Blame view
block/blk-throttle.c
32.5 KB
e43473b7f blkio: Core imple... |
1 2 3 4 5 6 7 8 9 10 11 12 |
/* * Interface for controlling IO bandwidth on a request queue * * Copyright (C) 2010 Vivek Goyal <vgoyal@redhat.com> */ #include <linux/module.h> #include <linux/slab.h> #include <linux/blkdev.h> #include <linux/bio.h> #include <linux/blktrace_api.h> #include "blk-cgroup.h" |
bc9fcbf9c block: move blk_t... |
13 |
#include "blk.h" |
e43473b7f blkio: Core imple... |
14 15 16 17 18 19 20 21 22 |
/* Max dispatch from a group in 1 round */ static int throtl_grp_quantum = 8; /* Total max dispatch from all groups in one round */ static int throtl_quantum = 32; /* Throttling is performed over 100ms slice and after that slice is renewed */ static unsigned long throtl_slice = HZ/10; /* 100 ms */ |
450adcbe5 blk-throttle: Do ... |
23 24 25 26 |
/* A workqueue to queue throttle related work */ static struct workqueue_struct *kthrotld_workqueue; static void throtl_schedule_delayed_work(struct throtl_data *td, unsigned long delay); |
e43473b7f blkio: Core imple... |
27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 |
struct throtl_rb_root { struct rb_root rb; struct rb_node *left; unsigned int count; unsigned long min_disptime; }; #define THROTL_RB_ROOT (struct throtl_rb_root) { .rb = RB_ROOT, .left = NULL, \ .count = 0, .min_disptime = 0} #define rb_entry_tg(node) rb_entry((node), struct throtl_grp, rb_node) struct throtl_grp { /* List of throtl groups on the request queue*/ struct hlist_node tg_node; /* active throtl group service_tree member */ struct rb_node rb_node; /* * Dispatch time in jiffies. This is the estimated time when group * will unthrottle and is ready to dispatch more bio. It is used as * key to sort active groups in service tree. */ unsigned long disptime; struct blkio_group blkg; atomic_t ref; unsigned int flags; /* Two lists for READ and WRITE */ struct bio_list bio_lists[2]; /* Number of queued bios on READ and WRITE lists */ unsigned int nr_queued[2]; /* bytes per second rate limits */ uint64_t bps[2]; |
8e89d13f4 blkio: Implementa... |
65 66 |
/* IOPS limits */ unsigned int iops[2]; |
e43473b7f blkio: Core imple... |
67 68 |
/* Number of bytes disptached in current slice */ uint64_t bytes_disp[2]; |
8e89d13f4 blkio: Implementa... |
69 70 |
/* Number of bio's dispatched in current slice */ unsigned int io_disp[2]; |
e43473b7f blkio: Core imple... |
71 72 73 74 |
/* When did we start a new slice */ unsigned long slice_start[2]; unsigned long slice_end[2]; |
fe0714377 blkio: Recalculat... |
75 76 |
/* Some throttle limits got updated for the group */ |
6f0379377 blk-throttle: don... |
77 |
int limits_changed; |
4843c69d4 blk-throttle: Fre... |
78 79 |
struct rcu_head rcu_head; |
e43473b7f blkio: Core imple... |
80 81 82 83 84 85 86 87 88 |
}; struct throtl_data { /* List of throtl groups */ struct hlist_head tg_list; /* service tree for active throtl groups */ struct throtl_rb_root tg_service_tree; |
29b125892 blk-throttle: Dyn... |
89 |
struct throtl_grp *root_tg; |
e43473b7f blkio: Core imple... |
90 91 92 93 94 95 |
struct request_queue *queue; /* Total Number of queued bios on READ and WRITE lists */ unsigned int nr_queued[2]; /* |
02977e4af blkio: Add root g... |
96 |
* number of total undestroyed groups |
e43473b7f blkio: Core imple... |
97 98 99 100 101 |
*/ unsigned int nr_undestroyed_grps; /* Work for dispatching throttled bios */ struct delayed_work throtl_work; |
fe0714377 blkio: Recalculat... |
102 |
|
6f0379377 blk-throttle: don... |
103 |
int limits_changed; |
e43473b7f blkio: Core imple... |
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 |
}; enum tg_state_flags { THROTL_TG_FLAG_on_rr = 0, /* on round-robin busy list */ }; #define THROTL_TG_FNS(name) \ static inline void throtl_mark_tg_##name(struct throtl_grp *tg) \ { \ (tg)->flags |= (1 << THROTL_TG_FLAG_##name); \ } \ static inline void throtl_clear_tg_##name(struct throtl_grp *tg) \ { \ (tg)->flags &= ~(1 << THROTL_TG_FLAG_##name); \ } \ static inline int throtl_tg_##name(const struct throtl_grp *tg) \ { \ return ((tg)->flags & (1 << THROTL_TG_FLAG_##name)) != 0; \ } THROTL_TG_FNS(on_rr); #define throtl_log_tg(td, tg, fmt, args...) \ blk_add_trace_msg((td)->queue, "throtl %s " fmt, \ blkg_path(&(tg)->blkg), ##args); \ #define throtl_log(td, fmt, args...) \ blk_add_trace_msg((td)->queue, "throtl " fmt, ##args) static inline struct throtl_grp *tg_of_blkg(struct blkio_group *blkg) { if (blkg) return container_of(blkg, struct throtl_grp, blkg); return NULL; } |
d2f31a5fd blk-throttle: Mak... |
140 |
static inline unsigned int total_nr_queued(struct throtl_data *td) |
e43473b7f blkio: Core imple... |
141 |
{ |
d2f31a5fd blk-throttle: Mak... |
142 |
return td->nr_queued[0] + td->nr_queued[1]; |
e43473b7f blkio: Core imple... |
143 144 145 146 147 148 149 |
