blk-cgroup.c 27.0 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13
/*
 * Common Block IO controller cgroup interface
 *
 * Based on ideas and code from CFQ, CFS and BFQ:
 * Copyright (C) 2003 Jens Axboe <axboe@kernel.dk>
 *
 * Copyright (C) 2008 Fabio Checconi <fabio@gandalf.sssup.it>
 *		      Paolo Valente <paolo.valente@unimore.it>
 *
 * Copyright (C) 2009 Vivek Goyal <vgoyal@redhat.com>
 * 	              Nauman Rafique <nauman@google.com>
 */
#include <linux/ioprio.h>
14 15
#include <linux/seq_file.h>
#include <linux/kdev_t.h>
16
#include <linux/module.h>
17
#include <linux/err.h>
18
#include <linux/blkdev.h>
19
#include <linux/slab.h>
20
#include "blk-cgroup.h"
21
#include <linux/genhd.h>
22

23 24
#define MAX_KEY_LEN 100

25 26
static DEFINE_SPINLOCK(blkio_list_lock);
static LIST_HEAD(blkio_list);
27

28
struct blkio_cgroup blkio_root_cgroup = { .weight = 2*BLKIO_WEIGHT_DEFAULT };
29 30
EXPORT_SYMBOL_GPL(blkio_root_cgroup);

B
Ben Blum 已提交
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
static struct cgroup_subsys_state *blkiocg_create(struct cgroup_subsys *,
						  struct cgroup *);
static int blkiocg_can_attach(struct cgroup_subsys *, struct cgroup *,
			      struct task_struct *, bool);
static void blkiocg_attach(struct cgroup_subsys *, struct cgroup *,
			   struct cgroup *, struct task_struct *, bool);
static void blkiocg_destroy(struct cgroup_subsys *, struct cgroup *);
static int blkiocg_populate(struct cgroup_subsys *, struct cgroup *);

struct cgroup_subsys blkio_subsys = {
	.name = "blkio",
	.create = blkiocg_create,
	.can_attach = blkiocg_can_attach,
	.attach = blkiocg_attach,
	.destroy = blkiocg_destroy,
	.populate = blkiocg_populate,
#ifdef CONFIG_BLK_CGROUP
	/* note: blkio_subsys_id is otherwise defined in blk-cgroup.h */
	.subsys_id = blkio_subsys_id,
#endif
	.use_id = 1,
	.module = THIS_MODULE,
};
EXPORT_SYMBOL_GPL(blkio_subsys);

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
static inline void blkio_policy_insert_node(struct blkio_cgroup *blkcg,
					    struct blkio_policy_node *pn)
{
	list_add(&pn->node, &blkcg->policy_list);
}

/* Must be called with blkcg->lock held */
static inline void blkio_policy_delete_node(struct blkio_policy_node *pn)
{
	list_del(&pn->node);
}

/* Must be called with blkcg->lock held */
static struct blkio_policy_node *
blkio_policy_search_node(const struct blkio_cgroup *blkcg, dev_t dev)
{
	struct blkio_policy_node *pn;

	list_for_each_entry(pn, &blkcg->policy_list, node) {
		if (pn->dev == dev)
			return pn;
	}

	return NULL;
}

82 83 84 85 86
struct blkio_cgroup *cgroup_to_blkio_cgroup(struct cgroup *cgroup)
{
	return container_of(cgroup_subsys_state(cgroup, blkio_subsys_id),
			    struct blkio_cgroup, css);
}
87
EXPORT_SYMBOL_GPL(cgroup_to_blkio_cgroup);
88

89 90 91 92 93 94
void blkio_group_init(struct blkio_group *blkg)
{
	spin_lock_init(&blkg->stats_lock);
}
EXPORT_SYMBOL_GPL(blkio_group_init);

95 96 97 98
/*
 * Add to the appropriate stat variable depending on the request type.
 * This should be called with the blkg->stats_lock held.
 */
99 100
static void blkio_add_stat(uint64_t *stat, uint64_t add, bool direction,
				bool sync)
101
{
102 103
	if (direction)
		stat[BLKIO_STAT_WRITE] += add;
104
	else
105 106 107
		stat[BLKIO_STAT_READ] += add;
	if (sync)
		stat[BLKIO_STAT_SYNC] += add;
108
	else
109
		stat[BLKIO_STAT_ASYNC] += add;
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
/*
 * Decrements the appropriate stat variable if non-zero depending on the
 * request type. Panics on value being zero.
 * This should be called with the blkg->stats_lock held.
 */
static void blkio_check_and_dec_stat(uint64_t *stat, bool direction, bool sync)
{
	if (direction) {
		BUG_ON(stat[BLKIO_STAT_WRITE] == 0);
		stat[BLKIO_STAT_WRITE]--;
	} else {
		BUG_ON(stat[BLKIO_STAT_READ] == 0);
		stat[BLKIO_STAT_READ]--;
	}
	if (sync) {
		BUG_ON(stat[BLKIO_STAT_SYNC] == 0);
		stat[BLKIO_STAT_SYNC]--;
	} else {
		BUG_ON(stat[BLKIO_STAT_ASYNC] == 0);
		stat[BLKIO_STAT_ASYNC]--;
	}
}

