/*
* linux/kernel/power/snapshot.c
*
- * This file provide system snapshot/restore functionality.
+ * This file provides system snapshot/restore functionality for swsusp.
*
* Copyright (C) 1998-2005 Pavel Machek <pavel@suse.cz>
+ * Copyright (C) 2006 Rafael J. Wysocki <rjw@sisk.pl>
*
- * This file is released under the GPLv2, and is based on swsusp.c.
+ * This file is released under the GPLv2.
*
*/
-
#include <linux/version.h>
#include <linux/module.h>
#include <linux/mm.h>
#include <linux/suspend.h>
-#include <linux/smp_lock.h>
#include <linux/delay.h>
#include <linux/bitops.h>
#include <linux/spinlock.h>
#include <linux/kernel.h>
#include <linux/pm.h>
#include <linux/device.h>
+#include <linux/init.h>
#include <linux/bootmem.h>
#include <linux/syscalls.h>
#include <linux/console.h>
#include <linux/highmem.h>
+#include <linux/list.h>
#include <asm/uaccess.h>
#include <asm/mmu_context.h>
#include "power.h"
-/* List of PBEs used for creating and restoring the suspend image */
-struct pbe *restore_pblist;
-
-static unsigned int nr_copy_pages;
-static unsigned int nr_meta_pages;
-static unsigned long *buffer;
-
-#ifdef CONFIG_HIGHMEM
-unsigned int count_highmem_pages(void)
-{
- struct zone *zone;
- unsigned long zone_pfn;
- unsigned int n = 0;
-
- for_each_zone (zone)
- if (is_highmem(zone)) {
- mark_free_pages(zone);
- for (zone_pfn = 0; zone_pfn < zone->spanned_pages; zone_pfn++) {
- struct page *page;
- unsigned long pfn = zone_pfn + zone->zone_start_pfn;
- if (!pfn_valid(pfn))
- continue;
- page = pfn_to_page(pfn);
- if (PageReserved(page))
- continue;
- if (PageNosaveFree(page))
- continue;
- n++;
- }
- }
- return n;
-}
-
-struct highmem_page {
- char *data;
- struct page *page;
- struct highmem_page *next;
-};
-
-static struct highmem_page *highmem_copy;
-
-static int save_highmem_zone(struct zone *zone)
-{
- unsigned long zone_pfn;
- mark_free_pages(zone);
- for (zone_pfn = 0; zone_pfn < zone->spanned_pages; ++zone_pfn) {
- struct page *page;
- struct highmem_page *save;
- void *kaddr;
- unsigned long pfn = zone_pfn + zone->zone_start_pfn;
-
- if (!(pfn%10000))
- printk(".");
- if (!pfn_valid(pfn))
- continue;
- page = pfn_to_page(pfn);
- /*
- * This condition results from rvmalloc() sans vmalloc_32()
- * and architectural memory reservations. This should be
- * corrected eventually when the cases giving rise to this
- * are better understood.
- */
- if (PageReserved(page))
- continue;
- BUG_ON(PageNosave(page));
- if (PageNosaveFree(page))
- continue;
- save = kmalloc(sizeof(struct highmem_page), GFP_ATOMIC);
- if (!save)
- return -ENOMEM;
- save->next = highmem_copy;
- save->page = page;
- save->data = (void *) get_zeroed_page(GFP_ATOMIC);
- if (!save->data) {
- kfree(save);
- return -ENOMEM;
- }
- kaddr = kmap_atomic(page, KM_USER0);
- memcpy(save->data, kaddr, PAGE_SIZE);
- kunmap_atomic(kaddr, KM_USER0);
- highmem_copy = save;
- }
- return 0;
-}
+static int swsusp_page_is_free(struct page *);
+static void swsusp_set_page_forbidden(struct page *);
+static void swsusp_unset_page_forbidden(struct page *);
-int save_highmem(void)
-{
- struct zone *zone;
- int res = 0;
+/*
+ * Preferred image size in bytes (tunable via /sys/power/image_size).
+ * When it is set to N, swsusp will do its best to ensure the image
+ * size will not exceed N bytes, but if that is impossible, it will
+ * try to create the smallest image possible.
+ */
+unsigned long image_size = 500 * 1024 * 1024;
- pr_debug("swsusp: Saving Highmem");
- drain_local_pages();
- for_each_zone (zone) {
- if (is_highmem(zone))
- res = save_highmem_zone(zone);
- if (res)
- return res;
- }
- printk("\n");
- return 0;
-}
+/* List of PBEs needed for restoring the pages that were allocated before
+ * the suspend and included in the suspend image, but have also been
+ * allocated by the "resume" kernel, so their contents cannot be written
+ * directly to their "original" page frames.
+ */
+struct pbe *restore_pblist;
-int restore_highmem(void)
-{
- printk("swsusp: Restoring Highmem\n");
- while (highmem_copy) {
- struct highmem_page *save = highmem_copy;
- void *kaddr;
- highmem_copy = save->next;
-
- kaddr = kmap_atomic(save->page, KM_USER0);
- memcpy(kaddr, save->data, PAGE_SIZE);
- kunmap_atomic(kaddr, KM_USER0);
- free_page((long) save->data);
- kfree(save);
- }
- return 0;
-}
-#else
-static inline unsigned int count_highmem_pages(void) {return 0;}
-static inline int save_highmem(void) {return 0;}
-static inline int restore_highmem(void) {return 0;}
-#endif
+/* Pointer to an auxiliary buffer (1 page) */
+static void *buffer;
/**
* @safe_needed - on resume, for storing the PBE list and the image,
* we can only use memory pages that do not conflict with the pages
- * used before suspend.
+ * used before suspend. The unsafe pages have PageNosaveFree set
+ * and we count them using unsafe_pages.
*
- * The unsafe pages are marked with the PG_nosave_free flag
- * and we count them using unsafe_pages
+ * Each allocated image page is marked as PageNosave and PageNosaveFree
+ * so that swsusp_free() can release it.
*/
#define PG_ANY 0
#define PG_UNSAFE_CLEAR 1
#define PG_UNSAFE_KEEP 0
-static unsigned int unsafe_pages;
+static unsigned int allocated_unsafe_pages;
-static void *alloc_image_page(gfp_t gfp_mask, int safe_needed)
+static void *get_image_page(gfp_t gfp_mask, int safe_needed)
{
void *res;
res = (void *)get_zeroed_page(gfp_mask);
if (safe_needed)
- while (res && PageNosaveFree(virt_to_page(res))) {
+ while (res && swsusp_page_is_free(virt_to_page(res))) {
/* The page is unsafe, mark it for swsusp_free() */
- SetPageNosave(virt_to_page(res));
- unsafe_pages++;
+ swsusp_set_page_forbidden(virt_to_page(res));
+ allocated_unsafe_pages++;
res = (void *)get_zeroed_page(gfp_mask);
}
if (res) {
- SetPageNosave(virt_to_page(res));
- SetPageNosaveFree(virt_to_page(res));
+ swsusp_set_page_forbidden(virt_to_page(res));
+ swsusp_set_page_free(virt_to_page(res));
}
return res;
}
unsigned long get_safe_page(gfp_t gfp_mask)
{
- return (unsigned long)alloc_image_page(gfp_mask, PG_SAFE);
+ return (unsigned long)get_image_page(gfp_mask, PG_SAFE);
+}
+
+static struct page *alloc_image_page(gfp_t gfp_mask)
+{
+ struct page *page;
+
+ page = alloc_page(gfp_mask);
+ if (page) {
+ swsusp_set_page_forbidden(page);
+ swsusp_set_page_free(page);
+ }
+ return page;
}
/**
* free_image_page - free page represented by @addr, allocated with
- * alloc_image_page (page flags set by it must be cleared)
+ * get_image_page (page flags set by it must be cleared)
*/
static inline void free_image_page(void *addr, int clear_nosave_free)
{
- ClearPageNosave(virt_to_page(addr));
+ struct page *page;
+
+ BUG_ON(!virt_addr_valid(addr));
+
+ page = virt_to_page(addr);
+
+ swsusp_unset_page_forbidden(page);
if (clear_nosave_free)
- ClearPageNosaveFree(virt_to_page(addr));
- free_page((unsigned long)addr);
+ swsusp_unset_page_free(page);
+
+ __free_page(page);
}
/* struct linked_page is used to build chains of pages */
if (LINKED_PAGE_DATA_SIZE - ca->used_space < size) {
struct linked_page *lp;
- lp = alloc_image_page(ca->gfp_mask, ca->safe_needed);
+ lp = get_image_page(ca->gfp_mask, ca->safe_needed);
if (!lp)
return NULL;
return ret;
}
-static void chain_free(struct chain_allocator *ca, int clear_page_nosave)
-{
- free_list_of_pages(ca->chain, clear_page_nosave);
- memset(ca, 0, sizeof(struct chain_allocator));
-}
-
/**
* Data types related to memory bitmaps.
*
* objects. The main list's elements are of type struct zone_bitmap
* and each of them corresonds to one zone. For each zone bitmap
* object there is a list of objects of type struct bm_block that
- * represent each blocks of bit chunks in which information is
- * stored.
+ * represent each blocks of bitmap in which information is stored.
*
* struct memory_bitmap contains a pointer to the main list of zone
* bitmap objects, a struct bm_position used for browsing the bitmap,
* pfns that correspond to the start and end of the represented zone.
*
* struct bm_block contains a pointer to the memory page in which
- * information is stored (in the form of a block of bit chunks
- * of type unsigned long each). It also contains the pfns that
- * correspond to the start and end of the represented memory area and
- * the number of bit chunks in the block.
- *
- * NOTE: Memory bitmaps are used for two types of operations only:
- * "set a bit" and "find the next bit set". Moreover, the searching
- * is always carried out after all of the "set a bit" operations
- * on given bitmap.
+ * information is stored (in the form of a block of bitmap)
+ * It also contains the pfns that correspond to the start and end of
+ * the represented memory area.
