[PATCH] swsusp: cleanup

 - use list_for_each in head_of_free_region
 - cleanups from 2.4
 - fix for usb
 - kill broken queueing

Documentation/kernel-parameters.txt

@@ -45,6 +45,7 @@ restrictions referred to are that the relevant option is valid if:
 	SERIAL	Serial support is enabled.
 	SMP 	The kernel is an SMP kernel.
 	SOUND	Appropriate sound system support is enabled.
+	SWSUSP  Software suspension is enabled.
 	V4L	Video For Linux support is enabled.
 	VGA 	The VGA console has been enabled.
 	VT	Virtual terminal support is enabled.
@@ -392,6 +393,8 @@ running once the system is up.
 			initial RAM disk.
 
 	nointroute	[IA-64]
+
+	noresume	[SWSUSP] Disables resume and restore original swap space.
  
 	no-scroll	[VGA]
 
@@ -519,6 +522,8 @@ running once the system is up.
 
 	reserve=	[KNL,BUGS] force the kernel to ignore some iomem area.
 
+	resume=		[SWSUSP] specify the partition device for software suspension.
+
 	riscom8=	[HW,SERIAL]
 
 	ro		[KNL] Mount root device read-only on boot.

arch/i386/kernel/acpi_wakeup.S

@@ -262,7 +262,6 @@ ENTRY(saved_esp)	.long	0
 
 ENTRY(saved_magic)	.long	0
 ENTRY(saved_magic2)	.long	0	
-ENTRY(saved_videomode)	.long	0
 
 ALIGN
 # saved registers

arch/i386/kernel/setup.c

@@ -168,6 +168,8 @@ void __init visws_get_board_type_and_rev(void);
 static int disable_x86_serial_nr __initdata = 1;
 static int disable_x86_fxsr __initdata = 0;
 
+unsigned long saved_videomode;
+
 extern unsigned long saved_videomode;
 
 /*
@@ -684,10 +686,8 @@ void __init setup_arch(char **cmdline_p)
  	drive_info = DRIVE_INFO;
  	screen_info = SCREEN_INFO;
 	apm_info.bios = APM_BIOS_INFO;
-#ifdef CONFIG_ACPI_SLEEP
 	saved_videomode = VIDEO_MODE;
 	printk("Video mode to be used for restore is %lx\n", saved_videomode);
-#endif
 	if( SYS_DESC_TABLE.length != 0 ) {
 		MCA_bus = SYS_DESC_TABLE.table[3] &0x2;
 		machine_id = SYS_DESC_TABLE.table[0];

drivers/acpi/acpi_system.c

@@ -709,11 +709,9 @@ acpi_system_write_sleep (
 	if (!system->states[state])
 		return_VALUE(-ENODEV);
 
-	
 #ifdef CONFIG_SOFTWARE_SUSPEND
 	if (state == 4) {
-		/* We are working from process context, that's why we may call it directly. */ 
-		do_software_suspend();
+		software_suspend();
 		return_VALUE(count);
 	}
 #endif

drivers/block/floppy.c

@@ -4220,14 +4220,15 @@ static int __init floppy_setup(char *str)
 
 static int have_no_fdc= -ENODEV;
 
-static struct device device_floppy;
+static struct device device_floppy = {
+	name:		"floppy",
+	bus_id:		"03?0",
+};
 
 int __init floppy_init(void)
 {
 	int i,unit,drive;
 
-	strcpy(device_floppy.name, "floppy");
-	strcpy(device_floppy.bus_id, "03?0");
 	register_sys_device(&device_floppy);
 
 	raw_cmd = NULL;

drivers/block/loop.c

@@ -577,7 +577,9 @@ static int loop_thread(void *data)
 	daemonize();
 
 	sprintf(current->comm, "loop%d", lo->lo_number);
-	current->flags |= PF_IOTHREAD;
+	current->flags |= PF_IOTHREAD;	/* loop can be used in an encrypted device
+					   hence, it mustn't be stopped at all because it could
+					   be indirectly used during suspension */
 
 	spin_lock_irq(&current->sigmask_lock);
 	sigfillset(&current->blocked);

drivers/usb/core/hub.c

@@ -25,6 +25,7 @@
 #endif
 #include <linux/usb.h>
 #include <linux/usbdevice_fs.h>
+#include <linux/suspend.h>
 