#ifdef CONFIG_DEBUG_BLK_CGROUP
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
/* This should be called with the blkg->stats_lock held. */
static void blkio_set_start_group_wait_time(struct blkio_group *blkg,
						struct blkio_group *curr_blkg)
{
	if (blkio_blkg_waiting(&blkg->stats))
		return;
	if (blkg == curr_blkg)
		return;
	blkg->stats.start_group_wait_time = sched_clock();
	blkio_mark_blkg_waiting(&blkg->stats);
}

/* This should be called with the blkg->stats_lock held. */
static void blkio_update_group_wait_time(struct blkio_group_stats *stats)
{
	unsigned long long now;

	if (!blkio_blkg_waiting(stats))
		return;

	now = sched_clock();
	if (time_after64(now, stats->start_group_wait_time))
		stats->group_wait_time += now - stats->start_group_wait_time;
	blkio_clear_blkg_waiting(stats);
}

/* This should be called with the blkg->stats_lock held. */
static void blkio_end_empty_time(struct blkio_group_stats *stats)
{
	unsigned long long now;

	if (!blkio_blkg_empty(stats))
		return;

	now = sched_clock();
	if (time_after64(now, stats->start_empty_time))
		stats->empty_time += now - stats->start_empty_time;
	blkio_clear_blkg_empty(stats);
}

void blkiocg_update_set_idle_time_stats(struct blkio_group *blkg)
{
	unsigned long flags;

	spin_lock_irqsave(&blkg->stats_lock, flags);
	BUG_ON(blkio_blkg_idling(&blkg->stats));
	blkg->stats.start_idle_time = sched_clock();
	blkio_mark_blkg_idling(&blkg->stats);
	spin_unlock_irqrestore(&blkg->stats_lock, flags);
}
EXPORT_SYMBOL_GPL(blkiocg_update_set_idle_time_stats);

void blkiocg_update_idle_time_stats(struct blkio_group *blkg)
{
	unsigned long flags;
	unsigned long long now;
	struct blkio_group_stats *stats;

	spin_lock_irqsave(&blkg->stats_lock, flags);
	stats = &blkg->stats;
	if (blkio_blkg_idling(stats)) {
		now = sched_clock();
		if (time_after64(now, stats->start_idle_time))
			stats->idle_time += now - stats->start_idle_time;
		blkio_clear_blkg_idling(stats);
	}
	spin_unlock_irqrestore(&blkg->stats_lock, flags);
}
EXPORT_SYMBOL_GPL(blkiocg_update_idle_time_stats);

206
void blkiocg_update_avg_queue_size_stats(struct blkio_group *blkg)
207 208 209 210 211 212 213 214 215 216
{
	unsigned long flags;
	struct blkio_group_stats *stats;

	spin_lock_irqsave(&blkg->stats_lock, flags);
	stats = &blkg->stats;
	stats->avg_queue_size_sum +=
			stats->stat_arr[BLKIO_STAT_QUEUED][BLKIO_STAT_READ] +
			stats->stat_arr[BLKIO_STAT_QUEUED][BLKIO_STAT_WRITE];
	stats->avg_queue_size_samples++;
217
	blkio_update_group_wait_time(stats);
218 219
	spin_unlock_irqrestore(&blkg->stats_lock, flags);
}
220 221 222 223 224 225 226 227
EXPORT_SYMBOL_GPL(blkiocg_update_avg_queue_size_stats);

void blkiocg_update_dequeue_stats(struct blkio_group *blkg,
			unsigned long dequeue)
{
	blkg->stats.dequeue += dequeue;
}
EXPORT_SYMBOL_GPL(blkiocg_update_dequeue_stats);
228 229 230 231
#else
static inline void blkio_set_start_group_wait_time(struct blkio_group *blkg,
					struct blkio_group *curr_blkg) {}
static inline void blkio_end_empty_time(struct blkio_group_stats *stats) {}
232 233
#endif

234
void blkiocg_update_io_add_stats(struct blkio_group *blkg,
235 236 237 238 239 240 241 242
			struct blkio_group *curr_blkg, bool direction,
			bool sync)
{
	unsigned long flags;

	spin_lock_irqsave(&blkg->stats_lock, flags);
	blkio_add_stat(blkg->stats.stat_arr[BLKIO_STAT_QUEUED], 1, direction,
			sync);
243 244
	blkio_end_empty_time(&blkg->stats);
	blkio_set_start_group_wait_time(blkg, curr_blkg);
245 246
	spin_unlock_irqrestore(&blkg->stats_lock, flags);
}
247
EXPORT_SYMBOL_GPL(blkiocg_update_io_add_stats);
248

249
void blkiocg_update_io_remove_stats(struct blkio_group *blkg,
250 251 252 253 254 255 256 257 258
						bool direction, bool sync)
{
	unsigned long flags;

	spin_lock_irqsave(&blkg->stats_lock, flags);
	blkio_check_and_dec_stat(blkg->stats.stat_arr[BLKIO_STAT_QUEUED],
					direction, sync);
	spin_unlock_irqrestore(&blkg->stats_lock, flags);
}
259
EXPORT_SYMBOL_GPL(blkiocg_update_io_remove_stats);
260

