4 #include <linux/spinlock.h>
5 #include <linux/linkage.h>
6 #include <linux/mmzone.h>
7 #include <linux/list.h>
8 #include <linux/memcontrol.h>
9 #include <linux/sched.h>
10 #include <linux/node.h>
12 #include <linux/atomic.h>
15 struct notifier_block;
19 #define SWAP_FLAG_PREFER 0x8000 /* set if swap priority specified */
20 #define SWAP_FLAG_PRIO_MASK 0x7fff
21 #define SWAP_FLAG_PRIO_SHIFT 0
22 #define SWAP_FLAG_DISCARD 0x10000 /* discard swap cluster after use */
24 #define SWAP_FLAGS_VALID (SWAP_FLAG_PRIO_MASK | SWAP_FLAG_PREFER | \
27 static inline int current_is_kswapd(void)
29 return current->flags & PF_KSWAPD;
33 * MAX_SWAPFILES defines the maximum number of swaptypes: things which can
34 * be swapped to. The swap type and the offset into that swap type are
35 * encoded into pte's and into pgoff_t's in the swapcache. Using five bits
36 * for the type means that the maximum number of swapcache pages is 27 bits
37 * on 32-bit-pgoff_t architectures. And that assumes that the architecture packs
38 * the type/offset into the pte as 5/27 as well.
40 #define MAX_SWAPFILES_SHIFT 5
43 * Use some of the swap files numbers for other purposes. This
44 * is a convenient way to hook into the VM to trigger special
49 * NUMA node memory migration support
51 #ifdef CONFIG_MIGRATION
52 #define SWP_MIGRATION_NUM 2
53 #define SWP_MIGRATION_READ (MAX_SWAPFILES + SWP_HWPOISON_NUM)
54 #define SWP_MIGRATION_WRITE (MAX_SWAPFILES + SWP_HWPOISON_NUM + 1)
56 #define SWP_MIGRATION_NUM 0
60 * Handling of hardware poisoned pages with memory corruption.
62 #ifdef CONFIG_MEMORY_FAILURE
63 #define SWP_HWPOISON_NUM 1
64 #define SWP_HWPOISON MAX_SWAPFILES
66 #define SWP_HWPOISON_NUM 0
69 #define MAX_SWAPFILES \
70 ((1 << MAX_SWAPFILES_SHIFT) - SWP_MIGRATION_NUM - SWP_HWPOISON_NUM)
73 * Magic header for a swap area. The first part of the union is
74 * what the swap magic looks like for the old (limited to 128MB)
75 * swap area format, the second part of the union adds - in the
76 * old reserved area - some extra information. Note that the first
77 * kilobyte is reserved for boot loader or disk label stuff...
79 * Having the magic at the end of the PAGE_SIZE makes detecting swap
80 * areas somewhat tricky on machines that support multiple page sizes.
81 * For 2.5 we'll probably want to move the magic to just beyond the
86 char reserved[PAGE_SIZE - 10];
87 char magic[10]; /* SWAP-SPACE or SWAPSPACE2 */
90 char bootbits[1024]; /* Space for disklabel etc. */
94 unsigned char sws_uuid[16];
95 unsigned char sws_volume[16];
101 /* A swap entry has to fit into a "unsigned long", as
102 * the entry is hidden in the "index" field of the
103 * swapper address space.
110 * current->reclaim_state points to one of these when a task is running
113 struct reclaim_state {
114 unsigned long reclaimed_slab;
119 struct address_space;
121 struct writeback_control;
125 * A swap extent maps a range of a swapfile's PAGE_SIZE pages onto a range of
126 * disk blocks. A list of swap extents maps the entire swapfile. (Where the
127 * term `swapfile' refers to either a blockdevice or an IS_REG file. Apart
128 * from setup, they're handled identically.
130 * We always assume that blocks are of size PAGE_SIZE.
133 struct list_head list;
136 sector_t start_block;
140 * Max bad pages in the new format..
