X-Git-Url: http://ftp.safe.ca/?a=blobdiff_plain;f=kernel%2Fpower%2Fsnapshot.c;h=33e2e4a819f97014a6e70e5f06d52e246ba783b3;hb=ee99c71c59f897436ec65debb99372b3146f9985;hp=7f511d89c6677701a7baa22b1871bf378e251f6c;hpb=ce4ab0012b32c1a4a1d6e934aeb73bf3151c48d9;p=safe%2Fjmp%2Flinux-2.6 diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c index 7f511d8..33e2e4a 100644 --- a/kernel/power/snapshot.c +++ b/kernel/power/snapshot.c @@ -1,30 +1,31 @@ /* * 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 + * Copyright (C) 2006 Rafael J. Wysocki * - * This file is released under the GPLv2, and is based on swsusp.c. + * This file is released under the GPLv2. * */ - #include #include #include #include -#include #include #include #include #include #include #include +#include #include #include #include #include +#include #include #include @@ -34,585 +35,1203 @@ #include "power.h" -struct pbe *pagedir_nosave; -static unsigned int nr_copy_pages; -static unsigned int nr_meta_pages; -static unsigned long *buffer; +static int swsusp_page_is_free(struct page *); +static void swsusp_set_page_forbidden(struct page *); +static void swsusp_unset_page_forbidden(struct page *); -struct arch_saveable_page { - unsigned long start; - unsigned long end; - char *data; - struct arch_saveable_page *next; -}; -static struct arch_saveable_page *arch_pages; +/* 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; + +/* 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. The unsafe pages have PageNosaveFree set + * and we count them using unsafe_pages. + * + * Each allocated image page is marked as PageNosave and PageNosaveFree + * so that swsusp_free() can release it. + */ -int swsusp_add_arch_pages(unsigned long start, unsigned long end) +#define PG_ANY 0 +#define PG_SAFE 1 +#define PG_UNSAFE_CLEAR 1 +#define PG_UNSAFE_KEEP 0 + +static unsigned int allocated_unsafe_pages; + +static void *get_image_page(gfp_t gfp_mask, int safe_needed) { - struct arch_saveable_page *tmp; + void *res; - while (start < end) { - tmp = kzalloc(sizeof(struct arch_saveable_page), GFP_KERNEL); - if (!tmp) - return -ENOMEM; - tmp->start = start; - tmp->end = ((start >> PAGE_SHIFT) + 1) << PAGE_SHIFT; - if (tmp->end > end) - tmp->end = end; - tmp->next = arch_pages; - start = tmp->end; - arch_pages = tmp; + res = (void *)get_zeroed_page(gfp_mask); + if (safe_needed) + while (res && swsusp_page_is_free(virt_to_page(res))) { + /* The page is unsafe, mark it for swsusp_free() */ + swsusp_set_page_forbidden(virt_to_page(res)); + allocated_unsafe_pages++; + res = (void *)get_zeroed_page(gfp_mask); + } + if (res) { + swsusp_set_page_forbidden(virt_to_page(res)); + swsusp_set_page_free(virt_to_page(res)); } - return 0; + return res; +} + +unsigned long get_safe_page(gfp_t gfp_mask) +{ + return (unsigned long)get_image_page(gfp_mask, PG_SAFE); } -static unsigned int count_arch_pages(void) +static struct page *alloc_image_page(gfp_t gfp_mask) { - unsigned int count = 0; - struct arch_saveable_page *tmp = arch_pages; - while (tmp) { - count++; - tmp = tmp->next; + struct page *page; + + page = alloc_page(gfp_mask); + if (page) { + swsusp_set_page_forbidden(page); + swsusp_set_page_free(page); } - return count; + return page; } -static int save_arch_mem(void) +/** + * free_image_page - free page represented by @addr, allocated with + * get_image_page (page flags set by it must be cleared) + */ + +static inline void free_image_page(void *addr, int clear_nosave_free) { - char *kaddr; - struct arch_saveable_page *tmp = arch_pages; - int offset; + struct page *page; - pr_debug("swsusp: Saving arch specific memory"); - while (tmp) { - tmp->data = (char *)__get_free_page(GFP_ATOMIC); - if (!tmp->data) - return -ENOMEM; - offset = tmp->start - (tmp->start & PAGE_MASK); - /* arch pages might haven't a 'struct page' */ - kaddr = kmap_atomic_pfn(tmp->start >> PAGE_SHIFT, KM_USER0); - memcpy(tmp->data + offset, kaddr + offset, - tmp->end - tmp->start); - kunmap_atomic(kaddr, KM_USER0); + BUG_ON(!virt_addr_valid(addr)); - tmp = tmp->next; - } - return 0; + page = virt_to_page(addr); + + swsusp_unset_page_forbidden(page); + if (clear_nosave_free) + swsusp_unset_page_free(page); + + __free_page(page); } -static int restore_arch_mem(void) +/* struct linked_page is used to build chains of pages */ + +#define LINKED_PAGE_DATA_SIZE (PAGE_SIZE - sizeof(void *)) + +struct linked_page { + struct linked_page *next; + char data[LINKED_PAGE_DATA_SIZE]; +} __attribute__((packed)); + +static