*/
#define BM_END_OF_MAP (~0UL)
-#define BM_CHUNKS_PER_BLOCK (PAGE_SIZE / sizeof(long))
-#define BM_BITS_PER_CHUNK (sizeof(long) << 3)
-#define BM_BITS_PER_BLOCK (PAGE_SIZE << 3)
+#define BM_BITS_PER_BLOCK (PAGE_SIZE * BITS_PER_BYTE)
struct bm_block {
- struct bm_block *next; /* next element of the list */
+ struct list_head hook; /* hook into a list of bitmap blocks */
unsigned long start_pfn; /* pfn represented by the first bit */
unsigned long end_pfn; /* pfn represented by the last bit plus 1 */
- unsigned int size; /* number of bit chunks */
- unsigned long *data; /* chunks of bits representing pages */
+ unsigned long *data; /* bitmap representing pages */
};
-struct zone_bitmap {
- struct zone_bitmap *next; /* next element of the list */
- unsigned long start_pfn; /* minimal pfn in this zone */
- unsigned long end_pfn; /* maximal pfn in this zone plus 1 */
- struct bm_block *bm_blocks; /* list of bitmap blocks */
- struct bm_block *cur_block; /* recently used bitmap block */
-};
+static inline unsigned long bm_block_bits(struct bm_block *bb)
+{
+ return bb->end_pfn - bb->start_pfn;
+}
/* strcut bm_position is used for browsing memory bitmaps */
struct bm_position {
- struct zone_bitmap *zone_bm;
struct bm_block *block;
- int chunk;
int bit;
};
struct memory_bitmap {
- struct zone_bitmap *zone_bm_list; /* list of zone bitmaps */
+ struct list_head blocks; /* list of bitmap blocks */
struct linked_page *p_list; /* list of pages used to store zone
* bitmap objects and bitmap block
* objects
/* Functions that operate on memory bitmaps */
-static inline void memory_bm_reset_chunk(struct memory_bitmap *bm)
-{
- bm->cur.chunk = 0;
- bm->cur.bit = -1;
-}
-
static void memory_bm_position_reset(struct memory_bitmap *bm)
{
- struct zone_bitmap *zone_bm;
-
- zone_bm = bm->zone_bm_list;
- bm->cur.zone_bm = zone_bm;
- bm->cur.block = zone_bm->bm_blocks;
- memory_bm_reset_chunk(bm);
+ bm->cur.block = list_entry(bm->blocks.next, struct bm_block, hook);
+ bm->cur.bit = 0;
}
static void memory_bm_free(struct memory_bitmap *bm, int clear_nosave_free);
/**
* create_bm_block_list - create a list of block bitmap objects
+ * @pages - number of pages to track
+ * @list - list to put the allocated blocks into
+ * @ca - chain allocator to be used for allocating memory
*/
-
-static inline struct bm_block *
-create_bm_block_list(unsigned int nr_blocks, struct chain_allocator *ca)
+static int create_bm_block_list(unsigned long pages,
+ struct list_head *list,
+ struct chain_allocator *ca)
{
- struct bm_block *bblist = NULL;
+ unsigned int nr_blocks = DIV_ROUND_UP(pages, BM_BITS_PER_BLOCK);
while (nr_blocks-- > 0) {
struct bm_block *bb;
bb = chain_alloc(ca, sizeof(struct bm_block));
if (!bb)
- return NULL;
-
- bb->next = bblist;
- bblist = bb;
+ return -ENOMEM;
+ list_add(&bb->hook, list);
}
- return bblist;
+
+ return 0;
}
+struct mem_extent {
+ struct list_head hook;
+ unsigned long start;
+ unsigned long end;
+};
+
/**
- * create_zone_bm_list - create a list of zone bitmap objects
+ * free_mem_extents - free a list of memory extents
+ * @list - list of extents to empty
*/
+static void free_mem_extents(struct list_head *list)
+{
+ struct mem_extent *ext, *aux;
+
+ list_for_each_entry_safe(ext, aux, list, hook) {
+ list_del(&ext->hook);
+ kfree(ext);
+ }
+}
-static inline struct zone_bitmap *
-create_zone_bm_list(unsigned int nr_zones, struct chain_allocator *ca)
+/**
+ * create_mem_extents - create a list of memory extents representing
+ * contiguous ranges of PFNs
+ * @list - list to put the extents into
+ * @gfp_mask - mask to use for memory allocations
+ */
+static int create_mem_extents(struct list_head *list, gfp_t gfp_mask)
{
- struct zone_bitmap *zbmlist = NULL;
+ struct zone *zone;
- while (nr_zones-- > 0) {
- struct zone_bitmap *zbm;
+ INIT_LIST_HEAD(list);
- zbm = chain_alloc(ca, sizeof(struct zone_bitmap));
- if (!zbm)
- return NULL;
+ for_each_populated_zone(zone) {
+ unsigned long zone_start, zone_end;
+ struct mem_extent *ext, *cur, *aux;
+
+ zone_start = zone->zone_start_pfn;
+ zone_end = zone->zone_start_pfn + zone->spanned_pages;
+
+ list_for_each_entry(ext, list, hook)
+ if (zone_start <= ext->end)
+ break;
+
+ if (&ext->hook == list || zone_end < ext->start) {
+ /* New extent is necessary */
+ struct mem_extent *new_ext;
+
+ new_ext = kzalloc(sizeof(struct mem_extent), gfp_mask);
+ if (!new_ext) {
+ free_mem_extents(list);
+ return -ENOMEM;
+ }
+ new_ext->start = zone_start;
+ new_ext->end = zone_end;
+ list_add_tail(&new_ext->hook, &ext->hook);
+ continue;
+ }
+
+ /* Merge this zone's range of PFNs with the existing one */
+ if (zone_start < ext->start)
+ ext->start = zone_start;
+ if (zone_end > ext->end)
+ ext->end = zone_end;
- zbm->next = zbmlist;
- zbmlist = zbm;
+ /* More merging may be possible */
+ cur = ext;
+ list_for_each_entry_safe_continue(cur, aux, list, hook) {
+ if (zone_end < cur->start)
+ break;
+ if (zone_end < cur->end)
+ ext->end = cur->end;
+ list_del(&cur->hook);
+ kfree(cur);
+ }
}
- return zbmlist;
+
+ return 0;
}
/**
* memory_bm_create - allocate memory for a memory bitmap
*/
-
static int
memory_bm_create(struct memory_bitmap *bm, gfp_t gfp_mask, int safe_needed)
{
struct chain_allocator ca;
- struct zone *zone;
- struct zone_bitmap *zone_bm;
- struct bm_block *bb;
- unsigned int nr;
+ struct list_head mem_extents;
+ struct mem_extent *ext;
+ int error;
chain_init(&ca, gfp_mask, safe_needed);
+ INIT_LIST_HEAD(&bm->blocks);
- /* Compute the number of zones */
- nr = 0;
- for_each_zone (zone)
- if (populated_zone(zone) && !is_highmem(zone))
- nr++;
-
- /* Allocate the list of zones bitmap objects */
- zone_bm = create_zone_bm_list(nr, &ca);
- bm->zone_bm_list = zone_bm;
- if (!zone_bm) {
- chain_free(&ca, PG_UNSAFE_CLEAR);
- return -ENOMEM;
- }
-
- /* Initialize the zone bitmap objects */
- for_each_zone (zone) {
- unsigned long pfn;
+ error = create_mem_extents(&mem_extents, gfp_mask);
+ if (error)
+ return error;
- if (!populated_zone(zone) || is_highmem(zone))
- continue;
+ list_for_each_entry(ext, &mem_extents, hook) {
+ struct bm_block *bb;
+ unsigned long pfn = ext->start;
+ unsigned long pages = ext->end - ext->start;
- zone_bm->start_pfn = zone->zone_start_pfn;
- zone_bm->end_pfn = zone->zone_start_pfn + zone->spanned_pages;
- /* Allocate the list of bitmap block objects */
- nr = DIV_ROUND_UP(zone->spanned_pages, BM_BITS_PER_BLOCK);
- bb = create_bm_block_list(nr, &ca);
- zone_bm->bm_blocks = bb;
- zone_bm->cur_block = bb;
- if (!bb)
- goto Free;
+ bb = list_entry(bm->blocks.prev, struct bm_block, hook);
- nr = zone->spanned_pages;
- pfn = zone->zone_start_pfn;
- /* Initialize the bitmap block objects */
- while (bb) {
- unsigned long *ptr;
+ error = create_bm_block_list(pages, bm->blocks.prev, &ca);
+ if (error)
+ goto Error;
- ptr = alloc_image_page(gfp_mask, safe_needed);
- bb->data = ptr;
- if (!ptr)
- goto Free;
+ list_for_each_entry_continue(bb, &bm->blocks, hook) {
+ bb->data = get_image_page(gfp_mask, safe_needed);
+ if (!bb->data) {
+ error = -ENOMEM;
+ goto Error;
+ }
bb->start_pfn = pfn;
- if (nr >= BM_BITS_PER_BLOCK) {
+ if (pages >= BM_BITS_PER_BLOCK) {
pfn += BM_BITS_PER_BLOCK;
- bb->size = BM_CHUNKS_PER_BLOCK;
- nr -= BM_BITS_PER_BLOCK;
+ pages -= BM_BITS_PER_BLOCK;
} else {
/* This is executed only once in the loop */
- pfn += nr;
- bb->size = DIV_ROUND_UP(nr, BM_BITS_PER_CHUNK);
+ pfn += pages;
}
bb->end_pfn = pfn;
- bb = bb->next;
}
- zone_bm = zone_bm->next;
}
+
bm->p_list = ca.chain;
memory_bm_position_reset(bm);
- return 0;
+ Exit:
+ free_mem_extents(&mem_extents);
+ return error;
-Free:
+ Error:
bm->p_list = ca.chain;
memory_bm_free(bm, PG_UNSAFE_CLEAR);
- return -ENOMEM;
+ goto Exit;
}
/**
* memory_bm_free - free memory occupied by the memory bitmap @bm
*/
-
static void memory_bm_free(struct memory_bitmap *bm, int clear_nosave_free)
{
- struct zone_bitmap *zone_bm;
+ struct bm_block *bb;
- /* Free the list of bit blocks for each zone_bitmap object */
- zone_bm = bm->zone_bm_list;
- while (zone_bm) {
- struct bm_block *bb;
+ list_for_each_entry(bb, &bm->blocks, hook)
+ if (bb->data)
+ free_image_page(bb->data, clear_nosave_free);
- bb = zone_bm->bm_blocks;
- while (bb) {
- if (bb->data)
- free_image_page(bb->data, clear_nosave_free);
- bb = bb->next;
- }
- zone_bm = zone_bm->next;
- }
free_list_of_pages(bm->p_list, clear_nosave_free);
- bm->zone_bm_list = NULL;
+
+ INIT_LIST_HEAD(&bm->blocks);
}
/**
- * memory_bm_set_bit - set the bit in the bitmap @bm that corresponds
+ * memory_bm_find_bit - find the bit in the bitmap @bm that corresponds
* to given pfn. The cur_zone_bm member of @bm and the cur_block member
* of @bm->cur_zone_bm are updated.
- *
- * If the bit cannot be set, the function returns -EINVAL .
*/
-
-static int
-memory_bm_set_bit(struct memory_bitmap *bm, unsigned long pfn)
+static int memory_bm_find_bit(struct memory_bitmap *bm, unsigned long pfn,
+ void **addr, unsigned int *bit_nr)
{
- struct zone_bitmap *zone_bm;
struct bm_block *bb;
- /* Check if the pfn is from the current zone */
- zone_bm = bm->cur.zone_bm;
- if (pfn < zone_bm->start_pfn || pfn >= zone_bm->end_pfn) {
- zone_bm = bm->zone_bm_list;
- /* We don't assume that the zones are sorted by pfns */
- while (pfn < zone_bm->start_pfn || pfn >= zone_bm->end_pfn) {
- zone_bm = zone_bm->next;
- if (unlikely(!zone_bm))
- return -EINVAL;
- }
- bm->cur.zone_bm = zone_bm;
- }
- /* Check if the pfn corresponds to the current bitmap block */
- bb = zone_bm->cur_block;
+ /*
+ * Check if the pfn corresponds to the current bitmap block and find
+ * the block where it fits if this is not the case.
