aboutsummaryrefslogtreecommitdiffstats
path: root/mm/readahead.c
blob: 9861e883fe57e069d43492ebdbd7d7671464f371 (plain) (blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
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
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
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
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
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
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
/*
 * mm/readahead.c - address_space-level file readahead.
 *
 * Copyright (C) 2002, Linus Torvalds
 *
 * 09Apr2002	akpm@zip.com.au
 *		Initial version.
 */

#include <linux/kernel.h>
#include <linux/fs.h>
#include <linux/mm.h>
#include <linux/module.h>
#include <linux/blkdev.h>
#include <linux/backing-dev.h>
#include <linux/task_io_accounting_ops.h>
#include <linux/pagevec.h>

void default_unplug_io_fn(struct backing_dev_info *bdi, struct page *page)
{
}
EXPORT_SYMBOL(default_unplug_io_fn);

struct backing_dev_info default_backing_dev_info = {
	.ra_pages	= (VM_MAX_READAHEAD * 1024) / PAGE_CACHE_SIZE,
	.state		= 0,
	.capabilities	= BDI_CAP_MAP_COPY,
	.unplug_io_fn	= default_unplug_io_fn,
};
EXPORT_SYMBOL_GPL(default_backing_dev_info);

/*
 * Initialise a struct file's readahead state.  Assumes that the caller has
 * memset *ra to zero.
 */
void
file_ra_state_init(struct file_ra_state *ra, struct address_space *mapping)
{
	ra->ra_pages = mapping->backing_dev_info->ra_pages;
	ra->prev_index = -1;
}
EXPORT_SYMBOL_GPL(file_ra_state_init);

/*
 * Return max readahead size for this inode in number-of-pages.
 */
static inline unsigned long get_max_readahead(struct file_ra_state *ra)
{
	return ra->ra_pages;
}

static inline unsigned long get_min_readahead(struct file_ra_state *ra)
{
	return (VM_MIN_READAHEAD * 1024) / PAGE_CACHE_SIZE;
}

static inline void reset_ahead_window(struct file_ra_state *ra)
{
	/*
	 * ... but preserve ahead_start + ahead_size value,
	 * see 'recheck:' label in page_cache_readahead().
	 * Note: We never use ->ahead_size as rvalue without
	 * checking ->ahead_start != 0 first.
	 */
	ra->ahead_size += ra->ahead_start;
	ra->ahead_start = 0;
}

static inline void ra_off(struct file_ra_state *ra)
{
	ra->start = 0;
	ra->flags = 0;
	ra->size = 0;
	reset_ahead_window(ra);
	return;
}

/*
 * Set the initial window size, round to next power of 2 and square
 * for small size, x 4 for medium, and x 2 for large
 * for 128k (32 page) max ra
 * 1-8 page = 32k initial, > 8 page = 128k initial
 */
static unsigned long get_init_ra_size(unsigned long size, unsigned long max)
{
	unsigned long newsize = roundup_pow_of_two(size);

	if (newsize <= max / 32)
		newsize = newsize * 4;
	else if (newsize <= max / 4)
		newsize = newsize * 2;
	else
		newsize = max;
	return newsize;
}

/*
 * Set the new window size, this is called only when I/O is to be submitted,
 * not for each call to readahead.  If a cache miss occured, reduce next I/O
 * size, else increase depending on how close to max we are.
 */
static inline unsigned long get_next_ra_size(struct file_ra_state *ra)
{
	unsigned long max = get_max_readahead(ra);
	unsigned long min = get_min_readahead(ra);
	unsigned long cur = ra->size;
	unsigned long newsize;

	if (ra->flags & RA_FLAG_MISS) {
		ra->flags &= ~RA_FLAG_MISS;
		newsize = max((cur - 2), min);
	} else if (cur < max / 16) {
		newsize = 4 * cur;
	} else {
		newsize = 2 * cur;
	}
	return min(newsize, max);
}

