Blame view
block/blk-cgroup.c
43.2 KB
31e4c28d9
|
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> |
220841906
|
14 15 |
#include <linux/seq_file.h> #include <linux/kdev_t.h> |
9d6a986c0
|
16 |
#include <linux/module.h> |
accee7854
|
17 |
#include <linux/err.h> |
9195291e5
|
18 |
#include <linux/blkdev.h> |
5a0e3ad6a
|
19 |
#include <linux/slab.h> |
31e4c28d9
|
20 |
#include "blk-cgroup.h" |
34d0f179d
|
21 |
#include <linux/genhd.h> |
3e2520668
|
22 |
|
84c124da9
|
23 |
#define MAX_KEY_LEN 100 |
3e2520668
|
24 25 |
static DEFINE_SPINLOCK(blkio_list_lock); static LIST_HEAD(blkio_list); |
b1c357696
|
26 |
|
31e4c28d9
|
27 |
struct blkio_cgroup blkio_root_cgroup = { .weight = 2*BLKIO_WEIGHT_DEFAULT }; |
9d6a986c0
|
28 |
EXPORT_SYMBOL_GPL(blkio_root_cgroup); |
67523c48a
|
29 30 |
static struct cgroup_subsys_state *blkiocg_create(struct cgroup_subsys *, struct cgroup *); |
bb9d97b6d
|
31 32 33 34 |
static int blkiocg_can_attach(struct cgroup_subsys *, struct cgroup *, struct cgroup_taskset *); static void blkiocg_attach(struct cgroup_subsys *, struct cgroup *, struct cgroup_taskset *); |
67523c48a
|
35 36 |
static void blkiocg_destroy(struct cgroup_subsys *, struct cgroup *); static int blkiocg_populate(struct cgroup_subsys *, struct cgroup *); |
062a644d6
|
37 38 39 40 41 |
/* for encoding cft->private value on file */ #define BLKIOFILE_PRIVATE(x, val) (((x) << 16) | (val)) /* What policy owns the file, proportional or throttle */ #define BLKIOFILE_POLICY(val) (((val) >> 16) & 0xffff) #define BLKIOFILE_ATTR(val) ((val) & 0xffff) |
67523c48a
|
42 43 44 |
struct cgroup_subsys blkio_subsys = { .name = "blkio", .create = blkiocg_create, |
bb9d97b6d
|
45 46 |
.can_attach = blkiocg_can_attach, .attach = blkiocg_attach, |
67523c48a
|
47 48 49 50 51 52 53 54 55 56 |
.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); |
34d0f179d
|
57 58 59 60 61 |
static inline void blkio_policy_insert_node(struct blkio_cgroup *blkcg, struct blkio_policy_node *pn) { list_add(&pn->node, &blkcg->policy_list); } |
062a644d6
|
62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 |
static inline bool cftype_blkg_same_policy(struct cftype *cft, struct blkio_group *blkg) { enum blkio_policy_id plid = BLKIOFILE_POLICY(cft->private); if (blkg->plid == plid) return 1; return 0; } /* Determines if policy node matches cgroup file being accessed */ static inline bool pn_matches_cftype(struct cftype *cft, struct blkio_policy_node *pn) { enum blkio_policy_id plid = BLKIOFILE_POLICY(cft->private); int fileid = BLKIOFILE_ATTR(cft->private); return (plid == pn->plid && fileid == pn->fileid); } |
34d0f179d
|
82 83 84 85 86 87 88 89 |
/* 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 * |
062a644d6
|
90 91 |
blkio_policy_search_node(const struct blkio_cgroup *blkcg, dev_t dev, enum blkio_policy_id plid, int fileid) |
34d0f179d
|
92 93 94 95 |
{ struct blkio_policy_node *pn; list_for_each_entry(pn, &blkcg->policy_list, node) { |
062a644d6
|
96 |
if (pn->dev == dev && pn->plid == plid && pn->fileid == fileid) |
34d0f179d
|
97 98 99 100 101 |
return pn; } return NULL; } |
31e4c28d9
|
102 103 104 105 106 |
struct blkio_cgroup *cgroup_to_blkio_cgroup(struct cgroup *cgroup) { return container_of(cgroup_subsys_state(cgroup, blkio_subsys_id), struct blkio_cgroup, css); } |
9d6a986c0
|
107 |
EXPORT_SYMBOL_GPL(cgroup_to_blkio_cgroup); |
31e4c28d9
|
108 |
|
70087dc38
|
109 110 111 112 113 114 |
struct blkio_cgroup *task_blkio_cgroup(struct task_struct *tsk) { return container_of(task_subsys_state(tsk, blkio_subsys_id), struct blkio_cgroup, css); } EXPORT_SYMBOL_GPL(task_blkio_cgroup); |
062a644d6
|
115 116 117 118 119 120 121 122 123 124 |
static inline void blkio_update_group_weight(struct blkio_group *blkg, unsigned int weight) { struct blkio_policy_type *blkiop; list_for_each_entry(blkiop, &blkio_list, list) { /* If this policy does not own the blkg, do not send updates */ if (blkiop->plid != blkg->plid) continue; if (blkiop->ops.blkio_update_group_weight_fn) |
fe0714377
|
125 126 |
blkiop->ops.blkio_update_group_weight_fn(blkg->key, blkg, weight); |
062a644d6
|
127 128 |
} } |
4c9eefa16
|
129 130 131 132 133 134 135 136 137 138 139 140 141 |
static inline void blkio_update_group_bps(struct blkio_group *blkg, u64 bps, int fileid) { struct blkio_policy_type *blkiop; list_for_each_entry(blkiop, &blkio_list, list) { /* If this policy does not own the blkg, do not send updates */ if (blkiop->plid != blkg->plid) continue; if (fileid == BLKIO_THROTL_read_bps_device && blkiop->ops.blkio_update_group_read_bps_fn) |
fe0714377
|
142 143 |
blkiop->ops.blkio_update_group_read_bps_fn(blkg->key, blkg, bps); |
4c9eefa16
|
144 145 146 |
if (fileid == BLKIO_THROTL_write_bps_device && blkiop->ops.blkio_update_group_write_bps_fn) |
fe0714377
|
147 148 |
blkiop->ops.blkio_update_group_write_bps_fn(blkg->key, blkg, bps); |
4c9eefa16
|
149 150 |
} } |
7702e8f45
|
151 152 153 154 155 156 157 158 159 160 161 162 163 |
static inline void blkio_update_group_iops(struct blkio_group *blkg, unsigned int iops, int fileid) { struct blkio_policy_type *blkiop; list_for_each_entry(blkiop, &blkio_list, list) { /* If this policy does not own the blkg, do not send updates */ if (blkiop->plid != blkg->plid) continue; if (fileid == BLKIO_THROTL_read_iops_device && blkiop->ops.blkio_update_group_read_iops_fn) |
fe0714377
|
164 165 |
blkiop->ops.blkio_update_group_read_iops_fn(blkg->key, blkg, iops); |
7702e8f45
|
166 167 168 |
if (fileid == BLKIO_THROTL_write_iops_device && blkiop->ops.blkio_update_group_write_iops_fn) |
fe0714377
|
169 170 |
blkiop->ops.blkio_update_group_write_iops_fn(blkg->key, blkg,iops); |
7702e8f45
|
171 172 |
} } |
9195291e5
|
173 174 175 176 |
/* * Add to the appropriate stat variable depending on the request type. * This should be called with the blkg->stats_lock held. */ |
84c124da9
|
177 178 |
static void blkio_add_stat(uint64_t *stat, uint64_t add, bool direction, bool sync) |