+ */
+ bb = bm->cur.block;
if (pfn < bb->start_pfn)
- bb = zone_bm->bm_blocks;
+ list_for_each_entry_continue_reverse(bb, &bm->blocks, hook)
+ if (pfn >= bb->start_pfn)
+ break;
- while (pfn >= bb->end_pfn) {
- bb = bb->next;
- if (unlikely(!bb))
- return -EINVAL;
- }
- zone_bm->cur_block = bb;
+ if (pfn >= bb->end_pfn)
+ list_for_each_entry_continue(bb, &bm->blocks, hook)
+ if (pfn >= bb->start_pfn && pfn < bb->end_pfn)
+ break;
+
+ if (&bb->hook == &bm->blocks)
+ return -EFAULT;
+
+ /* The block has been found */
+ bm->cur.block = bb;
pfn -= bb->start_pfn;
- set_bit(pfn % BM_BITS_PER_CHUNK, bb->data + pfn / BM_BITS_PER_CHUNK);
+ bm->cur.bit = pfn + 1;
+ *bit_nr = pfn;
+ *addr = bb->data;
return 0;
}
-/* Two auxiliary functions for memory_bm_next_pfn */
+static void memory_bm_set_bit(struct memory_bitmap *bm, unsigned long pfn)
+{
+ void *addr;
+ unsigned int bit;
+ int error;
-/* Find the first set bit in the given chunk, if there is one */
+ error = memory_bm_find_bit(bm, pfn, &addr, &bit);
+ BUG_ON(error);
+ set_bit(bit, addr);
+}
-static inline int next_bit_in_chunk(int bit, unsigned long *chunk_p)
+static int mem_bm_set_bit_check(struct memory_bitmap *bm, unsigned long pfn)
{
- bit++;
- while (bit < BM_BITS_PER_CHUNK) {
- if (test_bit(bit, chunk_p))
- return bit;
+ void *addr;
+ unsigned int bit;
+ int error;
- bit++;
- }
- return -1;
+ error = memory_bm_find_bit(bm, pfn, &addr, &bit);
+ if (!error)
+ set_bit(bit, addr);
+ return error;
}
-/* Find a chunk containing some bits set in given block of bits */
+static void memory_bm_clear_bit(struct memory_bitmap *bm, unsigned long pfn)
+{
+ void *addr;
+ unsigned int bit;
+ int error;
+
+ error = memory_bm_find_bit(bm, pfn, &addr, &bit);
+ BUG_ON(error);
+ clear_bit(bit, addr);
+}
-static inline int next_chunk_in_block(int n, struct bm_block *bb)
+static int memory_bm_test_bit(struct memory_bitmap *bm, unsigned long pfn)
{
- n++;
- while (n < bb->size) {
- if (bb->data[n])
- return n;
+ void *addr;
+ unsigned int bit;
+ int error;
- n++;
- }
- return -1;
+ error = memory_bm_find_bit(bm, pfn, &addr, &bit);
+ BUG_ON(error);
+ return test_bit(bit, addr);
+}
+
+static bool memory_bm_pfn_present(struct memory_bitmap *bm, unsigned long pfn)
+{
+ void *addr;
+ unsigned int bit;
+
+ return !memory_bm_find_bit(bm, pfn, &addr, &bit);
}
/**
static unsigned long memory_bm_next_pfn(struct memory_bitmap *bm)
{
- struct zone_bitmap *zone_bm;
struct bm_block *bb;
- int chunk;
int bit;
+ bb = bm->cur.block;
do {
- bb = bm->cur.block;
- do {
- chunk = bm->cur.chunk;
- bit = bm->cur.bit;
- do {
- bit = next_bit_in_chunk(bit, bb->data + chunk);
- if (bit >= 0)
- goto Return_pfn;
-
- chunk = next_chunk_in_block(chunk, bb);
- bit = -1;
- } while (chunk >= 0);
- bb = bb->next;
- bm->cur.block = bb;
- memory_bm_reset_chunk(bm);
- } while (bb);
- zone_bm = bm->cur.zone_bm->next;
- if (zone_bm) {
- bm->cur.zone_bm = zone_bm;
- bm->cur.block = zone_bm->bm_blocks;
- memory_bm_reset_chunk(bm);
- }
- } while (zone_bm);
+ bit = bm->cur.bit;
+ bit = find_next_bit(bb->data, bm_block_bits(bb), bit);
+ if (bit < bm_block_bits(bb))
+ goto Return_pfn;
+
+ bb = list_entry(bb->hook.next, struct bm_block, hook);
+ bm->cur.block = bb;
+ bm->cur.bit = 0;
+ } while (&bb->hook != &bm->blocks);
+
memory_bm_position_reset(bm);
return BM_END_OF_MAP;
-Return_pfn:
- bm->cur.chunk = chunk;
- bm->cur.bit = bit;
- return bb->start_pfn + chunk * BM_BITS_PER_CHUNK + bit;
+ Return_pfn:
+ bm->cur.bit = bit + 1;
+ return bb->start_pfn + bit;
}
/**
- * snapshot_additional_pages - estimate the number of additional pages
- * be needed for setting up the suspend image data structures for given
- * zone (usually the returned value is greater than the exact number)
+ * This structure represents a range of page frames the contents of which
+ * should not be saved during the suspend.
*/
-unsigned int snapshot_additional_pages(struct zone *zone)
-{
- unsigned int res;
+struct nosave_region {
+ struct list_head list;
+ unsigned long start_pfn;
+ unsigned long end_pfn;
+};
- res = DIV_ROUND_UP(zone->spanned_pages, BM_BITS_PER_BLOCK);
- res += DIV_ROUND_UP(res * sizeof(struct bm_block), PAGE_SIZE);
- return res;
-}
+static LIST_HEAD(nosave_regions);
/**
- * pfn_is_nosave - check if given pfn is in the 'nosave' section
+ * register_nosave_region - register a range of page frames the contents
+ * of which should not be saved during the suspend (to be used in the early
+ * initialization code)
*/
-static inline int pfn_is_nosave(unsigned long pfn)
+void __init
+__register_nosave_region(unsigned long start_pfn, unsigned long end_pfn,
+ int use_kmalloc)
{
- unsigned long nosave_begin_pfn = __pa(&__nosave_begin) >> PAGE_SHIFT;
- unsigned long nosave_end_pfn = PAGE_ALIGN(__pa(&__nosave_end)) >> PAGE_SHIFT;
- return (pfn >= nosave_begin_pfn) && (pfn < nosave_end_pfn);
+ struct nosave_region *region;
+
+ if (start_pfn >= end_pfn)
+ return;
+
+ if (!list_empty(&nosave_regions)) {
+ /* Try to extend the previous region (they should be sorted) */
+ region = list_entry(nosave_regions.prev,
+ struct nosave_region, list);
+ if (region->end_pfn == start_pfn) {
+ region->end_pfn = end_pfn;
+ goto Report;
+ }
+ }
+ if (use_kmalloc) {
+ /* during init, this shouldn't fail */
+ region = kmalloc(sizeof(struct nosave_region), GFP_KERNEL);
+ BUG_ON(!region);
+ } else
+ /* This allocation cannot fail */
+ region = alloc_bootmem(sizeof(struct nosave_region));
+ region->start_pfn = start_pfn;
+ region->end_pfn = end_pfn;
+ list_add_tail(®ion->list, &nosave_regions);
+ Report:
+ printk(KERN_INFO "PM: Registered nosave memory: %016lx - %016lx\n",
+ start_pfn << PAGE_SHIFT, end_pfn << PAGE_SHIFT);
}
-/**
- * saveable - Determine whether a page should be cloned or not.
- * @pfn: The page
- *
- * We save a page if it isn't Nosave, and is not in the range of pages
- * statically defined as 'unsaveable', and it
- * isn't a part of a free chunk of pages.
+/*
+ * Set bits in this map correspond to the page frames the contents of which
+ * should not be saved during the suspend.
*/
+static struct memory_bitmap *forbidden_pages_map;
-static struct page *saveable_page(unsigned long pfn)
-{
- struct page *page;
-
- if (!pfn_valid(pfn))
- return NULL;
-
- page = pfn_to_page(pfn);
+/* Set bits in this map correspond to free page frames. */
+static struct memory_bitmap *free_pages_map;
- if (PageNosave(page))
- return NULL;
- if (PageReserved(page) && pfn_is_nosave(pfn))
- return NULL;
- if (PageNosaveFree(page))
- return NULL;
+/*
+ * Each page frame allocated for creating the image is marked by setting the
+ * corresponding bits in forbidden_pages_map and free_pages_map simultaneously
+ */
- return page;
+void swsusp_set_page_free(struct page *page)
+{
+ if (free_pages_map)
+ memory_bm_set_bit(free_pages_map, page_to_pfn(page));
}
-unsigned int count_data_pages(void)
+static int swsusp_page_is_free(struct page *page)
{
- struct zone *zone;
- unsigned long pfn, max_zone_pfn;
- unsigned int n = 0;
-
- for_each_zone (zone) {
- if (is_highmem(zone))
- continue;
- mark_free_pages(zone);
- max_zone_pfn = zone->zone_start_pfn + zone->spanned_pages;
- for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++)
- n += !!saveable_page(pfn);
- }
- return n;
+ return free_pages_map ?
+ memory_bm_test_bit(free_pages_map, page_to_pfn(page)) : 0;
}
-static inline void copy_data_page(long *dst, long *src)
+void swsusp_unset_page_free(struct page *page)
{
- int n;
-
- /* copy_page and memcpy are not usable for copying task structs. */
- for (n = PAGE_SIZE / sizeof(long); n; n--)
- *dst++ = *src++;
+ if (free_pages_map)
+ memory_bm_clear_bit(free_pages_map, page_to_pfn(page));
}
-static void
-copy_data_pages(struct memory_bitmap *copy_bm, struct memory_bitmap *orig_bm)
+static void swsusp_set_page_forbidden(struct page *page)
{
- struct zone *zone;
- unsigned long pfn;
-
- for_each_zone (zone) {
- unsigned long max_zone_pfn;
-
- if (is_highmem(zone))
- continue;
+ if (forbidden_pages_map)
+ memory_bm_set_bit(forbidden_pages_map, page_to_pfn(page));
+}
- mark_free_pages(zone);
- max_zone_pfn = zone->zone_start_pfn + zone->spanned_pages;
- for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++)
- if (saveable_page(pfn))
- memory_bm_set_bit(orig_bm, pfn);
- }
- memory_bm_position_reset(orig_bm);
- memory_bm_position_reset(copy_bm);
- do {
- pfn = memory_bm_next_pfn(orig_bm);
- if (likely(pfn != BM_END_OF_MAP)) {
- struct page *page;
- void *src;
+int swsusp_page_is_forbidden(struct page *page)
+{
+ return forbidden_pages_map ?
+ memory_bm_test_bit(forbidden_pages_map, page_to_pfn(page)) : 0;
+}
- page = pfn_to_page(pfn);
- src = page_address(page);
- page = pfn_to_page(memory_bm_next_pfn(copy_bm));
- copy_data_page(page_address(page), src);
- }
- } while (pfn != BM_END_OF_MAP);
+static void swsusp_unset_page_forbidden(struct page *page)
+{
+ if (forbidden_pages_map)
+ memory_bm_clear_bit(forbidden_pages_map, page_to_pfn(page));
}
/**
- * free_pagedir - free pages allocated with alloc_pagedir()
+ * mark_nosave_pages - set bits corresponding to the page frames the
+ * contents of which should not be saved in a given bitmap.
*/
-static void free_pagedir(struct pbe *pblist, int clear_nosave_free)
+static void mark_nosave_pages(struct memory_bitmap *bm)
{
- struct pbe *pbe;
+ struct nosave_region *region;
+
+ if (list_empty(&nosave_regions))
+ return;
+
+ list_for_each_entry(region, &nosave_regions, list) {
+ unsigned long pfn;
+
+ pr_debug("PM: Marking nosave pages: %016lx - %016lx\n",
+ region->start_pfn << PAGE_SHIFT,
+ region->end_pfn << PAGE_SHIFT);
- while (pblist) {
- pbe = (pblist + PB_PAGE_SKIP)->next;
- free_image_page(pblist, clear_nosave_free);
- pblist = pbe;
+ for (pfn = region->start_pfn; pfn < region->end_pfn; pfn++)
+ if (pfn_valid(pfn)) {
+ /*
+ * It is safe to ignore the result of
+ * mem_bm_set_bit_check() here, since we won't
+ * touch the PFNs for which the error is
+ * returned anyway.
+ */
+ mem_bm_set_bit_check(bm, pfn);
+ }
}
}
/**
- * fill_pb_page - Create a list of PBEs on a given memory page
+ * create_basic_memory_bitmaps - create bitmaps needed for marking page
+ * frames that should not be saved and free page frames. The pointers
+ * forbidden_pages_map and free_pages_map are only modified if everything
+ * goes well, because we don't want the bits to be used before both bitmaps
+ * are set up.
