swap_state.c 10.4 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12
/*
 *  linux/mm/swap_state.c
 *
 *  Copyright (C) 1991, 1992, 1993, 1994  Linus Torvalds
 *  Swap reorganised 29.12.95, Stephen Tweedie
 *
 *  Rewritten to use page cache, (C) 1998 Stephen Tweedie
 */
#include <linux/module.h>
#include <linux/mm.h>
#include <linux/kernel_stat.h>
#include <linux/swap.h>
13
#include <linux/swapops.h>
L
Linus Torvalds 已提交
14 15 16 17
#include <linux/init.h>
#include <linux/pagemap.h>
#include <linux/buffer_head.h>
#include <linux/backing-dev.h>
18
#include <linux/pagevec.h>
C
Christoph Lameter 已提交
19
#include <linux/migrate.h>
L
Linus Torvalds 已提交
20 21 22 23 24

#include <asm/pgtable.h>

/*
 * swapper_space is a fiction, retained to simplify the path through
25
 * vmscan's shrink_page_list, to make sync_page look nicer, and to allow
L
Linus Torvalds 已提交
26 27
 * future use of radix_tree tags in the swap cache.
 */
28
static const struct address_space_operations swap_aops = {
L
Linus Torvalds 已提交
29 30 31
	.writepage	= swap_writepage,
	.sync_page	= block_sync_page,
	.set_page_dirty	= __set_page_dirty_nobuffers,
32
	.migratepage	= migrate_page,
L
Linus Torvalds 已提交
33 34 35 36 37 38 39 40 41
};

static struct backing_dev_info swap_backing_dev_info = {
	.capabilities	= BDI_CAP_NO_ACCT_DIRTY | BDI_CAP_NO_WRITEBACK,
	.unplug_io_fn	= swap_unplug_io_fn,
};

struct address_space swapper_space = {
	.page_tree	= RADIX_TREE_INIT(GFP_ATOMIC|__GFP_NOWARN),
42
	.tree_lock	= __RW_LOCK_UNLOCKED(swapper_space.tree_lock),
L
Linus Torvalds 已提交
43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58
	.a_ops		= &swap_aops,
	.i_mmap_nonlinear = LIST_HEAD_INIT(swapper_space.i_mmap_nonlinear),
	.backing_dev_info = &swap_backing_dev_info,
};

#define INC_CACHE_INFO(x)	do { swap_cache_info.x++; } while (0)

static struct {
	unsigned long add_total;
	unsigned long del_total;
	unsigned long find_success;
	unsigned long find_total;
} swap_cache_info;

void show_swap_cache_info(void)
{
59
	printk("Swap cache: add %lu, delete %lu, find %lu/%lu\n",
L
Linus Torvalds 已提交
60
		swap_cache_info.add_total, swap_cache_info.del_total,
61
		swap_cache_info.find_success, swap_cache_info.find_total);
L
Linus Torvalds 已提交
62 63 64 65 66
	printk("Free swap  = %lukB\n", nr_swap_pages << (PAGE_SHIFT - 10));
	printk("Total swap = %lukB\n", total_swap_pages << (PAGE_SHIFT - 10));
}

/*
H
Hugh Dickins 已提交
67
 * add_to_swap_cache resembles add_to_page_cache on swapper_space,
L
Linus Torvalds 已提交
68 69
 * but sets SwapCache flag and private instead of mapping and index.
 */
H
Hugh Dickins 已提交
70
static int add_to_swap_cache(struct page *page, swp_entry_t entry,
A
Al Viro 已提交
71
			       gfp_t gfp_mask)
L
Linus Torvalds 已提交
72 73 74
{
	int error;

75
	BUG_ON(!PageLocked(page));
L
Linus Torvalds 已提交
76 77 78 79 80 81 82 83 84 85
	BUG_ON(PageSwapCache(page));
	BUG_ON(PagePrivate(page));
	error = radix_tree_preload(gfp_mask);
	if (!error) {
		write_lock_irq(&swapper_space.tree_lock);
		error = radix_tree_insert(&swapper_space.page_tree,
						entry.val, page);
		if (!error) {
			page_cache_get(page);
			SetPageSwapCache(page);
H
Hugh Dickins 已提交
86
			set_page_private(page, entry.val);
L
Linus Torvalds 已提交
87
			total_swapcache_pages++;
88
			__inc_zone_page_state(page, NR_FILE_PAGES);
89
			INC_CACHE_INFO(add_total);
L
Linus Torvalds 已提交
90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105
		}
		write_unlock_irq(&swapper_space.tree_lock);
		radix_tree_preload_end();
	}
	return error;
}

/*
 * This must be called only on pages that have
 * been verified to be in the swap cache.
 */
void __delete_from_swap_cache(struct page *page)
{
	BUG_ON(!PageLocked(page));
	BUG_ON(!PageSwapCache(page));
	BUG_ON(PageWriteback(page));
106
	BUG_ON(PagePrivate(page));
L
Linus Torvalds 已提交
107

