Merge ../linus

Conflicts:

	drivers/cpufreq/cpufreq.c

10 files changed, 1168 insertions, 521 deletions

diff --git a/kernel/power/Kconfig b/kernel/power/Kconfig
index 825068ca347..710ed084e7c 100644
--- a/kernel/power/Kconfig
+++ b/kernel/power/Kconfig
@@ -78,7 +78,7 @@ config PM_SYSFS_DEPRECATED
 
 config SOFTWARE_SUSPEND
 	bool "Software Suspend"
-	depends on PM && SWAP && ((X86 && (!SMP || SUSPEND_SMP) && !X86_PAE) || ((FRV || PPC32) && !SMP))
+	depends on PM && SWAP && ((X86 && (!SMP || SUSPEND_SMP)) || ((FRV || PPC32) && !SMP))
 	---help---
 	  Enable the possibility of suspending the machine.
 	  It doesn't need ACPI or APM.
diff --git a/kernel/power/disk.c b/kernel/power/disk.c
index d3a158a6031..0b00f56c2ad 100644
--- a/kernel/power/disk.c
+++ b/kernel/power/disk.c
@@ -20,6 +20,7 @@
 #include <linux/pm.h>
 #include <linux/console.h>
 #include <linux/cpu.h>
+#include <linux/freezer.h>
 
 #include "power.h"
 
@@ -27,6 +28,23 @@
 static int noresume = 0;
 char resume_file[256] = CONFIG_PM_STD_PARTITION;
 dev_t swsusp_resume_device;
+sector_t swsusp_resume_block;
+
+/**
+ *	platform_prepare - prepare the machine for hibernation using the
+ *	platform driver if so configured and return an error code if it fails
+ */
+
+static inline int platform_prepare(void)
+{
+	int error = 0;
+
+	if (pm_disk_mode == PM_DISK_PLATFORM) {
+		if (pm_ops && pm_ops->prepare)
+			error = pm_ops->prepare(PM_SUSPEND_DISK);
+	}
+	return error;
+}
 
 /**
  *	power_down - Shut machine down for hibernate.
@@ -40,12 +58,10 @@ dev_t swsusp_resume_device;
 
 static void power_down(suspend_disk_method_t mode)
 {
-	int error = 0;
-
 	switch(mode) {
 	case PM_DISK_PLATFORM:
 		kernel_shutdown_prepare(SYSTEM_SUSPEND_DISK);
-		error = pm_ops->enter(PM_SUSPEND_DISK);
+		pm_ops->enter(PM_SUSPEND_DISK);
 		break;
 	case PM_DISK_SHUTDOWN:
 		kernel_power_off();
@@ -71,7 +87,7 @@ static inline void platform_finish(void)
 
 static int prepare_processes(void)
 {
-	int error;
+	int error = 0;
 
 	pm_prepare_console();
 
@@ -84,12 +100,24 @@ static int prepare_processes(void)
 		goto thaw;
 	}
 
+	if (pm_disk_mode == PM_DISK_TESTPROC) {
+		printk("swsusp debug: Waiting for 5 seconds.\n");
+		mdelay(5000);
+		goto thaw;
+	}
+
+	error = platform_prepare();
+	if (error)
+		goto thaw;
+
 	/* Free memory before shutting down devices. */
 	if (!(error = swsusp_shrink_memory()))
 		return 0;
-thaw:
+
+	platform_finish();
+ thaw:
 	thaw_processes();
-enable_cpus:
+ enable_cpus:
 	enable_nonboot_cpus();
 	pm_restore_console();
 	return error;
@@ -120,13 +148,21 @@ int pm_suspend_disk(void)
 	if (error)
 		return error;
 
+	if (pm_disk_mode == PM_DISK_TESTPROC)
+		return 0;
+
 	suspend_console();
 	error = device_suspend(PMSG_FREEZE);
 	if (error) {
 		resume_console();
 		printk("Some devices failed to suspend\n");
-		unprepare_processes();
-		return error;
+		goto Thaw;
+	}
+
+	if (pm_disk_mode == PM_DISK_TEST) {
+		printk("swsusp debug: Waiting for 5 seconds.\n");
+		mdelay(5000);
+		goto Done;
 	}
 
 	pr_debug("PM: snapshotting memory.\n");
@@ -143,16 +179,17 @@ int pm_suspend_disk(void)
 			power_down(pm_disk_mode);
 		else {
 			swsusp_free();
-			unprepare_processes();
-			return error;
+			goto Thaw;
 		}
-	} else
+	} else {
 		pr_debug("PM: Image restored successfully.\n");
+	}
 
 	swsusp_free();
  Done:
 	device_resume();
 	resume_console();
+ Thaw:
 	unprepare_processes();
 	return error;
 }
@@ -174,10 +211,10 @@ static int software_resume(void)
 {
 	int error;
 
-	down(&pm_sem);
+	mutex_lock(&pm_mutex);
 	if (!swsusp_resume_device) {
 		if (!strlen(resume_file)) {
-			up(&pm_sem);
+			mutex_unlock(&pm_mutex);
 			return -ENOENT;
 		}
 		swsusp_resume_device = name_to_dev_t(resume_file);
@@ -192,7 +229,7 @@ static int software_resume(void)
 		 * FIXME: If noresume is specified, we need to find the partition
 		 * and reset it back to normal swap space.
 		 */
-		up(&pm_sem);
+		mutex_unlock(&pm_mutex);
 		return 0;
 	}
 
@@ -236,7 +273,7 @@ static int software_resume(void)
 	unprepare_processes();
  Done:
 	/* For success case, the suspend path will release the lock */
-	up(&pm_sem);
+	mutex_unlock(&pm_mutex);
 	pr_debug("PM: Resume from disk failed.\n");
 	return 0;
 }
@@ -249,6 +286,8 @@ static const char * const pm_disk_modes[] = {
 	[PM_DISK_PLATFORM]	= "platform",
 	[PM_DISK_SHUTDOWN]	= "shutdown",
 	[PM_DISK_REBOOT]	= "reboot",
+	[PM_DISK_TEST]		= "test",
+	[PM_DISK_TESTPROC]	= "testproc",
 };
 
 /**
@@ -295,7 +334,7 @@ static ssize_t disk_store(struct subsystem * s, const char * buf, size_t n)
 	p = memchr(buf, '\n', n);
 	len = p ? p - buf : n;
 
-	down(&pm_sem);
+	mutex_lock(&pm_mutex);
 	for (i = PM_DISK_FIRMWARE; i < PM_DISK_MAX; i++) {
 		if (!strncmp(buf, pm_disk_modes[i], len)) {
 			mode = i;
@@ -303,21 +342,23 @@ static ssize_t disk_store(struct subsystem * s, const char * buf, size_t n)
 		}
 	}
 	if (mode) {
-		if (mode == PM_DISK_SHUTDOWN || mode == PM_DISK_REBOOT)
+		if (mode == PM_DISK_SHUTDOWN || mode == PM_DISK_REBOOT ||
+		     mode == PM_DISK_TEST || mode == PM_DISK_TESTPROC) {
 			pm_disk_mode = mode;
-		else {
+		} else {
 			if (pm_ops && pm_ops->enter &&
 			    (mode == pm_ops->pm_disk_mode))
 				pm_disk_mode = mode;
 			else
 				error = -EINVAL;
 		}
-	} else
+	} else {
 		error = -EINVAL;
+	}
 
 	pr_debug("PM: suspend-to-disk mode set to '%s'\n",
 		 pm_disk_modes[mode]);
-	up(&pm_sem);
+	mutex_unlock(&pm_mutex);
 	return error ? error : n;
 }
 
@@ -342,14 +383,14 @@ static ssize_t resume_store(struct subsystem *subsys, const char *buf, size_t n)
 	if (maj != MAJOR(res) || min != MINOR(res))
 		goto out;
 
-	down(&pm_sem);
+	mutex_lock(&pm_mutex);
 	swsusp_resume_device = res;
-	up(&pm_sem);
+	mutex_unlock(&pm_mutex);
 	printk("Attempting manual resume\n");
 	noresume = 0;
 	software_resume();
 	ret = n;
-out:
+ out:
 	return ret;
 }
 
@@ -404,6 +445,19 @@ static int __init resume_setup(char *str)
 	return 1;
 }
 
+static int __init resume_offset_setup(char *str)
+{
+	unsigned long long offset;
+
+	if (noresume)
+		return 1;
+
+	if (sscanf(str, "%llu", &offset) == 1)
+		swsusp_resume_block = offset;
+
+	return 1;
+}
+
 static int __init noresume_setup(char *str)
 {
 	noresume = 1;
@@ -411,4 +465,5 @@ static int __init noresume_setup(char *str)
 }
 
 __setup("noresume", noresume_setup);
+__setup("resume_offset=", resume_offset_setup);
 __setup("resume=", resume_setup);
diff --git a/kernel/power/main.c b/kernel/power/main.c
index 873228c71da..500eb87f643 100644
--- a/kernel/power/main.c
+++ b/kernel/power/main.c
@@ -8,6 +8,7 @@
  *
  */
 
+#include <linux/module.h>
 #include <linux/suspend.h>
 #include <linux/kobject.h>
 #include <linux/string.h>
@@ -18,13 +19,14 @@
 #include <linux/console.h>
 #include <linux/cpu.h>
 #include <linux/resume-trace.h>
+#include <linux/freezer.h>
 
 #include "power.h"
 
 /*This is just an arbitrary number */
 #define FREE_PAGE_NUMBER (100)
 
-DECLARE_MUTEX(pm_sem);
+DEFINE_MUTEX(pm_mutex);
 
 struct pm_ops *pm_ops;
 suspend_disk_method_t pm_disk_mode = PM_DISK_SHUTDOWN;
@@ -36,9 +38,9 @@ suspend_disk_method_t pm_disk_mode = PM_DISK_SHUTDOWN;
 
 void pm_set_ops(struct pm_ops * ops)
 {
-	down(&pm_sem);
+	mutex_lock(&pm_mutex);
 	pm_ops = ops;
-	up(&pm_sem);
+	mutex_unlock(&pm_mutex);
 }
 
 
@@ -182,7 +184,7 @@ static int enter_state(suspend_state_t state)
 
 	if (!valid_state(state))
 		return -ENODEV;
-	if (down_trylock(&pm_sem))
+	if (!mutex_trylock(&pm_mutex))
 		return -EBUSY;
 
 	if (state == PM_SUSPEND_DISK) {
@@ -200,7 +202,7 @@ static int enter_state(suspend_state_t state)
 	pr_debug("PM: Finishing wakeup.\n");
 	suspend_finish(state);
  Unlock:
-	up(&pm_sem);
+	mutex_unlock(&pm_mutex);
 	return error;
 }
 
@@ -229,7 +231,7 @@ int pm_suspend(suspend_state_t state)
 	return -EINVAL;
 }
 
-
+EXPORT_SYMBOL(pm_suspend);
 
 decl_subsys(power,NULL,NULL);
 
diff --git a/kernel/power/power.h b/kernel/power/power.h
index bfe999f7b27..eb461b816bf 100644
--- a/kernel/power/power.h
+++ b/kernel/power/power.h
@@ -22,7 +22,9 @@ static inline int pm_suspend_disk(void)
 	return -EPERM;
 }
 #endif
-extern struct semaphore pm_sem;
+
+extern struct mutex pm_mutex;
+
 #define power_attr(_name) \
 static struct subsys_attribute _name##_attr = {	\
 	.attr	= {				\
@@ -42,6 +44,7 @@ extern const void __nosave_begin, __nosave_end;
 extern unsigned long image_size;
 extern int in_suspend;
 extern dev_t swsusp_resume_device;
+extern sector_t swsusp_resume_block;
 
 extern asmlinkage int swsusp_arch_suspend(void);
 extern asmlinkage int swsusp_arch_resume(void);
@@ -102,8 +105,18 @@ struct snapshot_handle {
 extern unsigned int snapshot_additional_pages(struct zone *zone);
 extern int snapshot_read_next(struct snapshot_handle *handle, size_t count);
 extern int snapshot_write_next(struct snapshot_handle *handle, size_t count);
+extern void snapshot_write_finalize(struct snapshot_handle *handle);
 extern int snapshot_image_loaded(struct snapshot_handle *handle);
-extern void snapshot_free_unused_memory(struct snapshot_handle *handle);
+
+/*
+ * This structure is used to pass the values needed for the identification
+ * of the resume swap area from a user space to the kernel via the
+ * SNAPSHOT_SET_SWAP_AREA ioctl
+ */
+struct resume_swap_area {
+	loff_t offset;
+	u_int32_t dev;
+} __attribute__((packed));
 