inline void +free_list_of_pages(struct linked_page *list, int clear_page_nosave) { - char *kaddr; - struct arch_saveable_page *tmp = arch_pages; - int offset; + while (list) { + struct linked_page *lp = list->next; - while (tmp) { - if (!tmp->data) - continue; - offset = tmp->start - (tmp->start & PAGE_MASK); - kaddr = kmap_atomic_pfn(tmp->start >> PAGE_SHIFT, KM_USER0); - memcpy(kaddr + offset, tmp->data + offset, - tmp->end - tmp->start); - kunmap_atomic(kaddr, KM_USER0); - free_page((long)tmp->data); - tmp->data = NULL; - tmp = tmp->next; + free_image_page(list, clear_page_nosave); + list = lp; } - return 0; } -#ifdef CONFIG_HIGHMEM -unsigned int count_highmem_pages(void) +/** + * struct chain_allocator is used for allocating small objects out of + * a linked list of pages called 'the chain'. + * + * The chain grows each time when there is no room for a new object in + * the current page. The allocated objects cannot be freed individually. + * It is only possible to free them all at once, by freeing the entire + * chain. + * + * NOTE: The chain allocator may be inefficient if the allocated objects + * are not much smaller than PAGE_SIZE. + */ + +struct chain_allocator { + struct linked_page *chain; /* the chain */ + unsigned int used_space; /* total size of objects allocated out + * of the current page + */ + gfp_t gfp_mask; /* mask for allocating pages */ + int safe_needed; /* if set, only "safe" pages are allocated */ +}; + +static void +chain_init(struct chain_allocator *ca, gfp_t gfp_mask, int safe_needed) { - struct zone *zone; - unsigned long zone_pfn; - unsigned int n = 0; + ca->chain = NULL; + ca->used_space = LINKED_PAGE_DATA_SIZE; + ca->gfp_mask = gfp_mask; + ca->safe_needed = safe_needed; +} - 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; +static void *chain_alloc(struct chain_allocator *ca, unsigned int size) +{ + void *ret; + + if (LINKED_PAGE_DATA_SIZE - ca->used_space < size) { + struct linked_page *lp; + + lp = get_image_page(ca->gfp_mask, ca->safe_needed); + if (!lp) + return NULL; + + lp->next = ca->chain; + ca->chain = lp; + ca->used_space = 0; + } + ret = ca->chain->data + ca->used_space; + ca->used_space += size; + return ret; } -struct highmem_page { - char *data; - struct page *page; - struct highmem_page *next; +/** + * Data types related to memory bitmaps. + * + * Memory bitmap is a structure consiting of many linked lists of + * 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 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, + * and a pointer to the list of pages used for allocating all of the + * zone bitmap objects and bitmap block objects. + * + * NOTE: It has to be possible to lay out the bitmap in memory + * using only allocations of order 0. Additionally, the bitmap is + * designed to work with arbitrary number of zones (this is over the + * top for now, but let's avoid making unnecessary assumptions ;-). + * + * struct zone_bitmap contains a pointer to a list of bitmap block + * objects and a pointer to the bitmap block object that has been + * most recently used for setting bits. Additionally, it contains the + * 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 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_BITS_PER_BLOCK (PAGE_SIZE << 3) + +struct bm_block { + 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 long *data; /* bitmap representing pages */ }; -static struct highmem_page *highmem_copy; +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 */ -static int save_highmem_zone(struct zone *zone) +struct bm_position { + struct bm_block *block; + int bit; +}; + +struct memory_bitmap { + 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 + */ + struct bm_position cur; /* most recently used bit position */ +}; + +/* Functions that operate on memory bitmaps */ + +static void memory_bm_position_reset(struct memory_bitmap *bm) { - 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; + bm->cur.block = list_entry(bm->blocks.next, struct bm_block, hook); + bm->cur.bit = 0; +} - 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); +static void memory_bm_free(struct memory_bitmap *bm, int clear_nosave_free); + +/** + * create_bm_block_list - create a list of block bitmap objects + * @nr_blocks - number of blocks to allocate + * @list - list to put the allocated blocks into + * @ca - chain allocator to be used for allocating memory + */ +static int create_bm_block_list(unsigned long