9195291e5
|
179 |
{ |
84c124da9
|
180 181 |
if (direction) stat[BLKIO_STAT_WRITE] += add; |
9195291e5
|
182 |
else |
84c124da9
|
183 184 185 |
stat[BLKIO_STAT_READ] += add; if (sync) stat[BLKIO_STAT_SYNC] += add; |
9195291e5
|
186 |
else |
84c124da9
|
187 |
stat[BLKIO_STAT_ASYNC] += add; |
9195291e5
|
188 |
} |
cdc1184cf
|
189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 |
/* * 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 |
812df48d1
|
213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 |
/* 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); |
a11cdaa7a
|
282 |
void blkiocg_update_avg_queue_size_stats(struct blkio_group *blkg) |
cdc1184cf
|
283 284 285 286 287 288 289 290 291 292 |
{ 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++; |
812df48d1
|
293 |
blkio_update_group_wait_time(stats); |
cdc1184cf
|
294 295 |
spin_unlock_irqrestore(&blkg->stats_lock, flags); } |
a11cdaa7a
|
296 |
EXPORT_SYMBOL_GPL(blkiocg_update_avg_queue_size_stats); |
e5ff082e8
|
297 |
void blkiocg_set_start_empty_time(struct blkio_group *blkg) |
28baf4429
|
298 299 300 301 302 303 304 305 306 307 308 309 310 311 |
{ 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; } /* |
e5ff082e8
|
312 313 314 |
* group is already marked empty. This can happen if cfqq got new * request in parent group and moved to this group while being added * to service tree. Just ignore the event and move on. |
28baf4429
|
315 |
*/ |
e5ff082e8
|
316 317 318 319 |
if(blkio_blkg_empty(stats)) { spin_unlock_irqrestore(&blkg->stats_lock, flags); return; } |
28baf4429
|
320 321 322 323 324 |
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); |
a11cdaa7a
|
325 326 327 328 329 330 |
void blkiocg_update_dequeue_stats(struct blkio_group *blkg, unsigned long dequeue) { blkg->stats.dequeue += dequeue; } EXPORT_SYMBOL_GPL(blkiocg_update_dequeue_stats); |
812df48d1
|
331 332 333 334 |
#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) {} |
cdc1184cf
|
335 |
#endif |
a11cdaa7a
|
336 |
void blkiocg_update_io_add_stats(struct blkio_group *blkg, |
cdc1184cf
|
337 338 339 340 341 342 343 344 |
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); |
812df48d1
|
345 346 |
blkio_end_empty_time(&blkg->stats); blkio_set_start_group_wait_time(blkg, curr_blkg); |
cdc1184cf
|
347 348 |
spin_unlock_irqrestore(&blkg->stats_lock, flags); } |
a11cdaa7a
|
349 |
EXPORT_SYMBOL_GPL(blkiocg_update_io_add_stats); |
cdc1184cf
|
350 |
|
a11cdaa7a
|
351 |
void blkiocg_update_io_remove_stats(struct blkio_group *blkg, |
cdc1184cf
|
352 353 354 355 356 357 358 359 360 |
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); } |
a11cdaa7a
|
361 |
EXPORT_SYMBOL_GPL(blkiocg_update_io_remove_stats); |
cdc1184cf
|
362 |
|
167400d34
|
363 364 |
void blkiocg_update_timeslice_used(struct blkio_group *blkg, unsigned long time, unsigned long unaccounted_time) |
220841906
|
365 |
{ |
303a3acb2
|
366 367 368 369 |
unsigned long flags; spin_lock_irqsave(&blkg->stats_lock, flags); blkg->stats.time += time; |
a23e68695
|
370 |
#ifdef CONFIG_DEBUG_BLK_CGROUP |
167400d34
|
371 |
blkg->stats.unaccounted_time += unaccounted_time; |
a23e68695
|
372 |
#endif |
303a3acb2
|
373 |
spin_unlock_irqrestore(&blkg->stats_lock, flags); |
220841906
|
374 |
} |
303a3acb2
|
375 |
EXPORT_SYMBOL_GPL(blkiocg_update_timeslice_used); |
220841906
|
376 |
|
5624a4e44
|
377 378 379 380 |
/* * should be called under rcu read lock or queue lock to make sure blkg pointer * is valid. */ |
84c124da9
|
381 382 |
void blkiocg_update_dispatch_stats(struct blkio_group *blkg, uint64_t bytes, bool direction, bool sync) |
9195291e5
|
383 |
{ |
5624a4e44
|
384 |
struct blkio_group_stats_cpu *stats_cpu; |
575969a0d
|
385 386 387 388 389 390 391 392 |
unsigned long flags; /* * Disabling interrupts to provide mutual exclusion between two * writes on same cpu. It probably is not needed for 64bit. Not * optimizing that case yet. */ local_irq_save(flags); |
9195291e5
|
393 |
|
5624a4e44
|
394 |
stats_cpu = this_cpu_ptr(blkg->stats_cpu); |
575969a0d
|
395 |
u64_stats_update_begin(&stats_cpu->syncp); |
5624a4e44
|
396 397 398 399 400 |
stats_cpu->sectors += bytes >> 9; blkio_add_stat(stats_cpu->stat_arr_cpu[BLKIO_STAT_CPU_SERVICED], 1, direction, sync); blkio_add_stat(stats_cpu->stat_arr_cpu[BLKIO_STAT_CPU_SERVICE_BYTES], bytes, direction, sync); |
575969a0d
|
401 402 |
u64_stats_update_end(&stats_cpu->syncp); local_irq_restore(flags); |
9195291e5
|
403 |
} |
84c124da9
|
404 |
EXPORT_SYMBOL_GPL(blkiocg_update_dispatch_stats); |
9195291e5
|
405 |
|
84c124da9
|
406 407 |
void blkiocg_update_completion_stats(struct blkio_group *blkg, uint64_t start_time, uint64_t io_start_time, bool direction, bool sync) |
9195291e5
|
408 409 410 411 412 413 414 |
{ struct blkio_group_stats *stats; unsigned long flags; unsigned long long now = sched_clock(); spin_lock_irqsave(&blkg->stats_lock, flags); stats = &blkg->stats; |
84c124da9
|
415 416 417 418 419 420 |
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); |
9195291e5
|
421 422 |
spin_unlock_irqrestore(&blkg->stats_lock, flags); } |
84c124da9
|
423 |
EXPORT_SYMBOL_GPL(blkiocg_update_completion_stats); |
9195291e5
|
424 |
|
317389a77
|
425 |
/* Merged stats are per cpu. */ |
812d40264
|
426 427 428 |
void blkiocg_update_io_merged_stats(struct blkio_group *blkg, bool direction, bool sync) { |
317389a77
|
429 |
struct blkio_group_stats_cpu *stats_cpu; |
812d40264
|
430 |
unsigned long flags; |
317389a77
|
431 432 433 434 435 436 437 438 439 440 441 442 443 444 |
/* * Disabling interrupts to provide mutual exclusion between two * writes on same cpu. It probably is not needed for 64bit. Not * optimizing that case yet. */ local_irq_save(flags); stats_cpu = this_cpu_ptr(blkg->stats_cpu); u64_stats_update_begin(&stats_cpu->syncp); blkio_add_stat(stats_cpu->stat_arr_cpu[BLKIO_STAT_CPU_MERGED], 1, direction, sync); u64_stats_update_end(&stats_cpu->syncp); local_irq_restore(flags); |
812d40264
|
445 446 |