#define list_to_page(head) (list_entry((head)->prev, struct page, lru))

/**
 * read_cache_pages - populate an address space with some pages & start reads against them
 * @mapping: the address_space
 * @pages: The address of a list_head which contains the target pages.  These
 *   pages have their ->index populated and are otherwise uninitialised.
 * @filler: callback routine for filling a single page.
 * @data: private data for the callback routine.
 *
 * Hides the details of the LRU cache etc from the filesystems.
 */
int read_cache_pages(struct address_space *mapping, struct list_head *pages,
			int (*filler)(void *, struct page *), void *data)
{
	struct page *page;
	struct pagevec lru_pvec;
	int ret = 0;

	pagevec_init(&lru_pvec, 0);

	while (!list_empty(pages)) {
		page = list_to_page(pages);
		list_del(&page->lru);
		if (add_to_page_cache(page, mapping, page->index, GFP_KERNEL)) {
			page_cache_release(page);
			continue;
		}
		ret = filler(data, page);
		if (!pagevec_add(&lru_pvec, page))
			__pagevec_lru_add(&lru_pvec);
		if (ret) {
			put_pages_list(pages);
			break;
		}
		task_io_account_read(PAGE_CACHE_SIZE);
	}
	pagevec_lru_add(&lru_pvec);
	return ret;
}

EXPORT_SYMBOL(read_cache_pages);

static int read_pages(struct address_space *mapping, struct file *filp,
		struct list_head *pages, unsigned nr_pages)
{
	unsigned page_idx;
	struct pagevec lru_pvec;
	int ret;

	if (mapping->a_ops->readpages) {
		ret = mapping->a_ops->readpages(filp, mapping, pages, nr_pages);
		/* Clean up the remaining pages */
		put_pages_list(pages);
		goto out;
	}

	pagevec_init(&lru_pvec, 0);
	for (page_idx = 0; page_idx < nr_pages; page_idx++) {
		struct page *page = list_to_page(pages);
		list_del(&page->lru);
		if (!add_to_page_cache(page, mapping,
					page->index, GFP_KERNEL)) {
			mapping->a_ops->readpage(filp, page);
			if (!pagevec_add(&lru_pvec, page))
				__pagevec_lru_add(&lru_pvec);
		} else
			page_cache_release(page);
	}
	pagevec_lru_add(&lru_pvec);
	ret = 0;
out:
	return ret;
}

/*
 * Readahead design.
 *
 * The fields in struct file_ra_state represent the most-recently-executed
 * readahead attempt:
 *
 * start:	Page index at which we started the readahead
 * size:	Number of pages in that read
 *              Together, these form the "current window".
 *              Together, start and size represent the `readahead window'.
 * prev_index:  The page which the readahead algorithm most-recently inspected.
 *              It is mainly used to detect sequential file reading.
 *              If page_cache_readahead sees that it is again being called for
 *              a page which it just looked at, it can return immediately without
 *              making any state changes.
 * offset:      Offset in the prev_index where the last read ended - used for
 *              detection of sequential file reading.
 * ahead_start,
 * ahead_size:  Together, these form the "ahead window".
 * ra_pages:	The externally controlled max readahead for this fd.
 *
 * When readahead is in the off state (size == 0), readahead is disabled.
 * In this state, prev_index is used to detect the resumption of sequential I/O.
 *
 * The readahead code manages two windows - the "current" and the "ahead"
 * windows.  The intent is that while the application is walking the pages
 * in the current window, I/O is underway on the ahead window.  When the
 * current window is fully traversed, it is replaced by the ahead window
 * and the ahead window is invalidated.  When this copying happens, the
 * new current window's pages are probably still locked.  So
 * we submit a new batch of I/O immediately, creating a new ahead window.
 *
 * So:
 *
 *   ----|----------------|----------------|-----
 *       ^start           ^start+size
 *                        ^ahead_start     ^ahead_start+ahead_size
 *
 *         ^ When this page is read, we submit I/O for the
 *           ahead window.
 *
 * A `readahead hit' occurs when a read request is made against a page which is
 * the next sequential page. Ahead window calculations are done only when it
 * is time to submit a new IO.  The code ramps up the size agressively at first,
 * but slow down as it approaches max_readhead.
 *
 * Any seek/ramdom IO will result in readahead being turned off.  It will resume
 * at the first sequential access.
 *
 * There is a special-case: if the first page which the application tries to
 * read happens to be the first page of the file, it is assumed that a linear
 * read is about to happen and the window is immediately set to the initial size
 * based on I/O request size and the max_readahead.
 *
 * This function is to be called for every read request, rather than when
 * it is time to perform readahead.  It is called only once for the entire I/O
 * regardless of size unless readahead is unable to start enough I/O to satisfy
 * the request (I/O request > max_readahead).
 */

