Merge bk://kernel.bkbits.net//home/mochel/linux-2.6-power

into ppc970.osdl.org:/home/torvalds/v2.6/linux

Documentation/power/swsusp.txt

@@ -20,26 +20,6 @@ From kernel/suspend.c:
 You need to append resume=/dev/your_swap_partition to kernel command
 line. Then you suspend by echo 4 > /proc/acpi/sleep.
 
-Pavel's unreliable guide to swsusp mess
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-There are currently two versions of swap suspend in the kernel, the old
-"Pavel's" version in kernel/power/swsusp.c and the new "Patrick's"
-version in kernel/power/pmdisk.c. They provide the same functionality;
-the old version looks ugly but was tested, while the new version looks
-nicer but did not receive so much testing. echo 4 > /proc/acpi/sleep
-calls the old version, echo disk > /sys/power/state calls the new one.
-
-[In the future, when the new version is stable enough, two things can
-happen:
-
-* the new version is moved into swsusp.c, and swsusp is renamed to swap
-  suspend (Pavel prefers this)
-
-* pmdisk is kept as is and swsusp.c is removed from the kernel]
-
-
-
 Article about goals and implementation of Software Suspend for Linux
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 Author: G‚ábor Kuti
@@ -75,10 +55,6 @@ hardware!
 About the code
 
 Things to implement
-- SMP support. I've done an SMP support but since I don't have access to a kind
-  of this one I cannot test it. Please SMP people test it.  .. Tested it,
-  doesn't work. Had no time to figure out why. There is some mess with
-  interrupts AFAIK..
 - We should only make a copy of data related to kernel segment, since any
   process data won't be changed.
 - Should make more sanity checks. Or are these enough?
@@ -90,11 +66,6 @@ Not so important ideas for implementing
 - We should not free pages at the beginning so aggressively, most of them
   go there anyway..
 
-Drivers that need support
-- pc_keyb -- perhaps we can wait for vojtech's input patches
-- do IDE cdroms need some kind of support?
-- IDE CD-RW -- how to deal with that?
-
 Sleep states summary (thanx, Ducrot)
 ====================================
 
@@ -109,7 +80,8 @@ and perhaps
 echo 4b > /proc/acpi/sleep      # for suspend to disk via s4bios
 
 
-FAQ:
+Frequently Asked Questions
+==========================
 
 Q: well, suspending a server is IMHO a really stupid thing,
 but... (Diego Zuccato):

arch/i386/power/Makefile

 obj-$(CONFIG_PM)		+= cpu.o
-obj-$(CONFIG_PM_DISK)		+= pmdisk.o
 obj-$(CONFIG_SOFTWARE_SUSPEND)	+= swsusp.o

arch/i386/power/pmdisk.S

-/* Originally gcc generated, modified by hand */
-
-#include <linux/linkage.h>
-#include <asm/segment.h>
-#include <asm/page.h>
-
-	.text
-
-ENTRY(pmdisk_arch_suspend)
-	cmpl $0,4(%esp)
-	jne .L1450
-
-	movl %esp, saved_context_esp
-	movl %ebx, saved_context_ebx
-	movl %ebp, saved_context_ebp
-	movl %esi, saved_context_esi
-	movl %edi, saved_context_edi
-	pushfl ; popl saved_context_eflags
-
-	call pmdisk_suspend
-	jmp .L1449
-	.p2align 4,,7
-.L1450:
-	movl $swsusp_pg_dir-__PAGE_OFFSET,%ecx
-	movl %ecx,%cr3
-
-	movl	pm_pagedir_nosave,%ebx
-	xorl	%eax, %eax
-	xorl	%edx, %edx
-	.p2align 4,,7
-.L1455:
-	movl	4(%ebx,%edx),%edi
-	movl	(%ebx,%edx),%esi
-
-	movl	$1024, %ecx
-	rep
-	movsl
-
-	movl	%cr3, %ecx;
-	movl	%ecx, %cr3;  # flush TLB 
-
-	incl	%eax
-	addl	$16, %edx
-	cmpl	pmdisk_pages,%eax
-	jb .L1455
-	.p2align 4,,7
-.L1453:
-	movl saved_context_esp, %esp
-	movl saved_context_ebp, %ebp
-	movl saved_context_ebx, %ebx
-	movl saved_context_esi, %esi
-	movl saved_context_edi, %edi
-	pushl saved_context_eflags ; popfl
-	call pmdisk_resume
-.L1449:
-	ret

arch/i386/power/swsusp.S

@@ -15,83 +15,47 @@
 
 	.text
 
-ENTRY(do_magic)
-	pushl %ebx
-	cmpl $0,8(%esp)
-	jne resume
-	call do_magic_suspend_1
-	call save_processor_state
+ENTRY(swsusp_arch_suspend)
 
 	movl %esp, saved_context_esp
-	movl %eax, saved_context_eax
 	movl %ebx, saved_context_ebx
-	movl %ecx, saved_context_ecx
-	movl %edx, saved_context_edx
 	movl %ebp, saved_context_ebp
 	movl %esi, saved_context_esi
 	movl %edi, saved_context_edi
 	pushfl ; popl saved_context_eflags
 
-	call do_magic_suspend_2
-	popl %ebx
+	call swsusp_save
 	ret
 
-resume:
+ENTRY(swsusp_arch_resume)
 	movl $swsusp_pg_dir-__PAGE_OFFSET,%ecx
 	movl %ecx,%cr3
 
-	call do_magic_resume_1
-	movl $0,loop
-	cmpl $0,nr_copy_pages
-	je copy_done
-copy_loop:
-	movl $0,loop2
+	movl	pagedir_nosave, %ebx
+	xorl	%eax, %eax
+	xorl	%edx, %edx
 	.p2align 4,,7
-copy_one_page:
-	movl pagedir_nosave,%ecx
-	movl loop,%eax
-	movl loop2,%edx
-	sall $4,%eax
-	movl 4(%ecx,%eax),%ebx
-	movl (%ecx,%eax),%eax
-	movb (%edx,%eax),%al
-	movb %al,(%edx,%ebx)
 
-	movl loop2,%eax
-	leal 1(%eax),%edx
-	movl %edx,loop2
-	movl %edx,%eax
-	cmpl $4095,%eax
-	jbe copy_one_page
-	movl loop,%eax
-	leal 1(%eax),%edx
-	movl %edx,loop
-	movl %edx,%eax
-	cmpl nr_copy_pages,%eax
-	jb copy_loop
+copy_loop:
+	movl	4(%ebx,%edx),%edi
+	movl	(%ebx,%edx),%esi
+
+	movl	$1024, %ecx
+	rep
+	movsl
 
-copy_done:
-	movl $__USER_DS,%eax
+	incl	%eax
+	addl	$16, %edx
+	cmpl	nr_copy_pages,%eax
+	jb copy_loop
+	.p2align 4,,7
 
-	movw %ax, %ds
-	movw %ax, %es
 	movl saved_context_esp, %esp
 	movl saved_context_ebp, %ebp
-	movl saved_context_eax, %eax
 	movl saved_context_ebx, %ebx
-	movl saved_context_ecx, %ecx
-	movl saved_context_edx, %edx
 	movl saved_context_esi, %esi
 	movl saved_context_edi, %edi
-	call restore_processor_state
+
 	pushl saved_context_eflags ; popfl
-	call do_magic_resume_2
-	popl %ebx
+	call swsusp_restore
 	ret
-
-       .section .data.nosave
-loop:
-       .quad 0
-loop2:
-       .quad 0
-       .previous

arch/x86_64/kernel/suspend_asm.S

-/* originally gcc generated, but now changed. don't overwrite. */
+/* Originally gcc generated, modified by hand
+ *
+ * This may not use any stack, nor any variable that is not "NoSave":
+ *
+ * Its rewriting one kernel image with another. What is stack in "old"
+ * image could very well be data page in "new" image, and overwriting
+ * your own stack under you is bad idea.
+ */
 	
 	.text
 #include <linux/linkage.h>
 #include <asm/segment.h>
 #include <asm/page.h>
 
-/* Input:
- * rdi	resume flag 
- */	
-			
-ENTRY(do_magic)
-.LFB5:
-	subq	$8, %rsp
-.LCFI2:
-	testl	%edi, %edi
-	jne	.L90
-	call	do_magic_suspend_1
-	call	save_processor_state
+ENTRY(swsusp_arch_suspend)
 
 	movq %rsp, saved_context_esp(%rip)
 	movq %rax, saved_context_eax(%rip)
@@ -36,9 +32,10 @@ ENTRY(do_magic)
 	movq %r15, saved_context_r15(%rip)
 	pushfq ; popq saved_context_eflags(%rip)
 
-	addq	$8, %rsp
-	jmp	do_magic_suspend_2
-.L90:
+	call swsusp_save
+	ret
+
+ENTRY(swsusp_arch_resume)
 	/* set up cr3 */	
 	leaq	init_level4_pgt(%rip),%rax
 	subq	$__START_KERNEL_map,%rax
@@ -53,7 +50,6 @@ ENTRY(do_magic)
 	movq %rcx, %cr3;                     
 	movq %rax, %cr4;  # turn PGE back on 
 
-	call	do_magic_resume_1
 	movl	nr_copy_pages(%rip), %eax
 	xorl	%ecx, %ecx
 	movq	$0, loop(%rip)
@@ -113,9 +109,8 @@ ENTRY(do_magic)
 	movq saved_context_r14(%rip), %r14
 	movq saved_context_r15(%rip), %r15
 	pushq saved_context_eflags(%rip) ; popfq
-	call	restore_processor_state
-	addq	$8, %rsp
-	jmp	do_magic_resume_2
+	call	swsusp_restore
+	ret
 
 	.section .data.nosave
 loop:

drivers/acpi/sleep/main.c

@@ -217,7 +217,8 @@ static int __init acpi_sleep_init(void)
 				sleep_states[i] = 1;
 				printk(" S4bios");
 				acpi_pm_ops.pm_disk_mode = PM_DISK_FIRMWARE;
-			} else if (sleep_states[i])
+			}
+			if (sleep_states[i])
 				acpi_pm_ops.pm_disk_mode = PM_DISK_PLATFORM;
 		}
 	}

include/linux/pm.h

@@ -194,11 +194,12 @@ extern void (*pm_idle)(void);
 extern void (*pm_power_off)(void);
 
 enum {
-	PM_SUSPEND_ON,
-	PM_SUSPEND_STANDBY,
-	PM_SUSPEND_MEM,
-	PM_SUSPEND_DISK,
-	PM_SUSPEND_MAX,
+	PM_SUSPEND_ON = 0,
+	PM_SUSPEND_STANDBY = 1,
+	/* NOTE: PM_SUSPEND_MEM == PCI_D3hot */
+	PM_SUSPEND_MEM = 3,
+	PM_SUSPEND_DISK = 4,
+	PM_SUSPEND_MAX = 5,
 };
 
 enum {

include/linux/suspend.h

@@ -23,16 +23,6 @@ typedef struct pbe {
 
 #define SWAP_FILENAME_MAXLENGTH	32
 
-struct suspend_header {
-	u32 version_code;
-	unsigned long num_physpages;
-	char machine[8];
-	char version[20];
-	int num_cpus;
-	int page_size;
-	suspend_pagedir_t *suspend_pagedir;
-	unsigned int num_pbes;
-};
 
 #define SUSPEND_PD_PAGES(x)     (((x)*sizeof(struct pbe))/PAGE_SIZE+1)
    
@@ -45,16 +35,12 @@ extern void drain_local_pages(void);
 /* kernel/power/swsusp.c */
 extern int software_suspend(void);
 
-extern unsigned int nr_copy_pages __nosavedata;
-extern suspend_pagedir_t *pagedir_nosave __nosavedata;
-
 #else	/* CONFIG_SOFTWARE_SUSPEND */
 static inline int software_suspend(void)
 {
 	printk("Warning: fake suspend called\n");
 	return -EPERM;
 }
-#define software_resume()		do { } while(0)
 #endif	/* CONFIG_SOFTWARE_SUSPEND */
 
 
@@ -78,12 +64,6 @@ static inline void disable_nonboot_cpus(void) {}
 static inline void enable_nonboot_cpus(void) {}
 #endif
 
-asmlinkage void do_magic(int is_resume);
-asmlinkage void do_magic_resume_1(void);
-asmlinkage void do_magic_resume_2(void);
-asmlinkage void do_magic_suspend_1(void);
-asmlinkage void do_magic_suspend_2(void);
-
 void save_processor_state(void);
 void restore_processor_state(void);
 struct saved_context;

kernel/power/Kconfig

@@ -18,6 +18,13 @@ config PM
 	  will issue the hlt instruction if nothing is to be done, thereby
 	  sending the processor to sleep and saving power.
 
+config PM_DEBUG
+	bool "Power Management Debug Support"
+	---help---
+	This option enables verbose debugging support in the Power Management
+	code. This is helpful when debugging and reporting various PM bugs, 
+	like suspend support.
+
 config SOFTWARE_SUSPEND
 	bool "Software Suspend (EXPERIMENTAL)"
 	depends on EXPERIMENTAL && PM && SWAP
@@ -42,33 +49,12 @@ config SOFTWARE_SUSPEND
 
 	  For more information take a look at Documentation/power/swsusp.txt.
 
-config PM_DISK
-	bool "Suspend-to-Disk Support"
-	depends on PM && SWAP && X86 && !X86_64
-	---help---
-	  Suspend-to-disk is a power management state in which the contents
-	  of memory are stored on disk and the entire system is shut down or
-	  put into a low-power state (e.g. ACPI S4). When the computer is 
-	  turned back on, the stored image is loaded from disk and execution
-	  resumes from where it left off before suspending. 
-
-	  This config option enables the core infrastructure necessary to 
-	  perform the suspend and resume transition. 
-
-	  Currently, this suspend-to-disk implementation is based on a forked
-	  version of the swsusp code base. As such, it's still experimental,
-	  and still relies on CONFIG_SWAP. 
-
-	  More information can be found in Documentation/power/.
-
-	  If unsure, Say N.
-
-config PM_DISK_PARTITION
+config PM_STD_PARTITION
 	string "Default resume partition"
-	depends on PM_DISK
+	depends on SOFTWARE_SUSPEND
 	default ""
 	---help---
-	  The default resume partition is the partition that the pmdisk suspend-
+	  The default resume partition is the partition that the suspend-
 	  to-disk implementation will look for a suspended disk image. 
 
 	  The partition specified here will be different for almost every user. 
@@ -77,16 +63,10 @@ config PM_DISK_PARTITION
 
 	  The partition specified can be overridden by specifying:
 
-		pmdisk=/dev/<other device> 
+		resume=/dev/<other device> 
 
 	  which will set the resume partition to the device specified. 
 