 #include <asm/semaphore.h>
 #include <asm/uaccess.h>
@@ -1064,6 +1065,8 @@ static int usb_hub_thread(void *__hub)
 	/* Send me a signal to get me die (for debugging) */
 	do {
 		usb_hub_events();
+		if (current->flags & PF_FREEZE)
+			refrigerator(PF_IOTHREAD);
 		wait_event_interruptible(khubd_wait, !list_empty(&hub_event_list)); 
 	} while (!signal_pending(current));
 

fs/buffer.c

@@ -122,8 +122,6 @@ void unlock_buffer(struct buffer_head *bh)
 	wake_up_buffer(bh);
 }
 
-DECLARE_TASK_QUEUE(tq_bdflush);
-
 /*
  * Block until a buffer comes unlocked.  This doesn't stop it
  * from becoming locked again - you have to lock it yourself

fs/jbd/journal.c

@@ -226,7 +226,6 @@ int kjournald(void *arg)
 			journal->j_commit_interval / HZ);
 	list_add(&journal->j_all_journals, &all_journals);
 
-	current->flags |= PF_KERNTHREAD;
 	/* And now, wait forever for commit wakeup events. */
 	while (1) {
 		if (journal->j_flags & JFS_UNMOUNT)

fs/reiserfs/journal.c

@@ -1888,7 +1888,6 @@ static int reiserfs_journal_commit_thread(void *nullp) {
   spin_unlock_irq(&current->sigmask_lock);
 
   sprintf(current->comm, "kreiserfsd") ;
-  current->flags |= PF_KERNTHREAD;
   lock_kernel() ;
   while(1) {
 

include/asm-i386/suspend.h

-#ifndef __ASM_I386_SUSPEND_H
-#define __ASM_I386_SUSPEND_H
-#endif
-
 /*
  * Copyright 2001-2002 Pavel Machek <pavel@suse.cz>
  * Based on code
  * Copyright 2001 Patrick Mochel <mochel@osdl.org>
  */
-#if defined(SUSPEND_C) || defined(ACPI_C)
 #include <asm/desc.h>
 #include <asm/i387.h>
 
@@ -225,7 +220,6 @@ static inline void restore_processor_context (void)
 	do_fpu_end();
 }
 
-#endif
 #ifdef SUSPEND_C
 /* Local variables for do_magic */
 static int loop __nosavedata = 0;

include/linux/suspend.h

 #ifndef _LINUX_SWSUSP_H
 #define _LINUX_SWSUSP_H
 
+#if defined(SUSPEND_C) || defined(ACPI_C)
 #include <asm/suspend.h>
+#endif
 #include <linux/swap.h>
 #include <linux/notifier.h>
 #include <linux/config.h>

include/linux/tqueue.h

@@ -66,7 +66,7 @@ typedef struct list_head task_queue;
 #define DECLARE_TASK_QUEUE(q)	LIST_HEAD(q)
 #define TQ_ACTIVE(q)		(!list_empty(&q))
 
-extern task_queue tq_timer, tq_immediate, tq_bdflush;
+extern task_queue tq_timer, tq_immediate;
 
 /*
  * To implement your own list of active bottom halfs, use the following

kernel/suspend.c

 /*
- * linux/kernel/swsusp.c
+ * linux/kernel/suspend.c
  *
  * This file is to realize architecture-independent
  * machine suspend feature using pretty near only high-level routines
@@ -34,13 +34,6 @@
  * For TODOs,FIXMEs also look in Documentation/swsusp.txt
  */
 
-/*
- * TODO:
- *
- * - we should launch a kernel_thread to process suspend request, cleaning up
- * bdflush from this task. (check apm.c for something similar).
- */
-
 #include <linux/module.h>
 #include <linux/mm.h>
 #include <linux/swapctl.h>
@@ -65,6 +58,7 @@
 #include <linux/swap.h>
 #include <linux/pm.h>
 #include <linux/device.h>
+#include <linux/buffer_head.h>
 
 #include <asm/uaccess.h>
 #include <asm/mmu_context.h>
@@ -78,7 +72,7 @@ unsigned char software_suspend_enabled = 0;
    we probably do not take enough locks for switching consoles, etc,
    so bad things might happen.
 */
-#ifndef CONFIG_VT
+#if !defined(CONFIG_VT) || !defined(CONFIG_VT_CONSOLE)
 #undef SUSPEND_CONSOLE
 #endif
 
