Merge branch 'linus' into xen-64bit

4 files changed, 135 insertions, 99 deletions

diff --git a/kernel/power/disk.c b/kernel/power/disk.c
index 14a656c..f011e08 100644
--- a/kernel/power/disk.c
+++ b/kernel/power/disk.c
@@ -180,6 +180,17 @@ static void platform_restore_cleanup(int platform_mode)
 }
 
 /**
+ *	platform_recover - recover the platform from a failure to suspend
+ *	devices.
+ */
+
+static void platform_recover(int platform_mode)
+{
+	if (platform_mode && hibernation_ops && hibernation_ops->recover)
+		hibernation_ops->recover();
+}
+
+/**
  *	create_image - freeze devices that need to be frozen with interrupts
  *	off, create the hibernation image and thaw those devices.  Control
  *	reappears in this routine after a restore.
@@ -193,6 +204,7 @@ static int create_image(int platform_mode)
 	if (error)
 		return error;
 
+	device_pm_lock();
 	local_irq_disable();
 	/* At this point, device_suspend() has been called, but *not*
 	 * device_power_down(). We *must* call device_power_down() now.
@@ -224,9 +236,11 @@ static int create_image(int platform_mode)
 	/* NOTE:  device_power_up() is just a resume() for devices
 	 * that suspended with irqs off ... no overall powerup.
 	 */
-	device_power_up();
+	device_power_up(in_suspend ?
+		(error ? PMSG_RECOVER : PMSG_THAW) : PMSG_RESTORE);
  Enable_irqs:
 	local_irq_enable();
+	device_pm_unlock();
 	return error;
 }
 
@@ -255,10 +269,10 @@ int hibernation_snapshot(int platform_mode)
 	suspend_console();
 	error = device_suspend(PMSG_FREEZE);
 	if (error)
-		goto Resume_console;
+		goto Recover_platform;
 
 	if (hibernation_test(TEST_DEVICES))
-		goto Resume_devices;
+		goto Recover_platform;
 
 	error = platform_pre_snapshot(platform_mode);
 	if (error || hibernation_test(TEST_PLATFORM))
@@ -280,12 +294,16 @@ int hibernation_snapshot(int platform_mode)
  Finish:
 	platform_finish(platform_mode);
  Resume_devices:
-	device_resume();
- Resume_console:
+	device_resume(in_suspend ?
+		(error ? PMSG_RECOVER : PMSG_THAW) : PMSG_RESTORE);
 	resume_console();
  Close:
 	platform_end(platform_mode);
 	return error;
+
+ Recover_platform:
+	platform_recover(platform_mode);
+	goto Resume_devices;
 }
 
 /**
@@ -300,8 +318,9 @@ static int resume_target_kernel(void)
 {
 	int error;
 
+	device_pm_lock();
 	local_irq_disable();
-	error = device_power_down(PMSG_PRETHAW);
+	error = device_power_down(PMSG_QUIESCE);
 	if (error) {
 		printk(KERN_ERR "PM: Some devices failed to power down, "
 			"aborting resume\n");
@@ -329,9 +348,10 @@ static int resume_target_kernel(void)
 	swsusp_free();
 	restore_processor_state();
 	touch_softlockup_watchdog();
-	device_power_up();
+	device_power_up(PMSG_RECOVER);
  Enable_irqs:
 	local_irq_enable();
+	device_pm_unlock();
 	return error;
 }
 
@@ -350,7 +370,7 @@ int hibernation_restore(int platform_mode)
 
 	pm_prepare_console();
 	suspend_console();
-	error = device_suspend(PMSG_PRETHAW);
+	error = device_suspend(PMSG_QUIESCE);
 	if (error)
 		goto Finish;
 
