page-writeback.c 59.8 KB
Newer Older
Linus Torvalds's avatar
Linus Torvalds committed
1
/*
2
 * mm/page-writeback.c
Linus Torvalds's avatar
Linus Torvalds committed
3
4
 *
 * Copyright (C) 2002, Linus Torvalds.
5
 * Copyright (C) 2007 Red Hat, Inc., Peter Zijlstra <pzijlstr@redhat.com>
Linus Torvalds's avatar
Linus Torvalds committed
6
7
8
9
 *
 * Contains functions related to writing back dirty pages at the
 * address_space level.
 *
10
 * 10Apr2002	Andrew Morton
Linus Torvalds's avatar
Linus Torvalds committed
11
12
13
14
 *		Initial version
 */

#include <linux/kernel.h>
15
#include <linux/export.h>
Linus Torvalds's avatar
Linus Torvalds committed
16
17
18
19
20
21
22
23
24
#include <linux/spinlock.h>
#include <linux/fs.h>
#include <linux/mm.h>
#include <linux/swap.h>
#include <linux/slab.h>
#include <linux/pagemap.h>
#include <linux/writeback.h>
#include <linux/init.h>
#include <linux/backing-dev.h>
25
#include <linux/task_io_accounting_ops.h>
Linus Torvalds's avatar
Linus Torvalds committed
26
27
#include <linux/blkdev.h>
#include <linux/mpage.h>
28
#include <linux/rmap.h>
Linus Torvalds's avatar
Linus Torvalds committed
29
30
31
32
33
34
#include <linux/percpu.h>
#include <linux/notifier.h>
#include <linux/smp.h>
#include <linux/sysctl.h>
#include <linux/cpu.h>
#include <linux/syscalls.h>
Al Viro's avatar
Al Viro committed
35
#include <linux/buffer_head.h> /* __set_page_dirty_buffers */
36
#include <linux/pagevec.h>
37
#include <trace/events/writeback.h>
Linus Torvalds's avatar
Linus Torvalds committed
38

39
40
41
42
43
/*
 * Sleep at most 200ms at a time in balance_dirty_pages().
 */
#define MAX_PAUSE		max(HZ/5, 1)

44
45
46
47
48
/*
 * Estimate write bandwidth at 200ms intervals.
 */
#define BANDWIDTH_INTERVAL	max(HZ/5, 1)

49
50
#define RATELIMIT_CALC_SHIFT	10

Linus Torvalds's avatar
Linus Torvalds committed
51
52
53
54
55
56
57
58
59
/*
 * After a CPU has dirtied this many pages, balance_dirty_pages_ratelimited
 * will look to see if it needs to force writeback or throttling.
 */
static long ratelimit_pages = 32;

/* The following parameters are exported via /proc/sys/vm */

/*
60
 * Start background writeback (via writeback threads) at this percentage
Linus Torvalds's avatar
Linus Torvalds committed
61
 */
62
int dirty_background_ratio = 10;
Linus Torvalds's avatar
Linus Torvalds committed
63

64
65
66
67
68
69
/*
 * dirty_background_bytes starts at 0 (disabled) so that it is a function of
 * dirty_background_ratio * the amount of dirtyable memory
 */
unsigned long dirty_background_bytes;

70
71
72
73
74
75
/*
 * free highmem will not be subtracted from the total free memory
 * for calculating free ratios if vm_highmem_is_dirtyable is true
 */
int vm_highmem_is_dirtyable;

Linus Torvalds's avatar
Linus Torvalds committed
76
77
78
/*
 * The generator of dirty data starts writeback at this percentage
 */
79
int vm_dirty_ratio = 20;
Linus Torvalds's avatar
Linus Torvalds committed
80

81
82
83
84
85
86
/*
 * vm_dirty_bytes starts at 0 (disabled) so that it is a function of
 * vm_dirty_ratio * the amount of dirtyable memory
 */
unsigned long vm_dirty_bytes;

Linus Torvalds's avatar
Linus Torvalds committed
87
/*
88
 * The interval between `kupdate'-style writebacks
Linus Torvalds's avatar
Linus Torvalds committed
89
 */
90
unsigned int dirty_writeback_interval = 5 * 100; /* centiseconds */
Linus Torvalds's avatar
Linus Torvalds committed
91
92

/*
93
 * The longest time for which data is allowed to remain dirty
Linus Torvalds's avatar
Linus Torvalds committed
94
 */
95
unsigned int dirty_expire_interval = 30 * 100; /* centiseconds */
Linus Torvalds's avatar
Linus Torvalds committed
96
97
98
99
100
101
102

/*
 * Flag that makes the machine dump writes/reads and block dirtyings.
 */
int block_dump;

/*
103
104
 * Flag that puts the machine in "laptop mode". Doubles as a timeout in jiffies:
 * a full sync is triggered after this time elapses without any disk activity.
Linus Torvalds's avatar
Linus Torvalds committed
105
106
107
108
109
110
111
 */
int laptop_mode;

EXPORT_SYMBOL(laptop_mode);

/* End of sysctl-exported parameters */

112
unsigned long global_dirty_limit;
Linus Torvalds's avatar
Linus Torvalds committed
113

114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
/*
 * Scale the writeback cache size proportional to the relative writeout speeds.
 *
 * We do this by keeping a floating proportion between BDIs, based on page
 * writeback completions [end_page_writeback()]. Those devices that write out
 * pages fastest will get the larger share, while the slower will get a smaller
 * share.
 *
 * We use page writeout completions because we are interested in getting rid of
 * dirty pages. Having them written out is the primary goal.
 *
 * We introduce a concept of time, a period over which we measure these events,
 * because demand can/will vary over time. The length of this period itself is
 * measured in page writeback completions.
 *
 */
static struct prop_descriptor vm_completions;

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
/*
 * Work out the current dirty-memory clamping and background writeout
 * thresholds.
 *
 * The main aim here is to lower them aggressively if there is a lot of mapped
 * memory around.  To avoid stressing page reclaim with lots of unreclaimable
 * pages.  It is better to clamp down on writers than to start swapping, and
 * performing lots of scanning.
 *
 * We only allow 1/2 of the currently-unmapped memory to be dirtied.
 *
 * We don't permit the clamping level to fall below 5% - that is getting rather
 * excessive.
 *
 * We make sure that the background writeout level is below the adjusted
 * clamping level.
 */
static unsigned long highmem_dirtyable_memory(unsigned long total)
{
#ifdef CONFIG_HIGHMEM
	int node;
	unsigned long x = 0;

	for_each_node_state(node, N_HIGH_MEMORY) {
		struct zone *z =
			&NODE_DATA(node)->node_zones[ZONE_HIGHMEM];

		x += zone_page_state(z, NR_FREE_PAGES) +
160
		     zone_reclaimable_pages(z) - z->dirty_balance_reserve;
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
	}
	/*
	 * Make sure that the number of highmem pages is never larger
	 * than the number of the total dirtyable memory. This can only
	 * occur in very strange VM situations but we want to make sure
	 * that this does not occur.
	 */
	return min(x, total);
#else
	return 0;
#endif
}

/**
 * determine_dirtyable_memory - amount of memory that may be used
 *
 * Returns the numebr of pages that can currently be freed and used
 * by the kernel for direct mappings.
 */
static unsigned long determine_dirtyable_memory(void)
{
	unsigned long x;

184
185
	x = global_page_state(NR_FREE_PAGES) + global_reclaimable_pages() -
	    dirty_balance_reserve;
186
187
188
189
190
191
192

	if (!vm_highmem_is_dirtyable)
		x -= highmem_dirtyable_memory(x);

	return x + 1;	/* Ensure that we never return 0 */
}

193
194
195
196
197
198
199
200
201
/*
 * couple the period to the dirty_ratio:
 *
 *   period/2 ~ roundup_pow_of_two(dirty limit)
 */
static int calc_period_shift(void)
{
	unsigned long dirty_total;