} EXPORT_SYMBOL_GPL(blkiocg_update_io_merged_stats); |
5624a4e44
|
447 448 449 450 451 452 453 454 455 456 457 458 459 |
/* * This function allocates the per cpu stats for blkio_group. Should be called * from sleepable context as alloc_per_cpu() requires that. */ int blkio_alloc_blkg_stats(struct blkio_group *blkg) { /* Allocate memory for per cpu stats */ blkg->stats_cpu = alloc_percpu(struct blkio_group_stats_cpu); if (!blkg->stats_cpu) return -ENOMEM; return 0; } EXPORT_SYMBOL_GPL(blkio_alloc_blkg_stats); |
31e4c28d9
|
460 |
void blkiocg_add_blkio_group(struct blkio_cgroup *blkcg, |
062a644d6
|
461 462 |
struct blkio_group *blkg, void *key, dev_t dev, enum blkio_policy_id plid) |
31e4c28d9
|
463 464 465 466 |
{ unsigned long flags; spin_lock_irqsave(&blkcg->lock, flags); |
8d2a91f89
|
467 |
spin_lock_init(&blkg->stats_lock); |
31e4c28d9
|
468 |
rcu_assign_pointer(blkg->key, key); |
b1c357696
|
469 |
blkg->blkcg_id = css_id(&blkcg->css); |
31e4c28d9
|
470 |
hlist_add_head_rcu(&blkg->blkcg_node, &blkcg->blkg_list); |
062a644d6
|
471 |
blkg->plid = plid; |
31e4c28d9
|
472 |
spin_unlock_irqrestore(&blkcg->lock, flags); |
2868ef7b3
|
473 474 |
/* Need to take css reference ? */ cgroup_path(blkcg->css.cgroup, blkg->path, sizeof(blkg->path)); |
220841906
|
475 |
blkg->dev = dev; |
31e4c28d9
|
476 |
} |
9d6a986c0
|
477 |
EXPORT_SYMBOL_GPL(blkiocg_add_blkio_group); |
31e4c28d9
|
478 |
|
b1c357696
|
479 480 481 482 483 484 485 486 487 488 |
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. */ |
31e4c28d9
|
489 490 |
int blkiocg_del_blkio_group(struct blkio_group *blkg) { |
b1c357696
|
491 492 493 494 495 496 497 |
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); |
0f3942a39
|
498 499 500 501 502 503 504 505 |
if (css) { 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); |
b1c357696
|
506 |
} |
0f3942a39
|
507 |
|
b1c357696
|
508 509 |
rcu_read_unlock(); return ret; |
31e4c28d9
|
510 |
} |
9d6a986c0
|
511 |
EXPORT_SYMBOL_GPL(blkiocg_del_blkio_group); |
31e4c28d9
|
512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 |
/* 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; } |
9d6a986c0
|
528 |
EXPORT_SYMBOL_GPL(blkiocg_lookup_group); |
31e4c28d9
|
529 |
|
f0bdc8cdd
|
530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 |
static void blkio_reset_stats_cpu(struct blkio_group *blkg) { struct blkio_group_stats_cpu *stats_cpu; int i, j, k; /* * Note: On 64 bit arch this should not be an issue. This has the * possibility of returning some inconsistent value on 32bit arch * as 64bit update on 32bit is non atomic. Taking care of this * corner case makes code very complicated, like sending IPIs to * cpus, taking care of stats of offline cpus etc. * * reset stats is anyway more of a debug feature and this sounds a * corner case. So I am not complicating the code yet until and * unless this becomes a real issue. */ for_each_possible_cpu(i) { stats_cpu = per_cpu_ptr(blkg->stats_cpu, i); stats_cpu->sectors = 0; for(j = 0; j < BLKIO_STAT_CPU_NR; j++) for (k = 0; k < BLKIO_STAT_TOTAL; k++) stats_cpu->stat_arr_cpu[j][k] = 0; } } |
303a3acb2
|
553 |
static int |
84c124da9
|
554 |
blkiocg_reset_stats(struct cgroup *cgroup, struct cftype *cftype, u64 val) |
303a3acb2
|
555 556 557 |
{ struct blkio_cgroup *blkcg; struct blkio_group *blkg; |
812df48d1
|
558 |
struct blkio_group_stats *stats; |
303a3acb2
|
559 |
struct hlist_node *n; |
cdc1184cf
|
560 561 |
uint64_t queued[BLKIO_STAT_TOTAL]; int i; |
812df48d1
|
562 563 564 565 |
#ifdef CONFIG_DEBUG_BLK_CGROUP bool idling, waiting, empty; unsigned long long now = sched_clock(); #endif |
303a3acb2
|
566 567 568 569 570 |
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); |
812df48d1
|
571 572 573 574 575 576 |
stats = &blkg->stats; #ifdef CONFIG_DEBUG_BLK_CGROUP idling = blkio_blkg_idling(stats); waiting = blkio_blkg_waiting(stats); empty = blkio_blkg_empty(stats); #endif |
cdc1184cf
|
577 |
for (i = 0; i < BLKIO_STAT_TOTAL; i++) |
812df48d1
|
578 579 |
queued[i] = stats->stat_arr[BLKIO_STAT_QUEUED][i]; memset(stats, 0, sizeof(struct blkio_group_stats)); |
cdc1184cf
|
580 |
for (i = 0; i < BLKIO_STAT_TOTAL; i++) |
812df48d1
|
581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 |
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 |
303a3acb2
|
596 |
spin_unlock(&blkg->stats_lock); |
f0bdc8cdd
|
597 598 599 |
/* Reset Per cpu stats which don't take blkg->stats_lock */ blkio_reset_stats_cpu(blkg); |
303a3acb2
|
600 |
} |
f0bdc8cdd
|
601 |
|
303a3acb2
|
602 603 604 |
spin_unlock_irq(&blkcg->lock); return 0; } |
84c124da9
|
605 606 |
static void blkio_get_key_name(enum stat_sub_type type, dev_t dev, char *str, int chars_left, bool diskname_only) |
303a3acb2
|
607 |
{ |
84c124da9
|
608 |
snprintf(str, chars_left, "%d:%d", MAJOR(dev), MINOR(dev)); |
303a3acb2
|
609 610 611 612 613 614 |
chars_left -= strlen(str); if (chars_left <= 0) { printk(KERN_WARNING "Possibly incorrect cgroup stat display format"); return; } |
84c124da9
|
615 616 |
if (diskname_only) return; |
303a3acb2
|
617 |
switch (type) { |
84c124da9
|
618 |
case BLKIO_STAT_READ: |
303a3acb2
|
619 620 |
strlcat(str, " Read", chars_left); break; |
84c124da9
|
621 |
case BLKIO_STAT_WRITE: |
303a3acb2
|
622 623 |
strlcat(str, " Write", chars_left); break; |
84c124da9
|
624 |
case BLKIO_STAT_SYNC: |
303a3acb2
|
625 626 |
strlcat(str, " Sync", chars_left); break; |
84c124da9
|
627 |
case BLKIO_STAT_ASYNC: |
303a3acb2
|
628 629 |
strlcat(str, " Async", chars_left); break; |
84c124da9
|
630 |
case BLKIO_STAT_TOTAL: |
303a3acb2
|
631 632 633 634 635 636 |
strlcat(str, " Total", chars_left); break; default: strlcat(str, " Invalid", chars_left); } } |
84c124da9
|
637 638 639 640 641 642 643 |
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; } |
303a3acb2
|
644 |
|
5624a4e44
|
645 646 647 648 649 650 |
static uint64_t blkio_read_stat_cpu(struct blkio_group *blkg, enum stat_type_cpu type, enum stat_sub_type sub_type) { int cpu; struct blkio_group_stats_cpu *stats_cpu; |