261
void blkiocg_update_timeslice_used(struct blkio_group *blkg, unsigned long time)
262
{
263 264 265 266 267
	unsigned long flags;

	spin_lock_irqsave(&blkg->stats_lock, flags);
	blkg->stats.time += time;
	spin_unlock_irqrestore(&blkg->stats_lock, flags);
268
}
269
EXPORT_SYMBOL_GPL(blkiocg_update_timeslice_used);
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
void blkiocg_set_start_empty_time(struct blkio_group *blkg, bool ignore)
{
	unsigned long flags;
	struct blkio_group_stats *stats;

	spin_lock_irqsave(&blkg->stats_lock, flags);
	stats = &blkg->stats;

	if (stats->stat_arr[BLKIO_STAT_QUEUED][BLKIO_STAT_READ] ||
			stats->stat_arr[BLKIO_STAT_QUEUED][BLKIO_STAT_WRITE]) {
		spin_unlock_irqrestore(&blkg->stats_lock, flags);
		return;
	}

	/*
	 * If ignore is set, we do not panic on the empty flag being set
	 * already. This is to avoid cases where there are superfluous timeslice
	 * complete events (for eg., forced_dispatch in CFQ) when no IOs are
	 * served which could result in triggering the empty check incorrectly.
	 */
	BUG_ON(!ignore && blkio_blkg_empty(stats));
	stats->start_empty_time = sched_clock();
	blkio_mark_blkg_empty(stats);
	spin_unlock_irqrestore(&blkg->stats_lock, flags);
}
EXPORT_SYMBOL_GPL(blkiocg_set_start_empty_time);

298 299
void blkiocg_update_dispatch_stats(struct blkio_group *blkg,
				uint64_t bytes, bool direction, bool sync)
300 301 302 303 304 305
{
	struct blkio_group_stats *stats;
	unsigned long flags;

	spin_lock_irqsave(&blkg->stats_lock, flags);
	stats = &blkg->stats;
306 307 308 309 310
	stats->sectors += bytes >> 9;
	blkio_add_stat(stats->stat_arr[BLKIO_STAT_SERVICED], 1, direction,
			sync);
	blkio_add_stat(stats->stat_arr[BLKIO_STAT_SERVICE_BYTES], bytes,
			direction, sync);
311 312
	spin_unlock_irqrestore(&blkg->stats_lock, flags);
}
313
EXPORT_SYMBOL_GPL(blkiocg_update_dispatch_stats);
314

315 316
void blkiocg_update_completion_stats(struct blkio_group *blkg,
	uint64_t start_time, uint64_t io_start_time, bool direction, bool sync)
317 318 319 320 321 322 323
{
	struct blkio_group_stats *stats;
	unsigned long flags;
	unsigned long long now = sched_clock();

	spin_lock_irqsave(&blkg->stats_lock, flags);
	stats = &blkg->stats;
324 325 326 327 328 329
	if (time_after64(now, io_start_time))
		blkio_add_stat(stats->stat_arr[BLKIO_STAT_SERVICE_TIME],
				now - io_start_time, direction, sync);
	if (time_after64(io_start_time, start_time))
		blkio_add_stat(stats->stat_arr[BLKIO_STAT_WAIT_TIME],
				io_start_time - start_time, direction, sync);
330 331
	spin_unlock_irqrestore(&blkg->stats_lock, flags);
}
332
EXPORT_SYMBOL_GPL(blkiocg_update_completion_stats);
333

D
Divyesh Shah 已提交
334 335 336 337 338 339 340 341 342 343 344 345
void blkiocg_update_io_merged_stats(struct blkio_group *blkg, bool direction,
					bool sync)
{
	unsigned long flags;

	spin_lock_irqsave(&blkg->stats_lock, flags);
	blkio_add_stat(blkg->stats.stat_arr[BLKIO_STAT_MERGED], 1, direction,
			sync);
	spin_unlock_irqrestore(&blkg->stats_lock, flags);
}
EXPORT_SYMBOL_GPL(blkiocg_update_io_merged_stats);

346
void blkiocg_add_blkio_group(struct blkio_cgroup *blkcg,
347
			struct blkio_group *blkg, void *key, dev_t dev)
348 349 350 351 352
{
	unsigned long flags;

	spin_lock_irqsave(&blkcg->lock, flags);
	rcu_assign_pointer(blkg->key, key);
353
	blkg->blkcg_id = css_id(&blkcg->css);
354 355
	hlist_add_head_rcu(&blkg->blkcg_node, &blkcg->blkg_list);
	spin_unlock_irqrestore(&blkcg->lock, flags);
V
Vivek Goyal 已提交
356 357 358 359
#ifdef CONFIG_DEBUG_BLK_CGROUP
	/* Need to take css reference ? */
	cgroup_path(blkcg->css.cgroup, blkg->path, sizeof(blkg->path));
#endif
360
	blkg->dev = dev;
361
}
362
EXPORT_SYMBOL_GPL(blkiocg_add_blkio_group);
363

364 365 366 367 368 369 370 371 372 373
static void __blkiocg_del_blkio_group(struct blkio_group *blkg)
{
	hlist_del_init_rcu(&blkg->blkcg_node);
	blkg->blkcg_id = 0;
}