@@ -152,7 +146,7 @@ static const char *name_resume = "Resume Machine: ";
 #define	DEBUG_DEFAULT	1
 #undef	DEBUG_PROCESS
 #undef	DEBUG_SLOW
-#define TEST_SWSUSP 0		/* Set to 1 to reboot instead of halt machine after suspension */
+#define TEST_SWSUSP 1		/* Set to 1 to reboot instead of halt machine after suspension */
 
 #ifdef DEBUG_DEFAULT
 #define PRINTD(func, f, a...)	\
@@ -775,7 +769,10 @@ void suspend_power_down(void)
 		machine_restart(NULL);
 	else
 #endif
+	{
+		device_shutdown();
 		machine_power_off();
+	}
 
 	printk(KERN_EMERG "%sProbably not capable for powerdown. System halted.\n", name_suspend);
 	machine_halt();
@@ -783,9 +780,6 @@ void suspend_power_down(void)
 	/* NOTREACHED */
 }
 
-/* forward decl */
-void do_software_suspend(void);
-
 /*
  * Magic happens here
  */
@@ -822,7 +816,6 @@ static void do_magic_resume_2(void)
 #ifdef SUSPEND_CONSOLE
 	update_screen(fg_console);	/* Hmm, is this the problem? */
 #endif
-	suspend_tq.routine = (void *)do_software_suspend;
 }
 
 static void do_magic_suspend_1(void)
@@ -851,7 +844,6 @@ static void do_magic_suspend_2(void)
 	drivers_resume(RESUME_PHASE1);
 	spin_unlock_irq(&suspend_pagedir_lock);
 	mark_swapfiles(((swp_entry_t) {0}), MARK_SWAP_RESUME);
-	suspend_tq.routine = (void *)do_software_suspend;
 	printk(KERN_WARNING "%sLeaving do_magic_suspend_2...\n", name_suspend);	
 }
 
@@ -859,43 +851,38 @@ static void do_magic_suspend_2(void)
 #include <asm/suspend.h>
 
 /*
- * This function is triggered using process bdflush. We try to swap out as
- * much as we can then make a copy of the occupied pages in memory so we can
- * make a copy of kernel state atomically, the I/O needed by saving won't
- * bother us anymore.
+ * We try to swap out as much as we can then make a copy of the
+ * occupied pages in memory so we can make a copy of kernel state
+ * atomically, the I/O needed by saving won't bother us anymore. 
  */
 void do_software_suspend(void)
 {
 	arch_prepare_suspend();
-	if (!prepare_suspend_console()) {
-		if (!prepare_suspend_processes()) {
-			free_some_memory();
-
-			/* No need to invalidate any vfsmnt list -- they will be valid after resume, anyway.
-			 *
-			 * We sync here -- so you have consistent filesystem state when things go wrong.
-			 * -- so that noone writes to disk after we do atomic copy of data.
-			 */
-			PRINTS("Syncing disks before copy\n");
-			do_suspend_sync();
-			if(drivers_suspend()==0)
-				do_magic(0);			/* This function returns after machine woken up from resume */
-			PRINTR("Restarting processes...\n");
-			thaw_processes();
-		}
+	if (prepare_suspend_console())
+		printk( "Can't allocate a console... proceeding\n");
+	if (!prepare_suspend_processes()) {
+		free_some_memory();
+		
+		/* No need to invalidate any vfsmnt list -- they will be valid after resume, anyway.
+		 *
+		 * We sync here -- so you have consistent filesystem state when things go wrong.
+		 * -- so that noone writes to disk after we do atomic copy of data.
+		 */
+		PRINTS("Syncing disks before copy\n");
+		do_suspend_sync();
+		if(drivers_suspend()==0)
+			do_magic(0);			/* This function returns after machine woken up from resume */
+		PRINTR("Restarting processes...\n");
+		thaw_processes();
 	}
 	software_suspend_enabled = 1;
 	MDELAY(1000);
 	restore_console ();
 }
 