/*
 * do_page_cache_readahead actually reads a chunk of disk.  It allocates all
 * the pages first, then submits them all for I/O. This avoids the very bad
 * behaviour which would occur if page allocations are causing VM writeback.
 * We really don't want to intermingle reads and writes like that.
 *
 * Returns the number of pages requested, or the maximum amount of I/O allowed.
 *
 * do_page_cache_readahead() returns -1 if it encountered request queue
 * congestion.
 */
static int
__do_page_cache_readahead(struct address_space *mapping, struct file *filp,
			pgoff_t offset, unsigned long nr_to_read)
{
	struct inode *inode = mapping->host;
	struct page *page;
	unsigned long end_index;	/* The last page we want to read */
	LIST_HEAD(page_pool);
	int page_idx;
	int ret = 0;
	loff_t isize = i_size_read(inode);

	if (isize == 0)
		goto out;

 	end_index = ((isize - 1) >> PAGE_CACHE_SHIFT);

	/*
	 * Preallocate as many pages as we will need.
	 */
	read_lock_irq(&mapping->tree_lock);
	for (page_idx = 0; page_idx < nr_to_read; page_idx++) {
		pgoff_t page_offset = offset + page_idx;
		
		if (page_offset > end_index)
			break;

		page = radix_tree_lookup(&mapping->page_tree, page_offset);
		if (page)
			continue;

		read_unlock_irq(&mapping->tree_lock);
		page = page_cache_alloc_cold(mapping);
		read_lock_irq(&mapping->tree_lock);
		if (!page)
			break;
		page->index = page_offset;
		list_add(&page->lru, &page_pool);
		ret++;
	}
	read_unlock_irq(&mapping->tree_lock);

	/*
	 * Now start the IO.  We ignore I/O errors - if the page is not
	 * uptodate then the caller will launch readpage again, and
	 * will then handle the error.
	 */
	if (ret)
		read_pages(mapping, filp, &page_pool, ret);
	BUG_ON(!list_empty(&page_pool));
out:
	return ret;
}

/*
 * Chunk the readahead into 2 megabyte units, so that we don't pin too much
 * memory at once.
 */
int force_page_cache_readahead(struct address_space *mapping, struct file *filp,
		pgoff_t offset, unsigned long nr_to_read)
{
	int ret = 0;

	if (unlikely(!mapping->a_ops->readpage && !mapping->a_ops->readpages))
		return -EINVAL;

	while (nr_to_read) {
		int err;

		unsigned long this_chunk = (2 * 1024 * 1024) / PAGE_CACHE_SIZE;

		if (this_chunk > nr_to_read)
			this_chunk = nr_to_read;
		err = __do_page_cache_readahead(mapping, filp,
						offset, this_chunk);
		if (err < 0) {
			ret = err;
			break;
		}
		ret += err;
		offset += this_chunk;
		nr_to_read -= this_chunk;
	}
	return ret;
}