} static inline struct throtl_grp *throtl_ref_get_tg(struct throtl_grp *tg) { atomic_inc(&tg->ref); return tg; } |
4843c69d4 blk-throttle: Fre... |
150 151 152 153 154 |
static void throtl_free_tg(struct rcu_head *head) { struct throtl_grp *tg; tg = container_of(head, struct throtl_grp, rcu_head); |
5624a4e44 blk-throttle: Mak... |
155 |
free_percpu(tg->blkg.stats_cpu); |
4843c69d4 blk-throttle: Fre... |
156 157 |
kfree(tg); } |
e43473b7f blkio: Core imple... |
158 159 160 161 162 |
static void throtl_put_tg(struct throtl_grp *tg) { BUG_ON(atomic_read(&tg->ref) <= 0); if (!atomic_dec_and_test(&tg->ref)) return; |
4843c69d4 blk-throttle: Fre... |
163 164 165 166 167 168 169 170 171 172 173 |
/* * A group is freed in rcu manner. But having an rcu lock does not * mean that one can access all the fields of blkg and assume these * are valid. For example, don't try to follow throtl_data and * request queue links. * * Having a reference to blkg under an rcu allows acess to only * values local to groups like group stats and group rate limits */ call_rcu(&tg->rcu_head, throtl_free_tg); |
e43473b7f blkio: Core imple... |
174 |
} |
a29a171e7 blk-throttle: Do ... |
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 |
static void throtl_init_group(struct throtl_grp *tg) { INIT_HLIST_NODE(&tg->tg_node); RB_CLEAR_NODE(&tg->rb_node); bio_list_init(&tg->bio_lists[0]); bio_list_init(&tg->bio_lists[1]); tg->limits_changed = false; /* Practically unlimited BW */ tg->bps[0] = tg->bps[1] = -1; tg->iops[0] = tg->iops[1] = -1; /* * Take the initial reference that will be released on destroy * This can be thought of a joint reference by cgroup and * request queue which will be dropped by either request queue * exit or cgroup deletion path depending on who is exiting first. */ atomic_set(&tg->ref, 1); } /* Should be called with rcu read lock held (needed for blkcg) */ static void throtl_add_group_to_td_list(struct throtl_data *td, struct throtl_grp *tg) { hlist_add_head(&tg->tg_node, &td->tg_list); td->nr_undestroyed_grps++; } |
269f54155 blk-throttle: Int... |
203 204 |
static void __throtl_tg_fill_dev_details(struct throtl_data *td, struct throtl_grp *tg) |
f469a7b4d blk-cgroup: Allow... |
205 206 207 |
{ struct backing_dev_info *bdi = &td->queue->backing_dev_info; unsigned int major, minor; |
269f54155 blk-throttle: Int... |
208 209 210 211 212 213 214 215 216 217 218 219 220 |
if (!tg || tg->blkg.dev) return; /* * Fill in device details for a group which might not have been * filled at group creation time as queue was being instantiated * and driver had not attached a device yet */ if (bdi->dev && dev_name(bdi->dev)) { sscanf(dev_name(bdi->dev), "%u:%u", &major, &minor); tg->blkg.dev = MKDEV(major, minor); } } |
af75cd3c6 blk-throttle: Mak... |
221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 |
/* * Should be called with without queue lock held. Here queue lock will be * taken rarely. It will be taken only once during life time of a group * if need be */ static void throtl_tg_fill_dev_details(struct throtl_data *td, struct throtl_grp *tg) { if (!tg || tg->blkg.dev) return; spin_lock_irq(td->queue->queue_lock); __throtl_tg_fill_dev_details(td, tg); spin_unlock_irq(td->queue->queue_lock); } |
269f54155 blk-throttle: Int... |
236 237 238 239 |
static void throtl_init_add_tg_lists(struct throtl_data *td, struct throtl_grp *tg, struct blkio_cgroup *blkcg) { __throtl_tg_fill_dev_details(td, tg); |
f469a7b4d blk-cgroup: Allow... |
240 |
/* Add group onto cgroup list */ |
f469a7b4d blk-cgroup: Allow... |
241 |
blkiocg_add_blkio_group(blkcg, &tg->blkg, (void *)td, |
269f54155 blk-throttle: Int... |
242 |
tg->blkg.dev, BLKIO_POLICY_THROTL); |
f469a7b4d blk-cgroup: Allow... |
243 244 245 246 247 248 249 250 251 252 253 254 255 |
tg->bps[READ] = blkcg_get_read_bps(blkcg, tg->blkg.dev); tg->bps[WRITE] = blkcg_get_write_bps(blkcg, tg->blkg.dev); tg->iops[READ] = blkcg_get_read_iops(blkcg, tg->blkg.dev); tg->iops[WRITE] = blkcg_get_write_iops(blkcg, tg->blkg.dev); throtl_add_group_to_td_list(td, tg); } /* Should be called without queue lock and outside of rcu period */ static struct throtl_grp *throtl_alloc_tg(struct throtl_data *td) { struct throtl_grp *tg = NULL; |
5624a4e44 blk-throttle: Mak... |
256 |
int ret; |
f469a7b4d blk-cgroup: Allow... |
257 258 259 260 |
tg = kzalloc_node(sizeof(*tg), GFP_ATOMIC, td->queue->node); if (!tg) return NULL; |
5624a4e44 blk-throttle: Mak... |
261 262 263 264 265 266 |
ret = blkio_alloc_blkg_stats(&tg->blkg); if (ret) { kfree(tg); return NULL; } |
f469a7b4d blk-cgroup: Allow... |
267 268 269 270 271 272 |
throtl_init_group(tg); return tg; } static struct throtl_grp *throtl_find_tg(struct throtl_data *td, struct blkio_cgroup *blkcg) |
e43473b7f blkio: Core imple... |
273 |
{ |
e43473b7f blkio: Core imple... |