pages, + struct list_head *list, + struct chain_allocator *ca) +{ + 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 -ENOMEM; - } - kaddr = kmap_atomic(page, KM_USER0); - memcpy(save->data, kaddr, PAGE_SIZE); - kunmap_atomic(kaddr, KM_USER0); - highmem_copy = save; + list_add(&bb->hook, list); } + return 0; } -int save_highmem(void) +struct mem_extent { + struct list_head hook; + unsigned long start; + unsigned long end; +}; + +/** + * 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); + } +} + +/** + * 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 *zone; - int res = 0; - 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; + INIT_LIST_HEAD(list); + + 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; + + /* 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); + } } - printk("\n"); + return 0; } -int restore_highmem(void) +/** + * 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) { - printk("swsusp: Restoring Highmem\n"); - while (highmem_copy) { - struct highmem_page *save = highmem_copy; - void *kaddr; - highmem_copy = save->next; + struct chain_allocator ca; + struct list_head mem_extents; + struct mem_extent *ext; + int error; + + chain_init(&ca, gfp_mask, safe_needed); + INIT_LIST_HEAD(&bm->blocks); - 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); + error = create_mem_extents(&mem_extents, gfp_mask); + if (error) + return error; + + list_for_each_entry(ext, &mem_extents, hook) { + struct bm_block *bb; + unsigned long pfn = ext->start; + unsigned long pages = ext->end - ext->start; + + bb = list_entry(bm->blocks.prev, struct bm_block, hook); + + error = create_bm_block_list(pages, bm->blocks.prev, &ca); + if (error) + goto Error; + + 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 (pages >= BM_BITS_PER_BLOCK) { + pfn += BM_BITS_PER_BLOCK; + pages -= BM_BITS_PER_BLOCK; + } else { + /* This is executed only once in the loop */ + pfn += pages; + } + bb->end_pfn = pfn; + } } + + bm->p_list = ca.chain; + memory_bm_position_reset(bm); + Exit: + free_mem_extents(&mem_extents); + return error; + + Error: + bm->p_list = ca.chain; + memory_bm_free(bm, PG_UNSAFE_CLEAR); + 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 bm_block *bb; + + list_for_each_entry(bb, &bm->blocks, hook) + if (bb->data) + free_image_page(bb->data, clear_nosave_free); + + free_list_of_pages(bm->p_list, clear_nosave_free); + + INIT_LIST_HEAD(&bm->blocks); +} + +/** + * 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. + */ +static int memory_bm_find_bit(struct memory_bitmap *bm, unsigned long pfn, + void **addr, unsigned int *bit_nr) +{ + struct bm_block *bb; + + /* + * 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) + list_for_each_entry_continue_reverse(bb, &bm->blocks, hook) + if (pfn >= bb->start_pfn) + break; + + 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; + bm->cur.bit = pfn + 1; + *bit_nr = pfn; + *addr = bb->data; return 0; } -#else -static unsigned int count_highmem_pages(void) {return 0;} -static int save_highmem(void) {return 0;} -static int restore_highmem(void) {return 0;} -#endif -unsigned int count_special_pages(void) +static void memory_bm_set_bit(struct memory_bitmap *bm, unsigned long pfn) { - return count_arch_pages() + count_highmem_pages(); + void *addr; + unsigned int bit; + int error; + + error = memory_bm_find_bit(bm, pfn, &addr, &bit); + BUG_ON(error); + set_bit(bit, addr); } -int save_special_mem(void) +static int mem_bm_set_bit_check(struct memory_bitmap *bm, unsigned long pfn) { - int ret; - ret = save_arch_mem(); - if (!ret) - ret = save_highmem(); - return ret; + void *addr; + unsigned int bit; + int error; + + error = memory_bm_find_bit(bm, pfn, &addr, &bit); + if (!error) + set_bit(bit, addr); + return error; } -int restore_special_mem(void) +static void memory_bm_clear_bit(struct memory_bitmap *bm, unsigned long pfn) { - int ret; - ret = restore_arch_mem(); - if (!ret) - ret = restore_highmem(); - return ret; + void *addr; + unsigned int bit; + int error; + + error = memory_bm_find_bit(bm, pfn, &addr, &bit); + BUG_ON(error); + clear_bit(bit, addr); } -static int pfn_is_nosave(unsigned long pfn) +static int memory_bm_test_bit(struct memory_bitmap *bm, unsigned long pfn) { - 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); + void *addr; + unsigned int bit; + int error; + + 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); } /** - * saveable - Determine whether a page should be cloned or not. - * @pfn: The