202
203
204
205
206
	if (vm_dirty_bytes)
		dirty_total = vm_dirty_bytes / PAGE_SIZE;
	else
		dirty_total = (vm_dirty_ratio * determine_dirtyable_memory()) /
				100;
207
208
209
210
	return 2 + ilog2(dirty_total - 1);
}

/*
211
 * update the period when the dirty threshold changes.
212
 */
213
214
215
216
static void update_completion_period(void)
{
	int shift = calc_period_shift();
	prop_change_shift(&vm_completions, shift);
217
218

	writeback_set_ratelimit();
219
220
221
}

int dirty_background_ratio_handler(struct ctl_table *table, int write,
222
		void __user *buffer, size_t *lenp,
223
224
225
226
		loff_t *ppos)
{
	int ret;

227
	ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos);
228
229
230
231
232
233
	if (ret == 0 && write)
		dirty_background_bytes = 0;
	return ret;
}

int dirty_background_bytes_handler(struct ctl_table *table, int write,
234
		void __user *buffer, size_t *lenp,
235
236
237
238
		loff_t *ppos)
{
	int ret;

239
	ret = proc_doulongvec_minmax(table, write, buffer, lenp, ppos);
240
241
242
243
244
	if (ret == 0 && write)
		dirty_background_ratio = 0;
	return ret;
}

245
int dirty_ratio_handler(struct ctl_table *table, int write,
246
		void __user *buffer, size_t *lenp,
247
248
249
		loff_t *ppos)
{
	int old_ratio = vm_dirty_ratio;
250
251
	int ret;

252
	ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos);
253
	if (ret == 0 && write && vm_dirty_ratio != old_ratio) {
254
255
256
257
258
259
260
		update_completion_period();
		vm_dirty_bytes = 0;
	}
	return ret;
}

int dirty_bytes_handler(struct ctl_table *table, int write,
261
		void __user *buffer, size_t *lenp,
262
263
		loff_t *ppos)
{
264
	unsigned long old_bytes = vm_dirty_bytes;
265
266
	int ret;

267
	ret = proc_doulongvec_minmax(table, write, buffer, lenp, ppos);
268
269
270
	if (ret == 0 && write && vm_dirty_bytes != old_bytes) {
		update_completion_period();
		vm_dirty_ratio = 0;
271
272
273
274
275
276
277
278
279
280
	}
	return ret;
}

/*
 * Increment the BDI's writeout completion count and the global writeout
 * completion count. Called from test_clear_page_writeback().
 */
static inline void __bdi_writeout_inc(struct backing_dev_info *bdi)
{
281
	__inc_bdi_stat(bdi, BDI_WRITTEN);
282
283
	__prop_inc_percpu_max(&vm_completions, &bdi->completions,
			      bdi->max_prop_frac);
284
285
}

286
287
288
289
290
291
292
293
294
295
void bdi_writeout_inc(struct backing_dev_info *bdi)
{
	unsigned long flags;

	local_irq_save(flags);
	__bdi_writeout_inc(bdi);
	local_irq_restore(flags);
}
EXPORT_SYMBOL_GPL(bdi_writeout_inc);

296
297
298
299
300
301
/*
 * Obtain an accurate fraction of the BDI's portion.
 */
static void bdi_writeout_fraction(struct backing_dev_info *bdi,
		long *numerator, long *denominator)
{
302
	prop_fraction_percpu(&vm_completions, &bdi->completions,
303
304
305
				numerator, denominator);
}

306
/*
307
308
309
 * bdi_min_ratio keeps the sum of the minimum dirty shares of all
 * registered backing devices, which, for obvious reasons, can not
 * exceed 100%.
310
311
312
313
314
315
316
 */
static unsigned int bdi_min_ratio;

int bdi_set_min_ratio(struct backing_dev_info *bdi, unsigned int min_ratio)
{
	int ret = 0;

317
	spin_lock_bh(&bdi_lock);
318
	if (min_ratio > bdi->max_ratio) {
319
		ret = -EINVAL;
320
321
322
323
324
325
326
327
328
	} else {
		min_ratio -= bdi->min_ratio;
		if (bdi_min_ratio + min_ratio < 100) {
			bdi_min_ratio += min_ratio;
			bdi->min_ratio += min_ratio;
		} else {
			ret = -EINVAL;
		}
	}
329
	spin_unlock_bh(&bdi_lock);
330
331
332
333
334
335
336
337
338
339
340

	return ret;
}

int bdi_set_max_ratio(struct backing_dev_info *bdi, unsigned max_ratio)
{
	int ret = 0;

	if (max_ratio > 100)
		return -EINVAL;

341
	spin_lock_bh(&bdi_lock);
342
343
344
345
346
347
	if (bdi->min_ratio > max_ratio) {
		ret = -EINVAL;
	} else {
		bdi->max_ratio = max_ratio;
		bdi->max_prop_frac = (PROP_FRAC_BASE * max_ratio) / 100;
	}
348
	spin_unlock_bh(&bdi_lock);
349
350
351

	return ret;
}
352
EXPORT_SYMBOL(bdi_set_max_ratio);
353

354
355
356
357
358
359
static unsigned long dirty_freerun_ceiling(unsigned long thresh,
					   unsigned long bg_thresh)
{
	return (thresh + bg_thresh) / 2;
}

360
361
362
363
364
static unsigned long hard_dirty_limit(unsigned long thresh)
{
	return max(thresh, global_dirty_limit);
}

365
/*
366
367
368
369
370
371
 * global_dirty_limits - background-writeback and dirty-throttling thresholds
 *
 * Calculate the dirty thresholds based on sysctl parameters
 * - vm.dirty_background_ratio  or  vm.dirty_background_bytes
 * - vm.dirty_ratio             or  vm.dirty_bytes
 * The dirty limits will be lifted by 1/4 for PF_LESS_THROTTLE (ie. nfsd) and
372
 * real-time tasks.
373
 */
374
void global_dirty_limits(unsigned long *pbackground, unsigned long *pdirty)
Linus Torvalds's avatar
Linus Torvalds committed
375
{
376
377
	unsigned long background;
	unsigned long dirty;
378
	unsigned long uninitialized_var(available_memory);
Linus Torvalds's avatar
Linus Torvalds committed
379
380
	struct task_struct *tsk;

381
382
383
	if (!vm_dirty_bytes || !dirty_background_bytes)
		available_memory = determine_dirtyable_memory();

384
385
	if (vm_dirty_bytes)
		dirty = DIV_ROUND_UP(vm_dirty_bytes, PAGE_SIZE);
386
387
	else
		dirty = (vm_dirty_ratio * available_memory) / 100;
Linus Torvalds's avatar
Linus Torvalds committed
388

389
390
391
392
	if (dirty_background_bytes)
		background = DIV_ROUND_UP(dirty_background_bytes, PAGE_SIZE);
	else
		background = (dirty_background_ratio * available_memory) / 100;
Linus Torvalds's avatar
Linus Torvalds committed
393

394
395
	if (background >= dirty)
		background = dirty / 2;
Linus Torvalds's avatar
Linus Torvalds committed
396
397
398
399
400
401
402
	tsk = current;
	if (tsk->flags & PF_LESS_THROTTLE || rt_task(tsk)) {
		background += background / 4;
		dirty += dirty / 4;
	}
	*pbackground = background;
	*pdirty = dirty;
403
	trace_global_dirty_state(background, dirty);
404
}
405