274 275 |
struct throtl_grp *tg = NULL; void *key = td; |
e43473b7f blkio: Core imple... |
276 277 |
/* |
be2c6b199 blkio-throttle: A... |
278 279 280 281 |
* This is the common case when there are no blkio cgroups. * Avoid lookup in this case */ if (blkcg == &blkio_root_cgroup) |
29b125892 blk-throttle: Dyn... |
282 |
tg = td->root_tg; |
be2c6b199 blkio-throttle: A... |
283 284 |
else tg = tg_of_blkg(blkiocg_lookup_group(blkcg, key)); |
e43473b7f blkio: Core imple... |
285 |
|
269f54155 blk-throttle: Int... |
286 |
__throtl_tg_fill_dev_details(td, tg); |
e43473b7f blkio: Core imple... |
287 288 289 290 291 |
return tg; } static struct throtl_grp * throtl_get_tg(struct throtl_data *td) { |
f469a7b4d blk-cgroup: Allow... |
292 |
struct throtl_grp *tg = NULL, *__tg = NULL; |
70087dc38 blk-throttle: Use... |
293 |
struct blkio_cgroup *blkcg; |
f469a7b4d blk-cgroup: Allow... |
294 |
struct request_queue *q = td->queue; |
e43473b7f blkio: Core imple... |
295 |
|
c9a929dde block: fix reques... |
296 |
/* no throttling for dead queue */ |
34f6055c8 block: add blk_qu... |
297 |
if (unlikely(blk_queue_dead(q))) |
c9a929dde block: fix reques... |
298 |
return NULL; |
e43473b7f blkio: Core imple... |
299 |
rcu_read_lock(); |
70087dc38 blk-throttle: Use... |
300 |
blkcg = task_blkio_cgroup(current); |
f469a7b4d blk-cgroup: Allow... |
301 302 303 304 305 306 307 308 309 |
tg = throtl_find_tg(td, blkcg); if (tg) { rcu_read_unlock(); return tg; } /* * Need to allocate a group. Allocation of group also needs allocation * of per cpu stats which in-turn takes a mutex() and can block. Hence |
315fceee8 block: drop unnec... |
310 |
* we need to drop rcu lock and queue_lock before we call alloc. |
f469a7b4d blk-cgroup: Allow... |
311 |
*/ |
f469a7b4d blk-cgroup: Allow... |
312 313 314 315 |
rcu_read_unlock(); spin_unlock_irq(q->queue_lock); tg = throtl_alloc_tg(td); |
f469a7b4d blk-cgroup: Allow... |
316 317 318 |
/* Group allocated and queue is still alive. take the lock */ spin_lock_irq(q->queue_lock); |
bc16a4f93 block: reorganize... |
319 |
/* Make sure @q is still alive */ |
34f6055c8 block: add blk_qu... |
320 |
if (unlikely(blk_queue_dead(q))) { |
bc16a4f93 block: reorganize... |
321 322 323 |
kfree(tg); return NULL; } |
f469a7b4d blk-cgroup: Allow... |
324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 |
/* * Initialize the new group. After sleeping, read the blkcg again. */ rcu_read_lock(); blkcg = task_blkio_cgroup(current); /* * If some other thread already allocated the group while we were * not holding queue lock, free up the group */ __tg = throtl_find_tg(td, blkcg); if (__tg) { kfree(tg); rcu_read_unlock(); return __tg; } /* Group allocation failed. Account the IO to root group */ if (!tg) { |
29b125892 blk-throttle: Dyn... |
344 |
tg = td->root_tg; |
f469a7b4d blk-cgroup: Allow... |
345 346 347 348 |
return tg; } throtl_init_add_tg_lists(td, tg, blkcg); |
e43473b7f blkio: Core imple... |
349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 |
rcu_read_unlock(); return tg; } static struct throtl_grp *throtl_rb_first(struct throtl_rb_root *root) { /* Service tree is empty */ if (!root->count) return NULL; if (!root->left) root->left = rb_first(&root->rb); if (root->left) return rb_entry_tg(root->left); return NULL; } static void rb_erase_init(struct rb_node *n, struct rb_root *root) { rb_erase(n, root); RB_CLEAR_NODE(n); } static void throtl_rb_erase(struct rb_node *n, struct throtl_rb_root *root) { if (root->left == n) root->left = NULL; rb_erase_init(n, &root->rb); --root->count; } static void update_min_dispatch_time(struct throtl_rb_root *st) { struct throtl_grp *tg; tg = throtl_rb_first(st); if (!tg) return; st->min_disptime = tg->disptime; } static void tg_service_tree_add(struct throtl_rb_root *st, struct throtl_grp *tg) { struct rb_node **node = &st->rb.rb_node; struct rb_node *parent = NULL; struct throtl_grp *__tg; unsigned long key = tg->disptime; int left = 1; while (*node != NULL) { parent = *node; __tg = rb_entry_tg(parent); if (time_before(key, __tg->disptime)) node = &parent->rb_left; else { node = &parent->rb_right; left = 0; } } if (left) st->left = &tg->rb_node; rb_link_node(&tg->rb_node, parent, node); rb_insert_color(&tg->rb_node, &st->rb); } static void __throtl_enqueue_tg(struct throtl_data *td, struct throtl_grp *tg) { struct throtl_rb_root *st = &td->tg_service_tree; tg_service_tree_add(st, tg); throtl_mark_tg_on_rr(tg); st->count++; } static void throtl_enqueue_tg(struct throtl_data *td, struct throtl_grp *tg) { if (!throtl_tg_on_rr(tg)) __throtl_enqueue_tg(td, tg); } static void __throtl_dequeue_tg(struct throtl_data *td, struct throtl_grp *tg) { throtl_rb_erase(&tg->rb_node, &td->tg_service_tree); throtl_clear_tg_on_rr(tg); } static void throtl_dequeue_tg(struct throtl_data *td, struct throtl_grp *tg) { if (throtl_tg_on_rr(tg)) __throtl_dequeue_tg(td, tg); } static void throtl_schedule_next_dispatch(struct throtl_data *td) { struct throtl_rb_root *st = &td->tg_service_tree; /* * If there are more bios pending, schedule more work. */ if (!total_nr_queued(td)) return; BUG_ON(!st->count); update_min_dispatch_time(st); if (time_before_eq(st->min_disptime, jiffies)) |