@@ -362,7 +382,7 @@ int hibernation_restore(int platform_mode)
 		enable_nonboot_cpus();
 	}
 	platform_restore_cleanup(platform_mode);
-	device_resume();
+	device_resume(PMSG_RECOVER);
  Finish:
 	resume_console();
 	pm_restore_console();
@@ -392,8 +412,11 @@ int hibernation_platform_enter(void)
 
 	suspend_console();
 	error = device_suspend(PMSG_HIBERNATE);
-	if (error)
-		goto Resume_console;
+	if (error) {
+		if (hibernation_ops->recover)
+			hibernation_ops->recover();
+		goto Resume_devices;
+	}
 
 	error = hibernation_ops->prepare();
 	if (error)
@@ -403,6 +426,7 @@ int hibernation_platform_enter(void)
 	if (error)
 		goto Finish;
 
+	device_pm_lock();
 	local_irq_disable();
 	error = device_power_down(PMSG_HIBERNATE);
 	if (!error) {
@@ -411,6 +435,7 @@ int hibernation_platform_enter(void)
 		while (1);
 	}
 	local_irq_enable();
+	device_pm_unlock();
 
 	/*
 	 * We don't need to reenable the nonboot CPUs or resume consoles, since
@@ -419,8 +444,7 @@ int hibernation_platform_enter(void)
  Finish:
 	hibernation_ops->finish();
  Resume_devices:
-	device_resume();
- Resume_console:
+	device_resume(PMSG_RESTORE);
 	resume_console();
  Close:
 	hibernation_ops->end();
diff --git a/kernel/power/main.c b/kernel/power/main.c
index 6a6d5eb..3398f46 100644
--- a/kernel/power/main.c
+++ b/kernel/power/main.c
@@ -228,6 +228,7 @@ static int suspend_enter(suspend_state_t state)
 {
 	int error = 0;
 
+	device_pm_lock();
 	arch_suspend_disable_irqs();
 	BUG_ON(!irqs_disabled());
 
@@ -239,10 +240,11 @@ static int suspend_enter(suspend_state_t state)
 	if (!suspend_test(TEST_CORE))
 		error = suspend_ops->enter(state);
 
-	device_power_up();
+	device_power_up(PMSG_RESUME);
  Done:
 	arch_suspend_enable_irqs();
 	BUG_ON(irqs_disabled());
+	device_pm_unlock();
 	return error;
 }
 
@@ -267,11 +269,11 @@ int suspend_devices_and_enter(suspend_state_t state)
 	error = device_suspend(PMSG_SUSPEND);
 	if (error) {
 		printk(KERN_ERR "PM: Some devices failed to suspend\n");
-		goto Resume_console;
+		goto Recover_platform;
 	}
 
 	if (suspend_test(TEST_DEVICES))
-		goto Resume_devices;
+		goto Recover_platform;
 
 	if (suspend_ops->prepare) {
 		error = suspend_ops->prepare();
@@ -291,13 +293,17 @@ int suspend_devices_and_enter(suspend_state_t state)
 	if (suspend_ops->finish)
 		suspend_ops->finish();
  Resume_devices:
-	device_resume();
- Resume_console:
+	device_resume(PMSG_RESUME);
 	resume_console();
  Close:
 	if (suspend_ops->end)
 		suspend_ops->end();
 	return error;
+
+ Recover_platform:
+	if (suspend_ops->recover)
+		suspend_ops->recover();
+	goto Resume_devices;
 }
 
 /**
diff --git a/kernel/power/process.c b/kernel/power/process.c
index f1d0b34..5fb8765 100644
--- a/kernel/power/process.c
+++ b/kernel/power/process.c
@@ -19,9 +19,6 @@
  */
 #define TIMEOUT	(20 * HZ)
 
-#define FREEZER_KERNEL_THREADS 0
-#define FREEZER_USER_SPACE 1
-
 static inline int freezeable(struct task_struct * p)
 {
 	if ((p == current) ||
@@ -84,63 +81,53 @@ static void fake_signal_wake_up(struct task_struct *p)
 	spin_unlock_irqrestore(&p->sighand->siglock, flags);
 }
 
