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
+copy_loop:
+	movl	4(%ebx,%edx),%edi
+	movl	(%ebx,%edx),%esi
+
+	movl	$1024, %ecx
+	rep
+	movsl
+
+	incl	%eax
+	addl	$16, %edx
 	cmpl	nr_copy_pages,%eax
 	jb copy_loop
+	.p2align 4,,7
 
-copy_done:
-	movl $__USER_DS,%eax
-
-	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"
@@ -23,12 +21,15 @@
 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.
@@ -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/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 = {	\