450adcbe5 blk-throttle: Do ... |
463 |
throtl_schedule_delayed_work(td, 0); |
e43473b7f blkio: Core imple... |
464 |
else |
450adcbe5 blk-throttle: Do ... |
465 |
throtl_schedule_delayed_work(td, (st->min_disptime - jiffies)); |
e43473b7f blkio: Core imple... |
466 467 468 469 470 471 |
} static inline void throtl_start_new_slice(struct throtl_data *td, struct throtl_grp *tg, bool rw) { tg->bytes_disp[rw] = 0; |
8e89d13f4 blkio: Implementa... |
472 |
tg->io_disp[rw] = 0; |
e43473b7f blkio: Core imple... |
473 474 475 476 477 478 |
tg->slice_start[rw] = jiffies; tg->slice_end[rw] = jiffies + throtl_slice; throtl_log_tg(td, tg, "[%c] new slice start=%lu end=%lu jiffies=%lu", rw == READ ? 'R' : 'W', tg->slice_start[rw], tg->slice_end[rw], jiffies); } |
d1ae8ffdf blk-throttle: Tri... |
479 480 481 482 483 |
static inline void throtl_set_slice_end(struct throtl_data *td, struct throtl_grp *tg, bool rw, unsigned long jiffy_end) { tg->slice_end[rw] = roundup(jiffy_end, throtl_slice); } |
e43473b7f blkio: Core imple... |
484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 |
static inline void throtl_extend_slice(struct throtl_data *td, struct throtl_grp *tg, bool rw, unsigned long jiffy_end) { tg->slice_end[rw] = roundup(jiffy_end, throtl_slice); throtl_log_tg(td, tg, "[%c] extend slice start=%lu end=%lu jiffies=%lu", rw == READ ? 'R' : 'W', tg->slice_start[rw], tg->slice_end[rw], jiffies); } /* Determine if previously allocated or extended slice is complete or not */ static bool throtl_slice_used(struct throtl_data *td, struct throtl_grp *tg, bool rw) { if (time_in_range(jiffies, tg->slice_start[rw], tg->slice_end[rw])) return 0; return 1; } /* Trim the used slices and adjust slice start accordingly */ static inline void throtl_trim_slice(struct throtl_data *td, struct throtl_grp *tg, bool rw) { |
3aad5d3ee blkio-throttle: F... |
507 508 |
unsigned long nr_slices, time_elapsed, io_trim; u64 bytes_trim, tmp; |
e43473b7f blkio: Core imple... |
509 510 511 512 513 514 515 516 517 518 |
BUG_ON(time_before(tg->slice_end[rw], tg->slice_start[rw])); /* * If bps are unlimited (-1), then time slice don't get * renewed. Don't try to trim the slice if slice is used. A new * slice will start when appropriate. */ if (throtl_slice_used(td, tg, rw)) return; |
d1ae8ffdf blk-throttle: Tri... |
519 520 521 522 523 524 525 526 527 |
/* * A bio has been dispatched. Also adjust slice_end. It might happen * that initially cgroup limit was very low resulting in high * slice_end, but later limit was bumped up and bio was dispached * sooner, then we need to reduce slice_end. A high bogus slice_end * is bad because it does not allow new slice to start. */ throtl_set_slice_end(td, tg, rw, jiffies + throtl_slice); |
e43473b7f blkio: Core imple... |
528 529 530 531 532 533 |
time_elapsed = jiffies - tg->slice_start[rw]; nr_slices = time_elapsed / throtl_slice; if (!nr_slices) return; |
3aad5d3ee blkio-throttle: F... |
534 535 536 |
tmp = tg->bps[rw] * throtl_slice * nr_slices; do_div(tmp, HZ); bytes_trim = tmp; |
e43473b7f blkio: Core imple... |
537 |
|
8e89d13f4 blkio: Implementa... |
538 |
io_trim = (tg->iops[rw] * throtl_slice * nr_slices)/HZ; |
e43473b7f blkio: Core imple... |
539 |
|
8e89d13f4 blkio: Implementa... |
540 |
if (!bytes_trim && !io_trim) |
e43473b7f blkio: Core imple... |
541 542 543 544 545 546 |
return; if (tg->bytes_disp[rw] >= bytes_trim) tg->bytes_disp[rw] -= bytes_trim; else tg->bytes_disp[rw] = 0; |
8e89d13f4 blkio: Implementa... |
547 548 549 550 |
if (tg->io_disp[rw] >= io_trim) tg->io_disp[rw] -= io_trim; else tg->io_disp[rw] = 0; |
e43473b7f blkio: Core imple... |
551 |
tg->slice_start[rw] += nr_slices * throtl_slice; |
3aad5d3ee blkio-throttle: F... |
552 |
throtl_log_tg(td, tg, "[%c] trim slice nr=%lu bytes=%llu io=%lu" |
e43473b7f blkio: Core imple... |
553 |
" start=%lu end=%lu jiffies=%lu", |
8e89d13f4 blkio: Implementa... |
554 |
rw == READ ? 'R' : 'W', nr_slices, bytes_trim, io_trim, |
e43473b7f blkio: Core imple... |
555 556 |
tg->slice_start[rw], tg->slice_end[rw], jiffies); } |
8e89d13f4 blkio: Implementa... |
557 558 |
static bool tg_with_in_iops_limit(struct throtl_data *td, struct throtl_grp *tg, struct bio *bio, unsigned long *wait) |
e43473b7f blkio: Core imple... |
559 560 |
{ bool rw = bio_data_dir(bio); |
8e89d13f4 blkio: Implementa... |
561 |
unsigned int io_allowed; |
e43473b7f blkio: Core imple... |
562 |
unsigned long jiffy_elapsed, jiffy_wait, jiffy_elapsed_rnd; |