 #define SNAPSHOT_IOC_MAGIC	'3'
 #define SNAPSHOT_FREEZE			_IO(SNAPSHOT_IOC_MAGIC, 1)
@@ -117,7 +130,14 @@ extern void snapshot_free_unused_memory(struct snapshot_handle *handle);
 #define SNAPSHOT_FREE_SWAP_PAGES	_IO(SNAPSHOT_IOC_MAGIC, 9)
 #define SNAPSHOT_SET_SWAP_FILE		_IOW(SNAPSHOT_IOC_MAGIC, 10, unsigned int)
 #define SNAPSHOT_S2RAM			_IO(SNAPSHOT_IOC_MAGIC, 11)
-#define SNAPSHOT_IOC_MAXNR	11
+#define SNAPSHOT_PMOPS			_IOW(SNAPSHOT_IOC_MAGIC, 12, unsigned int)
+#define SNAPSHOT_SET_SWAP_AREA		_IOW(SNAPSHOT_IOC_MAGIC, 13, \
+							struct resume_swap_area)
+#define SNAPSHOT_IOC_MAXNR	13
+
+#define PMOPS_PREPARE	1
+#define PMOPS_ENTER	2
+#define PMOPS_FINISH	3
 
 /**
  *	The bitmap is used for tracing allocated swap pages
@@ -141,7 +161,7 @@ struct bitmap_page {
 
 extern void free_bitmap(struct bitmap_page *bitmap);
 extern struct bitmap_page *alloc_bitmap(unsigned int nr_bits);
-extern unsigned long alloc_swap_page(int swap, struct bitmap_page *bitmap);
+extern sector_t alloc_swapdev_block(int swap, struct bitmap_page *bitmap);
 extern void free_all_swap_pages(int swap, struct bitmap_page *bitmap);
 
 extern int swsusp_check(void);
@@ -153,3 +173,7 @@ extern int swsusp_read(void);
 extern int swsusp_write(void);
 extern void swsusp_close(void);
 extern int suspend_enter(suspend_state_t state);
+
+struct timeval;
+extern void swsusp_show_speed(struct timeval *, struct timeval *,
+				unsigned int, char *);
diff --git a/kernel/power/poweroff.c b/kernel/power/poweroff.c
index f1f900ac316..678ec736076 100644
--- a/kernel/power/poweroff.c
+++ b/kernel/power/poweroff.c
@@ -16,12 +16,12 @@
  * callback we use.
  */
 
-static void do_poweroff(void *dummy)
+static void do_poweroff(struct work_struct *dummy)
 {
 	kernel_power_off();
 }
 
-static DECLARE_WORK(poweroff_work, do_poweroff, NULL);
+static DECLARE_WORK(poweroff_work, do_poweroff);
 
 static void handle_poweroff(int key, struct tty_struct *tty)
 {
diff --git a/kernel/power/process.c b/kernel/power/process.c
index 72e72d2c61e..99eeb119b06 100644
--- a/kernel/power/process.c
+++ b/kernel/power/process.c
@@ -13,12 +13,15 @@
 #include <linux/suspend.h>
 #include <linux/module.h>
 #include <linux/syscalls.h>
+#include <linux/freezer.h>
 
 /* 
  * Timeout for stopping processes
  */
 #define TIMEOUT	(20 * HZ)
 
+#define FREEZER_KERNEL_THREADS 0
+#define FREEZER_USER_SPACE 1
 
 static inline int freezeable(struct task_struct * p)
 {
@@ -39,7 +42,6 @@ void refrigerator(void)
 	long save;
 	save = current->state;
 	pr_debug("%s entered refrigerator\n", current->comm);
-	printk("=");
 
 	frozen_process(current);
 	spin_lock_irq(&current->sighand->siglock);
@@ -79,96 +81,136 @@ static void cancel_freezing(struct task_struct *p)
 	}
 }
 
-/* 0 = success, else # of processes that we failed to stop */
-int freeze_processes(void)
+static inline int is_user_space(struct task_struct *p)
+{
+	return p->mm && !(p->flags & PF_BORROWED_MM);
+}
+
+static unsigned int try_to_freeze_tasks(int freeze_user_space)
 {
-	int todo, nr_user, user_frozen;
-	unsigned long start_time;
 	struct task_struct *g, *p;
+	unsigned long end_time;
+	unsigned int todo;
 
-	printk( "Stopping tasks: " );
-	start_time = jiffies;
-	user_frozen = 0;
+	end_time = jiffies + TIMEOUT;
 	do {
-		nr_user = todo = 0;
+		todo = 0;
 		read_lock(&tasklist_lock);
 		do_each_thread(g, p) {
 			if (!freezeable(p))
 				continue;
+
 			if (frozen(p))
 				continue;
-			if (p->state == TASK_TRACED && frozen(p->parent)) {
+
+			if (p->state == TASK_TRACED &&
+			    (frozen(p->parent) ||
+			     p->parent->state == TASK_STOPPED)) {
 				cancel_freezing(p);
 				continue;
 			}
-			if (p->mm && !(p->flags & PF_BORROWED_MM)) {
-				/* The task is a user-space one.
-				 * Freeze it unless there's a vfork completion
-				 * pending
+			if (is_user_space(p)) {
+				if (!freeze_user_space)
+					continue;
+
+				/* Freeze the task unless there is a vfork
+				 * completion pending
 				 */
 				if (!p->vfork_done)
 					freeze_process(p);
-				nr_user++;
 			} else {
-				/* Freeze only if the user space is frozen */
-				if (user_frozen)
-					freeze_process(p);
-				todo++;
+				if (freeze_user_space)
+					continue;
+
+				freeze_process(p);
 			}
+			todo++;
 		} while_each_thread(g, p);
 		read_unlock(&tasklist_lock);
-		todo += nr_user;
-		if (!user_frozen && !nr_user) {
-			sys_sync();
-			start_time = jiffies;
-		}
-		user_frozen = !nr_user;
 		yield();			/* Yield is okay here */
-		if (todo && time_after(jiffies, start_time + TIMEOUT))
+		if (todo && time_after(jiffies, end_time))
 			break;
-	} while(todo);
+	} while (todo);
 
-	/* This does not unfreeze processes that are already frozen
-	 * (we have slightly ugly calling convention in that respect,
-	 * and caller must call thaw_processes() if something fails),
-	 * but it cleans up leftover PF_FREEZE requests.
-	 */
 	if (todo) {
-		printk( "\n" );
-		printk(KERN_ERR " stopping tasks timed out "
-			"after %d seconds (%d tasks remaining):\n",
-			TIMEOUT / HZ, todo);
+		/* This does not unfreeze processes that are already frozen
+		 * (we have slightly ugly calling convention in that respect,
+		 * and caller must call thaw_processes() if something fails),
+		 * but it cleans up leftover PF_FREEZE requests.
+		 */
+		printk("\n");
+		printk(KERN_ERR "Stopping %s timed out after %d seconds "
+				"(%d tasks refusing to freeze):\n",
+				freeze_user_space ? "user space processes" :
+					"kernel threads",
+				TIMEOUT / HZ, todo);
 		read_lock(&tasklist_lock);
 		do_each_thread(g, p) {
+			if (is_user_space(p) == !freeze_user_space)
+				continue;
+
 			if (freezeable(p) && !frozen(p))
-				printk(KERN_ERR "  %s\n", p->comm);
+				printk(KERN_ERR " %s\n", p->comm);
+
 			cancel_freezing(p);
 		} while_each_thread(g, p);
 		read_unlock(&tasklist_lock);
-		return todo;
 	}
 
-	printk( "|\n" );
+	return todo;
+}
+
+/**
+ *	freeze_processes - tell processes to enter the refrigerator
+ *
+ *	Returns 0 on success, or the number of processes that didn't freeze,
+ *	although they were told to.
+ */
+int freeze_processes(void)
+{
+	unsigned int nr_unfrozen;
+
+	printk("Stopping tasks ... ");
+	nr_unfrozen = try_to_freeze_tasks(FREEZER_USER_SPACE);
+	if (nr_unfrozen)
+		return nr_unfrozen;
+
+	sys_sync();
+	nr_unfrozen = try_to_freeze_tasks(FREEZER_KERNEL_THREADS);
+	if (nr_unfrozen)
+		return nr_unfrozen;
+
+	printk("done.\n");
 	BUG_ON(in_atomic());
 	return 0;
 }
 
-void thaw_processes(void)
+static void thaw_tasks(int thaw_user_space)
 {
 	struct task_struct *g, *p;
 
-	printk( "Restarting tasks..." );
 	read_lock(&tasklist_lock);
 	do_each_thread(g, p) {
 		if (!freezeable(p))
 			continue;
+
+		if (is_user_space(p) == !thaw_user_space)
+			continue;
+
 		if (!thaw_process(p))
-			printk(KERN_INFO " Strange, %s not stopped\n", p->comm );
+			printk(KERN_WARNING " Strange, %s not stopped\n",
+				p->comm );
 	} while_each_thread(g, p);
-
 	read_unlock(&tasklist_lock);
+}
+
+void thaw_processes(void)
+{
+	printk("Restarting tasks ... ");
+	thaw_tasks(FREEZER_KERNEL_THREADS);
+	thaw_tasks(FREEZER_USER_SPACE);
 	schedule();
-	printk( " done\n" );
+	printk("done.\n");
 }
 
 EXPORT_SYMBOL(refrigerator);
diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c
index 99f9b7d177d..c024606221c 100644
--- a/kernel/power/snapshot.c
+++ b/kernel/power/snapshot.c
@@ -1,15 +1,15 @@
 /*
  * linux/kernel/power/snapshot.c
  *
- * This file provide system snapshot/restore functionality.
+ * This file provides system snapshot/restore functionality for swsusp.
  *
  * Copyright (C) 1998-2005 Pavel Machek <pavel@suse.cz>
+ * Copyright (C) 2006 Rafael J. Wysocki <rjw@sisk.pl>
  *
- * This file is released under the GPLv2, and is based on swsusp.c.
+ * This file is released under the GPLv2.
  *
  */
 
-
 #include <linux/version.h>
 #include <linux/module.h>
 #include <linux/mm.h>
@@ -34,137 +34,24 @@
 
 #include "power.h"
 
-/* List of PBEs used for creating and restoring the suspend image */
+/* List of PBEs needed for restoring the pages that were allocated before
+ * the suspend and included in the suspend image, but have also been
+ * allocated by the "resume" kernel, so their contents cannot be written
+ * directly to their "original" page frames.
+ */
 struct pbe *restore_pblist;
 
-static unsigned int nr_copy_pages;
-static unsigned int nr_meta_pages;
+/* Pointer to an auxiliary buffer (1 page) */
 static void *buffer;
 