575969a0d
|
651 |
u64 val = 0, tval; |
5624a4e44
|
652 653 |
for_each_possible_cpu(cpu) { |
575969a0d
|
654 |
unsigned int start; |
5624a4e44
|
655 |
stats_cpu = per_cpu_ptr(blkg->stats_cpu, cpu); |
575969a0d
|
656 657 658 659 660 661 662 663 664 |
do { start = u64_stats_fetch_begin(&stats_cpu->syncp); if (type == BLKIO_STAT_CPU_SECTORS) tval = stats_cpu->sectors; else tval = stats_cpu->stat_arr_cpu[type][sub_type]; } while(u64_stats_fetch_retry(&stats_cpu->syncp, start)); val += tval; |
5624a4e44
|
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 |
} return val; } static uint64_t blkio_get_stat_cpu(struct blkio_group *blkg, struct cgroup_map_cb *cb, dev_t dev, enum stat_type_cpu type) { uint64_t disk_total, val; char key_str[MAX_KEY_LEN]; enum stat_sub_type sub_type; if (type == BLKIO_STAT_CPU_SECTORS) { val = blkio_read_stat_cpu(blkg, type, 0); return blkio_fill_stat(key_str, MAX_KEY_LEN - 1, val, cb, dev); } 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); val = blkio_read_stat_cpu(blkg, type, sub_type); cb->fill(cb, key_str, val); } disk_total = blkio_read_stat_cpu(blkg, type, BLKIO_STAT_READ) + blkio_read_stat_cpu(blkg, type, BLKIO_STAT_WRITE); blkio_get_key_name(BLKIO_STAT_TOTAL, dev, key_str, MAX_KEY_LEN, false); cb->fill(cb, key_str, disk_total); return disk_total; } |
84c124da9
|
696 697 698 |
/* 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) |
303a3acb2
|
699 700 701 |
{ uint64_t disk_total; char key_str[MAX_KEY_LEN]; |
84c124da9
|
702 703 704 705 706 |
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); |
9026e521c
|
707 |
#ifdef CONFIG_DEBUG_BLK_CGROUP |
167400d34
|
708 709 710 |
if (type == BLKIO_STAT_UNACCOUNTED_TIME) return blkio_fill_stat(key_str, MAX_KEY_LEN - 1, blkg->stats.unaccounted_time, cb, dev); |
cdc1184cf
|
711 712 713 714 715 716 717 718 719 |
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); } |
812df48d1
|
720 721 722 723 724 725 726 727 728 |
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); |
84c124da9
|
729 730 731 732 |
if (type == BLKIO_STAT_DEQUEUE) return blkio_fill_stat(key_str, MAX_KEY_LEN - 1, blkg->stats.dequeue, cb, dev); #endif |
303a3acb2
|
733 |
|
84c124da9
|
734 735 736 737 |
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]); |
303a3acb2
|
738 |
} |
84c124da9
|
739 740 741 |
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); |
303a3acb2
|
742 743 744 |
cb->fill(cb, key_str, disk_total); return disk_total; } |
34d0f179d
|
745 |
static int blkio_policy_parse_and_set(char *buf, |
062a644d6
|
746 |
struct blkio_policy_node *newpn, enum blkio_policy_id plid, int fileid) |
34d0f179d
|
747 |
{ |
ece84241b
|
748 |
struct gendisk *disk = NULL; |
34d0f179d
|
749 |
char *s[4], *p, *major_s = NULL, *minor_s = NULL; |
d11bb4462
|
750 |
unsigned long major, minor; |
ece84241b
|
751 752 |
int i = 0, ret = -EINVAL; int part; |
34d0f179d
|
753 |
dev_t dev; |
d11bb4462
|
754 |
u64 temp; |
34d0f179d
|
755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 |
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) |
ece84241b
|
770 |
goto out; |
34d0f179d
|
771 772 773 774 775 |
p = strsep(&s[0], ":"); if (p != NULL) major_s = p; else |
ece84241b
|
776 |
goto out; |
34d0f179d
|
777 778 779 |
minor_s = s[0]; if (!minor_s) |
ece84241b
|
780 |
goto out; |
34d0f179d
|
781 |
|
ece84241b
|
782 783 |
if (strict_strtoul(major_s, 10, &major)) goto out; |
34d0f179d
|
784 |
|
ece84241b
|
785 786 |
if (strict_strtoul(minor_s, 10, &minor)) goto out; |
34d0f179d
|
787 788 |
dev = MKDEV(major, minor); |
ece84241b
|
789 790 |
if (strict_strtoull(s[1], 10, &temp)) goto out; |
34d0f179d
|
791 |
|
d11bb4462
|
792 793 |
/* For rule removal, do not check for device presence. */ if (temp) { |
ece84241b
|
794 795 796 797 798 |
disk = get_gendisk(dev, &part); if (!disk || part) { ret = -ENODEV; goto out; } |
d11bb4462
|
799 |
} |
34d0f179d
|
800 |
|
d11bb4462
|
801 |
newpn->dev = dev; |
34d0f179d
|
802 |
|
062a644d6
|
803 804 |
switch (plid) { case BLKIO_POLICY_PROP: |
d11bb4462
|
805 806 |
if ((temp < BLKIO_WEIGHT_MIN && temp > 0) || temp > BLKIO_WEIGHT_MAX) |
ece84241b
|
807 |
goto out; |
34d0f179d
|
808 |
|
062a644d6
|
809 810 |
newpn->plid = plid; newpn->fileid = fileid; |
4c9eefa16
|
811 812 813 |
newpn->val.weight = temp; break; case BLKIO_POLICY_THROTL: |
7702e8f45
|
814 815 816 |
switch(fileid) { case BLKIO_THROTL_read_bps_device: case BLKIO_THROTL_write_bps_device: |
7702e8f45
|
817 818 |
newpn->plid = plid; newpn->fileid = fileid; |
d11bb4462
|
819 |
newpn->val.bps = temp; |
7702e8f45
|
820 821 822 |
break; case BLKIO_THROTL_read_iops_device: case BLKIO_THROTL_write_iops_device: |
d11bb4462
|
823 |
if (temp > THROTL_IOPS_MAX) |
ece84241b
|
824 |
goto out; |
9355aede5
|
825 |
|
7702e8f45
|
826 827 |
newpn->plid = plid; newpn->fileid = fileid; |
d11bb4462
|
828 |
newpn->val.iops = (unsigned int)temp; |
7702e8f45
|
829 830 |
break; } |
062a644d6
|
831 832 833 834 |
break; default: BUG(); } |
ece84241b
|
835 836 837 838 |
ret = 0; out: put_disk(disk); return ret; |
34d0f179d
|
839 840 841 842 843 844 |
} unsigned int blkcg_get_weight(struct blkio_cgroup *blkcg, dev_t dev) { struct blkio_policy_node *pn; |
a38eb630f
|
845 846 847 848 |
unsigned long flags; unsigned int weight; spin_lock_irqsave(&blkcg->lock, flags); |
34d0f179d
|
849 |
|
062a644d6
|
850 851 |
pn = blkio_policy_search_node(blkcg, dev, BLKIO_POLICY_PROP, BLKIO_PROP_weight_device); |
34d0f179d
|
852 |
if (pn) |
a38eb630f
|
853 |
weight = pn->val.weight; |
34d0f179d
|
854 |
else |
a38eb630f
|
855 856 857 858 859 |
weight = blkcg->weight; spin_unlock_irqrestore(&blkcg->lock, flags); return weight; |
34d0f179d
|
860 861 |
} EXPORT_SYMBOL_GPL(blkcg_get_weight); |
4c9eefa16
|
862 863 864 |
uint64_t blkcg_get_read_bps(struct blkio_cgroup *blkcg, dev_t dev) { struct blkio_policy_node *pn; |
a38eb630f
|
865 866 |
unsigned long flags; uint64_t bps = -1; |
4c9eefa16
|
867 |
|
a38eb630f
|
868 |
