truncate.c 12.9 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10
/*
 * mm/truncate.c - code for taking down pages from address_spaces
 *
 * Copyright (C) 2002, Linus Torvalds
 *
 * 10Sep2002	akpm@zip.com.au
 *		Initial version.
 */

#include <linux/kernel.h>
A
Alexey Dobriyan 已提交
11
#include <linux/backing-dev.h>
L
Linus Torvalds 已提交
12
#include <linux/mm.h>
N
Nick Piggin 已提交
13
#include <linux/swap.h>
L
Linus Torvalds 已提交
14 15
#include <linux/module.h>
#include <linux/pagemap.h>
16
#include <linux/highmem.h>
L
Linus Torvalds 已提交
17
#include <linux/pagevec.h>
18
#include <linux/task_io_accounting_ops.h>
L
Linus Torvalds 已提交
19
#include <linux/buffer_head.h>	/* grr. try_to_release_page,
20
				   do_invalidatepage */
L
Linus Torvalds 已提交
21 22


23
/**
24
 * do_invalidatepage - invalidate part or all of a page
25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40
 * @page: the page which is affected
 * @offset: the index of the truncation point
 *
 * do_invalidatepage() is called when all or part of the page has become
 * invalidated by a truncate operation.
 *
 * do_invalidatepage() does not have to release all buffers, but it must
 * ensure that no dirty buffer is left outside @offset and that no I/O
 * is underway against any of the blocks which are outside the truncation
 * point.  Because the caller is about to free (and possibly reuse) those
 * blocks on-disk.
 */
void do_invalidatepage(struct page *page, unsigned long offset)
{
	void (*invalidatepage)(struct page *, unsigned long);
	invalidatepage = page->mapping->a_ops->invalidatepage;
41
#ifdef CONFIG_BLOCK
42 43
	if (!invalidatepage)
		invalidatepage = block_invalidatepage;
44
#endif
45 46 47 48
	if (invalidatepage)
		(*invalidatepage)(page, offset);
}

L
Linus Torvalds 已提交
49 50
static inline void truncate_partial_page(struct page *page, unsigned partial)
{
51
	zero_user_segment(page, partial, PAGE_CACHE_SIZE);
L
Linus Torvalds 已提交
52 53 54 55
	if (PagePrivate(page))
		do_invalidatepage(page, partial);
}

56 57 58 59 60 61 62 63 64 65 66 67 68 69
/*
 * This cancels just the dirty bit on the kernel page itself, it
 * does NOT actually remove dirty bits on any mmap's that may be
 * around. It also leaves the page tagged dirty, so any sync
 * activity will still find it on the dirty lists, and in particular,
 * clear_page_dirty_for_io() will still look at the dirty bits in
 * the VM.
 *
 * Doing this should *normally* only ever be done when a page
 * is truncated, and is not actually mapped anywhere at all. However,
 * fs/buffer.c does this when it notices that somebody has cleaned
 * out all the buffers on a page without actually doing it through
 * the VM. Can you say "ext3 is horribly ugly"? Tought you could.
 */
70 71
void cancel_dirty_page(struct page *page, unsigned int account_size)
{
72 73 74 75
	if (TestClearPageDirty(page)) {
		struct address_space *mapping = page->mapping;
		if (mapping && mapping_cap_account_dirty(mapping)) {
			dec_zone_page_state(page, NR_FILE_DIRTY);
76 77
			dec_bdi_stat(mapping->backing_dev_info,
					BDI_RECLAIMABLE);
78 79 80
			if (account_size)
				task_io_account_cancelled_write(account_size);
		}
81
	}
82
}
83
EXPORT_SYMBOL(cancel_dirty_page);
84

L
Linus Torvalds 已提交
85 86 87
/*
 * If truncate cannot remove the fs-private metadata from the page, the page
 * becomes anonymous.  It will be left on the LRU and may even be mapped into
88
 * user pagetables if we're racing with filemap_fault().
L
Linus Torvalds 已提交
89 90 91
 *
 * We need to bale out if page->mapping is no longer equal to the original
 * mapping.  This happens a) when the VM reclaimed the page while we waited on
92
 * its lock, b) when a concurrent invalidate_mapping_pages got there first and
L
Linus Torvalds 已提交
93 94 95 96 97 98 99 100
 * c) when tmpfs swizzles a page between a tmpfs inode and swapper_space.
 */
static void
truncate_complete_page(struct address_space *mapping, struct page *page)
{
	if (page->mapping != mapping)
		return;

101 102
	cancel_dirty_page(page, PAGE_CACHE_SIZE);

L
Linus Torvalds 已提交
103 104 105
	if (PagePrivate(page))
		do_invalidatepage(page, 0);