-#ifdef CONFIG_HIGHMEM
-unsigned int count_highmem_pages(void)
-{
-	struct zone *zone;
-	unsigned long zone_pfn;
-	unsigned int n = 0;
-
-	for_each_zone (zone)
-		if (is_highmem(zone)) {
-			mark_free_pages(zone);
-			for (zone_pfn = 0; zone_pfn < zone->spanned_pages; zone_pfn++) {
-				struct page *page;
-				unsigned long pfn = zone_pfn + zone->zone_start_pfn;
-				if (!pfn_valid(pfn))
-					continue;
-				page = pfn_to_page(pfn);
-				if (PageReserved(page))
-					continue;
-				if (PageNosaveFree(page))
-					continue;
-				n++;
-			}
-		}
-	return n;
-}
-
-struct highmem_page {
-	char *data;
-	struct page *page;
-	struct highmem_page *next;
-};
-
-static struct highmem_page *highmem_copy;
-
-static int save_highmem_zone(struct zone *zone)
-{
-	unsigned long zone_pfn;
-	mark_free_pages(zone);
-	for (zone_pfn = 0; zone_pfn < zone->spanned_pages; ++zone_pfn) {
-		struct page *page;
-		struct highmem_page *save;
-		void *kaddr;
-		unsigned long pfn = zone_pfn + zone->zone_start_pfn;
-
-		if (!(pfn%10000))
-			printk(".");
-		if (!pfn_valid(pfn))
-			continue;
-		page = pfn_to_page(pfn);
-		/*
-		 * This condition results from rvmalloc() sans vmalloc_32()
-		 * and architectural memory reservations. This should be
-		 * corrected eventually when the cases giving rise to this
-		 * are better understood.
-		 */
-		if (PageReserved(page))
-			continue;
-		BUG_ON(PageNosave(page));
-		if (PageNosaveFree(page))
-			continue;
-		save = kmalloc(sizeof(struct highmem_page), GFP_ATOMIC);
-		if (!save)
-			return -ENOMEM;
-		save->next = highmem_copy;
-		save->page = page;
-		save->data = (void *) get_zeroed_page(GFP_ATOMIC);
-		if (!save->data) {
-			kfree(save);
-			return -ENOMEM;
-		}
-		kaddr = kmap_atomic(page, KM_USER0);
-		memcpy(save->data, kaddr, PAGE_SIZE);
-		kunmap_atomic(kaddr, KM_USER0);
-		highmem_copy = save;
-	}
-	return 0;
-}
-
-int save_highmem(void)
-{
-	struct zone *zone;
-	int res = 0;
-
-	pr_debug("swsusp: Saving Highmem");
-	drain_local_pages();
-	for_each_zone (zone) {
-		if (is_highmem(zone))
-			res = save_highmem_zone(zone);
-		if (res)
-			return res;
-	}
-	printk("\n");
-	return 0;
-}
-
-int restore_highmem(void)
-{
-	printk("swsusp: Restoring Highmem\n");
-	while (highmem_copy) {
-		struct highmem_page *save = highmem_copy;
-		void *kaddr;
-		highmem_copy = save->next;
-
-		kaddr = kmap_atomic(save->page, KM_USER0);
-		memcpy(kaddr, save->data, PAGE_SIZE);
-		kunmap_atomic(kaddr, KM_USER0);
-		free_page((long) save->data);
-		kfree(save);
-	}
-	return 0;
-}
-#else
-static inline unsigned int count_highmem_pages(void) {return 0;}
-static inline int save_highmem(void) {return 0;}
-static inline int restore_highmem(void) {return 0;}
-#endif
-
 /**
  *	@safe_needed - on resume, for storing the PBE list and the image,
  *	we can only use memory pages that do not conflict with the pages
- *	used before suspend.
+ *	used before suspend.  The unsafe pages have PageNosaveFree set
+ *	and we count them using unsafe_pages.
  *
- *	The unsafe pages are marked with the PG_nosave_free flag
- *	and we count them using unsafe_pages
+ *	Each allocated image page is marked as PageNosave and PageNosaveFree
+ *	so that swsusp_free() can release it.
  */
 
 #define PG_ANY		0
@@ -174,7 +61,7 @@ static inline int restore_highmem(void) {return 0;}
 
 static unsigned int allocated_unsafe_pages;
 
-static void *alloc_image_page(gfp_t gfp_mask, int safe_needed)
+static void *get_image_page(gfp_t gfp_mask, int safe_needed)
 {
 	void *res;
 
@@ -195,20 +82,39 @@ static void *alloc_image_page(gfp_t gfp_mask, int safe_needed)
 
 unsigned long get_safe_page(gfp_t gfp_mask)
 {
-	return (unsigned long)alloc_image_page(gfp_mask, PG_SAFE);
+	return (unsigned long)get_image_page(gfp_mask, PG_SAFE);
+}
+
+static struct page *alloc_image_page(gfp_t gfp_mask)
+{
+	struct page *page;
+
+	page = alloc_page(gfp_mask);
+	if (page) {
+		SetPageNosave(page);
+		SetPageNosaveFree(page);
+	}
+	return page;
 }
 
 /**
  *	free_image_page - free page represented by @addr, allocated with
- *	alloc_image_page (page flags set by it must be cleared)
+ *	get_image_page (page flags set by it must be cleared)
  */
 
 static inline void free_image_page(void *addr, int clear_nosave_free)
 {
-	ClearPageNosave(virt_to_page(addr));
+	struct page *page;
+
+	BUG_ON(!virt_addr_valid(addr));
+
+	page = virt_to_page(addr);
+
+	ClearPageNosave(page);
 	if (clear_nosave_free)
-		ClearPageNosaveFree(virt_to_page(addr));
-	free_page((unsigned long)addr);
+		ClearPageNosaveFree(page);
+
+	__free_page(page);
 }
 
 /* struct linked_page is used to build chains of pages */
@@ -269,7 +175,7 @@ static void *chain_alloc(struct chain_allocator *ca, unsigned int size)
 	if (LINKED_PAGE_DATA_SIZE - ca->used_space < size) {
 		struct linked_page *lp;
 
-		lp = alloc_image_page(ca->gfp_mask, ca->safe_needed);
+		lp = get_image_page(ca->gfp_mask, ca->safe_needed);
 		if (!lp)
 			return NULL;
 
@@ -446,8 +352,8 @@ memory_bm_create(struct memory_bitmap *bm, gfp_t gfp_mask, int safe_needed)
 
 	/* Compute the number of zones */
 	nr = 0;
-	for_each_zone (zone)
-		if (populated_zone(zone) && !is_highmem(zone))
+	for_each_zone(zone)
+		if (populated_zone(zone))
 			nr++;
 
 	/* Allocate the list of zones bitmap objects */
@@ -459,10 +365,10 @@ memory_bm_create(struct memory_bitmap *bm, gfp_t gfp_mask, int safe_needed)
 	}
 
 	/* Initialize the zone bitmap objects */
-	for_each_zone (zone) {
+	for_each_zone(zone) {
 		unsigned long pfn;
 
-		if (!populated_zone(zone) || is_highmem(zone))
+		if (!populated_zone(zone))
 			continue;
 
 		zone_bm->start_pfn = zone->zone_start_pfn;
@@ -481,7 +387,7 @@ memory_bm_create(struct memory_bitmap *bm, gfp_t gfp_mask, int safe_needed)
 		while (bb) {
 			unsigned long *ptr;
 
-			ptr = alloc_image_page(gfp_mask, safe_needed);
+			ptr = get_image_page(gfp_mask, safe_needed);
 			bb->data = ptr;
 			if (!ptr)
 				goto Free;
@@ -505,7 +411,7 @@ memory_bm_create(struct memory_bitmap *bm, gfp_t gfp_mask, int safe_needed)
 	memory_bm_position_reset(bm);
 	return 0;
 
-Free:
+ Free:
 	bm->p_list = ca.chain;
 	memory_bm_free(bm, PG_UNSAFE_CLEAR);
 	return -ENOMEM;
@@ -651,7 +557,7 @@ static unsigned long memory_bm_next_pfn(struct memory_bitmap *bm)
 	memory_bm_position_reset(bm);
 	return BM_END_OF_MAP;
 
-Return_pfn:
+ Return_pfn:
 	bm->cur.chunk = chunk;
 	bm->cur.bit = bit;
 	return bb->start_pfn + chunk * BM_BITS_PER_CHUNK + bit;
@@ -669,10 +575,82 @@ unsigned int snapshot_additional_pages(struct zone *zone)
 
 	res = DIV_ROUND_UP(zone->spanned_pages, BM_BITS_PER_BLOCK);
 	res += DIV_ROUND_UP(res * sizeof(struct bm_block), PAGE_SIZE);
-	return res;
+	return 2 * res;
+}
+
+#ifdef CONFIG_HIGHMEM
+/**
+ *	count_free_highmem_pages - compute the total number of free highmem
+ *	pages, system-wide.
+ */
+
+static unsigned int count_free_highmem_pages(void)
+{
+	struct zone *zone;
+	unsigned int cnt = 0;
+
+	for_each_zone(zone)
+		if (populated_zone(zone) && is_highmem(zone))
+			cnt += zone->free_pages;
+
+	return cnt;
+}
+
+/**
+ *	saveable_highmem_page - Determine whether a highmem page should be
+ *	included in the suspend image.
+ *
+ *	We should save the page if it isn't Nosave or NosaveFree, or Reserved,
+ *	and it isn't a part of a free chunk of pages.
+ */
+
+static struct page *saveable_highmem_page(unsigned long pfn)
+{
+	struct page *page;
+
+	if (!pfn_valid(pfn))
+		return NULL;
+
+	page = pfn_to_page(pfn);
+
+	BUG_ON(!PageHighMem(page));
+
+	if (PageNosave(page) || PageReserved(page) || PageNosaveFree(page))
+		return NULL;
+
+	return page;
 }
 
 /**
+ *	count_highmem_pages - compute the total number of saveable highmem
+ *	pages.
+ */
+
+unsigned int count_highmem_pages(void)
+{
+	struct zone *zone;
+	unsigned int n = 0;
+
+	for_each_zone(zone) {
+		unsigned long pfn, max_zone_pfn;
+
+		if (!is_highmem(zone))
+			continue;
+
+		mark_free_pages(zone);
+		max_zone_pfn = zone->zone_start_pfn + zone->spanned_pages;
+		for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++)
+			if (saveable_highmem_page(pfn))
+				n++;
+	}
+	return n;
+}
+#else
+static inline void *saveable_highmem_page(unsigned long pfn) { return NULL; }
+static inline unsigned int count_highmem_pages(void) { return 0; }
+#endif /* CONFIG_HIGHMEM */
+
+/**
  *	pfn_is_nosave - check if given pfn is in the 'nosave' section
  */
 
@@ -684,12 +662,12 @@ static inline int pfn_is_nosave(unsigned long pfn)
 }
 
 /**
- *	saveable - Determine whether a page should be cloned or not.
- *	@pfn:	The page
+ *	saveable - Determine whether a non-highmem page should be included in
+ *	the suspend image.
  *
- *	We save a page if it isn't Nosave, and is not in the range of pages
- *	statically defined as 'unsaveable', and it
- *	isn't a part of a free chunk of pages.
+ *	We should save the page if it isn't Nosave, and is not in the range
+ *	of pages statically defined as 'unsaveable', and it isn't a part of
+ *	a free chunk of pages.
  */
 
 static struct page *saveable_page(unsigned long pfn)
@@ -701,76 +679,130 @@ static struct page *saveable_page(unsigned long pfn)
 
 	page = pfn_to_page(pfn);
 
-	if (PageNosave(page))
+	BUG_ON(PageHighMem(page));
+
+	if (PageNosave(page) || PageNosaveFree(page))
 		return NULL;
+
 	if (PageReserved(page) && pfn_is_nosave(pfn))
 		return NULL;
-	if (PageNosaveFree(page))
-		return NULL;
 
 	return page;
 }
 
+/**
+ *	count_data_pages - compute the total number of saveable non-highmem
+ *	pages.
+ */
+
 unsigned int count_data_pages(void)
 {
 	struct zone *zone;
 	unsigned long pfn, max_zone_pfn;
 	unsigned int n = 0;
 
-	for_each_zone (zone) {
+	for_each_zone(zone) {
 		if (is_highmem(zone))
 			continue;
+
 		mark_free_pages(zone);
 		max_zone_pfn = zone->zone_start_pfn + zone->spanned_pages;
 		for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++)
-			n += !!saveable_page(pfn);
+			if(saveable_page(pfn))
+				n++;
 	}
 	return n;
 }
 
-static inline void copy_data_page(long *dst, long *src)
+/* This is needed, because copy_page and memcpy are not usable for copying
+ * task structs.
+ */
+static inline void do_copy_page(long *dst, long *src)
 {
 	int n;
 
-	/* copy_page and memcpy are not usable for copying task structs. */
 	for (n = PAGE_SIZE / sizeof(long); n; n--)
 		*dst++ = *src++;
 }
 
