fedora core 6 1.2949 + vserver 2.2.0
[linux-2.6.git] / kernel / power / swsusp.c
index 2d9d08f..31aa039 100644 (file)
  * Fixed runaway init
  *
  * Rafael J. Wysocki <rjw@sisk.pl>
- * Added the swap map data structure and reworked the handling of swap
+ * Reworked the freeing of memory and the handling of swap
  *
  * More state savers are welcome. Especially for the scsi layer...
  *
  * For TODOs,FIXMEs also look in Documentation/power/swsusp.txt
  */
 
-#include <linux/module.h>
 #include <linux/mm.h>
 #include <linux/suspend.h>
-#include <linux/smp_lock.h>
-#include <linux/file.h>
-#include <linux/utsname.h>
-#include <linux/version.h>
-#include <linux/delay.h>
-#include <linux/bitops.h>
 #include <linux/spinlock.h>
-#include <linux/genhd.h>
 #include <linux/kernel.h>
 #include <linux/major.h>
 #include <linux/swap.h>
 #include <linux/pm.h>
-#include <linux/device.h>
-#include <linux/buffer_head.h>
 #include <linux/swapops.h>
 #include <linux/bootmem.h>
 #include <linux/syscalls.h>
 #include <linux/highmem.h>
-#include <linux/bio.h>
-
-#include <asm/uaccess.h>
-#include <asm/mmu_context.h>
-#include <asm/pgtable.h>
-#include <asm/tlbflush.h>
-#include <asm/io.h>
+#include <linux/time.h>
 
 #include "power.h"
 
  */
 unsigned long image_size = 500 * 1024 * 1024;
 
+int in_suspend __nosavedata = 0;
+
 #ifdef CONFIG_HIGHMEM
 unsigned int count_highmem_pages(void);
-int save_highmem(void);
 int restore_highmem(void);
 #else
-static int save_highmem(void) { return 0; }
-static int restore_highmem(void) { return 0; }
-static unsigned int count_highmem_pages(void) { return 0; }
+static inline int restore_highmem(void) { return 0; }
+static inline unsigned int count_highmem_pages(void) { return 0; }
 #endif
 
-extern char resume_file[];
-
-#define SWSUSP_SIG     "S1SUSPEND"
-
-static struct swsusp_header {
-       char reserved[PAGE_SIZE - 20 - sizeof(swp_entry_t)];
-       swp_entry_t image;
-       char    orig_sig[10];
-       char    sig[10];
-} __attribute__((packed, aligned(PAGE_SIZE))) swsusp_header;
-
-static struct swsusp_info swsusp_info;
-
-/*
- * Saving part...
- */
-
-static unsigned short root_swap = 0xffff;
-
-static int mark_swapfiles(swp_entry_t start)
-{
-       int error;
-
-       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.image = start;
-               error = rw_swap_page_sync(WRITE,
-                                         swp_entry(root_swap, 0),
-                                         virt_to_page((unsigned long)
-                                                      &swsusp_header));
-       } else {
-               pr_debug("swsusp: Partition is not swap space.\n");
-               error = -ENODEV;
-       }
-       return error;
-}
-
-/*
- * Check whether the swap device is the specified resume
- * device, irrespective of whether they are specified by
- * identical names.
- *
- * (Thus, device inode aliasing is allowed.  You can say /dev/hda4
- * instead of /dev/ide/host0/bus0/target0/lun0/part4 [if using devfs]
- * and they'll be considered the same device.  This is *necessary* for
- * devfs, since the resume code can only recognize the form /dev/hda4,
- * but the suspend code would see the long name.)
- */
-static inline int is_resume_device(const struct swap_info_struct *swap_info)
-{
-       struct file *file = swap_info->swap_file;
-       struct inode *inode = file->f_dentry->d_inode;
-
-       return S_ISBLK(inode->i_mode) &&
-               swsusp_resume_device == MKDEV(imajor(inode), iminor(inode));
-}
-
-static int swsusp_swap_check(void) /* This is called before saving image */
-{
-       int i;
-
-       spin_lock(&swap_lock);
-       for (i = 0; i < MAX_SWAPFILES; i++) {
-               if (!(swap_info[i].flags & SWP_WRITEOK))
-                       continue;
-               if (!swsusp_resume_device || is_resume_device(swap_info + i)) {
-                       spin_unlock(&swap_lock);
-                       root_swap = i;
-                       return 0;
-               }
-       }
-       spin_unlock(&swap_lock);
-       return -ENODEV;
-}
-
 /**
- *     write_page - Write one page to a fresh swap location.
- *     @addr:  Address we're writing.
- *     @loc:   Place to store the entry we used.
+ *     The following functions are used for tracing the allocated
+ *     swap pages, so that they can be freed in case of an error.
  *
- *     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.
+ *     The functions operate on a linked bitmap structure defined
+ *     in power.h
  */
-static int write_page(unsigned long addr, swp_entry_t *loc)
-{
-       swp_entry_t entry;
-       int error = -ENOSPC;
 