c49c06e49 blkio-throttle: F... |
563 |
u64 tmp; |
e43473b7f blkio: Core imple... |
564 |
|
8e89d13f4 blkio: Implementa... |
565 |
jiffy_elapsed = jiffy_elapsed_rnd = jiffies - tg->slice_start[rw]; |
e43473b7f blkio: Core imple... |
566 |
|
8e89d13f4 blkio: Implementa... |
567 568 569 570 571 |
/* Slice has just started. Consider one slice interval */ if (!jiffy_elapsed) jiffy_elapsed_rnd = throtl_slice; jiffy_elapsed_rnd = roundup(jiffy_elapsed_rnd, throtl_slice); |
c49c06e49 blkio-throttle: F... |
572 573 574 575 576 577 578 579 580 581 582 583 584 585 |
/* * jiffy_elapsed_rnd should not be a big value as minimum iops can be * 1 then at max jiffy elapsed should be equivalent of 1 second as we * will allow dispatch after 1 second and after that slice should * have been trimmed. */ tmp = (u64)tg->iops[rw] * jiffy_elapsed_rnd; do_div(tmp, HZ); if (tmp > UINT_MAX) io_allowed = UINT_MAX; else io_allowed = tmp; |
8e89d13f4 blkio: Implementa... |
586 587 |
if (tg->io_disp[rw] + 1 <= io_allowed) { |
e43473b7f blkio: Core imple... |
588 589 590 591 |
if (wait) *wait = 0; return 1; } |
8e89d13f4 blkio: Implementa... |
592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 |
/* Calc approx time to dispatch */ jiffy_wait = ((tg->io_disp[rw] + 1) * HZ)/tg->iops[rw] + 1; if (jiffy_wait > jiffy_elapsed) jiffy_wait = jiffy_wait - jiffy_elapsed; else jiffy_wait = 1; if (wait) *wait = jiffy_wait; return 0; } static bool tg_with_in_bps_limit(struct throtl_data *td, struct throtl_grp *tg, struct bio *bio, unsigned long *wait) { bool rw = bio_data_dir(bio); |
3aad5d3ee blkio-throttle: F... |
609 |
u64 bytes_allowed, extra_bytes, tmp; |
8e89d13f4 blkio: Implementa... |
610 |
unsigned long jiffy_elapsed, jiffy_wait, jiffy_elapsed_rnd; |
e43473b7f blkio: Core imple... |
611 612 613 614 615 616 617 618 |
jiffy_elapsed = jiffy_elapsed_rnd = jiffies - tg->slice_start[rw]; /* Slice has just started. Consider one slice interval */ if (!jiffy_elapsed) jiffy_elapsed_rnd = throtl_slice; jiffy_elapsed_rnd = roundup(jiffy_elapsed_rnd, throtl_slice); |
5e901a2b9 blkio-throttle: T... |
619 620 |
tmp = tg->bps[rw] * jiffy_elapsed_rnd; do_div(tmp, HZ); |
3aad5d3ee blkio-throttle: F... |
621 |
bytes_allowed = tmp; |
e43473b7f blkio: Core imple... |
622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 |
if (tg->bytes_disp[rw] + bio->bi_size <= bytes_allowed) { if (wait) *wait = 0; return 1; } /* Calc approx time to dispatch */ extra_bytes = tg->bytes_disp[rw] + bio->bi_size - bytes_allowed; jiffy_wait = div64_u64(extra_bytes * HZ, tg->bps[rw]); if (!jiffy_wait) jiffy_wait = 1; /* * This wait time is without taking into consideration the rounding * up we did. Add that time also. */ jiffy_wait = jiffy_wait + (jiffy_elapsed_rnd - jiffy_elapsed); |
e43473b7f blkio: Core imple... |
641 642 |
if (wait) *wait = jiffy_wait; |
8e89d13f4 blkio: Implementa... |
643 644 |
return 0; } |
af75cd3c6 blk-throttle: Mak... |
645 646 647 648 649 |
static bool tg_no_rule_group(struct throtl_grp *tg, bool rw) { if (tg->bps[rw] == -1 && tg->iops[rw] == -1) return 1; return 0; } |
8e89d13f4 blkio: Implementa... |
650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 |
/* * Returns whether one can dispatch a bio or not. Also returns approx number * of jiffies to wait before this bio is with-in IO rate and can be dispatched */ static bool tg_may_dispatch(struct throtl_data *td, struct throtl_grp *tg, struct bio *bio, unsigned long *wait) { bool rw = bio_data_dir(bio); unsigned long bps_wait = 0, iops_wait = 0, max_wait = 0; /* * Currently whole state machine of group depends on first bio * queued in the group bio list. So one should not be calling * this function with a different bio if there are other bios * queued. */ BUG_ON(tg->nr_queued[rw] && bio != bio_list_peek(&tg->bio_lists[rw])); |
e43473b7f blkio: Core imple... |
667 |
|
8e89d13f4 blkio: Implementa... |
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 |
/* If tg->bps = -1, then BW is unlimited */ if (tg->bps[rw] == -1 && tg->iops[rw] == -1) { if (wait) *wait = 0; return 1; } /* * If previous slice expired, start a new one otherwise renew/extend * existing slice to make sure it is at least throtl_slice interval * long since now. */ if (throtl_slice_used(td, tg, rw)) throtl_start_new_slice(td, tg, rw); else { if (time_before(tg->slice_end[rw], jiffies + throtl_slice)) throtl_extend_slice(td, tg, rw, jiffies + throtl_slice); } if (tg_with_in_bps_limit(td, tg, bio, &bps_wait) && tg_with_in_iops_limit(td, tg, bio, &iops_wait)) { if (wait) *wait = 0; return 1; } max_wait = max(bps_wait, iops_wait); if (wait) *wait = max_wait; if (time_before(tg->slice_end[rw], jiffies + max_wait)) throtl_extend_slice(td, tg, rw, jiffies + max_wait); |
e43473b7f blkio: Core imple... |
701 702 703 704 705 706 707 |