page + * memory_bm_next_pfn - find the pfn that corresponds to the next set bit + * in the bitmap @bm. If the pfn cannot be found, BM_END_OF_MAP is + * returned. * - * We save a page if it's Reserved, and not in the range of pages - * statically defined as 'unsaveable', or if it isn't reserved, and - * isn't part of a free chunk of pages. + * It is required to run memory_bm_position_reset() before the first call to + * this function. */ -static int saveable(struct zone *zone, unsigned long *zone_pfn) +static unsigned long memory_bm_next_pfn(struct memory_bitmap *bm) { - unsigned long pfn = *zone_pfn + zone->zone_start_pfn; - struct page *page; + struct bm_block *bb; + int bit; + + bb = bm->cur.block; + do { + 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.bit = bit + 1; + return bb->start_pfn + bit; +} - if (!pfn_valid(pfn)) - return 0; +/** + * This structure represents a range of page frames the contents of which + * should not be saved during the suspend. + */ - page = pfn_to_page(pfn); - if (PageNosave(page)) - return 0; - if (PageReserved(page) && pfn_is_nosave(pfn)) - return 0; - if (PageNosaveFree(page)) - return 0; +struct nosave_region { + struct list_head list; + unsigned long start_pfn; + unsigned long end_pfn; +}; - return 1; +static LIST_HEAD(nosave_regions); + +/** + * 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) + */ + +void __init +__register_nosave_region(unsigned long start_pfn, unsigned long end_pfn, + int use_kmalloc) +{ + 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_low(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); } -unsigned int count_data_pages(void) +/* + * 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; + +/* Set bits in this map correspond to free page frames. */ +static struct memory_bitmap *free_pages_map; + +/* + * Each page frame allocated for creating the image is marked by setting the + * corresponding bits in forbidden_pages_map and free_pages_map simultaneously + */ + +void swsusp_set_page_free(struct page *page) { - struct zone *zone; - unsigned long zone_pfn; - unsigned int n = 0; + if (free_pages_map) + memory_bm_set_bit(free_pages_map, page_to_pfn(page)); +} - for_each_zone (zone) { - if (is_highmem(zone)) - continue; - mark_free_pages(zone); - for (zone_pfn = 0; zone_pfn < zone->spanned_pages; ++zone_pfn) - n += saveable(zone, &zone_pfn); - } - return n; +static int swsusp_page_is_free(struct page *page) +{ + return free_pages_map ? + memory_bm_test_bit(free_pages_map, page_to_pfn(page)) : 0; } -static void copy_data_pages(struct pbe *pblist) +void swsusp_unset_page_free(struct page *page) { - struct zone *zone; - unsigned long zone_pfn; - struct pbe *pbe, *p; + if (free_pages_map) + memory_bm_clear_bit(free_pages_map, page_to_pfn(page)); +} - pbe = pblist; - for_each_zone (zone) { - if (is_highmem(zone)) - continue; - mark_free_pages(zone); - /* This is necessary for swsusp_free() */ - for_each_pb_page (p, pblist) - SetPageNosaveFree(virt_to_page(p)); - for_each_pbe (p, pblist) - SetPageNosaveFree(virt_to_page(p->address)); - for (zone_pfn = 0; zone_pfn < zone->spanned_pages; ++zone_pfn) { - if (saveable(zone, &zone_pfn)) { - struct page *page; - page = pfn_to_page(zone_pfn + zone->zone_start_pfn); - BUG_ON(!pbe); - pbe->orig_address = (unsigned long)page_address(page); - /* copy_page is not usable for copying task structs. */ - memcpy((void *)pbe->address, (void *)pbe->orig_address, PAGE_SIZE); - pbe = pbe->next; +static void swsusp_set_page_forbidden(struct page *page) +{ + if (forbidden_pages_map) + memory_bm_set_bit(forbidden_pages_map, page_to_pfn(page)); +} + +int swsusp_page_is_forbidden(struct page *page) +{ + return forbidden_pages_map ? + memory_bm_test_bit(forbidden_pages_map, page_to_pfn(page)) : 0; +} + +static void swsusp_unset_page_forbidden(struct page *page) +{ + if (forbidden_pages_map) + memory_bm_clear_bit(forbidden_pages_map, page_to_pfn(page)); +} + +/** + * mark_nosave_pages - set bits corresponding to the page frames the + * contents of which should not be saved in a given bitmap. + */ + +static void mark_nosave_pages(struct memory_bitmap *bm) +{ + 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); + + 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); } - } } - BUG_ON(pbe); } +/** + * 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. + */ + +int create_basic_memory_bitmaps(void) +{ + 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; + + 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; +} /** - * free_pagedir - free pages allocated with alloc_pagedir() + * 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 void free_pagedir(struct pbe *pblist, int clear_nosave_free) +void free_basic_memory_bitmaps(void) { - struct pbe *pbe; + struct memory_bitmap *bm1, *bm2; - while (pblist) { - pbe = (pblist + PB_PAGE_SKIP)->next; - ClearPageNosave(virt_to_page(pblist)); - if (clear_nosave_free) - ClearPageNosaveFree(virt_to_page(pblist)); - free_page((unsigned long)pblist); - pblist = pbe; - } + BUG_ON(!