-	  One may also do: 
-
-		pmdisk=off
-
-	  to inform the kernel not to perform a resume transition. 
-
 	  Note there is currently not a way to specify which device to save the
 	  suspended image to. It will simply pick the first available swap 
 	  device.

kernel/power/Makefile

 
+ifeq ($(CONFIG_PM_DEBUG),y)
+EXTRA_CFLAGS	+=	-DDEBUG
+endif
+
 swsusp-smp-$(CONFIG_SMP)	+= smp.o
 
 obj-y				:= main.o process.o console.o pm.o
-obj-$(CONFIG_SOFTWARE_SUSPEND)	+= swsusp.o $(swsusp-smp-y)
-obj-$(CONFIG_PM_DISK)		+= disk.o pmdisk.o
+obj-$(CONFIG_SOFTWARE_SUSPEND)	+= swsusp.o $(swsusp-smp-y) disk.o
 
 obj-$(CONFIG_MAGIC_SYSRQ)	+= poweroff.o

kernel/power/disk.c

@@ -8,13 +8,11 @@
  *
  */
 
-#define DEBUG
-
-
 #include <linux/suspend.h>
 #include <linux/syscalls.h>
 #include <linux/reboot.h>
 #include <linux/string.h>
+#include <linux/device.h>
 #include <linux/delay.h>
 #include <linux/fs.h>
 #include "power.h"
@@ -23,13 +21,16 @@
 extern u32 pm_disk_mode;
 extern struct pm_ops * pm_ops;
 
-extern int pmdisk_save(void);
-extern int pmdisk_write(void);
-extern int pmdisk_read(void);
-extern int pmdisk_restore(void);
-extern int pmdisk_free(void);
+extern int swsusp_suspend(void);
+extern int swsusp_write(void);
+extern int swsusp_read(void);
+extern int swsusp_resume(void);
+extern int swsusp_free(void);
 
 
+static int noresume = 0;
+char resume_file[256] = CONFIG_PM_STD_PARTITION;
+
 /**
  *	power_down - Shut machine down for hibernate.
  *	@mode:		Suspend-to-disk mode
@@ -46,22 +47,26 @@ static int power_down(u32 mode)
 	int error = 0;
 
 	local_irq_save(flags);
-	device_power_down(PM_SUSPEND_DISK);
 	switch(mode) {
 	case PM_DISK_PLATFORM:
+		device_power_down(PM_SUSPEND_DISK);
 		error = pm_ops->enter(PM_SUSPEND_DISK);
 		break;
 	case PM_DISK_SHUTDOWN:
 		printk("Powering off system\n");
+		device_shutdown();
 		machine_power_off();
 		break;
 	case PM_DISK_REBOOT:
+		device_shutdown();
 		machine_restart(NULL);
 		break;
 	}
 	machine_halt();
-	device_power_up();
-	local_irq_restore(flags);
+	/* Valid image is on the disk, if we continue we risk serious data corruption
+	   after resume. */
+	printk(KERN_CRIT "Please power me down manually\n");
+	while(1);
 	return 0;
 }
 
@@ -99,6 +104,7 @@ static void finish(void)
 {
 	device_resume();
 	platform_finish();
+	enable_nonboot_cpus();
 	thaw_processes();
 	pm_restore_console();
 }
@@ -126,6 +132,7 @@ static int prepare(void)
 	/* Free memory before shutting down devices. */
 	free_some_memory();
 
+	disable_nonboot_cpus();
 	if ((error = device_suspend(PM_SUSPEND_DISK)))
 		goto Finish;
 
@@ -133,6 +140,7 @@ static int prepare(void)
  Finish:
 	platform_finish();
  Thaw:
+	enable_nonboot_cpus();
 	thaw_processes();
 	pm_restore_console();
 	return error;
@@ -161,7 +169,7 @@ int pm_suspend_disk(void)
 
 	pr_debug("PM: snapshotting memory.\n");
 	in_suspend = 1;
-	if ((error = pmdisk_save()))
+	if ((error = swsusp_suspend()))
 		goto Done;
 
 	if (in_suspend) {
@@ -173,14 +181,14 @@ int pm_suspend_disk(void)
 		mb();
 		barrier();
 
-		error = pmdisk_write();
+		error = swsusp_write();
 		if (!error) {
 			error = power_down(pm_disk_mode);
 			pr_debug("PM: Power down failed.\n");
 		}
 	} else
 		pr_debug("PM: Image restored successfully.\n");
-	pmdisk_free();
+	swsusp_free();
  Done:
 	finish();
 	return error;
@@ -188,7 +196,7 @@ int pm_suspend_disk(void)
 
 
 /**
- *	pm_resume - Resume from a saved image.
+ *	software_resume - Resume from a saved image.
  *
  *	Called as a late_initcall (so all devices are discovered and
  *	initialized), we call pmdisk to see if we have a saved image or not.
@@ -199,13 +207,21 @@ int pm_suspend_disk(void)
  *
  */
 
-static int pm_resume(void)
+static int software_resume(void)
 {
 	int error;
 
+	if (noresume) {
+		/**
+		 * FIXME: If noresume is specified, we need to find the partition
+		 * and reset it back to normal swap space.
+		 */
+		return 0;
+	}
+
 	pr_debug("PM: Reading pmdisk image.\n");
 
-	if ((error = pmdisk_read()))
+	if ((error = swsusp_read()))
 		goto Done;
 
 	pr_debug("PM: Preparing system for restore.\n");
@@ -216,28 +232,18 @@ static int pm_resume(void)
 	barrier();
 	mb();
 
-	/* FIXME: The following (comment and mdelay()) are from swsusp.
-	 * Are they really necessary?
-	 *
-	 * We do not want some readahead with DMA to corrupt our memory, right?
-	 * Do it with disabled interrupts for best effect. That way, if some
-	 * driver scheduled DMA, we have good chance for DMA to finish ;-).
-	 */
-	pr_debug("PM: Waiting for DMAs to settle down.\n");
-	mdelay(1000);
-
 	pr_debug("PM: Restoring saved image.\n");
-	pmdisk_restore();
+	swsusp_resume();
 	pr_debug("PM: Restore failed, recovering.n");
 	finish();
  Free:
-	pmdisk_free();
+	swsusp_free();
  Done:
 	pr_debug("PM: Resume from disk failed.\n");
 	return 0;
 }
 
-late_initcall(pm_resume);
+late_initcall(software_resume);
 
 
 static char * pm_disk_modes[] = {
@@ -336,3 +342,22 @@ static int __init pm_disk_init(void)
 }
 
 core_initcall(pm_disk_init);
+
+
+static int __init resume_setup(char *str)
+{
+	if (noresume)
+		return 1;
+
+	strncpy( resume_file, str, 255 );
+	return 1;
+}
+
+static int __init noresume_setup(char *str)
+{
+	noresume = 1;
+	return 1;
+}
+
+__setup("noresume", noresume_setup);
+__setup("resume=", resume_setup);

kernel/power/main.c

@@ -8,8 +8,6 @@
  *
  */
 
-#define DEBUG
-
 #include <linux/suspend.h>
 #include <linux/kobject.h>
 #include <linux/string.h>
@@ -35,8 +33,6 @@ void pm_set_ops(struct pm_ops * ops)
 {
 	down(&pm_sem);
 	pm_ops = ops;
-	if (ops->pm_disk_mode && ops->pm_disk_mode < PM_DISK_MAX)
-		pm_disk_mode = ops->pm_disk_mode;
 	up(&pm_sem);
 }
 
@@ -169,6 +165,15 @@ static int enter_state(u32 state)
 	return error;
 }
 
+/*
+ * This is main interface to the outside world. It needs to be
+ * called from process context.
+ */
+int software_suspend(void)
+{
+	return enter_state(PM_SUSPEND_DISK);
+}
+
 
 /**
  *	pm_suspend - Externally visible function for suspending system.
@@ -225,8 +230,8 @@ static ssize_t state_store(struct subsystem * subsys, const char * buf, size_t n
 	p = memchr(buf, '\n', n);
 	len = p ? p - buf : n;
 
-	for (s = &pm_states[state]; *s; s++, state++) {
-		if (!strncmp(buf, *s, len))
+	for (s = &pm_states[state]; state < PM_SUSPEND_MAX; s++, state++) {
+		if (*s && !strncmp(buf, *s, len))
 			break;
 	}
 	if (*s)