/*
 * Check how effective readahead is being.  If the amount of started IO is
 * less than expected then the file is partly or fully in pagecache and
 * readahead isn't helping.
 *
 */
static inline int check_ra_success(struct file_ra_state *ra,
			unsigned long nr_to_read, unsigned long actual)
{
	if (actual == 0) {
		ra->cache_hit += nr_to_read;
		if (ra->cache_hit >= VM_MAX_CACHE_HIT) {
			ra_off(ra);
			ra->flags |= RA_FLAG_INCACHE;
			return 0;
		}
	} else {
		ra->cache_hit=0;
	}
	return 1;
}

/*
 * This version skips the IO if the queue is read-congested, and will tell the
 * block layer to abandon the readahead if request allocation would block.
 *
 * force_page_cache_readahead() will ignore queue congestion and will block on
 * request queues.
 */
int do_page_cache_readahead(struct address_space *mapping, struct file *filp,
			pgoff_t offset, unsigned long nr_to_read)
{
	if (bdi_read_congested(mapping->backing_dev_info))
		return -1;

	return __do_page_cache_readahead(mapping, filp, offset, nr_to_read);
}

/*
 * Read 'nr_to_read' pages starting at page 'offset'. If the flag 'block'
 * is set wait till the read completes.  Otherwise attempt to read without
 * blocking.
 * Returns 1 meaning 'success' if read is successful without switching off
 * readahead mode. Otherwise return failure.
 */
static int
blockable_page_cache_readahead(struct address_space *mapping, struct file *filp,
			pgoff_t offset, unsigned long nr_to_read,
			struct file_ra_state *ra, int block)
{
	int actual;

	if (!block && bdi_read_congested(mapping->backing_dev_info))
		return 0;

	actual = __do_page_cache_readahead(mapping, filp, offset, nr_to_read);

	return check_ra_success(ra, nr_to_read, actual);
}

static int make_ahead_window(struct address_space *mapping, struct file *filp,
				struct file_ra_state *ra, int force)
{
	int block, ret;

	ra->ahead_size = get_next_ra_size(ra);
	ra->ahead_start = ra->start + ra->size;

	block = force || (ra->prev_index >= ra->ahead_start);
	ret = blockable_page_cache_readahead(mapping, filp,
			ra->ahead_start, ra->ahead_size, ra, block);

	if (!ret && !force) {
		/* A read failure in blocking mode, implies pages are
		 * all cached. So we can safely assume we have taken
		 * care of all the pages requested in this call.
		 * A read failure in non-blocking mode, implies we are
		 * reading more pages than requested in this call.  So
		 * we safely assume we have taken care of all the pages
		 * requested in this call.
		 *
		 * Just reset the ahead window in case we failed due to
		 * congestion.  The ahead window will any way be closed
		 * in case we failed due to excessive page cache hits.
		 */
		reset_ahead_window(ra);
	}

	return ret;
}

/**
 * page_cache_readahead - generic adaptive readahead
 * @mapping: address_space which holds the pagecache and I/O vectors
 * @ra: file_ra_state which holds the readahead state
 * @filp: passed on to ->readpage() and ->readpages()
 * @offset: start offset into @mapping, in PAGE_CACHE_SIZE units
 * @req_size: hint: total size of the read which the caller is performing in
 *            PAGE_CACHE_SIZE units
 *
 * page_cache_readahead() is the main function.  If performs the adaptive
 * readahead window size management and submits the readahead I/O.
 *
 * Note that @filp is purely used for passing on to the ->readpage[s]()
 * handler: it may refer to a different file from @mapping (so we may not use
 * @filp->f_mapping or @filp->f_path.dentry->d_inode here).
 * Also, @ra may not be equal to &@filp->f_ra.
 *
 */
unsigned long
page_cache_readahead(struct address_space *mapping, struct file_ra_state *ra,
		     struct file *filp, pgoff_t offset, unsigned long req_size)
{
	unsigned long max, newsize;
	int sequential;