142 #define __swapoffset(x) ((unsigned long)&((union swap_header *)0)->x)
143 #define MAX_SWAP_BADPAGES \
144 ((__swapoffset(magic.magic) - __swapoffset(info.badpages)) / sizeof(int))
147 SWP_USED = (1 << 0), /* is slot in swap_info[] used? */
148 SWP_WRITEOK = (1 << 1), /* ok to write to this swap? */
149 SWP_DISCARDABLE = (1 << 2), /* swapon+blkdev support discard */
150 SWP_DISCARDING = (1 << 3), /* now discarding a free cluster */
151 SWP_SOLIDSTATE = (1 << 4), /* blkdev seeks are cheap */
152 SWP_CONTINUED = (1 << 5), /* swap_map has count continuation */
153 SWP_BLKDEV = (1 << 6), /* its a block device */
154 SWP_FILE = (1 << 7), /* set after swap_activate success */
155 /* add others here before... */
156 SWP_SCANNING = (1 << 8), /* refcount in scan_swap_map */
159 #define SWAP_CLUSTER_MAX 32
160 #define COMPACT_CLUSTER_MAX SWAP_CLUSTER_MAX
163 * Ratio between the present memory in the zone and the "gap" that
164 * we're allowing kswapd to shrink in addition to the per-zone high
165 * wmark, even for zones that already have the high wmark satisfied,
166 * in order to provide better per-zone lru behavior. We are ok to
167 * spend not more than 1% of the memory for this zone balancing "gap".
169 #define KSWAPD_ZONE_BALANCE_GAP_RATIO 100
171 #define SWAP_MAP_MAX 0x3e /* Max duplication count, in first swap_map */
172 #define SWAP_MAP_BAD 0x3f /* Note pageblock is bad, in first swap_map */
173 #define SWAP_HAS_CACHE 0x40 /* Flag page is cached, in first swap_map */
174 #define SWAP_CONT_MAX 0x7f /* Max count, in each swap_map continuation */
175 #define COUNT_CONTINUED 0x80 /* See swap_map continuation for full count */
176 #define SWAP_MAP_SHMEM 0xbf /* Owned by shmem/tmpfs, in first swap_map */
179 * The in-memory structure used to track swap areas.
181 struct swap_info_struct {
182 unsigned long flags; /* SWP_USED etc: see above */
183 signed short prio; /* swap priority of this type */
184 signed char type; /* strange name for an index */
185 signed char next; /* next type on the swap list */
186 unsigned int max; /* extent of the swap_map */
187 unsigned char *swap_map; /* vmalloc'ed array of usage counts */
188 unsigned int lowest_bit; /* index of first free in swap_map */
189 unsigned int highest_bit; /* index of last free in swap_map */
190 unsigned int pages; /* total of usable pages of swap */
191 unsigned int inuse_pages; /* number of those currently in use */
192 unsigned int cluster_next; /* likely index for next allocation */
193 unsigned int cluster_nr; /* countdown to next cluster search */
194 unsigned int lowest_alloc; /* while preparing discard cluster */
195 unsigned int highest_alloc; /* while preparing discard cluster */
196 struct swap_extent *curr_swap_extent;
197 struct swap_extent first_swap_extent;
198 struct block_device *bdev; /* swap device or bdev of swap file */
199 struct file *swap_file; /* seldom referenced */
200 unsigned int old_block_size; /* seldom referenced */
201 #ifdef CONFIG_FRONTSWAP
202 unsigned long *frontswap_map; /* frontswap in-use, one bit per page */
203 atomic_t frontswap_pages; /* frontswap pages in-use counter */
208 int head; /* head of priority-ordered swapfile list */
209 int next; /* swapfile to be used next */
212 /* Swap 50% full? Release swapcache more aggressively.. */
213 #define vm_swap_full() (nr_swap_pages*2 < total_swap_pages)
215 /* linux/mm/page_alloc.c */
216 extern unsigned long totalram_pages;
217 extern unsigned long totalreserve_pages;
218 extern unsigned long dirty_balance_reserve;
219 extern unsigned int nr_free_buffer_pages(void);
220 extern unsigned int nr_free_pagecache_pages(void);