H
Hugh Dickins 已提交
108 109
	radix_tree_delete(&swapper_space.page_tree, page_private(page));
	set_page_private(page, 0);
L
Linus Torvalds 已提交
110 111
	ClearPageSwapCache(page);
	total_swapcache_pages--;
112
	__dec_zone_page_state(page, NR_FILE_PAGES);
L
Linus Torvalds 已提交
113 114 115 116 117 118 119 120 121 122
	INC_CACHE_INFO(del_total);
}

/**
 * add_to_swap - allocate swap space for a page
 * @page: page we want to move to swap
 *
 * Allocate swap space for the page and add the page to the
 * swap cache.  Caller needs to hold the page lock. 
 */
123
int add_to_swap(struct page * page, gfp_t gfp_mask)
L
Linus Torvalds 已提交
124 125 126 127
{
	swp_entry_t entry;
	int err;

128
	BUG_ON(!PageLocked(page));
L
Linus Torvalds 已提交
129 130 131 132 133 134

	for (;;) {
		entry = get_swap_page();
		if (!entry.val)
			return 0;

N
Nick Piggin 已提交
135 136 137 138
		/*
		 * Radix-tree node allocations from PF_MEMALLOC contexts could
		 * completely exhaust the page allocator. __GFP_NOMEMALLOC
		 * stops emergency reserves from being allocated.
L
Linus Torvalds 已提交
139
		 *
N
Nick Piggin 已提交
140 141
		 * TODO: this could cause a theoretical memory reclaim
		 * deadlock in the swap out path.
L
Linus Torvalds 已提交
142 143 144 145
		 */
		/*
		 * Add it to the swap cache and mark it dirty
		 */
H
Hugh Dickins 已提交
146
		err = add_to_swap_cache(page, entry,
147
				gfp_mask|__GFP_NOMEMALLOC|__GFP_NOWARN);
L
Linus Torvalds 已提交
148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175

		switch (err) {
		case 0:				/* Success */
			SetPageUptodate(page);
			SetPageDirty(page);
			return 1;
		case -EEXIST:
			/* Raced with "speculative" read_swap_cache_async */
			swap_free(entry);
			continue;
		default:
			/* -ENOMEM radix-tree allocation failure */
			swap_free(entry);
			return 0;
		}
	}
}

/*
 * This must be called only on pages that have
 * been verified to be in the swap cache and locked.
 * It will never put the page into the free list,
 * the caller has a reference on the page.
 */
void delete_from_swap_cache(struct page *page)
{
	swp_entry_t entry;

H
Hugh Dickins 已提交
176
	entry.val = page_private(page);
L
Linus Torvalds 已提交
177 178 179 180 181 182 183 184 185 186 187 188 189 190

	write_lock_irq(&swapper_space.tree_lock);
	__delete_from_swap_cache(page);
	write_unlock_irq(&swapper_space.tree_lock);

	swap_free(entry);
	page_cache_release(page);
}

/*
 * Strange swizzling function only for use by shmem_writepage
 */
int move_to_swap_cache(struct page *page, swp_entry_t entry)
{
H
Hugh Dickins 已提交
191
	int err = add_to_swap_cache(page, entry, GFP_ATOMIC);
L
Linus Torvalds 已提交
192 193 194 195 196 197
	if (!err) {
		remove_from_page_cache(page);
		page_cache_release(page);	/* pagecache ref */
		if (!swap_duplicate(entry))
			BUG();
		SetPageDirty(page);
198
	}
L
Linus Torvalds 已提交
199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235
	return err;
}

/*
 * Strange swizzling function for shmem_getpage (and shmem_unuse)
 */
int move_from_swap_cache(struct page *page, unsigned long index,
		struct address_space *mapping)
{
	int err = add_to_page_cache(page, mapping, index, GFP_ATOMIC);
	if (!err) {
		delete_from_swap_cache(page);
		/* shift page from clean_pages to dirty_pages list */
		ClearPageDirty(page);
		set_page_dirty(page);
	}
	return err;
}

/* 
 * If we are the only user, then try to free up the swap cache. 
 * 
 * Its ok to check for PageSwapCache without the page lock
 * here because we are going to recheck again inside 
 * exclusive_swap_page() _with_ the lock. 
 * 					- Marcelo
 */
static inline void free_swap_cache(struct page *page)
{
	if (PageSwapCache(page) && !TestSetPageLocked(page)) {
		remove_exclusive_swap_page(page);
		unlock_page(page);
	}
}

/* 
 * Perform a free_page(), also freeing any swap cache associated with
236
 * this page if it is the last user of the page.
L
Linus Torvalds 已提交
237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253
 */
void free_page_and_swap_cache(struct page *page)
{
	free_swap_cache(page);
	page_cache_release(page);
}

/*
 * Passed an array of pages, drop them all from swapcache and then release
 * them.  They are removed from the LRU and freed if this is their last use.
 */
void free_pages_and_swap_cache(struct page **pages, int nr)
{
	struct page **pagep = pages;

	lru_add_drain();
	while (nr) {
254
		int todo = min(nr, PAGEVEC_SIZE);
L
Linus Torvalds 已提交
255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289
		int i;

		for (i = 0; i < todo; i++)
			free_swap_cache(pagep[i]);
		release_pages(pagep, todo, 0);
		pagep += todo;
		nr -= todo;
	}
}