-static int has_mm(struct task_struct *p)
+static inline bool should_send_signal(struct task_struct *p)
 {
-	return (p->mm && !(p->flags & PF_BORROWED_MM));
+	return !(p->flags & PF_FREEZER_NOSIG);
 }
 
 /**
  *	freeze_task - send a freeze request to given task
  *	@p: task to send the request to
- *	@with_mm_only: if set, the request will only be sent if the task has its
- *		own mm
- *	Return value: 0, if @with_mm_only is set and the task has no mm of its
- *		own or the task is frozen, 1, otherwise
+ *	@sig_only: if set, the request will only be sent if the task has the
+ *		PF_FREEZER_NOSIG flag unset
+ *	Return value: 'false', if @sig_only is set and the task has
+ *		PF_FREEZER_NOSIG set or the task is frozen, 'true', otherwise
  *
- *	The freeze request is sent by seting the tasks's TIF_FREEZE flag and
+ *	The freeze request is sent by setting the tasks's TIF_FREEZE flag and
  *	either sending a fake signal to it or waking it up, depending on whether
- *	or not it has its own mm (ie. it is a user land task).  If @with_mm_only
- *	is set and the task has no mm of its own (ie. it is a kernel thread),
- *	its TIF_FREEZE flag should not be set.
- *
- *	The task_lock() is necessary to prevent races with exit_mm() or
- *	use_mm()/unuse_mm() from occuring.
+ *	or not it has PF_FREEZER_NOSIG set.  If @sig_only is set and the task
+ *	has PF_FREEZER_NOSIG set (ie. it is a typical kernel thread), its
+ *	TIF_FREEZE flag will not be set.
  */
-static int freeze_task(struct task_struct *p, int with_mm_only)
+static bool freeze_task(struct task_struct *p, bool sig_only)
 {
-	int ret = 1;
+	/*
+	 * We first check if the task is freezing and next if it has already
+	 * been frozen to avoid the race with frozen_process() which first marks
+	 * the task as frozen and next clears its TIF_FREEZE.
+	 */
+	if (!freezing(p)) {
+		rmb();
+		if (frozen(p))
+			return false;
 
-	task_lock(p);
-	if (freezing(p)) {
-		if (has_mm(p)) {
-			if (!signal_pending(p))
-				fake_signal_wake_up(p);
-		} else {
-			if (with_mm_only)
-				ret = 0;
-			else
-				wake_up_state(p, TASK_INTERRUPTIBLE);
-		}
+		if (!sig_only || should_send_signal(p))
+			set_freeze_flag(p);
+		else
+			return false;
+	}
+
+	if (should_send_signal(p)) {
+		if (!signal_pending(p))
+			fake_signal_wake_up(p);
+	} else if (sig_only) {
+		return false;
 	} else {
-		rmb();
-		if (frozen(p)) {
-			ret = 0;
-		} else {
-			if (has_mm(p)) {
-				set_freeze_flag(p);
-				fake_signal_wake_up(p);
-			} else {
-				if (with_mm_only) {
-					ret = 0;
-				} else {
-					set_freeze_flag(p);
-					wake_up_state(p, TASK_INTERRUPTIBLE);
-				}
-			}
-		}
+		wake_up_state(p, TASK_INTERRUPTIBLE);
 	}
-	task_unlock(p);
-	return ret;
+
+	return true;
 }
 
 static void cancel_freezing(struct task_struct *p)
@@ -156,7 +143,7 @@ static void cancel_freezing(struct task_struct *p)
 	}
 }
 
-static int try_to_freeze_tasks(int freeze_user_space)
+static int try_to_freeze_tasks(bool sig_only)
 {
 	struct task_struct *g, *p;
 	unsigned long end_time;
@@ -175,7 +162,7 @@ static int try_to_freeze_tasks(int freeze_user_space)
 			if (frozen(p) || !freezeable(p))
 				continue;
 
-			if (!freeze_task(p, freeze_user_space))
+			if (!freeze_task(p, sig_only))
 				continue;
 