222 /* Definition of global_page_state not available yet */
223 #define nr_free_pages() global_page_state(NR_FREE_PAGES)
226 /* linux/mm/swap.c */
227 extern void __lru_cache_add(struct page *, enum lru_list lru);
228 extern void lru_cache_add_lru(struct page *, enum lru_list lru);
229 extern void lru_add_page_tail(struct page *page, struct page *page_tail,
230 struct lruvec *lruvec);
231 extern void activate_page(struct page *);
232 extern void mark_page_accessed(struct page *);
233 extern void lru_add_drain(void);
234 extern void lru_add_drain_cpu(int cpu);
235 extern int lru_add_drain_all(void);
236 extern void rotate_reclaimable_page(struct page *page);
237 extern void deactivate_page(struct page *page);
238 extern void swap_setup(void);
240 extern void add_page_to_unevictable_list(struct page *page);
243 * lru_cache_add: add a page to the page lists
244 * @page: the page to add
246 static inline void lru_cache_add_anon(struct page *page)
248 __lru_cache_add(page, LRU_INACTIVE_ANON);
251 static inline void lru_cache_add_file(struct page *page)
253 __lru_cache_add(page, LRU_INACTIVE_FILE);
256 /* linux/mm/vmscan.c */
257 extern unsigned long try_to_free_pages(struct zonelist *zonelist, int order,
258 gfp_t gfp_mask, nodemask_t *mask);
259 extern int __isolate_lru_page(struct page *page, isolate_mode_t mode);
260 extern unsigned long try_to_free_mem_cgroup_pages(struct mem_cgroup *mem,
261 gfp_t gfp_mask, bool noswap);
262 extern unsigned long mem_cgroup_shrink_node_zone(struct mem_cgroup *mem,
263 gfp_t gfp_mask, bool noswap,
265 unsigned long *nr_scanned);
266 extern unsigned long shrink_all_memory(unsigned long nr_pages);
267 extern int vm_swappiness;
268 extern int remove_mapping(struct address_space *mapping, struct page *page);
269 extern long vm_total_pages;
272 extern int zone_reclaim_mode;
273 extern int sysctl_min_unmapped_ratio;
274 extern int sysctl_min_slab_ratio;
275 extern int zone_reclaim(struct zone *, gfp_t, unsigned int);
277 #define zone_reclaim_mode 0
278 static inline int zone_reclaim(struct zone *z, gfp_t mask, unsigned int order)
284 extern int page_evictable(struct page *page, struct vm_area_struct *vma);
285 extern void check_move_unevictable_pages(struct page **, int nr_pages);
287 extern unsigned long scan_unevictable_pages;
288 extern int scan_unevictable_handler(struct ctl_table *, int,
289 void __user *, size_t *, loff_t *);
291 extern int scan_unevictable_register_node(struct node *node);
292 extern void scan_unevictable_unregister_node(struct node *node);
294 static inline int scan_unevictable_register_node(struct node *node)
298 static inline void scan_unevictable_unregister_node(struct node *node)
303 extern int kswapd_run(int nid);
304 extern void kswapd_stop(int nid);
306 extern int mem_cgroup_swappiness(struct mem_cgroup *mem);
308 static inline int mem_cgroup_swappiness(struct mem_cgroup *mem)
310 return vm_swappiness;
313 #ifdef CONFIG_MEMCG_SWAP
314 extern void mem_cgroup_uncharge_swap(swp_entry_t ent);
316 static inline void mem_cgroup_uncharge_swap(swp_entry_t ent)
321 /* linux/mm/page_io.c */
322 extern int swap_readpage(struct page *);
323 extern int swap_writepage(struct page *page, struct writeback_control *wbc);
324 extern int swap_set_page_dirty(struct page *page);
325 extern void end_swap_bio_read(struct bio *bio, int err);
327 /* linux/mm/swap_state.c */
328 extern struct address_space swapper_space;
329 #define total_swapcache_pages swapper_space.nrpages
330 extern void show_swap_cache_info(void);
331 extern int add_to_swap(struct page *);
332 extern int add_to_swap_cache(struct page *, swp_entry_t, gfp_t);
333 extern void __delete_from_swap_cache(struct page *);