106
	remove_from_page_cache(page);
L
Linus Torvalds 已提交
107 108 109 110 111 112
	ClearPageUptodate(page);
	ClearPageMappedToDisk(page);
	page_cache_release(page);	/* pagecache ref */
}

/*
113
 * This is for invalidate_mapping_pages().  That function can be called at
L
Linus Torvalds 已提交
114
 * any time, and is not supposed to throw away dirty pages.  But pages can
N
Nick Piggin 已提交
115 116
 * be marked dirty at any time too, so use remove_mapping which safely
 * discards clean, unused pages.
L
Linus Torvalds 已提交
117 118 119 120 121 122
 *
 * Returns non-zero if the page was successfully invalidated.
 */
static int
invalidate_complete_page(struct address_space *mapping, struct page *page)
{
N
Nick Piggin 已提交
123 124
	int ret;

L
Linus Torvalds 已提交
125 126 127 128 129 130
	if (page->mapping != mapping)
		return 0;

	if (PagePrivate(page) && !try_to_release_page(page, 0))
		return 0;

N
Nick Piggin 已提交
131 132 133
	ret = remove_mapping(mapping, page);

	return ret;
L
Linus Torvalds 已提交
134 135 136
}

/**
137 138
 * truncate_inode_pages - truncate range of pages specified by start and
 * end byte offsets
L
Linus Torvalds 已提交
139 140
 * @mapping: mapping to truncate
 * @lstart: offset from which to truncate
141
 * @lend: offset to which to truncate
L
Linus Torvalds 已提交
142
 *
143 144 145
 * Truncate the page cache, removing the pages that are between
 * specified offsets (and zeroing out partial page
 * (if lstart is not page aligned)).
L
Linus Torvalds 已提交
146 147 148 149 150 151 152 153 154 155 156 157 158 159
 *
 * Truncate takes two passes - the first pass is nonblocking.  It will not
 * block on page locks and it will not block on writeback.  The second pass
 * will wait.  This is to prevent as much IO as possible in the affected region.
 * The first pass will remove most pages, so the search cost of the second pass
 * is low.
 *
 * When looking at page->index outside the page lock we need to be careful to
 * copy it into a local to avoid races (it could change at any time).
 *
 * We pass down the cache-hot hint to the page freeing code.  Even if the
 * mapping is large, it is probably the case that the final pages are the most
 * recently touched, and freeing happens in ascending file offset order.
 */
160 161
void truncate_inode_pages_range(struct address_space *mapping,
				loff_t lstart, loff_t lend)
L
Linus Torvalds 已提交
162 163
{
	const pgoff_t start = (lstart + PAGE_CACHE_SIZE-1) >> PAGE_CACHE_SHIFT;
164
	pgoff_t end;
L
Linus Torvalds 已提交
165 166 167 168 169 170 171 172
	const unsigned partial = lstart & (PAGE_CACHE_SIZE - 1);
	struct pagevec pvec;
	pgoff_t next;
	int i;

	if (mapping->nrpages == 0)
		return;

173 174 175
	BUG_ON((lend & (PAGE_CACHE_SIZE - 1)) != (PAGE_CACHE_SIZE - 1));
	end = (lend >> PAGE_CACHE_SHIFT);

L
Linus Torvalds 已提交
176 177
	pagevec_init(&pvec, 0);
	next = start;
178 179
	while (next <= end &&
	       pagevec_lookup(&pvec, mapping, next, PAGEVEC_SIZE)) {
L
Linus Torvalds 已提交
180 181 182 183
		for (i = 0; i < pagevec_count(&pvec); i++) {
			struct page *page = pvec.pages[i];
			pgoff_t page_index = page->index;

184 185 186 187 188
			if (page_index > end) {
				next = page_index;
				break;
			}

L
Linus Torvalds 已提交
189 190 191 192 193 194 195 196 197
			if (page_index > next)
				next = page_index;
			next++;
			if (TestSetPageLocked(page))
				continue;
			if (PageWriteback(page)) {
				unlock_page(page);
				continue;
			}
198 199 200 201 202
			if (page_mapped(page)) {
				unmap_mapping_range(mapping,
				  (loff_t)page_index<<PAGE_CACHE_SHIFT,
				  PAGE_CACHE_SIZE, 0);
			}
L
Linus Torvalds 已提交
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
			truncate_complete_page(mapping, page);
			unlock_page(page);
		}
		pagevec_release(&pvec);
		cond_resched();
	}

	if (partial) {
		struct page *page = find_lock_page(mapping, start - 1);
		if (page) {
			wait_on_page_writeback(page);
			truncate_partial_page(page, partial);
			unlock_page(page);
			page_cache_release(page);
		}
	}

	next = start;
	for ( ; ; ) {
		cond_resched();
		if (!pagevec_lookup(&pvec, mapping, next, PAGEVEC_SIZE)) {
			if (next == start)
				break;
			next = start;
			continue;
		}
229 230 231 232
		if (pvec.pages[0]->index > end) {
			pagevec_release(&pvec);
			break;
		}
L
Linus Torvalds 已提交
233 234 235
		for (i = 0; i < pagevec_count(&pvec); i++) {
			struct page *page = pvec.pages[i];