-       entry = get_swap_page_of_type(root_swap);
-       if (swp_offset(entry)) {
-               error = rw_swap_page_sync(WRITE, entry, virt_to_page(addr));
-               if (!error || error == -EIO)
-                       *loc = entry;
-       }
-       return error;
-}
-
-/**
- *     Swap map-handling functions
- *
- *     The swap map is a data structure used for keeping track of each page
- *     written to the swap.  It consists of many swap_map_page structures
- *     that contain each an array of MAP_PAGE_SIZE swap entries.
- *     These structures are linked together with the help of either the
- *     .next (in memory) or the .next_swap (in swap) member.
- *
- *     The swap map is created during suspend.  At that time we need to keep
- *     it in memory, because we have to free all of the allocated swap
- *     entries if an error occurs.  The memory needed is preallocated
- *     so that we know in advance if there's enough of it.
- *
- *     The first swap_map_page structure is filled with the swap entries that
- *     correspond to the first MAP_PAGE_SIZE data pages written to swap and
- *     so on.  After the all of the data pages have been written, the order
- *     of the swap_map_page structures in the map is reversed so that they
- *     can be read from swap in the original order.  This causes the data
- *     pages to be loaded in exactly the same order in which they have been
- *     saved.
- *
- *     During resume we only need to use one swap_map_page structure
- *     at a time, which means that we only need to use two memory pages for
- *     reading the image - one for reading the swap_map_page structures
- *     and the second for reading the data pages from swap.
- */
-
-#define MAP_PAGE_SIZE  ((PAGE_SIZE - sizeof(swp_entry_t) - sizeof(void *)) \
-                       / sizeof(swp_entry_t))
-
-struct swap_map_page {
-       swp_entry_t             entries[MAP_PAGE_SIZE];
-       swp_entry_t             next_swap;
-       struct swap_map_page    *next;
-};
-
-static inline void free_swap_map(struct swap_map_page *swap_map)
+void free_bitmap(struct bitmap_page *bitmap)
 {
-       struct swap_map_page *swp;
+       struct bitmap_page *bp;
 
-       while (swap_map) {
-               swp = swap_map->next;
-               free_page((unsigned long)swap_map);
-               swap_map = swp;
+       while (bitmap) {
+               bp = bitmap->next;
+               free_page((unsigned long)bitmap);
+               bitmap = bp;
        }
 }
 
-static struct swap_map_page *alloc_swap_map(unsigned int nr_pages)
+struct bitmap_page *alloc_bitmap(unsigned int nr_bits)
 {
-       struct swap_map_page *swap_map, *swp;
-       unsigned n = 0;
+       struct bitmap_page *bitmap, *bp;
+       unsigned int n;
 
-       if (!nr_pages)
+       if (!nr_bits)
                return NULL;
 
-       pr_debug("alloc_swap_map(): nr_pages = %d\n", nr_pages);
-       swap_map = (struct swap_map_page *)get_zeroed_page(GFP_ATOMIC);
-       swp = swap_map;
-       for (n = MAP_PAGE_SIZE; n < nr_pages; n += MAP_PAGE_SIZE) {
-               swp->next = (struct swap_map_page *)get_zeroed_page(GFP_ATOMIC);
-               swp = swp->next;
-               if (!swp) {
-                       free_swap_map(swap_map);
+       bitmap = (struct bitmap_page *)get_zeroed_page(GFP_KERNEL);
+       bp = bitmap;
+       for (n = BITMAP_PAGE_BITS; n < nr_bits; n += BITMAP_PAGE_BITS) {
+               bp->next = (struct bitmap_page *)get_zeroed_page(GFP_KERNEL);
+               bp = bp->next;
+               if (!bp) {
+                       free_bitmap(bitmap);
                        return NULL;
                }
        }
-       return swap_map;
+       return bitmap;
 }
 