/*
 * returns 0 if blkio_group was still on cgroup list. Otherwise returns 1
 * indicating that blk_group was unhashed by the time we got to it.
 */
374 375
int blkiocg_del_blkio_group(struct blkio_group *blkg)
{
376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395
	struct blkio_cgroup *blkcg;
	unsigned long flags;
	struct cgroup_subsys_state *css;
	int ret = 1;

	rcu_read_lock();
	css = css_lookup(&blkio_subsys, blkg->blkcg_id);
	if (!css)
		goto out;

	blkcg = container_of(css, struct blkio_cgroup, css);
	spin_lock_irqsave(&blkcg->lock, flags);
	if (!hlist_unhashed(&blkg->blkcg_node)) {
		__blkiocg_del_blkio_group(blkg);
		ret = 0;
	}
	spin_unlock_irqrestore(&blkcg->lock, flags);
out:
	rcu_read_unlock();
	return ret;
396
}
397
EXPORT_SYMBOL_GPL(blkiocg_del_blkio_group);
398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413

/* called under rcu_read_lock(). */
struct blkio_group *blkiocg_lookup_group(struct blkio_cgroup *blkcg, void *key)
{
	struct blkio_group *blkg;
	struct hlist_node *n;
	void *__key;

	hlist_for_each_entry_rcu(blkg, n, &blkcg->blkg_list, blkcg_node) {
		__key = blkg->key;
		if (__key == key)
			return blkg;
	}

	return NULL;
}
414
EXPORT_SYMBOL_GPL(blkiocg_lookup_group);
415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432

#define SHOW_FUNCTION(__VAR)						\
static u64 blkiocg_##__VAR##_read(struct cgroup *cgroup,		\
				       struct cftype *cftype)		\
{									\
	struct blkio_cgroup *blkcg;					\
									\
	blkcg = cgroup_to_blkio_cgroup(cgroup);				\
	return (u64)blkcg->__VAR;					\
}

SHOW_FUNCTION(weight);
#undef SHOW_FUNCTION

static int
blkiocg_weight_write(struct cgroup *cgroup, struct cftype *cftype, u64 val)
{
	struct blkio_cgroup *blkcg;
433 434
	struct blkio_group *blkg;
	struct hlist_node *n;
435
	struct blkio_policy_type *blkiop;
436
	struct blkio_policy_node *pn;
437 438 439 440 441

	if (val < BLKIO_WEIGHT_MIN || val > BLKIO_WEIGHT_MAX)
		return -EINVAL;

	blkcg = cgroup_to_blkio_cgroup(cgroup);
442
	spin_lock(&blkio_list_lock);
443
	spin_lock_irq(&blkcg->lock);
444
	blkcg->weight = (unsigned int)val;
445

446
	hlist_for_each_entry(blkg, n, &blkcg->blkg_list, blkcg_node) {
447 448 449 450 451
		pn = blkio_policy_search_node(blkcg, blkg->dev);

		if (pn)
			continue;

452 453 454 455
		list_for_each_entry(blkiop, &blkio_list, list)
			blkiop->ops.blkio_update_group_weight_fn(blkg,
					blkcg->weight);
	}
456
	spin_unlock_irq(&blkcg->lock);
457
	spin_unlock(&blkio_list_lock);
458 459 460
	return 0;
}

461
static int
462
blkiocg_reset_stats(struct cgroup *cgroup, struct cftype *cftype, u64 val)
463 464 465
{
	struct blkio_cgroup *blkcg;
	struct blkio_group *blkg;
466
	struct blkio_group_stats *stats;
467
	struct hlist_node *n;
468 469
	uint64_t queued[BLKIO_STAT_TOTAL];
	int i;
470 471 472 473
#ifdef CONFIG_DEBUG_BLK_CGROUP
	bool idling, waiting, empty;
	unsigned long long now = sched_clock();
#endif
474 475 476 477 478

	blkcg = cgroup_to_blkio_cgroup(cgroup);
	spin_lock_irq(&blkcg->lock);
	hlist_for_each_entry(blkg, n, &blkcg->blkg_list, blkcg_node) {
		spin_lock(&blkg->stats_lock);
479 480 481 482 483 484
		stats = &blkg->stats;
#ifdef CONFIG_DEBUG_BLK_CGROUP
		idling = blkio_blkg_idling(stats);
		waiting = blkio_blkg_waiting(stats);
		empty = blkio_blkg_empty(stats);
#endif
485
		for (i = 0; i < BLKIO_STAT_TOTAL; i++)
486 487
			queued[i] = stats->stat_arr[BLKIO_STAT_QUEUED][i];
		memset(stats, 0, sizeof(struct blkio_group_stats));
488
		for (i = 0; i < BLKIO_STAT_TOTAL; i++)
489 490 491 492 493 494 495 496 497 498 499 500 501 502 503
			stats->stat_arr[BLKIO_STAT_QUEUED][i] = queued[i];
#ifdef CONFIG_DEBUG_BLK_CGROUP
		if (idling) {
			blkio_mark_blkg_idling(stats);
			stats->start_idle_time = now;
		}
		if (waiting) {
			blkio_mark_blkg_waiting(stats);
			stats->start_group_wait_time = now;
		}
		if (empty) {
			blkio_mark_blkg_empty(stats);
			stats->start_empty_time = now;
		}
#endif
504 505 506 507 508 509
		spin_unlock(&blkg->stats_lock);
	}
	spin_unlock_irq(&blkcg->lock);
	return 0;
}