*/
-static inline void fill_pb_page(struct pbe *pbpage, unsigned int n)
+int create_basic_memory_bitmaps(void)
{
- struct pbe *p;
+ struct memory_bitmap *bm1, *bm2;
+ int error = 0;
+
+ BUG_ON(forbidden_pages_map || free_pages_map);
+
+ bm1 = kzalloc(sizeof(struct memory_bitmap), GFP_KERNEL);
+ if (!bm1)
+ return -ENOMEM;
- p = pbpage;
- pbpage += n - 1;
- do
- p->next = p + 1;
- while (++p < pbpage);
+ error = memory_bm_create(bm1, GFP_KERNEL, PG_ANY);
+ if (error)
+ goto Free_first_object;
+
+ bm2 = kzalloc(sizeof(struct memory_bitmap), GFP_KERNEL);
+ if (!bm2)
+ goto Free_first_bitmap;
+
+ error = memory_bm_create(bm2, GFP_KERNEL, PG_ANY);
+ if (error)
+ goto Free_second_object;
+
+ forbidden_pages_map = bm1;
+ free_pages_map = bm2;
+ mark_nosave_pages(forbidden_pages_map);
+
+ pr_debug("PM: Basic memory bitmaps created\n");
+
+ return 0;
+
+ Free_second_object:
+ kfree(bm2);
+ Free_first_bitmap:
+ memory_bm_free(bm1, PG_UNSAFE_CLEAR);
+ Free_first_object:
+ kfree(bm1);
+ return -ENOMEM;
}
/**
- * create_pbe_list - Create a list of PBEs on top of a given chain
- * of memory pages allocated with alloc_pagedir()
- *
- * This function assumes that pages allocated by alloc_image_page() will
- * always be zeroed.
+ * free_basic_memory_bitmaps - free memory bitmaps allocated by
+ * create_basic_memory_bitmaps(). The auxiliary pointers are necessary
+ * so that the bitmaps themselves are not referred to while they are being
+ * freed.
*/
-static inline void create_pbe_list(struct pbe *pblist, unsigned int nr_pages)
+void free_basic_memory_bitmaps(void)
{
- struct pbe *pbpage;
- unsigned int num = PBES_PER_PAGE;
+ struct memory_bitmap *bm1, *bm2;
- for_each_pb_page (pbpage, pblist) {
- if (num >= nr_pages)
- break;
+ BUG_ON(!(forbidden_pages_map && free_pages_map));
- fill_pb_page(pbpage, PBES_PER_PAGE);
- num += PBES_PER_PAGE;
- }
- if (pbpage) {
- num -= PBES_PER_PAGE;
- fill_pb_page(pbpage, nr_pages - num);
- }
+ bm1 = forbidden_pages_map;
+ bm2 = free_pages_map;
+ forbidden_pages_map = NULL;
+ free_pages_map = NULL;
+ memory_bm_free(bm1, PG_UNSAFE_CLEAR);
+ kfree(bm1);
+ memory_bm_free(bm2, PG_UNSAFE_CLEAR);
+ kfree(bm2);
+
+ pr_debug("PM: Basic memory bitmaps freed\n");
}
/**
- * alloc_pagedir - Allocate the page directory.
- *
- * First, determine exactly how many pages we need and
- * allocate them.
- *
- * We arrange the pages in a chain: each page is an array of PBES_PER_PAGE
- * struct pbe elements (pbes) and the last element in the page points
- * to the next page.
+ * snapshot_additional_pages - estimate the number of additional pages
+ * be needed for setting up the suspend image data structures for given
+ * zone (usually the returned value is greater than the exact number)
+ */
+
+unsigned int snapshot_additional_pages(struct zone *zone)
+{
+ unsigned int res;
+
+ res = DIV_ROUND_UP(zone->spanned_pages, BM_BITS_PER_BLOCK);
+ res += DIV_ROUND_UP(res * sizeof(struct bm_block), PAGE_SIZE);
+ return 2 * res;
+}
+
+#ifdef CONFIG_HIGHMEM
+/**
+ * count_free_highmem_pages - compute the total number of free highmem
+ * pages, system-wide.
+ */
+
+static unsigned int count_free_highmem_pages(void)
+{
+ struct zone *zone;
+ unsigned int cnt = 0;
+
+ for_each_populated_zone(zone)
+ if (is_highmem(zone))
+ cnt += zone_page_state(zone, NR_FREE_PAGES);
+
+ return cnt;
+}
+
+/**
+ * saveable_highmem_page - Determine whether a highmem page should be
+ * included in the suspend image.
*
- * On each page we set up a list of struct_pbe elements.
+ * We should save the page if it isn't Nosave or NosaveFree, or Reserved,
+ * and it isn't a part of a free chunk of pages.
+ */
+static struct page *saveable_highmem_page(struct zone *zone, unsigned long pfn)
+{
+ struct page *page;
+
+ if (!pfn_valid(pfn))
+ return NULL;
+
+ page = pfn_to_page(pfn);
+ if (page_zone(page) != zone)
+ return NULL;
+
+ BUG_ON(!PageHighMem(page));
+
+ if (swsusp_page_is_forbidden(page) || swsusp_page_is_free(page) ||
+ PageReserved(page))
+ return NULL;
+
+ return page;
+}
+
+/**
+ * count_highmem_pages - compute the total number of saveable highmem
+ * pages.
*/
-static struct pbe *alloc_pagedir(unsigned int nr_pages, gfp_t gfp_mask,
- int safe_needed)
+static unsigned int count_highmem_pages(void)
{
- unsigned int num;
- struct pbe *pblist, *pbe;
+ struct zone *zone;
+ unsigned int n = 0;
- if (!nr_pages)
+ for_each_populated_zone(zone) {
+ unsigned long pfn, max_zone_pfn;
+
+ if (!is_highmem(zone))
+ continue;
+
+ mark_free_pages(zone);
+ max_zone_pfn = zone->zone_start_pfn + zone->spanned_pages;
+ for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++)
+ if (saveable_highmem_page(zone, pfn))
+ n++;
+ }
+ return n;
+}
+#else
+static inline void *saveable_highmem_page(struct zone *z, unsigned long p)
+{
+ return NULL;
+}
+#endif /* CONFIG_HIGHMEM */
+
+/**
+ * saveable_page - Determine whether a non-highmem page should be included
+ * in the suspend image.
+ *
+ * We should save the page if it isn't Nosave, and is not in the range
+ * of pages statically defined as 'unsaveable', and it isn't a part of
+ * a free chunk of pages.
+ */
+static struct page *saveable_page(struct zone *zone, unsigned long pfn)
+{
+ struct page *page;
+
+ if (!pfn_valid(pfn))
return NULL;
- pblist = alloc_image_page(gfp_mask, safe_needed);
- pbe = pblist;
- for (num = PBES_PER_PAGE; num < nr_pages; num += PBES_PER_PAGE) {
- if (!pbe) {
- free_pagedir(pblist, PG_UNSAFE_CLEAR);
- return NULL;
+ page = pfn_to_page(pfn);
+ if (page_zone(page) != zone)
+ return NULL;
+
+ BUG_ON(PageHighMem(page));
+
+ if (swsusp_page_is_forbidden(page) || swsusp_page_is_free(page))
+ return NULL;
+
+ if (PageReserved(page)
+ && (!kernel_page_present(page) || pfn_is_nosave(pfn)))
+ return NULL;
+
+ return page;
+}
+
+/**
+ * count_data_pages - compute the total number of saveable non-highmem
+ * pages.
+ */
+
+static unsigned int count_data_pages(void)
+{
+ struct zone *zone;
+ unsigned long pfn, max_zone_pfn;
+ unsigned int n = 0;
+
+ for_each_populated_zone(zone) {
+ if (is_highmem(zone))
+ continue;
+
+ mark_free_pages(zone);
+ max_zone_pfn = zone->zone_start_pfn + zone->spanned_pages;
+ for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++)
+ if (saveable_page(zone, pfn))
+ n++;
+ }
+ return n;
+}
+
+/* This is needed, because copy_page and memcpy are not usable for copying
+ * task structs.
+ */
+static inline void do_copy_page(long *dst, long *src)
+{
+ int n;
+
+ for (n = PAGE_SIZE / sizeof(long); n; n--)
+ *dst++ = *src++;
+}
+
+
+/**
+ * safe_copy_page - check if the page we are going to copy is marked as
+ * present in the kernel page tables (this always is the case if
+ * CONFIG_DEBUG_PAGEALLOC is not set and in that case
+ * kernel_page_present() always returns 'true').
+ */
+static void safe_copy_page(void *dst, struct page *s_page)
+{
+ if (kernel_page_present(s_page)) {
+ do_copy_page(dst, page_address(s_page));
+ } else {
+ kernel_map_pages(s_page, 1, 1);
+ do_copy_page(dst, page_address(s_page));
+ kernel_map_pages(s_page, 1, 0);
+ }
+}
+
+
+#ifdef CONFIG_HIGHMEM
+static inline struct page *
+page_is_saveable(struct zone *zone, unsigned long pfn)
+{
+ return is_highmem(zone) ?
+ saveable_highmem_page(zone, pfn) : saveable_page(zone, pfn);
+}
+
+static void copy_data_page(unsigned long dst_pfn, unsigned long src_pfn)
+{
+ struct page *s_page, *d_page;
+ void *src, *dst;
+
+ s_page = pfn_to_page(src_pfn);
+ d_page = pfn_to_page(dst_pfn);
+ if (PageHighMem(s_page)) {
+ src = kmap_atomic(s_page, KM_USER0);
+ dst = kmap_atomic(d_page, KM_USER1);
+ do_copy_page(dst, src);
+ kunmap_atomic(src, KM_USER0);
+ kunmap_atomic(dst, KM_USER1);
+ } else {
+ if (PageHighMem(d_page)) {
+ /* Page pointed to by src may contain some kernel
+ * data modified by kmap_atomic()
+ */
+ safe_copy_page(buffer, s_page);
+ dst = kmap_atomic(d_page, KM_USER0);
+ memcpy(dst, buffer, PAGE_SIZE);
+ kunmap_atomic(dst, KM_USER0);
+ } else {
+ safe_copy_page(page_address(d_page), s_page);
}
- pbe += PB_PAGE_SKIP;
- pbe->next = alloc_image_page(gfp_mask, safe_needed);
- pbe = pbe->next;
}
- create_pbe_list(pblist, nr_pages);
- return pblist;
+}
+#else
+#define page_is_saveable(zone, pfn) saveable_page(zone, pfn)
+
+static inline void copy_data_page(unsigned long dst_pfn, unsigned long src_pfn)
+{
+ safe_copy_page(page_address(pfn_to_page(dst_pfn)),
+ pfn_to_page(src_pfn));
+}
+#endif /* CONFIG_HIGHMEM */
+
+static void
+copy_data_pages(struct memory_bitmap *copy_bm, struct memory_bitmap *orig_bm)
+{
+ struct zone *zone;
+ unsigned long pfn;
+
+ for_each_populated_zone(zone) {
+ unsigned long max_zone_pfn;
+
+ mark_free_pages(zone);
+ max_zone_pfn = zone->zone_start_pfn + zone->spanned_pages;
+ for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++)
+ if (page_is_saveable(zone, pfn))
+ memory_bm_set_bit(orig_bm, pfn);
+ }
+ memory_bm_position_reset(orig_bm);
+ memory_bm_position_reset(copy_bm);
+ for(;;) {
+ pfn = memory_bm_next_pfn(orig_bm);
+ if (unlikely(pfn == BM_END_OF_MAP))
+ break;
+ copy_data_page(memory_bm_next_pfn(copy_bm), pfn);
+ }
}
+/* Total number of image pages */
+static unsigned int nr_copy_pages;
+/* Number of pages needed for saving the original pfns of the image pages */
+static unsigned int nr_meta_pages;
+/*
+ * Numbers of normal and highmem page frames allocated for hibernation image
+ * before suspending devices.
+ */
+unsigned int alloc_normal, alloc_highmem;
+/*
+ * Memory bitmap used for marking saveable pages (during hibernation) or
+ * hibernation image pages (during restore)
+ */
+static struct memory_bitmap orig_bm;
+/*
+ * Memory bitmap used during hibernation for marking allocated page frames that
+ * will contain copies of saveable pages. During restore it is initially used
+ * for marking hibernation image pages, but then the set bits from it are
+ * duplicated in @orig_bm and it is released. On highmem systems it is next
+ * used for marking "safe" highmem pages, but it has to be reinitialized for
+ * this purpose.