-/**
- *     reverse_swap_map - reverse the order of pages in the swap map
- *     @swap_map
- */
-
-static inline struct swap_map_page *reverse_swap_map(struct swap_map_page *swap_map)
+static int bitmap_set(struct bitmap_page *bitmap, unsigned long bit)
 {
-       struct swap_map_page *prev, *next;
+       unsigned int n;
 
-       prev = NULL;
-       while (swap_map) {
-               next = swap_map->next;
-               swap_map->next = prev;
-               prev = swap_map;
-               swap_map = next;
+       n = BITMAP_PAGE_BITS;
+       while (bitmap && n <= bit) {
+               n += BITMAP_PAGE_BITS;
+               bitmap = bitmap->next;
        }
-       return prev;
-}
-
-/**
- *     free_swap_map_entries - free the swap entries allocated to store
- *     the swap map @swap_map (this is only called in case of an error)
- */
-static inline void free_swap_map_entries(struct swap_map_page *swap_map)
-{
-       while (swap_map) {
-               if (swap_map->next_swap.val)
-                       swap_free(swap_map->next_swap);
-               swap_map = swap_map->next;
-       }
-}
-
-/**
- *     save_swap_map - save the swap map used for tracing the data pages
- *     stored in the swap
- */
-
-static int save_swap_map(struct swap_map_page *swap_map, swp_entry_t *start)
-{
-       swp_entry_t entry = (swp_entry_t){0};
-       int error;
-
-       while (swap_map) {
-               swap_map->next_swap = entry;
-               if ((error = write_page((unsigned long)swap_map, &entry)))
-                       return error;
-               swap_map = swap_map->next;
+       if (!bitmap)
+               return -EINVAL;
+       n -= BITMAP_PAGE_BITS;
+       bit -= n;
+       n = 0;
+       while (bit >= BITS_PER_CHUNK) {
+               bit -= BITS_PER_CHUNK;
+               n++;
        }
-       *start = entry;
+       bitmap->chunks[n] |= (1UL << bit);
        return 0;
 }
 
-/**
- *     free_image_entries - free the swap entries allocated to store
- *     the image data pages (this is only called in case of an error)
- */
-
-static inline void free_image_entries(struct swap_map_page *swp)
-{
-       unsigned k;
-
-       while (swp) {
-               for (k = 0; k < MAP_PAGE_SIZE; k++)
-                       if (swp->entries[k].val)
-                               swap_free(swp->entries[k]);
-               swp = swp->next;
-       }
-}
-
-/**
- *     The swap_map_handle structure is used for handling the swap map in
- *     a file-alike way
- */
-
-struct swap_map_handle {
-       struct swap_map_page *cur;
-       unsigned int k;
-};
-
-static inline void init_swap_map_handle(struct swap_map_handle *handle,
-                                        struct swap_map_page *map)
+sector_t alloc_swapdev_block(int swap, struct bitmap_page *bitmap)
 {
-       handle->cur = map;
-       handle->k = 0;
-}
-
-static inline int swap_map_write_page(struct swap_map_handle *handle,
-                                      unsigned long addr)
-{
-       int error;
+       unsigned long offset;
 
-       error = write_page(addr, handle->cur->entries + handle->k);
-       if (error)
-               return error;
-       if (++handle->k >= MAP_PAGE_SIZE) {
-               handle->cur = handle->cur->next;
-               handle->k = 0;
+       offset = swp_offset(get_swap_page_of_type(swap));
+       if (offset) {
+               if (bitmap_set(bitmap, offset))
+                       swap_free(swp_entry(swap, offset));
+               else
+                       return swapdev_block(swap, offset);
        }
        return 0;
 }
 