kernel/power/pmdisk.c

-/*
- * kernel/power/pmdisk.c - Suspend-to-disk implmentation
- *
- * This STD implementation is initially derived from swsusp (suspend-to-swap).
- * The original copyright on that was: 
- *
- * Copyright (C) 1998-2001 Gabor Kuti <seasons@fornax.hu>
- * Copyright (C) 1998,2001,2002 Pavel Machek <pavel@suse.cz>
- *
- * The additional parts are: 
- * 
- * Copyright (C) 2003 Patrick Mochel
- * Copyright (C) 2003 Open Source Development Lab
- * 
- * This file is released under the GPLv2. 
- *
- * For more information, please see the text files in Documentation/power/
- *
- */
-
-#undef DEBUG
-
-#include <linux/mm.h>
-#include <linux/bio.h>
-#include <linux/suspend.h>
-#include <linux/version.h>
-#include <linux/reboot.h>
-#include <linux/device.h>
-#include <linux/swapops.h>
-#include <linux/bootmem.h>
-#include <linux/utsname.h>
-
-#include <asm/mmu_context.h>
-
-#include "power.h"
-
-
-extern asmlinkage int pmdisk_arch_suspend(int resume);
-
-#define __ADDRESS(x)  ((unsigned long) phys_to_virt(x))
-#define ADDRESS(x) __ADDRESS((x) << PAGE_SHIFT)
-#define ADDRESS2(x) __ADDRESS(__pa(x))		/* Needed for x86-64 where some pages are in memory twice */
-
-/* References to section boundaries */
-extern char __nosave_begin, __nosave_end;
-
-extern int is_head_of_free_region(struct page *);
-
-/* Variables to be preserved over suspend */
-static int pagedir_order_check;
-static int nr_copy_pages_check;
-
-/* For resume= kernel option */
-static char resume_file[256] = CONFIG_PM_DISK_PARTITION;
-
-static dev_t resume_device;
-/* Local variables that should not be affected by save */
-unsigned int pmdisk_pages __nosavedata = 0;
-
-/* Suspend pagedir is allocated before final copy, therefore it
-   must be freed after resume 
-
-   Warning: this is evil. There are actually two pagedirs at time of
-   resume. One is "pagedir_save", which is empty frame allocated at
-   time of suspend, that must be freed. Second is "pagedir_nosave", 
-   allocated at time of resume, that travels through memory not to
-   collide with anything.
- */
-suspend_pagedir_t *pm_pagedir_nosave __nosavedata = NULL;
-static suspend_pagedir_t *pagedir_save;
-static int pagedir_order __nosavedata = 0;
-
-
-struct pmdisk_info {
-	struct new_utsname	uts;
-	u32			version_code;
-	unsigned long		num_physpages;
-	int			cpus;
-	unsigned long		image_pages;
-	unsigned long		pagedir_pages;
-	swp_entry_t		pagedir[768];
-} __attribute__((aligned(PAGE_SIZE))) pmdisk_info;
-
-
-
-#define PMDISK_SIG	"pmdisk-swap1"
-
-struct pmdisk_header {
-	char reserved[PAGE_SIZE - 20 - sizeof(swp_entry_t)];
-	swp_entry_t pmdisk_info;
-	char	orig_sig[10];
-	char	sig[10];
-} __attribute__((packed, aligned(PAGE_SIZE))) pmdisk_header;
-
-/*
- * XXX: We try to keep some more pages free so that I/O operations succeed
- * without paging. Might this be more?
- */
-#define PAGES_FOR_IO	512
-
-
-/*
- * Saving part...
- */
-
-
-/* We memorize in swapfile_used what swap devices are used for suspension */
-#define SWAPFILE_UNUSED    0
-#define SWAPFILE_SUSPEND   1	/* This is the suspending device */
-#define SWAPFILE_IGNORED   2	/* Those are other swap devices ignored for suspension */
-
-static unsigned short swapfile_used[MAX_SWAPFILES];
-static unsigned short root_swap;
-
-
-static int mark_swapfiles(swp_entry_t prev)
-{
-	int error;
-
-	rw_swap_page_sync(READ, 
-			  swp_entry(root_swap, 0),
-			  virt_to_page((unsigned long)&pmdisk_header));
-	if (!memcmp("SWAP-SPACE",pmdisk_header.sig,10) ||
-	    !memcmp("SWAPSPACE2",pmdisk_header.sig,10)) {
-		memcpy(pmdisk_header.orig_sig,pmdisk_header.sig,10);
-		memcpy(pmdisk_header.sig,PMDISK_SIG,10);
-		pmdisk_header.pmdisk_info = prev;
-		error = rw_swap_page_sync(WRITE, 
-					  swp_entry(root_swap, 0),
-					  virt_to_page((unsigned long)
-						       &pmdisk_header));
-	} else {
-		pr_debug("pmdisk: Partition is not swap space.\n");
-		error = -ENODEV;
-	}
-	return error;
-}
-
-static int read_swapfiles(void) /* This is called before saving image */
-{
-	int i, len;
-	
-	len=strlen(resume_file);
-	root_swap = 0xFFFF;
-	
-	swap_list_lock();
-	for(i=0; i<MAX_SWAPFILES; i++) {
-		if (swap_info[i].flags == 0) {
-			swapfile_used[i]=SWAPFILE_UNUSED;
-		} else {
-			if(!len) {
-				pr_debug("pmdisk: Default resume partition not set.\n");
-				if(root_swap == 0xFFFF) {
-					swapfile_used[i] = SWAPFILE_SUSPEND;
-					root_swap = i;
-				} else
-					swapfile_used[i] = SWAPFILE_IGNORED;				  
-			} else {
-	  			/* we ignore all swap devices that are not the resume_file */
-				if (1) {
-// FIXME				if(resume_device == swap_info[i].swap_device) {
-					swapfile_used[i] = SWAPFILE_SUSPEND;
-					root_swap = i;
-				} else
-				  	swapfile_used[i] = SWAPFILE_IGNORED;
-			}
-		}
-	}
-	swap_list_unlock();
-	return (root_swap != 0xffff) ? 0 : -ENODEV;
-}
-
-
-/* This is called after saving image so modification
-   will be lost after resume... and that's what we want. */
-static void lock_swapdevices(void)
-{
-	int i;
-
-	swap_list_lock();
-	for(i = 0; i< MAX_SWAPFILES; i++)
-		if(swapfile_used[i] == SWAPFILE_IGNORED) {
-			swap_info[i].flags ^= 0xFF; /* we make the device unusable. A new call to
-						       lock_swapdevices can unlock the devices. */
-		}
-	swap_list_unlock();
-}
-
-
-
-/**
- *	write_swap_page - Write one page to a fresh swap location.
- *	@addr:	Address we're writing.
- *	@loc:	Place to store the entry we used.
- *
- *	Allocate a new swap entry and 'sync' it. Note we discard -EIO
- *	errors. That is an artifact left over from swsusp. It did not 
- *	check the return of rw_swap_page_sync() at all, since most pages
- *	written back to swap would return -EIO.
- *	This is a partial improvement, since we will at least return other
- *	errors, though we need to eventually fix the damn code.
- */
-
-static int write_swap_page(unsigned long addr, swp_entry_t * loc)
-{
-	swp_entry_t entry;
-	int error = 0;
-
-	entry = get_swap_page();
-	if (swp_offset(entry) && 
-	    swapfile_used[swp_type(entry)] == SWAPFILE_SUSPEND) {
-		error = rw_swap_page_sync(WRITE, entry,
-					  virt_to_page(addr));
-		if (error == -EIO)
-			error = 0;
-		if (!error)
-			*loc = entry;
-	} else
-		error = -ENOSPC;
-	return error;
-}
-
-
-/**
- *	free_data - Free the swap entries used by the saved image.
- *
- *	Walk the list of used swap entries and free each one. 
- */
-
-static void free_data(void)
-{
-	swp_entry_t entry;
-	int i;
-
-	for (i = 0; i < pmdisk_pages; i++) {
-		entry = (pm_pagedir_nosave + i)->swap_address;
-		if (entry.val)
-			swap_free(entry);
-		else
-			break;
-		(pm_pagedir_nosave + i)->swap_address = (swp_entry_t){0};
-	}
-}
-
-
-/**
- *	write_data - Write saved image to swap.
- *
- *	Walk the list of pages in the image and sync each one to swap.
- */
-
-static int write_data(void)
-{
-	int error = 0;
-	int i;
-
-	printk( "Writing data to swap (%d pages): ", pmdisk_pages );
-	for (i = 0; i < pmdisk_pages && !error; i++) {
-		if (!(i%100))
-			printk( "." );
-		error = write_swap_page((pm_pagedir_nosave+i)->address,
-					&((pm_pagedir_nosave+i)->swap_address));
-	}
-	printk(" %d Pages done.\n",i);
-	return error;
-}
-
-
-/**
- *	free_pagedir - Free pages used by the page directory.
- */
-
-static void free_pagedir_entries(void)
-{
-	int num = pmdisk_info.pagedir_pages;
-	int i;
-
-	for (i = 0; i < num; i++)
-		swap_free(pmdisk_info.pagedir[i]);
-}
-
-
-/**
- *	write_pagedir - Write the array of pages holding the page directory.
- *	@last:	Last swap entry we write (needed for header).
- */
-
-static int write_pagedir(void)
-{
-	unsigned long addr = (unsigned long)pm_pagedir_nosave;
-	int error = 0;
-	int n = SUSPEND_PD_PAGES(pmdisk_pages);
-	int i;
-
-	pmdisk_info.pagedir_pages = n;
-	printk( "Writing pagedir (%d pages)\n", n);
-	for (i = 0; i < n && !error; i++, addr += PAGE_SIZE)
-		error = write_swap_page(addr,&pmdisk_info.pagedir[i]);
-	return error;
-}
-
-
-#ifdef DEBUG
-static void dump_pmdisk_info(void)
-{
-	printk(" pmdisk: Version: %u\n",pmdisk_info.version_code);
-	printk(" pmdisk: Num Pages: %ld\n",pmdisk_info.num_physpages);
-	printk(" pmdisk: UTS Sys: %s\n",pmdisk_info.uts.sysname);
-	printk(" pmdisk: UTS Node: %s\n",pmdisk_info.uts.nodename);
-	printk(" pmdisk: UTS Release: %s\n",pmdisk_info.uts.release);
-	printk(" pmdisk: UTS Version: %s\n",pmdisk_info.uts.version);
-	printk(" pmdisk: UTS Machine: %s\n",pmdisk_info.uts.machine);
-	printk(" pmdisk: UTS Domain: %s\n",pmdisk_info.uts.domainname);
-	printk(" pmdisk: CPUs: %d\n",pmdisk_info.cpus);
-	printk(" pmdisk: Image: %ld Pages\n",pmdisk_info.image_pages);
-	printk(" pmdisk: Pagedir: %ld Pages\n",pmdisk_info.pagedir_pages);
-}
-#else
-static void dump_pmdisk_info(void)
-{
-
-}
-#endif
-
-static void init_header(void)
-{
-	memset(&pmdisk_info,0,sizeof(pmdisk_info));
-	pmdisk_info.version_code = LINUX_VERSION_CODE;
-	pmdisk_info.num_physpages = num_physpages;
-	memcpy(&pmdisk_info.uts,&system_utsname,sizeof(system_utsname));
-
-	pmdisk_info.cpus = num_online_cpus();
-	pmdisk_info.image_pages = pmdisk_pages;
-}
-
-/**
- *	write_header - Fill and write the suspend header.
- *	@entry:	Location of the last swap entry used.
- *
- *	Allocate a page, fill header, write header. 
- *
- *	@entry is the location of the last pagedir entry written on 
- *	entrance. On exit, it contains the location of the header. 
- */
-
-static int write_header(swp_entry_t * entry)
-{
-	dump_pmdisk_info();
-	return write_swap_page((unsigned long)&pmdisk_info,entry);
-}
-
-
-
-/**
- *	write_suspend_image - Write entire image and metadata.
- *
- */
-
-static int write_suspend_image(void)
-{
-	int error;
-	swp_entry_t prev = { 0 };
-
-	init_header();
-
-	if ((error = write_data()))
-		goto FreeData;
-
-	if ((error = write_pagedir()))
-		goto FreePagedir;
-
-	if ((error = write_header(&prev)))
-		goto FreePagedir;
-
-	error = mark_swapfiles(prev);
- Done:
-	return error;
- FreePagedir:
-	free_pagedir_entries();
- FreeData:
-	free_data();
-	goto Done;
-}
-
-
-
-/**
- *	saveable - Determine whether a page should be cloned or not.
- *	@pfn:	The page
- *
- *	We save a page if it's Reserved, and not in the range of pages
- *	statically defined as 'unsaveable', or if it isn't reserved, and
- *	isn't part of a free chunk of pages.
- *	If it is part of a free chunk, we update @pfn to point to the last 
- *	page of the chunk.
- */
-
-static int saveable(unsigned long * pfn)
-{
-	struct page * page = pfn_to_page(*pfn);
-
-	if (PageNosave(page))
-		return 0;
-
-	if (!PageReserved(page)) {
-		int chunk_size;
-
-		if ((chunk_size = is_head_of_free_region(page))) {
-			*pfn += chunk_size - 1;
-			return 0;
-		}
-	} else if (PageReserved(page)) {
-		/* Just copy whole code segment. 
-		 * Hopefully it is not that big.
-		 */
-		if ((ADDRESS(*pfn) >= (unsigned long) ADDRESS2(&__nosave_begin)) && 
-		    (ADDRESS(*pfn) <  (unsigned long) ADDRESS2(&__nosave_end))) {
-			pr_debug("[nosave %lx]\n", ADDRESS(*pfn));
-			return 0;
-		}
-		/* Hmm, perhaps copying all reserved pages is not 
-		 * too healthy as they may contain 
-		 * critical bios data? 
-		 */
-	}
-	return 1;
-}
-
-
-
-/**
- *	count_pages - Determine size of page directory.
- *	
- *	Iterate over all the pages in the system and tally the number
- *	we need to clone.
- */
-
-static void count_pages(void)
-{
-	unsigned long pfn;
-	int n = 0;
-	
-	for (pfn = 0; pfn < max_pfn; pfn++) {
-		if (saveable(&pfn))
-			n++;
-	}
-	pmdisk_pages = n;
-}
-
-
-/**
- *	copy_pages - Atomically snapshot memory.
- *
- *	Iterate over all the pages in the system and copy each one 
- *	into its corresponding location in the pagedir.
- *	We rely on the fact that the number of pages that we're snap-
- *	shotting hasn't changed since we counted them. 
- */
-
-static void copy_pages(void)
-{
-	struct pbe * p = pagedir_save;
-	unsigned long pfn;
-	int n = 0;
-
-	for (pfn = 0; pfn < max_pfn; pfn++) {
-		if (saveable(&pfn)) {
-			n++;
-			p->orig_address = ADDRESS(pfn);
-			copy_page((void *) p->address, 
-				  (void *) p->orig_address);
-			p++;
-		}
-	}
-	BUG_ON(n != pmdisk_pages);
-}
-
-
-/**
- *	free_image_pages - Free each page allocated for snapshot.
- */
-
-static void free_image_pages(void)
-{
-	struct pbe * p;
-	int i;
-
-	for (i = 0, p = pagedir_save; i < pmdisk_pages; i++, p++) {
-		ClearPageNosave(virt_to_page(p->address));
-		free_page(p->address);
-	}
-}
-
-
-/**
- *	free_pagedir - Free the page directory.
- */
-
-static void free_pagedir(void)
-{
-	free_image_pages();
-	free_pages((unsigned long)pagedir_save, pagedir_order);
-}
-
-
-static void calc_order(void)
-{
-	int diff;
-	int order;
-
-	order = get_bitmask_order(SUSPEND_PD_PAGES(pmdisk_pages));
-	pmdisk_pages += 1 << order;
-	do {
-		diff = get_bitmask_order(SUSPEND_PD_PAGES(pmdisk_pages)) - order;
-		if (diff) {
-			order += diff;
-			pmdisk_pages += 1 << diff;
-		}
-	} while(diff);
-	pagedir_order = order;
-}
-
-
-/**
- *	alloc_pagedir - Allocate the page directory.
- *
- *	First, determine exactly how many contiguous pages we need, 
- *	allocate them, then mark each 'unsavable'.
- */
-
-static int alloc_pagedir(void)
-{
-	calc_order();
-	pagedir_save = (suspend_pagedir_t *)__get_free_pages(GFP_ATOMIC | __GFP_COLD, 
-							     pagedir_order);
-	if(!pagedir_save)
-		return -ENOMEM;
-	memset(pagedir_save,0,(1 << pagedir_order) * PAGE_SIZE);
-	pm_pagedir_nosave = pagedir_save;
-	return 0;
-}
-
-
-/**
- *	alloc_image_pages - Allocate pages for the snapshot.
- *
- */
-
-static int alloc_image_pages(void)
-{
-	struct pbe * p;
-	int i;
-
-	for (i = 0, p = pagedir_save; i < pmdisk_pages; i++, p++) {
-		p->address = get_zeroed_page(GFP_ATOMIC | __GFP_COLD);
-		if(!p->address)
-			goto Error;
-		SetPageNosave(virt_to_page(p->address));
-	}
-	return 0;
- Error:
-	do { 
-		if (p->address)
-			free_page(p->address);
-		p->address = 0;
-	} while (p-- > pagedir_save);
-	return -ENOMEM;
-}
-
-
-/**
- *	enough_free_mem - Make sure we enough free memory to snapshot.
- *
- *	Returns TRUE or FALSE after checking the number of available 
- *	free pages.
- */
-
-static int enough_free_mem(void)
-{
-	if(nr_free_pages() < (pmdisk_pages + PAGES_FOR_IO)) {
-		pr_debug("pmdisk: Not enough free pages: Have %d\n",
-			 nr_free_pages());
-		return 0;
-	}
-	return 1;
-}
-
-
-/**
- *	enough_swap - Make sure we have enough swap to save the image.
- *
- *	Returns TRUE or FALSE after checking the total amount of swap 
- *	space avaiable.
- *
- *	FIXME: si_swapinfo(&i) returns all swap devices information.
- *	We should only consider resume_device. 
- */
-
-static int enough_swap(void)
-{
-	struct sysinfo i;
-
-	si_swapinfo(&i);
-	if (i.freeswap < (pmdisk_pages + PAGES_FOR_IO))  {
-		pr_debug("pmdisk: Not enough swap. Need %ld\n",i.freeswap);
-		return 0;
-	}
-	return 1;
-}
-
-
-/**
- *	pmdisk_suspend - Atomically snapshot the system.
- *
- *	This must be called with interrupts disabled, to prevent the 
- *	system changing at all from underneath us. 
- *
- *	To do this, we count the number of pages in the system that we 
- *	need to save; make sure	we have enough memory and swap to clone
- *	the pages and save them in swap, allocate the space to hold them,
- *	and then snapshot them all.
- */
-
-int pmdisk_suspend(void)
-{
-	int error = 0;
-
-	if ((error = read_swapfiles()))
-		return error;
-
-	drain_local_pages();
-
-	pm_pagedir_nosave = NULL;
-	pr_debug("pmdisk: Counting pages to copy.\n" );
-	count_pages();
-	
-	pr_debug("pmdisk: (pages needed: %d + %d free: %d)\n",
-		 pmdisk_pages,PAGES_FOR_IO,nr_free_pages());
-
-	if (!enough_free_mem())
-		return -ENOMEM;
-
-	if (!enough_swap())
-		return -ENOSPC;
-
-	if ((error = alloc_pagedir())) {
-		pr_debug("pmdisk: Allocating pagedir failed.\n");
-		return error;
-	}
-	if ((error = alloc_image_pages())) {
-		pr_debug("pmdisk: Allocating image pages failed.\n");
-		free_pagedir();
-		return error;
-	}
-
-	nr_copy_pages_check = pmdisk_pages;
-	pagedir_order_check = pagedir_order;
-
-	/* During allocating of suspend pagedir, new cold pages may appear. 
-	 * Kill them 
-	 */
-	drain_local_pages();
-
-	/* copy */
-	copy_pages();
-
-	/*
-	 * End of critical section. From now on, we can write to memory,
-	 * but we should not touch disk. This specially means we must _not_
-	 * touch swap space! Except we must write out our image of course.
-	 */
-
-	pr_debug("pmdisk: %d pages copied\n", pmdisk_pages );
-	return 0;
-}
-
-
-/**
- *	suspend_save_image - Prepare and write saved image to swap.
- *
- *	IRQs are re-enabled here so we can resume devices and safely write
- *	to the swap devices. We disable them again before we leave.
- *
- *	The second lock_swapdevices() will unlock ignored swap devices since
- *	writing is finished.
- *	It is important _NOT_ to umount filesystems at this point. We want
- *	them synced (in case something goes wrong) but we DO not want to mark
- *	filesystem clean: it is not. (And it does not matter, if we resume
- *	correctly, we'll mark system clean, anyway.)
- */
-
-static int suspend_save_image(void)
-{
-	int error;
-	device_resume();
-	lock_swapdevices();
-	error = write_suspend_image();
-	lock_swapdevices();
-	return error;
-}
-
-/*
- * Magic happens here
- */
-
-int pmdisk_resume(void)
-{
-	BUG_ON (nr_copy_pages_check != pmdisk_pages);
-	BUG_ON (pagedir_order_check != pagedir_order);
-	
-	/* Even mappings of "global" things (vmalloc) need to be fixed */
-	__flush_tlb_global();
-	return 0;
-}
-
-/* pmdisk_arch_suspend() is implemented in arch/?/power/pmdisk.S,
-   and basically does:
-
-	if (!resume) {
-		save_processor_state();
-		SAVE_REGISTERS
-		return pmdisk_suspend();
-	}
-	GO_TO_SWAPPER_PAGE_TABLES
-	COPY_PAGES_BACK
-	RESTORE_REGISTERS
-	restore_processor_state();
-	return pmdisk_resume();
-
- */
-
-
-/* More restore stuff */
-
-#define does_collide(addr) does_collide_order(pm_pagedir_nosave, addr, 0)
-
-/*
- * Returns true if given address/order collides with any orig_address 
- */
-static int __init does_collide_order(suspend_pagedir_t *pagedir, 
-				     unsigned long addr, int order)
-{
-	int i;
-	unsigned long addre = addr + (PAGE_SIZE<<order);
-	
-	for(i=0; i < pmdisk_pages; i++)
-		if((pagedir+i)->orig_address >= addr &&
-			(pagedir+i)->orig_address < addre)
-			return 1;
-
-	return 0;
-}
-
-/*
- * We check here that pagedir & pages it points to won't collide with pages
- * where we're going to restore from the loaded pages later
- */
-static int __init check_pagedir(void)
-{
-	int i;
-
-	for(i=0; i < pmdisk_pages; i++) {
-		unsigned long addr;
-
-		do {
-			addr = get_zeroed_page(GFP_ATOMIC);
-			if(!addr)
-				return -ENOMEM;
-		} while (does_collide(addr));
-
-		(pm_pagedir_nosave+i)->address = addr;
-	}
-	return 0;
-}
-
-static int __init relocate_pagedir(void)
-{
-	/*
-	 * We have to avoid recursion (not to overflow kernel stack),
-	 * and that's why code looks pretty cryptic 
-	 */
-	suspend_pagedir_t *old_pagedir = pm_pagedir_nosave;
-	void **eaten_memory = NULL;
-	void **c = eaten_memory, *m, *f;
-	int err;
-
-	pr_debug("pmdisk: Relocating pagedir\n");
-
-	if(!does_collide_order(old_pagedir, (unsigned long)old_pagedir, pagedir_order)) {
-		pr_debug("pmdisk: Relocation not necessary\n");
-		return 0;
-	}
-
-	err = -ENOMEM;
-	while ((m = (void *) __get_free_pages(GFP_ATOMIC, pagedir_order)) != NULL) {
-		if (!does_collide_order(old_pagedir, (unsigned long)m,
-					pagedir_order)) {
-			pm_pagedir_nosave =
-				memcpy(m, old_pagedir,
-				       PAGE_SIZE << pagedir_order);
-			err = 0;
-			break;
-		}
-		eaten_memory = m;
-		printk( "." ); 
-		*eaten_memory = c;
-		c = eaten_memory;
-	}
-
-	c = eaten_memory;
-	while(c) {
-		printk(":");
-		f = c;
-		c = *c;
-		free_pages((unsigned long)f, pagedir_order);
-	}
-	printk("|\n");
-	return err;
-}
-
-
-static struct block_device * resume_bdev;
-
-
-/**
- *	Using bio to read from swap.
- *	This code requires a bit more work than just using buffer heads
- *	but, it is the recommended way for 2.5/2.6.
- *	The following are to signal the beginning and end of I/O. Bios
- *	finish asynchronously, while we want them to happen synchronously.
- *	A simple atomic_t, and a wait loop take care of this problem.
- */
-
-static atomic_t io_done = ATOMIC_INIT(0);
-
-static void start_io(void)
-{
-	atomic_set(&io_done,1);
-}
-
-static int end_io(struct bio * bio, unsigned int num, int err)
-{
-	atomic_set(&io_done,0);
-	return 0;
-}
-
-static void wait_io(void)
-{
-	while(atomic_read(&io_done))
-		io_schedule();
-}
-
-
-/**
- *	submit - submit BIO request.
- *	@rw:	READ or WRITE.
- *	@off	physical offset of page.
- *	@page:	page we're reading or writing.
- *
- *	Straight from the textbook - allocate and initialize the bio.
- *	If we're writing, make sure the page is marked as dirty.
- *	Then submit it and wait.
- */
-
-static int submit(int rw, pgoff_t page_off, void * page)
-{
-	int error = 0;
-	struct bio * bio;
-
-	bio = bio_alloc(GFP_ATOMIC,1);
-	if (!bio)
-		return -ENOMEM;
-	bio->bi_sector = page_off * (PAGE_SIZE >> 9);
-	bio_get(bio);
-	bio->bi_bdev = resume_bdev;
-	bio->bi_end_io = end_io;
-
-	if (bio_add_page(bio, virt_to_page(page), PAGE_SIZE, 0) < PAGE_SIZE) {
-		printk("pmdisk: ERROR: adding page to bio at %ld\n",page_off);
-		error = -EFAULT;
-		goto Done;
-	}
-
-	if (rw == WRITE)
-		bio_set_pages_dirty(bio);
-	start_io();
-	submit_bio(rw | (1 << BIO_RW_SYNC), bio);
-	wait_io();
- Done:
-	bio_put(bio);
-	return error;
-}
-
-static int
-read_page(pgoff_t page_off, void * page)
-{
-	return submit(READ,page_off,page);
-}
-
-static int
-write_page(pgoff_t page_off, void * page)
-{
-	return submit(WRITE,page_off,page);
-}
-
-
-extern dev_t __init name_to_dev_t(const char *line);
-
-
-static int __init check_sig(void)
-{
-	int error;
-
-	memset(&pmdisk_header,0,sizeof(pmdisk_header));
-	if ((error = read_page(0,&pmdisk_header)))
-		return error;
-	if (!memcmp(PMDISK_SIG,pmdisk_header.sig,10)) {
-		memcpy(pmdisk_header.sig,pmdisk_header.orig_sig,10);
-
-		/*
-		 * Reset swap signature now.
-		 */
-		error = write_page(0,&pmdisk_header);
-	} else { 
-		pr_debug(KERN_ERR "pmdisk: Invalid partition type.\n");
-		return -EINVAL;
-	}
-	if (!error)
-		pr_debug("pmdisk: Signature found, resuming\n");
-	return error;
-}
-
-
-/*
- * Sanity check if this image makes sense with this kernel/swap context
- * I really don't think that it's foolproof but more than nothing..
- */
-
-static const char * __init sanity_check(void)
-{
-	dump_pmdisk_info();
-	if(pmdisk_info.version_code != LINUX_VERSION_CODE)
-		return "kernel version";
-	if(pmdisk_info.num_physpages != num_physpages)
-		return "memory size";
-	if (strcmp(pmdisk_info.uts.sysname,system_utsname.sysname))
-		return "system type";
-	if (strcmp(pmdisk_info.uts.release,system_utsname.release))
-		return "kernel release";
-	if (strcmp(pmdisk_info.uts.version,system_utsname.version))
-		return "version";
-	if (strcmp(pmdisk_info.uts.machine,system_utsname.machine))
-		return "machine";
-	if(pmdisk_info.cpus != num_online_cpus())
-		return "number of cpus";
-	return NULL;
-}
-
-
-static int __init check_header(void)
-{
-	const char * reason = NULL;
-	int error;
-
-	init_header();
-
-	if ((error = read_page(swp_offset(pmdisk_header.pmdisk_info), 
-			       &pmdisk_info)))
-		return error;
-
- 	/* Is this same machine? */
-	if ((reason = sanity_check())) {
-		printk(KERN_ERR "pmdisk: Resume mismatch: %s\n",reason);
-		return -EPERM;
-	}
-	pmdisk_pages = pmdisk_info.image_pages;
-	return error;
-}
-
-
-static int __init read_pagedir(void)
-{
-	unsigned long addr;
-	int i, n = pmdisk_info.pagedir_pages;
-	int error = 0;
-
-	pagedir_order = get_bitmask_order(n);
-
-	addr =__get_free_pages(GFP_ATOMIC, pagedir_order);
-	if (!addr)
-		return -ENOMEM;
-	pm_pagedir_nosave = (struct pbe *)addr;
-
-	pr_debug("pmdisk: Reading pagedir (%d Pages)\n",n);
-
-	for (i = 0; i < n && !error; i++, addr += PAGE_SIZE) {
-		unsigned long offset = swp_offset(pmdisk_info.pagedir[i]);
-		if (offset)
-			error = read_page(offset, (void *)addr);
-		else
-			error = -EFAULT;
-	}
-	if (error)
-		free_pages((unsigned long)pm_pagedir_nosave,pagedir_order);
-	return error;
-}
-
-
-/**
- *	read_image_data - Read image pages from swap.
- *
- *	You do not need to check for overlaps, check_pagedir()
- *	already did that.
- */
-
-static int __init read_image_data(void)
-{
-	struct pbe * p;
-	int error = 0;
-	int i;
-
-	printk( "Reading image data (%d pages): ", pmdisk_pages );
-	for(i = 0, p = pm_pagedir_nosave; i < pmdisk_pages && !error; i++, p++) {
-		if (!(i%100))
-			printk( "." );
-		error = read_page(swp_offset(p->swap_address),
-				  (void *)p->address);
-	}
-	printk(" %d done.\n",i);
-	return error;
-}
-
-
-static int __init read_suspend_image(void)
-{
-	int error = 0;
-
-	if ((error = check_sig()))
-		return error;
-	if ((error = check_header()))
-		return error;
-	if ((error = read_pagedir()))
-		return error;
-	if ((error = relocate_pagedir()))
-		goto FreePagedir;
-	if ((error = check_pagedir()))
-		goto FreePagedir;
-	if ((error = read_image_data()))
-		goto FreePagedir;
- Done:
-	return error;
- FreePagedir:
-	free_pages((unsigned long)pm_pagedir_nosave,pagedir_order);
-	goto Done;
-}
-
-/**
- *	pmdisk_save - Snapshot memory
- */
-
-int pmdisk_save(void) 
-{
-	int error;
-
-#if defined (CONFIG_HIGHMEM) || defined (CONFIG_DISCONTIGMEM)
-	pr_debug("pmdisk: not supported with high- or discontig-mem.\n");
-	return -EPERM;
-#endif
-	if ((error = arch_prepare_suspend()))
-		return error;
-	local_irq_disable();
-	save_processor_state();
-	error = pmdisk_arch_suspend(0);
-	restore_processor_state();
-	local_irq_enable();
-	return error;
-}
-
-
-/**
- *	pmdisk_write - Write saved memory image to swap.
- *
- *	pmdisk_arch_suspend(0) returns after system is resumed.
- *
- *	pmdisk_arch_suspend() copies all "used" memory to "free" memory,
- *	then unsuspends all device drivers, and writes memory to disk
- *	using normal kernel mechanism.
- */
-
-int pmdisk_write(void)
-{
-	return suspend_save_image();
-}
-
-
-/**
- *	pmdisk_read - Read saved image from swap.
- */
-
-int __init pmdisk_read(void)
-{
-	int error;
-
-	if (!strlen(resume_file))
-		return -ENOENT;
-
-	resume_device = name_to_dev_t(resume_file);
-	pr_debug("pmdisk: Resume From Partition: %s\n", resume_file);
-
-	resume_bdev = open_by_devnum(resume_device, FMODE_READ);
-	if (!IS_ERR(resume_bdev)) {
-		set_blocksize(resume_bdev, PAGE_SIZE);
-		error = read_suspend_image();
-		blkdev_put(resume_bdev);
-	} else
-		error = PTR_ERR(resume_bdev);
-
-	if (!error)
-		pr_debug("Reading resume file was successful\n");
-	else
-		pr_debug("pmdisk: Error %d resuming\n", error);
-	return error;
-}
-
-
-/**
- *	pmdisk_restore - Replace running kernel with saved image.
- */
-
-int __init pmdisk_restore(void)
-{
-	int error;
-	local_irq_disable();
-	save_processor_state();
-	error = pmdisk_arch_suspend(1);
-	restore_processor_state();
-	local_irq_enable();
-	return error;
-}
-
-
-/**
- *	pmdisk_free - Free memory allocated to hold snapshot.
- */
-
-int pmdisk_free(void)
-{
-	pr_debug( "Freeing prev allocated pagedir\n" );
-	free_pagedir();
-	return 0;
-}
-
-static int __init pmdisk_setup(char *str)
-{
-	if (strlen(str)) {
-		if (!strcmp(str,"off"))
-			resume_file[0] = '\0';
-		else
-			strncpy(resume_file, str, 255);
-	} else
-		resume_file[0] = '\0';
-	return 1;
-}
-
-__setup("pmdisk=", pmdisk_setup);
-