510 511
static void blkio_get_key_name(enum stat_sub_type type, dev_t dev, char *str,
				int chars_left, bool diskname_only)
512
{
513
	snprintf(str, chars_left, "%d:%d", MAJOR(dev), MINOR(dev));
514 515 516 517 518 519
	chars_left -= strlen(str);
	if (chars_left <= 0) {
		printk(KERN_WARNING
			"Possibly incorrect cgroup stat display format");
		return;
	}
520 521
	if (diskname_only)
		return;
522
	switch (type) {
523
	case BLKIO_STAT_READ:
524 525
		strlcat(str, " Read", chars_left);
		break;
526
	case BLKIO_STAT_WRITE:
527 528
		strlcat(str, " Write", chars_left);
		break;
529
	case BLKIO_STAT_SYNC:
530 531
		strlcat(str, " Sync", chars_left);
		break;
532
	case BLKIO_STAT_ASYNC:
533 534
		strlcat(str, " Async", chars_left);
		break;
535
	case BLKIO_STAT_TOTAL:
536 537 538 539 540 541 542
		strlcat(str, " Total", chars_left);
		break;
	default:
		strlcat(str, " Invalid", chars_left);
	}
}

543 544 545 546 547 548 549
static uint64_t blkio_fill_stat(char *str, int chars_left, uint64_t val,
				struct cgroup_map_cb *cb, dev_t dev)
{
	blkio_get_key_name(0, dev, str, chars_left, true);
	cb->fill(cb, str, val);
	return val;
}
550

551 552 553
/* This should be called with blkg->stats_lock held */
static uint64_t blkio_get_stat(struct blkio_group *blkg,
		struct cgroup_map_cb *cb, dev_t dev, enum stat_type type)
554 555 556
{
	uint64_t disk_total;
	char key_str[MAX_KEY_LEN];
557 558 559 560 561 562 563 564 565
	enum stat_sub_type sub_type;

	if (type == BLKIO_STAT_TIME)
		return blkio_fill_stat(key_str, MAX_KEY_LEN - 1,
					blkg->stats.time, cb, dev);
	if (type == BLKIO_STAT_SECTORS)
		return blkio_fill_stat(key_str, MAX_KEY_LEN - 1,
					blkg->stats.sectors, cb, dev);
#ifdef CONFIG_DEBUG_BLK_CGROUP
566 567 568 569 570 571 572 573 574
	if (type == BLKIO_STAT_AVG_QUEUE_SIZE) {
		uint64_t sum = blkg->stats.avg_queue_size_sum;
		uint64_t samples = blkg->stats.avg_queue_size_samples;
		if (samples)
			do_div(sum, samples);
		else
			sum = 0;
		return blkio_fill_stat(key_str, MAX_KEY_LEN - 1, sum, cb, dev);
	}
575 576 577 578 579 580 581 582 583
	if (type == BLKIO_STAT_GROUP_WAIT_TIME)
		return blkio_fill_stat(key_str, MAX_KEY_LEN - 1,
					blkg->stats.group_wait_time, cb, dev);
	if (type == BLKIO_STAT_IDLE_TIME)
		return blkio_fill_stat(key_str, MAX_KEY_LEN - 1,
					blkg->stats.idle_time, cb, dev);
	if (type == BLKIO_STAT_EMPTY_TIME)
		return blkio_fill_stat(key_str, MAX_KEY_LEN - 1,
					blkg->stats.empty_time, cb, dev);
584 585 586 587
	if (type == BLKIO_STAT_DEQUEUE)
		return blkio_fill_stat(key_str, MAX_KEY_LEN - 1,
					blkg->stats.dequeue, cb, dev);
#endif
588

589 590 591 592
	for (sub_type = BLKIO_STAT_READ; sub_type < BLKIO_STAT_TOTAL;
			sub_type++) {
		blkio_get_key_name(sub_type, dev, key_str, MAX_KEY_LEN, false);
		cb->fill(cb, key_str, blkg->stats.stat_arr[type][sub_type]);
593
	}
594 595 596
	disk_total = blkg->stats.stat_arr[type][BLKIO_STAT_READ] +
			blkg->stats.stat_arr[type][BLKIO_STAT_WRITE];
	blkio_get_key_name(BLKIO_STAT_TOTAL, dev, key_str, MAX_KEY_LEN, false);
597 598 599 600
	cb->fill(cb, key_str, disk_total);
	return disk_total;
}

