blk-cgroup.c 18.0 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 "blk-cgroup.h"
20

21 22
#define MAX_KEY_LEN 100

23 24
static DEFINE_SPINLOCK(blkio_list_lock);
static LIST_HEAD(blkio_list);
25

26
struct blkio_cgroup blkio_root_cgroup = { .weight = 2*BLKIO_WEIGHT_DEFAULT };
27 28
EXPORT_SYMBOL_GPL(blkio_root_cgroup);

B
Ben Blum 已提交
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
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);

54 55 56 57 58
struct blkio_cgroup *cgroup_to_blkio_cgroup(struct cgroup *cgroup)
{
	return container_of(cgroup_subsys_state(cgroup, blkio_subsys_id),
			    struct blkio_cgroup, css);
}
59
EXPORT_SYMBOL_GPL(cgroup_to_blkio_cgroup);
60

61 62 63 64 65 66
void blkio_group_init(struct blkio_group *blkg)
{
	spin_lock_init(&blkg->stats_lock);
}
EXPORT_SYMBOL_GPL(blkio_group_init);

67 68 69 70
/*
 * Add to the appropriate stat variable depending on the request type.
 * This should be called with the blkg->stats_lock held.
 */
71 72
static void blkio_add_stat(uint64_t *stat, uint64_t add, bool direction,
				bool sync)
73
{
74 75
	if (direction)
		stat[BLKIO_STAT_WRITE] += add;
76
	else
77 78 79
		stat[BLKIO_STAT_READ] += add;
	if (sync)
		stat[BLKIO_STAT_SYNC] += add;
80
	else
81
		stat[BLKIO_STAT_ASYNC] += add;
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
/*
 * 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
void blkiocg_update_set_active_queue_stats(struct blkio_group *blkg)
{
	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++;
	spin_unlock_irqrestore(&blkg->stats_lock, flags);
}
EXPORT_SYMBOL_GPL(blkiocg_update_set_active_queue_stats);
#endif

void blkiocg_update_request_add_stats(struct blkio_group *blkg,
			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);
	spin_unlock_irqrestore(&blkg->stats_lock, flags);
}
EXPORT_SYMBOL_GPL(blkiocg_update_request_add_stats);

void blkiocg_update_request_remove_stats(struct blkio_group *blkg,
						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);
}
EXPORT_SYMBOL_GPL(blkiocg_update_request_remove_stats);

149
void blkiocg_update_timeslice_used(struct blkio_group *blkg, unsigned long time)
150
{
151 152 153 154 155
	unsigned long flags;

	spin_lock_irqsave(&blkg->stats_lock, flags);
	blkg->stats.time += time;
	spin_unlock_irqrestore(&blkg->stats_lock, flags);
156
}
157
EXPORT_SYMBOL_GPL(blkiocg_update_timeslice_used);
158

159 160
void blkiocg_update_dispatch_stats(struct blkio_group *blkg,
				uint64_t bytes, bool direction, bool sync)
161 162 163 164 165 166
{
	struct blkio_group_stats *stats;
	unsigned long flags;

	spin_lock_irqsave(&blkg->stats_lock, flags);
	stats = &blkg->stats;
167 168 169 170 171
	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);
172 173
	spin_unlock_irqrestore(&blkg->stats_lock, flags);
}
174
EXPORT_SYMBOL_GPL(blkiocg_update_dispatch_stats);
175

176 177
void blkiocg_update_completion_stats(struct blkio_group *blkg,
	uint64_t start_time, uint64_t io_start_time, bool direction, bool sync)
178 179 180 181 182 183 184
{
	struct blkio_group_stats *stats;
	unsigned long flags;
	unsigned long long now = sched_clock();

	spin_lock_irqsave(&blkg->stats_lock, flags);
	stats = &blkg->stats;
185 186 187 188 189 190
	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);
191 192
	spin_unlock_irqrestore(&blkg->stats_lock, flags);
}
193
EXPORT_SYMBOL_GPL(blkiocg_update_completion_stats);
194

D
Divyesh Shah 已提交
195 196 197 198 199 200 201 202 203 204 205 206
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);