/*
 * Lookup a swap entry in the swap cache. A found page will be returned
 * unlocked and with its refcount incremented - we rely on the kernel
 * lock getting page table operations atomic even if we drop the page
 * lock before returning.
 */
struct page * lookup_swap_cache(swp_entry_t entry)
{
	struct page *page;

	page = find_get_page(&swapper_space, entry.val);

	if (page)
		INC_CACHE_INFO(find_success);

	INC_CACHE_INFO(find_total);
	return page;
}

/* 
 * Locate a page of swap in physical memory, reserving swap cache space
 * and reading the disk if it is not already cached.
 * A failure return means that either the page allocation failed or that
 * the swap entry is no longer in use.
 */
290
struct page *read_swap_cache_async(swp_entry_t entry, gfp_t gfp_mask,
L
Linus Torvalds 已提交
291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309
			struct vm_area_struct *vma, unsigned long addr)
{
	struct page *found_page, *new_page = NULL;
	int err;

	do {
		/*
		 * First check the swap cache.  Since this is normally
		 * called after lookup_swap_cache() failed, re-calling
		 * that would confuse statistics.
		 */
		found_page = find_get_page(&swapper_space, entry.val);
		if (found_page)
			break;

		/*
		 * Get a new page to read into from swap.
		 */
		if (!new_page) {
310
			new_page = alloc_page_vma(gfp_mask, vma, addr);
L
Linus Torvalds 已提交
311 312 313 314
			if (!new_page)
				break;		/* Out of memory */
		}

H
Hugh Dickins 已提交
315 316 317 318 319 320
		/*
		 * Swap entry may have been freed since our caller observed it.
		 */
		if (!swap_duplicate(entry))
			break;

L
Linus Torvalds 已提交
321 322
		/*
		 * Associate the page with swap entry in the swap cache.
H
Hugh Dickins 已提交
323 324 325 326
		 * May fail (-EEXIST) if there is already a page associated
		 * with this entry in the swap cache: added by a racing
		 * read_swap_cache_async, or add_to_swap or shmem_writepage
		 * re-using the just freed swap entry for an existing page.
L
Linus Torvalds 已提交
327 328
		 * May fail (-ENOMEM) if radix-tree node allocation failed.
		 */
H
Hugh Dickins 已提交
329 330
		SetPageLocked(new_page);
		err = add_to_swap_cache(new_page, entry, gfp_mask & GFP_KERNEL);
L
Linus Torvalds 已提交
331 332 333 334 335 336 337 338
		if (!err) {
			/*
			 * Initiate read into locked page and return.
			 */
			lru_cache_add_active(new_page);
			swap_readpage(NULL, new_page);
			return new_page;
		}
H
Hugh Dickins 已提交
339 340 341
		ClearPageLocked(new_page);
		swap_free(entry);
	} while (err != -ENOMEM);
L
Linus Torvalds 已提交
342 343 344 345 346

	if (new_page)
		page_cache_release(new_page);
	return found_page;
}
347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365

/**
 * swapin_readahead - swap in pages in hope we need them soon
 * @entry: swap entry of this memory
 * @vma: user vma this address belongs to
 * @addr: target address for mempolicy
 *
 * Returns the struct page for entry and addr, after queueing swapin.
 *
 * Primitive swap readahead code. We simply read an aligned block of
 * (1 << page_cluster) entries in the swap area. This method is chosen
 * because it doesn't cost us any seek time.  We also make sure to queue
 * the 'original' request together with the readahead ones...
 *
 * This has been extended to use the NUMA policies from the mm triggering
 * the readahead.
 *
 * Caller must hold down_read on the vma->vm_mm if vma is not NULL.
 */
366
struct page *swapin_readahead(swp_entry_t entry, gfp_t gfp_mask,
367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384
			struct vm_area_struct *vma, unsigned long addr)
{
	int nr_pages;
	struct page *page;
	unsigned long offset;
	unsigned long end_offset;

	/*
	 * Get starting offset for readaround, and number of pages to read.
	 * Adjust starting address by readbehind (for NUMA interleave case)?
	 * No, it's very unlikely that swap layout would follow vma layout,
	 * more likely that neighbouring swap pages came from the same node:
	 * so use the same "addr" to choose the same node for each swap read.
	 */
	nr_pages = valid_swaphandles(entry, &offset);
	for (end_offset = offset + nr_pages; offset < end_offset; offset++) {
		/* Ok, do the async read-ahead now */
		page = read_swap_cache_async(swp_entry(swp_type(entry), offset),
385
						gfp_mask, vma, addr);
386 387 388 389 390
		if (!page)
			break;
		page_cache_release(page);
	}
	lru_add_drain();	/* Push any new pages onto the LRU now */
391
	return read_swap_cache_async(entry, gfp_mask, vma, addr);
392
}