-struct tq_struct suspend_tq =
-	{ routine: (void *)(void *)do_software_suspend, 
-	  data: 0 };
-
 /*
- * This is the trigger function, we must queue ourself since we
- * can be called from interrupt && bdflush context is needed
+ * This is main interface to the outside world. It needs to be
+ * called from process context.
  */
 void software_suspend(void)
 {
@@ -903,8 +890,8 @@ void software_suspend(void)
 		return;
 
 	software_suspend_enabled = 0;
-	queue_task(&suspend_tq, &tq_bdflush);
-	wakeup_bdflush();
+	BUG_ON(in_interrupt());
+	do_software_suspend();
 }
 
 /* More restore stuff */
@@ -1123,6 +1110,7 @@ static int resume_try_to_read(const char * specialfile, int noresume)
 	pagedir_order = get_bitmask_order(nr_pgdir_pages);
 
 	error = -ENOMEM;
+	free_page((unsigned long) cur);
 	pagedir_nosave = (suspend_pagedir_t *)__get_free_pages(GFP_ATOMIC, pagedir_order);
 	if(!pagedir_nosave)
 		goto resume_read_error;

mm/page_alloc.c

@@ -247,42 +247,26 @@ static struct page * rmqueue(zone_t *zone, unsigned int order)
 }
 
 #ifdef CONFIG_SOFTWARE_SUSPEND
-int is_head_of_free_region(struct page *p)
+int is_head_of_free_region(struct page *page)
 {
-	pg_data_t *pgdat = pgdat_list;
-	unsigned type;
-	unsigned long flags;
-
-	for (type=0;type < MAX_NR_ZONES; type++) {
-		zone_t *zone = pgdat->node_zones + type;
-		int order = MAX_ORDER - 1;
-		free_area_t *area;
-		struct list_head *head, *curr;
-		spin_lock_irqsave(&zone->lock, flags);	/* Should not matter as we need quiescent system for suspend anyway, but... */
-
-		do {
-			area = zone->free_area + order;
-			head = &area->free_list;
-			curr = head;
+        zone_t *zone = page_zone(page);
+        unsigned long flags;
+	int order;
+	list_t *curr;
 
-			for(;;) {
-				if(!curr) {
-//					printk("FIXME: this should not happen but it does!!!");
-					break;
-				}
-				if(p != memlist_entry(curr, struct page, list)) {
-					curr = memlist_next(curr);
-					if (curr == head)
-						break;
-					continue;
-				}
+	/*
+	 * Should not matter as we need quiescent system for
+	 * suspend anyway, but...
+	 */
+	spin_lock_irqsave(&zone->lock, flags);
+	for (order = MAX_ORDER - 1; order >= 0; --order)
+		list_for_each(curr, &zone->free_area[order].free_list)
+			if (page == list_entry(curr, struct page, list)) {
+				spin_unlock_irqrestore(&zone->lock, flags);
 				return 1 << order;
 			}
-		} while(order--);
-		spin_unlock_irqrestore(&zone->lock, flags);
-
-	}
-	return 0;
+	spin_unlock_irqrestore(&zone->lock, flags);
+        return 0;
 }
 #endif /* CONFIG_SOFTWARE_SUSPEND */
 

mm/pdflush.c

@@ -91,7 +91,7 @@ static int __pdflush(struct pdflush_work *my_work)
 	recalc_sigpending();
 	spin_unlock_irq(&current->sigmask_lock);
 
-	current->flags |= PF_FLUSHER | PF_KERNTHREAD;
+	current->flags |= PF_FLUSHER;
 	my_work->fn = NULL;
 	my_work->who = current;
 
@@ -106,11 +106,8 @@ static int __pdflush(struct pdflush_work *my_work)
 		set_current_state(TASK_INTERRUPTIBLE);
 		spin_unlock_irq(&pdflush_lock);
 
-#ifdef CONFIG_SOFTWARE_SUSPEND
-		run_task_queue(&tq_bdflush);
 		if (current->flags & PF_FREEZE)
 			refrigerator(PF_IOTHREAD);
-#endif
 		schedule();
 
 		if (my_work->fn)

mm/vmscan.c

@@ -791,7 +791,7 @@ int kswapd(void *unused)
 	 * us from recursively trying to free more memory as we're
 	 * trying to free the first piece of memory in the first place).
 	 */
-	tsk->flags |= PF_MEMALLOC | PF_KERNTHREAD;
+	tsk->flags |= PF_MEMALLOC;
 
 	/*
 	 * Kswapd main loop.