nonboot_cpus();
+	error = hibernate_resume_nonboot_cpu_disable();
 	if (error)
 		goto Enable_cpus;
 
@@ -642,12 +648,39 @@ static void power_down(void)
 		cpu_relax();
 }
 
+static int load_image_and_restore(void)
+{
+	int error;
+	unsigned int flags;
+
+	pr_debug("PM: Loading hibernation image.\n");
+
+	lock_device_hotplug();
+	error = create_basic_memory_bitmaps();
+	if (error)
+		goto Unlock;
+
+	error = swsusp_read(&flags);
+	swsusp_close(FMODE_READ);
+	if (!error)
+		hibernation_restore(flags & SF_PLATFORM_MODE);
+
+	printk(KERN_ERR "PM: Failed to load hibernation image, recovering.\n");
+	swsusp_free();
+	free_basic_memory_bitmaps();
+ Unlock:
+	unlock_device_hotplug();
+
+	return error;
+}
+
 /**
  * hibernate - Carry out system hibernation, including saving the image.
  */
 int hibernate(void)
 {
-	int error;
+	int error, nr_calls = 0;
+	bool snapshot_test = false;
 
 	if (!hibernation_available()) {
 		pr_debug("PM: Hibernation not available.\n");
@@ -662,9 +695,11 @@ int hibernate(void)
 	}
 
 	pm_prepare_console();
-	error = pm_notifier_call_chain(PM_HIBERNATION_PREPARE);
+	error = __pm_notifier_call_chain(PM_HIBERNATION_PREPARE, -1, &nr_calls);
-	if (error)
+	if (error) {
+		nr_calls--;
 		goto Exit;
+	}
 
 	printk(KERN_INFO "PM: Syncing filesystems ... ");
 	sys_sync();
@@ -697,8 +732,12 @@ int hibernate(void)
 		pr_debug("PM: writing image.\n");
 		error = swsusp_write(flags);
 		swsusp_free();
-		if (!error)
+		if (!error) {
-			power_down();
+			if (hibernation_mode == HIBERNATION_TEST_RESUME)
+				snapshot_test = true;
+			else
+				power_down();
+		}
 		in_suspend = 0;
 		pm_restore_gfp_mask();
 	} else {
@@ -709,12 +748,18 @@ int hibernate(void)
 	free_basic_memory_bitmaps();
  Thaw:
 	unlock_device_hotplug();
+	if (snapshot_test) {
+		pr_debug("PM: Checking hibernation image\n");
+		error = swsusp_check();
+		if (!error)
+			error = load_image_and_restore();
+	}
 	thaw_processes();
 
 	/* Don't bother checking whether freezer_test_done is true */
 	freezer_test_done = false;
  Exit:
-	pm_notifier_call_chain(PM_POST_HIBERNATION);
+	__pm_notifier_call_chain(PM_POST_HIBERNATION, nr_calls, NULL);
 	pm_restore_console();
 	atomic_inc(&snapshot_device_available);
  Unlock:
@@ -740,8 +785,7 @@ int hibernate(void)
  */
 static int software_resume(void)
 {
-	int error;
+	int error, nr_calls = 0;
-	unsigned int flags;
 
 	/*
 	 * If the user said "noresume".. bail out early.
@@ -827,35 +871,20 @@ static int software_resume(void)
 	}
 
 	pm_prepare_console();
-	error = pm_notifier_call_chain(PM_RESTORE_PREPARE);
+	error = __pm_notifier_call_chain(PM_RESTORE_PREPARE, -1, &nr_calls);
-	if (error)
+	if (error) {
+		nr_calls--;
 		goto Close_Finish;
+	}
 
 	pr_debug("PM: Preparing processes for restore.\n");
 	error = freeze_processes();
 	if (error)
 		goto Close_Finish;
-
+	error = load_image_and_restore();
-	pr_debug("PM: Loading hibernation image.\n");
-
-	lock_device_hotplug();
-	error = create_basic_memory_bitmaps();
-	if (error)
-		goto Thaw;
-
-	error = swsusp_read(&flags);
-	swsusp_close(FMODE_READ);
-	if (!error)
-		hibernation_restore(flags & SF_PLATFORM_MODE);
-
-	printk(KERN_ERR "PM: Failed to load hibernation image, recovering.\n");
-	swsusp_free();
-	free_basic_memory_bitmaps();
- Thaw:
-	unlock_device_hotplug();
 	thaw_processes();
  Finish:
-	pm_notifier_call_chain(PM_POST_RESTORE);
+	__pm_notifier_call_chain(PM_POST_RESTORE, nr_calls, NULL);
 	pm_restore_console();
 	atomic_inc(&snapshot_device_available);
 	/* For success case, the suspend path will release the lock */
@@ -878,6 +907,7 @@ static const char * const hibernation_modes[] = {
 #ifdef CONFIG_SUSPEND
 	[HIBERNATION_SUSPEND]	= "suspend",
 #endif
+	[HIBERNATION_TEST_RESUME]	= "test_resume",
 };
 