 			/*
@@ -235,13 +222,13 @@ int freeze_processes(void)
 	int error;
 
 	printk("Freezing user space processes ... ");
-	error = try_to_freeze_tasks(FREEZER_USER_SPACE);
+	error = try_to_freeze_tasks(true);
 	if (error)
 		goto Exit;
 	printk("done.\n");
 
 	printk("Freezing remaining freezable tasks ... ");
-	error = try_to_freeze_tasks(FREEZER_KERNEL_THREADS);
+	error = try_to_freeze_tasks(false);
 	if (error)
 		goto Exit;
 	printk("done.");
@@ -251,7 +238,7 @@ int freeze_processes(void)
 	return error;
 }
 
-static void thaw_tasks(int thaw_user_space)
+static void thaw_tasks(bool nosig_only)
 {
 	struct task_struct *g, *p;
 
@@ -260,7 +247,7 @@ static void thaw_tasks(int thaw_user_space)
 		if (!freezeable(p))
 			continue;
 
-		if (!p->mm == thaw_user_space)
+		if (nosig_only && should_send_signal(p))
 			continue;
 
 		thaw_process(p);
@@ -271,8 +258,8 @@ static void thaw_tasks(int thaw_user_space)
 void thaw_processes(void)
 {
 	printk("Restarting tasks ... ");
-	thaw_tasks(FREEZER_KERNEL_THREADS);
-	thaw_tasks(FREEZER_USER_SPACE);
+	thaw_tasks(true);
+	thaw_tasks(false);
 	schedule();
 	printk("done.\n");
 }
diff --git a/kernel/power/user.c b/kernel/power/user.c
index f5512cb..a6332a3 100644
--- a/kernel/power/user.c
+++ b/kernel/power/user.c
@@ -23,6 +23,7 @@
 #include <linux/console.h>
 #include <linux/cpu.h>
 #include <linux/freezer.h>
+#include <linux/smp_lock.h>
 
 #include <asm/uaccess.h>
 
@@ -69,16 +70,22 @@ static int snapshot_open(struct inode *inode, struct file *filp)
 	struct snapshot_data *data;
 	int error;
 
-	if (!atomic_add_unless(&snapshot_device_available, -1, 0))
-		return -EBUSY;
+	mutex_lock(&pm_mutex);
+
+	if (!atomic_add_unless(&snapshot_device_available, -1, 0)) {
+		error = -EBUSY;
+		goto Unlock;
+	}
 
 	if ((filp->f_flags & O_ACCMODE) == O_RDWR) {
 		atomic_inc(&snapshot_device_available);
-		return -ENOSYS;
+		error = -ENOSYS;
+		goto Unlock;
 	}
 	if(create_basic_memory_bitmaps()) {
 		atomic_inc(&snapshot_device_available);
-		return -ENOMEM;
+		error = -ENOMEM;
+		goto Unlock;
 	}
 	nonseekable_open(inode, filp);
 	data = &snapshot_state;
@@ -98,33 +105,36 @@ static int snapshot_open(struct inode *inode, struct file *filp)
 		if (error)
 			pm_notifier_call_chain(PM_POST_HIBERNATION);
 	}
-	if (error) {
+	if (error)
 		atomic_inc(&snapshot_device_available);
-		return error;
-	}
 	data->frozen = 0;
 	data->ready = 0;
 	data->platform_support = 0;
 
-	return 0;
+ Unlock:
+	mutex_unlock(&pm_mutex);
+
+	return error;
 }
 
 static int snapshot_release(struct inode *inode, struct file *filp)
 {
 	struct snapshot_data *data;
 
+	mutex_lock(&pm_mutex);
+
 	swsusp_free();
 	free_basic_memory_bitmaps();
 	data = filp->private_data;
 	free_all_swap_pages(data->swap);
-	if (data->frozen) {
-		mutex_lock(&pm_mutex);
+	if (data->frozen)
 		thaw_processes();
-		mutex_unlock(&pm_mutex);
-	}
 	pm_notifier_call_chain(data->mode == O_WRONLY ?
 			PM_POST_HIBERNATION : PM_POST_RESTORE);
 	atomic_inc(&snapshot_device_available);
+
+	mutex_unlock(&pm_mutex);
+
 	return 0;
 }
 