406
/**
407
 * bdi_dirty_limit - @bdi's share of dirty throttling threshold
408
409
 * @bdi: the backing_dev_info to query
 * @dirty: global dirty limit in pages
410
 *
411
412
 * Returns @bdi's dirty limit in pages. The term "dirty" in the context of
 * dirty balancing includes all PG_dirty, PG_writeback and NFS unstable pages.
413
414
415
416
417
418
419
 *
 * Note that balance_dirty_pages() will only seriously take it as a hard limit
 * when sleeping max_pause per page is not enough to keep the dirty pages under
 * control. For example, when the device is completely stalled due to some error
 * conditions, or when there are 1000 dd tasks writing to a slow 10MB/s USB key.
 * In the other normal situations, it acts more gently by throttling the tasks
 * more (rather than completely block them) when the bdi dirty pages go high.
420
 *
421
 * It allocates high/low dirty limits to fast/slow devices, in order to prevent
422
423
424
425
426
427
428
 * - starving fast devices
 * - piling up dirty pages (that will take long time to sync) on slow devices
 *
 * The bdi's share of dirty limit will be adapting to its throughput and
 * bounded by the bdi->min_ratio and/or bdi->max_ratio parameters, if set.
 */
unsigned long bdi_dirty_limit(struct backing_dev_info *bdi, unsigned long dirty)
429
430
431
{
	u64 bdi_dirty;
	long numerator, denominator;
432

433
434
435
436
	/*
	 * Calculate this BDI's share of the dirty ratio.
	 */
	bdi_writeout_fraction(bdi, &numerator, &denominator);
437

438
439
440
	bdi_dirty = (dirty * (100 - bdi_min_ratio)) / 100;
	bdi_dirty *= numerator;
	do_div(bdi_dirty, denominator);
441

442
443
444
445
446
	bdi_dirty += (dirty * bdi->min_ratio) / 100;
	if (bdi_dirty > (dirty * bdi->max_ratio) / 100)
		bdi_dirty = dirty * bdi->max_ratio / 100;

	return bdi_dirty;
Linus Torvalds's avatar
Linus Torvalds committed
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
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
/*
 * Dirty position control.
 *
 * (o) global/bdi setpoints
 *
 * We want the dirty pages be balanced around the global/bdi setpoints.
 * When the number of dirty pages is higher/lower than the setpoint, the
 * dirty position control ratio (and hence task dirty ratelimit) will be
 * decreased/increased to bring the dirty pages back to the setpoint.
 *
 *     pos_ratio = 1 << RATELIMIT_CALC_SHIFT
 *
 *     if (dirty < setpoint) scale up   pos_ratio
 *     if (dirty > setpoint) scale down pos_ratio
 *
 *     if (bdi_dirty < bdi_setpoint) scale up   pos_ratio
 *     if (bdi_dirty > bdi_setpoint) scale down pos_ratio
 *
 *     task_ratelimit = dirty_ratelimit * pos_ratio >> RATELIMIT_CALC_SHIFT
 *
 * (o) global control line
 *
 *     ^ pos_ratio
 *     |
 *     |            |<===== global dirty control scope ======>|
 * 2.0 .............*
 *     |            .*
 *     |            . *
 *     |            .   *
 *     |            .     *
 *     |            .        *
 *     |            .            *
 * 1.0 ................................*
 *     |            .                  .     *
 *     |            .                  .          *
 *     |            .                  .              *
 *     |            .                  .                 *
 *     |            .                  .                    *
 *   0 +------------.------------------.----------------------*------------->
 *           freerun^          setpoint^                 limit^   dirty pages
 *
 * (o) bdi control line
 *
 *     ^ pos_ratio
 *     |
 *     |            *
 *     |              *
 *     |                *
 *     |                  *
 *     |                    * |<=========== span ============>|
 * 1.0 .......................*
 *     |                      . *
 *     |                      .   *
 *     |                      .     *
 *     |                      .       *
 *     |                      .         *
 *     |                      .           *
 *     |                      .             *
 *     |                      .               *
 *     |                      .                 *
 *     |                      .                   *
 *     |                      .                     *
 * 1/4 ...............................................* * * * * * * * * * * *
 *     |                      .                         .
 *     |                      .                           .
 *     |                      .                             .
 *   0 +----------------------.-------------------------------.------------->
 *                bdi_setpoint^                    x_intercept^
 *
 * The bdi control line won't drop below pos_ratio=1/4, so that bdi_dirty can
 * be smoothly throttled down to normal if it starts high in situations like
 * - start writing to a slow SD card and a fast disk at the same time. The SD
 *   card's bdi_dirty may rush to many times higher than bdi_setpoint.
 * - the bdi dirty thresh drops quickly due to change of JBOD workload
 */
static unsigned long bdi_position_ratio(struct backing_dev_info *bdi,
					unsigned long thresh,
					unsigned long bg_thresh,
					unsigned long dirty,
					unsigned long bdi_thresh,
					unsigned long bdi_dirty)
{
	unsigned long write_bw = bdi->avg_write_bandwidth;
	unsigned long freerun = dirty_freerun_ceiling(thresh, bg_thresh);
	unsigned long limit = hard_dirty_limit(thresh);
	unsigned long x_intercept;
	unsigned long setpoint;		/* dirty pages' target balance point */
	unsigned long bdi_setpoint;
	unsigned long span;
	long long pos_ratio;		/* for scaling up/down the rate limit */
	long x;

	if (unlikely(dirty >= limit))
		return 0;

	/*
	 * global setpoint
	 *
	 *                           setpoint - dirty 3
	 *        f(dirty) := 1.0 + (----------------)
	 *                           limit - setpoint
	 *
	 * it's a 3rd order polynomial that subjects to
	 *
	 * (1) f(freerun)  = 2.0 => rampup dirty_ratelimit reasonably fast
	 * (2) f(setpoint) = 1.0 => the balance point
	 * (3) f(limit)    = 0   => the hard limit
	 * (4) df/dx      <= 0	 => negative feedback control
	 * (5) the closer to setpoint, the smaller |df/dx| (and the reverse)
	 *     => fast response on large errors; small oscillation near setpoint
	 */
	setpoint = (freerun + limit) / 2;
	x = div_s64((setpoint - dirty) << RATELIMIT_CALC_SHIFT,
		    limit - setpoint + 1);
	pos_ratio = x;
	pos_ratio = pos_ratio * x >> RATELIMIT_CALC_SHIFT;
	pos_ratio = pos_ratio * x >> RATELIMIT_CALC_SHIFT;
	pos_ratio += 1 << RATELIMIT_CALC_SHIFT;

	/*
	 * We have computed basic pos_ratio above based on global situation. If
	 * the bdi is over/under its share of dirty pages, we want to scale
	 * pos_ratio further down/up. That is done by the following mechanism.
	 */

	/*
	 * bdi setpoint
	 *
	 *        f(bdi_dirty) := 1.0 + k * (bdi_dirty - bdi_setpoint)
	 *
	 *                        x_intercept - bdi_dirty
	 *                     := --------------------------
	 *                        x_intercept - bdi_setpoint
	 *
	 * The main bdi control line is a linear function that subjects to
	 *
	 * (1) f(bdi_setpoint) = 1.0
	 * (2) k = - 1 / (8 * write_bw)  (in single bdi case)
	 *     or equally: x_intercept = bdi_setpoint + 8 * write_bw
	 *
	 * For single bdi case, the dirty pages are observed to fluctuate
	 * regularly within range
	 *        [bdi_setpoint - write_bw/2, bdi_setpoint + write_bw/2]
	 * for various filesystems, where (2) can yield in a reasonable 12.5%
	 * fluctuation range for pos_ratio.
	 *
	 * For JBOD case, bdi_thresh (not bdi_dirty!) could fluctuate up to its
	 * own size, so move the slope over accordingly and choose a slope that
	 * yields 100% pos_ratio fluctuation on suddenly doubled bdi_thresh.
	 */
	if (unlikely(bdi_thresh > thresh))
		bdi_thresh = thresh;
601
602
603
604
605
606
607
	/*
	 * It's very possible that bdi_thresh is close to 0 not because the
	 * device is slow, but that it has remained inactive for long time.
	 * Honour such devices a reasonable good (hopefully IO efficient)
	 * threshold, so that the occasional writes won't be blocked and active
	 * writes can rampup the threshold quickly.
	 */
608
	bdi_thresh = max(bdi_thresh, (limit - dirty) / 8);
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
	/*
	 * scale global setpoint to bdi's:
	 *	bdi_setpoint = setpoint * bdi_thresh / thresh
	 */
	x = div_u64((u64)bdi_thresh << 16, thresh + 1);
	bdi_setpoint = setpoint * (u64)x >> 16;
	/*
	 * Use span=(8*write_bw) in single bdi case as indicated by
	 * (thresh - bdi_thresh ~= 0) and transit to bdi_thresh in JBOD case.
	 *
	 *        bdi_thresh                    thresh - bdi_thresh
	 * span = ---------- * (8 * write_bw) + ------------------- * bdi_thresh
	 *          thresh                            thresh
	 */
	span = (thresh - bdi_thresh + 8 * write_bw) * (u64)x >> 16;
	x_intercept = bdi_setpoint + span;

	if (bdi_dirty < x_intercept - span / 4) {
627
628
		pos_ratio = div_u64(pos_ratio * (x_intercept - bdi_dirty),
				    x_intercept - bdi_setpoint + 1);
629
630
631
	} else
		pos_ratio /= 4;