 /*
@@ -924,6 +954,7 @@ static ssize_t disk_show(struct kobject *kobj, struct kobj_attribute *attr,
 #ifdef CONFIG_SUSPEND
 		case HIBERNATION_SUSPEND:
 #endif
+		case HIBERNATION_TEST_RESUME:
 			break;
 		case HIBERNATION_PLATFORM:
 			if (hibernation_ops)
@@ -970,6 +1001,7 @@ static ssize_t disk_store(struct kobject *kobj, struct kobj_attribute *attr,
 #ifdef CONFIG_SUSPEND
 		case HIBERNATION_SUSPEND:
 #endif
+		case HIBERNATION_TEST_RESUME:
 			hibernation_mode = mode;
 			break;
 		case HIBERNATION_PLATFORM:
@@ -1115,13 +1147,16 @@ static int __init resume_offset_setup(char *str)
 
 static int __init hibernate_setup(char *str)
 {
-	if (!strncmp(str, "noresume", 8))
+	if (!strncmp(str, "noresume", 8)) {
 		noresume = 1;
-	else if (!strncmp(str, "nocompress", 10))
+	} else if (!strncmp(str, "nocompress", 10)) {
 		nocompress = 1;
-	else if (!strncmp(str, "no", 2)) {
+	} else if (!strncmp(str, "no", 2)) {
 		noresume = 1;
 		nohibernate = 1;
+	} else if (IS_ENABLED(CONFIG_DEBUG_RODATA)
+		   && !strncmp(str, "protect_image", 13)) {
+		enable_restore_image_protection();
 	}
 	return 1;
 }

kernel/power/main.c

@@ -38,12 +38,19 @@ int unregister_pm_notifier(struct notifier_block *nb)
 }
 EXPORT_SYMBOL_GPL(unregister_pm_notifier);
 
-int pm_notifier_call_chain(unsigned long val)
+int __pm_notifier_call_chain(unsigned long val, int nr_to_call, int *nr_calls)
 {
-	int ret = blocking_notifier_call_chain(&pm_chain_head, val, NULL);
+	int ret;
+
+	ret = __blocking_notifier_call_chain(&pm_chain_head, val, NULL,
+						nr_to_call, nr_calls);
 
 	return notifier_to_errno(ret);
 }
+int pm_notifier_call_chain(unsigned long val)
+{
+	return __pm_notifier_call_chain(val, -1, NULL);
+}
 
 /* If set, devices may be suspended and resumed asynchronously. */
 int pm_async_enabled = 1;

kernel/power/power.h

@@ -38,6 +38,8 @@ static inline char *check_image_kernel(struct swsusp_info *info)
 }
 #endif /* CONFIG_ARCH_HIBERNATION_HEADER */
 
+extern int hibernate_resume_nonboot_cpu_disable(void);
+
 /*
  * Keep some memory free so that I/O operations can succeed without paging
  * [Might this be more than 4 MB?]
@@ -59,6 +61,13 @@ extern int hibernation_snapshot(int platform_mode);
 extern int hibernation_restore(int platform_mode);
 extern int hibernation_platform_enter(void);
 
+#ifdef CONFIG_DEBUG_RODATA
+/* kernel/power/snapshot.c */
+extern void enable_restore_image_protection(void);
+#else
+static inline void enable_restore_image_protection(void) {}
+#endif /* CONFIG_DEBUG_RODATA */
+
 #else /* !CONFIG_HIBERNATION */
 
 static inline void hibernate_reserved_size_init(void) {}
@@ -200,6 +209,8 @@ static inline void suspend_test_finish(const char *label) {}
 
 #ifdef CONFIG_PM_SLEEP
 /* kernel/power/main.c */
+extern int __pm_notifier_call_chain(unsigned long val, int nr_to_call,
+				    int *nr_calls);
 extern int pm_notifier_call_chain(unsigned long val);
 #endif
 

kernel/power/process.c

@@ -89,6 +89,9 @@ static int try_to_freeze_tasks(bool user_only)
 		       elapsed_msecs / 1000, elapsed_msecs % 1000,
 		       todo - wq_busy, wq_busy);
 