+#ifdef CONFIG_HIGHMEM
+static inline struct page *
+page_is_saveable(struct zone *zone, unsigned long pfn)
+{
+	return is_highmem(zone) ?
+			saveable_highmem_page(pfn) : saveable_page(pfn);
+}
+
+static inline void
+copy_data_page(unsigned long dst_pfn, unsigned long src_pfn)
+{
+	struct page *s_page, *d_page;
+	void *src, *dst;
+
+	s_page = pfn_to_page(src_pfn);
+	d_page = pfn_to_page(dst_pfn);
+	if (PageHighMem(s_page)) {
+		src = kmap_atomic(s_page, KM_USER0);
+		dst = kmap_atomic(d_page, KM_USER1);
+		do_copy_page(dst, src);
+		kunmap_atomic(src, KM_USER0);
+		kunmap_atomic(dst, KM_USER1);
+	} else {
+		src = page_address(s_page);
+		if (PageHighMem(d_page)) {
+			/* Page pointed to by src may contain some kernel
+			 * data modified by kmap_atomic()
+			 */
+			do_copy_page(buffer, src);
+			dst = kmap_atomic(pfn_to_page(dst_pfn), KM_USER0);
+			memcpy(dst, buffer, PAGE_SIZE);
+			kunmap_atomic(dst, KM_USER0);
+		} else {
+			dst = page_address(d_page);
+			do_copy_page(dst, src);
+		}
+	}
+}
+#else
+#define page_is_saveable(zone, pfn)	saveable_page(pfn)
+
+static inline void
+copy_data_page(unsigned long dst_pfn, unsigned long src_pfn)
+{
+	do_copy_page(page_address(pfn_to_page(dst_pfn)),
+			page_address(pfn_to_page(src_pfn)));
+}
+#endif /* CONFIG_HIGHMEM */
+
 static void
 copy_data_pages(struct memory_bitmap *copy_bm, struct memory_bitmap *orig_bm)
 {
 	struct zone *zone;
 	unsigned long pfn;
 
-	for_each_zone (zone) {
+	for_each_zone(zone) {
 		unsigned long max_zone_pfn;
 
-		if (is_highmem(zone))
-			continue;
-
 		mark_free_pages(zone);
 		max_zone_pfn = zone->zone_start_pfn + zone->spanned_pages;
 		for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++)
-			if (saveable_page(pfn))
+			if (page_is_saveable(zone, pfn))
 				memory_bm_set_bit(orig_bm, pfn);
 	}
 	memory_bm_position_reset(orig_bm);
 	memory_bm_position_reset(copy_bm);
 	do {
 		pfn = memory_bm_next_pfn(orig_bm);
-		if (likely(pfn != BM_END_OF_MAP)) {
-			struct page *page;
-			void *src;
-
-			page = pfn_to_page(pfn);
-			src = page_address(page);
-			page = pfn_to_page(memory_bm_next_pfn(copy_bm));
-			copy_data_page(page_address(page), src);
-		}
+		if (likely(pfn != BM_END_OF_MAP))
+			copy_data_page(memory_bm_next_pfn(copy_bm), pfn);
 	} while (pfn != BM_END_OF_MAP);
 }
 
+/* Total number of image pages */
+static unsigned int nr_copy_pages;
+/* Number of pages needed for saving the original pfns of the image pages */
+static unsigned int nr_meta_pages;
+
 /**
  *	swsusp_free - free pages allocated for the suspend.
  *
@@ -792,7 +824,7 @@ void swsusp_free(void)
 				if (PageNosave(page) && PageNosaveFree(page)) {
 					ClearPageNosave(page);
 					ClearPageNosaveFree(page);
-					free_page((long) page_address(page));
+					__free_page(page);
 				}
 			}
 	}
@@ -802,34 +834,108 @@ void swsusp_free(void)
 	buffer = NULL;
 }
 
+#ifdef CONFIG_HIGHMEM
+/**
+  *	count_pages_for_highmem - compute the number of non-highmem pages
+  *	that will be necessary for creating copies of highmem pages.
+  */
+
+static unsigned int count_pages_for_highmem(unsigned int nr_highmem)
+{
+	unsigned int free_highmem = count_free_highmem_pages();
+
+	if (free_highmem >= nr_highmem)
+		nr_highmem = 0;
+	else
+		nr_highmem -= free_highmem;
+
+	return nr_highmem;
+}
+#else
+static unsigned int
+count_pages_for_highmem(unsigned int nr_highmem) { return 0; }
+#endif /* CONFIG_HIGHMEM */
 
 /**
- *	enough_free_mem - Make sure we enough free memory to snapshot.
- *
- *	Returns TRUE or FALSE after checking the number of available
- *	free pages.
+ *	enough_free_mem - Make sure we have enough free memory for the
+ *	snapshot image.
  */
 
-static int enough_free_mem(unsigned int nr_pages)
+static int enough_free_mem(unsigned int nr_pages, unsigned int nr_highmem)
 {
 	struct zone *zone;
 	unsigned int free = 0, meta = 0;
 
-	for_each_zone (zone)
-		if (!is_highmem(zone)) {
+	for_each_zone(zone) {
+		meta += snapshot_additional_pages(zone);
+		if (!is_highmem(zone))
 			free += zone->free_pages;
-			meta += snapshot_additional_pages(zone);
-		}
+	}
 
-	pr_debug("swsusp: pages needed: %u + %u + %u, available pages: %u\n",
+	nr_pages += count_pages_for_highmem(nr_highmem);
+	pr_debug("swsusp: Normal pages needed: %u + %u + %u, available pages: %u\n",
 		nr_pages, PAGES_FOR_IO, meta, free);
 
 	return free > nr_pages + PAGES_FOR_IO + meta;
 }
 
+#ifdef CONFIG_HIGHMEM
+/**
+ *	get_highmem_buffer - if there are some highmem pages in the suspend
+ *	image, we may need the buffer to copy them and/or load their data.
+ */
+
+static inline int get_highmem_buffer(int safe_needed)
+{
+	buffer = get_image_page(GFP_ATOMIC | __GFP_COLD, safe_needed);
+	return buffer ? 0 : -ENOMEM;
+}
+
+/**
+ *	alloc_highmem_image_pages - allocate some highmem pages for the image.
+ *	Try to allocate as many pages as needed, but if the number of free
+ *	highmem pages is lesser than that, allocate them all.
+ */
+
+static inline unsigned int
+alloc_highmem_image_pages(struct memory_bitmap *bm, unsigned int nr_highmem)
+{
+	unsigned int to_alloc = count_free_highmem_pages();
+
+	if (to_alloc > nr_highmem)
+		to_alloc = nr_highmem;
+
+	nr_highmem -= to_alloc;
+	while (to_alloc-- > 0) {
+		struct page *page;
+
+		page = alloc_image_page(__GFP_HIGHMEM);
+		memory_bm_set_bit(bm, page_to_pfn(page));
+	}
+	return nr_highmem;
+}
+#else
+static inline int get_highmem_buffer(int safe_needed) { return 0; }
+
+static inline unsigned int
+alloc_highmem_image_pages(struct memory_bitmap *bm, unsigned int n) { return 0; }
+#endif /* CONFIG_HIGHMEM */
+
+/**
+ *	swsusp_alloc - allocate memory for the suspend image
+ *
+ *	We first try to allocate as many highmem pages as there are
+ *	saveable highmem pages in the system.  If that fails, we allocate
+ *	non-highmem pages for the copies of the remaining highmem ones.
+ *
+ *	In this approach it is likely that the copies of highmem pages will
+ *	also be located in the high memory, because of the way in which
+ *	copy_data_pages() works.
+ */
+
 static int
 swsusp_alloc(struct memory_bitmap *orig_bm, struct memory_bitmap *copy_bm,
-		unsigned int nr_pages)
+		unsigned int nr_pages, unsigned int nr_highmem)
 {
 	int error;
 
@@ -841,46 +947,61 @@ swsusp_alloc(struct memory_bitmap *orig_bm, struct memory_bitmap *copy_bm,
 	if (error)
 		goto Free;
 
+	if (nr_highmem > 0) {
+		error = get_highmem_buffer(PG_ANY);
+		if (error)
+			goto Free;
+
+		nr_pages += alloc_highmem_image_pages(copy_bm, nr_highmem);
+	}
 	while (nr_pages-- > 0) {
-		struct page *page = alloc_page(GFP_ATOMIC | __GFP_COLD);
+		struct page *page = alloc_image_page(GFP_ATOMIC | __GFP_COLD);
+
 		if (!page)
 			goto Free;
 
-		SetPageNosave(page);
-		SetPageNosaveFree(page);
 		memory_bm_set_bit(copy_bm, page_to_pfn(page));
 	}
 	return 0;
 
-Free:
+ Free:
 	swsusp_free();
 	return -ENOMEM;
 }
 
-/* Memory bitmap used for marking saveable pages */
+/* Memory bitmap used for marking saveable pages (during suspend) or the
+ * suspend image pages (during resume)
+ */
 static struct memory_bitmap orig_bm;
-/* Memory bitmap used for marking allocated pages that will contain the copies
- * of saveable pages
+/* Memory bitmap used on suspend for marking allocated pages that will contain
+ * the copies of saveable pages.  During resume it is initially used for
+ * marking the suspend image pages, but then its set bits are duplicated in
+ * @orig_bm and it is released.  Next, on systems with high memory, it may be
+ * used for marking "safe" highmem pages, but it has to be reinitialized for
+ * this purpose.
  */
 static struct memory_bitmap copy_bm;
 
 asmlinkage int swsusp_save(void)
 {
-	unsigned int nr_pages;
+	unsigned int nr_pages, nr_highmem;
 
-	pr_debug("swsusp: critical section: \n");
+	printk("swsusp: critical section: \n");
 
 	drain_local_pages();
 	nr_pages = count_data_pages();
-	printk("swsusp: Need to copy %u pages\n", nr_pages);
+	nr_highmem = count_highmem_pages();
+	printk("swsusp: Need to copy %u pages\n", nr_pages + nr_highmem);
 
-	if (!enough_free_mem(nr_pages)) {
+	if (!enough_free_mem(nr_pages, nr_highmem)) {
 		printk(KERN_ERR "swsusp: Not enough free memory\n");
 		return -ENOMEM;
 	}
 
-	if (swsusp_alloc(&orig_bm, &copy_bm, nr_pages))
+	if (swsusp_alloc(&orig_bm, &copy_bm, nr_pages, nr_highmem)) {
+		printk(KERN_ERR "swsusp: Memory allocation failed\n");
 		return -ENOMEM;
+	}
 
 	/* During allocating of suspend pagedir, new cold pages may appear.
 	 * Kill them.
@@ -894,10 +1015,12 @@ asmlinkage int swsusp_save(void)
 	 * touch swap space! Except we must write out our image of course.
 	 */
 
+	nr_pages += nr_highmem;
 	nr_copy_pages = nr_pages;
-	nr_meta_pages = (nr_pages * sizeof(long) + PAGE_SIZE - 1) >> PAGE_SHIFT;
+	nr_meta_pages = DIV_ROUND_UP(nr_pages * sizeof(long), PAGE_SIZE);
 
 	printk("swsusp: critical section/: done (%d pages copied)\n", nr_pages);
+
 	return 0;
 }
 
@@ -960,7 +1083,7 @@ int snapshot_read_next(struct snapshot_handle *handle, size_t count)
 
 	if (!buffer) {
 		/* This makes the buffer be freed by swsusp_free() */
-		buffer = alloc_image_page(GFP_ATOMIC, PG_ANY);
+		buffer = get_image_page(GFP_ATOMIC, PG_ANY);
 		if (!buffer)
 			return -ENOMEM;
 	}
@@ -975,9 +1098,23 @@ int snapshot_read_next(struct snapshot_handle *handle, size_t count)
 			memset(buffer, 0, PAGE_SIZE);
 			pack_pfns(buffer, &orig_bm);
 		} else {
-			unsigned long pfn = memory_bm_next_pfn(&copy_bm);
+			struct page *page;
 
-			handle->buffer = page_address(pfn_to_page(pfn));
+			page = pfn_to_page(memory_bm_next_pfn(&copy_bm));
+			if (PageHighMem(page)) {
+				/* Highmem pages are copied to the buffer,
+				 * because we can't return with a kmapped
+				 * highmem page (we may not be called again).
+				 */
+				void *kaddr;
+
+				kaddr = kmap_atomic(page, KM_USER0);
+				memcpy(buffer, kaddr, PAGE_SIZE);
+				kunmap_atomic(kaddr, KM_USER0);
+				handle->buffer = buffer;
+			} else {
+				handle->buffer = page_address(page);
+			}
 		}
 		handle->prev = handle->cur;
 	}
@@ -1005,7 +1142,7 @@ static int mark_unsafe_pages(struct memory_bitmap *bm)
 	unsigned long pfn, max_zone_pfn;
 
 	/* Clear page flags */
-	for_each_zone (zone) {
+	for_each_zone(zone) {
 		max_zone_pfn = zone->zone_start_pfn + zone->spanned_pages;
 		for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++)
 			if (pfn_valid(pfn))
@@ -1101,6 +1238,218 @@ unpack_orig_pfns(unsigned long *buf, struct memory_bitmap *bm)
 	}
 }
 