kernel/power/power.h

-
+#include <linux/suspend.h>
+#include <linux/utsname.h>
 
 /* With SUSPEND_CONSOLE defined, it suspend looks *really* cool, but
    we probably do not take enough locks for switching consoles, etc,
@@ -9,7 +10,20 @@
 #endif
 
 
-#ifdef CONFIG_PM_DISK
+struct swsusp_info {
+	struct new_utsname	uts;
+	u32			version_code;
+	unsigned long		num_physpages;
+	int			cpus;
+	unsigned long		image_pages;
+	unsigned long		pagedir_pages;
+	suspend_pagedir_t	* suspend_pagedir;
+	swp_entry_t		pagedir[768];
+} __attribute__((aligned(PAGE_SIZE)));
+
+
+
+#ifdef CONFIG_SOFTWARE_SUSPEND
 extern int pm_suspend_disk(void);
 
 #else
@@ -18,7 +32,6 @@ static inline int pm_suspend_disk(void)
 	return -EPERM;
 }
 #endif
-
 extern struct semaphore pm_sem;
 #define power_attr(_name) \
 static struct subsys_attribute _name##_attr = {	\

kernel/power/swsusp.c

@@ -62,6 +62,7 @@
 #include <linux/syscalls.h>
 #include <linux/console.h>
 #include <linux/highmem.h>
+#include <linux/bio.h>
 
 #include <asm/uaccess.h>
 #include <asm/mmu_context.h>
@@ -70,25 +71,16 @@
 
 #include "power.h"
 
-unsigned char software_suspend_enabled = 0;
-
-#define NORESUME		1
-#define RESUME_SPECIFIED	2
-
 /* References to section boundaries */
 extern char __nosave_begin, __nosave_end;
 
 extern int is_head_of_free_region(struct page *);
 
-/* Locks */
-spinlock_t suspend_pagedir_lock __nosavedata = SPIN_LOCK_UNLOCKED;
-
 /* Variables to be preserved over suspend */
-static int pagedir_order_check;
-static int nr_copy_pages_check;
+int pagedir_order_check;
+int nr_copy_pages_check;
 