	/*
	 * We avoid doing extra work and bogusly perturbing the readahead
	 * window expansion logic.
	 */
	if (offset == ra->prev_index && --req_size)
		++offset;

	/* Note that prev_index == -1 if it is a first read */
	sequential = (offset == ra->prev_index + 1);
	ra->prev_index = offset;
	ra->prev_offset = 0;

	max = get_max_readahead(ra);
	newsize = min(req_size, max);

	/* No readahead or sub-page sized read or file already in cache */
	if (newsize == 0 || (ra->flags & RA_FLAG_INCACHE))
		goto out;

	ra->prev_index += newsize - 1;

	/*
	 * Special case - first read at start of file. We'll assume it's
	 * a whole-file read and grow the window fast.  Or detect first
	 * sequential access
	 */
	if (sequential && ra->size == 0) {
		ra->size = get_init_ra_size(newsize, max);
		ra->start = offset;
		if (!blockable_page_cache_readahead(mapping, filp, offset,
							 ra->size, ra, 1))
			goto out;

		/*
		 * If the request size is larger than our max readahead, we
		 * at least want to be sure that we get 2 IOs in flight and
		 * we know that we will definitly need the new I/O.
		 * once we do this, subsequent calls should be able to overlap
		 * IOs,* thus preventing stalls. so issue the ahead window
		 * immediately.
		 */
		if (req_size >= max)
			make_ahead_window(mapping, filp, ra, 1);

		goto out;
	}

	/*
	 * Now handle the random case:
	 * partial page reads and first access were handled above,
	 * so this must be the next page otherwise it is random
	 */
	if (!sequential) {
		ra_off(ra);
		blockable_page_cache_readahead(mapping, filp, offset,
				 newsize, ra, 1);
		goto out;
	}

	/*
	 * If we get here we are doing sequential IO and this was not the first
	 * occurence (ie we have an existing window)
	 */
	if (ra->ahead_start == 0) {	 /* no ahead window yet */
		if (!make_ahead_window(mapping, filp, ra, 0))
			goto recheck;
	}

	/*
	 * Already have an ahead window, check if we crossed into it.
	 * If so, shift windows and issue a new ahead window.
	 * Only return the #pages that are in the current window, so that
	 * we get called back on the first page of the ahead window which
	 * will allow us to submit more IO.
	 */
	if (ra->prev_index >= ra->ahead_start) {
		ra->start = ra->ahead_start;
		ra->size = ra->ahead_size;
		make_ahead_window(mapping, filp, ra, 0);
recheck:
		/* prev_index shouldn't overrun the ahead window */
		ra->prev_index = min(ra->prev_index,
			ra->ahead_start + ra->ahead_size - 1);
	}

out:
	return ra->prev_index + 1;
}
EXPORT_SYMBOL_GPL(page_cache_readahead);

/*
 * handle_ra_miss() is called when it is known that a page which should have
 * been present in the pagecache (we just did some readahead there) was in fact
 * not found.  This will happen if it was evicted by the VM (readahead
 * thrashing)
 *
 * Turn on the cache miss flag in the RA struct, this will cause the RA code
 * to reduce the RA size on the next read.
 */
void handle_ra_miss(struct address_space *mapping,
		struct file_ra_state *ra, pgoff_t offset)
{
	ra->flags |= RA_FLAG_MISS;
	ra->flags &= ~RA_FLAG_INCACHE;
	ra->cache_hit = 0;
}

/*
 * Given a desired number of PAGE_CACHE_SIZE readahead pages, return a
 * sensible upper limit.
 */
unsigned long max_sane_readahead(unsigned long nr)
{
	return min(nr, (node_page_state(numa_node_id(), NR_INACTIVE)
		+ node_page_state(numa_node_id(), NR_FREE_PAGES)) / 2);
}