return 0; } static void throtl_charge_bio(struct throtl_grp *tg, struct bio *bio) { bool rw = bio_data_dir(bio); |
e5a94f568 blk-throttle: cor... |
708 |
bool sync = rw_is_sync(bio->bi_rw); |
e43473b7f blkio: Core imple... |
709 710 711 |
/* Charge the bio to the group */ tg->bytes_disp[rw] += bio->bi_size; |
8e89d13f4 blkio: Implementa... |
712 |
tg->io_disp[rw]++; |
e43473b7f blkio: Core imple... |
713 |
|
e43473b7f blkio: Core imple... |
714 |
blkiocg_update_dispatch_stats(&tg->blkg, bio->bi_size, rw, sync); |
e43473b7f blkio: Core imple... |
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 |
} static void throtl_add_bio_tg(struct throtl_data *td, struct throtl_grp *tg, struct bio *bio) { bool rw = bio_data_dir(bio); bio_list_add(&tg->bio_lists[rw], bio); /* Take a bio reference on tg */ throtl_ref_get_tg(tg); tg->nr_queued[rw]++; td->nr_queued[rw]++; throtl_enqueue_tg(td, tg); } static void tg_update_disptime(struct throtl_data *td, struct throtl_grp *tg) { unsigned long read_wait = -1, write_wait = -1, min_wait = -1, disptime; struct bio *bio; if ((bio = bio_list_peek(&tg->bio_lists[READ]))) tg_may_dispatch(td, tg, bio, &read_wait); if ((bio = bio_list_peek(&tg->bio_lists[WRITE]))) tg_may_dispatch(td, tg, bio, &write_wait); min_wait = min(read_wait, write_wait); disptime = jiffies + min_wait; |
e43473b7f blkio: Core imple... |
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 |
/* Update dispatch time */ throtl_dequeue_tg(td, tg); tg->disptime = disptime; throtl_enqueue_tg(td, tg); } static void tg_dispatch_one_bio(struct throtl_data *td, struct throtl_grp *tg, bool rw, struct bio_list *bl) { struct bio *bio; bio = bio_list_pop(&tg->bio_lists[rw]); tg->nr_queued[rw]--; /* Drop bio reference on tg */ throtl_put_tg(tg); BUG_ON(td->nr_queued[rw] <= 0); td->nr_queued[rw]--; throtl_charge_bio(tg, bio); bio_list_add(bl, bio); bio->bi_rw |= REQ_THROTTLED; throtl_trim_slice(td, tg, rw); } static int throtl_dispatch_tg(struct throtl_data *td, struct throtl_grp *tg, struct bio_list *bl) { unsigned int nr_reads = 0, nr_writes = 0; unsigned int max_nr_reads = throtl_grp_quantum*3/4; |
c2f6805d4 blk-throttle: Fix... |
774 |
unsigned int max_nr_writes = throtl_grp_quantum - max_nr_reads; |
e43473b7f blkio: Core imple... |
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 |
struct bio *bio; /* Try to dispatch 75% READS and 25% WRITES */ while ((bio = bio_list_peek(&tg->bio_lists[READ])) && tg_may_dispatch(td, tg, bio, NULL)) { tg_dispatch_one_bio(td, tg, bio_data_dir(bio), bl); nr_reads++; if (nr_reads >= max_nr_reads) break; } while ((bio = bio_list_peek(&tg->bio_lists[WRITE])) && tg_may_dispatch(td, tg, bio, NULL)) { tg_dispatch_one_bio(td, tg, bio_data_dir(bio), bl); nr_writes++; if (nr_writes >= max_nr_writes) break; } return nr_reads + nr_writes; } static int throtl_select_dispatch(struct throtl_data *td, struct bio_list *bl) { unsigned int nr_disp = 0; struct throtl_grp *tg; struct throtl_rb_root *st = &td->tg_service_tree; while (1) { tg = throtl_rb_first(st); if (!tg) break; if (time_before(jiffies, tg->disptime)) break; throtl_dequeue_tg(td, tg); nr_disp += throtl_dispatch_tg(td, tg, bl); if (tg->nr_queued[0] || tg->nr_queued[1]) { tg_update_disptime(td, tg); throtl_enqueue_tg(td, tg); } if (nr_disp >= throtl_quantum) break; } return nr_disp; } |
fe0714377 blkio: Recalculat... |
832 833 834 835 |
static void throtl_process_limit_change(struct throtl_data *td) { struct throtl_grp *tg; struct hlist_node *pos, *n; |
de701c74a blk-throttle: Som... |
836 |
if (!td->limits_changed) |
fe0714377 blkio: Recalculat... |
837 |
return; |
de701c74a blk-throttle: Som... |
838 |
xchg(&td->limits_changed, false); |
fe0714377 blkio: Recalculat... |
839 |
|
de701c74a blk-throttle: Som... |
840 |
throtl_log(td, "limits changed"); |
fe0714377 blkio: Recalculat... |
841 |
|
04a6b516c blk-throttle: Cor... |
842 |
hlist_for_each_entry_safe(tg, pos, n, &td->tg_list, tg_node) { |
de701c74a blk-throttle: Som... |
843 844 845 846 847 848 849 850 851 |
if (!tg->limits_changed) continue; if (!xchg(&tg->limits_changed, false)) continue; throtl_log_tg(td, tg, "limit change rbps=%llu wbps=%llu" " riops=%u wiops=%u", tg->bps[READ], tg->bps[WRITE], tg->iops[READ], tg->iops[WRITE]); |
04521db04 blk-throttle: Res... |
852 853 854 855 856 857 858 859 |
/* * Restart the slices for both READ and WRITES. It * might happen that a group's limit are dropped * suddenly and we don't want to account recently * dispatched IO with new low rate */ throtl_start_new_slice(td, tg, 0); throtl_start_new_slice(td, tg, 1); |
de701c74a blk-throttle: Som... |
860 |
if (throtl_tg_on_rr(tg)) |
fe0714377 blkio: Recalculat... |
861 |
tg_update_disptime(td, tg); |
fe0714377 blkio: Recalculat... |
862 |
} |
fe0714377 blkio: Recalculat... |
863 |
} |
e43473b7f blkio: Core imple... |
864 865 866 867 868 869 870 |