+ */
+static struct memory_bitmap copy_bm;
+
/**
- * Free pages we allocated for suspend. Suspend pages are alocated
- * before atomic copy, so we need to free them after resume.
+ * swsusp_free - free pages allocated for the suspend.
+ *
+ * Suspend pages are alocated before the atomic copy is made, so we
+ * need to release them after the resume.
*/
void swsusp_free(void)
struct zone *zone;
unsigned long pfn, max_zone_pfn;
- for_each_zone(zone) {
+ for_each_populated_zone(zone) {
max_zone_pfn = zone->zone_start_pfn + zone->spanned_pages;
for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++)
if (pfn_valid(pfn)) {
struct page *page = pfn_to_page(pfn);
- if (PageNosave(page) && PageNosaveFree(page)) {
- ClearPageNosave(page);
- ClearPageNosaveFree(page);
- free_page((long) page_address(page));
+ if (swsusp_page_is_forbidden(page) &&
+ swsusp_page_is_free(page)) {
+ swsusp_unset_page_forbidden(page);
+ swsusp_unset_page_free(page);
+ __free_page(page);
}
}
}
nr_meta_pages = 0;
restore_pblist = NULL;
buffer = NULL;
+ alloc_normal = 0;
+ alloc_highmem = 0;
+}
+
+/* Helper functions used for the shrinking of memory. */
+
+#define GFP_IMAGE (GFP_KERNEL | __GFP_NOWARN)
+
+/**
+ * preallocate_image_pages - Allocate a number of pages for hibernation image
+ * @nr_pages: Number of page frames to allocate.
+ * @mask: GFP flags to use for the allocation.
+ *
+ * Return value: Number of page frames actually allocated
+ */
+static unsigned long preallocate_image_pages(unsigned long nr_pages, gfp_t mask)
+{
+ unsigned long nr_alloc = 0;
+
+ while (nr_pages > 0) {
+ struct page *page;
+
+ page = alloc_image_page(mask);
+ if (!page)
+ break;
+ memory_bm_set_bit(©_bm, page_to_pfn(page));
+ if (PageHighMem(page))
+ alloc_highmem++;
+ else
+ alloc_normal++;
+ nr_pages--;
+ nr_alloc++;
+ }
+
+ return nr_alloc;
+}
+
+static unsigned long preallocate_image_memory(unsigned long nr_pages)
+{
+ return preallocate_image_pages(nr_pages, GFP_IMAGE);
+}
+
+#ifdef CONFIG_HIGHMEM
+static unsigned long preallocate_image_highmem(unsigned long nr_pages)
+{
+ return preallocate_image_pages(nr_pages, GFP_IMAGE | __GFP_HIGHMEM);
+}
+
+/**
+ * __fraction - Compute (an approximation of) x * (multiplier / base)
+ */
+static unsigned long __fraction(u64 x, u64 multiplier, u64 base)
+{
+ x *= multiplier;
+ do_div(x, base);
+ return (unsigned long)x;
+}
+
+static unsigned long preallocate_highmem_fraction(unsigned long nr_pages,
+ unsigned long highmem,
+ unsigned long total)
+{
+ unsigned long alloc = __fraction(nr_pages, highmem, total);
+
+ return preallocate_image_pages(alloc, GFP_IMAGE | __GFP_HIGHMEM);
+}
+#else /* CONFIG_HIGHMEM */
+static inline unsigned long preallocate_image_highmem(unsigned long nr_pages)
+{
+ return 0;
+}
+
+static inline unsigned long preallocate_highmem_fraction(unsigned long nr_pages,
+ unsigned long highmem,
+ unsigned long total)
+{
+ return 0;
+}
+#endif /* CONFIG_HIGHMEM */
+
+/**
+ * free_unnecessary_pages - Release preallocated pages not needed for the image
+ */
+static void free_unnecessary_pages(void)
+{
+ unsigned long save_highmem, to_free_normal, to_free_highmem;
+
+ to_free_normal = alloc_normal - count_data_pages();
+ save_highmem = count_highmem_pages();
+ if (alloc_highmem > save_highmem) {
+ to_free_highmem = alloc_highmem - save_highmem;
+ } else {
+ to_free_highmem = 0;
+ to_free_normal -= save_highmem - alloc_highmem;
+ }
+
+ memory_bm_position_reset(©_bm);
+
+ while (to_free_normal > 0 && to_free_highmem > 0) {
+ unsigned long pfn = memory_bm_next_pfn(©_bm);
+ struct page *page = pfn_to_page(pfn);
+
+ if (PageHighMem(page)) {
+ if (!to_free_highmem)
+ continue;
+ to_free_highmem--;
+ alloc_highmem--;
+ } else {
+ if (!to_free_normal)
+ continue;
+ to_free_normal--;
+ alloc_normal--;
+ }
+ memory_bm_clear_bit(©_bm, pfn);
+ swsusp_unset_page_forbidden(page);
+ swsusp_unset_page_free(page);
+ __free_page(page);
+ }
+}
+
+/**
+ * minimum_image_size - Estimate the minimum acceptable size of an image
+ * @saveable: Number of saveable pages in the system.
+ *
+ * We want to avoid attempting to free too much memory too hard, so estimate the
+ * minimum acceptable size of a hibernation image to use as the lower limit for
+ * preallocating memory.
+ *
+ * We assume that the minimum image size should be proportional to
+ *
+ * [number of saveable pages] - [number of pages that can be freed in theory]
+ *
+ * where the second term is the sum of (1) reclaimable slab pages, (2) active
+ * and (3) inactive anonymouns pages, (4) active and (5) inactive file pages,
+ * minus mapped file pages.
+ */
+static unsigned long minimum_image_size(unsigned long saveable)
+{
+ unsigned long size;
+
+ size = global_page_state(NR_SLAB_RECLAIMABLE)
+ + global_page_state(NR_ACTIVE_ANON)
+ + global_page_state(NR_INACTIVE_ANON)
+ + global_page_state(NR_ACTIVE_FILE)
+ + global_page_state(NR_INACTIVE_FILE)
+ - global_page_state(NR_FILE_MAPPED);
+
+ return saveable <= size ? 0 : saveable - size;
+}
+
+/**
+ * hibernate_preallocate_memory - Preallocate memory for hibernation image
+ *
+ * To create a hibernation image it is necessary to make a copy of every page
+ * frame in use. We also need a number of page frames to be free during
+ * hibernation for allocations made while saving the image and for device
+ * drivers, in case they need to allocate memory from their hibernation
+ * callbacks (these two numbers are given by PAGES_FOR_IO and SPARE_PAGES,
+ * respectively, both of which are rough estimates). To make this happen, we
+ * compute the total number of available page frames and allocate at least
+ *
+ * ([page frames total] + PAGES_FOR_IO + [metadata pages]) / 2 + 2 * SPARE_PAGES
+ *
+ * of them, which corresponds to the maximum size of a hibernation image.
+ *
+ * If image_size is set below the number following from the above formula,
+ * the preallocation of memory is continued until the total number of saveable
+ * pages in the system is below the requested image size or the minimum
+ * acceptable image size returned by minimum_image_size(), whichever is greater.
+ */
+int hibernate_preallocate_memory(void)
+{
+ struct zone *zone;
+ unsigned long saveable, size, max_size, count, highmem, pages = 0;
+ unsigned long alloc, save_highmem, pages_highmem;
+ struct timeval start, stop;
+ int error;
+
+ printk(KERN_INFO "PM: Preallocating image memory... ");
+ do_gettimeofday(&start);
+
+ error = memory_bm_create(&orig_bm, GFP_IMAGE, PG_ANY);
+ if (error)
+ goto err_out;
+
+ error = memory_bm_create(©_bm, GFP_IMAGE, PG_ANY);
+ if (error)
+ goto err_out;
+
+ alloc_normal = 0;
+ alloc_highmem = 0;
+
+ /* Count the number of saveable data pages. */
+ save_highmem = count_highmem_pages();
+ saveable = count_data_pages();
+
+ /*
+ * Compute the total number of page frames we can use (count) and the
+ * number of pages needed for image metadata (size).
+ */
+ count = saveable;
+ saveable += save_highmem;
+ highmem = save_highmem;
+ size = 0;
+ for_each_populated_zone(zone) {
+ size += snapshot_additional_pages(zone);
+ if (is_highmem(zone))
+ highmem += zone_page_state(zone, NR_FREE_PAGES);
+ else
+ count += zone_page_state(zone, NR_FREE_PAGES);
+ }
+ count += highmem;
+ count -= totalreserve_pages;
+
+ /* Compute the maximum number of saveable pages to leave in memory. */
+ max_size = (count - (size + PAGES_FOR_IO)) / 2 - 2 * SPARE_PAGES;
+ size = DIV_ROUND_UP(image_size, PAGE_SIZE);
+ if (size > max_size)
+ size = max_size;
+ /*
+ * If the maximum is not less than the current number of saveable pages
+ * in memory, allocate page frames for the image and we're done.
+ */
+ if (size >= saveable) {
+ pages = preallocate_image_highmem(save_highmem);
+ pages += preallocate_image_memory(saveable - pages);
+ goto out;
+ }
+
+ /* Estimate the minimum size of the image. */
+ pages = minimum_image_size(saveable);
+ if (size < pages)
+ size = min_t(unsigned long, pages, max_size);
+
+ /*
+ * Let the memory management subsystem know that we're going to need a
+ * large number of page frames to allocate and make it free some memory.
+ * NOTE: If this is not done, performance will be hurt badly in some
+ * test cases.
+ */
+ shrink_all_memory(saveable - size);
+
+ /*
+ * The number of saveable pages in memory was too high, so apply some
+ * pressure to decrease it. First, make room for the largest possible
+ * image and fail if that doesn't work. Next, try to decrease the size
+ * of the image as much as indicated by 'size' using allocations from
+ * highmem and non-highmem zones separately.
+ */
+ pages_highmem = preallocate_image_highmem(highmem / 2);
+ alloc = (count - max_size) - pages_highmem;
+ pages = preallocate_image_memory(alloc);
+ if (pages < alloc)
+ goto err_out;
+ size = max_size - size;
+ alloc = size;
+ size = preallocate_highmem_fraction(size, highmem, count);
+ pages_highmem += size;
+ alloc -= size;
+ pages += preallocate_image_memory(alloc);
+ pages += pages_highmem;
+
+ /*
+ * We only need as many page frames for the image as there are saveable
+ * pages in memory, but we have allocated more. Release the excessive
+ * ones now.
+ */
+ free_unnecessary_pages();
+
+ out:
+ do_gettimeofday(&stop);
+ printk(KERN_CONT "done (allocated %lu pages)\n", pages);
+ swsusp_show_speed(&start, &stop, pages, "Allocated");
+
+ return 0;
+
+ err_out:
+ printk(KERN_CONT "\n");
+ swsusp_free();
+ return -ENOMEM;
+}
+
+#ifdef CONFIG_HIGHMEM
+/**
+ * count_pages_for_highmem - compute the number of non-highmem pages
+ * that will be necessary for creating copies of highmem pages.
+ */
+
+static unsigned int count_pages_for_highmem(unsigned int nr_highmem)
+{
+ unsigned int free_highmem = count_free_highmem_pages() + alloc_highmem;
+
+ if (free_highmem >= nr_highmem)
+ nr_highmem = 0;
+ else
+ nr_highmem -= free_highmem;
+
+ return nr_highmem;
+}
+#else
+static unsigned int
+count_pages_for_highmem(unsigned int nr_highmem) { return 0; }
+#endif /* CONFIG_HIGHMEM */
+
+/**
+ * enough_free_mem - Make sure we have enough free memory for the
+ * snapshot image.