334 extern void delete_from_swap_cache(struct page *);
335 extern void free_page_and_swap_cache(struct page *);
336 extern void free_pages_and_swap_cache(struct page **, int);
337 extern struct page *lookup_swap_cache(swp_entry_t);
338 extern struct page *read_swap_cache_async(swp_entry_t, gfp_t,
339 struct vm_area_struct *vma, unsigned long addr);
340 extern struct page *swapin_readahead(swp_entry_t, gfp_t,
341 struct vm_area_struct *vma, unsigned long addr);
343 /* linux/mm/swapfile.c */
344 extern long nr_swap_pages;
345 extern long total_swap_pages;
346 extern void si_swapinfo(struct sysinfo *);
347 extern swp_entry_t get_swap_page(void);
348 extern swp_entry_t get_swap_page_of_type(int);
349 extern int add_swap_count_continuation(swp_entry_t, gfp_t);
350 extern void swap_shmem_alloc(swp_entry_t);
351 extern int swap_duplicate(swp_entry_t);
352 extern int swapcache_prepare(swp_entry_t);
353 extern void swap_free(swp_entry_t);
354 extern void swapcache_free(swp_entry_t, struct page *page);
355 extern int free_swap_and_cache(swp_entry_t);
356 extern int swap_type_of(dev_t, sector_t, struct block_device **);
357 extern unsigned int count_swap_pages(int, int);
358 extern sector_t map_swap_page(struct page *, struct block_device **);
359 extern sector_t swapdev_block(int, pgoff_t);
360 extern int page_swapcount(struct page *);
361 extern struct swap_info_struct *page_swap_info(struct page *);
362 extern int reuse_swap_page(struct page *);
363 extern int try_to_free_swap(struct page *);
364 struct backing_dev_info;
368 mem_cgroup_uncharge_swapcache(struct page *page, swp_entry_t ent, bool swapout);
371 mem_cgroup_uncharge_swapcache(struct page *page, swp_entry_t ent, bool swapout)
376 #else /* CONFIG_SWAP */
378 #define nr_swap_pages 0L
379 #define total_swap_pages 0L
380 #define total_swapcache_pages 0UL
382 #define si_swapinfo(val) \
383 do { (val)->freeswap = (val)->totalswap = 0; } while (0)
384 /* only sparc can not include linux/pagemap.h in this file
385 * so leave page_cache_release and release_pages undeclared... */
386 #define free_page_and_swap_cache(page) \
387 page_cache_release(page)
388 #define free_pages_and_swap_cache(pages, nr) \
389 release_pages((pages), (nr), 0);
391 static inline void show_swap_cache_info(void)
395 #define free_swap_and_cache(swp) is_migration_entry(swp)
396 #define swapcache_prepare(swp) is_migration_entry(swp)
398 static inline int add_swap_count_continuation(swp_entry_t swp, gfp_t gfp_mask)
403 static inline void swap_shmem_alloc(swp_entry_t swp)
407 static inline int swap_duplicate(swp_entry_t swp)
412 static inline void swap_free(swp_entry_t swp)
416 static inline void swapcache_free(swp_entry_t swp, struct page *page)
420 static inline struct page *swapin_readahead(swp_entry_t swp, gfp_t gfp_mask,
421 struct vm_area_struct *vma, unsigned long addr)
426 static inline int swap_writepage(struct page *p, struct writeback_control *wbc)
431 static inline struct page *lookup_swap_cache(swp_entry_t swp)
436 static inline int add_to_swap(struct page *page)
441 static inline int add_to_swap_cache(struct page *page, swp_entry_t entry,
447 static inline void __delete_from_swap_cache(struct page *page)
451 static inline void delete_from_swap_cache(struct page *page)
455 static inline int page_swapcount(struct page *page)
460 #define reuse_swap_page(page) (page_mapcount(page) == 1)
462 static inline int try_to_free_swap(struct page *page)
467 static inline swp_entry_t get_swap_page(void)
475 mem_cgroup_uncharge_swapcache(struct page *page, swp_entry_t ent)
479 #endif /* CONFIG_SWAP */
480 #endif /* __KERNEL__*/
481 #endif /* _LINUX_SWAP_H */