-static int resume_status;
-static char resume_file[256] = "";			/* For resume= kernel option */
+extern char resume_file[];
 static dev_t resume_device;
 /* Local variables that should not be affected by save */
 unsigned int nr_copy_pages __nosavedata = 0;
@@ -110,16 +102,16 @@ suspend_pagedir_t *pagedir_nosave __nosavedata = NULL;
 static suspend_pagedir_t *pagedir_save;
 static int pagedir_order __nosavedata = 0;
 
-struct link {
-	char dummy[PAGE_SIZE - sizeof(swp_entry_t)];
-	swp_entry_t next;
-};
+#define SWSUSP_SIG	"S1SUSPEND"
 
-union diskpage {
-	union swap_header swh;
-	struct link link;
-	struct suspend_header sh;
-};
+struct swsusp_header {
+	char reserved[PAGE_SIZE - 20 - sizeof(swp_entry_t)];
+	swp_entry_t swsusp_info;
+	char	orig_sig[10];
+	char	sig[10];
+} __attribute__((packed, aligned(PAGE_SIZE))) swsusp_header;
+
+struct swsusp_info swsusp_info;
 
 /*
  * XXX: We try to keep some more pages free so that I/O operations succeed
@@ -127,54 +119,10 @@ union diskpage {
  */
 #define PAGES_FOR_IO	512
 
-static const char name_suspend[] = "Suspend Machine: ";
-static const char name_resume[] = "Resume Machine: ";
-
-/*
- * Debug
- */
-#define	DEBUG_DEFAULT
-#undef	DEBUG_PROCESS
-#undef	DEBUG_SLOW
-#define TEST_SWSUSP 0		/* Set to 1 to reboot instead of halt machine after suspension */
-
-#ifdef DEBUG_DEFAULT
-# define PRINTK(f, a...)	printk(f, ## a)
-#else
-# define PRINTK(f, a...)       	do { } while(0)
-#endif
-
-#ifdef DEBUG_SLOW
-#define MDELAY(a) mdelay(a)
-#else
-#define MDELAY(a) do { } while(0)
-#endif
-
 /*
  * Saving part...
  */
 
-static __inline__ int fill_suspend_header(struct suspend_header *sh)
-{
-	memset((char *)sh, 0, sizeof(*sh));
-
-	sh->version_code = LINUX_VERSION_CODE;
-	sh->num_physpages = num_physpages;
-	strncpy(sh->machine, system_utsname.machine, 8);
-	strncpy(sh->version, system_utsname.version, 20);
-	/* FIXME: Is this bogus? --RR */
-	sh->num_cpus = num_online_cpus();
-	sh->page_size = PAGE_SIZE;
-	sh->suspend_pagedir = pagedir_nosave;
-	BUG_ON (pagedir_save != pagedir_nosave);
-	sh->num_pbes = nr_copy_pages;
-	/* TODO: needed? mounted fs' last mounted date comparison
-	 * [so they haven't been mounted since last suspend.
-	 * Maybe it isn't.] [we'd need to do this for _all_ fs-es]
-	 */
-	return 0;
-}
-
 /* We memorize in swapfile_used what swap devices are used for suspension */
 #define SWAPFILE_UNUSED    0
 #define SWAPFILE_SUSPEND   1	/* This is the suspending device */
@@ -182,47 +130,30 @@ static __inline__ int fill_suspend_header(struct suspend_header *sh)
 
 static unsigned short swapfile_used[MAX_SWAPFILES];
 static unsigned short root_swap;
-#define MARK_SWAP_SUSPEND 0
-#define MARK_SWAP_RESUME 2
 
-static void mark_swapfiles(swp_entry_t prev, int mode)
+static int mark_swapfiles(swp_entry_t prev)
 {
-	swp_entry_t entry;
-	union diskpage *cur;
-	struct page *page;
+	int error;
 
-	if (root_swap == 0xFFFF)  /* ignored */
-		return;
-
-	page = alloc_page(GFP_ATOMIC);
-	if (!page)
-		panic("Out of memory in mark_swapfiles");
-	cur = page_address(page);
-	/* XXX: this is dirty hack to get first page of swap file */
-	entry = swp_entry(root_swap, 0);
-	rw_swap_page_sync(READ, entry, page);
-
-	if (mode == MARK_SWAP_RESUME) {
-	  	if (!memcmp("S1",cur->swh.magic.magic,2))
-		  	memcpy(cur->swh.magic.magic,"SWAP-SPACE",10);
-		else if (!memcmp("S2",cur->swh.magic.magic,2))
-			memcpy(cur->swh.magic.magic,"SWAPSPACE2",10);
-		else printk("%sUnable to find suspended-data signature (%.10s - misspelled?\n", 
-		      	name_resume, cur->swh.magic.magic);
+	rw_swap_page_sync(READ, 
+			  swp_entry(root_swap, 0),
+			  virt_to_page((unsigned long)&swsusp_header));
+	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.swsusp_info = prev;
+		error = rw_swap_page_sync(WRITE, 
+					  swp_entry(root_swap, 0),
+					  virt_to_page((unsigned long)
+						       &swsusp_header));
 	} else {
-	  	if ((!memcmp("SWAP-SPACE",cur->swh.magic.magic,10)))
-		  	memcpy(cur->swh.magic.magic,"S1SUSP....",10);
-		else if ((!memcmp("SWAPSPACE2",cur->swh.magic.magic,10)))
-			memcpy(cur->swh.magic.magic,"S2SUSP....",10);
-		else panic("\nSwapspace is not swapspace (%.10s)\n", cur->swh.magic.magic);
-		cur->link.next = prev; /* prev is the first/last swap page of the resume area */
-		/* link.next lies *no more* in last 4/8 bytes of magic */
+		pr_debug("swsusp: Partition is not swap space.\n");
+		error = -ENODEV;
 	}
-	rw_swap_page_sync(WRITE, entry, page);
-	__free_page(page);
+	return error;
 }
 
-
 /*
  * Check whether the swap device is the specified resume
  * device, irrespective of whether they are specified by
@@ -243,7 +174,7 @@ static int is_resume_device(const struct swap_info_struct *swap_info)
 		resume_device == MKDEV(imajor(inode), iminor(inode));
 }
 
-static void read_swapfiles(void) /* This is called before saving image */
+int swsusp_swap_check(void) /* This is called before saving image */
 {
 	int i, len;
 	
@@ -274,114 +205,211 @@ static void read_swapfiles(void) /* This is called before saving image */
 		}
 	}
 	swap_list_unlock();
+	return (root_swap != 0xffff) ? 0 : -ENODEV;
 }
 
-static void lock_swapdevices(void) /* This is called after saving image so modification
-				      will be lost after resume... and that's what we want. */
+/**
+ * This is called after saving image so modification
+ * will be lost after resume... and that's what we want.
+ * we make the device unusable. A new call to
+ * lock_swapdevices can unlock the devices. 
+ */
+static void lock_swapdevices(void)
 {
 	int i;
 
 	swap_list_lock();
 	for(i = 0; i< MAX_SWAPFILES; i++)
 		if(swapfile_used[i] == SWAPFILE_IGNORED) {
-			swap_info[i].flags ^= 0xFF; /* we make the device unusable. A new call to
-						       lock_swapdevices can unlock the devices. */
+			swap_info[i].flags ^= 0xFF;
 		}
 	swap_list_unlock();
 }
 
+
+
 /**
- *    write_suspend_image - Write entire image to disk.
+ *	write_swap_page - Write one page to a fresh swap location.
+ *	@addr:	Address we're writing.
+ *	@loc:	Place to store the entry we used.
  *
- *    After writing suspend signature to the disk, suspend may no
- *    longer fail: we have ready-to-run image in swap, and rollback
- *    would happen on next reboot -- corrupting data.
+ *	Allocate a new swap entry and 'sync' it. Note we discard -EIO
+ *	errors. That is an artifact left over from swsusp. It did not 
+ *	check the return of rw_swap_page_sync() at all, since most pages
+ *	written back to swap would return -EIO.
+ *	This is a partial improvement, since we will at least return other
+ *	errors, though we need to eventually fix the damn code.
+ */
+
+static int write_page(unsigned long addr, swp_entry_t * loc)
+{
+	swp_entry_t entry;
+	int error = 0;
+
+	entry = get_swap_page();
+	if (swp_offset(entry) && 
+	    swapfile_used[swp_type(entry)] == SWAPFILE_SUSPEND) {
+		error = rw_swap_page_sync(WRITE, entry,
+					  virt_to_page(addr));
+		if (error == -EIO)
+			error = 0;
+		if (!error)
+			*loc = entry;
+	} else
+		error = -ENOSPC;
+	return error;
+}
+
+
+/**
+ *	data_free - Free the swap entries used by the saved image.
  *
- *    Note: The buffer we allocate to use to write the suspend header is
- *    not freed; its not needed since the system is going down anyway
- *    (plus it causes an oops and I'm lazy^H^H^H^Htoo busy).
+ *	Walk the list of used swap entries and free each one. 
+ *	This is only used for cleanup when suspend fails.
  */
-static int write_suspend_image(void)
+
+static void data_free(void)
 {
+	swp_entry_t entry;
 	int i;
-	swp_entry_t entry, prev = { 0 };
-	int nr_pgdir_pages = SUSPEND_PD_PAGES(nr_copy_pages);
-	union diskpage *cur,  *buffer = (union diskpage *)get_zeroed_page(GFP_ATOMIC);
-	unsigned long address;
-	struct page *page;
 
-	if (!buffer)
-		return -ENOMEM;
+	for (i = 0; i < nr_copy_pages; i++) {
+		entry = (pagedir_nosave + i)->swap_address;
+		if (entry.val)
+			swap_free(entry);
+		else
+			break;
+		(pagedir_nosave + i)->swap_address = (swp_entry_t){0};
+	}
+}
+
+
+/**
+ *	data_write - Write saved image to swap.
+ *
+ *	Walk the list of pages in the image and sync each one to swap.
+ */
+
+static int data_write(void)
+{
+	int error = 0;
+	int i;
 
 	printk( "Writing data to swap (%d pages): ", nr_copy_pages );
-	for (i=0; i<nr_copy_pages; i++) {
+	for (i = 0; i < nr_copy_pages && !error; i++) {
 		if (!(i%100))
 			printk( "." );
-		entry = get_swap_page();
-		if (!entry.val)
-			panic("\nNot enough swapspace when writing data" );
-		
-		if (swapfile_used[swp_type(entry)] != SWAPFILE_SUSPEND)
-			panic("\nPage %d: not enough swapspace on suspend device", i );
-	    
-		address = (pagedir_nosave+i)->address;
-		page = virt_to_page(address);
-		rw_swap_page_sync(WRITE, entry, page);
-		(pagedir_nosave+i)->swap_address = entry;
+		error = write_page((pagedir_nosave+i)->address,
+					  &((pagedir_nosave+i)->swap_address));
 	}
-	printk( "|\n" );
-	printk( "Writing pagedir (%d pages): ", nr_pgdir_pages);
-	for (i=0; i<nr_pgdir_pages; i++) {
-		cur = (union diskpage *)((char *) pagedir_nosave)+i;
-		BUG_ON ((char *) cur != (((char *) pagedir_nosave) + i*PAGE_SIZE));
-		printk( "." );
-		entry = get_swap_page();
-		if (!entry.val) {
-			printk(KERN_CRIT "Not enough swapspace when writing pgdir\n" );
-			panic("Don't know how to recover");
-			free_page((unsigned long) buffer);
-			return -ENOSPC;
-		}
+	printk(" %d Pages done.\n",i);
+	return error;
+}
 
-		if(swapfile_used[swp_type(entry)] != SWAPFILE_SUSPEND)
-			panic("\nNot enough swapspace for pagedir on suspend device" );
+static void dump_info(void)
+{
+	pr_debug(" swsusp: Version: %u\n",swsusp_info.version_code);
+	pr_debug(" swsusp: Num Pages: %ld\n",swsusp_info.num_physpages);
+	pr_debug(" swsusp: UTS Sys: %s\n",swsusp_info.uts.sysname);
+	pr_debug(" swsusp: UTS Node: %s\n",swsusp_info.uts.nodename);
+	pr_debug(" swsusp: UTS Release: %s\n",swsusp_info.uts.release);
+	pr_debug(" swsusp: UTS Version: %s\n",swsusp_info.uts.version);
+	pr_debug(" swsusp: UTS Machine: %s\n",swsusp_info.uts.machine);
+	pr_debug(" swsusp: UTS Domain: %s\n",swsusp_info.uts.domainname);
+	pr_debug(" swsusp: CPUs: %d\n",swsusp_info.cpus);
+	pr_debug(" swsusp: Image: %ld Pages\n",swsusp_info.image_pages);
+	pr_debug(" swsusp: Pagedir: %ld Pages\n",swsusp_info.pagedir_pages);
+}
 
-		BUG_ON (sizeof(swp_entry_t) != sizeof(long));
-		BUG_ON (PAGE_SIZE % sizeof(struct pbe));
+static void init_header(void)
+{
+	memset(&swsusp_info,0,sizeof(swsusp_info));
+	swsusp_info.version_code = LINUX_VERSION_CODE;
+	swsusp_info.num_physpages = num_physpages;
+	memcpy(&swsusp_info.uts,&system_utsname,sizeof(system_utsname));
+
+	swsusp_info.suspend_pagedir = pagedir_nosave;
+	swsusp_info.cpus = num_online_cpus();
+	swsusp_info.image_pages = nr_copy_pages;
+	dump_info();
+}
 
-		cur->link.next = prev;				
-		page = virt_to_page((unsigned long)cur);
-		rw_swap_page_sync(WRITE, entry, page);
-		prev = entry;
+static int close_swap(void)
+{
+	swp_entry_t entry;
+	int error;
+
+	error = write_page((unsigned long)&swsusp_info,&entry);
+	if (!error) { 
+		printk( "S" );
+		error = mark_swapfiles(entry);
+		printk( "|\n" );
 	}
-	printk("H");
-	BUG_ON (sizeof(struct suspend_header) > PAGE_SIZE-sizeof(swp_entry_t));
-	BUG_ON (sizeof(union diskpage) != PAGE_SIZE);
-	BUG_ON (sizeof(struct link) != PAGE_SIZE);
-	entry = get_swap_page();
-	if (!entry.val)
-		panic( "\nNot enough swapspace when writing header" );
-	if (swapfile_used[swp_type(entry)] != SWAPFILE_SUSPEND)
-		panic("\nNot enough swapspace for header on suspend device" );
-
-	cur = (void *) buffer;
-	if (fill_suspend_header(&cur->sh))
-		BUG();		/* Not a BUG_ON(): we want fill_suspend_header to be called, always */
-		
-	cur->link.next = prev;
-
-	page = virt_to_page((unsigned long)cur);
-	rw_swap_page_sync(WRITE, entry, page);
-	prev = entry;
-
-	printk( "S" );
-	mark_swapfiles(prev, MARK_SWAP_SUSPEND);
-	printk( "|\n" );
-
-	MDELAY(1000);
-	return 0;
+	return error;
 }
 