spin_lock_irqsave(&blkcg->lock, flags); |
4c9eefa16
|
869 870 871 |
pn = blkio_policy_search_node(blkcg, dev, BLKIO_POLICY_THROTL, BLKIO_THROTL_read_bps_device); if (pn) |
a38eb630f
|
872 873 874 875 |
bps = pn->val.bps; spin_unlock_irqrestore(&blkcg->lock, flags); return bps; |
4c9eefa16
|
876 877 878 879 880 |
} uint64_t blkcg_get_write_bps(struct blkio_cgroup *blkcg, dev_t dev) { struct blkio_policy_node *pn; |
a38eb630f
|
881 882 883 884 |
unsigned long flags; uint64_t bps = -1; spin_lock_irqsave(&blkcg->lock, flags); |
4c9eefa16
|
885 886 887 |
pn = blkio_policy_search_node(blkcg, dev, BLKIO_POLICY_THROTL, BLKIO_THROTL_write_bps_device); if (pn) |
a38eb630f
|
888 889 890 891 |
bps = pn->val.bps; spin_unlock_irqrestore(&blkcg->lock, flags); return bps; |
4c9eefa16
|
892 |
} |
7702e8f45
|
893 894 895 |
unsigned int blkcg_get_read_iops(struct blkio_cgroup *blkcg, dev_t dev) { struct blkio_policy_node *pn; |
a38eb630f
|
896 897 |
unsigned long flags; unsigned int iops = -1; |
7702e8f45
|
898 |
|
a38eb630f
|
899 |
spin_lock_irqsave(&blkcg->lock, flags); |
7702e8f45
|
900 901 902 |
pn = blkio_policy_search_node(blkcg, dev, BLKIO_POLICY_THROTL, BLKIO_THROTL_read_iops_device); if (pn) |
a38eb630f
|
903 904 905 906 |
iops = pn->val.iops; spin_unlock_irqrestore(&blkcg->lock, flags); return iops; |
7702e8f45
|
907 908 909 910 911 |
} unsigned int blkcg_get_write_iops(struct blkio_cgroup *blkcg, dev_t dev) { struct blkio_policy_node *pn; |
a38eb630f
|
912 913 914 915 |
unsigned long flags; unsigned int iops = -1; spin_lock_irqsave(&blkcg->lock, flags); |
7702e8f45
|
916 917 918 |
pn = blkio_policy_search_node(blkcg, dev, BLKIO_POLICY_THROTL, BLKIO_THROTL_write_iops_device); if (pn) |
a38eb630f
|
919 920 921 922 |
iops = pn->val.iops; spin_unlock_irqrestore(&blkcg->lock, flags); return iops; |
7702e8f45
|
923 |
} |
062a644d6
|
924 925 926 927 928 |
/* Checks whether user asked for deleting a policy rule */ static bool blkio_delete_rule_command(struct blkio_policy_node *pn) { switch(pn->plid) { case BLKIO_POLICY_PROP: |
4c9eefa16
|
929 930 931 932 |
if (pn->val.weight == 0) return 1; break; case BLKIO_POLICY_THROTL: |
7702e8f45
|
933 934 935 936 937 938 939 940 941 942 943 |
switch(pn->fileid) { case BLKIO_THROTL_read_bps_device: case BLKIO_THROTL_write_bps_device: if (pn->val.bps == 0) return 1; break; case BLKIO_THROTL_read_iops_device: case BLKIO_THROTL_write_iops_device: if (pn->val.iops == 0) return 1; } |
062a644d6
|
944 945 946 947 948 949 950 951 952 953 954 955 956 |
break; default: BUG(); } return 0; } static void blkio_update_policy_rule(struct blkio_policy_node *oldpn, struct blkio_policy_node *newpn) { switch(oldpn->plid) { case BLKIO_POLICY_PROP: |
4c9eefa16
|
957 958 959 |
oldpn->val.weight = newpn->val.weight; break; case BLKIO_POLICY_THROTL: |
7702e8f45
|
960 961 962 963 964 965 966 967 968 |
switch(newpn->fileid) { case BLKIO_THROTL_read_bps_device: case BLKIO_THROTL_write_bps_device: oldpn->val.bps = newpn->val.bps; break; case BLKIO_THROTL_read_iops_device: case BLKIO_THROTL_write_iops_device: oldpn->val.iops = newpn->val.iops; } |
062a644d6
|
969 970 971 972 973 974 975 |
break; default: BUG(); } } /* |
25985edce
|
976 |
* Some rules/values in blkg have changed. Propagate those to respective |
062a644d6
|
977 978 979 980 981 |
* policies. */ static void blkio_update_blkg_policy(struct blkio_cgroup *blkcg, struct blkio_group *blkg, struct blkio_policy_node *pn) { |
7702e8f45
|
982 |
unsigned int weight, iops; |
4c9eefa16
|
983 |
u64 bps; |
062a644d6
|
984 985 986 |
switch(pn->plid) { case BLKIO_POLICY_PROP: |
4c9eefa16
|
987 |
weight = pn->val.weight ? pn->val.weight : |
062a644d6
|
988 989 990 |
blkcg->weight; blkio_update_group_weight(blkg, weight); break; |
4c9eefa16
|
991 992 993 994 995 996 997 |
case BLKIO_POLICY_THROTL: switch(pn->fileid) { case BLKIO_THROTL_read_bps_device: case BLKIO_THROTL_write_bps_device: bps = pn->val.bps ? pn->val.bps : (-1); blkio_update_group_bps(blkg, bps, pn->fileid); break; |
7702e8f45
|
998 999 1000 1001 1002 |
case BLKIO_THROTL_read_iops_device: case BLKIO_THROTL_write_iops_device: iops = pn->val.iops ? pn->val.iops : (-1); blkio_update_group_iops(blkg, iops, pn->fileid); break; |
4c9eefa16
|
1003 1004 |
} break; |
062a644d6
|
1005 1006 1007 1008 1009 1010 |
default: BUG(); } } /* |
25985edce
|
1011 |
* A policy node rule has been updated. Propagate this update to all the |
062a644d6
|
1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 |
* block groups which might be affected by this update. */ static void blkio_update_policy_node_blkg(struct blkio_cgroup *blkcg, struct blkio_policy_node *pn) { struct blkio_group *blkg; struct hlist_node *n; spin_lock(&blkio_list_lock); spin_lock_irq(&blkcg->lock); |
34d0f179d
|
1022 |
|
062a644d6
|
1023 1024 1025 1026 1027 1028 1029 1030 1031 |
hlist_for_each_entry(blkg, n, &blkcg->blkg_list, blkcg_node) { if (pn->dev != blkg->dev || pn->plid != blkg->plid) continue; blkio_update_blkg_policy(blkcg, blkg, pn); } spin_unlock_irq(&blkcg->lock); spin_unlock(&blkio_list_lock); } |
34d0f179d
|
1032 |
|
062a644d6
|
1033 1034 |
static int blkiocg_file_write(struct cgroup *cgrp, struct cftype *cft, const char *buffer) |
34d0f179d
|
1035 1036 1037 1038 1039 |
{ int ret = 0; char *buf; struct blkio_policy_node *newpn, *pn; struct blkio_cgroup *blkcg; |
34d0f179d
|
1040 |
int keep_newpn = 0; |
062a644d6
|
1041 1042 |
enum blkio_policy_id plid = BLKIOFILE_POLICY(cft->private); int fileid = BLKIOFILE_ATTR(cft->private); |
34d0f179d
|
1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 |
buf = kstrdup(buffer, GFP_KERNEL); if (!buf) return -ENOMEM; newpn = kzalloc(sizeof(*newpn), GFP_KERNEL); if (!newpn) { ret = -ENOMEM; goto free_buf; } |
062a644d6
|
1053 |
ret = blkio_policy_parse_and_set(buf, newpn, plid, fileid); |
34d0f179d
|
1054 1055 1056 1057 1058 1059 |
if (ret) goto free_newpn; blkcg = cgroup_to_blkio_cgroup(cgrp); spin_lock_irq(&blkcg->lock); |