207
void blkiocg_add_blkio_group(struct blkio_cgroup *blkcg,
208
			struct blkio_group *blkg, void *key, dev_t dev)
209 210 211 212 213
{
	unsigned long flags;

	spin_lock_irqsave(&blkcg->lock, flags);
	rcu_assign_pointer(blkg->key, key);
214
	blkg->blkcg_id = css_id(&blkcg->css);
215 216
	hlist_add_head_rcu(&blkg->blkcg_node, &blkcg->blkg_list);
	spin_unlock_irqrestore(&blkcg->lock, flags);
V
Vivek Goyal 已提交
217 218 219 220
#ifdef CONFIG_DEBUG_BLK_CGROUP
	/* Need to take css reference ? */
	cgroup_path(blkcg->css.cgroup, blkg->path, sizeof(blkg->path));
#endif
221
	blkg->dev = dev;
222
}
223
EXPORT_SYMBOL_GPL(blkiocg_add_blkio_group);
224

225 226 227 228 229 230 231 232 233 234
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.
 */
235 236
int blkiocg_del_blkio_group(struct blkio_group *blkg)
{
237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256
	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;
257
}
258
EXPORT_SYMBOL_GPL(blkiocg_del_blkio_group);
259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274

/* 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;
}
275
EXPORT_SYMBOL_GPL(blkiocg_lookup_group);
276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293

#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;
294 295
	struct blkio_group *blkg;
	struct hlist_node *n;
296
	struct blkio_policy_type *blkiop;
297 298 299 300 301

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

	blkcg = cgroup_to_blkio_cgroup(cgroup);
302
	spin_lock(&blkio_list_lock);
303
	spin_lock_irq(&blkcg->lock);
304
	blkcg->weight = (unsigned int)val;
305 306 307 308 309
	hlist_for_each_entry(blkg, n, &blkcg->blkg_list, blkcg_node) {
		list_for_each_entry(blkiop, &blkio_list, list)
			blkiop->ops.blkio_update_group_weight_fn(blkg,
					blkcg->weight);
	}
310
	spin_unlock_irq(&blkcg->lock);
311
	spin_unlock(&blkio_list_lock);
312 313 314
	return 0;
}

315
static int
316
blkiocg_reset_stats(struct cgroup *cgroup, struct cftype *cftype, u64 val)
317 318 319 320
{
	struct blkio_cgroup *blkcg;
	struct blkio_group *blkg;
	struct hlist_node *n;
321 322
	uint64_t queued[BLKIO_STAT_TOTAL];
	int i;
323 324 325 326 327

	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);
328 329 330 331 332
		for (i = 0; i < BLKIO_STAT_TOTAL; i++)
			queued[i] = blkg->stats.stat_arr[BLKIO_STAT_QUEUED][i];
		memset(&blkg->stats, 0, sizeof(struct blkio_group_stats));
		for (i = 0; i < BLKIO_STAT_TOTAL; i++)
			blkg->stats.stat_arr[BLKIO_STAT_QUEUED][i] = queued[i];
333 334 335 336 337 338
		spin_unlock(&blkg->stats_lock);
	}
	spin_unlock_irq(&blkcg->lock);
	return 0;
}

339 340
static void blkio_get_key_name(enum stat_sub_type type, dev_t dev, char *str,
				int chars_left, bool diskname_only)
341
{
342
	snprintf(str, chars_left, "%d:%d", MAJOR(dev), MINOR(dev));
343 344 345 346 347 348
	chars_left -= strlen(str);
	if (chars_left <= 0) {
		printk(KERN_WARNING
			"Possibly incorrect cgroup stat display format");
		return;
	}
349 350
	if (diskname_only)
		return;
351
	switch (type) {
352
	case BLKIO_STAT_READ:
353 354
		strlcat(str, " Read", chars_left);
		break;
355
	case BLKIO_STAT_WRITE:
356 357
		strlcat(str, " Write", chars_left);
		break;
358
	case BLKIO_STAT_SYNC:
359 360
		strlcat(str, " Sync", chars_left);
		break;
361
	case BLKIO_STAT_ASYNC:
362 363
		strlcat(str, " Async", chars_left);
		break;
364
	case BLKIO_STAT_TOTAL:
365 366 367 368 369 370 371
		strlcat(str, " Total", chars_left);
		break;
	default:
		strlcat(str, " Invalid", chars_left);
	}
}

372 373 374 375 376 377 378
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;
}
379

380 381 382
/* 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)
383 384 385
{
	uint64_t disk_total;
	char key_str[MAX_KEY_LEN];
386 387 388 389 390 391 392 393 394
	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
395 396 397 398 399 400 401 402 403
	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);
	}
404 405 406 407
	if (type == BLKIO_STAT_DEQUEUE)
		return blkio_fill_stat(key_str, MAX_KEY_LEN - 1,
					blkg->stats.dequeue, cb, dev);
#endif
408