+ */
+
+static int enough_free_mem(unsigned int nr_pages, unsigned int nr_highmem)
+{
+ struct zone *zone;
+ unsigned int free = alloc_normal;
+
+ for_each_populated_zone(zone)
+ if (!is_highmem(zone))
+ free += zone_page_state(zone, NR_FREE_PAGES);
+
+ nr_pages += count_pages_for_highmem(nr_highmem);
+ pr_debug("PM: Normal pages needed: %u + %u, available pages: %u\n",
+ nr_pages, PAGES_FOR_IO, free);
+
+ return free > nr_pages + PAGES_FOR_IO;
+}
+
+#ifdef CONFIG_HIGHMEM
+/**
+ * get_highmem_buffer - if there are some highmem pages in the suspend
+ * image, we may need the buffer to copy them and/or load their data.
+ */
+
+static inline int get_highmem_buffer(int safe_needed)
+{
+ buffer = get_image_page(GFP_ATOMIC | __GFP_COLD, safe_needed);
+ return buffer ? 0 : -ENOMEM;
}
-
/**
- * enough_free_mem - Make sure we enough free memory to snapshot.
- *
- * Returns TRUE or FALSE after checking the number of available
- * free pages.
+ * alloc_highmem_image_pages - allocate some highmem pages for the image.
+ * Try to allocate as many pages as needed, but if the number of free
+ * highmem pages is lesser than that, allocate them all.
*/
-static int enough_free_mem(unsigned int nr_pages)
+static inline unsigned int
+alloc_highmem_pages(struct memory_bitmap *bm, unsigned int nr_highmem)
{
- struct zone *zone;
- unsigned int n = 0;
+ unsigned int to_alloc = count_free_highmem_pages();
- for_each_zone (zone)
- if (!is_highmem(zone))
- n += zone->free_pages;
- pr_debug("swsusp: available memory: %u pages\n", n);
- return n > (nr_pages + PAGES_FOR_IO +
- (nr_pages + PBES_PER_PAGE - 1) / PBES_PER_PAGE);
+ if (to_alloc > nr_highmem)
+ to_alloc = nr_highmem;
+
+ nr_highmem -= to_alloc;
+ while (to_alloc-- > 0) {
+ struct page *page;
+
+ page = alloc_image_page(__GFP_HIGHMEM);
+ memory_bm_set_bit(bm, page_to_pfn(page));
+ }
+ return nr_highmem;
}
+#else
+static inline int get_highmem_buffer(int safe_needed) { return 0; }
+
+static inline unsigned int
+alloc_highmem_pages(struct memory_bitmap *bm, unsigned int n) { return 0; }
+#endif /* CONFIG_HIGHMEM */
+
+/**
+ * swsusp_alloc - allocate memory for the suspend image
+ *
+ * We first try to allocate as many highmem pages as there are
+ * saveable highmem pages in the system. If that fails, we allocate
+ * non-highmem pages for the copies of the remaining highmem ones.
+ *
+ * In this approach it is likely that the copies of highmem pages will
+ * also be located in the high memory, because of the way in which
+ * copy_data_pages() works.
+ */
static int
swsusp_alloc(struct memory_bitmap *orig_bm, struct memory_bitmap *copy_bm,
- unsigned int nr_pages)
+ unsigned int nr_pages, unsigned int nr_highmem)
{
- int error;
-
- error = memory_bm_create(orig_bm, GFP_ATOMIC | __GFP_COLD, PG_ANY);
- if (error)
- goto Free;
-
- error = memory_bm_create(copy_bm, GFP_ATOMIC | __GFP_COLD, PG_ANY);
- if (error)
- goto Free;
+ int error = 0;
- while (nr_pages-- > 0) {
- struct page *page = alloc_page(GFP_ATOMIC | __GFP_COLD);
- if (!page)
- goto Free;
+ if (nr_highmem > 0) {
+ error = get_highmem_buffer(PG_ANY);
+ if (error)
+ goto err_out;
+ if (nr_highmem > alloc_highmem) {
+ nr_highmem -= alloc_highmem;
+ nr_pages += alloc_highmem_pages(copy_bm, nr_highmem);
+ }
+ }
+ if (nr_pages > alloc_normal) {
+ nr_pages -= alloc_normal;
+ while (nr_pages-- > 0) {
+ struct page *page;
- SetPageNosave(page);
- SetPageNosaveFree(page);
- memory_bm_set_bit(copy_bm, page_to_pfn(page));
+ page = alloc_image_page(GFP_ATOMIC | __GFP_COLD);
+ if (!page)
+ goto err_out;
+ memory_bm_set_bit(copy_bm, page_to_pfn(page));
+ }
}
+
return 0;
-Free:
+ err_out:
swsusp_free();
- return -ENOMEM;
+ return error;
}
-/* Memory bitmap used for marking saveable pages */
-static struct memory_bitmap orig_bm;
-/* Memory bitmap used for marking allocated pages that will contain the copies
- * of saveable pages
- */
-static struct memory_bitmap copy_bm;
-
asmlinkage int swsusp_save(void)
{
- unsigned int nr_pages;
+ unsigned int nr_pages, nr_highmem;
- pr_debug("swsusp: critical section: \n");
+ printk(KERN_INFO "PM: Creating hibernation image: \n");
- drain_local_pages();
+ drain_local_pages(NULL);
nr_pages = count_data_pages();
- printk("swsusp: Need to copy %u pages\n", nr_pages);
-
- pr_debug("swsusp: pages needed: %u + %lu + %u, free: %u\n",
- nr_pages,
- (nr_pages + PBES_PER_PAGE - 1) / PBES_PER_PAGE,
- PAGES_FOR_IO, nr_free_pages());
+ nr_highmem = count_highmem_pages();
+ printk(KERN_INFO "PM: Need to copy %u pages\n", nr_pages + nr_highmem);
- if (!enough_free_mem(nr_pages)) {
- printk(KERN_ERR "swsusp: Not enough free memory\n");
+ if (!enough_free_mem(nr_pages, nr_highmem)) {
+ printk(KERN_ERR "PM: Not enough free memory\n");
return -ENOMEM;
}
- if (swsusp_alloc(&orig_bm, ©_bm, nr_pages))
+ if (swsusp_alloc(&orig_bm, ©_bm, nr_pages, nr_highmem)) {
+ printk(KERN_ERR "PM: Memory allocation failed\n");
return -ENOMEM;
+ }
/* During allocating of suspend pagedir, new cold pages may appear.
* Kill them.
*/
- drain_local_pages();
+ drain_local_pages(NULL);
copy_data_pages(©_bm, &orig_bm);
/*
* touch swap space! Except we must write out our image of course.
*/
+ nr_pages += nr_highmem;
nr_copy_pages = nr_pages;
- nr_meta_pages = (nr_pages * sizeof(long) + PAGE_SIZE - 1) >> PAGE_SHIFT;
+ nr_meta_pages = DIV_ROUND_UP(nr_pages * sizeof(long), PAGE_SIZE);
+
+ printk(KERN_INFO "PM: Hibernation image created (%d pages copied)\n",
+ nr_pages);
- printk("swsusp: critical section/: done (%d pages copied)\n", nr_pages);
return 0;
}
-static void init_header(struct swsusp_info *info)
+#ifndef CONFIG_ARCH_HIBERNATION_HEADER
+static int init_header_complete(struct swsusp_info *info)
{
- memset(info, 0, sizeof(struct swsusp_info));
+ memcpy(&info->uts, init_utsname(), sizeof(struct new_utsname));
info->version_code = LINUX_VERSION_CODE;
+ return 0;
+}
+
+static char *check_image_kernel(struct swsusp_info *info)
+{
+ if (info->version_code != LINUX_VERSION_CODE)
+ return "kernel version";
+ if (strcmp(info->uts.sysname,init_utsname()->sysname))
+ return "system type";
+ if (strcmp(info->uts.release,init_utsname()->release))
+ return "kernel release";
+ if (strcmp(info->uts.version,init_utsname()->version))
+ return "version";
+ if (strcmp(info->uts.machine,init_utsname()->machine))
+ return "machine";
+ return NULL;
+}
+#endif /* CONFIG_ARCH_HIBERNATION_HEADER */
+
+unsigned long snapshot_get_image_size(void)
+{
+ return nr_copy_pages + nr_meta_pages + 1;
+}
+
+static int init_header(struct swsusp_info *info)
+{
+ memset(info, 0, sizeof(struct swsusp_info));
info->num_physpages = num_physpages;
- memcpy(&info->uts, &system_utsname, sizeof(system_utsname));
- info->cpus = num_online_cpus();
info->image_pages = nr_copy_pages;
- info->pages = nr_copy_pages + nr_meta_pages + 1;
+ info->pages = snapshot_get_image_size();
info->size = info->pages;
info->size <<= PAGE_SHIFT;
+ return init_header_complete(info);
}
/**
- * pack_addresses - the addresses corresponding to pfns found in the
- * bitmap @bm are stored in the array @buf[] (1 page)
+ * pack_pfns - pfns corresponding to the set bits found in the bitmap @bm
+ * are stored in the array @buf[] (1 page at a time)
*/
static inline void
-pack_addresses(unsigned long *buf, struct memory_bitmap *bm)
+pack_pfns(unsigned long *buf, struct memory_bitmap *bm)
{
int j;
for (j = 0; j < PAGE_SIZE / sizeof(long); j++) {
- unsigned long pfn = memory_bm_next_pfn(bm);
-
- if (unlikely(pfn == BM_END_OF_MAP))
+ buf[j] = memory_bm_next_pfn(bm);
+ if (unlikely(buf[j] == BM_END_OF_MAP))
break;
-
- buf[j] = (unsigned long)page_address(pfn_to_page(pfn));
}
}
if (!buffer) {
/* This makes the buffer be freed by swsusp_free() */
- buffer = alloc_image_page(GFP_ATOMIC, PG_ANY);
+ buffer = get_image_page(GFP_ATOMIC, PG_ANY);
if (!buffer)
return -ENOMEM;
}
if (!handle->offset) {
- init_header((struct swsusp_info *)buffer);
+ int error;
+
+ error = init_header((struct swsusp_info *)buffer);
+ if (error)
+ return error;
handle->buffer = buffer;
memory_bm_position_reset(&orig_bm);
memory_bm_position_reset(©_bm);
if (handle->prev < handle->cur) {
if (handle->cur <= nr_meta_pages) {
memset(buffer, 0, PAGE_SIZE);
- pack_addresses(buffer, &orig_bm);
+ pack_pfns(buffer, &orig_bm);
} else {
- unsigned long pfn = memory_bm_next_pfn(©_bm);
+ struct page *page;
- handle->buffer = page_address(pfn_to_page(pfn));
+ page = pfn_to_page(memory_bm_next_pfn(©_bm));
+ if (PageHighMem(page)) {
+ /* Highmem pages are copied to the buffer,
+ * because we can't return with a kmapped
+ * highmem page (we may not be called again).