-/**
- *     save_image_data - save the data pages pointed to by the PBEs
- *     from the list @pblist using the swap map handle @handle
- *     (assume there are @nr_pages data pages to save)
- */
-
-static int save_image_data(struct pbe *pblist,
-                           struct swap_map_handle *handle,
-                           unsigned int nr_pages)
-{
-       unsigned int m;
-       struct pbe *p;
-       int error = 0;
-
-       printk("Saving image data pages (%u pages) ...     ", nr_pages);
-       m = nr_pages / 100;
-       if (!m)
-               m = 1;
-       nr_pages = 0;
-       for_each_pbe (p, pblist) {
-               error = swap_map_write_page(handle, p->address);
-               if (error)
-                       break;
-               if (!(nr_pages % m))
-                       printk("\b\b\b\b%3d%%", nr_pages / m);
-               nr_pages++;
-       }
-       if (!error)
-               printk("\b\b\b\bdone\n");
-       return error;
-}
-
-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: Total: %ld Pages\n", swsusp_info.pages);
-}
-
-static void init_header(unsigned int nr_pages)
-{
-       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.cpus = num_online_cpus();
-       swsusp_info.image_pages = nr_pages;
-       swsusp_info.pages = nr_pages +
-               ((nr_pages * sizeof(long) + PAGE_SIZE - 1) >> PAGE_SHIFT) + 1;
-}
-
-/**
- *     pack_orig_addresses - the .orig_address fields of the PBEs from the
- *     list starting at @pbe are stored in the array @buf[] (1 page)
- */
-
-static inline struct pbe *pack_orig_addresses(unsigned long *buf,
-                                              struct pbe *pbe)
+void free_all_swap_pages(int swap, struct bitmap_page *bitmap)
 {
-       int j;
+       unsigned int bit, n;
+       unsigned long test;
 
-       for (j = 0; j < PAGE_SIZE / sizeof(long) && pbe; j++) {
-               buf[j] = pbe->orig_address;
-               pbe = pbe->next;
+       bit = 0;
+       while (bitmap) {
+               for (n = 0; n < BITMAP_PAGE_CHUNKS; n++)
+                       for (test = 1UL; test; test <<= 1) {
+                               if (bitmap->chunks[n] & test)
+                                       swap_free(swp_entry(swap, bit));
+                               bit++;
+                       }
+               bitmap = bitmap->next;
        }
-       if (!pbe)
-               for (; j < PAGE_SIZE / sizeof(long); j++)
-                       buf[j] = 0;
-       return pbe;
 }
 
 /**
- *     save_image_metadata - save the .orig_address fields of the PBEs
- *     from the list @pblist using the swap map handle @handle
- */
-
-static int save_image_metadata(struct pbe *pblist,
-                               struct swap_map_handle *handle)
-{
-       unsigned long *buf;
-       unsigned int n = 0;
-       struct pbe *p;
-       int error = 0;
-
-       printk("Saving image metadata ... ");
-       buf = (unsigned long *)get_zeroed_page(GFP_ATOMIC);
-       if (!buf)
-               return -ENOMEM;
-       p = pblist;
-       while (p) {
-               p = pack_orig_addresses(buf, p);
-               error = swap_map_write_page(handle, (unsigned long)buf);
-               if (error)
-                       break;
-               n++;
-       }
-       free_page((unsigned long)buf);
-       if (!error)
-               printk("done (%u pages saved)\n", n);
-       return error;
-}
-
-/**
- *     enough_swap - Make sure we have enough swap to save the image.
+ *     swsusp_show_speed - print the time elapsed between two events represented by
+ *     @start and @stop
  *
- *     Returns TRUE or FALSE after checking the total amount of swap
- *     space avaiable from the resume partition.
+ *     @nr_pages -     number of pages processed between @start and @stop
+ *     @msg -          introductory message to print
  */
 
-static int enough_swap(unsigned int nr_pages)
+void swsusp_show_speed(struct timeval *start, struct timeval *stop,
+                       unsigned nr_pages, char *msg)
 {
-       unsigned int free_swap = swap_info[root_swap].pages -
-               swap_info[root_swap].inuse_pages;
+       s64 elapsed_centisecs64;
+       int centisecs;
+       int k;
+       int kps;
 