409 410 411 412
	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]);
413
	}
414 415 416
	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);
417 418 419 420
	cb->fill(cb, key_str, disk_total);
	return disk_total;
}

421
#define SHOW_FUNCTION_PER_GROUP(__VAR, type, show_total)		\
422
static int blkiocg_##__VAR##_read(struct cgroup *cgroup,		\
423
		struct cftype *cftype, struct cgroup_map_cb *cb)	\
424 425 426 427
{									\
	struct blkio_cgroup *blkcg;					\
	struct blkio_group *blkg;					\
	struct hlist_node *n;						\
428
	uint64_t cgroup_total = 0;					\
429 430 431 432 433 434 435
									\
	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) {\
436 437
		if (blkg->dev) {					\
			spin_lock_irq(&blkg->stats_lock);		\
438 439
			cgroup_total += blkio_get_stat(blkg, cb,	\
						blkg->dev, type);	\
440 441
			spin_unlock_irq(&blkg->stats_lock);		\
		}							\
442
	}								\
443 444
	if (show_total)							\
		cb->fill(cb, "Total", cgroup_total);			\
445 446 447 448 449
	rcu_read_unlock();						\
	cgroup_unlock();						\
	return 0;							\
}

450 451 452 453 454 455
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
456
SHOW_FUNCTION_PER_GROUP(io_merged, BLKIO_STAT_MERGED, 1);
457
SHOW_FUNCTION_PER_GROUP(io_queued, BLKIO_STAT_QUEUED, 1);
458
#ifdef CONFIG_DEBUG_BLK_CGROUP
459
SHOW_FUNCTION_PER_GROUP(dequeue, BLKIO_STAT_DEQUEUE, 0);
460
SHOW_FUNCTION_PER_GROUP(avg_queue_size, BLKIO_STAT_AVG_QUEUE_SIZE, 0);
461 462 463 464
#endif
#undef SHOW_FUNCTION_PER_GROUP

#ifdef CONFIG_DEBUG_BLK_CGROUP
465
void blkiocg_update_dequeue_stats(struct blkio_group *blkg,
466 467
			unsigned long dequeue)
{
468
	blkg->stats.dequeue += dequeue;
469
}
470
EXPORT_SYMBOL_GPL(blkiocg_update_dequeue_stats);
471 472
#endif

473 474 475 476 477 478
struct cftype blkio_files[] = {
	{
		.name = "weight",
		.read_u64 = blkiocg_weight_read,
		.write_u64 = blkiocg_weight_write,
	},
479 480
	{
		.name = "time",
481
		.read_map = blkiocg_time_read,
482 483 484
	},
	{
		.name = "sectors",
485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501
		.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,
502
	},
D
Divyesh Shah 已提交
503 504 505 506
	{
		.name = "io_merged",
		.read_map = blkiocg_io_merged_read,
	},
507 508 509 510
	{
		.name = "io_queued",
		.read_map = blkiocg_io_queued_read,
	},
511 512 513
	{
		.name = "reset_stats",
		.write_u64 = blkiocg_reset_stats,
514 515
	},
#ifdef CONFIG_DEBUG_BLK_CGROUP
516 517 518 519 520
	{
		.name = "avg_queue_size",
		.read_map = blkiocg_avg_queue_size_read,
	},
	{
521
		.name = "dequeue",
522
		.read_map = blkiocg_dequeue_read,
523
	},
524
#endif
525 526 527 528 529 530 531 532 533 534 535
};

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);
536 537 538
	unsigned long flags;
	struct blkio_group *blkg;
	void *key;
539
	struct blkio_policy_type *blkiop;
540 541 542 543 544 545 546 547 548 549 550 551 552 553

	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);
554

555 556 557 558 559 560 561 562 563 564
	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.
	 */
565 566 567 568
	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);
569 570
	goto remove_entry;
done:
571
	free_css_id(&blkio_subsys, &blkcg->css);
572
	rcu_read_unlock();
B
Ben Blum 已提交
573 574
	if (blkcg != &blkio_root_cgroup)
		kfree(blkcg);
575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639
}

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);

	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);
}

640 641 642 643 644 645 646 647 648 649 650 651 652 653 654
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
655 656 657 658 659 660 661 662 663 664 665 666 667 668

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");