601
#define SHOW_FUNCTION_PER_GROUP(__VAR, type, show_total)		\
602
static int blkiocg_##__VAR##_read(struct cgroup *cgroup,		\
603
		struct cftype *cftype, struct cgroup_map_cb *cb)	\
604 605 606 607
{									\
	struct blkio_cgroup *blkcg;					\
	struct blkio_group *blkg;					\
	struct hlist_node *n;						\
608
	uint64_t cgroup_total = 0;					\
609 610 611 612 613 614 615
									\
	if (!cgroup_lock_live_group(cgroup))				\
		return -ENODEV;						\
									\
	blkcg = cgroup_to_blkio_cgroup(cgroup);				\
	rcu_read_lock();						\
	hlist_for_each_entry_rcu(blkg, n, &blkcg->blkg_list, blkcg_node) {\
616 617
		if (blkg->dev) {					\
			spin_lock_irq(&blkg->stats_lock);		\
618 619
			cgroup_total += blkio_get_stat(blkg, cb,	\
						blkg->dev, type);	\
620 621
			spin_unlock_irq(&blkg->stats_lock);		\
		}							\
622
	}								\
623 624
	if (show_total)							\
		cb->fill(cb, "Total", cgroup_total);			\
625 626 627 628 629
	rcu_read_unlock();						\
	cgroup_unlock();						\
	return 0;							\
}

630 631 632 633 634 635
SHOW_FUNCTION_PER_GROUP(time, BLKIO_STAT_TIME, 0);
SHOW_FUNCTION_PER_GROUP(sectors, BLKIO_STAT_SECTORS, 0);
SHOW_FUNCTION_PER_GROUP(io_service_bytes, BLKIO_STAT_SERVICE_BYTES, 1);
SHOW_FUNCTION_PER_GROUP(io_serviced, BLKIO_STAT_SERVICED, 1);
SHOW_FUNCTION_PER_GROUP(io_service_time, BLKIO_STAT_SERVICE_TIME, 1);
SHOW_FUNCTION_PER_GROUP(io_wait_time, BLKIO_STAT_WAIT_TIME, 1);
D
Divyesh Shah 已提交
636
SHOW_FUNCTION_PER_GROUP(io_merged, BLKIO_STAT_MERGED, 1);
637
SHOW_FUNCTION_PER_GROUP(io_queued, BLKIO_STAT_QUEUED, 1);
638
#ifdef CONFIG_DEBUG_BLK_CGROUP
639
SHOW_FUNCTION_PER_GROUP(dequeue, BLKIO_STAT_DEQUEUE, 0);
640
SHOW_FUNCTION_PER_GROUP(avg_queue_size, BLKIO_STAT_AVG_QUEUE_SIZE, 0);
641 642 643
SHOW_FUNCTION_PER_GROUP(group_wait_time, BLKIO_STAT_GROUP_WAIT_TIME, 0);
SHOW_FUNCTION_PER_GROUP(idle_time, BLKIO_STAT_IDLE_TIME, 0);
SHOW_FUNCTION_PER_GROUP(empty_time, BLKIO_STAT_EMPTY_TIME, 0);
644 645 646
#endif
#undef SHOW_FUNCTION_PER_GROUP

647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835
static int blkio_check_dev_num(dev_t dev)
{
	int part = 0;
	struct gendisk *disk;

	disk = get_gendisk(dev, &part);
	if (!disk || part)
		return -ENODEV;

	return 0;
}

static int blkio_policy_parse_and_set(char *buf,
				      struct blkio_policy_node *newpn)
{
	char *s[4], *p, *major_s = NULL, *minor_s = NULL;
	int ret;
	unsigned long major, minor, temp;
	int i = 0;
	dev_t dev;

	memset(s, 0, sizeof(s));

	while ((p = strsep(&buf, " ")) != NULL) {
		if (!*p)
			continue;

		s[i++] = p;

		/* Prevent from inputing too many things */
		if (i == 3)
			break;
	}

	if (i != 2)
		return -EINVAL;

	p = strsep(&s[0], ":");
	if (p != NULL)
		major_s = p;
	else
		return -EINVAL;

	minor_s = s[0];
	if (!minor_s)
		return -EINVAL;

	ret = strict_strtoul(major_s, 10, &major);
	if (ret)
		return -EINVAL;

	ret = strict_strtoul(minor_s, 10, &minor);
	if (ret)
		return -EINVAL;

	dev = MKDEV(major, minor);

	ret = blkio_check_dev_num(dev);
	if (ret)
		return ret;

	newpn->dev = dev;

	if (s[1] == NULL)
		return -EINVAL;

	ret = strict_strtoul(s[1], 10, &temp);
	if (ret || (temp < BLKIO_WEIGHT_MIN && temp > 0) ||
	    temp > BLKIO_WEIGHT_MAX)
		return -EINVAL;

	newpn->weight =  temp;

	return 0;
}

unsigned int blkcg_get_weight(struct blkio_cgroup *blkcg,
			      dev_t dev)
{
	struct blkio_policy_node *pn;

	pn = blkio_policy_search_node(blkcg, dev);
	if (pn)
		return pn->weight;
	else
		return blkcg->weight;
}
EXPORT_SYMBOL_GPL(blkcg_get_weight);


static int blkiocg_weight_device_write(struct cgroup *cgrp, struct cftype *cft,
				       const char *buffer)
{
	int ret = 0;
	char *buf;
	struct blkio_policy_node *newpn, *pn;
	struct blkio_cgroup *blkcg;
	struct blkio_group *blkg;
	int keep_newpn = 0;
	struct hlist_node *n;
	struct blkio_policy_type *blkiop;