+/**
+ *	free_pagedir_entries - Free pages used by the page directory.
+ *
+ *	This is used during suspend for error recovery.
+ */
+
+static void free_pagedir_entries(void)
+{
+	int i;
+
+	for (i = 0; i < swsusp_info.pagedir_pages; i++)
+		swap_free(swsusp_info.pagedir[i]);
+}
+
+
+/**
+ *	write_pagedir - Write the array of pages holding the page directory.
+ *	@last:	Last swap entry we write (needed for header).
+ */
+
+static int write_pagedir(void)
+{
+	unsigned long addr = (unsigned long)pagedir_nosave;
+	int error = 0;
+	int n = SUSPEND_PD_PAGES(nr_copy_pages);
+	int i;
+
+	swsusp_info.pagedir_pages = n;
+	printk( "Writing pagedir (%d pages)\n", n);
+	for (i = 0; i < n && !error; i++, addr += PAGE_SIZE)
+		error = write_page(addr, &swsusp_info.pagedir[i]);
+	return error;
+}
+
+/**
+ *	write_suspend_image - Write entire image and metadata.
+ *
+ */
+
+static int write_suspend_image(void)
+{
+	int error;
+
+	init_header();
+	if ((error = data_write()))
+		goto FreeData;
+
+	if ((error = write_pagedir()))
+		goto FreePagedir;
+
+	if ((error = close_swap()))
+		goto FreePagedir;
+ Done:
+	return error;
+ FreePagedir:
+	free_pagedir_entries();
+ FreeData:
+	data_free();
+	goto Done;
+}
+
+
 #ifdef CONFIG_HIGHMEM
 struct highmem_page {
 	char *data;
@@ -438,22 +466,30 @@ static int save_highmem_zone(struct zone *zone)
 	}
 	return 0;
 }
+#endif /* CONFIG_HIGHMEM */
+
 
 static int save_highmem(void)
 {
+#ifdef CONFIG_HIGHMEM
 	struct zone *zone;
 	int res = 0;
+
+	pr_debug("swsusp: Saving Highmem\n");
 	for_each_zone(zone) {
 		if (is_highmem(zone))
 			res = save_highmem_zone(zone);
 		if (res)
 			return res;
 	}
+#endif
 	return 0;
 }
 
 static int restore_highmem(void)
 {
+#ifdef CONFIG_HIGHMEM
+	printk("swsusp: Restoring Highmem\n");
 	while (highmem_copy) {
 		struct highmem_page *save = highmem_copy;
 		void *kaddr;
@@ -465,9 +501,10 @@ static int restore_highmem(void)
 		free_page((long) save->data);
 		kfree(save);
 	}
+#endif
 	return 0;
 }
-#endif
+
 
 static int pfn_is_nosave(unsigned long pfn)
 {
@@ -476,57 +513,82 @@ static int pfn_is_nosave(unsigned long pfn)
 	return (pfn >= nosave_begin_pfn) && (pfn < nosave_end_pfn);
 }
 
-/* if *pagedir_p != NULL it also copies the counted pages */
-static int count_and_copy_zone(struct zone *zone, struct pbe **pagedir_p)
+/**
+ *	saveable - Determine whether a page should be cloned or not.
+ *	@pfn:	The page
+ *
+ *	We save a page if it's Reserved, and not in the range of pages
+ *	statically defined as 'unsaveable', or if it isn't reserved, and
+ *	isn't part of a free chunk of pages.
+ *	If it is part of a free chunk, we update @pfn to point to the last 
+ *	page of the chunk.
+ */
+
+static int saveable(struct zone * zone, unsigned long * zone_pfn)
 {
-	unsigned long zone_pfn, chunk_size, nr_copy_pages = 0;
-	struct pbe *pbe = *pagedir_p;
-	for (zone_pfn = 0; zone_pfn < zone->spanned_pages; ++zone_pfn) {
-		struct page *page;
-		unsigned long pfn = zone_pfn + zone->zone_start_pfn;
+	unsigned long pfn = *zone_pfn + zone->zone_start_pfn;
+	unsigned long chunk_size;
+	struct page * page;
 
-		if (!(pfn%1000))
-			printk(".");
-		if (!pfn_valid(pfn))
-			continue;
-		page = pfn_to_page(pfn);
-		BUG_ON(PageReserved(page) && PageNosave(page));
-		if (PageNosave(page))
-			continue;
-		if (PageReserved(page) && pfn_is_nosave(pfn)) {
-			PRINTK("[nosave pfn 0x%lx]", pfn);
-			continue;
-		}
-		if ((chunk_size = is_head_of_free_region(page))) {
-			pfn += chunk_size - 1;
-			zone_pfn += chunk_size - 1;
-			continue;
+	if (!pfn_valid(pfn))
+		return 0;
+
+	if (!(pfn%1000))
+		printk(".");
+	page = pfn_to_page(pfn);
+	BUG_ON(PageReserved(page) && PageNosave(page));
+	if (PageNosave(page))
+		return 0;
+	if (PageReserved(page) && pfn_is_nosave(pfn)) {
+		pr_debug("[nosave pfn 0x%lx]", pfn);
+		return 0;
+	}
+	if ((chunk_size = is_head_of_free_region(page))) {
+		*zone_pfn += chunk_size - 1;
+		return 0;
+	}
+
+	return 1;
+}
+
+static void count_data_pages(void)
+{
+	struct zone *zone;
+	unsigned long zone_pfn;
+
+	nr_copy_pages = 0;
+
+	for_each_zone(zone) {
+		if (!is_highmem(zone)) {
+			for (zone_pfn = 0; zone_pfn < zone->spanned_pages; ++zone_pfn)
+				nr_copy_pages += saveable(zone, &zone_pfn);
 		}
-		nr_copy_pages++;
-		if (!pbe)
-			continue;
-		pbe->orig_address = (long) page_address(page);
-		/* Copy page is dangerous: it likes to mess with
-		   preempt count on specific cpus. Wrong preempt count is then copied,
-		   oops. */
-		copy_page((void *)pbe->address, (void *)pbe->orig_address);
-		pbe++;
 	}
-	*pagedir_p = pbe;
-	return nr_copy_pages;
 }
 
-static int count_and_copy_data_pages(struct pbe *pagedir_p)
+
+static void copy_data_pages(void)
 {
-	int nr_copy_pages = 0;
 	struct zone *zone;
+	unsigned long zone_pfn;
+	struct pbe * pbe = pagedir_nosave;
+	
 	for_each_zone(zone) {
 		if (!is_highmem(zone))
-			nr_copy_pages += count_and_copy_zone(zone, &pagedir_p);
+			for (zone_pfn = 0; zone_pfn < zone->spanned_pages; ++zone_pfn) {
+				if (saveable(zone, &zone_pfn)) {
+					struct page * page;
+					page = pfn_to_page(zone_pfn + zone->zone_start_pfn);
+					pbe->orig_address = (long) page_address(page);
+					/* copy_page is no usable for copying task structs. */
+					memcpy((void *)pbe->address, (void *)pbe->orig_address, PAGE_SIZE);
+					pbe++;
+				}
+			}
 	}
-	return nr_copy_pages;
 }
 
+
 static void free_suspend_pagedir_zone(struct zone *zone, unsigned long pagedir)
 {
 	unsigned long zone_pfn, pagedir_end, pagedir_pfn, pagedir_end_pfn;
@@ -547,119 +609,202 @@ static void free_suspend_pagedir_zone(struct zone *zone, unsigned long pagedir)
 	}
 }
 
-static void free_suspend_pagedir(unsigned long this_pagedir)
+void swsusp_free(void)
 {
+	unsigned long p = (unsigned long)pagedir_save;
 	struct zone *zone;
 	for_each_zone(zone) {
 		if (!is_highmem(zone))
-			free_suspend_pagedir_zone(zone, this_pagedir);
+			free_suspend_pagedir_zone(zone, p);
 	}
-	free_pages(this_pagedir, pagedir_order);
+	free_pages(p, pagedir_order);
 }
 
-static suspend_pagedir_t *create_suspend_pagedir(int nr_copy_pages)
+
+/**
+ *	calc_order - Determine the order of allocation needed for pagedir_save.
+ *
+ *	This looks tricky, but is just subtle. Please fix it some time.
+ *	Since there are %nr_copy_pages worth of pages in the snapshot, we need
+ *	to allocate enough contiguous space to hold 
+ *		(%nr_copy_pages * sizeof(struct pbe)), 
+ *	which has the saved/orig locations of the page.. 
+ *
+ *	SUSPEND_PD_PAGES() tells us how many pages we need to hold those 
+ *	structures, then we call get_bitmask_order(), which will tell us the
+ *	last bit set in the number, starting with 1. (If we need 30 pages, that
+ *	is 0x0000001e in hex. The last bit is the 5th, which is the order we 
+ *	would use to allocate 32 contiguous pages).
+ *
+ *	Since we also need to save those pages, we add the number of pages that
+ *	we need to nr_copy_pages, and in case of an overflow, do the 
+ *	calculation again to update the number of pages needed. 
+ *
+ *	With this model, we will tend to waste a lot of memory if we just cross
+ *	an order boundary. Plus, the higher the order of allocation that we try
+ *	to do, the more likely we are to fail in a low-memory situtation 
+ *	(though	we're unlikely to get this far in such a case, since swsusp 
+ *	requires half of memory to be free anyway).
+ */
+
+
+static void calc_order(void)
 {
-	int i;
-	suspend_pagedir_t *pagedir;
-	struct pbe *p;
-	struct page *page;
+	int diff = 0;
+	int order = 0;
+
+	do {
+		diff = get_bitmask_order(SUSPEND_PD_PAGES(nr_copy_pages)) - order;
+		if (diff) {
+			order += diff;
+			nr_copy_pages += 1 << diff;
+		}
+	} while(diff);
+	pagedir_order = order;
+}
 
-	pagedir_order = get_bitmask_order(SUSPEND_PD_PAGES(nr_copy_pages));
 
-	p = pagedir = (suspend_pagedir_t *)__get_free_pages(GFP_ATOMIC | __GFP_COLD, pagedir_order);
-	if (!pagedir)
-		return NULL;
+/**
+ *	alloc_pagedir - Allocate the page directory.
+ *
+ *	First, determine exactly how many contiguous pages we need and
+ *	allocate them.
+ */
 
-	page = virt_to_page(pagedir);
-	for(i=0; i < 1<<pagedir_order; i++)
-		SetPageNosave(page++);
-		
-	while(nr_copy_pages--) {
+static int alloc_pagedir(void)
+{
+	calc_order();
+	pagedir_save = (suspend_pagedir_t *)__get_free_pages(GFP_ATOMIC | __GFP_COLD,
+							     pagedir_order);
+	if (!pagedir_save)
+		return -ENOMEM;
+	memset(pagedir_save, 0, (1 << pagedir_order) * PAGE_SIZE);
+	pagedir_nosave = pagedir_save;
+	return 0;
+}
+
+
+/**
+ *	alloc_image_pages - Allocate pages for the snapshot.
+ *
+ */
+
+static int alloc_image_pages(void)
+{
+	struct pbe * p;
+	int i;
+
+	for (i = 0, p = pagedir_save; i < nr_copy_pages; i++, p++) {
 		p->address = get_zeroed_page(GFP_ATOMIC | __GFP_COLD);
-		if (!p->address) {
-			free_suspend_pagedir((unsigned long) pagedir);
-			return NULL;
-		}
+		if(!p->address)
+			goto Error;
 		SetPageNosave(virt_to_page(p->address));
-		p->orig_address = 0;
-		p++;
 	}
-	return pagedir;
+	return 0;
+ Error:
+	do { 
+		if (p->address)
+			free_page(p->address);
+		p->address = 0;
+	} while (p-- > pagedir_save);
+	return -ENOMEM;
 }
 
-static int prepare_suspend_processes(void)
+
+/**
+ *	enough_free_mem - Make sure we enough free memory to snapshot.
+ *
+ *	Returns TRUE or FALSE after checking the number of available 
+ *	free pages.
+ */
+
+static int enough_free_mem(void)
 {
-	sys_sync();	/* Syncing needs pdflushd, so do it before stopping processes */
-	if (freeze_processes()) {
-		printk( KERN_ERR "Suspend failed: Not all processes stopped!\n" );
-		thaw_processes();
-		return 1;
+	if (nr_free_pages() < (nr_copy_pages + PAGES_FOR_IO)) {
+		pr_debug("swsusp: Not enough free pages: Have %d\n",
+			 nr_free_pages());
+		return 0;
 	}
-	return 0;
+	return 1;
 }
 
-/*
- * Try to free as much memory as possible, but do not OOM-kill anyone
+
+/**
+ *	enough_swap - Make sure we have enough swap to save the image.
+ *
+ *	Returns TRUE or FALSE after checking the total amount of swap 
+ *	space avaiable.
  *
- * Notice: all userland should be stopped at this point, or livelock is possible.
+ *	FIXME: si_swapinfo(&i) returns all swap devices information.
+ *	We should only consider resume_device. 
  */
-static void free_some_memory(void)
+
+static int enough_swap(void)
 {
-	printk("Freeing memory: ");
-	while (shrink_all_memory(10000))
-		printk(".");
-	printk("|\n");
+	struct sysinfo i;
+
+	si_swapinfo(&i);
+	if (i.freeswap < (nr_copy_pages + PAGES_FOR_IO))  {
+		pr_debug("swsusp: Not enough swap. Need %ld\n",i.freeswap);
+		return 0;
+	}
+	return 1;
 }
 
-static int suspend_prepare_image(void)
+static int swsusp_alloc(void)
 {
-	struct sysinfo i;
-	unsigned int nr_needed_pages = 0;
+	int error;
+
+	pr_debug("suspend: (pages needed: %d + %d free: %d)\n",
+		 nr_copy_pages, PAGES_FOR_IO, nr_free_pages());
 
 	pagedir_nosave = NULL;
-	printk( "/critical section: ");
-#ifdef CONFIG_HIGHMEM
-	printk( "handling highmem" );
-	if (save_highmem()) {
-		printk(KERN_CRIT "%sNot enough free pages for highmem\n", name_suspend);
+	if (!enough_free_mem())
 		return -ENOMEM;
-	}
-	printk(", ");
-#endif
 
-	printk("counting pages to copy" );
-	drain_local_pages();
-	nr_copy_pages = count_and_copy_data_pages(NULL);
-	nr_needed_pages = nr_copy_pages + PAGES_FOR_IO;
-	
-	printk(" (pages needed: %d+%d=%d free: %d)\n",nr_copy_pages,PAGES_FOR_IO,nr_needed_pages,nr_free_pages());
-	if(nr_free_pages() < nr_needed_pages) {
-		printk(KERN_CRIT "%sCouldn't get enough free pages, on %d pages short\n",
-		       name_suspend, nr_needed_pages-nr_free_pages());
-		root_swap = 0xFFFF;
-		return -ENOMEM;
-	}
-	si_swapinfo(&i);	/* FIXME: si_swapinfo(&i) returns all swap devices information.
-				   We should only consider resume_device. */
-	if (i.freeswap < nr_needed_pages)  {
-		printk(KERN_CRIT "%sThere's not enough swap space available, on %ld pages short\n",
-		       name_suspend, nr_needed_pages-i.freeswap);
+	if (!enough_swap())
 		return -ENOSPC;
-	}
 