+ */
+ void *kaddr;
+
+ kaddr = kmap_atomic(page, KM_USER0);
+ memcpy(buffer, kaddr, PAGE_SIZE);
+ kunmap_atomic(kaddr, KM_USER0);
+ handle->buffer = buffer;
+ } else {
+ handle->buffer = page_address(page);
+ }
}
handle->prev = handle->cur;
}
* had been used before suspend
*/
-static int mark_unsafe_pages(struct pbe *pblist)
+static int mark_unsafe_pages(struct memory_bitmap *bm)
{
struct zone *zone;
unsigned long pfn, max_zone_pfn;
- struct pbe *p;
-
- if (!pblist) /* a sanity check */
- return -EINVAL;
/* Clear page flags */
- for_each_zone (zone) {
+ for_each_populated_zone(zone) {
max_zone_pfn = zone->zone_start_pfn + zone->spanned_pages;
for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++)
if (pfn_valid(pfn))
- ClearPageNosaveFree(pfn_to_page(pfn));
+ swsusp_unset_page_free(pfn_to_page(pfn));
}
- /* Mark orig addresses */
- for_each_pbe (p, pblist) {
- if (virt_addr_valid(p->orig_address))
- SetPageNosaveFree(virt_to_page(p->orig_address));
- else
- return -EFAULT;
- }
+ /* Mark pages that correspond to the "original" pfns as "unsafe" */
+ memory_bm_position_reset(bm);
+ do {
+ pfn = memory_bm_next_pfn(bm);
+ if (likely(pfn != BM_END_OF_MAP)) {
+ if (likely(pfn_valid(pfn)))
+ swsusp_set_page_free(pfn_to_page(pfn));
+ else
+ return -EFAULT;
+ }
+ } while (pfn != BM_END_OF_MAP);
- unsafe_pages = 0;
+ allocated_unsafe_pages = 0;
return 0;
}
-static void copy_page_backup_list(struct pbe *dst, struct pbe *src)
+static void
+duplicate_memory_bitmap(struct memory_bitmap *dst, struct memory_bitmap *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;
+ unsigned long pfn;
+
+ memory_bm_position_reset(src);
+ pfn = memory_bm_next_pfn(src);
+ while (pfn != BM_END_OF_MAP) {
+ memory_bm_set_bit(dst, pfn);
+ pfn = memory_bm_next_pfn(src);
}
}
static int check_header(struct swsusp_info *info)
{
- char *reason = NULL;
+ char *reason;
- if (info->version_code != LINUX_VERSION_CODE)
- reason = "kernel version";
- if (info->num_physpages != num_physpages)
+ reason = check_image_kernel(info);
+ if (!reason && info->num_physpages != num_physpages)
reason = "memory size";
- if (strcmp(info->uts.sysname,system_utsname.sysname))
- reason = "system type";
- if (strcmp(info->uts.release,system_utsname.release))
- reason = "kernel release";
- if (strcmp(info->uts.version,system_utsname.version))
- reason = "version";
- if (strcmp(info->uts.machine,system_utsname.machine))
- reason = "machine";
if (reason) {
- printk(KERN_ERR "swsusp: Resume mismatch: %s\n", reason);
+ printk(KERN_ERR "PM: Image mismatch: %s\n", reason);
return -EPERM;
}
return 0;
* load header - check the image header and copy data from it
*/
-static int load_header(struct snapshot_handle *handle,
- struct swsusp_info *info)
+static int
+load_header(struct swsusp_info *info)
{
int error;
- struct pbe *pblist;
+ restore_pblist = NULL;
error = check_header(info);
if (!error) {
- pblist = alloc_pagedir(info->image_pages, GFP_ATOMIC, PG_ANY);
- if (!pblist)
- return -ENOMEM;
- restore_pblist = pblist;
- handle->pbe = pblist;
nr_copy_pages = info->image_pages;
nr_meta_pages = info->pages - info->image_pages - 1;
}
}
/**
- * unpack_orig_addresses - copy the elements of @buf[] (1 page) to
- * the PBEs in the list starting at @pbe
+ * unpack_orig_pfns - for each element of @buf[] (1 page at a time) set
+ * the corresponding bit in the memory bitmap @bm
*/
-
-static inline struct pbe *unpack_orig_addresses(unsigned long *buf,
- struct pbe *pbe)
+static int unpack_orig_pfns(unsigned long *buf, struct memory_bitmap *bm)
{
int j;
- for (j = 0; j < PAGE_SIZE / sizeof(long) && pbe; j++) {
- pbe->orig_address = buf[j];
- pbe = pbe->next;
+ for (j = 0; j < PAGE_SIZE / sizeof(long); j++) {
+ if (unlikely(buf[j] == BM_END_OF_MAP))
+ break;
+
+ if (memory_bm_pfn_present(bm, buf[j]))
+ memory_bm_set_bit(bm, buf[j]);
+ else
+ return -EFAULT;
+ }
+
+ return 0;
+}
+
+/* List of "safe" pages that may be used to store data loaded from the suspend
+ * image
+ */
+static struct linked_page *safe_pages_list;
+
+#ifdef CONFIG_HIGHMEM
+/* struct highmem_pbe is used for creating the list of highmem pages that
+ * should be restored atomically during the resume from disk, because the page
+ * frames they have occupied before the suspend are in use.
+ */
+struct highmem_pbe {
+ struct page *copy_page; /* data is here now */
+ struct page *orig_page; /* data was here before the suspend */
+ struct highmem_pbe *next;
+};
+
+/* List of highmem PBEs needed for restoring the highmem pages that were
+ * allocated before the suspend and included in the suspend image, but have
+ * also been allocated by the "resume" kernel, so their contents cannot be
+ * written directly to their "original" page frames.
+ */
+static struct highmem_pbe *highmem_pblist;
+
+/**
+ * count_highmem_image_pages - compute the number of highmem pages in the
+ * suspend image. The bits in the memory bitmap @bm that correspond to the
+ * image pages are assumed to be set.
+ */
+
+static unsigned int count_highmem_image_pages(struct memory_bitmap *bm)
+{
+ unsigned long pfn;
+ unsigned int cnt = 0;
+
+ memory_bm_position_reset(bm);
+ pfn = memory_bm_next_pfn(bm);
+ while (pfn != BM_END_OF_MAP) {
+ if (PageHighMem(pfn_to_page(pfn)))
+ cnt++;
+
+ pfn = memory_bm_next_pfn(bm);
}
- return pbe;
+ return cnt;
}
/**
- * prepare_image - use metadata contained in the PBE list
- * pointed to by restore_pblist to mark the pages that will
- * be overwritten in the process of restoring the system
- * memory state from the image ("unsafe" pages) and allocate
- * memory for the image
+ * prepare_highmem_image - try to allocate as many highmem pages as
+ * there are highmem image pages (@nr_highmem_p points to the variable
+ * containing the number of highmem image pages). The pages that are
+ * "safe" (ie. will not be overwritten when the suspend image is
+ * restored) have the corresponding bits set in @bm (it must be
+ * unitialized).
*
- * The idea is to allocate the PBE list first and then
- * allocate as many pages as it's needed for the image data,
- * but not to assign these pages to the PBEs initially.
- * Instead, we just mark them as allocated and create a list
- * of "safe" which will be used later
+ * NOTE: This function should not be called if there are no highmem
+ * image pages.
*/
-static struct linked_page *safe_pages;
+static unsigned int safe_highmem_pages;
-static int prepare_image(struct snapshot_handle *handle)
+static struct memory_bitmap *safe_highmem_bm;
+
+static int
+prepare_highmem_image(struct memory_bitmap *bm, unsigned int *nr_highmem_p)
{
- int error = 0;
- unsigned int nr_pages = nr_copy_pages;
- struct pbe *p, *pblist = NULL;
+ unsigned int to_alloc;
- p = restore_pblist;
- error = mark_unsafe_pages(p);
- if (!error) {
- pblist = alloc_pagedir(nr_pages, GFP_ATOMIC, PG_SAFE);
- if (pblist)
- copy_page_backup_list(pblist, p);
- free_pagedir(p, PG_UNSAFE_KEEP);
- if (!pblist)
+ if (memory_bm_create(bm, GFP_ATOMIC, PG_SAFE))
+ return -ENOMEM;
+
+ if (get_highmem_buffer(PG_SAFE))
+ return -ENOMEM;
+
+ to_alloc = count_free_highmem_pages();
+ if (to_alloc > *nr_highmem_p)
+ to_alloc = *nr_highmem_p;
+ else
+ *nr_highmem_p = to_alloc;
+
+ safe_highmem_pages = 0;
+ while (to_alloc-- > 0) {
+ struct page *page;
+
+ page = alloc_page(__GFP_HIGHMEM);
+ if (!swsusp_page_is_free(page)) {
+ /* The page is "safe", set its bit the bitmap */
+ memory_bm_set_bit(bm, page_to_pfn(page));
+ safe_highmem_pages++;
+ }
+ /* Mark the page as allocated */
+ swsusp_set_page_forbidden(page);
+ swsusp_set_page_free(page);
+ }
+ memory_bm_position_reset(bm);
+ safe_highmem_bm = bm;
+ return 0;
+}
+
+/**
+ * get_highmem_page_buffer - for given highmem image page find the buffer
+ * that suspend_write_next() should set for its caller to write to.
+ *
+ * If the page is to be saved to its "original" page frame or a copy of
+ * the page is to be made in the highmem, @buffer is returned. Otherwise,
+ * the copy of the page is to be made in normal memory, so the address of
+ * the copy is returned.
+ *
+ * If @buffer is returned, the caller of suspend_write_next() will write
+ * the page's contents to @buffer, so they will have to be copied to the
+ * right location on the next call to suspend_write_next() and it is done
+ * with the help of copy_last_highmem_page(). For this purpose, if
+ * @buffer is returned, @last_highmem page is set to the page to which
+ * the data will have to be copied from @buffer.
+ */
+
+static struct page *last_highmem_page;
+
+static void *
+get_highmem_page_buffer(struct page *page, struct chain_allocator *ca)
+{
+ struct highmem_pbe *pbe;
+ void *kaddr;
+
+ if (swsusp_page_is_forbidden(page) && swsusp_page_is_free(page)) {
+ /* We have allocated the "original" page frame and we can
+ * use it directly to store the loaded page.
+ */
+ last_highmem_page = page;
+ return buffer;
+ }
+ /* The "original" page frame has not been allocated and we have to
+ * use a "safe" page frame to store the loaded page.
+ */
+ pbe = chain_alloc(ca, sizeof(struct highmem_pbe));
+ if (!pbe) {
+ swsusp_free();
+ return ERR_PTR(-ENOMEM);
+ }
+ pbe->orig_page = page;
+ if (safe_highmem_pages > 0) {
+ struct page *tmp;
+
+ /* Copy of the page will be stored in high memory */
+ kaddr = buffer;
+ tmp = pfn_to_page(memory_bm_next_pfn(safe_highmem_bm));
+ safe_highmem_pages--;
+ last_highmem_page = tmp;
+ pbe->copy_page = tmp;
+ } else {
+ /* Copy of the page will be stored in normal memory */
+ kaddr = safe_pages_list;
+ safe_pages_list = safe_pages_list->next;
+ pbe->copy_page = virt_to_page(kaddr);
+ }
+ pbe->next = highmem_pblist;
+ highmem_pblist = pbe;
+ return kaddr;
+}
+
+/**
+ * copy_last_highmem_page - copy the contents of a highmem image from
+ * @buffer, where the caller of snapshot_write_next() has place them,
+ * to the right location represented by @last_highmem_page .
+ */
+
+static void copy_last_highmem_page(void)
+{
+ if (last_highmem_page) {
+ void *dst;
+
+ dst = kmap_atomic(last_highmem_page, KM_USER0);
+ memcpy(dst, buffer, PAGE_SIZE);
+ kunmap_atomic(dst, KM_USER0);
+ last_highmem_page = NULL;
+ }
+}
+
+static inline int last_highmem_page_copied(void)
+{
+ return !last_highmem_page;
+}
+
+static inline void free_highmem_data(void)
+{
+ if (safe_highmem_bm)
+ memory_bm_free(safe_highmem_bm, PG_UNSAFE_CLEAR);
+
+ if (buffer)
+ free_image_page(buffer, PG_UNSAFE_CLEAR);
+}
+#else
+static inline int get_safe_write_buffer(void) { return 0; }
+
+static unsigned int
+count_highmem_image_pages(struct memory_bitmap *bm) { return 0; }
+
+static inline int
+prepare_highmem_image(struct memory_bitmap *bm, unsigned int *nr_highmem_p)
+{
+ return 0;
+}
+
+static inline void *
+get_highmem_page_buffer(struct page *page, struct chain_allocator *ca)
+{
+ return ERR_PTR(-EINVAL);
+}
+
+static inline void copy_last_highmem_page(void) {}
+static inline int last_highmem_page_copied(void) { return 1; }
+static inline void free_highmem_data(void) {}
+#endif /* CONFIG_HIGHMEM */
+
+/**
+ * prepare_image - use the memory bitmap @bm to mark the pages that will
+ * be overwritten in the process of restoring the system memory state
+ * from the suspend image ("unsafe" pages) and allocate memory for the
+ * image.