632
633
634
635
636
637
638
	/*
	 * bdi reserve area, safeguard against dirty pool underrun and disk idle
	 * It may push the desired control point of global dirty pages higher
	 * than setpoint.
	 */
	x_intercept = bdi_thresh / 2;
	if (bdi_dirty < x_intercept) {
639
640
641
		if (bdi_dirty > x_intercept / 8)
			pos_ratio = div_u64(pos_ratio * x_intercept, bdi_dirty);
		else
642
643
644
			pos_ratio *= 8;
	}

645
646
647
	return pos_ratio;
}

648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
static void bdi_update_write_bandwidth(struct backing_dev_info *bdi,
				       unsigned long elapsed,
				       unsigned long written)
{
	const unsigned long period = roundup_pow_of_two(3 * HZ);
	unsigned long avg = bdi->avg_write_bandwidth;
	unsigned long old = bdi->write_bandwidth;
	u64 bw;

	/*
	 * bw = written * HZ / elapsed
	 *
	 *                   bw * elapsed + write_bandwidth * (period - elapsed)
	 * write_bandwidth = ---------------------------------------------------
	 *                                          period
	 */
	bw = written - bdi->written_stamp;
	bw *= HZ;
	if (unlikely(elapsed > period)) {
		do_div(bw, elapsed);
		avg = bw;
		goto out;
	}
	bw += (u64)bdi->write_bandwidth * (period - elapsed);
	bw >>= ilog2(period);

	/*
	 * one more level of smoothing, for filtering out sudden spikes
	 */
	if (avg > old && old >= (unsigned long)bw)
		avg -= (avg - old) >> 3;

	if (avg < old && old <= (unsigned long)bw)
		avg += (old - avg) >> 3;

out:
	bdi->write_bandwidth = bw;
	bdi->avg_write_bandwidth = avg;
}

688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
/*
 * The global dirtyable memory and dirty threshold could be suddenly knocked
 * down by a large amount (eg. on the startup of KVM in a swapless system).
 * This may throw the system into deep dirty exceeded state and throttle
 * heavy/light dirtiers alike. To retain good responsiveness, maintain
 * global_dirty_limit for tracking slowly down to the knocked down dirty
 * threshold.
 */
static void update_dirty_limit(unsigned long thresh, unsigned long dirty)
{
	unsigned long limit = global_dirty_limit;

	/*
	 * Follow up in one step.
	 */
	if (limit < thresh) {
		limit = thresh;
		goto update;
	}

	/*
	 * Follow down slowly. Use the higher one as the target, because thresh
	 * may drop below dirty. This is exactly the reason to introduce
	 * global_dirty_limit which is guaranteed to lie above the dirty pages.
	 */
	thresh = max(thresh, dirty);
	if (limit > thresh) {
		limit -= (limit - thresh) >> 5;
		goto update;
	}
	return;
update:
	global_dirty_limit = limit;
}

static void global_update_bandwidth(unsigned long thresh,
				    unsigned long dirty,
				    unsigned long now)
{
	static DEFINE_SPINLOCK(dirty_lock);
	static unsigned long update_time;

	/*
	 * check locklessly first to optimize away locking for the most time
	 */
	if (time_before(now, update_time + BANDWIDTH_INTERVAL))
		return;

	spin_lock(&dirty_lock);
	if (time_after_eq(now, update_time + BANDWIDTH_INTERVAL)) {
		update_dirty_limit(thresh, dirty);
		update_time = now;
	}
	spin_unlock(&dirty_lock);
}

Wu Fengguang's avatar
Wu Fengguang committed
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
/*
 * Maintain bdi->dirty_ratelimit, the base dirty throttle rate.
 *
 * Normal bdi tasks will be curbed at or below it in long term.
 * Obviously it should be around (write_bw / N) when there are N dd tasks.
 */
static void bdi_update_dirty_ratelimit(struct backing_dev_info *bdi,
				       unsigned long thresh,
				       unsigned long bg_thresh,
				       unsigned long dirty,
				       unsigned long bdi_thresh,
				       unsigned long bdi_dirty,
				       unsigned long dirtied,
				       unsigned long elapsed)
{
759
760
761
	unsigned long freerun = dirty_freerun_ceiling(thresh, bg_thresh);
	unsigned long limit = hard_dirty_limit(thresh);
	unsigned long setpoint = (freerun + limit) / 2;
Wu Fengguang's avatar
Wu Fengguang committed
762
763
764
765
766
767
	unsigned long write_bw = bdi->avg_write_bandwidth;
	unsigned long dirty_ratelimit = bdi->dirty_ratelimit;
	unsigned long dirty_rate;
	unsigned long task_ratelimit;
	unsigned long balanced_dirty_ratelimit;
	unsigned long pos_ratio;
768
769
	unsigned long step;
	unsigned long x;
Wu Fengguang's avatar
Wu Fengguang committed
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818

	/*
	 * The dirty rate will match the writeout rate in long term, except
	 * when dirty pages are truncated by userspace or re-dirtied by FS.
	 */
	dirty_rate = (dirtied - bdi->dirtied_stamp) * HZ / elapsed;

	pos_ratio = bdi_position_ratio(bdi, thresh, bg_thresh, dirty,
				       bdi_thresh, bdi_dirty);
	/*
	 * task_ratelimit reflects each dd's dirty rate for the past 200ms.
	 */
	task_ratelimit = (u64)dirty_ratelimit *
					pos_ratio >> RATELIMIT_CALC_SHIFT;
	task_ratelimit++; /* it helps rampup dirty_ratelimit from tiny values */

	/*
	 * A linear estimation of the "balanced" throttle rate. The theory is,
	 * if there are N dd tasks, each throttled at task_ratelimit, the bdi's
	 * dirty_rate will be measured to be (N * task_ratelimit). So the below
	 * formula will yield the balanced rate limit (write_bw / N).
	 *
	 * Note that the expanded form is not a pure rate feedback:
	 *	rate_(i+1) = rate_(i) * (write_bw / dirty_rate)		     (1)
	 * but also takes pos_ratio into account:
	 *	rate_(i+1) = rate_(i) * (write_bw / dirty_rate) * pos_ratio  (2)
	 *
	 * (1) is not realistic because pos_ratio also takes part in balancing
	 * the dirty rate.  Consider the state
	 *	pos_ratio = 0.5						     (3)
	 *	rate = 2 * (write_bw / N)				     (4)
	 * If (1) is used, it will stuck in that state! Because each dd will
	 * be throttled at
	 *	task_ratelimit = pos_ratio * rate = (write_bw / N)	     (5)
	 * yielding
	 *	dirty_rate = N * task_ratelimit = write_bw		     (6)
	 * put (6) into (1) we get
	 *	rate_(i+1) = rate_(i)					     (7)
	 *
	 * So we end up using (2) to always keep
	 *	rate_(i+1) ~= (write_bw / N)				     (8)
	 * regardless of the value of pos_ratio. As long as (8) is satisfied,
	 * pos_ratio is able to drive itself to 1.0, which is not only where
	 * the dirty count meet the setpoint, but also where the slope of
	 * pos_ratio is most flat and hence task_ratelimit is least fluctuated.
	 */
	balanced_dirty_ratelimit = div_u64((u64)task_ratelimit * write_bw,
					   dirty_rate | 1);