	buf = kstrdup(buffer, GFP_KERNEL);
	if (!buf)
		return -ENOMEM;

	newpn = kzalloc(sizeof(*newpn), GFP_KERNEL);
	if (!newpn) {
		ret = -ENOMEM;
		goto free_buf;
	}

	ret = blkio_policy_parse_and_set(buf, newpn);
	if (ret)
		goto free_newpn;

	blkcg = cgroup_to_blkio_cgroup(cgrp);

	spin_lock_irq(&blkcg->lock);

	pn = blkio_policy_search_node(blkcg, newpn->dev);
	if (!pn) {
		if (newpn->weight != 0) {
			blkio_policy_insert_node(blkcg, newpn);
			keep_newpn = 1;
		}
		spin_unlock_irq(&blkcg->lock);
		goto update_io_group;
	}

	if (newpn->weight == 0) {
		/* weight == 0 means deleteing a specific weight */
		blkio_policy_delete_node(pn);
		spin_unlock_irq(&blkcg->lock);
		goto update_io_group;
	}
	spin_unlock_irq(&blkcg->lock);

	pn->weight = newpn->weight;

update_io_group:
	/* update weight for each cfqg */
	spin_lock(&blkio_list_lock);
	spin_lock_irq(&blkcg->lock);

	hlist_for_each_entry(blkg, n, &blkcg->blkg_list, blkcg_node) {
		if (newpn->dev == blkg->dev) {
			list_for_each_entry(blkiop, &blkio_list, list)
				blkiop->ops.blkio_update_group_weight_fn(blkg,
							 newpn->weight ?
							 newpn->weight :
							 blkcg->weight);
		}
	}

	spin_unlock_irq(&blkcg->lock);
	spin_unlock(&blkio_list_lock);

free_newpn:
	if (!keep_newpn)
		kfree(newpn);
free_buf:
	kfree(buf);
	return ret;
}

static int blkiocg_weight_device_read(struct cgroup *cgrp, struct cftype *cft,
				      struct seq_file *m)
{
	struct blkio_cgroup *blkcg;
	struct blkio_policy_node *pn;

	seq_printf(m, "dev\tweight\n");

	blkcg = cgroup_to_blkio_cgroup(cgrp);
	if (list_empty(&blkcg->policy_list))
		goto out;

	spin_lock_irq(&blkcg->lock);
	list_for_each_entry(pn, &blkcg->policy_list, node) {
		seq_printf(m, "%u:%u\t%u\n", MAJOR(pn->dev),
			   MINOR(pn->dev), pn->weight);
	}
	spin_unlock_irq(&blkcg->lock);

out:
	return 0;
}

836
struct cftype blkio_files[] = {
837 838 839 840 841 842
	{
		.name = "weight_device",
		.read_seq_string = blkiocg_weight_device_read,
		.write_string = blkiocg_weight_device_write,
		.max_write_len = 256,
	},
843 844 845 846 847
	{
		.name = "weight",
		.read_u64 = blkiocg_weight_read,
		.write_u64 = blkiocg_weight_write,
	},
848 849
	{
		.name = "time",
850
		.read_map = blkiocg_time_read,
851 852 853
	},
	{
		.name = "sectors",
854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870
		.read_map = blkiocg_sectors_read,
	},
	{
		.name = "io_service_bytes",
		.read_map = blkiocg_io_service_bytes_read,
	},
	{
		.name = "io_serviced",
		.read_map = blkiocg_io_serviced_read,
	},
	{
		.name = "io_service_time",
		.read_map = blkiocg_io_service_time_read,
	},
	{
		.name = "io_wait_time",
		.read_map = blkiocg_io_wait_time_read,
871
	},
D
Divyesh Shah 已提交
872 873 874 875
	{
		.name = "io_merged",
		.read_map = blkiocg_io_merged_read,
	},
876 877 878 879
	{
		.name = "io_queued",
		.read_map = blkiocg_io_queued_read,
	},
880 881 882
	{
		.name = "reset_stats",
		.write_u64 = blkiocg_reset_stats,
883 884
	},
#ifdef CONFIG_DEBUG_BLK_CGROUP
885 886 887 888
	{
		.name = "avg_queue_size",
		.read_map = blkiocg_avg_queue_size_read,
	},
889 890 891 892 893 894 895 896 897 898 899 900
	{
		.name = "group_wait_time",
		.read_map = blkiocg_group_wait_time_read,
	},
	{
		.name = "idle_time",
		.read_map = blkiocg_idle_time_read,
	},
	{
		.name = "empty_time",
		.read_map = blkiocg_empty_time_read,
	},
901
	{
902
		.name = "dequeue",
903
		.read_map = blkiocg_dequeue_read,
904
	},
905
#endif
906 907 908 909 910 911 912 913 914 915 916
};

static int blkiocg_populate(struct cgroup_subsys *subsys, struct cgroup *cgroup)
{
	return cgroup_add_files(cgroup, subsys, blkio_files,
				ARRAY_SIZE(blkio_files));
}