+/* List of "safe" pages that may be used to store data loaded from the suspend
+ * image
+ */
+static struct linked_page *safe_pages_list;
+
+#ifdef CONFIG_HIGHMEM
+/* struct highmem_pbe is used for creating the list of highmem pages that
+ * should be restored atomically during the resume from disk, because the page
+ * frames they have occupied before the suspend are in use.
+ */
+struct highmem_pbe {
+	struct page *copy_page;	/* data is here now */
+	struct page *orig_page;	/* data was here before the suspend */
+	struct highmem_pbe *next;
+};
+
+/* List of highmem PBEs needed for restoring the highmem pages that were
+ * allocated before the suspend and included in the suspend image, but have
+ * also been allocated by the "resume" kernel, so their contents cannot be
+ * written directly to their "original" page frames.
+ */
+static struct highmem_pbe *highmem_pblist;
+
+/**
+ *	count_highmem_image_pages - compute the number of highmem pages in the
+ *	suspend image.  The bits in the memory bitmap @bm that correspond to the
+ *	image pages are assumed to be set.
+ */
+
+static unsigned int count_highmem_image_pages(struct memory_bitmap *bm)
+{
+	unsigned long pfn;
+	unsigned int cnt = 0;
+
+	memory_bm_position_reset(bm);
+	pfn = memory_bm_next_pfn(bm);
+	while (pfn != BM_END_OF_MAP) {
+		if (PageHighMem(pfn_to_page(pfn)))
+			cnt++;
+
+		pfn = memory_bm_next_pfn(bm);
+	}
+	return cnt;
+}
+
+/**
+ *	prepare_highmem_image - try to allocate as many highmem pages as
+ *	there are highmem image pages (@nr_highmem_p points to the variable
+ *	containing the number of highmem image pages).  The pages that are
+ *	"safe" (ie. will not be overwritten when the suspend image is
+ *	restored) have the corresponding bits set in @bm (it must be
+ *	unitialized).
+ *
+ *	NOTE: This function should not be called if there are no highmem
+ *	image pages.
+ */
+
+static unsigned int safe_highmem_pages;
+
+static struct memory_bitmap *safe_highmem_bm;
+
+static int
+prepare_highmem_image(struct memory_bitmap *bm, unsigned int *nr_highmem_p)
+{
+	unsigned int to_alloc;
+
+	if (memory_bm_create(bm, GFP_ATOMIC, PG_SAFE))
+		return -ENOMEM;
+
+	if (get_highmem_buffer(PG_SAFE))
+		return -ENOMEM;
+
+	to_alloc = count_free_highmem_pages();
+	if (to_alloc > *nr_highmem_p)
+		to_alloc = *nr_highmem_p;
+	else
+		*nr_highmem_p = to_alloc;
+
+	safe_highmem_pages = 0;
+	while (to_alloc-- > 0) {
+		struct page *page;
+
+		page = alloc_page(__GFP_HIGHMEM);
+		if (!PageNosaveFree(page)) {
+			/* The page is "safe", set its bit the bitmap */
+			memory_bm_set_bit(bm, page_to_pfn(page));
+			safe_highmem_pages++;
+		}
+		/* Mark the page as allocated */
+		SetPageNosave(page);
+		SetPageNosaveFree(page);
+	}
+	memory_bm_position_reset(bm);
+	safe_highmem_bm = bm;
+	return 0;
+}
+
+/**
+ *	get_highmem_page_buffer - for given highmem image page find the buffer
+ *	that suspend_write_next() should set for its caller to write to.
+ *
+ *	If the page is to be saved to its "original" page frame or a copy of
+ *	the page is to be made in the highmem, @buffer is returned.  Otherwise,
+ *	the copy of the page is to be made in normal memory, so the address of
+ *	the copy is returned.
+ *
+ *	If @buffer is returned, the caller of suspend_write_next() will write
+ *	the page's contents to @buffer, so they will have to be copied to the
+ *	right location on the next call to suspend_write_next() and it is done
+ *	with the help of copy_last_highmem_page().  For this purpose, if
+ *	@buffer is returned, @last_highmem page is set to the page to which
+ *	the data will have to be copied from @buffer.
+ */
+
+static struct page *last_highmem_page;
+
+static void *
+get_highmem_page_buffer(struct page *page, struct chain_allocator *ca)
+{
+	struct highmem_pbe *pbe;
+	void *kaddr;
+
+	if (PageNosave(page) && PageNosaveFree(page)) {
+		/* We have allocated the "original" page frame and we can
+		 * use it directly to store the loaded page.
+		 */
+		last_highmem_page = page;
+		return buffer;
+	}
+	/* The "original" page frame has not been allocated and we have to
+	 * use a "safe" page frame to store the loaded page.
+	 */
+	pbe = chain_alloc(ca, sizeof(struct highmem_pbe));
+	if (!pbe) {
+		swsusp_free();
+		return NULL;
+	}
+	pbe->orig_page = page;
+	if (safe_highmem_pages > 0) {
+		struct page *tmp;
+
+		/* Copy of the page will be stored in high memory */
+		kaddr = buffer;
+		tmp = pfn_to_page(memory_bm_next_pfn(safe_highmem_bm));
+		safe_highmem_pages--;
+		last_highmem_page = tmp;
+		pbe->copy_page = tmp;
+	} else {
+		/* Copy of the page will be stored in normal memory */
+		kaddr = safe_pages_list;
+		safe_pages_list = safe_pages_list->next;
+		pbe->copy_page = virt_to_page(kaddr);
+	}
+	pbe->next = highmem_pblist;
+	highmem_pblist = pbe;
+	return kaddr;
+}
+
+/**
+ *	copy_last_highmem_page - copy the contents of a highmem image from
+ *	@buffer, where the caller of snapshot_write_next() has place them,
+ *	to the right location represented by @last_highmem_page .
+ */
+
+static void copy_last_highmem_page(void)
+{
+	if (last_highmem_page) {
+		void *dst;
+
+		dst = kmap_atomic(last_highmem_page, KM_USER0);
+		memcpy(dst, buffer, PAGE_SIZE);
+		kunmap_atomic(dst, KM_USER0);
+		last_highmem_page = NULL;
+	}
+}
+
+static inline int last_highmem_page_copied(void)
+{
+	return !last_highmem_page;
+}
+
+static inline void free_highmem_data(void)
+{
+	if (safe_highmem_bm)
+		memory_bm_free(safe_highmem_bm, PG_UNSAFE_CLEAR);
+
+	if (buffer)
+		free_image_page(buffer, PG_UNSAFE_CLEAR);
+}
+#else
+static inline int get_safe_write_buffer(void) { return 0; }
+
+static unsigned int
+count_highmem_image_pages(struct memory_bitmap *bm) { return 0; }
+
+static inline int
+prepare_highmem_image(struct memory_bitmap *bm, unsigned int *nr_highmem_p)
+{
+	return 0;
+}
+
+static inline void *
+get_highmem_page_buffer(struct page *page, struct chain_allocator *ca)
+{
+	return NULL;
+}
+
+static inline void copy_last_highmem_page(void) {}
+static inline int last_highmem_page_copied(void) { return 1; }
+static inline void free_highmem_data(void) {}
+#endif /* CONFIG_HIGHMEM */
+
 /**
  *	prepare_image - use the memory bitmap @bm to mark the pages that will
  *	be overwritten in the process of restoring the system memory state
@@ -1110,20 +1459,25 @@ unpack_orig_pfns(unsigned long *buf, struct memory_bitmap *bm)
  *	The idea is to allocate a new memory bitmap first and then allocate
  *	as many pages as needed for the image data, but not to assign these
  *	pages to specific tasks initially.  Instead, we just mark them as
- *	allocated and create a list of "safe" pages that will be used later.
+ *	allocated and create a lists of "safe" pages that will be used
+ *	later.  On systems with high memory a list of "safe" highmem pages is
+ *	also created.
  */
 
 #define PBES_PER_LINKED_PAGE	(LINKED_PAGE_DATA_SIZE / sizeof(struct pbe))
 
-static struct linked_page *safe_pages_list;
-
 static int
 prepare_image(struct memory_bitmap *new_bm, struct memory_bitmap *bm)
 {
-	unsigned int nr_pages;
+	unsigned int nr_pages, nr_highmem;
 	struct linked_page *sp_list, *lp;
 	int error;
 
+	/* If there is no highmem, the buffer will not be necessary */
+	free_image_page(buffer, PG_UNSAFE_CLEAR);
+	buffer = NULL;
+
+	nr_highmem = count_highmem_image_pages(bm);
 	error = mark_unsafe_pages(bm);
 	if (error)
 		goto Free;
@@ -1134,6 +1488,11 @@ prepare_image(struct memory_bitmap *new_bm, struct memory_bitmap *bm)
 
 	duplicate_memory_bitmap(new_bm, bm);
 	memory_bm_free(bm, PG_UNSAFE_KEEP);
+	if (nr_highmem > 0) {
+		error = prepare_highmem_image(bm, &nr_highmem);
+		if (error)
+			goto Free;
+	}
 	/* Reserve some safe pages for potential later use.
 	 *
 	 * NOTE: This way we make sure there will be enough safe pages for the
@@ -1142,10 +1501,10 @@ prepare_image(struct memory_bitmap *new_bm, struct memory_bitmap *bm)
 	 */
 	sp_list = NULL;
 	/* nr_copy_pages cannot be lesser than allocated_unsafe_pages */
-	nr_pages = nr_copy_pages - allocated_unsafe_pages;
+	nr_pages = nr_copy_pages - nr_highmem - allocated_unsafe_pages;
 	nr_pages = DIV_ROUND_UP(nr_pages, PBES_PER_LINKED_PAGE);
 	while (nr_pages > 0) {
-		lp = alloc_image_page(GFP_ATOMIC, PG_SAFE);
+		lp = get_image_page(GFP_ATOMIC, PG_SAFE);
 		if (!lp) {
 			error = -ENOMEM;
 			goto Free;
@@ -1156,7 +1515,7 @@ prepare_image(struct memory_bitmap *new_bm, struct memory_bitmap *bm)
 	}
 	/* Preallocate memory for the image */
 	safe_pages_list = NULL;
-	nr_pages = nr_copy_pages - allocated_unsafe_pages;
+	nr_pages = nr_copy_pages - nr_highmem - allocated_unsafe_pages;
 	while (nr_pages > 0) {
 		lp = (struct linked_page *)get_zeroed_page(GFP_ATOMIC);
 		if (!lp) {
@@ -1181,7 +1540,7 @@ prepare_image(struct memory_bitmap *new_bm, struct memory_bitmap *bm)
 	}
 	return 0;
 
-Free:
+ Free:
 	swsusp_free();
 	return error;
 }
@@ -1196,6 +1555,9 @@ static void *get_buffer(struct memory_bitmap *bm, struct chain_allocator *ca)
 	struct pbe *pbe;
 	struct page *page = pfn_to_page(memory_bm_next_pfn(bm));
 
+	if (PageHighMem(page))
+		return get_highmem_page_buffer(page, ca);
+
 	if (PageNosave(page) && PageNosaveFree(page))
 		/* We have allocated the "original" page frame and we can
 		 * use it directly to store the loaded page.
@@ -1210,12 +1572,12 @@ static void *get_buffer(struct memory_bitmap *bm, struct chain_allocator *ca)
 		swsusp_free();
 		return NULL;
 	}
-	pbe->orig_address = (unsigned long)page_address(page);
-	pbe->address = (unsigned long)safe_pages_list;
+	pbe->orig_address = page_address(page);
+	pbe->address = safe_pages_list;
 	safe_pages_list = safe_pages_list->next;
 	pbe->next = restore_pblist;
 	restore_pblist = pbe;
-	return (void *)pbe->address;
+	return pbe->address;
 }
 
 /**
@@ -1249,14 +1611,16 @@ int snapshot_write_next(struct snapshot_handle *handle, size_t count)
 	if (handle->prev && handle->cur > nr_meta_pages + nr_copy_pages)
 		return 0;
 