-       pr_debug("swsusp: free swap pages: %u\n", free_swap);
-       return free_swap > (nr_pages + PAGES_FOR_IO +
-               (nr_pages + PBES_PER_PAGE - 1) / PBES_PER_PAGE);
-}
-
-/**
- *     swsusp_write - Write entire image and metadata.
- *
- *     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(struct pbe *pblist, unsigned int nr_pages)
-{
-       struct swap_map_page *swap_map;
-       struct swap_map_handle handle;
-       swp_entry_t start;
-       int error;
-
-       if ((error = swsusp_swap_check())) {
-               printk(KERN_ERR "swsusp: Cannot find swap device, try swapon -a.\n");
-               return error;
-       }
-       if (!enough_swap(nr_pages)) {
-               printk(KERN_ERR "swsusp: Not enough free swap\n");
-               return -ENOSPC;
-       }
-
-       init_header(nr_pages);
-       swap_map = alloc_swap_map(swsusp_info.pages);
-       if (!swap_map)
-               return -ENOMEM;
-       init_swap_map_handle(&handle, swap_map);
-
-       error = swap_map_write_page(&handle, (unsigned long)&swsusp_info);
-       if (!error)
-               error = save_image_metadata(pblist, &handle);
-       if (!error)
-               error = save_image_data(pblist, &handle, nr_pages);
-       if (error)
-               goto Free_image_entries;
-
-       swap_map = reverse_swap_map(swap_map);
-       error = save_swap_map(swap_map, &start);
-       if (error)
-               goto Free_map_entries;
-
-       dump_info();
-       printk( "S" );
-       error = mark_swapfiles(start);
-       printk( "|\n" );
-       if (error)
-               goto Free_map_entries;
-
-Free_swap_map:
-       free_swap_map(swap_map);
-       return error;
-
-Free_map_entries:
-       free_swap_map_entries(swap_map);
-Free_image_entries:
-       free_image_entries(swap_map);
-       goto Free_swap_map;
+       elapsed_centisecs64 = timeval_to_ns(stop) - timeval_to_ns(start);
+       do_div(elapsed_centisecs64, NSEC_PER_SEC / 100);
+       centisecs = elapsed_centisecs64;
+       if (centisecs == 0)
+               centisecs = 1;  /* avoid div-by-zero */
+       k = nr_pages * (PAGE_SIZE / 1024);
+       kps = (k * 100) / centisecs;
+       printk("%s %d kbytes in %d.%02d seconds (%d.%02d MB/s)\n", msg, k,
+                       centisecs / 100, centisecs % 100,
+                       kps / 1000, (kps % 1000) / 10);
 }
 
 /**
@@ -564,37 +202,60 @@ Free_image_entries:
  */
 
 #define SHRINK_BITE    10000
+static inline unsigned long __shrink_memory(long tmp)
+{
+       if (tmp > SHRINK_BITE)
+               tmp = SHRINK_BITE;
+       return shrink_all_memory(tmp);
+}
 
 int swsusp_shrink_memory(void)
 {
-       long size, tmp;
+       long tmp;
        struct zone *zone;
        unsigned long pages = 0;
        unsigned int i = 0;
        char *p = "-\\|/";
+       struct timeval start, stop;
 
        printk("Shrinking memory...  ");
+       do_gettimeofday(&start);
        do {
-               size = 2 * count_highmem_pages();
-               size += size / 50 + count_data_pages();
-               size += (size + PBES_PER_PAGE - 1) / PBES_PER_PAGE +
-                       PAGES_FOR_IO;
+               long size, highmem_size;
+
+               highmem_size = count_highmem_pages();
+               size = count_data_pages() + PAGES_FOR_IO;
                tmp = size;
+               size += highmem_size;
                for_each_zone (zone)
-                       if (!is_highmem(zone))
-                               tmp -= zone->free_pages;
+                       if (populated_zone(zone)) {
+                               if (is_highmem(zone)) {
+                                       highmem_size -= zone->free_pages;
+                               } else {
+                                       tmp -= zone->free_pages;
+                                       tmp += zone->lowmem_reserve[ZONE_NORMAL];
+                                       tmp += snapshot_additional_pages(zone);
+                               }
+                       }
+
+               if (highmem_size < 0)
+                       highmem_size = 0;
+
+               tmp += highmem_size;
                if (tmp > 0) {
-                       tmp = shrink_all_memory(SHRINK_BITE);
+                       tmp = __shrink_memory(tmp);
                        if (!tmp)
                                return -ENOMEM;
                        pages += tmp;
                } else if (size > image_size / PAGE_SIZE) {
-                       tmp = shrink_all_memory(SHRINK_BITE);
+                       tmp = __shrink_memory(size - (image_size / PAGE_SIZE));
                        pages += tmp;
                }
                printk("\b%c", p[i++%4]);
        } while (tmp > 0);
+       do_gettimeofday(&stop);
        printk("\bdone (%lu pages freed)\n", pages);
+       swsusp_show_speed(&start, &stop, pages, "Freed");
 
        return 0;
 }
@@ -605,6 +266,7 @@ int swsusp_suspend(void)
 
        if ((error = arch_prepare_suspend()))
                return error;
+
        local_irq_disable();
        /* At this point, device_suspend() has been called, but *not*
         * device_power_down(). We *must* device_power_down() now.
@@ -617,20 +279,16 @@ int swsusp_suspend(void)
                goto Enable_irqs;
        }
 