(forbidden_pages_map && free_pages_map)); + + 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"); } /** - * fill_pb_page - Create a list of PBEs on a given memory 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) */ -static inline void fill_pb_page(struct pbe *pbpage) +unsigned int snapshot_additional_pages(struct zone *zone) { - struct pbe *p; + unsigned int res; - p = pbpage; - pbpage += PB_PAGE_SKIP; - do - p->next = p + 1; - while (++p < pbpage); + 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 /** - * create_pbe_list - Create a list of PBEs on top of a given chain - * of memory pages allocated with alloc_pagedir() + * count_free_highmem_pages - compute the total number of free highmem + * pages, system-wide. */ -static inline void create_pbe_list(struct pbe *pblist, unsigned int nr_pages) +static unsigned int count_free_highmem_pages(void) { - struct pbe *pbpage, *p; - unsigned int num = PBES_PER_PAGE; + struct zone *zone; + unsigned int cnt = 0; - for_each_pb_page (pbpage, pblist) { - if (num >= nr_pages) - break; + for_each_populated_zone(zone) + if (is_highmem(zone)) + cnt += zone_page_state(zone, NR_FREE_PAGES); - fill_pb_page(pbpage); - num += PBES_PER_PAGE; - } - if (pbpage) { - for (num -= PBES_PER_PAGE - 1, p = pbpage; num < nr_pages; p++, num++) - p->next = p + 1; - p->next = NULL; - } + return cnt; } /** - * On resume it is necessary to trace and eventually free the unsafe - * pages that have been allocated, because they are needed for I/O - * (on x86-64 we likely will "eat" these pages once again while - * creating the temporary page translation tables) + * saveable_highmem_page - Determine whether a highmem page should be + * included in the suspend image. + * + * 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; -struct eaten_page { - struct eaten_page *next; - char padding[PAGE_SIZE - sizeof(void *)]; -}; + 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; +} -static struct eaten_page *eaten_pages = NULL; +/** + * count_highmem_pages - compute the total number of saveable highmem + * pages. + */ -static void release_eaten_pages(void) +unsigned int count_highmem_pages(void) { - struct eaten_page *p, *q; + struct zone *zone; + unsigned int n = 0; + + for_each_zone(zone) { + unsigned long pfn, max_zone_pfn; + + if (!is_highmem(zone)) + continue; - p = eaten_pages; - while (p) { - q = p->next; - /* We don't want swsusp_free() to free this page again */ - ClearPageNosave(virt_to_page(p)); - free_page((unsigned long)p); - p = q; + 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++; } - eaten_pages = NULL; + return n; } +#else +static inline void *saveable_highmem_page(struct zone *z, unsigned long p) +{ + return NULL; +} +#endif /* CONFIG_HIGHMEM */ /** - * @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 - * which had been used before suspend. - * - * The unsafe pages are marked with the PG_nosave_free flag + * saveable_page - Determine whether a non-highmem page should be included + * in the suspend image. * - * Allocated but unusable (ie eaten) memory pages should be marked - * so that swsusp_free() can release them + * 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; + + 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; -static inline void *alloc_image_page(gfp_t gfp_mask, int safe_needed) + 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. + */ + +unsigned int count_data_pages(void) { - void *res; + struct zone *zone; + unsigned long pfn, max_zone_pfn; + unsigned int n = 0; - if (safe_needed) - do { - res = (void *)get_zeroed_page(gfp_mask); - if (res && PageNosaveFree(virt_to_page(res))) { - /* This is for swsusp_free() */ - SetPageNosave(virt_to_page(res)); - ((struct eaten_page *)res)->next = eaten_pages; - eaten_pages = res; - } - } while (res && PageNosaveFree(virt_to_page(res))); - else - res = (void *)get_zeroed_page(gfp_mask); - if (res) { - SetPageNosave(virt_to_page(res)); - SetPageNosaveFree(virt_to_page(res)); + 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++) + if (saveable_page(zone, pfn)) + n++; } - return res; + return n; } -unsigned long get_safe_page(gfp_t gfp_mask) +/* 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) { - return (unsigned long)alloc_image_page(gfp_mask, 1); + int n; + + for (n = PAGE_SIZE / sizeof(long); n; n--) + *dst++ = *src++; } + /** - * 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. - * - * On each page we set up a list of struct_pbe elements. + * 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); + } +} + -struct pbe *alloc_pagedir(unsigned int nr_pages, gfp_t gfp_mask, int safe_needed) +#ifdef CONFIG_HIGHMEM +static inline struct page * +page_is_saveable(struct zone *zone, unsigned long pfn) { - unsigned int num; - struct pbe *pblist, *pbe; + return is_highmem(zone) ? + saveable_highmem_page(zone, pfn) : saveable_page(zone, pfn); +} - if (!nr_pages) - return NULL; +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); + } + } +} +#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_zone(zone) { + unsigned long max_zone_pfn; - pblist = alloc_image_page(gfp_mask, safe_needed); - /* FIXME: rewrite this ugly loop */ - for (pbe = pblist, num = PBES_PER_PAGE; pbe && num < nr_pages; - pbe = pbe->next, num += PBES_PER_PAGE) { - pbe += PB_PAGE_SKIP; - pbe->next = alloc_image_page(gfp_mask, safe_needed); + 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); } - if (!pbe) { /* get_zeroed_page() failed */ - free_pagedir(pblist, 1); - pblist = NULL; - } else - create_pbe_list(pblist, nr_pages); - return pblist; } +/* 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; + /** - * 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 zone_pfn; + unsigned long pfn, max_zone_pfn; for_each_zone(zone) { - for (zone_pfn = 0; zone_pfn < zone->spanned_pages; ++zone_pfn) - if (pfn_valid(zone_pfn + zone->zone_start_pfn)) { - struct page *page; - page = pfn_to_page(zone_pfn + zone->zone_start_pfn); - if (PageNosave(page) && PageNosaveFree(page)) { - ClearPageNosave(page); - ClearPageNosaveFree(page); - free_page((long) page_address(page)); + 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 (swsusp_page_is_forbidden(page) && + swsusp_page_is_free(page)) { + swsusp_unset_page_forbidden(page); + swsusp_unset_page_free(page); + __free_page(page); } } } nr_copy_pages = 0; nr_meta_pages = 0; - pagedir_nosave = NULL; + restore_pblist = NULL; buffer = NULL; } +#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(); + + 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 enough free memory to snapshot. - * - * Returns TRUE or FALSE after checking the number of available - * free pages. + * enough_free_mem - Make sure we have enough free memory for the + * snapshot image. */ -static int enough_free_mem(unsigned int nr_pages) +static int enough_free_mem(unsigned int nr_pages, unsigned int nr_highmem) { struct zone *zone; - unsigned int n = 0; + unsigned int free = 0, meta = 0; - for_each_zone (zone) + for_each_zone(zone) { + meta += snapshot_additional_pages(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); + free += zone_page_state(zone, NR_FREE_PAGES); + } + + nr_pages += count_pages_for_highmem(nr_highmem); + pr_debug("PM: Normal pages needed: %u + %u + %u, available pages: %u\n", + nr_pages, PAGES_FOR_IO, meta, free); + + return free > nr_pages + PAGES_FOR_IO + meta; } -static int alloc_data_pages(struct pbe *pblist, gfp_t gfp_mask, int safe_needed) +#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) { - struct pbe *p; + buffer = get_image_page(GFP_ATOMIC | __GFP_COLD, safe_needed); + return buffer ? 0 : -ENOMEM; +} - for_each_pbe (p, pblist) { - p->address = (unsigned long)alloc_image_page(gfp_mask, safe_needed); - if (!p->address) - return -ENOMEM; +/** + * 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 inline unsigned int +alloc_highmem_image_pages(struct memory_bitmap *bm, unsigned int nr_highmem) +{ + unsigned int to_alloc = count_free_highmem_pages(); + + 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 0; + return nr_highmem; } +#else +static inline int get_highmem_buffer(int safe_needed) { return 0; } + +static inline unsigned int +alloc_highmem_image_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 struct pbe *swsusp_alloc(unsigned int nr_pages) +static int +swsusp_alloc(struct memory_bitmap *orig_bm, struct memory_bitmap *copy_bm, + unsigned int nr_pages, unsigned int nr_highmem) { - struct pbe *pblist; + int error; - if (!