819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
	/*
	 * We could safely do this and return immediately:
	 *
	 *	bdi->dirty_ratelimit = balanced_dirty_ratelimit;
	 *
	 * However to get a more stable dirty_ratelimit, the below elaborated
	 * code makes use of task_ratelimit to filter out sigular points and
	 * limit the step size.
	 *
	 * The below code essentially only uses the relative value of
	 *
	 *	task_ratelimit - dirty_ratelimit
	 *	= (pos_ratio - 1) * dirty_ratelimit
	 *
	 * which reflects the direction and size of dirty position error.
	 */

	/*
	 * dirty_ratelimit will follow balanced_dirty_ratelimit iff
	 * task_ratelimit is on the same side of dirty_ratelimit, too.
	 * For example, when
	 * - dirty_ratelimit > balanced_dirty_ratelimit
	 * - dirty_ratelimit > task_ratelimit (dirty pages are above setpoint)
	 * lowering dirty_ratelimit will help meet both the position and rate
	 * control targets. Otherwise, don't update dirty_ratelimit if it will
	 * only help meet the rate target. After all, what the users ultimately
	 * feel and care are stable dirty rate and small position error.
	 *
	 * |task_ratelimit - dirty_ratelimit| is used to limit the step size
	 * and filter out the sigular points of balanced_dirty_ratelimit. Which
	 * keeps jumping around randomly and can even leap far away at times
	 * due to the small 200ms estimation period of dirty_rate (we want to
	 * keep that period small to reduce time lags).
	 */
	step = 0;
	if (dirty < setpoint) {
		x = min(bdi->balanced_dirty_ratelimit,
			 min(balanced_dirty_ratelimit, task_ratelimit));
		if (dirty_ratelimit < x)
			step = x - dirty_ratelimit;
	} else {
		x = max(bdi->balanced_dirty_ratelimit,
			 max(balanced_dirty_ratelimit, task_ratelimit));
		if (dirty_ratelimit > x)
			step = dirty_ratelimit - x;
	}

	/*
	 * Don't pursue 100% rate matching. It's impossible since the balanced
	 * rate itself is constantly fluctuating. So decrease the track speed
	 * when it gets close to the target. Helps eliminate pointless tremors.
	 */
	step >>= dirty_ratelimit / (2 * step + 1);
	/*
	 * Limit the tracking speed to avoid overshooting.
	 */
	step = (step + 7) / 8;

	if (dirty_ratelimit < balanced_dirty_ratelimit)
		dirty_ratelimit += step;
	else
		dirty_ratelimit -= step;

	bdi->dirty_ratelimit = max(dirty_ratelimit, 1UL);
	bdi->balanced_dirty_ratelimit = balanced_dirty_ratelimit;
884
885

	trace_bdi_dirty_ratelimit(bdi, dirty_rate, task_ratelimit);
Wu Fengguang's avatar
Wu Fengguang committed
886
887
}

888
void __bdi_update_bandwidth(struct backing_dev_info *bdi,
889
			    unsigned long thresh,
890
			    unsigned long bg_thresh,
891
892
893
			    unsigned long dirty,
			    unsigned long bdi_thresh,
			    unsigned long bdi_dirty,
894
895
896
897
			    unsigned long start_time)
{
	unsigned long now = jiffies;
	unsigned long elapsed = now - bdi->bw_time_stamp;
Wu Fengguang's avatar
Wu Fengguang committed
898
	unsigned long dirtied;
899
900
901
902
903
904
905
906
	unsigned long written;

	/*
	 * rate-limit, only update once every 200ms.
	 */
	if (elapsed < BANDWIDTH_INTERVAL)
		return;

Wu Fengguang's avatar
Wu Fengguang committed
907
	dirtied = percpu_counter_read(&bdi->bdi_stat[BDI_DIRTIED]);
908
909
910
911
912
913
914
915
916
	written = percpu_counter_read(&bdi->bdi_stat[BDI_WRITTEN]);

	/*
	 * Skip quiet periods when disk bandwidth is under-utilized.
	 * (at least 1s idle time between two flusher runs)
	 */
	if (elapsed > HZ && time_before(bdi->bw_time_stamp, start_time))
		goto snapshot;

Wu Fengguang's avatar
Wu Fengguang committed
917
	if (thresh) {
918
		global_update_bandwidth(thresh, dirty, now);
Wu Fengguang's avatar
Wu Fengguang committed
919
920
921
922
		bdi_update_dirty_ratelimit(bdi, thresh, bg_thresh, dirty,
					   bdi_thresh, bdi_dirty,
					   dirtied, elapsed);
	}
923
924
925
	bdi_update_write_bandwidth(bdi, elapsed, written);

snapshot:
Wu Fengguang's avatar
Wu Fengguang committed
926
	bdi->dirtied_stamp = dirtied;
927
928
929
930
931
	bdi->written_stamp = written;
	bdi->bw_time_stamp = now;
}

static void bdi_update_bandwidth(struct backing_dev_info *bdi,
932
				 unsigned long thresh,
933
				 unsigned long bg_thresh,
934
935
936
				 unsigned long dirty,
				 unsigned long bdi_thresh,
				 unsigned long bdi_dirty,
937
938
939
940
941
				 unsigned long start_time)
{
	if (time_is_after_eq_jiffies(bdi->bw_time_stamp + BANDWIDTH_INTERVAL))
		return;
	spin_lock(&bdi->wb.list_lock);
942
943
	__bdi_update_bandwidth(bdi, thresh, bg_thresh, dirty,
			       bdi_thresh, bdi_dirty, start_time);
944
945
946
	spin_unlock(&bdi->wb.list_lock);
}

947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
/*
 * After a task dirtied this many pages, balance_dirty_pages_ratelimited_nr()
 * will look to see if it needs to start dirty throttling.
 *
 * If dirty_poll_interval is too low, big NUMA machines will call the expensive
 * global_page_state() too often. So scale it near-sqrt to the safety margin
 * (the number of pages we may dirty without exceeding the dirty limits).
 */
static unsigned long dirty_poll_interval(unsigned long dirty,
					 unsigned long thresh)
{
	if (thresh > dirty)
		return 1UL << (ilog2(thresh - dirty) >> 1);

	return 1;
}

964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
static unsigned long bdi_max_pause(struct backing_dev_info *bdi,
				   unsigned long bdi_dirty)
{
	unsigned long bw = bdi->avg_write_bandwidth;
	unsigned long hi = ilog2(bw);
	unsigned long lo = ilog2(bdi->dirty_ratelimit);
	unsigned long t;

	/* target for 20ms max pause on 1-dd case */
	t = HZ / 50;

	/*
	 * Scale up pause time for concurrent dirtiers in order to reduce CPU
	 * overheads.
	 *
	 * (N * 20ms) on 2^N concurrent tasks.
	 */
	if (hi > lo)
		t += (hi - lo) * (20 * HZ) / 1024;

	/*
	 * Limit pause time for small memory systems. If sleeping for too long
	 * time, a small pool of dirty/writeback pages may go empty and disk go
	 * idle.
	 *
	 * 8 serves as the safety ratio.
	 */
991
	t = min(t, bdi_dirty * HZ / (8 * bw + 1));
992
993
994
995
996
997
998
999

	/*
	 * The pause time will be settled within range (max_pause/4, max_pause).
	 * Apply a minimal value of 4 to get a non-zero max_pause/4.
	 */
	return clamp_val(t, 4, MAX_PAUSE);
}