+ *
+ * The idea is to allocate a new memory bitmap first and then allocate
+ * as many pages as needed for the image data, but not to assign these
+ * pages to specific tasks initially. Instead, we just mark them as
+ * allocated and create a lists of "safe" pages that will be used
+ * later. On systems with high memory a list of "safe" highmem pages is
+ * also created.
+ */
+
+#define PBES_PER_LINKED_PAGE (LINKED_PAGE_DATA_SIZE / sizeof(struct pbe))
+
+static int
+prepare_image(struct memory_bitmap *new_bm, struct memory_bitmap *bm)
+{
+ unsigned int nr_pages, nr_highmem;
+ struct linked_page *sp_list, *lp;
+ int error;
+
+ /* If there is no highmem, the buffer will not be necessary */
+ free_image_page(buffer, PG_UNSAFE_CLEAR);
+ buffer = NULL;
+
+ nr_highmem = count_highmem_image_pages(bm);
+ error = mark_unsafe_pages(bm);
+ if (error)
+ goto Free;
+
+ error = memory_bm_create(new_bm, GFP_ATOMIC, PG_SAFE);
+ if (error)
+ goto Free;
+
+ duplicate_memory_bitmap(new_bm, bm);
+ memory_bm_free(bm, PG_UNSAFE_KEEP);
+ if (nr_highmem > 0) {
+ error = prepare_highmem_image(bm, &nr_highmem);
+ if (error)
+ goto Free;
+ }
+ /* Reserve some safe pages for potential later use.
+ *
+ * NOTE: This way we make sure there will be enough safe pages for the
+ * chain_alloc() in get_buffer(). It is a bit wasteful, but
+ * nr_copy_pages cannot be greater than 50% of the memory anyway.
+ */
+ sp_list = NULL;
+ /* nr_copy_pages cannot be lesser than allocated_unsafe_pages */
+ nr_pages = nr_copy_pages - nr_highmem - allocated_unsafe_pages;
+ nr_pages = DIV_ROUND_UP(nr_pages, PBES_PER_LINKED_PAGE);
+ while (nr_pages > 0) {
+ lp = get_image_page(GFP_ATOMIC, PG_SAFE);
+ if (!lp) {
error = -ENOMEM;
+ goto Free;
+ }
+ lp->next = sp_list;
+ sp_list = lp;
+ nr_pages--;
}
- safe_pages = NULL;
- if (!error && nr_pages > unsafe_pages) {
- nr_pages -= unsafe_pages;
- while (nr_pages--) {
- struct linked_page *ptr;
-
- ptr = (void *)get_zeroed_page(GFP_ATOMIC);
- if (!ptr) {
- error = -ENOMEM;
- break;
- }
- if (!PageNosaveFree(virt_to_page(ptr))) {
- /* The page is "safe", add it to the list */
- ptr->next = safe_pages;
- safe_pages = ptr;
- }
- /* Mark the page as allocated */
- SetPageNosave(virt_to_page(ptr));
- SetPageNosaveFree(virt_to_page(ptr));
+ /* Preallocate memory for the image */
+ safe_pages_list = NULL;
+ nr_pages = nr_copy_pages - nr_highmem - allocated_unsafe_pages;
+ while (nr_pages > 0) {
+ lp = (struct linked_page *)get_zeroed_page(GFP_ATOMIC);
+ if (!lp) {
+ error = -ENOMEM;
+ goto Free;
+ }
+ if (!swsusp_page_is_free(virt_to_page(lp))) {
+ /* The page is "safe", add it to the list */
+ lp->next = safe_pages_list;
+ safe_pages_list = lp;
}
+ /* Mark the page as allocated */
+ swsusp_set_page_forbidden(virt_to_page(lp));
+ swsusp_set_page_free(virt_to_page(lp));
+ nr_pages--;
}
- if (!error) {
- restore_pblist = pblist;
- } else {
- handle->pbe = NULL;
- swsusp_free();
+ /* Free the reserved safe pages so that chain_alloc() can use them */
+ while (sp_list) {
+ lp = sp_list->next;
+ free_image_page(sp_list, PG_UNSAFE_CLEAR);
+ sp_list = lp;
}
+ return 0;
+
+ Free:
+ swsusp_free();
return error;
}
-static void *get_buffer(struct snapshot_handle *handle)
+/**
+ * get_buffer - compute the address that snapshot_write_next() should
+ * set for its caller to write to.
+ */
+
+static void *get_buffer(struct memory_bitmap *bm, struct chain_allocator *ca)
{
- struct pbe *pbe = handle->pbe, *last = handle->last_pbe;
- struct page *page = virt_to_page(pbe->orig_address);
+ struct pbe *pbe;
+ struct page *page;
+ unsigned long pfn = memory_bm_next_pfn(bm);
+
+ if (pfn == BM_END_OF_MAP)
+ return ERR_PTR(-EFAULT);
- if (PageNosave(page) && PageNosaveFree(page)) {
- /*
- * We have allocated the "original" page frame and we can
- * use it directly to store the read page
+ page = pfn_to_page(pfn);
+ if (PageHighMem(page))
+ return get_highmem_page_buffer(page, ca);
+
+ if (swsusp_page_is_forbidden(page) && swsusp_page_is_free(page))
+ /* We have allocated the "original" page frame and we can
+ * use it directly to store the loaded page.
*/
- pbe->address = 0;
- if (last && last->next)
- last->next = NULL;
- return (void *)pbe->orig_address;
- }
- /*
- * The "original" page frame has not been allocated and we have to
- * use a "safe" page frame to store the read page
+ return page_address(page);
+
+ /* The "original" page frame has not been allocated and we have to
+ * use a "safe" page frame to store the loaded page.
*/
- pbe->address = (unsigned long)safe_pages;
- safe_pages = safe_pages->next;
- if (last)
- last->next = pbe;
- handle->last_pbe = pbe;
- return (void *)pbe->address;
+ pbe = chain_alloc(ca, sizeof(struct pbe));
+ if (!pbe) {
+ swsusp_free();
+ return ERR_PTR(-ENOMEM);
+ }
+ pbe->orig_address = page_address(page);
+ pbe->address = safe_pages_list;
+ safe_pages_list = safe_pages_list->next;
+ pbe->next = restore_pblist;
+ restore_pblist = pbe;
+ return pbe->address;
}
/**
int snapshot_write_next(struct snapshot_handle *handle, size_t count)
{
+ static struct chain_allocator ca;
int error = 0;
+ /* Check if we have already loaded the entire image */
if (handle->prev && handle->cur > nr_meta_pages + nr_copy_pages)
return 0;
- if (!buffer) {
- /* This makes the buffer be freed by swsusp_free() */
- buffer = alloc_image_page(GFP_ATOMIC, PG_ANY);
+
+ if (handle->offset == 0) {
+ if (!buffer)
+ /* This makes the buffer be freed by swsusp_free() */
+ buffer = get_image_page(GFP_ATOMIC, PG_ANY);
+
if (!buffer)
return -ENOMEM;
- }
- if (!handle->offset)
+
handle->buffer = buffer;
+ }
handle->sync_read = 1;
if (handle->prev < handle->cur) {
- if (!handle->prev) {
- error = load_header(handle,
- (struct swsusp_info *)buffer);
+ if (handle->prev == 0) {
+ error = load_header(buffer);
+ if (error)
+ return error;
+
+ error = memory_bm_create(©_bm, GFP_ATOMIC, PG_ANY);
if (error)
return error;
+
} else if (handle->prev <= nr_meta_pages) {
- handle->pbe = unpack_orig_addresses(buffer,
- handle->pbe);
- if (!handle->pbe) {
- error = prepare_image(handle);
+ error = unpack_orig_pfns(buffer, ©_bm);
+ if (error)
+ return error;
+
+ if (handle->prev == nr_meta_pages) {
+ error = prepare_image(&orig_bm, ©_bm);
if (error)
return error;
- handle->pbe = restore_pblist;
- handle->last_pbe = NULL;
- handle->buffer = get_buffer(handle);
+
+ chain_init(&ca, GFP_ATOMIC, PG_SAFE);
+ memory_bm_position_reset(&orig_bm);
+ restore_pblist = NULL;
+ handle->buffer = get_buffer(&orig_bm, &ca);
handle->sync_read = 0;
+ if (IS_ERR(handle->buffer))
+ return PTR_ERR(handle->buffer);
}
} else {
- handle->pbe = handle->pbe->next;
- handle->buffer = get_buffer(handle);
- handle->sync_read = 0;
+ copy_last_highmem_page();
+ handle->buffer = get_buffer(&orig_bm, &ca);
+ if (IS_ERR(handle->buffer))
+ return PTR_ERR(handle->buffer);
+ if (handle->buffer != buffer)
+ handle->sync_read = 0;
}
handle->prev = handle->cur;
}
return count;
}
+/**
+ * snapshot_write_finalize - must be called after the last call to
+ * snapshot_write_next() in case the last page in the image happens
+ * to be a highmem page and its contents should be stored in the
+ * highmem. Additionally, it releases the memory that will not be
+ * used any more.
+ */
+
+void snapshot_write_finalize(struct snapshot_handle *handle)
+{
+ copy_last_highmem_page();
+ /* Free only if we have loaded the image entirely */
+ if (handle->prev && handle->cur > nr_meta_pages + nr_copy_pages) {
+ memory_bm_free(&orig_bm, PG_UNSAFE_CLEAR);
+ free_highmem_data();
+ }
+}
+
int snapshot_image_loaded(struct snapshot_handle *handle)
{
- return !(!handle->pbe || handle->pbe->next || !nr_copy_pages ||
- handle->cur <= nr_meta_pages + nr_copy_pages);
+ return !(!nr_copy_pages || !last_highmem_page_copied() ||
+ handle->cur <= nr_meta_pages + nr_copy_pages);
+}
+
+#ifdef CONFIG_HIGHMEM
+/* Assumes that @buf is ready and points to a "safe" page */
+static inline void
+swap_two_pages_data(struct page *p1, struct page *p2, void *buf)
+{
+ void *kaddr1, *kaddr2;
+
+ kaddr1 = kmap_atomic(p1, KM_USER0);
+ kaddr2 = kmap_atomic(p2, KM_USER1);
+ memcpy(buf, kaddr1, PAGE_SIZE);
+ memcpy(kaddr1, kaddr2, PAGE_SIZE);
+ memcpy(kaddr2, buf, PAGE_SIZE);
+ kunmap_atomic(kaddr1, KM_USER0);
+ kunmap_atomic(kaddr2, KM_USER1);
+}
+
+/**
+ * restore_highmem - for each highmem page that was allocated before
+ * the suspend and included in the suspend image, and also has been
+ * allocated by the "resume" kernel swap its current (ie. "before
+ * resume") contents with the previous (ie. "before suspend") one.
+ *
+ * If the resume eventually fails, we can call this function once
+ * again and restore the "before resume" highmem state.
+ */
+
+int restore_highmem(void)
+{
+ struct highmem_pbe *pbe = highmem_pblist;
+ void *buf;
+
+ if (!pbe)
+ return 0;
+
+ buf = get_image_page(GFP_ATOMIC, PG_SAFE);
+ if (!buf)
+ return -ENOMEM;
+
+ while (pbe) {
+ swap_two_pages_data(pbe->copy_page, pbe->orig_page, buf);
+ pbe = pbe->next;
+ }
+ free_image_page(buf, PG_UNSAFE_CLEAR);
+ return 0;
}
+#endif /* CONFIG_HIGHMEM */
git://ftp.safe.ca / safe / jmp / linux-2.6 / blobdiff