/* Dispatch throttled bios. Should be called without queue lock held. */ static int throtl_dispatch(struct request_queue *q) { struct throtl_data *td = q->td; unsigned int nr_disp = 0; struct bio_list bio_list_on_stack; struct bio *bio; |
69d60eb96 blk-throttle: Use... |
871 |
struct blk_plug plug; |
e43473b7f blkio: Core imple... |
872 873 |
spin_lock_irq(q->queue_lock); |
fe0714377 blkio: Recalculat... |
874 |
throtl_process_limit_change(td); |
e43473b7f blkio: Core imple... |
875 876 877 878 |
if (!total_nr_queued(td)) goto out; bio_list_init(&bio_list_on_stack); |
d2f31a5fd blk-throttle: Mak... |
879 |
throtl_log(td, "dispatch nr_queued=%u read=%u write=%u", |
e43473b7f blkio: Core imple... |
880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 |
total_nr_queued(td), td->nr_queued[READ], td->nr_queued[WRITE]); nr_disp = throtl_select_dispatch(td, &bio_list_on_stack); if (nr_disp) throtl_log(td, "bios disp=%u", nr_disp); throtl_schedule_next_dispatch(td); out: spin_unlock_irq(q->queue_lock); /* * If we dispatched some requests, unplug the queue to make sure * immediate dispatch */ if (nr_disp) { |
69d60eb96 blk-throttle: Use... |
897 |
blk_start_plug(&plug); |
e43473b7f blkio: Core imple... |
898 899 |
while((bio = bio_list_pop(&bio_list_on_stack))) generic_make_request(bio); |
69d60eb96 blk-throttle: Use... |
900 |
blk_finish_plug(&plug); |
e43473b7f blkio: Core imple... |
901 902 903 904 905 906 907 908 909 910 911 912 913 914 |
} return nr_disp; } void blk_throtl_work(struct work_struct *work) { struct throtl_data *td = container_of(work, struct throtl_data, throtl_work.work); struct request_queue *q = td->queue; throtl_dispatch(q); } /* Call with queue lock held */ |
450adcbe5 blk-throttle: Do ... |
915 916 |
static void throtl_schedule_delayed_work(struct throtl_data *td, unsigned long delay) |
e43473b7f blkio: Core imple... |
917 |
{ |
e43473b7f blkio: Core imple... |
918 |
struct delayed_work *dwork = &td->throtl_work; |
04521db04 blk-throttle: Res... |
919 |
/* schedule work if limits changed even if no bio is queued */ |
d2f31a5fd blk-throttle: Mak... |
920 |
if (total_nr_queued(td) || td->limits_changed) { |
e43473b7f blkio: Core imple... |
921 922 923 924 925 |
/* * We might have a work scheduled to be executed in future. * Cancel that and schedule a new one. */ __cancel_delayed_work(dwork); |
450adcbe5 blk-throttle: Do ... |
926 |
queue_delayed_work(kthrotld_workqueue, dwork, delay); |
e43473b7f blkio: Core imple... |
927 928 929 930 |
throtl_log(td, "schedule work. delay=%lu jiffies=%lu", delay, jiffies); } } |
e43473b7f blkio: Core imple... |
931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 |
static void throtl_destroy_tg(struct throtl_data *td, struct throtl_grp *tg) { /* Something wrong if we are trying to remove same group twice */ BUG_ON(hlist_unhashed(&tg->tg_node)); hlist_del_init(&tg->tg_node); /* * Put the reference taken at the time of creation so that when all * queues are gone, group can be destroyed. */ throtl_put_tg(tg); td->nr_undestroyed_grps--; } static void throtl_release_tgs(struct throtl_data *td) { struct hlist_node *pos, *n; struct throtl_grp *tg; hlist_for_each_entry_safe(tg, pos, n, &td->tg_list, tg_node) { /* * If cgroup removal path got to blk_group first and removed * it from cgroup list, then it will take care of destroying * cfqg also. */ if (!blkiocg_del_blkio_group(&tg->blkg)) throtl_destroy_tg(td, tg); } } |
e43473b7f blkio: Core imple... |
963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 |
/* * Blk cgroup controller notification saying that blkio_group object is being * delinked as associated cgroup object is going away. That also means that * no new IO will come in this group. So get rid of this group as soon as * any pending IO in the group is finished. * * This function is called under rcu_read_lock(). key is the rcu protected * pointer. That means "key" is a valid throtl_data pointer as long as we are * rcu read lock. * * "key" was fetched from blkio_group under blkio_cgroup->lock. That means * it should not be NULL as even if queue was going away, cgroup deltion * path got to it first. */ void throtl_unlink_blkio_group(void *key, struct blkio_group *blkg) { unsigned long flags; struct throtl_data *td = key; spin_lock_irqsave(td->queue->queue_lock, flags); throtl_destroy_tg(td, tg_of_blkg(blkg)); spin_unlock_irqrestore(td->queue->queue_lock, flags); } |
de701c74a blk-throttle: Som... |
986 987 988 989 990 991 992 993 |
static void throtl_update_blkio_group_common(struct throtl_data *td, struct throtl_grp *tg) { xchg(&tg->limits_changed, true); xchg(&td->limits_changed, true); /* Schedule a work now to process the limit change */ throtl_schedule_delayed_work(td, 0); } |
fe0714377 blkio: Recalculat... |
994 995 996 |