-       if ((error = save_highmem())) {
-               printk(KERN_ERR "swsusp: Not enough free pages for highmem\n");
-               goto Restore_highmem;
-       }
-
        save_processor_state();
        if ((error = swsusp_arch_suspend()))
                printk(KERN_ERR "Error %d suspending\n", error);
        /* Restore control flow magically appears here */
        restore_processor_state();
-Restore_highmem:
-       restore_highmem();
+       /* NOTE:  device_power_up() is just a resume() for devices
+        * that suspended with irqs off ... no overall powerup.
+        */
        device_power_up();
-Enable_irqs:
+ Enable_irqs:
        local_irq_enable();
        return error;
 }
@@ -638,401 +296,34 @@ Enable_irqs:
 int swsusp_resume(void)
 {
        int error;
+
        local_irq_disable();
-       if (device_power_down(PMSG_FREEZE))
+       /* NOTE:  device_power_down() is just a suspend() with irqs off;
+        * it has no special "power things down" semantics
+        */
+       if (device_power_down(PMSG_PRETHAW))
                printk(KERN_ERR "Some devices failed to power down, very bad\n");
        /* We'll ignore saved state, but this gets preempt count (etc) right */
        save_processor_state();
-       error = swsusp_arch_resume();
-       /* Code below is only ever reached in case of failure. Otherwise
-        * execution continues at place where swsusp_arch_suspend was called
-         */
-       BUG_ON(!error);
+       error = restore_highmem();
+       if (!error) {
+               error = swsusp_arch_resume();
+               /* The code below is only ever reached in case of a failure.
+                * Otherwise execution continues at place where
+                * swsusp_arch_suspend() was called
+                */
+               BUG_ON(!error);
+               /* This call to restore_highmem() undos the previous one */
+               restore_highmem();
+       }
        /* The only reason why swsusp_arch_resume() can fail is memory being
         * very tight, so we have to free it as soon as we can to avoid
         * subsequent failures
         */
        swsusp_free();
        restore_processor_state();
-       restore_highmem();
        touch_softlockup_watchdog();
        device_power_up();
        local_irq_enable();
        return error;
 }
-
-/**
- *     mark_unsafe_pages - mark the pages that cannot be used for storing
- *     the image during resume, because they conflict with the pages that
- *     had been used before suspend
- */
-
-static void mark_unsafe_pages(struct pbe *pblist)
-{
-       struct zone *zone;
-       unsigned long zone_pfn;
-       struct pbe *p;
-
-       if (!pblist) /* a sanity check */
-               return;
-
-       /* Clear page flags */
-       for_each_zone (zone) {
-               for (zone_pfn = 0; zone_pfn < zone->spanned_pages; ++zone_pfn)
-                       if (pfn_valid(zone_pfn + zone->zone_start_pfn))
-                               ClearPageNosaveFree(pfn_to_page(zone_pfn +
-                                       zone->zone_start_pfn));
-       }
-
-       /* Mark orig addresses */
-       for_each_pbe (p, pblist)
-               SetPageNosaveFree(virt_to_page(p->orig_address));
-
-}
-
-static void copy_page_backup_list(struct pbe *dst, struct pbe *src)
-{
-       /* We assume both lists contain the same number of elements */
-       while (src) {
-               dst->orig_address = src->orig_address;
-               dst = dst->next;
-               src = src->next;
-       }
-}
-
-/*
- *     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 int end_io(struct bio *bio, unsigned int num, int err)
-{
-       if (!test_bit(BIO_UPTODATE, &bio->bi_flags))
-               panic("I/O error reading memory image");
-       atomic_set(&io_done, 0);
-       return 0;
-}
-
-static struct block_device *resume_bdev;
-
-/**
- *     submit - submit BIO request.
- *     @rw:    READ or WRITE.
- *     @off    physical offset of page.
- *     @page:  page we're reading or writing.
- *
- *     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->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;
-       }
-
-
-       atomic_set(&io_done, 1);