-	if (!buffer) {
-		/* This makes the buffer be freed by swsusp_free() */
-		buffer = alloc_image_page(GFP_ATOMIC, PG_ANY);
+	if (handle->offset == 0) {
+		if (!buffer)
+			/* This makes the buffer be freed by swsusp_free() */
+			buffer = get_image_page(GFP_ATOMIC, PG_ANY);
+
 		if (!buffer)
 			return -ENOMEM;
-	}
-	if (!handle->offset)
+
 		handle->buffer = buffer;
+	}
 	handle->sync_read = 1;
 	if (handle->prev < handle->cur) {
 		if (handle->prev == 0) {
@@ -1284,8 +1648,10 @@ int snapshot_write_next(struct snapshot_handle *handle, size_t count)
 					return -ENOMEM;
 			}
 		} else {
+			copy_last_highmem_page();
 			handle->buffer = get_buffer(&orig_bm, &ca);
-			handle->sync_read = 0;
+			if (handle->buffer != buffer)
+				handle->sync_read = 0;
 		}
 		handle->prev = handle->cur;
 	}
@@ -1301,15 +1667,73 @@ int snapshot_write_next(struct snapshot_handle *handle, size_t count)
 	return count;
 }
 
+/**
+ *	snapshot_write_finalize - must be called after the last call to
+ *	snapshot_write_next() in case the last page in the image happens
+ *	to be a highmem page and its contents should be stored in the
+ *	highmem.  Additionally, it releases the memory that will not be
+ *	used any more.
+ */
+
+void snapshot_write_finalize(struct snapshot_handle *handle)
+{
+	copy_last_highmem_page();
+	/* Free only if we have loaded the image entirely */
+	if (handle->prev && handle->cur > nr_meta_pages + nr_copy_pages) {
+		memory_bm_free(&orig_bm, PG_UNSAFE_CLEAR);
+		free_highmem_data();
+	}
+}
+
 int snapshot_image_loaded(struct snapshot_handle *handle)
 {
-	return !(!nr_copy_pages ||
+	return !(!nr_copy_pages || !last_highmem_page_copied() ||
 			handle->cur <= nr_meta_pages + nr_copy_pages);
 }
 
-void snapshot_free_unused_memory(struct snapshot_handle *handle)
+#ifdef CONFIG_HIGHMEM
+/* Assumes that @buf is ready and points to a "safe" page */
+static inline void
+swap_two_pages_data(struct page *p1, struct page *p2, void *buf)
 {
-	/* Free only if we have loaded the image entirely */
-	if (handle->prev && handle->cur > nr_meta_pages + nr_copy_pages)
-		memory_bm_free(&orig_bm, PG_UNSAFE_CLEAR);
+	void *kaddr1, *kaddr2;
+
+	kaddr1 = kmap_atomic(p1, KM_USER0);
+	kaddr2 = kmap_atomic(p2, KM_USER1);
+	memcpy(buf, kaddr1, PAGE_SIZE);
+	memcpy(kaddr1, kaddr2, PAGE_SIZE);
+	memcpy(kaddr2, buf, PAGE_SIZE);
+	kunmap_atomic(kaddr1, KM_USER0);
+	kunmap_atomic(kaddr2, KM_USER1);
+}
+
+/**
+ *	restore_highmem - for each highmem page that was allocated before
+ *	the suspend and included in the suspend image, and also has been
+ *	allocated by the "resume" kernel swap its current (ie. "before
+ *	resume") contents with the previous (ie. "before suspend") one.
+ *
+ *	If the resume eventually fails, we can call this function once
+ *	again and restore the "before resume" highmem state.
+ */
+
+int restore_highmem(void)
+{
+	struct highmem_pbe *pbe = highmem_pblist;
+	void *buf;
+
+	if (!pbe)
+		return 0;
+
+	buf = get_image_page(GFP_ATOMIC, PG_SAFE);
+	if (!buf)
+		return -ENOMEM;
+
+	while (pbe) {
+		swap_two_pages_data(pbe->copy_page, pbe->orig_page, buf);
+		pbe = pbe->next;
+	}
+	free_image_page(buf, PG_UNSAFE_CLEAR);
+	return 0;
 }
+#endif /* CONFIG_HIGHMEM */
diff --git a/kernel/power/swap.c b/kernel/power/swap.c
index 9b2ee5344de..f133d4a6d81 100644
--- a/kernel/power/swap.c
+++ b/kernel/power/swap.c
@@ -34,34 +34,123 @@ extern char resume_file[];
 #define SWSUSP_SIG	"S1SUSPEND"
 
 static struct swsusp_header {
-	char reserved[PAGE_SIZE - 20 - sizeof(swp_entry_t)];
-	swp_entry_t image;
+	char reserved[PAGE_SIZE - 20 - sizeof(sector_t)];
+	sector_t image;
 	char	orig_sig[10];
 	char	sig[10];
 } __attribute__((packed, aligned(PAGE_SIZE))) swsusp_header;
 
 /*
- * Saving part...
+ * General things
  */
 
 static unsigned short root_swap = 0xffff;
+static struct block_device *resume_bdev;
+
+/**
+ *	submit - submit BIO request.
+ *	@rw:	READ or WRITE.
+ *	@off	physical offset of page.
+ *	@page:	page we're reading or writing.
+ *	@bio_chain: list of pending biod (for async reading)
+ *
+ *	Straight from the textbook - allocate and initialize the bio.
+ *	If we're reading, make sure the page is marked as dirty.
+ *	Then submit it and, if @bio_chain == NULL, wait.
+ */
+static int submit(int rw, pgoff_t page_off, struct page *page,
+			struct bio **bio_chain)
+{
+	struct bio *bio;
+
+	bio = bio_alloc(__GFP_WAIT | __GFP_HIGH, 1);
+	if (!bio)
+		return -ENOMEM;
+	bio->bi_sector = page_off * (PAGE_SIZE >> 9);
+	bio->bi_bdev = resume_bdev;
+	bio->bi_end_io = end_swap_bio_read;
+
+	if (bio_add_page(bio, page, PAGE_SIZE, 0) < PAGE_SIZE) {
+		printk("swsusp: ERROR: adding page to bio at %ld\n", page_off);
+		bio_put(bio);
+		return -EFAULT;
+	}
+
+	lock_page(page);
+	bio_get(bio);
 
-static int mark_swapfiles(swp_entry_t start)
+	if (bio_chain == NULL) {
+		submit_bio(rw | (1 << BIO_RW_SYNC), bio);
+		wait_on_page_locked(page);
+		if (rw == READ)
+			bio_set_pages_dirty(bio);
+		bio_put(bio);
+	} else {
+		if (rw == READ)
+			get_page(page);	/* These pages are freed later */
+		bio->bi_private = *bio_chain;
+		*bio_chain = bio;
+		submit_bio(rw | (1 << BIO_RW_SYNC), bio);
+	}
+	return 0;
+}
+
+static int bio_read_page(pgoff_t page_off, void *addr, struct bio **bio_chain)
+{
+	return submit(READ, page_off, virt_to_page(addr), bio_chain);
+}
+
+static int bio_write_page(pgoff_t page_off, void *addr, struct bio **bio_chain)
+{
+	return submit(WRITE, page_off, virt_to_page(addr), bio_chain);
+}
+
+static int wait_on_bio_chain(struct bio **bio_chain)
+{
+	struct bio *bio;
+	struct bio *next_bio;
+	int ret = 0;
+
+	if (bio_chain == NULL)
+		return 0;
+
+	bio = *bio_chain;
+	if (bio == NULL)
+		return 0;
+	while (bio) {
+		struct page *page;
+
+		next_bio = bio->bi_private;
+		page = bio->bi_io_vec[0].bv_page;
+		wait_on_page_locked(page);
+		if (!PageUptodate(page) || PageError(page))
+			ret = -EIO;
+		put_page(page);
+		bio_put(bio);
+		bio = next_bio;
+	}
+	*bio_chain = NULL;
+	return ret;
+}
+
+/*
+ * Saving part
+ */
+
+static int mark_swapfiles(sector_t start)
 {
 	int error;
 
-	rw_swap_page_sync(READ, swp_entry(root_swap, 0),
-			  virt_to_page((unsigned long)&swsusp_header), NULL);
+	bio_read_page(swsusp_resume_block, &swsusp_header, NULL);
 	if (!memcmp("SWAP-SPACE",swsusp_header.sig, 10) ||
 	    !memcmp("SWAPSPACE2",swsusp_header.sig, 10)) {
 		memcpy(swsusp_header.orig_sig,swsusp_header.sig, 10);
 		memcpy(swsusp_header.sig,SWSUSP_SIG, 10);
 		swsusp_header.image = start;
-		error = rw_swap_page_sync(WRITE, swp_entry(root_swap, 0),
-				virt_to_page((unsigned long)&swsusp_header),
-				NULL);
+		error = bio_write_page(swsusp_resume_block,
+					&swsusp_header, NULL);
 	} else {
-		pr_debug("swsusp: Partition is not swap space.\n");
+		printk(KERN_ERR "swsusp: Swap header not found!\n");
 		error = -ENODEV;
 	}
 	return error;
@@ -74,12 +163,21 @@ static int mark_swapfiles(swp_entry_t start)
 
 static int swsusp_swap_check(void) /* This is called before saving image */
 {
-	int res = swap_type_of(swsusp_resume_device);
+	int res;
+
+	res = swap_type_of(swsusp_resume_device, swsusp_resume_block);
+	if (res < 0)
+		return res;
+
+	root_swap = res;
+	resume_bdev = open_by_devnum(swsusp_resume_device, FMODE_WRITE);
+	if (IS_ERR(resume_bdev))
+		return PTR_ERR(resume_bdev);
+
+	res = set_blocksize(resume_bdev, PAGE_SIZE);
+	if (res < 0)
+		blkdev_put(resume_bdev);
 
-	if (res >= 0) {
-		root_swap = res;
-		return 0;
-	}
 	return res;
 }
 
@@ -90,36 +188,26 @@ static int swsusp_swap_check(void) /* This is called before saving image */
  *	@bio_chain:	Link the next write BIO here
  */
 
-static int write_page(void *buf, unsigned long offset, struct bio **bio_chain)
+static int write_page(void *buf, sector_t offset, struct bio **bio_chain)
 {
-	swp_entry_t entry;
-	int error = -ENOSPC;
-
-	if (offset) {
-		struct page *page = virt_to_page(buf);
-
-		if (bio_chain) {
-			/*
-			 * Whether or not we successfully allocated a copy page,
-			 * we take a ref on the page here.  It gets undone in
-			 * wait_on_bio_chain().
-			 */
-			struct page *page_copy;
-			page_copy = alloc_page(GFP_ATOMIC);
-			if (page_copy == NULL) {
-				WARN_ON_ONCE(1);
-				bio_chain = NULL;	/* Go synchronous */
-				get_page(page);
-			} else {
-				memcpy(page_address(page_copy),
-					page_address(page), PAGE_SIZE);
-				page = page_copy;
-			}
+	void *src;
+
+	if (!offset)
+		return -ENOSPC;
+
+	if (bio_chain) {
+		src = (void *)__get_free_page(__GFP_WAIT | __GFP_HIGH);
+		if (src) {
+			memcpy(src, buf, PAGE_SIZE);
+		} else {
+			WARN_ON_ONCE(1);
+			bio_chain = NULL;	/* Go synchronous */
+			src = buf;
 		}
-		entry = swp_entry(root_swap, offset);
-		error = rw_swap_page_sync(WRITE, entry, page, bio_chain);
+	} else {
+		src = buf;
 	}
-	return error;
+	return bio_write_page(offset, src, bio_chain);
 }
 
 /*
@@ -137,11 +225,11 @@ static int write_page(void *buf, unsigned long offset, struct bio **bio_chain)
  *	at a time.
  */
 
-#define MAP_PAGE_ENTRIES	(PAGE_SIZE / sizeof(long) - 1)
+#define MAP_PAGE_ENTRIES	(PAGE_SIZE / sizeof(sector_t) - 1)
 
 struct swap_map_page {
-	unsigned long		entries[MAP_PAGE_ENTRIES];
-	unsigned long		next_swap;
+	sector_t entries[MAP_PAGE_ENTRIES];
+	sector_t next_swap;
 };
 
 /**
@@ -151,7 +239,7 @@ struct swap_map_page {
 
 struct swap_map_handle {
 	struct swap_map_page *cur;
-	unsigned long cur_swap;
+	sector_t cur_swap;
 	struct bitmap_page *bitmap;
 	unsigned int k;
 };
@@ -166,26 +254,6 @@ static void release_swap_writer(struct swap_map_handle *handle)
 	handle->bitmap = NULL;
 }
 