-	PRINTK( "Alloc pagedir\n" ); 
-	pagedir_save = pagedir_nosave = create_suspend_pagedir(nr_copy_pages);
-	if (!pagedir_nosave) {
-		/* Pagedir is big, one-chunk allocation. It is easily possible for this allocation to fail */
-		printk(KERN_CRIT "%sCouldn't allocate continuous pagedir\n", name_suspend);
-		return -ENOMEM;
+	if ((error = alloc_pagedir())) {
+		pr_debug("suspend: Allocating pagedir failed.\n");
+		return error;
+	}
+	if ((error = alloc_image_pages())) {
+		pr_debug("suspend: Allocating image pages failed.\n");
+		swsusp_free();
+		return error;
 	}
+
 	nr_copy_pages_check = nr_copy_pages;
 	pagedir_order_check = pagedir_order;
+	return 0;
+}
 
-	drain_local_pages();	/* During allocating of suspend pagedir, new cold pages may appear. Kill them */
-	if (nr_copy_pages != count_and_copy_data_pages(pagedir_nosave))	/* copy */
-		BUG();
+int suspend_prepare_image(void)
+{
+	unsigned int nr_needed_pages = 0;
+	int error;
+
+	pr_debug("swsusp: critical section: \n");
+	if (save_highmem()) {
+		printk(KERN_CRIT "Suspend machine: Not enough free pages for highmem\n");
+		return -ENOMEM;
+	}
+
+	drain_local_pages();
+	count_data_pages();
+	printk("swsusp: Need to copy %u pages\n",nr_copy_pages);
+	nr_needed_pages = nr_copy_pages + PAGES_FOR_IO;
+
+	error = swsusp_alloc();
+	if (error)
+		return error;
+	
+	/* During allocating of suspend pagedir, new cold pages may appear. 
+	 * Kill them.
+	 */
+	drain_local_pages();
+	copy_data_pages();
 
 	/*
 	 * End of critical section. From now on, we can write to memory,
@@ -667,205 +812,79 @@ static int suspend_prepare_image(void)
 	 * touch swap space! Except we must write out our image of course.
 	 */
 
-	printk( "critical section/: done (%d pages copied)\n", nr_copy_pages );
+	printk("swsusp: critical section/: done (%d pages copied)\n", nr_copy_pages );
 	return 0;
 }
 
-static void suspend_save_image(void)
+
+/* It is important _NOT_ to umount filesystems at this point. We want
+ * them synced (in case something goes wrong) but we DO not want to mark
+ * filesystem clean: it is not. (And it does not matter, if we resume
+ * correctly, we'll mark system clean, anyway.)
+ */
+int swsusp_write(void)
 {
+	int error;
 	device_resume();
-
 	lock_swapdevices();
-	write_suspend_image();
-	lock_swapdevices();	/* This will unlock ignored swap devices since writing is finished */
+	error = write_suspend_image();
+	/* This will unlock ignored swap devices since writing is finished */
+	lock_swapdevices();
+	return error;
 
-	/* It is important _NOT_ to umount filesystems at this point. We want
-	 * them synced (in case something goes wrong) but we DO not want to mark
-	 * filesystem clean: it is not. (And it does not matter, if we resume
-	 * correctly, we'll mark system clean, anyway.)
-	 */
 }
 
-static void suspend_power_down(void)
-{
-	extern int C_A_D;
-	C_A_D = 0;
-	printk(KERN_EMERG "%s%s Trying to power down.\n", name_suspend, TEST_SWSUSP ? "Disable TEST_SWSUSP. NOT ": "");
-#ifdef CONFIG_VT
-	PRINTK(KERN_EMERG "shift_state: %04x\n", shift_state);
-	mdelay(1000);
-	if (TEST_SWSUSP ^ (!!(shift_state & (1 << KG_CTRL))))
-		machine_restart(NULL);
-	else
-#endif
-	{
-		device_suspend(3);
-		device_shutdown();
-		machine_power_off();
-	}
 
-	printk(KERN_EMERG "%sProbably not capable for powerdown. System halted.\n", name_suspend);
-	machine_halt();
-	while (1);
-	/* NOTREACHED */
-}
+extern asmlinkage int swsusp_arch_suspend(void);
+extern asmlinkage int swsusp_arch_resume(void);
 
-/*
- * Magic happens here
- */
 
-asmlinkage void do_magic_resume_1(void)
+asmlinkage int swsusp_save(void)
 {
-	barrier();
-	mb();
-	spin_lock_irq(&suspend_pagedir_lock);	/* Done to disable interrupts */ 
-
-	device_power_down(3);
-	PRINTK( "Waiting for DMAs to settle down...\n");
-	mdelay(1000);	/* We do not want some readahead with DMA to corrupt our memory, right?
-			   Do it with disabled interrupts for best effect. That way, if some
-			   driver scheduled DMA, we have good chance for DMA to finish ;-). */
-}
+	int error = 0;
 
-asmlinkage void do_magic_resume_2(void)
-{
-	BUG_ON (nr_copy_pages_check != nr_copy_pages);
-	BUG_ON (pagedir_order_check != pagedir_order);
-
-	__flush_tlb_global();		/* Even mappings of "global" things (vmalloc) need to be fixed */
-
-	PRINTK( "Freeing prev allocated pagedir\n" );
-	free_suspend_pagedir((unsigned long) pagedir_save);
-
-#ifdef CONFIG_HIGHMEM
-	printk( "Restoring highmem\n" );
-	restore_highmem();
-#endif
-	printk("done, devices\n");
-
-	device_power_up();
-	spin_unlock_irq(&suspend_pagedir_lock);
-	device_resume();
-
-	/* Fixme: this is too late; we should do this ASAP to avoid "infinite reboots" problem */
-	PRINTK( "Fixing swap signatures... " );
-	mark_swapfiles(((swp_entry_t) {0}), MARK_SWAP_RESUME);
-	PRINTK( "ok\n" );
-
-#ifdef SUSPEND_CONSOLE
-	acquire_console_sem();
-	update_screen(fg_console);
-	release_console_sem();
-#endif
+	if ((error = swsusp_swap_check()))
+		return error;
+	return suspend_prepare_image();
 }
 
-/* do_magic() is implemented in arch/?/kernel/suspend_asm.S, and basically does:
-
-	if (!resume) {
-		do_magic_suspend_1();
-		save_processor_state();
-		SAVE_REGISTERS
-		do_magic_suspend_2();
-		return;
-	}
-	GO_TO_SWAPPER_PAGE_TABLES
-	do_magic_resume_1();
-	COPY_PAGES_BACK
-	RESTORE_REGISTERS
+int swsusp_suspend(void)
+{
+	int error;
+	if ((error = arch_prepare_suspend()))
+		return error;
+	local_irq_disable();
+	save_processor_state();
+	error = swsusp_arch_suspend();
 	restore_processor_state();
-	do_magic_resume_2();
+	local_irq_enable();
+	return error;
+}
 
- */
 
-asmlinkage void do_magic_suspend_1(void)
+asmlinkage int swsusp_restore(void)
 {
-	mb();
-	barrier();
-	BUG_ON(in_atomic());
-	spin_lock_irq(&suspend_pagedir_lock);
+	BUG_ON (nr_copy_pages_check != nr_copy_pages);
+	BUG_ON (pagedir_order_check != pagedir_order);
+	
+	/* Even mappings of "global" things (vmalloc) need to be fixed */
+	__flush_tlb_global();
+	return 0;
 }
 
-asmlinkage void do_magic_suspend_2(void)
+int swsusp_resume(void)
 {
-	int is_problem;
-	read_swapfiles();
-	device_power_down(3);
-	is_problem = suspend_prepare_image();
-	device_power_up();
-	spin_unlock_irq(&suspend_pagedir_lock);
-	if (!is_problem) {
-		kernel_fpu_end();	/* save_processor_state() does kernel_fpu_begin, and we need to revert it in order to pass in_atomic() checks */
-		BUG_ON(in_atomic());
-		suspend_save_image();
-		suspend_power_down();	/* FIXME: if suspend_power_down is commented out, console is lost after few suspends ?! */
-	}
-
-	printk(KERN_EMERG "%sSuspend failed, trying to recover...\n", name_suspend);
-	MDELAY(1000); /* So user can wait and report us messages if armageddon comes :-) */
-
-	barrier();
-	mb();
-	spin_lock_irq(&suspend_pagedir_lock);	/* Done to disable interrupts */ 
-
-	free_pages((unsigned long) pagedir_nosave, pagedir_order);
-	spin_unlock_irq(&suspend_pagedir_lock);
-
-	device_resume();
-	PRINTK( "Fixing swap signatures... " );
-	mark_swapfiles(((swp_entry_t) {0}), MARK_SWAP_RESUME);
-	PRINTK( "ok\n" );
+	int error;
+	local_irq_disable();
+	save_processor_state();
+	error = swsusp_arch_resume();
+	restore_processor_state();
+	restore_highmem();
+	local_irq_enable();
+	return error;
 }
 
-/*
- * This is main interface to the outside world. It needs to be
- * called from process context.
- */
-int software_suspend(void)
-{
-	int res;
-	if (!software_suspend_enabled)
-		return -EAGAIN;
-
-	software_suspend_enabled = 0;
-	might_sleep();
 
-	if (arch_prepare_suspend()) {
-		printk("%sArchitecture failed to prepare\n", name_suspend);
-		return -EPERM;
-	}		
-	if (pm_prepare_console())
-		printk( "%sCan't allocate a console... proceeding\n", name_suspend);
-	if (!prepare_suspend_processes()) {
-
-		/* At this point, all user processes and "dangerous"
-                   kernel threads are stopped. Free some memory, as we
-                   need half of memory free. */
-
-		free_some_memory();
-		disable_nonboot_cpus();
-		/* Save state of all device drivers, and stop them. */
-		printk("Suspending devices... ");
-		if ((res = device_suspend(3))==0) {
-			/* If stopping device drivers worked, we proceed basically into
-			 * suspend_save_image.
-			 *
-			 * do_magic(0) returns after system is resumed.
-			 *
-			 * do_magic() copies all "used" memory to "free" memory, then
-			 * unsuspends all device drivers, and writes memory to disk
-			 * using normal kernel mechanism.
-			 */
-			do_magic(0);
-		}
-		thaw_processes();
-		enable_nonboot_cpus();
-	} else
-		res = -EBUSY;
-	software_suspend_enabled = 1;
-	MDELAY(1000);
-	pm_restore_console();
-	return res;
-}
 
 /* More restore stuff */
 
@@ -874,14 +893,14 @@ int software_suspend(void)
 /*
  * Returns true if given address/order collides with any orig_address 
  */
-static int does_collide_order(suspend_pagedir_t *pagedir, unsigned long addr,
+static int __init does_collide_order(suspend_pagedir_t *pagedir, unsigned long addr,
 		int order)
 {
 	int i;
 	unsigned long addre = addr + (PAGE_SIZE<<order);
 	
-	for(i=0; i < nr_copy_pages; i++)
-		if((pagedir+i)->orig_address >= addr &&
+	for (i=0; i < nr_copy_pages; i++)
+		if ((pagedir+i)->orig_address >= addr &&
 			(pagedir+i)->orig_address < addre)
 			return 1;
 
@@ -892,7 +911,7 @@ static int does_collide_order(suspend_pagedir_t *pagedir, unsigned long addr,
  * We check here that pagedir & pages it points to won't collide with pages
  * where we're going to restore from the loaded pages later
  */
-static int check_pagedir(void)
+static int __init check_pagedir(void)
 {
 	int i;
 
@@ -910,7 +929,7 @@ static int check_pagedir(void)
 	return 0;
 }
 
-static int relocate_pagedir(void)
+static int __init swsusp_pagedir_relocate(void)
 {
 	/*
 	 * We have to avoid recursion (not to overflow kernel stack),
@@ -923,9 +942,9 @@ static int relocate_pagedir(void)
 
 	printk("Relocating pagedir ");
 
-	if(!does_collide_order(old_pagedir, (unsigned long)old_pagedir, pagedir_order)) {
+	if (!does_collide_order(old_pagedir, (unsigned long)old_pagedir, pagedir_order)) {
 		printk("not necessary\n");
-		return 0;
+		return check_pagedir();
 	}
 
 	while ((m = (void *) __get_free_pages(GFP_ATOMIC, pagedir_order)) != NULL) {
@@ -953,283 +972,253 @@ static int relocate_pagedir(void)
 		free_pages((unsigned long)f, pagedir_order);
 	}
 	printk("|\n");
-	return ret;
+	return check_pagedir();
 }
 
-/*
- * Sanity check if this image makes sense with this kernel/swap context
- * I really don't think that it's foolproof but more than nothing..
+/**
+ *	Using bio to read from swap.
+ *	This code requires a bit more work than just using buffer heads
+ *	but, it is the recommended way for 2.5/2.6.
+ *	The following are to signal the beginning and end of I/O. Bios
+ *	finish asynchronously, while we want them to happen synchronously.
+ *	A simple atomic_t, and a wait loop take care of this problem.
  */
 
-static int sanity_check_failed(char *reason)
-{
-	printk(KERN_ERR "%s%s\n", name_resume, reason);
-	return -EPERM;
-}
+static atomic_t io_done = ATOMIC_INIT(0);
 
-static int sanity_check(struct suspend_header *sh)
+static void start_io(void)
 {
-	if (sh->version_code != LINUX_VERSION_CODE)
-		return sanity_check_failed("Incorrect kernel version");
-	if (sh->num_physpages != num_physpages)
-		return sanity_check_failed("Incorrect memory size");
-	if (strncmp(sh->machine, system_utsname.machine, 8))
-		return sanity_check_failed("Incorrect machine type");
-	if (strncmp(sh->version, system_utsname.version, 20))
-		return sanity_check_failed("Incorrect version");
-	if (sh->num_cpus != num_online_cpus())
-		return sanity_check_failed("Incorrect number of cpus");
-	if (sh->page_size != PAGE_SIZE)
-		return sanity_check_failed("Incorrect PAGE_SIZE");
-	return 0;
+	atomic_set(&io_done,1);
 }
 
-static int bdev_read_page(struct block_device *bdev, long pos, void *buf)
+static int end_io(struct bio * bio, unsigned int num, int err)
 {
-	struct buffer_head *bh;
-	BUG_ON (pos%PAGE_SIZE);
-	bh = __bread(bdev, pos/PAGE_SIZE, PAGE_SIZE);
-	if (!bh || (!bh->b_data)) {
-		return -1;
-	}
-	memcpy(buf, bh->b_data, PAGE_SIZE);	/* FIXME: may need kmap() */
-	BUG_ON(!buffer_uptodate(bh));
-	brelse(bh);
+	atomic_set(&io_done,0);
 	return 0;
-} 
+}
 