(pblist = alloc_pagedir(nr_pages, GFP_ATOMIC | __GFP_COLD, 0))) { - printk(KERN_ERR "suspend: Allocating pagedir failed.\n"); - return NULL; + 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; + + if (nr_highmem > 0) { + error = get_highmem_buffer(PG_ANY); + if (error) + goto Free; + + nr_pages += alloc_highmem_image_pages(copy_bm, nr_highmem); } + while (nr_pages-- > 0) { + struct page *page = alloc_image_page(GFP_ATOMIC | __GFP_COLD); - if (alloc_data_pages(pblist, GFP_ATOMIC | __GFP_COLD, 0)) { - printk(KERN_ERR "suspend: Allocating image pages failed.\n"); - swsusp_free(); - return NULL; + if (!page) + goto Free; + + memory_bm_set_bit(copy_bm, page_to_pfn(page)); } + return 0; - return pblist; + Free: + swsusp_free(); + return -ENOMEM; } +/* Memory bitmap used for marking saveable pages (during suspend) or the + * suspend image pages (during resume) + */ +static struct memory_bitmap orig_bm; +/* Memory bitmap used on suspend for marking allocated pages that will contain + * the copies of saveable pages. During resume it is initially used for + * marking the suspend image pages, but then its set bits are duplicated in + * @orig_bm and it is released. Next, on systems with high memory, it may be + * used for marking "safe" highmem pages, but it has to be reinitialized for + * this purpose. + */ +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); + nr_highmem = count_highmem_pages(); + printk(KERN_INFO "PM: Need to copy %u pages\n", nr_pages + nr_highmem); - 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()); - - 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; } - pagedir_nosave = swsusp_alloc(nr_pages); - if (!pagedir_nosave) + 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(); - copy_data_pages(pagedir_nosave); + drain_local_pages(NULL); + copy_data_pages(©_bm, &orig_bm); /* * End of critical section. From now on, we can write to memory, @@ -620,43 +1239,71 @@ asmlinkage int swsusp_save(void) * 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_orig_addresses - the .orig_address fields of the PBEs from the - * list starting at @pbe 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 struct pbe *pack_orig_addresses(unsigned long *buf, struct pbe *pbe) +static inline void +pack_pfns(unsigned long *buf, struct memory_bitmap *bm) { int j; - for (j = 0; j < PAGE_SIZE / sizeof(long) && pbe; j++) { - buf[j] = pbe->orig_address; - pbe = pbe->next; + for (j = 0; j < PAGE_SIZE / sizeof(long); j++) { + buf[j] = memory_bm_next_pfn(bm); + if (unlikely(buf[j] == BM_END_OF_MAP)) + break; } - if (!pbe) - for (; j < PAGE_SIZE / sizeof(long); j++) - buf[j] = 0; - return pbe; } /** @@ -683,37 +1330,57 @@ static inline struct pbe *pack_orig_addresses(unsigned long *buf, struct pbe *pb int snapshot_read_next(struct snapshot_handle *handle, size_t count) { - if (handle->page > nr_meta_pages + nr_copy_pages) + if (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, 0); + 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; - handle->pbe = pagedir_nosave; + memory_bm_position_reset(&orig_bm); + memory_bm_position_reset(©_bm); } - if (handle->prev < handle->page) { - if (handle->page <= nr_meta_pages) { - handle->pbe = pack_orig_addresses(buffer, handle->pbe); - if (!handle->pbe) - handle->pbe = pagedir_nosave; + if (handle->prev < handle->cur) { + if (handle->cur <= nr_meta_pages) { + memset(buffer, 0, PAGE_SIZE); + pack_pfns(buffer, &orig_bm); } else { - handle->buffer = (void *)handle->pbe->address; - handle->pbe = handle->pbe->next; + struct page *page; + + 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->page; + handle->prev = handle->cur; } - handle->buf_offset = handle->page_offset; - if (handle->page_offset + count >= PAGE_SIZE) { - count = PAGE_SIZE - handle->page_offset; - handle->page_offset = 0; - handle->page++; + handle->buf_offset = handle->cur_offset; + if (handle->cur_offset + count >= PAGE_SIZE) { + count = PAGE_SIZE - handle->cur_offset; + handle->cur_offset = 0; + handle->cur++; } else { - handle->page_offset += count; + handle->cur_offset += count; } handle->offset += count; return count; @@ -725,62 +1392,58 @@ int snapshot_read_next(struct snapshot_handle *handle, size_t count) * 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 zone_pfn; - struct pbe *p; - - if (!pblist) /* a sanity check */ - return -EINVAL; + unsigned long pfn, max_zone_pfn; /* 