static void blkiocg_destroy(struct cgroup_subsys *subsys, struct cgroup *cgroup)
{
	struct blkio_cgroup *blkcg = cgroup_to_blkio_cgroup(cgroup);
917 918 919
	unsigned long flags;
	struct blkio_group *blkg;
	void *key;
920
	struct blkio_policy_type *blkiop;
921
	struct blkio_policy_node *pn, *pntmp;
922 923 924 925 926 927 928 929 930 931 932 933 934 935

	rcu_read_lock();
remove_entry:
	spin_lock_irqsave(&blkcg->lock, flags);

	if (hlist_empty(&blkcg->blkg_list)) {
		spin_unlock_irqrestore(&blkcg->lock, flags);
		goto done;
	}

	blkg = hlist_entry(blkcg->blkg_list.first, struct blkio_group,
				blkcg_node);
	key = rcu_dereference(blkg->key);
	__blkiocg_del_blkio_group(blkg);
936

937 938 939 940 941 942 943 944 945 946
	spin_unlock_irqrestore(&blkcg->lock, flags);

	/*
	 * This blkio_group is being unlinked as associated cgroup is going
	 * away. Let all the IO controlling policies know about this event.
	 *
	 * Currently this is static call to one io controlling policy. Once
	 * we have more policies in place, we need some dynamic registration
	 * of callback function.
	 */
947 948 949 950
	spin_lock(&blkio_list_lock);
	list_for_each_entry(blkiop, &blkio_list, list)
		blkiop->ops.blkio_unlink_group_fn(key, blkg);
	spin_unlock(&blkio_list_lock);
951
	goto remove_entry;
952

953
done:
954 955 956 957
	list_for_each_entry_safe(pn, pntmp, &blkcg->policy_list, node) {
		blkio_policy_delete_node(pn);
		kfree(pn);
	}
958
	free_css_id(&blkio_subsys, &blkcg->css);
959
	rcu_read_unlock();
B
Ben Blum 已提交
960 961
	if (blkcg != &blkio_root_cgroup)
		kfree(blkcg);
962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987
}

static struct cgroup_subsys_state *
blkiocg_create(struct cgroup_subsys *subsys, struct cgroup *cgroup)
{
	struct blkio_cgroup *blkcg, *parent_blkcg;

	if (!cgroup->parent) {
		blkcg = &blkio_root_cgroup;
		goto done;
	}

	/* Currently we do not support hierarchy deeper than two level (0,1) */
	parent_blkcg = cgroup_to_blkio_cgroup(cgroup->parent);
	if (css_depth(&parent_blkcg->css) > 0)
		return ERR_PTR(-EINVAL);

	blkcg = kzalloc(sizeof(*blkcg), GFP_KERNEL);
	if (!blkcg)
		return ERR_PTR(-ENOMEM);

	blkcg->weight = BLKIO_WEIGHT_DEFAULT;
done:
	spin_lock_init(&blkcg->lock);
	INIT_HLIST_HEAD(&blkcg->blkg_list);

988
	INIT_LIST_HEAD(&blkcg->policy_list);
989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027
	return &blkcg->css;
}

/*
 * We cannot support shared io contexts, as we have no mean to support
 * two tasks with the same ioc in two different groups without major rework
 * of the main cic data structures.  For now we allow a task to change
 * its cgroup only if it's the only owner of its ioc.
 */
static int blkiocg_can_attach(struct cgroup_subsys *subsys,
				struct cgroup *cgroup, struct task_struct *tsk,
				bool threadgroup)
{
	struct io_context *ioc;
	int ret = 0;

	/* task_lock() is needed to avoid races with exit_io_context() */
	task_lock(tsk);
	ioc = tsk->io_context;
	if (ioc && atomic_read(&ioc->nr_tasks) > 1)
		ret = -EINVAL;
	task_unlock(tsk);

	return ret;
}

static void blkiocg_attach(struct cgroup_subsys *subsys, struct cgroup *cgroup,
				struct cgroup *prev, struct task_struct *tsk,
				bool threadgroup)
{
	struct io_context *ioc;

	task_lock(tsk);
	ioc = tsk->io_context;
	if (ioc)
		ioc->cgroup_changed = 1;
	task_unlock(tsk);
}

1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042
void blkio_policy_register(struct blkio_policy_type *blkiop)
{
	spin_lock(&blkio_list_lock);
	list_add_tail(&blkiop->list, &blkio_list);
	spin_unlock(&blkio_list_lock);
}
EXPORT_SYMBOL_GPL(blkio_policy_register);

void blkio_policy_unregister(struct blkio_policy_type *blkiop)
{
	spin_lock(&blkio_list_lock);
	list_del_init(&blkiop->list);
	spin_unlock(&blkio_list_lock);
}
EXPORT_SYMBOL_GPL(blkio_policy_unregister);
B
Ben Blum 已提交
1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056

static int __init init_cgroup_blkio(void)
{
	return cgroup_load_subsys(&blkio_subsys);
}

static void __exit exit_cgroup_blkio(void)
{
	cgroup_unload_subsys(&blkio_subsys);
}

module_init(init_cgroup_blkio);
module_exit(exit_cgroup_blkio);
MODULE_LICENSE("GPL");