-static void show_speed(struct timeval *start, struct timeval *stop,
-			unsigned nr_pages, char *msg)
-{
-	s64 elapsed_centisecs64;
-	int centisecs;
-	int k;
-	int kps;
-
-	elapsed_centisecs64 = timeval_to_ns(stop) - timeval_to_ns(start);
-	do_div(elapsed_centisecs64, NSEC_PER_SEC / 100);
-	centisecs = elapsed_centisecs64;
-	if (centisecs == 0)
-		centisecs = 1;	/* avoid div-by-zero */
-	k = nr_pages * (PAGE_SIZE / 1024);
-	kps = (k * 100) / centisecs;
-	printk("%s %d kbytes in %d.%02d seconds (%d.%02d MB/s)\n", msg, k,
-			centisecs / 100, centisecs % 100,
-			kps / 1000, (kps % 1000) / 10);
-}
-
 static int get_swap_writer(struct swap_map_handle *handle)
 {
 	handle->cur = (struct swap_map_page *)get_zeroed_page(GFP_KERNEL);
@@ -196,7 +264,7 @@ static int get_swap_writer(struct swap_map_handle *handle)
 		release_swap_writer(handle);
 		return -ENOMEM;
 	}
-	handle->cur_swap = alloc_swap_page(root_swap, handle->bitmap);
+	handle->cur_swap = alloc_swapdev_block(root_swap, handle->bitmap);
 	if (!handle->cur_swap) {
 		release_swap_writer(handle);
 		return -ENOSPC;
@@ -205,43 +273,15 @@ static int get_swap_writer(struct swap_map_handle *handle)
 	return 0;
 }
 
-static int wait_on_bio_chain(struct bio **bio_chain)
-{
-	struct bio *bio;
-	struct bio *next_bio;
-	int ret = 0;
-
-	if (bio_chain == NULL)
-		return 0;
-
-	bio = *bio_chain;
-	if (bio == NULL)
-		return 0;
-	while (bio) {
-		struct page *page;
-
-		next_bio = bio->bi_private;
-		page = bio->bi_io_vec[0].bv_page;
-		wait_on_page_locked(page);
-		if (!PageUptodate(page) || PageError(page))
-			ret = -EIO;
-		put_page(page);
-		bio_put(bio);
-		bio = next_bio;
-	}
-	*bio_chain = NULL;
-	return ret;
-}
-
 static int swap_write_page(struct swap_map_handle *handle, void *buf,
 				struct bio **bio_chain)
 {
 	int error = 0;
-	unsigned long offset;
+	sector_t offset;
 
 	if (!handle->cur)
 		return -EINVAL;
-	offset = alloc_swap_page(root_swap, handle->bitmap);
+	offset = alloc_swapdev_block(root_swap, handle->bitmap);
 	error = write_page(buf, offset, bio_chain);
 	if (error)
 		return error;
@@ -250,7 +290,7 @@ static int swap_write_page(struct swap_map_handle *handle, void *buf,
 		error = wait_on_bio_chain(bio_chain);
 		if (error)
 			goto out;
-		offset = alloc_swap_page(root_swap, handle->bitmap);
+		offset = alloc_swapdev_block(root_swap, handle->bitmap);
 		if (!offset)
 			return -ENOSPC;
 		handle->cur->next_swap = offset;
@@ -261,7 +301,7 @@ static int swap_write_page(struct swap_map_handle *handle, void *buf,
 		handle->cur_swap = offset;
 		handle->k = 0;
 	}
-out:
+ out:
 	return error;
 }
 
@@ -315,7 +355,7 @@ static int save_image(struct swap_map_handle *handle,
 		error = err2;
 	if (!error)
 		printk("\b\b\b\bdone\n");
-	show_speed(&start, &stop, nr_to_write, "Wrote");
+	swsusp_show_speed(&start, &stop, nr_to_write, "Wrote");
 	return error;
 }
 
@@ -350,99 +390,50 @@ int swsusp_write(void)
 	struct swsusp_info *header;
 	int error;
 
-	if ((error = swsusp_swap_check())) {
+	error = swsusp_swap_check();
+	if (error) {
 		printk(KERN_ERR "swsusp: Cannot find swap device, try "
 				"swapon -a.\n");
 		return error;
 	}
 	memset(&snapshot, 0, sizeof(struct snapshot_handle));
 	error = snapshot_read_next(&snapshot, PAGE_SIZE);
-	if (error < PAGE_SIZE)
-		return error < 0 ? error : -EFAULT;
+	if (error < PAGE_SIZE) {
+		if (error >= 0)
+			error = -EFAULT;
+
+		goto out;
+	}
 	header = (struct swsusp_info *)data_of(snapshot);
 	if (!enough_swap(header->pages)) {
 		printk(KERN_ERR "swsusp: Not enough free swap\n");
-		return -ENOSPC;
+		error = -ENOSPC;
+		goto out;
 	}
 	error = get_swap_writer(&handle);
 	if (!error) {
-		unsigned long start = handle.cur_swap;
+		sector_t start = handle.cur_swap;
+
 		error = swap_write_page(&handle, header, NULL);
 		if (!error)
 			error = save_image(&handle, &snapshot,
 					header->pages - 1);
+
 		if (!error) {
 			flush_swap_writer(&handle);
 			printk("S");
-			error = mark_swapfiles(swp_entry(root_swap, start));
+			error = mark_swapfiles(start);
 			printk("|\n");
 		}
 	}
 	if (error)
 		free_all_swap_pages(root_swap, handle.bitmap);
 	release_swap_writer(&handle);
+ out:
+	swsusp_close();
 	return error;
 }
 
-static struct block_device *resume_bdev;
-
-/**
- *	submit - submit BIO request.
- *	@rw:	READ or WRITE.
- *	@off	physical offset of page.
- *	@page:	page we're reading or writing.
- *	@bio_chain: list of pending biod (for async reading)
- *
- *	Straight from the textbook - allocate and initialize the bio.
- *	If we're reading, make sure the page is marked as dirty.
- *	Then submit it and, if @bio_chain == NULL, wait.
- */
-static int submit(int rw, pgoff_t page_off, struct page *page,
-			struct bio **bio_chain)
-{
-	struct bio *bio;
-
-	bio = bio_alloc(GFP_ATOMIC, 1);
-	if (!bio)
-		return -ENOMEM;
-	bio->bi_sector = page_off * (PAGE_SIZE >> 9);
-	bio->bi_bdev = resume_bdev;
-	bio->bi_end_io = end_swap_bio_read;
-
-	if (bio_add_page(bio, page, PAGE_SIZE, 0) < PAGE_SIZE) {
-		printk("swsusp: ERROR: adding page to bio at %ld\n", page_off);
-		bio_put(bio);
-		return -EFAULT;
-	}
-
-	lock_page(page);
-	bio_get(bio);
-
-	if (bio_chain == NULL) {
-		submit_bio(rw | (1 << BIO_RW_SYNC), bio);
-		wait_on_page_locked(page);
-		if (rw == READ)
-			bio_set_pages_dirty(bio);
-		bio_put(bio);
-	} else {
-		get_page(page);
-		bio->bi_private = *bio_chain;
-		*bio_chain = bio;
-		submit_bio(rw | (1 << BIO_RW_SYNC), bio);
-	}
-	return 0;
-}
-
-static int bio_read_page(pgoff_t page_off, void *addr, struct bio **bio_chain)
-{
-	return submit(READ, page_off, virt_to_page(addr), bio_chain);
-}
-
-static int bio_write_page(pgoff_t page_off, void *addr)
-{
-	return submit(WRITE, page_off, virt_to_page(addr), NULL);
-}
-
 /**
  *	The following functions allow us to read data using a swap map
  *	in a file-alike way
@@ -455,17 +446,18 @@ static void release_swap_reader(struct swap_map_handle *handle)
 	handle->cur = NULL;
 }
 
-static int get_swap_reader(struct swap_map_handle *handle,
-                                      swp_entry_t start)
+static int get_swap_reader(struct swap_map_handle *handle, sector_t start)
 {
 	int error;
 
-	if (!swp_offset(start))
+	if (!start)
 		return -EINVAL;
-	handle->cur = (struct swap_map_page *)get_zeroed_page(GFP_ATOMIC);
+
+	handle->cur = (struct swap_map_page *)get_zeroed_page(__GFP_WAIT | __GFP_HIGH);
 	if (!handle->cur)
 		return -ENOMEM;
-	error = bio_read_page(swp_offset(start), handle->cur, NULL);
+
+	error = bio_read_page(start, handle->cur, NULL);
 	if (error) {
 		release_swap_reader(handle);
 		return error;
@@ -477,7 +469,7 @@ static int get_swap_reader(struct swap_map_handle *handle,
 static int swap_read_page(struct swap_map_handle *handle, void *buf,
 				struct bio **bio_chain)
 {
-	unsigned long offset;
+	sector_t offset;
 	int error;
 
 	if (!handle->cur)
@@ -546,11 +538,11 @@ static int load_image(struct swap_map_handle *handle,
 		error = err2;
 	if (!error) {
 		printk("\b\b\b\bdone\n");
-		snapshot_free_unused_memory(snapshot);
+		snapshot_write_finalize(snapshot);
 		if (!snapshot_image_loaded(snapshot))
 			error = -ENODATA;
 	}
-	show_speed(&start, &stop, nr_to_read, "Read");
+	swsusp_show_speed(&start, &stop, nr_to_read, "Read");
 	return error;
 }
 
@@ -599,12 +591,16 @@ int swsusp_check(void)
 	if (!IS_ERR(resume_bdev)) {
 		set_blocksize(resume_bdev, PAGE_SIZE);
 		memset(&swsusp_header, 0, sizeof(swsusp_header));
-		if ((error = bio_read_page(0, &swsusp_header, NULL)))
+		error = bio_read_page(swsusp_resume_block,
+					&swsusp_header, NULL);
+		if (error)
 			return error;
+
 		if (!memcmp(SWSUSP_SIG, swsusp_header.sig, 10)) {
 			memcpy(swsusp_header.sig, swsusp_header.orig_sig, 10);
 			/* Reset swap signature now */
-			error = bio_write_page(0, &swsusp_header);
+			error = bio_write_page(swsusp_resume_block,
+						&swsusp_header, NULL);
 		} else {
 			return -EINVAL;
 		}
diff --git a/kernel/power/swsusp.c b/kernel/power/swsusp.c
index 0b66659dc51..31aa0390c77 100644
--- a/kernel/power/swsusp.c
+++ b/kernel/power/swsusp.c
@@ -49,6 +49,7 @@
 #include <linux/bootmem.h>
 #include <linux/syscalls.h>
 #include <linux/highmem.h>
+#include <linux/time.h>
 
 #include "power.h"
 
@@ -64,10 +65,8 @@ int in_suspend __nosavedata = 0;
 
 #ifdef CONFIG_HIGHMEM
 unsigned int count_highmem_pages(void);
-int save_highmem(void);
 int restore_highmem(void);
 #else
-static inline int save_highmem(void) { return 0; }
 static inline int restore_highmem(void) { return 0; }
 static inline unsigned int count_highmem_pages(void) { return 0; }
 #endif
@@ -134,18 +133,18 @@ static int bitmap_set(struct bitmap_page *bitmap, unsigned long bit)
 	return 0;
 }
 
-unsigned long alloc_swap_page(int swap, struct bitmap_page *bitmap)
+sector_t alloc_swapdev_block(int swap, struct bitmap_page *bitmap)
 {
 	unsigned long offset;
 
 	offset = swp_offset(get_swap_page_of_type(swap));
 	if (offset) {
-		if (bitmap_set(bitmap, offset)) {
+		if (bitmap_set(bitmap, offset))
 			swap_free(swp_entry(swap, offset));
-			offset = 0;
-		}
+		else
+			return swapdev_block(swap, offset);
 	}
-	return offset;
+	return 0;
 }
 
 void free_all_swap_pages(int swap, struct bitmap_page *bitmap)
@@ -166,6 +165,34 @@ void free_all_swap_pages(int swap, struct bitmap_page *bitmap)
 }
 