236 237
			if (page->index > end)
				break;
L
Linus Torvalds 已提交
238 239
			lock_page(page);
			wait_on_page_writeback(page);
240 241 242 243 244
			if (page_mapped(page)) {
				unmap_mapping_range(mapping,
				  (loff_t)page->index<<PAGE_CACHE_SHIFT,
				  PAGE_CACHE_SIZE, 0);
			}
L
Linus Torvalds 已提交
245 246 247 248 249 250 251 252 253
			if (page->index > next)
				next = page->index;
			next++;
			truncate_complete_page(mapping, page);
			unlock_page(page);
		}
		pagevec_release(&pvec);
	}
}
254
EXPORT_SYMBOL(truncate_inode_pages_range);
L
Linus Torvalds 已提交
255

256 257 258 259 260
/**
 * truncate_inode_pages - truncate *all* the pages from an offset
 * @mapping: mapping to truncate
 * @lstart: offset from which to truncate
 *
261
 * Called under (and serialised by) inode->i_mutex.
262 263 264 265 266
 */
void truncate_inode_pages(struct address_space *mapping, loff_t lstart)
{
	truncate_inode_pages_range(mapping, lstart, (loff_t)-1);
}
L
Linus Torvalds 已提交
267 268
EXPORT_SYMBOL(truncate_inode_pages);

269 270
unsigned long __invalidate_mapping_pages(struct address_space *mapping,
				pgoff_t start, pgoff_t end, bool be_atomic)
L
Linus Torvalds 已提交
271 272 273 274 275 276 277 278 279 280 281
{
	struct pagevec pvec;
	pgoff_t next = start;
	unsigned long ret = 0;
	int i;

	pagevec_init(&pvec, 0);
	while (next <= end &&
			pagevec_lookup(&pvec, mapping, next, PAGEVEC_SIZE)) {
		for (i = 0; i < pagevec_count(&pvec); i++) {
			struct page *page = pvec.pages[i];
282 283
			pgoff_t index;
			int lock_failed;
L
Linus Torvalds 已提交
284

285 286 287 288 289 290 291 292 293 294 295
			lock_failed = TestSetPageLocked(page);

			/*
			 * We really shouldn't be looking at the ->index of an
			 * unlocked page.  But we're not allowed to lock these
			 * pages.  So we rely upon nobody altering the ->index
			 * of this (pinned-by-us) page.
			 */
			index = page->index;
			if (index > next)
				next = index;
L
Linus Torvalds 已提交
296
			next++;
297 298 299
			if (lock_failed)
				continue;

L
Linus Torvalds 已提交
300 301 302 303 304 305 306 307 308 309 310
			if (PageDirty(page) || PageWriteback(page))
				goto unlock;
			if (page_mapped(page))
				goto unlock;
			ret += invalidate_complete_page(mapping, page);
unlock:
			unlock_page(page);
			if (next > end)
				break;
		}
		pagevec_release(&pvec);
311 312
		if (likely(!be_atomic))
			cond_resched();
L
Linus Torvalds 已提交
313 314 315
	}
	return ret;
}
316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334

/**
 * invalidate_mapping_pages - Invalidate all the unlocked pages of one inode
 * @mapping: the address_space which holds the pages to invalidate
 * @start: the offset 'from' which to invalidate
 * @end: the offset 'to' which to invalidate (inclusive)
 *
 * This function only removes the unlocked pages, if you want to
 * remove all the pages of one inode, you must call truncate_inode_pages.
 *
 * invalidate_mapping_pages() will not block on IO activity. It will not
 * invalidate pages which are dirty, locked, under writeback or mapped into
 * pagetables.
 */
unsigned long invalidate_mapping_pages(struct address_space *mapping,
				pgoff_t start, pgoff_t end)
{
	return __invalidate_mapping_pages(mapping, start, end, false);
}
335
EXPORT_SYMBOL(invalidate_mapping_pages);
L
Linus Torvalds 已提交
336