062a644d6
|
1060 |
pn = blkio_policy_search_node(blkcg, newpn->dev, plid, fileid); |
34d0f179d
|
1061 |
if (!pn) { |
062a644d6
|
1062 |
if (!blkio_delete_rule_command(newpn)) { |
34d0f179d
|
1063 1064 1065 1066 1067 1068 |
blkio_policy_insert_node(blkcg, newpn); keep_newpn = 1; } spin_unlock_irq(&blkcg->lock); goto update_io_group; } |
062a644d6
|
1069 |
if (blkio_delete_rule_command(newpn)) { |
34d0f179d
|
1070 |
blkio_policy_delete_node(pn); |
e060f00be
|
1071 |
kfree(pn); |
34d0f179d
|
1072 1073 1074 1075 |
spin_unlock_irq(&blkcg->lock); goto update_io_group; } spin_unlock_irq(&blkcg->lock); |
062a644d6
|
1076 |
blkio_update_policy_rule(pn, newpn); |
34d0f179d
|
1077 1078 |
update_io_group: |
062a644d6
|
1079 |
blkio_update_policy_node_blkg(blkcg, newpn); |
34d0f179d
|
1080 1081 1082 1083 1084 1085 1086 1087 |
free_newpn: if (!keep_newpn) kfree(newpn); free_buf: kfree(buf); return ret; } |
062a644d6
|
1088 1089 |
static void blkio_print_policy_node(struct seq_file *m, struct blkio_policy_node *pn) |
34d0f179d
|
1090 |
{ |
062a644d6
|
1091 1092 1093 1094 1095 |
switch(pn->plid) { case BLKIO_POLICY_PROP: if (pn->fileid == BLKIO_PROP_weight_device) seq_printf(m, "%u:%u\t%u ", MAJOR(pn->dev), |
4c9eefa16
|
1096 1097 1098 |
MINOR(pn->dev), pn->val.weight); break; case BLKIO_POLICY_THROTL: |
7702e8f45
|
1099 1100 1101 |
switch(pn->fileid) { case BLKIO_THROTL_read_bps_device: case BLKIO_THROTL_write_bps_device: |
4c9eefa16
|
1102 1103 1104 |
seq_printf(m, "%u:%u\t%llu ", MAJOR(pn->dev), MINOR(pn->dev), pn->val.bps); |
7702e8f45
|
1105 1106 1107 1108 1109 1110 1111 1112 |
break; case BLKIO_THROTL_read_iops_device: case BLKIO_THROTL_write_iops_device: seq_printf(m, "%u:%u\t%u ", MAJOR(pn->dev), MINOR(pn->dev), pn->val.iops); break; } |
062a644d6
|
1113 1114 1115 1116 1117 |
break; default: BUG(); } } |
34d0f179d
|
1118 |
|
062a644d6
|
1119 1120 1121 |
/* cgroup files which read their data from policy nodes end up here */ static void blkio_read_policy_node_files(struct cftype *cft, struct blkio_cgroup *blkcg, struct seq_file *m) |
34d0f179d
|
1122 |
{ |
34d0f179d
|
1123 |
struct blkio_policy_node *pn; |
34d0f179d
|
1124 |
|
0f3942a39
|
1125 1126 1127 |
if (!list_empty(&blkcg->policy_list)) { spin_lock_irq(&blkcg->lock); list_for_each_entry(pn, &blkcg->policy_list, node) { |
062a644d6
|
1128 1129 1130 |
if (!pn_matches_cftype(cft, pn)) continue; blkio_print_policy_node(m, pn); |
0f3942a39
|
1131 1132 |
} spin_unlock_irq(&blkcg->lock); |
34d0f179d
|
1133 |
} |
062a644d6
|
1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 |
} static int blkiocg_file_read(struct cgroup *cgrp, struct cftype *cft, struct seq_file *m) { struct blkio_cgroup *blkcg; enum blkio_policy_id plid = BLKIOFILE_POLICY(cft->private); int name = BLKIOFILE_ATTR(cft->private); blkcg = cgroup_to_blkio_cgroup(cgrp); switch(plid) { case BLKIO_POLICY_PROP: switch(name) { case BLKIO_PROP_weight_device: blkio_read_policy_node_files(cft, blkcg, m); return 0; default: BUG(); } break; |
4c9eefa16
|
1155 1156 1157 1158 |
case BLKIO_POLICY_THROTL: switch(name){ case BLKIO_THROTL_read_bps_device: case BLKIO_THROTL_write_bps_device: |
7702e8f45
|
1159 1160 |
case BLKIO_THROTL_read_iops_device: case BLKIO_THROTL_write_iops_device: |
4c9eefa16
|
1161 1162 1163 1164 1165 1166 |
blkio_read_policy_node_files(cft, blkcg, m); return 0; default: BUG(); } break; |
062a644d6
|
1167 1168 1169 1170 1171 1172 1173 1174 |
default: BUG(); } return 0; } static int blkio_read_blkg_stats(struct blkio_cgroup *blkcg, |
5624a4e44
|
1175 1176 |
struct cftype *cft, struct cgroup_map_cb *cb, enum stat_type type, bool show_total, bool pcpu) |
062a644d6
|
1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 |
{ struct blkio_group *blkg; struct hlist_node *n; uint64_t cgroup_total = 0; rcu_read_lock(); hlist_for_each_entry_rcu(blkg, n, &blkcg->blkg_list, blkcg_node) { if (blkg->dev) { if (!cftype_blkg_same_policy(cft, blkg)) continue; |
5624a4e44
|
1187 1188 1189 1190 1191 1192 1193 1194 1195 |
if (pcpu) cgroup_total += blkio_get_stat_cpu(blkg, cb, blkg->dev, type); else { spin_lock_irq(&blkg->stats_lock); cgroup_total += blkio_get_stat(blkg, cb, blkg->dev, type); spin_unlock_irq(&blkg->stats_lock); } |
062a644d6
|
1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 |
} } if (show_total) cb->fill(cb, "Total", cgroup_total); rcu_read_unlock(); return 0; } /* All map kind of cgroup file get serviced by this function */ static int blkiocg_file_read_map(struct cgroup *cgrp, struct cftype *cft, struct cgroup_map_cb *cb) { struct blkio_cgroup *blkcg; enum blkio_policy_id plid = BLKIOFILE_POLICY(cft->private); int name = BLKIOFILE_ATTR(cft->private); blkcg = cgroup_to_blkio_cgroup(cgrp); switch(plid) { case BLKIO_POLICY_PROP: switch(name) { case BLKIO_PROP_time: return blkio_read_blkg_stats(blkcg, cft, cb, |
5624a4e44
|
1219 |
BLKIO_STAT_TIME, 0, 0); |
062a644d6
|
1220 1221 |
case BLKIO_PROP_sectors: return blkio_read_blkg_stats(blkcg, cft, cb, |
5624a4e44
|
1222 |
BLKIO_STAT_CPU_SECTORS, 0, 1); |
062a644d6
|
1223 1224 |
case BLKIO_PROP_io_service_bytes: return blkio_read_blkg_stats(blkcg, cft, cb, |
5624a4e44
|
1225 |
BLKIO_STAT_CPU_SERVICE_BYTES, 1, 1); |
062a644d6
|
1226 1227 |
case BLKIO_PROP_io_serviced: return blkio_read_blkg_stats(blkcg, cft, cb, |
5624a4e44
|
1228 |
BLKIO_STAT_CPU_SERVICED, 1, 1); |
062a644d6
|
1229 1230 |
case BLKIO_PROP_io_service_time: return blkio_read_blkg_stats(blkcg, cft, cb, |
5624a4e44
|
1231 |
BLKIO_STAT_SERVICE_TIME, 1, 0); |
062a644d6
|
1232 1233 |
case BLKIO_PROP_io_wait_time: return blkio_read_blkg_stats(blkcg, cft, cb, |
5624a4e44
|
1234 |
BLKIO_STAT_WAIT_TIME, 1, 0); |
062a644d6
|
1235 1236 |
case BLKIO_PROP_io_merged: return blkio_read_blkg_stats(blkcg, cft, cb, |
317389a77
|
1237 |
BLKIO_STAT_CPU_MERGED, 1, 1); |
062a644d6
|
1238 1239 |
case BLKIO_PROP_io_queued: return blkio_read_blkg_stats(blkcg, cft, cb, |
5624a4e44
|
1240 |
BLKIO_STAT_QUEUED, 1, 0); |
062a644d6
|
1241 |