Linus Torvalds's avatar
Linus Torvalds committed
1000
1001
1002
/*
 * balance_dirty_pages() must be called by processes which are generating dirty
 * data.  It looks at the number of dirty pages in the machine and will force
1003
 * the caller to wait once crossing the (background_thresh + dirty_thresh) / 2.
1004
1005
 * If we're over `background_thresh' then the writeback threads are woken to
 * perform some writeout.
Linus Torvalds's avatar
Linus Torvalds committed
1006
 */
1007
static void balance_dirty_pages(struct address_space *mapping,
1008
				unsigned long pages_dirtied)
Linus Torvalds's avatar
Linus Torvalds committed
1009
{
1010
1011
	unsigned long nr_reclaimable;	/* = file_dirty + unstable_nfs */
	unsigned long bdi_reclaimable;
1012
1013
	unsigned long nr_dirty;  /* = file_dirty + writeback + unstable_nfs */
	unsigned long bdi_dirty;
1014
	unsigned long freerun;
1015
1016
1017
	unsigned long background_thresh;
	unsigned long dirty_thresh;
	unsigned long bdi_thresh;
1018
	long pause = 0;
1019
	long uninitialized_var(max_pause);
1020
	bool dirty_exceeded = false;
1021
	unsigned long task_ratelimit;
1022
	unsigned long uninitialized_var(dirty_ratelimit);
1023
	unsigned long pos_ratio;
Linus Torvalds's avatar
Linus Torvalds committed
1024
	struct backing_dev_info *bdi = mapping->backing_dev_info;
1025
	unsigned long start_time = jiffies;
Linus Torvalds's avatar
Linus Torvalds committed
1026
1027

	for (;;) {
1028
1029
1030
1031
1032
1033
		/*
		 * Unstable writes are a feature of certain networked
		 * filesystems (i.e. NFS) in which data may have been
		 * written to the server's write cache, but has not yet
		 * been flushed to permanent storage.
		 */
1034
1035
		nr_reclaimable = global_page_state(NR_FILE_DIRTY) +
					global_page_state(NR_UNSTABLE_NFS);
1036
		nr_dirty = nr_reclaimable + global_page_state(NR_WRITEBACK);
1037

1038
1039
1040
1041
1042
1043
1044
		global_dirty_limits(&background_thresh, &dirty_thresh);

		/*
		 * Throttle it only when the background writeback cannot
		 * catch-up. This avoids (excessively) small writeouts
		 * when the bdi limits are ramping up.
		 */
1045
1046
1047
		freerun = dirty_freerun_ceiling(dirty_thresh,
						background_thresh);
		if (nr_dirty <= freerun)
1048
1049
			break;

1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
		if (unlikely(!writeback_in_progress(bdi)))
			bdi_start_background_writeback(bdi);

		/*
		 * bdi_thresh is not treated as some limiting factor as
		 * dirty_thresh, due to reasons
		 * - in JBOD setup, bdi_thresh can fluctuate a lot
		 * - in a system with HDD and USB key, the USB key may somehow
		 *   go into state (bdi_dirty >> bdi_thresh) either because
		 *   bdi_dirty starts high, or because bdi_thresh drops low.
		 *   In this case we don't want to hard throttle the USB key
		 *   dirtiers for 100 seconds until bdi_dirty drops under
		 *   bdi_thresh. Instead the auxiliary bdi control line in
		 *   bdi_position_ratio() will let the dirtier task progress
		 *   at some rate <= (write_bw / 2) for bringing down bdi_dirty.
		 */
1066
1067
		bdi_thresh = bdi_dirty_limit(bdi, dirty_thresh);

1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
		/*
		 * In order to avoid the stacked BDI deadlock we need
		 * to ensure we accurately count the 'dirty' pages when
		 * the threshold is low.
		 *
		 * Otherwise it would be possible to get thresh+n pages
		 * reported dirty, even though there are thresh-m pages
		 * actually dirty; with m+n sitting in the percpu
		 * deltas.
		 */
1078
1079
1080
		if (bdi_thresh < 2 * bdi_stat_error(bdi)) {
			bdi_reclaimable = bdi_stat_sum(bdi, BDI_RECLAIMABLE);
			bdi_dirty = bdi_reclaimable +
1081
				    bdi_stat_sum(bdi, BDI_WRITEBACK);
1082
		} else {
1083
1084
			bdi_reclaimable = bdi_stat(bdi, BDI_RECLAIMABLE);
			bdi_dirty = bdi_reclaimable +
1085
				    bdi_stat(bdi, BDI_WRITEBACK);
1086
		}
1087

1088
		dirty_exceeded = (bdi_dirty > bdi_thresh) ||
1089
				  (nr_dirty > dirty_thresh);
1090
		if (dirty_exceeded && !bdi->dirty_exceeded)
1091
			bdi->dirty_exceeded = 1;
Linus Torvalds's avatar
Linus Torvalds committed
1092

1093
1094
1095
		bdi_update_bandwidth(bdi, dirty_thresh, background_thresh,
				     nr_dirty, bdi_thresh, bdi_dirty,
				     start_time);
1096

1097
1098
		max_pause = bdi_max_pause(bdi, bdi_dirty);

1099
1100
1101
1102
		dirty_ratelimit = bdi->dirty_ratelimit;
		pos_ratio = bdi_position_ratio(bdi, dirty_thresh,
					       background_thresh, nr_dirty,
					       bdi_thresh, bdi_dirty);
1103
1104
1105
		task_ratelimit = ((u64)dirty_ratelimit * pos_ratio) >>
							RATELIMIT_CALC_SHIFT;
		if (unlikely(task_ratelimit == 0)) {
1106
			pause = max_pause;
1107
			goto pause;
1108
		}
1109
		pause = HZ * pages_dirtied / task_ratelimit;
1110
		if (unlikely(pause <= 0)) {
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
			trace_balance_dirty_pages(bdi,
						  dirty_thresh,
						  background_thresh,
						  nr_dirty,
						  bdi_thresh,
						  bdi_dirty,
						  dirty_ratelimit,
						  task_ratelimit,
						  pages_dirtied,
						  pause,
						  start_time);
1122
1123
			pause = 1; /* avoid resetting nr_dirtied_pause below */
			break;
1124
		}
1125
		pause = min(pause, max_pause);
1126
1127

pause:
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
		trace_balance_dirty_pages(bdi,
					  dirty_thresh,
					  background_thresh,
					  nr_dirty,
					  bdi_thresh,
					  bdi_dirty,
					  dirty_ratelimit,
					  task_ratelimit,
					  pages_dirtied,
					  pause,
					  start_time);
1139
		__set_current_state(TASK_KILLABLE);
1140
		io_schedule_timeout(pause);
1141

1142
		/*
1143
1144
		 * This is typically equal to (nr_dirty < dirty_thresh) and can
		 * also keep "1000+ dd on a slow USB stick" under control.
1145
		 */
1146
		if (task_ratelimit)
1147
			break;
1148

1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
		/*
		 * In the case of an unresponding NFS server and the NFS dirty
		 * pages exceeds dirty_thresh, give the other good bdi's a pipe
		 * to go through, so that tasks on them still remain responsive.
		 *
		 * In theory 1 page is enough to keep the comsumer-producer
		 * pipe going: the flusher cleans 1 page => the task dirties 1
		 * more page. However bdi_dirty has accounting errors.  So use
		 * the larger and more IO friendly bdi_stat_error.
		 */
		if (bdi_dirty <= bdi_stat_error(bdi))
			break;

1162
1163
		if (fatal_signal_pending(current))
			break;
Linus Torvalds's avatar
Linus Torvalds committed
1164
1165
	}

1166
	if (!dirty_exceeded && bdi->dirty_exceeded)
1167
		bdi->dirty_exceeded = 0;
Linus Torvalds's avatar
Linus Torvalds committed
1168