+		if (wq_busy)
+			show_workqueue_state();
+
 		if (!wakeup) {
 			read_lock(&tasklist_lock);
 			for_each_process_thread(g, p) {

kernel/power/suspend.c

@@ -266,16 +266,18 @@ static int suspend_test(int level)
  */
 static int suspend_prepare(suspend_state_t state)
 {
-	int error;
+	int error, nr_calls = 0;
 
 	if (!sleep_state_supported(state))
 		return -EPERM;
 
 	pm_prepare_console();
 
-	error = pm_notifier_call_chain(PM_SUSPEND_PREPARE);
+	error = __pm_notifier_call_chain(PM_SUSPEND_PREPARE, -1, &nr_calls);
-	if (error)
+	if (error) {
+		nr_calls--;
 		goto Finish;
+	}
 
 	trace_suspend_resume(TPS("freeze_processes"), 0, true);
 	error = suspend_freeze_processes();
@@ -286,7 +288,7 @@ static int suspend_prepare(suspend_state_t state)
 	suspend_stats.failed_freeze++;
 	dpm_save_failed_step(SUSPEND_FREEZE);
  Finish:
-	pm_notifier_call_chain(PM_POST_SUSPEND);
+	__pm_notifier_call_chain(PM_POST_SUSPEND, nr_calls, NULL);
 	pm_restore_console();
 	return error;
 }

kernel/power/swap.c

@@ -348,6 +348,12 @@ static int swsusp_swap_check(void)
 	if (res < 0)
 		blkdev_put(hib_resume_bdev, FMODE_WRITE);
 
+	/*
+	 * Update the resume device to the one actually used,
+	 * so the test_resume mode can use it in case it is
+	 * invoked from hibernate() to test the snapshot.
+	 */
+	swsusp_resume_device = hib_resume_bdev->bd_dev;
 	return res;
 }
 

kernel/power/user.c

@@ -47,7 +47,7 @@ atomic_t snapshot_device_available = ATOMIC_INIT(1);
 static int snapshot_open(struct inode *inode, struct file *filp)
 {
 	struct snapshot_data *data;
-	int error;
+	int error, nr_calls = 0;
 
 	if (!hibernation_available())
 		return -EPERM;
@@ -74,9 +74,9 @@ static int snapshot_open(struct inode *inode, struct file *filp)
 			swap_type_of(swsusp_resume_device, 0, NULL) : -1;
 		data->mode = O_RDONLY;
 		data->free_bitmaps = false;
-		error = pm_notifier_call_chain(PM_HIBERNATION_PREPARE);
+		error = __pm_notifier_call_chain(PM_HIBERNATION_PREPARE, -1, &nr_calls);
 		if (error)
-			pm_notifier_call_chain(PM_POST_HIBERNATION);
+			__pm_notifier_call_chain(PM_POST_HIBERNATION, --nr_calls, NULL);
 	} else {
 		/*
 		 * Resuming.  We may need to wait for the image device to
@@ -86,13 +86,15 @@ static int snapshot_open(struct inode *inode, struct file *filp)
 
 		data->swap = -1;
 		data->mode = O_WRONLY;
-		error = pm_notifier_call_chain(PM_RESTORE_PREPARE);
+		error = __pm_notifier_call_chain(PM_RESTORE_PREPARE, -1, &nr_calls);
 		if (!error) {
 			error = create_basic_memory_bitmaps();
 			data->free_bitmaps = !error;
-		}
+		} else
+			nr_calls--;
+
 		if (error)
-			pm_notifier_call_chain(PM_POST_RESTORE);
+			__pm_notifier_call_chain(PM_POST_RESTORE, nr_calls, NULL);
 	}
 	if (error)
 		atomic_inc(&snapshot_device_available);

kernel/workqueue.c

@@ -4369,8 +4369,8 @@ static void show_pwq(struct pool_workqueue *pwq)
 /**
  * show_workqueue_state - dump workqueue state
  *
- * Called from a sysrq handler and prints out all busy workqueues and
+ * Called from a sysrq handler or try_to_freeze_tasks() and prints out
- * pools.
+ * all busy workqueues and pools.
  */
 void show_workqueue_state(void)
 {

tools/power/x86/turbostat/Makefile

@@ -1,7 +1,7 @@
 CC		= $(CROSS_COMPILE)gcc
 BUILD_OUTPUT	:= $(CURDIR)
-PREFIX		:= /usr
+PREFIX		?= /usr
-DESTDIR		:=
+DESTDIR		?=
 
 ifeq ("$(origin O)", "command line")
 	BUILD_OUTPUT := $(O)