#ifdef CONFIG_DEBUG_BLK_CGROUP |
9026e521c
|
1242 1243 |
case BLKIO_PROP_unaccounted_time: return blkio_read_blkg_stats(blkcg, cft, cb, |
5624a4e44
|
1244 |
BLKIO_STAT_UNACCOUNTED_TIME, 0, 0); |
062a644d6
|
1245 1246 |
case BLKIO_PROP_dequeue: return blkio_read_blkg_stats(blkcg, cft, cb, |
5624a4e44
|
1247 |
BLKIO_STAT_DEQUEUE, 0, 0); |
062a644d6
|
1248 1249 |
case BLKIO_PROP_avg_queue_size: return blkio_read_blkg_stats(blkcg, cft, cb, |
5624a4e44
|
1250 |
BLKIO_STAT_AVG_QUEUE_SIZE, 0, 0); |
062a644d6
|
1251 1252 |
case BLKIO_PROP_group_wait_time: return blkio_read_blkg_stats(blkcg, cft, cb, |
5624a4e44
|
1253 |
BLKIO_STAT_GROUP_WAIT_TIME, 0, 0); |
062a644d6
|
1254 1255 |
case BLKIO_PROP_idle_time: return blkio_read_blkg_stats(blkcg, cft, cb, |
5624a4e44
|
1256 |
BLKIO_STAT_IDLE_TIME, 0, 0); |
062a644d6
|
1257 1258 |
case BLKIO_PROP_empty_time: return blkio_read_blkg_stats(blkcg, cft, cb, |
5624a4e44
|
1259 |
BLKIO_STAT_EMPTY_TIME, 0, 0); |
062a644d6
|
1260 1261 1262 1263 1264 |
#endif default: BUG(); } break; |
4c9eefa16
|
1265 1266 1267 1268 |
case BLKIO_POLICY_THROTL: switch(name){ case BLKIO_THROTL_io_service_bytes: return blkio_read_blkg_stats(blkcg, cft, cb, |
5624a4e44
|
1269 |
BLKIO_STAT_CPU_SERVICE_BYTES, 1, 1); |
4c9eefa16
|
1270 1271 |
case BLKIO_THROTL_io_serviced: return blkio_read_blkg_stats(blkcg, cft, cb, |
5624a4e44
|
1272 |
BLKIO_STAT_CPU_SERVICED, 1, 1); |
4c9eefa16
|
1273 1274 1275 1276 |
default: BUG(); } break; |
062a644d6
|
1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 |
default: BUG(); } return 0; } static int blkio_weight_write(struct blkio_cgroup *blkcg, u64 val) { struct blkio_group *blkg; struct hlist_node *n; struct blkio_policy_node *pn; if (val < BLKIO_WEIGHT_MIN || val > BLKIO_WEIGHT_MAX) return -EINVAL; spin_lock(&blkio_list_lock); spin_lock_irq(&blkcg->lock); blkcg->weight = (unsigned int)val; hlist_for_each_entry(blkg, n, &blkcg->blkg_list, blkcg_node) { pn = blkio_policy_search_node(blkcg, blkg->dev, BLKIO_POLICY_PROP, BLKIO_PROP_weight_device); if (pn) continue; blkio_update_group_weight(blkg, blkcg->weight); } spin_unlock_irq(&blkcg->lock); spin_unlock(&blkio_list_lock); return 0; } static u64 blkiocg_file_read_u64 (struct cgroup *cgrp, struct cftype *cft) { struct blkio_cgroup *blkcg; enum blkio_policy_id plid = BLKIOFILE_POLICY(cft->private); int name = BLKIOFILE_ATTR(cft->private); blkcg = cgroup_to_blkio_cgroup(cgrp); switch(plid) { case BLKIO_POLICY_PROP: switch(name) { case BLKIO_PROP_weight: return (u64)blkcg->weight; } break; default: BUG(); } return 0; } static int blkiocg_file_write_u64(struct cgroup *cgrp, struct cftype *cft, u64 val) { struct blkio_cgroup *blkcg; enum blkio_policy_id plid = BLKIOFILE_POLICY(cft->private); int name = BLKIOFILE_ATTR(cft->private); blkcg = cgroup_to_blkio_cgroup(cgrp); switch(plid) { case BLKIO_POLICY_PROP: switch(name) { case BLKIO_PROP_weight: return blkio_weight_write(blkcg, val); } break; default: BUG(); } |
34d0f179d
|
1349 |
|
34d0f179d
|
1350 1351 |
return 0; } |
31e4c28d9
|
1352 1353 |
struct cftype blkio_files[] = { { |
34d0f179d
|
1354 |
.name = "weight_device", |
062a644d6
|
1355 1356 1357 1358 |
.private = BLKIOFILE_PRIVATE(BLKIO_POLICY_PROP, BLKIO_PROP_weight_device), .read_seq_string = blkiocg_file_read, .write_string = blkiocg_file_write, |
34d0f179d
|
1359 1360 1361 |
.max_write_len = 256, }, { |
31e4c28d9
|
1362 |
.name = "weight", |
062a644d6
|
1363 1364 1365 1366 |
.private = BLKIOFILE_PRIVATE(BLKIO_POLICY_PROP, BLKIO_PROP_weight), .read_u64 = blkiocg_file_read_u64, .write_u64 = blkiocg_file_write_u64, |
31e4c28d9
|
1367 |
}, |
220841906
|
1368 1369 |
{ .name = "time", |
062a644d6
|
1370 1371 1372 |
.private = BLKIOFILE_PRIVATE(BLKIO_POLICY_PROP, BLKIO_PROP_time), .read_map = blkiocg_file_read_map, |
220841906
|
1373 1374 1375 |
}, { .name = "sectors", |
062a644d6
|
1376 1377 1378 |
.private = BLKIOFILE_PRIVATE(BLKIO_POLICY_PROP, BLKIO_PROP_sectors), .read_map = blkiocg_file_read_map, |
303a3acb2
|
1379 1380 1381 |
}, { .name = "io_service_bytes", |
062a644d6
|
1382 1383 1384 |
.private = BLKIOFILE_PRIVATE(BLKIO_POLICY_PROP, BLKIO_PROP_io_service_bytes), .read_map = blkiocg_file_read_map, |
303a3acb2
|
1385 1386 1387 |
}, { .name = "io_serviced", |
062a644d6
|
1388 1389 1390 |
.private = BLKIOFILE_PRIVATE(BLKIO_POLICY_PROP, BLKIO_PROP_io_serviced), .read_map = blkiocg_file_read_map, |
303a3acb2
|
1391 1392 1393 |
}, { .name = "io_service_time", |
062a644d6
|
1394 1395 1396 |
.private = BLKIOFILE_PRIVATE(BLKIO_POLICY_PROP, BLKIO_PROP_io_service_time), .read_map = blkiocg_file_read_map, |
303a3acb2
|
1397 1398 1399 |
}, { .name = "io_wait_time", |
062a644d6
|
1400 1401 1402 |
.private = BLKIOFILE_PRIVATE(BLKIO_POLICY_PROP, BLKIO_PROP_io_wait_time), .read_map = blkiocg_file_read_map, |
84c124da9
|
1403 1404 |
}, { |
812d40264
|
1405 |
.name = "io_merged", |
062a644d6
|
1406 1407 1408 |
.private = BLKIOFILE_PRIVATE(BLKIO_POLICY_PROP, BLKIO_PROP_io_merged), .read_map = blkiocg_file_read_map, |
812d40264
|
1409 1410 |
}, { |
cdc1184cf
|
1411 |
.name = "io_queued", |
062a644d6
|
1412 1413 1414 |
.private = BLKIOFILE_PRIVATE(BLKIO_POLICY_PROP, BLKIO_PROP_io_queued), .read_map = blkiocg_file_read_map, |
cdc1184cf
|
1415 1416 |
}, { |
84c124da9
|
1417 1418 |
.name = "reset_stats", .write_u64 = blkiocg_reset_stats, |
220841906
|
1419 |
}, |
13f98250f
|
1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 |