-static int bdev_write_page(struct block_device *bdev, long pos, void *buf)
+static void wait_io(void)
 {
-#if 0
-	struct buffer_head *bh;
-	BUG_ON (pos%PAGE_SIZE);
-	bh = __bread(bdev, pos/PAGE_SIZE, PAGE_SIZE);
-	if (!bh || (!bh->b_data)) {
-		return -1;
-	}
-	memcpy(bh->b_data, buf, PAGE_SIZE);	/* FIXME: may need kmap() */
-	BUG_ON(!buffer_uptodate(bh));
-	generic_make_request(WRITE, bh);
-	if (!buffer_uptodate(bh))
-		printk(KERN_CRIT "%sWarning %s: Fixing swap signatures unsuccessful...\n", name_resume, resume_file);
-	wait_on_buffer(bh);
-	brelse(bh);
-	return 0;
-#endif
-	printk(KERN_CRIT "%sWarning %s: Fixing swap signatures unimplemented...\n", name_resume, resume_file);
-	return 0;
+	while(atomic_read(&io_done))
+		io_schedule();
 }
 
-extern dev_t __init name_to_dev_t(const char *line);
 
-static int __init __read_suspend_image(struct block_device *bdev, union diskpage *cur, int noresume)
-{
-	swp_entry_t next;
-	int i, nr_pgdir_pages;
+static struct block_device * resume_bdev;
 
-#define PREPARENEXT \
-	{	next = cur->link.next; \
-		next.val = swp_offset(next) * PAGE_SIZE; \
-        }
+/**
+ *	submit - submit BIO request.
+ *	@rw:	READ or WRITE.
+ *	@off	physical offset of page.
+ *	@page:	page we're reading or writing.
+ *
+ *	Straight from the textbook - allocate and initialize the bio.
+ *	If we're writing, make sure the page is marked as dirty.
+ *	Then submit it and wait.
+ */
 
-	if (bdev_read_page(bdev, 0, cur)) return -EIO;
+static int submit(int rw, pgoff_t page_off, void * page)
+{
+	int error = 0;
+	struct bio * bio;
 
-	if ((!memcmp("SWAP-SPACE",cur->swh.magic.magic,10)) ||
-	    (!memcmp("SWAPSPACE2",cur->swh.magic.magic,10))) {
-		printk(KERN_ERR "%sThis is normal swap space\n", name_resume );
-		return -EINVAL;
+	bio = bio_alloc(GFP_ATOMIC, 1);
+	if (!bio)
+		return -ENOMEM;
+	bio->bi_sector = page_off * (PAGE_SIZE >> 9);
+	bio_get(bio);
+	bio->bi_bdev = resume_bdev;
+	bio->bi_end_io = end_io;
+
+	if (bio_add_page(bio, virt_to_page(page), PAGE_SIZE, 0) < PAGE_SIZE) {
+		printk("swsusp: ERROR: adding page to bio at %ld\n",page_off);
+		error = -EFAULT;
+		goto Done;
 	}
 
-	PREPARENEXT; /* We have to read next position before we overwrite it */
-
-	if (!memcmp("S1",cur->swh.magic.magic,2))
-		memcpy(cur->swh.magic.magic,"SWAP-SPACE",10);
-	else if (!memcmp("S2",cur->swh.magic.magic,2))
-		memcpy(cur->swh.magic.magic,"SWAPSPACE2",10);
-	else {
-		if (noresume)
-			return -EINVAL;
-		panic("%sUnable to find suspended-data signature (%.10s - misspelled?\n", 
-			name_resume, cur->swh.magic.magic);
-	}
-	if (noresume) {
-		/* We don't do a sanity check here: we want to restore the swap
-		   whatever version of kernel made the suspend image;
-		   We need to write swap, but swap is *not* enabled so
-		   we must write the device directly */
-		printk("%s: Fixing swap signatures %s...\n", name_resume, resume_file);
-		bdev_write_page(bdev, 0, cur);
-	}
+	if (rw == WRITE)
+		bio_set_pages_dirty(bio);
+	start_io();
+	submit_bio(rw | (1 << BIO_RW_SYNC), bio);
+	wait_io();
+ Done:
+	bio_put(bio);
+	return error;
+}
 
-	printk( "%sSignature found, resuming\n", name_resume );
-	MDELAY(1000);
+int bio_read_page(pgoff_t page_off, void * page)
+{
+	return submit(READ, page_off, page);
+}
 
-	if (bdev_read_page(bdev, next.val, cur)) return -EIO;
-	if (sanity_check(&cur->sh)) 	/* Is this same machine? */	
-		return -EPERM;
-	PREPARENEXT;
+int bio_write_page(pgoff_t page_off, void * page)
+{
+	return submit(WRITE, page_off, page);
+}
 
-	pagedir_save = cur->sh.suspend_pagedir;
-	nr_copy_pages = cur->sh.num_pbes;
-	nr_pgdir_pages = SUSPEND_PD_PAGES(nr_copy_pages);
-	pagedir_order = get_bitmask_order(nr_pgdir_pages);
+/*
+ * Sanity check if this image makes sense with this kernel/swap context
+ * I really don't think that it's foolproof but more than nothing..
+ */
 
-	pagedir_nosave = (suspend_pagedir_t *)__get_free_pages(GFP_ATOMIC, pagedir_order);
-	if (!pagedir_nosave)
-		return -ENOMEM;
+static const char * __init sanity_check(void)
+{
+	dump_info();
+	if(swsusp_info.version_code != LINUX_VERSION_CODE)
+		return "kernel version";
+	if(swsusp_info.num_physpages != num_physpages)
+		return "memory size";
+	if (strcmp(swsusp_info.uts.sysname,system_utsname.sysname))
+		return "system type";
+	if (strcmp(swsusp_info.uts.release,system_utsname.release))
+		return "kernel release";
+	if (strcmp(swsusp_info.uts.version,system_utsname.version))
+		return "version";
+	if (strcmp(swsusp_info.uts.machine,system_utsname.machine))
+		return "machine";
+	if(swsusp_info.cpus != num_online_cpus())
+		return "number of cpus";
+	return NULL;
+}
 
-	PRINTK( "%sReading pagedir, ", name_resume );
 
-	/* We get pages in reverse order of saving! */
-	for (i=nr_pgdir_pages-1; i>=0; i--) {
-		BUG_ON (!next.val);
-		cur = (union diskpage *)((char *) pagedir_nosave)+i;
-		if (bdev_read_page(bdev, next.val, cur)) return -EIO;
-		PREPARENEXT;
-	}
-	BUG_ON (next.val);
+static int __init check_header(void)
+{
+	const char * reason = NULL;
+	int error;
 
-	if (relocate_pagedir())
-		return -ENOMEM;
-	if (check_pagedir())
-		return -ENOMEM;
+	if ((error = bio_read_page(swp_offset(swsusp_header.swsusp_info), &swsusp_info)))
+		return error;
 
-	printk( "Reading image data (%d pages): ", nr_copy_pages );
-	for(i=0; i < nr_copy_pages; i++) {
-		swp_entry_t swap_address = (pagedir_nosave+i)->swap_address;
-		if (!(i%100))
-			printk( "." );
-		/* You do not need to check for overlaps...
-		   ... check_pagedir already did this work */
-		if (bdev_read_page(bdev, swp_offset(swap_address) * PAGE_SIZE, (char *)((pagedir_nosave+i)->address)))
-			return -EIO;
+ 	/* Is this same machine? */
+	if ((reason = sanity_check())) {
+		printk(KERN_ERR "swsusp: Resume mismatch: %s\n",reason);
+		return -EPERM;
 	}
-	printk( "|\n" );
-	return 0;
+	nr_copy_pages = swsusp_info.image_pages;
+	return error;
 }
 
-static int __init read_suspend_image(const char * specialfile, int noresume)
+static int __init check_sig(void)
 {
-	union diskpage *cur;
-	unsigned long scratch_page = 0;
 	int error;
-	char b[BDEVNAME_SIZE];
-
-	resume_device = name_to_dev_t(specialfile);
-	scratch_page = get_zeroed_page(GFP_ATOMIC);
-	cur = (void *) scratch_page;
-	if (cur) {
-		struct block_device *bdev;
-		printk("Resuming from device %s\n",
-				__bdevname(resume_device, b));
-		bdev = open_by_devnum(resume_device, FMODE_READ);
-		if (IS_ERR(bdev)) {
-			error = PTR_ERR(bdev);
-		} else {
-			set_blocksize(bdev, PAGE_SIZE);
-			error = __read_suspend_image(bdev, cur, noresume);
-			blkdev_put(bdev);
-		}
-	} else error = -ENOMEM;
 
-	if (scratch_page)
-		free_page(scratch_page);
-	switch (error) {
-		case 0:
-			PRINTK("Reading resume file was successful\n");
-			break;
-		case -EINVAL:
-			break;
-		case -EIO:
-			printk( "%sI/O error\n", name_resume);
-			break;
-		case -ENOENT:
-			printk( "%s%s: No such file or directory\n", name_resume, specialfile);
-			break;
-		case -ENOMEM:
-			printk( "%sNot enough memory\n", name_resume);
-			break;
-		default:
-			printk( "%sError %d resuming\n", name_resume, error );
+	memset(&swsusp_header, 0, sizeof(swsusp_header));
+	if ((error = bio_read_page(0, &swsusp_header)))
+		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);
+	} else { 
+		pr_debug(KERN_ERR "swsusp: Invalid partition type.\n");
+		return -EINVAL;
 	}
-	MDELAY(1000);
+	if (!error)
+		pr_debug("swsusp: Signature found, resuming\n");
 	return error;
 }
 
 /**
- *	software_resume - Resume from a saved image.
- *
- *	Called as a late_initcall (so all devices are discovered and 
- *	initialized), we call swsusp to see if we have a saved image or not.
- *	If so, we quiesce devices, then restore the saved image. We will 
- *	return above (in pm_suspend_disk() ) if everything goes well. 
- *	Otherwise, we fail gracefully and return to the normally 
- *	scheduled program.
+ *	swsusp_read_data - Read image pages from swap.
  *
+ *	You do not need to check for overlaps, check_pagedir()
+ *	already did that.
  */
-static int __init software_resume(void)
-{
-	if (num_online_cpus() > 1) {
-		printk(KERN_WARNING "Software Suspend has malfunctioning SMP support. Disabled :(\n");	
-		return -EINVAL;
-	}
-	/* We enable the possibility of machine suspend */
-	software_suspend_enabled = 1;
-	if (!resume_status)
-		return 0;
 
-	printk( "%s", name_resume );
-	if (resume_status == NORESUME) {
-		if(resume_file[0])
-			read_suspend_image(resume_file, 1);
-		printk( "disabled\n" );
-		return 0;
-	}
-	MDELAY(1000);
+static int __init data_read(void)
+{
+	struct pbe * p;
+	int error;
+	int i;
 
-	if (pm_prepare_console())
-		printk("swsusp: Can't allocate a console... proceeding\n");
+	if ((error = swsusp_pagedir_relocate()))
+		return error;
 
-	if (!resume_file[0] && resume_status == RESUME_SPECIFIED) {
-		printk( "suspension device unspecified\n" );
-		return -EINVAL;
+	printk( "Reading image data (%d pages): ", nr_copy_pages );
+	for(i = 0, p = pagedir_nosave; i < nr_copy_pages && !error; i++, p++) {
+		if (!(i%100))
+			printk( "." );
+		error = bio_read_page(swp_offset(p->swap_address),
+				  (void *)p->address);
 	}
+	printk(" %d done.\n",i);
+	return error;
 
-	printk( "resuming from %s\n", resume_file);
-	if (read_suspend_image(resume_file, 0))
-		goto read_failure;
-	/* FIXME: Should we stop processes here, just to be safer? */
-	disable_nonboot_cpus();
-	device_suspend(3);
-	do_magic(1);
-	panic("This never returns");
-
-read_failure:
-	pm_restore_console();
-	return 0;
 }
 
-late_initcall(software_resume);
+extern dev_t __init name_to_dev_t(const char *line);
 
-static int __init resume_setup(char *str)
+static int __init read_pagedir(void)
 {
-	if (resume_status == NORESUME)
-		return 1;
+	unsigned long addr;
+	int i, n = swsusp_info.pagedir_pages;
+	int error = 0;
 
-	strncpy( resume_file, str, 255 );
-	resume_status = RESUME_SPECIFIED;
+	pagedir_order = get_bitmask_order(n);
 
-	return 1;
+	addr =__get_free_pages(GFP_ATOMIC, pagedir_order);
+	if (!addr)
+		return -ENOMEM;
+	pagedir_nosave = (struct pbe *)addr;
+
+	pr_debug("pmdisk: Reading pagedir (%d Pages)\n",n);
+
+	for (i = 0; i < n && !error; i++, addr += PAGE_SIZE) {
+		unsigned long offset = swp_offset(swsusp_info.pagedir[i]);
+		if (offset)
+			error = bio_read_page(offset, (void *)addr);
+		else
+			error = -EFAULT;
+	}
+	if (error)
+		free_pages((unsigned long)pagedir_nosave, pagedir_order);
+	return error;
 }
 
-static int __init noresume_setup(char *str)
+static int __init read_suspend_image(void)
 {
-	resume_status = NORESUME;
-	return 1;
+	int error = 0;
+
+	if ((error = check_sig()))
+		return error;
+	if ((error = check_header()))
+		return error;
+	if ((error = read_pagedir()))
+		return error;
+	if ((error = data_read()))
+		free_pages((unsigned long)pagedir_nosave, pagedir_order);
+	return error;
 }
 
-__setup("noresume", noresume_setup);
-__setup("resume=", resume_setup);
+/**
+ *	pmdisk_read - Read saved image from swap.
+ */
+
+int __init swsusp_read(void)
+{
+	int error;
+
+	if (!strlen(resume_file))
+		return -ENOENT;
 
-EXPORT_SYMBOL(software_suspend);
-EXPORT_SYMBOL(software_suspend_enabled);
+	resume_device = name_to_dev_t(resume_file);
+	pr_debug("swsusp: Resume From Partition: %s\n", resume_file);
+
+	resume_bdev = open_by_devnum(resume_device, FMODE_READ);
+	if (!IS_ERR(resume_bdev)) {
+		set_blocksize(resume_bdev, PAGE_SIZE);
+		error = read_suspend_image();
+		blkdev_put(resume_bdev);
+	} else
+		error = PTR_ERR(resume_bdev);
+
+	if (!error)
+		pr_debug("Reading resume file was successful\n");
+	else
+		pr_debug("pmdisk: Error %d resuming\n", error);
+	return error;
+}