 /**
+ *	swsusp_show_speed - print the time elapsed between two events represented by
+ *	@start and @stop
+ *
+ *	@nr_pages -	number of pages processed between @start and @stop
+ *	@msg -		introductory message to print
+ */
+
+void swsusp_show_speed(struct timeval *start, struct timeval *stop,
+			unsigned nr_pages, char *msg)
+{
+	s64 elapsed_centisecs64;
+	int centisecs;
+	int k;
+	int kps;
+
+	elapsed_centisecs64 = timeval_to_ns(stop) - timeval_to_ns(start);
+	do_div(elapsed_centisecs64, NSEC_PER_SEC / 100);
+	centisecs = elapsed_centisecs64;
+	if (centisecs == 0)
+		centisecs = 1;	/* avoid div-by-zero */
+	k = nr_pages * (PAGE_SIZE / 1024);
+	kps = (k * 100) / centisecs;
+	printk("%s %d kbytes in %d.%02d seconds (%d.%02d MB/s)\n", msg, k,
+			centisecs / 100, centisecs % 100,
+			kps / 1000, (kps % 1000) / 10);
+}
+
+/**
  *	swsusp_shrink_memory -  Try to free as much memory as needed
  *
  *	... but do not OOM-kill anyone
@@ -184,23 +211,37 @@ static inline unsigned long __shrink_memory(long tmp)
 
 int swsusp_shrink_memory(void)
 {
-	long size, tmp;
+	long tmp;
 	struct zone *zone;
 	unsigned long pages = 0;
 	unsigned int i = 0;
 	char *p = "-\\|/";
+	struct timeval start, stop;
 
 	printk("Shrinking memory...  ");
+	do_gettimeofday(&start);
 	do {
-		size = 2 * count_highmem_pages();
-		size += size / 50 + count_data_pages() + PAGES_FOR_IO;
+		long size, highmem_size;
+
+		highmem_size = count_highmem_pages();
+		size = count_data_pages() + PAGES_FOR_IO;
 		tmp = size;
+		size += highmem_size;
 		for_each_zone (zone)
-			if (!is_highmem(zone) && populated_zone(zone)) {
-				tmp -= zone->free_pages;
-				tmp += zone->lowmem_reserve[ZONE_NORMAL];
-				tmp += snapshot_additional_pages(zone);
+			if (populated_zone(zone)) {
+				if (is_highmem(zone)) {
+					highmem_size -= zone->free_pages;
+				} else {
+					tmp -= zone->free_pages;
+					tmp += zone->lowmem_reserve[ZONE_NORMAL];
+					tmp += snapshot_additional_pages(zone);
+				}
 			}
+
+		if (highmem_size < 0)
+			highmem_size = 0;
+
+		tmp += highmem_size;
 		if (tmp > 0) {
 			tmp = __shrink_memory(tmp);
 			if (!tmp)
@@ -212,7 +253,9 @@ int swsusp_shrink_memory(void)
 		}
 		printk("\b%c", p[i++%4]);
 	} while (tmp > 0);
+	do_gettimeofday(&stop);
 	printk("\bdone (%lu pages freed)\n", pages);
+	swsusp_show_speed(&start, &stop, pages, "Freed");
 
 	return 0;
 }
@@ -223,6 +266,7 @@ int swsusp_suspend(void)
 
 	if ((error = arch_prepare_suspend()))
 		return error;
+
 	local_irq_disable();
 	/* At this point, device_suspend() has been called, but *not*
 	 * device_power_down(). We *must* device_power_down() now.
@@ -235,23 +279,16 @@ int swsusp_suspend(void)
 		goto Enable_irqs;
 	}
 
-	if ((error = save_highmem())) {
-		printk(KERN_ERR "swsusp: Not enough free pages for highmem\n");
-		goto Restore_highmem;
-	}
-
 	save_processor_state();
 	if ((error = swsusp_arch_suspend()))
 		printk(KERN_ERR "Error %d suspending\n", error);
 	/* Restore control flow magically appears here */
 	restore_processor_state();
-Restore_highmem:
-	restore_highmem();
 	/* NOTE:  device_power_up() is just a resume() for devices
 	 * that suspended with irqs off ... no overall powerup.
 	 */
 	device_power_up();
-Enable_irqs:
+ Enable_irqs:
 	local_irq_enable();
 	return error;
 }
@@ -268,18 +305,23 @@ int swsusp_resume(void)
 		printk(KERN_ERR "Some devices failed to power down, very bad\n");
 	/* We'll ignore saved state, but this gets preempt count (etc) right */
 	save_processor_state();
-	error = swsusp_arch_resume();
-	/* Code below is only ever reached in case of failure. Otherwise
-	 * execution continues at place where swsusp_arch_suspend was called
-         */
-	BUG_ON(!error);
+	error = restore_highmem();
+	if (!error) {
+		error = swsusp_arch_resume();
+		/* The code below is only ever reached in case of a failure.
+		 * Otherwise execution continues at place where
+		 * swsusp_arch_suspend() was called
+        	 */
+		BUG_ON(!error);
+		/* This call to restore_highmem() undos the previous one */
+		restore_highmem();
+	}
 	/* The only reason why swsusp_arch_resume() can fail is memory being
 	 * very tight, so we have to free it as soon as we can to avoid
 	 * subsequent failures
 	 */
 	swsusp_free();
 	restore_processor_state();
-	restore_highmem();
 	touch_softlockup_watchdog();
 	device_power_up();
 	local_irq_enable();
diff --git a/kernel/power/user.c b/kernel/power/user.c
index d991d3b0e5a..89443b85163 100644
--- a/kernel/power/user.c
+++ b/kernel/power/user.c
@@ -11,6 +11,7 @@
 
 #include <linux/suspend.h>
 #include <linux/syscalls.h>
+#include <linux/reboot.h>
 #include <linux/string.h>
 #include <linux/device.h>
 #include <linux/miscdevice.h>
@@ -21,6 +22,7 @@
 #include <linux/fs.h>
 #include <linux/console.h>
 #include <linux/cpu.h>
+#include <linux/freezer.h>
 
 #include <asm/uaccess.h>
 
@@ -54,7 +56,8 @@ static int snapshot_open(struct inode *inode, struct file *filp)
 	filp->private_data = data;
 	memset(&data->handle, 0, sizeof(struct snapshot_handle));
 	if ((filp->f_flags & O_ACCMODE) == O_RDONLY) {
-		data->swap = swsusp_resume_device ? swap_type_of(swsusp_resume_device) : -1;
+		data->swap = swsusp_resume_device ?
+				swap_type_of(swsusp_resume_device, 0) : -1;
 		data->mode = O_RDONLY;
 	} else {
 		data->swap = -1;
@@ -76,10 +79,10 @@ static int snapshot_release(struct inode *inode, struct file *filp)
 	free_all_swap_pages(data->swap, data->bitmap);
 	free_bitmap(data->bitmap);
 	if (data->frozen) {
-		down(&pm_sem);
+		mutex_lock(&pm_mutex);
 		thaw_processes();
 		enable_nonboot_cpus();
-		up(&pm_sem);
+		mutex_unlock(&pm_mutex);
 	}
 	atomic_inc(&device_available);
 	return 0;
@@ -124,7 +127,8 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp,
 {
 	int error = 0;
 	struct snapshot_data *data;
-	loff_t offset, avail;
+	loff_t avail;
+	sector_t offset;
 
 	if (_IOC_TYPE(cmd) != SNAPSHOT_IOC_MAGIC)
 		return -ENOTTY;
@@ -140,7 +144,7 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp,
 	case SNAPSHOT_FREEZE:
 		if (data->frozen)
 			break;
-		down(&pm_sem);
+		mutex_lock(&pm_mutex);
 		error = disable_nonboot_cpus();
 		if (!error) {
 			error = freeze_processes();
@@ -150,7 +154,7 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp,
 				error = -EBUSY;
 			}
 		}
-		up(&pm_sem);
+		mutex_unlock(&pm_mutex);
 		if (!error)
 			data->frozen = 1;
 		break;
@@ -158,10 +162,10 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp,
 	case SNAPSHOT_UNFREEZE:
 		if (!data->frozen)
 			break;
-		down(&pm_sem);
+		mutex_lock(&pm_mutex);
 		thaw_processes();
 		enable_nonboot_cpus();
-		up(&pm_sem);
+		mutex_unlock(&pm_mutex);
 		data->frozen = 0;
 		break;
 
@@ -170,7 +174,7 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp,
 			error = -EPERM;
 			break;
 		}
-		down(&pm_sem);
+		mutex_lock(&pm_mutex);
 		/* Free memory before shutting down devices. */
 		error = swsusp_shrink_memory();
 		if (!error) {
@@ -183,7 +187,7 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp,
 			}
 			resume_console();
 		}
-		up(&pm_sem);
+		mutex_unlock(&pm_mutex);
 		if (!error)
 			error = put_user(in_suspend, (unsigned int __user *)arg);
 		if (!error)
@@ -191,13 +195,13 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp,
 		break;
 
 	case SNAPSHOT_ATOMIC_RESTORE:
+		snapshot_write_finalize(&data->handle);
 		if (data->mode != O_WRONLY || !data->frozen ||
 		    !snapshot_image_loaded(&data->handle)) {
 			error = -EPERM;
 			break;
 		}
-		snapshot_free_unused_memory(&data->handle);
-		down(&pm_sem);
+		mutex_lock(&pm_mutex);
 		pm_prepare_console();
 		suspend_console();
 		error = device_suspend(PMSG_PRETHAW);
@@ -207,7 +211,7 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp,
 		}
 		resume_console();
 		pm_restore_console();
-		up(&pm_sem);
+		mutex_unlock(&pm_mutex);
 		break;
 
 	case SNAPSHOT_FREE:
@@ -238,10 +242,10 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp,
 				break;
 			}
 		}
-		offset = alloc_swap_page(data->swap, data->bitmap);
+		offset = alloc_swapdev_block(data->swap, data->bitmap);
 		if (offset) {
 			offset <<= PAGE_SHIFT;
-			error = put_user(offset, (loff_t __user *)arg);
+			error = put_user(offset, (sector_t __user *)arg);
 		} else {
 			error = -ENOSPC;
 		}
@@ -264,7 +268,7 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp,
 			 * so we need to recode them
 			 */
 			if (old_decode_dev(arg)) {
-				data->swap = swap_type_of(old_decode_dev(arg));
+				data->swap = swap_type_of(old_decode_dev(arg), 0);
 				if (data->swap < 0)
 					error = -ENODEV;
 			} else {
@@ -282,7 +286,7 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp,
 			break;
 		}
 
-		if (down_trylock(&pm_sem)) {
+		if (!mutex_trylock(&pm_mutex)) {
 			error = -EBUSY;
 			break;
 		}
@@ -309,8 +313,66 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp,
 		if (pm_ops->finish)
 			pm_ops->finish(PM_SUSPEND_MEM);
 
-OutS3:
-		up(&pm_sem);
+ OutS3:
+		mutex_unlock(&pm_mutex);
+		break;
+
+	case SNAPSHOT_PMOPS:
+		switch (arg) {
+
+		case PMOPS_PREPARE:
+			if (pm_ops->prepare) {
+				error = pm_ops->prepare(PM_SUSPEND_DISK);
+			}
+			break;
+
+		case PMOPS_ENTER:
+			kernel_shutdown_prepare(SYSTEM_SUSPEND_DISK);
+			error = pm_ops->enter(PM_SUSPEND_DISK);
+			break;
+
+		case PMOPS_FINISH:
+			if (pm_ops && pm_ops->finish) {
+				pm_ops->finish(PM_SUSPEND_DISK);
+			}
+			break;
+
+		default:
+			printk(KERN_ERR "SNAPSHOT_PMOPS: invalid argument %ld\n", arg);
+			error = -EINVAL;
+
+		}
+		break;
+
+	case SNAPSHOT_SET_SWAP_AREA:
+		if (data->bitmap) {
+			error = -EPERM;
+		} else {
+			struct resume_swap_area swap_area;
+			dev_t swdev;
+
+			error = copy_from_user(&swap_area, (void __user *)arg,
+					sizeof(struct resume_swap_area));
+			if (error) {
+				error = -EFAULT;
+				break;
+			}
+
+			/*
+			 * User space encodes device types as two-byte values,
+			 * so we need to recode them
+			 */
+			swdev = old_decode_dev(swap_area.dev);
+			if (swdev) {
+				offset = swap_area.offset;
+				data->swap = swap_type_of(swdev, offset);
+				if (data->swap < 0)
+					error = -ENODEV;
+			} else {
+				data->swap = -1;
+				error = -EINVAL;
+			}
+		}
 		break;
 
 	default:
@@ -321,7 +383,7 @@ OutS3:
 	return error;
 }
 
-static struct file_operations snapshot_fops = {
+static const struct file_operations snapshot_fops = {
 	.open = snapshot_open,
 	.release = snapshot_release,
 	.read = snapshot_read,