-       submit_bio(rw | (1 << BIO_RW_SYNC), bio);
-       while (atomic_read(&io_done))
-               yield();
-       if (rw == READ)
-               bio_set_pages_dirty(bio);
- Done:
-       bio_put(bio);
-       return error;
-}
-
-static int bio_read_page(pgoff_t page_off, void *page)
-{
-       return submit(READ, page_off, page);
-}
-
-static int bio_write_page(pgoff_t page_off, void *page)
-{
-       return submit(WRITE, page_off, page);
-}
-
-/**
- *     The following functions allow us to read data using a swap map
- *     in a file-alike way
- */
-
-static inline void release_swap_map_reader(struct swap_map_handle *handle)
-{
-       if (handle->cur)
-               free_page((unsigned long)handle->cur);
-       handle->cur = NULL;
-}
-
-static inline int get_swap_map_reader(struct swap_map_handle *handle,
-                                      swp_entry_t start)
-{
-       int error;
-
-       if (!swp_offset(start))
-               return -EINVAL;
-       handle->cur = (struct swap_map_page *)get_zeroed_page(GFP_ATOMIC);
-       if (!handle->cur)
-               return -ENOMEM;
-       error = bio_read_page(swp_offset(start), handle->cur);
-       if (error) {
-               release_swap_map_reader(handle);
-               return error;
-       }
-       handle->k = 0;
-       return 0;
-}
-
-static inline int swap_map_read_page(struct swap_map_handle *handle, void *buf)
-{
-       unsigned long offset;
-       int error;
-
-       if (!handle->cur)
-               return -EINVAL;
-       offset = swp_offset(handle->cur->entries[handle->k]);
-       if (!offset)
-               return -EINVAL;
-       error = bio_read_page(offset, buf);
-       if (error)
-               return error;
-       if (++handle->k >= MAP_PAGE_SIZE) {
-               handle->k = 0;
-               offset = swp_offset(handle->cur->next_swap);
-               if (!offset)
-                       release_swap_map_reader(handle);
-               else
-                       error = bio_read_page(offset, handle->cur);
-       }
-       return error;
-}
-
-static int check_header(void)
-{
-       char *reason = NULL;
-
-       dump_info();
-       if (swsusp_info.version_code != LINUX_VERSION_CODE)
-               reason = "kernel version";
-       if (swsusp_info.num_physpages != num_physpages)
-               reason = "memory size";
-       if (strcmp(swsusp_info.uts.sysname,system_utsname.sysname))
-               reason = "system type";
-       if (strcmp(swsusp_info.uts.release,system_utsname.release))
-               reason = "kernel release";
-       if (strcmp(swsusp_info.uts.version,system_utsname.version))
-               reason = "version";
-       if (strcmp(swsusp_info.uts.machine,system_utsname.machine))
-               reason = "machine";
-       if (reason) {
-               printk(KERN_ERR "swsusp: Resume mismatch: %s\n", reason);
-               return -EPERM;
-       }
-       return 0;
-}
-
-/**
- *     load_image_data - load the image data using the swap map handle
- *     @handle and store them using the page backup list @pblist
- *     (assume there are @nr_pages pages to load)
- */
-
-static int load_image_data(struct pbe *pblist,
-                           struct swap_map_handle *handle,
-                           unsigned int nr_pages)
-{
-       int error;
-       unsigned int m;
-       struct pbe *p;
-
-       if (!pblist)
-               return -EINVAL;
-       printk("Loading image data pages (%u pages) ...     ", nr_pages);
-       m = nr_pages / 100;
-       if (!m)
-               m = 1;
-       nr_pages = 0;
-       p = pblist;
-       while (p) {
-               error = swap_map_read_page(handle, (void *)p->address);
-               if (error)
-                       break;
-               p = p->next;
-               if (!(nr_pages % m))
-                       printk("\b\b\b\b%3d%%", nr_pages / m);
-               nr_pages++;
-       }
-       if (!error)
-               printk("\b\b\b\bdone\n");
-       return error;
-}
-
-/**
- *     unpack_orig_addresses - copy the elements of @buf[] (1 page) to
- *     the PBEs in the list starting at @pbe
- */
-