#ifdef CONFIG_BLK_DEV_THROTTLING { .name = "throttle.read_bps_device", .private = BLKIOFILE_PRIVATE(BLKIO_POLICY_THROTL, BLKIO_THROTL_read_bps_device), .read_seq_string = blkiocg_file_read, .write_string = blkiocg_file_write, .max_write_len = 256, }, { .name = "throttle.write_bps_device", .private = BLKIOFILE_PRIVATE(BLKIO_POLICY_THROTL, BLKIO_THROTL_write_bps_device), .read_seq_string = blkiocg_file_read, .write_string = blkiocg_file_write, .max_write_len = 256, }, { .name = "throttle.read_iops_device", .private = BLKIOFILE_PRIVATE(BLKIO_POLICY_THROTL, BLKIO_THROTL_read_iops_device), .read_seq_string = blkiocg_file_read, .write_string = blkiocg_file_write, .max_write_len = 256, }, { .name = "throttle.write_iops_device", .private = BLKIOFILE_PRIVATE(BLKIO_POLICY_THROTL, BLKIO_THROTL_write_iops_device), .read_seq_string = blkiocg_file_read, .write_string = blkiocg_file_write, .max_write_len = 256, }, { .name = "throttle.io_service_bytes", .private = BLKIOFILE_PRIVATE(BLKIO_POLICY_THROTL, BLKIO_THROTL_io_service_bytes), .read_map = blkiocg_file_read_map, }, { .name = "throttle.io_serviced", .private = BLKIOFILE_PRIVATE(BLKIO_POLICY_THROTL, BLKIO_THROTL_io_serviced), .read_map = blkiocg_file_read_map, }, #endif /* CONFIG_BLK_DEV_THROTTLING */ |
220841906
|
1469 |
#ifdef CONFIG_DEBUG_BLK_CGROUP |
cdc1184cf
|
1470 1471 |
{ .name = "avg_queue_size", |
062a644d6
|
1472 1473 1474 |
.private = BLKIOFILE_PRIVATE(BLKIO_POLICY_PROP, BLKIO_PROP_avg_queue_size), .read_map = blkiocg_file_read_map, |
cdc1184cf
|
1475 1476 |
}, { |
812df48d1
|
1477 |
.name = "group_wait_time", |
062a644d6
|
1478 1479 1480 |
.private = BLKIOFILE_PRIVATE(BLKIO_POLICY_PROP, BLKIO_PROP_group_wait_time), .read_map = blkiocg_file_read_map, |
812df48d1
|
1481 1482 1483 |
}, { .name = "idle_time", |
062a644d6
|
1484 1485 1486 |
.private = BLKIOFILE_PRIVATE(BLKIO_POLICY_PROP, BLKIO_PROP_idle_time), .read_map = blkiocg_file_read_map, |
812df48d1
|
1487 1488 1489 |
}, { .name = "empty_time", |
062a644d6
|
1490 1491 1492 |
.private = BLKIOFILE_PRIVATE(BLKIO_POLICY_PROP, BLKIO_PROP_empty_time), .read_map = blkiocg_file_read_map, |
812df48d1
|
1493 1494 |
}, { |
220841906
|
1495 |
.name = "dequeue", |
062a644d6
|
1496 1497 1498 |
.private = BLKIOFILE_PRIVATE(BLKIO_POLICY_PROP, BLKIO_PROP_dequeue), .read_map = blkiocg_file_read_map, |
cdc1184cf
|
1499 |
}, |
9026e521c
|
1500 1501 1502 1503 1504 1505 |
{ .name = "unaccounted_time", .private = BLKIOFILE_PRIVATE(BLKIO_POLICY_PROP, BLKIO_PROP_unaccounted_time), .read_map = blkiocg_file_read_map, }, |
220841906
|
1506 |
#endif |
31e4c28d9
|
1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 |
}; 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); |
b1c357696
|
1518 1519 1520 |
unsigned long flags; struct blkio_group *blkg; void *key; |
3e2520668
|
1521 |
struct blkio_policy_type *blkiop; |
34d0f179d
|
1522 |
struct blkio_policy_node *pn, *pntmp; |
b1c357696
|
1523 1524 |
rcu_read_lock(); |
0f3942a39
|
1525 1526 |
do { spin_lock_irqsave(&blkcg->lock, flags); |
b1c357696
|
1527 |
|
0f3942a39
|
1528 1529 1530 1531 |
if (hlist_empty(&blkcg->blkg_list)) { spin_unlock_irqrestore(&blkcg->lock, flags); break; } |
b1c357696
|
1532 |
|
0f3942a39
|
1533 1534 1535 1536 |
blkg = hlist_entry(blkcg->blkg_list.first, struct blkio_group, blkcg_node); key = rcu_dereference(blkg->key); __blkiocg_del_blkio_group(blkg); |
31e4c28d9
|
1537 |
|
0f3942a39
|
1538 |
spin_unlock_irqrestore(&blkcg->lock, flags); |
b1c357696
|
1539 |
|
0f3942a39
|
1540 1541 1542 |
/* * This blkio_group is being unlinked as associated cgroup is * going away. Let all the IO controlling policies know about |
61014e96e
|
1543 |
* this event. |
0f3942a39
|
1544 1545 |
*/ spin_lock(&blkio_list_lock); |
61014e96e
|
1546 1547 1548 |
list_for_each_entry(blkiop, &blkio_list, list) { if (blkiop->plid != blkg->plid) continue; |
0f3942a39
|
1549 |
blkiop->ops.blkio_unlink_group_fn(key, blkg); |
61014e96e
|
1550 |
} |
0f3942a39
|
1551 1552 |
spin_unlock(&blkio_list_lock); } while (1); |
34d0f179d
|
1553 |
|
34d0f179d
|
1554 1555 1556 1557 |
list_for_each_entry_safe(pn, pntmp, &blkcg->policy_list, node) { blkio_policy_delete_node(pn); kfree(pn); } |
0f3942a39
|
1558 |
|
31e4c28d9
|
1559 |
free_css_id(&blkio_subsys, &blkcg->css); |
b1c357696
|
1560 |
rcu_read_unlock(); |
67523c48a
|
1561 1562 |
if (blkcg != &blkio_root_cgroup) kfree(blkcg); |
31e4c28d9
|
1563 1564 1565 1566 1567 |
} static struct cgroup_subsys_state * blkiocg_create(struct cgroup_subsys *subsys, struct cgroup *cgroup) { |
0341509fd
|
1568 1569 |
struct blkio_cgroup *blkcg; struct cgroup *parent = cgroup->parent; |
31e4c28d9
|
1570 |
|
0341509fd
|
1571 |
if (!parent) { |
31e4c28d9
|
1572 1573 1574 |
blkcg = &blkio_root_cgroup; goto done; } |
31e4c28d9
|
1575 1576 1577 1578 1579 1580 1581 1582 |
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); |
34d0f179d
|
1583 |
INIT_LIST_HEAD(&blkcg->policy_list); |
31e4c28d9
|
1584 1585 1586 1587 1588 1589 1590 1591 1592 |
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. */ |
bb9d97b6d
|
1593 1594 |
static int blkiocg_can_attach(struct cgroup_subsys *ss, struct cgroup *cgrp, struct cgroup_taskset *tset) |
31e4c28d9
|
1595 |
{ |
bb9d97b6d
|
1596 |
struct task_struct *task; |
31e4c28d9
|
1597 1598 1599 1600 |
struct io_context *ioc; int ret = 0; /* task_lock() is needed to avoid races with exit_io_context() */ |
bb9d97b6d
|
1601 1602 1603 1604 1605 1606 1607 1608 1609 |
cgroup_taskset_for_each(task, cgrp, tset) { task_lock(task); ioc = task->io_context; if (ioc && atomic_read(&ioc->nr_tasks) > 1) ret = -EINVAL; task_unlock(task); if (ret) break; } |
31e4c28d9
|
1610 1611 |
return ret; } |
bb9d97b6d
|
1612 1613 |
static void blkiocg_attach(struct cgroup_subsys *ss, struct cgroup *cgrp, struct cgroup_taskset *tset) |
31e4c28d9
|
1614 |
{ |
bb9d97b6d
|
1615 |
struct task_struct *task; |
31e4c28d9
|
1616 |
struct io_context *ioc; |
bb9d97b6d
|
1617 |
cgroup_taskset_for_each(task, cgrp, tset) { |
b3c9dd182
|
1618 1619 1620 1621 1622 1623 |
/* we don't lose anything even if ioc allocation fails */ ioc = get_task_io_context(task, GFP_ATOMIC, NUMA_NO_NODE); if (ioc) { ioc_cgroup_changed(ioc); put_io_context(ioc, NULL); } |
bb9d97b6d
|
1624 |
} |
31e4c28d9
|
1625 |
} |
3e2520668
|
1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 |
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); |
67523c48a
|
1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 |
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"); |