1169
	current->nr_dirtied = 0;
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
	if (pause == 0) { /* in freerun area */
		current->nr_dirtied_pause =
				dirty_poll_interval(nr_dirty, dirty_thresh);
	} else if (pause <= max_pause / 4 &&
		   pages_dirtied >= current->nr_dirtied_pause) {
		current->nr_dirtied_pause = clamp_val(
					dirty_ratelimit * (max_pause / 2) / HZ,
					pages_dirtied + pages_dirtied / 8,
					pages_dirtied * 4);
	} else if (pause >= max_pause) {
		current->nr_dirtied_pause = 1 | clamp_val(
					dirty_ratelimit * (max_pause / 2) / HZ,
					pages_dirtied / 4,
					pages_dirtied - pages_dirtied / 8);
	}
1185

Linus Torvalds's avatar
Linus Torvalds committed
1186
	if (writeback_in_progress(bdi))
1187
		return;
Linus Torvalds's avatar
Linus Torvalds committed
1188
1189
1190
1191
1192
1193
1194
1195
1196

	/*
	 * In laptop mode, we wait until hitting the higher threshold before
	 * starting background writeout, and then write out all the way down
	 * to the lower threshold.  So slow writers cause minimal disk activity.
	 *
	 * In normal mode, we start background writeout at the lower
	 * background_thresh, to keep the amount of dirty memory low.
	 */
1197
1198
1199
1200
	if (laptop_mode)
		return;

	if (nr_reclaimable > background_thresh)
1201
		bdi_start_background_writeback(bdi);
Linus Torvalds's avatar
Linus Torvalds committed
1202
1203
}

1204
void set_page_dirty_balance(struct page *page, int page_mkwrite)
1205
{
1206
	if (set_page_dirty(page) || page_mkwrite) {
1207
1208
1209
1210
1211
1212
1213
		struct address_space *mapping = page_mapping(page);

		if (mapping)
			balance_dirty_pages_ratelimited(mapping);
	}
}

1214
static DEFINE_PER_CPU(int, bdp_ratelimits);
1215

Linus Torvalds's avatar
Linus Torvalds committed
1216
/**
1217
 * balance_dirty_pages_ratelimited_nr - balance dirty memory state
1218
 * @mapping: address_space which was dirtied
1219
 * @nr_pages_dirtied: number of pages which the caller has just dirtied
Linus Torvalds's avatar
Linus Torvalds committed
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
 *
 * Processes which are dirtying memory should call in here once for each page
 * which was newly dirtied.  The function will periodically check the system's
 * dirty state and will initiate writeback if needed.
 *
 * On really big machines, get_writeback_state is expensive, so try to avoid
 * calling it too often (ratelimiting).  But once we're over the dirty memory
 * limit we decrease the ratelimiting by a lot, to prevent individual processes
 * from overshooting the limit by (ratelimit_pages) each.
 */
1230
1231
void balance_dirty_pages_ratelimited_nr(struct address_space *mapping,
					unsigned long nr_pages_dirtied)
Linus Torvalds's avatar
Linus Torvalds committed
1232
{
1233
	struct backing_dev_info *bdi = mapping->backing_dev_info;
1234
1235
	int ratelimit;
	int *p;
Linus Torvalds's avatar
Linus Torvalds committed
1236

1237
1238
1239
	if (!bdi_cap_account_dirty(bdi))
		return;

1240
1241
1242
1243
1244
	ratelimit = current->nr_dirtied_pause;
	if (bdi->dirty_exceeded)
		ratelimit = min(ratelimit, 32 >> (PAGE_SHIFT - 10));

	current->nr_dirtied += nr_pages_dirtied;
Linus Torvalds's avatar
Linus Torvalds committed
1245

1246
	preempt_disable();
Linus Torvalds's avatar
Linus Torvalds committed
1247
	/*
1248
1249
1250
1251
	 * This prevents one CPU to accumulate too many dirtied pages without
	 * calling into balance_dirty_pages(), which can happen when there are
	 * 1000+ tasks, all of them start dirtying pages at exactly the same
	 * time, hence all honoured too large initial task->nr_dirtied_pause.
Linus Torvalds's avatar
Linus Torvalds committed
1252
	 */
1253
	p =  &__get_cpu_var(bdp_ratelimits);
1254
	if (unlikely(current->nr_dirtied >= ratelimit))
1255
		*p = 0;
1256
1257
1258
1259
1260
1261
	else {
		*p += nr_pages_dirtied;
		if (unlikely(*p >= ratelimit_pages)) {
			*p = 0;
			ratelimit = 0;
		}
Linus Torvalds's avatar
Linus Torvalds committed
1262
	}
1263
	preempt_enable();
1264
1265
1266

	if (unlikely(current->nr_dirtied >= ratelimit))
		balance_dirty_pages(mapping, current->nr_dirtied);
Linus Torvalds's avatar
Linus Torvalds committed
1267
}
1268
EXPORT_SYMBOL(balance_dirty_pages_ratelimited_nr);
Linus Torvalds's avatar
Linus Torvalds committed
1269

1270
void throttle_vm_writeout(gfp_t gfp_mask)
Linus Torvalds's avatar
Linus Torvalds committed
1271
{
1272
1273
	unsigned long background_thresh;
	unsigned long dirty_thresh;
Linus Torvalds's avatar
Linus Torvalds committed
1274
1275

        for ( ; ; ) {
1276
		global_dirty_limits(&background_thresh, &dirty_thresh);
Linus Torvalds's avatar
Linus Torvalds committed
1277
1278
1279
1280
1281
1282
1283

                /*
                 * Boost the allowable dirty threshold a bit for page
                 * allocators so they don't get DoS'ed by heavy writers
                 */
                dirty_thresh += dirty_thresh / 10;      /* wheeee... */

1284
1285
1286
                if (global_page_state(NR_UNSTABLE_NFS) +
			global_page_state(NR_WRITEBACK) <= dirty_thresh)
                        	break;
1287
                congestion_wait(BLK_RW_ASYNC, HZ/10);
1288
1289
1290
1291
1292
1293
1294
1295

		/*
		 * The caller might hold locks which can prevent IO completion
		 * or progress in the filesystem.  So we cannot just sit here
		 * waiting for IO to complete.
		 */
		if ((gfp_mask & (__GFP_FS|__GFP_IO)) != (__GFP_FS|__GFP_IO))
			break;
Linus Torvalds's avatar
Linus Torvalds committed
1296
1297
1298
1299
1300
1301
1302
        }
}

/*
 * sysctl handler for /proc/sys/vm/dirty_writeback_centisecs
 */
int dirty_writeback_centisecs_handler(ctl_table *table, int write,
1303
	void __user *buffer, size_t *length, loff_t *ppos)
Linus Torvalds's avatar
Linus Torvalds committed
1304
{
1305
	proc_dointvec(table, write, buffer, length, ppos);
1306
	bdi_arm_supers_timer();
Linus Torvalds's avatar
Linus Torvalds committed
1307
1308
1309
	return 0;
}

1310
#ifdef CONFIG_BLOCK
1311
void laptop_mode_timer_fn(unsigned long data)
Linus Torvalds's avatar
Linus Torvalds committed
1312
{
1313
1314
1315
	struct request_queue *q = (struct request_queue *)data;
	int nr_pages = global_page_state(NR_FILE_DIRTY) +
		global_page_state(NR_UNSTABLE_NFS);
Linus Torvalds's avatar
Linus Torvalds committed
1316

1317
1318
1319
1320
1321
	/*
	 * We want to write everything out, not just down to the dirty
	 * threshold
	 */
	if (bdi_has_dirty_io(&q->backing_dev_info))
1322
1323
		bdi_start_writeback(&q->backing_dev_info, nr_pages,
					WB_REASON_LAPTOP_TIMER);
Linus Torvalds's avatar
Linus Torvalds committed
1324
1325
1326
1327
1328
1329
1330
}

/*
 * We've spun up the disk and we're in laptop mode: schedule writeback
 * of all dirty data a few seconds from now.  If the flush is already scheduled
 * then push it back - the user is still using the disk.
 */
1331
void laptop_io_completion(struct backing_dev_info *info)
Linus Torvalds's avatar
Linus Torvalds committed
1332
{
1333
	mod_timer(&info->laptop_mode_wb_timer, jiffies + laptop_mode);
Linus Torvalds's avatar
Linus Torvalds committed
1334
1335
1336
1337
1338
1339
1340
1341
1342
}