337 338 339 340
/*
 * This is like invalidate_complete_page(), except it ignores the page's
 * refcount.  We do this because invalidate_inode_pages2() needs stronger
 * invalidation guarantees, and cannot afford to leave pages behind because
341 342
 * shrink_page_list() has a temp ref on them, or because they're transiently
 * sitting in the lru_cache_add() pagevecs.
343 344 345 346 347 348 349
 */
static int
invalidate_complete_page2(struct address_space *mapping, struct page *page)
{
	if (page->mapping != mapping)
		return 0;

350
	if (PagePrivate(page) && !try_to_release_page(page, GFP_KERNEL))
351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367
		return 0;

	write_lock_irq(&mapping->tree_lock);
	if (PageDirty(page))
		goto failed;

	BUG_ON(PagePrivate(page));
	__remove_from_page_cache(page);
	write_unlock_irq(&mapping->tree_lock);
	ClearPageUptodate(page);
	page_cache_release(page);	/* pagecache ref */
	return 1;
failed:
	write_unlock_irq(&mapping->tree_lock);
	return 0;
}

368 369 370 371 372 373 374 375 376
static int do_launder_page(struct address_space *mapping, struct page *page)
{
	if (!PageDirty(page))
		return 0;
	if (page->mapping != mapping || mapping->a_ops->launder_page == NULL)
		return 0;
	return mapping->a_ops->launder_page(page);
}

L
Linus Torvalds 已提交
377 378
/**
 * invalidate_inode_pages2_range - remove range of pages from an address_space
379
 * @mapping: the address_space
L
Linus Torvalds 已提交
380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399
 * @start: the page offset 'from' which to invalidate
 * @end: the page offset 'to' which to invalidate (inclusive)
 *
 * Any pages which are found to be mapped into pagetables are unmapped prior to
 * invalidation.
 *
 * Returns -EIO if any pages could not be invalidated.
 */
int invalidate_inode_pages2_range(struct address_space *mapping,
				  pgoff_t start, pgoff_t end)
{
	struct pagevec pvec;
	pgoff_t next;
	int i;
	int ret = 0;
	int did_range_unmap = 0;
	int wrapped = 0;

	pagevec_init(&pvec, 0);
	next = start;
400
	while (next <= end && !wrapped &&
L
Linus Torvalds 已提交
401 402
		pagevec_lookup(&pvec, mapping, next,
			min(end - next, (pgoff_t)PAGEVEC_SIZE - 1) + 1)) {
403
		for (i = 0; i < pagevec_count(&pvec); i++) {
L
Linus Torvalds 已提交
404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420
			struct page *page = pvec.pages[i];
			pgoff_t page_index;

			lock_page(page);
			if (page->mapping != mapping) {
				unlock_page(page);
				continue;
			}
			page_index = page->index;
			next = page_index + 1;
			if (next == 0)
				wrapped = 1;
			if (page_index > end) {
				unlock_page(page);
				break;
			}
			wait_on_page_writeback(page);
421
			if (page_mapped(page)) {
L
Linus Torvalds 已提交
422 423 424 425 426
				if (!did_range_unmap) {
					/*
					 * Zap the rest of the file in one hit.
					 */
					unmap_mapping_range(mapping,
427 428
					   (loff_t)page_index<<PAGE_CACHE_SHIFT,
					   (loff_t)(end - page_index + 1)
L
Linus Torvalds 已提交
429 430 431 432 433 434 435 436
							<< PAGE_CACHE_SHIFT,
					    0);
					did_range_unmap = 1;
				} else {
					/*
					 * Just zap this page
					 */
					unmap_mapping_range(mapping,
437
					  (loff_t)page_index<<PAGE_CACHE_SHIFT,
L
Linus Torvalds 已提交
438 439 440
					  PAGE_CACHE_SIZE, 0);
				}
			}
441
			BUG_ON(page_mapped(page));
442 443
			ret = do_launder_page(mapping, page);
			if (ret == 0 && !invalidate_complete_page2(mapping, page))
L
Linus Torvalds 已提交
444 445 446 447 448 449 450 451 452 453 454 455
				ret = -EIO;
			unlock_page(page);
		}
		pagevec_release(&pvec);
		cond_resched();
	}
	return ret;
}
EXPORT_SYMBOL_GPL(invalidate_inode_pages2_range);

/**
 * invalidate_inode_pages2 - remove all pages from an address_space
456
 * @mapping: the address_space
L
Linus Torvalds 已提交
457 458 459 460 461 462 463 464 465 466 467
 *
 * Any pages which are found to be mapped into pagetables are unmapped prior to
 * invalidation.
 *
 * Returns -EIO if any pages could not be invalidated.
 */
int invalidate_inode_pages2(struct address_space *mapping)
{
	return invalidate_inode_pages2_range(mapping, 0, -1);
}
EXPORT_SYMBOL_GPL(invalidate_inode_pages2);