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)); + for_each_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)) + 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); + + 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; @@ -790,19 +1453,14 @@ static int check_header(struct swsusp_info *info) * 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, 0); - if (!pblist) - return -ENOMEM; - pagedir_nosave = pblist; - handle->pbe = pblist; nr_copy_pages = info->image_pages; nr_meta_pages = info->pages - info->image_pages - 1; } @@ -810,60 +1468,374 @@ static int load_header(struct snapshot_handle *handle, } /** - * 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 pbe; + + 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; + /** - * create_image - use metadata contained in the PBE list - * pointed to by pagedir_nosave to mark the pages that will - * be overwritten in the process of restoring the system - * memory state from the image and allocate memory for - * the image avoiding these pages + * 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 int create_image(struct snapshot_handle *handle) +static unsigned int count_highmem_image_pages(struct memory_bitmap *bm) { - int error = 0; - struct pbe *p, *pblist; + unsigned long pfn; + unsigned int cnt = 0; - p = pagedir_nosave; - error = mark_unsafe_pages(p); - if (!error) { - pblist = alloc_pagedir(nr_copy_pages, GFP_ATOMIC, 1); - if (pblist) - copy_page_backup_list(pblist, p); - free_pagedir(p, 0); - if (!pblist) - error = -ENOMEM; + 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); } - if (!error) - error = alloc_data_pages(pblist, GFP_ATOMIC, 1); - if (!error) { - release_eaten_pages(); - pagedir_nosave = pblist; + return cnt; +} + +/** + * 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). + * + * NOTE: This function should not be called if there are no highmem + * image pages. + */ + +static unsigned int safe_highmem_pages; + +static struct memory_bitmap *safe_highmem_bm; + +static int +prepare_highmem_image(struct memory_bitmap *bm, unsigned int *nr_highmem_p) +{ + unsigned int to_alloc; + + 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 { - pagedir_nosave = NULL; - handle->pbe = NULL; - nr_copy_pages = 0; - nr_meta_pages = 0; + /* 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--; + } + /* 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--; } + /* 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; } /** + * 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; + struct page *page; + unsigned long pfn = memory_bm_next_pfn(bm); + + if (pfn == BM_END_OF_MAP) + return ERR_PTR(-EFAULT); + + 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. + */ + 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 = 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; +} + +/** * snapshot_write_next - used for writing the system memory snapshot. * * On the first call to it @handle should point to a zeroed @@ -887,52 +1859,141 @@ static int create_image(struct snapshot_handle *handle) int snapshot_write_next(struct snapshot_handle *handle, size_t count) { + static struct chain_allocator ca; int error = 0; - if (handle->prev && handle->page > nr_meta_pages + nr_copy_pages) + /* 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, 0); + + 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; - if (handle->prev < handle->page) { - if (!handle->prev) { - error = load_header(handle, (struct swsusp_info *)buffer); + } + handle->sync_read = 1; + if (handle->prev < handle->cur) { + 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 = create_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 = pagedir_nosave; - handle->buffer = (void *)handle->pbe->address; + + 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 = (void *)handle->pbe->address; + 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->page; + handle->prev = handle->cur; } - handle->buf_offset = handle->page_offset; - if (handle->page_offset + count >= PAGE_SIZE) { - count = PAGE_SIZE - handle->page_offset; - handle->page_offset = 0; - handle->page++; + handle->buf_offset = handle->cur_offset; + if (handle->cur_offset + count >= PAGE_SIZE) { + count = PAGE_SIZE - handle->cur_offset; + handle->cur_offset = 0; + handle->cur++; } else { - handle->page_offset += count; + handle->cur_offset += count; } handle->offset += count; 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->page <= 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 */