-static inline struct pbe *unpack_orig_addresses(unsigned long *buf,
-                                                struct pbe *pbe)
-{
-       int j;
-
-       for (j = 0; j < PAGE_SIZE / sizeof(long) && pbe; j++) {
-               pbe->orig_address = buf[j];
-               pbe = pbe->next;
-       }
-       return pbe;
-}
-
-/**
- *     load_image_metadata - load the image metadata using the swap map
- *     handle @handle and put them into the PBEs in the list @pblist
- */
-
-static int load_image_metadata(struct pbe *pblist, struct swap_map_handle *handle)
-{
-       struct pbe *p;
-       unsigned long *buf;
-       unsigned int n = 0;
-       int error = 0;
-
-       printk("Loading image metadata ... ");
-       buf = (unsigned long *)get_zeroed_page(GFP_ATOMIC);
-       if (!buf)
-               return -ENOMEM;
-       p = pblist;
-       while (p) {
-               error = swap_map_read_page(handle, buf);
-               if (error)
-                       break;
-               p = unpack_orig_addresses(buf, p);
-               n++;
-       }
-       free_page((unsigned long)buf);
-       if (!error)
-               printk("done (%u pages loaded)\n", n);
-       return error;
-}
-
-int swsusp_read(struct pbe **pblist_ptr)
-{
-       int error;
-       struct pbe *p, *pblist;
-       struct swap_map_handle handle;
-       unsigned int nr_pages;
-
-       if (IS_ERR(resume_bdev)) {
-               pr_debug("swsusp: block device not initialised\n");
-               return PTR_ERR(resume_bdev);
-       }
-
-       error = get_swap_map_reader(&handle, swsusp_header.image);
-       if (!error)
-               error = swap_map_read_page(&handle, &swsusp_info);
-       if (!error)
-               error = check_header();
-       if (error)
-               return error;
-       nr_pages = swsusp_info.image_pages;
-       p = alloc_pagedir(nr_pages, GFP_ATOMIC, 0);
-       if (!p)
-               return -ENOMEM;
-       error = load_image_metadata(p, &handle);
-       if (!error) {
-               mark_unsafe_pages(p);
-               pblist = alloc_pagedir(nr_pages, GFP_ATOMIC, 1);
-               if (pblist)
-                       copy_page_backup_list(pblist, p);
-               free_pagedir(p);
-               if (!pblist)
-                       error = -ENOMEM;
-
-               /* Allocate memory for the image and read the data from swap */
-               if (!error)
-                       error = alloc_data_pages(pblist, GFP_ATOMIC, 1);
-               if (!error) {
-                       release_eaten_pages();
-                       error = load_image_data(pblist, &handle, nr_pages);
-               }
-               if (!error)
-                       *pblist_ptr = pblist;
-       }
-       release_swap_map_reader(&handle);
-
-       blkdev_put(resume_bdev);
-
-       if (!error)
-               pr_debug("swsusp: Reading resume file was successful\n");
-       else
-               pr_debug("swsusp: Error %d resuming\n", error);
-       return error;
-}
-
-/**
- *      swsusp_check - Check for swsusp signature in the resume device
- */
-
-int swsusp_check(void)
-{
-       int error;
-
-       resume_bdev = open_by_devnum(swsusp_resume_device, FMODE_READ);
-       if (!IS_ERR(resume_bdev)) {
-               set_blocksize(resume_bdev, PAGE_SIZE);
-               memset(&swsusp_header, 0, sizeof(swsusp_header));
-               if ((error = bio_read_page(0, &swsusp_header)))
-                       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 {
-                       return -EINVAL;
-               }
-               if (error)
-                       blkdev_put(resume_bdev);
-               else
-                       pr_debug("swsusp: Signature found, resuming\n");
-       } else {
-               error = PTR_ERR(resume_bdev);
-       }
-
-       if (error)
-               pr_debug("swsusp: Error %d check for resume file\n", error);
-
-       return error;
-}
-
-/**
- *     swsusp_close - close swap device.
- */
-
-void swsusp_close(void)
-{
-       if (IS_ERR(resume_bdev)) {
-               pr_debug("swsusp: block device not initialised\n");
-               return;
-       }
-
-       blkdev_put(resume_bdev);
-}