@@ -134,9 +144,13 @@ static ssize_t snapshot_read(struct file *filp, char __user *buf,
 	struct snapshot_data *data;
 	ssize_t res;
 
+	mutex_lock(&pm_mutex);
+
 	data = filp->private_data;
-	if (!data->ready)
-		return -ENODATA;
+	if (!data->ready) {
+		res = -ENODATA;
+		goto Unlock;
+	}
 	res = snapshot_read_next(&data->handle, count);
 	if (res > 0) {
 		if (copy_to_user(buf, data_of(data->handle), res))
@@ -144,6 +158,10 @@ static ssize_t snapshot_read(struct file *filp, char __user *buf,
 		else
 			*offp = data->handle.offset;
 	}
+
+ Unlock:
+	mutex_unlock(&pm_mutex);
+
 	return res;
 }
 
@@ -153,6 +171,8 @@ static ssize_t snapshot_write(struct file *filp, const char __user *buf,
 	struct snapshot_data *data;
 	ssize_t res;
 
+	mutex_lock(&pm_mutex);
+
 	data = filp->private_data;
 	res = snapshot_write_next(&data->handle, count);
 	if (res > 0) {
@@ -161,11 +181,14 @@ static ssize_t snapshot_write(struct file *filp, const char __user *buf,
 		else
 			*offp = data->handle.offset;
 	}
+
+	mutex_unlock(&pm_mutex);
+
 	return res;
 }
 
-static int snapshot_ioctl(struct inode *inode, struct file *filp,
-                          unsigned int cmd, unsigned long arg)
+static long snapshot_ioctl(struct file *filp, unsigned int cmd,
+							unsigned long arg)
 {
 	int error = 0;
 	struct snapshot_data *data;
@@ -179,6 +202,9 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp,
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
 
+	if (!mutex_trylock(&pm_mutex))
+		return -EBUSY;
+
 	data = filp->private_data;
 
 	switch (cmd) {
@@ -186,7 +212,6 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp,
 	case SNAPSHOT_FREEZE:
 		if (data->frozen)
 			break;
-		mutex_lock(&pm_mutex);
 		printk("Syncing filesystems ... ");
 		sys_sync();
 		printk("done.\n");
@@ -194,7 +219,6 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp,
 		error = freeze_processes();
 		if (error)
 			thaw_processes();
-		mutex_unlock(&pm_mutex);
 		if (!error)
 			data->frozen = 1;
 		break;
@@ -202,9 +226,7 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp,
 	case SNAPSHOT_UNFREEZE:
 		if (!data->frozen || data->ready)
 			break;
-		mutex_lock(&pm_mutex);
 		thaw_processes();
-		mutex_unlock(&pm_mutex);
 		data->frozen = 0;
 		break;
 
@@ -307,16 +329,11 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp,
 			error = -EPERM;
 			break;
 		}
-		if (!mutex_trylock(&pm_mutex)) {
-			error = -EBUSY;
-			break;
-		}
 		/*
 		 * Tasks are frozen and the notifiers have been called with
 		 * PM_HIBERNATION_PREPARE
 		 */
 		error = suspend_devices_and_enter(PM_SUSPEND_MEM);
-		mutex_unlock(&pm_mutex);
 		break;
 
 	case SNAPSHOT_PLATFORM_SUPPORT:
@@ -390,6 +407,8 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp,
 
 	}
 
+	mutex_unlock(&pm_mutex);
+
 	return error;
 }
 
@@ -399,7 +418,7 @@ static const struct file_operations snapshot_fops = {
 	.read = snapshot_read,
 	.write = snapshot_write,
 	.llseek = no_llseek,
-	.ioctl = snapshot_ioctl,
+	.unlocked_ioctl = snapshot_ioctl,
 };
 
 static struct miscdevice snapshot_device = {