/*
 * We're in laptop mode and we've just synced. The sync's writes will have
 * caused another writeback to be scheduled by laptop_io_completion.
 * Nothing needs to be written back anymore, so we unschedule the writeback.
 */
void laptop_sync_completion(void)
{
1343
1344
1345
1346
1347
1348
1349
1350
	struct backing_dev_info *bdi;

	rcu_read_lock();

	list_for_each_entry_rcu(bdi, &bdi_list, bdi_list)
		del_timer(&bdi->laptop_mode_wb_timer);

	rcu_read_unlock();
Linus Torvalds's avatar
Linus Torvalds committed
1351
}
1352
#endif
Linus Torvalds's avatar
Linus Torvalds committed
1353
1354
1355
1356
1357
1358
1359
1360
1361

/*
 * If ratelimit_pages is too high then we can get into dirty-data overload
 * if a large number of processes all perform writes at the same time.
 * If it is too low then SMP machines will call the (expensive)
 * get_writeback_state too often.
 *
 * Here we set ratelimit_pages to a level which ensures that when all CPUs are
 * dirtying in parallel, we cannot go more than 3% (1/32) over the dirty memory
1362
 * thresholds.
Linus Torvalds's avatar
Linus Torvalds committed
1363
1364
 */

1365
void writeback_set_ratelimit(void)
Linus Torvalds's avatar
Linus Torvalds committed
1366
{
1367
1368
1369
1370
	unsigned long background_thresh;
	unsigned long dirty_thresh;
	global_dirty_limits(&background_thresh, &dirty_thresh);
	ratelimit_pages = dirty_thresh / (num_online_cpus() * 32);
Linus Torvalds's avatar
Linus Torvalds committed
1371
1372
1373
1374
	if (ratelimit_pages < 16)
		ratelimit_pages = 16;
}

1375
static int __cpuinit
Linus Torvalds's avatar
Linus Torvalds committed
1376
1377
ratelimit_handler(struct notifier_block *self, unsigned long u, void *v)
{
1378
	writeback_set_ratelimit();
1379
	return NOTIFY_DONE;
Linus Torvalds's avatar
Linus Torvalds committed
1380
1381
}

1382
static struct notifier_block __cpuinitdata ratelimit_nb = {
Linus Torvalds's avatar
Linus Torvalds committed
1383
1384
1385
1386
1387
	.notifier_call	= ratelimit_handler,
	.next		= NULL,
};

/*
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
 * Called early on to tune the page writeback dirty limits.
 *
 * We used to scale dirty pages according to how total memory
 * related to pages that could be allocated for buffers (by
 * comparing nr_free_buffer_pages() to vm_total_pages.
 *
 * However, that was when we used "dirty_ratio" to scale with
 * all memory, and we don't do that any more. "dirty_ratio"
 * is now applied to total non-HIGHPAGE memory (by subtracting
 * totalhigh_pages from vm_total_pages), and as such we can't
 * get into the old insane situation any more where we had
 * large amounts of dirty pages compared to a small amount of
 * non-HIGHMEM memory.
 *
 * But we might still want to scale the dirty_ratio by how
 * much memory the box has..
Linus Torvalds's avatar
Linus Torvalds committed
1404
1405
1406
 */
void __init page_writeback_init(void)
{
1407
1408
	int shift;

1409
	writeback_set_ratelimit();
Linus Torvalds's avatar
Linus Torvalds committed
1410
	register_cpu_notifier(&ratelimit_nb);
1411
1412
1413

	shift = calc_period_shift();
	prop_descriptor_init(&vm_completions, shift);
Linus Torvalds's avatar
Linus Torvalds committed
1414
1415
}

1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
/**
 * tag_pages_for_writeback - tag pages to be written by write_cache_pages
 * @mapping: address space structure to write
 * @start: starting page index
 * @end: ending page index (inclusive)
 *
 * This function scans the page range from @start to @end (inclusive) and tags
 * all pages that have DIRTY tag set with a special TOWRITE tag. The idea is
 * that write_cache_pages (or whoever calls this function) will then use
 * TOWRITE tag to identify pages eligible for writeback.  This mechanism is
 * used to avoid livelocking of writeback by a process steadily creating new
 * dirty pages in the file (thus it is important for this function to be quick
 * so that it can tag pages faster than a dirtying process can create them).
 */
/*
 * We tag pages in batches of WRITEBACK_TAG_BATCH to reduce tree_lock latency.
 */
void tag_pages_for_writeback(struct address_space *mapping,
			     pgoff_t start, pgoff_t end)
{
Randy Dunlap's avatar
Randy Dunlap committed
1436
#define WRITEBACK_TAG_BATCH 4096
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
	unsigned long tagged;

	do {
		spin_lock_irq(&mapping->tree_lock);
		tagged = radix_tree_range_tag_if_tagged(&mapping->page_tree,
				&start, end, WRITEBACK_TAG_BATCH,
				PAGECACHE_TAG_DIRTY, PAGECACHE_TAG_TOWRITE);
		spin_unlock_irq(&mapping->tree_lock);
		WARN_ON_ONCE(tagged > WRITEBACK_TAG_BATCH);
		cond_resched();
1447
1448
		/* We check 'start' to handle wrapping when end == ~0UL */
	} while (tagged >= WRITEBACK_TAG_BATCH && start);
1449
1450
1451
}
EXPORT_SYMBOL(tag_pages_for_writeback);

1452
/**
1453
 * write_cache_pages - walk the list of dirty pages of the given address space and write all of them.
1454
1455
 * @mapping: address space structure to write
 * @wbc: subtract the number of written pages from *@wbc->nr_to_write
1456
1457
 * @writepage: function called for each page
 * @data: data passed to writepage function
1458
 *
1459
 * If a page is already under I/O, write_cache_pages() skips it, even
1460
1461
1462
1463
1464
1465
 * if it's dirty.  This is desirable behaviour for memory-cleaning writeback,
 * but it is INCORRECT for data-integrity system calls such as fsync().  fsync()
 * and msync() need to guarantee that all the data which was dirty at the time
 * the call was made get new I/O started against them.  If wbc->sync_mode is
 * WB_SYNC_ALL then we were called for data integrity and we must wait for
 * existing IO to complete.
1466
1467
1468
1469
1470
1471
1472
 *
 * To avoid livelocks (when other process dirties new pages), we first tag
 * pages which should be written back with TOWRITE tag and only then start
 * writing them. For data-integrity sync we have to be careful so that we do
 * not miss some pages (e.g., because some other process has cleared TOWRITE
 * tag we set). The rule we follow is that TOWRITE tag can be cleared only
 * by the process clearing the DIRTY tag (and submitting the page for IO).
1473
 */
1474
1475
1476
int write_cache_pages(struct address_space *mapping,
		      struct writeback_control *wbc, writepage_t writepage,
		      void *data)
1477
1478
1479
1480
1481
{
	int ret = 0;
	int done = 0;
	struct pagevec pvec;
	int nr_pages;
Nick Piggin's avatar
Nick Piggin committed
1482
	pgoff_t uninitialized_var(writeback_index);
1483
1484
	pgoff_t index;
	pgoff_t end;		/* Inclusive */
1485
	pgoff_t done_index;
Nick Piggin's avatar
Nick Piggin committed
1486
	int cycled;
1487
	int range_whole = 0;
1488
	int tag;
1489
1490
1491

	pagevec_init(&pvec, 0);
	if (wbc->range_cyclic) {
Nick Piggin's avatar
Nick Piggin committed
1492
1493
1494
1495
1496
1497
		writeback_index = mapping->writeback_index; /* prev offset */
		index = writeback_index;
		if (index == 0)
			cycled = 1;
		else
			cycled = 0;