/* * For all update functions, key should be a valid pointer because these * update functions are called under blkcg_lock, that means, blkg is |
25985edce Fix common misspe... |
997 |
* valid and in turn key is valid. queue exit path can not race because |
fe0714377 blkio: Recalculat... |
998 999 1000 1001 1002 1003 1004 |
* of blkcg_lock * * Can not take queue lock in update functions as queue lock under blkcg_lock * is not allowed. Under other paths we take blkcg_lock under queue_lock. */ static void throtl_update_blkio_group_read_bps(void *key, struct blkio_group *blkg, u64 read_bps) |
e43473b7f blkio: Core imple... |
1005 |
{ |
fe0714377 blkio: Recalculat... |
1006 |
struct throtl_data *td = key; |
de701c74a blk-throttle: Som... |
1007 |
struct throtl_grp *tg = tg_of_blkg(blkg); |
fe0714377 blkio: Recalculat... |
1008 |
|
de701c74a blk-throttle: Som... |
1009 1010 |
tg->bps[READ] = read_bps; throtl_update_blkio_group_common(td, tg); |
e43473b7f blkio: Core imple... |
1011 |
} |
fe0714377 blkio: Recalculat... |
1012 1013 |
static void throtl_update_blkio_group_write_bps(void *key, struct blkio_group *blkg, u64 write_bps) |
e43473b7f blkio: Core imple... |
1014 |
{ |
fe0714377 blkio: Recalculat... |
1015 |
struct throtl_data *td = key; |
de701c74a blk-throttle: Som... |
1016 |
struct throtl_grp *tg = tg_of_blkg(blkg); |
fe0714377 blkio: Recalculat... |
1017 |
|
de701c74a blk-throttle: Som... |
1018 1019 |
tg->bps[WRITE] = write_bps; throtl_update_blkio_group_common(td, tg); |
e43473b7f blkio: Core imple... |
1020 |
} |
fe0714377 blkio: Recalculat... |
1021 1022 |
static void throtl_update_blkio_group_read_iops(void *key, struct blkio_group *blkg, unsigned int read_iops) |
8e89d13f4 blkio: Implementa... |
1023 |
{ |
fe0714377 blkio: Recalculat... |
1024 |
struct throtl_data *td = key; |
de701c74a blk-throttle: Som... |
1025 |
struct throtl_grp *tg = tg_of_blkg(blkg); |
fe0714377 blkio: Recalculat... |
1026 |
|
de701c74a blk-throttle: Som... |
1027 1028 |
tg->iops[READ] = read_iops; throtl_update_blkio_group_common(td, tg); |
8e89d13f4 blkio: Implementa... |
1029 |
} |
fe0714377 blkio: Recalculat... |
1030 1031 |
static void throtl_update_blkio_group_write_iops(void *key, struct blkio_group *blkg, unsigned int write_iops) |
8e89d13f4 blkio: Implementa... |
1032 |
{ |
fe0714377 blkio: Recalculat... |
1033 |
struct throtl_data *td = key; |
de701c74a blk-throttle: Som... |
1034 |
struct throtl_grp *tg = tg_of_blkg(blkg); |
fe0714377 blkio: Recalculat... |
1035 |
|
de701c74a blk-throttle: Som... |
1036 1037 |
tg->iops[WRITE] = write_iops; throtl_update_blkio_group_common(td, tg); |
8e89d13f4 blkio: Implementa... |
1038 |
} |
da5277700 block: Move blk_t... |
1039 |
static void throtl_shutdown_wq(struct request_queue *q) |
e43473b7f blkio: Core imple... |
1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 |
{ struct throtl_data *td = q->td; cancel_delayed_work_sync(&td->throtl_work); } static struct blkio_policy_type blkio_policy_throtl = { .ops = { .blkio_unlink_group_fn = throtl_unlink_blkio_group, .blkio_update_group_read_bps_fn = throtl_update_blkio_group_read_bps, .blkio_update_group_write_bps_fn = throtl_update_blkio_group_write_bps, |
8e89d13f4 blkio: Implementa... |
1053 1054 1055 1056 |
.blkio_update_group_read_iops_fn = throtl_update_blkio_group_read_iops, .blkio_update_group_write_iops_fn = throtl_update_blkio_group_write_iops, |
e43473b7f blkio: Core imple... |
1057 |
}, |
8e89d13f4 blkio: Implementa... |
1058 |
.plid = BLKIO_POLICY_THROTL, |
e43473b7f blkio: Core imple... |
1059 |
}; |
bc16a4f93 block: reorganize... |
1060 |
bool blk_throtl_bio(struct request_queue *q, struct bio *bio) |
e43473b7f blkio: Core imple... |
1061 1062 1063 |
{ struct throtl_data *td = q->td; struct throtl_grp *tg; |
e43473b7f blkio: Core imple... |
1064 |
bool rw = bio_data_dir(bio), update_disptime = true; |
af75cd3c6 blk-throttle: Mak... |
1065 |
struct blkio_cgroup *blkcg; |
bc16a4f93 block: reorganize... |
1066 |
bool throttled = false; |
e43473b7f blkio: Core imple... |
1067 1068 1069 |
if (bio->bi_rw & REQ_THROTTLED) { bio->bi_rw &= ~REQ_THROTTLED; |
bc16a4f93 block: reorganize... |
1070 |
goto out; |
e43473b7f blkio: Core imple... |
1071 |
} |
af75cd3c6 blk-throttle: Mak... |
1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 |
/* * A throtl_grp pointer retrieved under rcu can be used to access * basic fields like stats and io rates. If a group has no rules, * just update the dispatch stats in lockless manner and return. */ rcu_read_lock(); blkcg = task_blkio_cgroup(current); tg = throtl_find_tg(td, blkcg); if (tg) { throtl_tg_fill_dev_details(td, tg); if (tg_no_rule_group(tg, rw)) { blkiocg_update_dispatch_stats(&tg->blkg, bio->bi_size, |
e5a94f568 blk-throttle: cor... |
1086 |
rw, rw_is_sync(bio->bi_rw)); |
af75cd3c6 blk-throttle: Mak... |
1087 |
rcu_read_unlock(); |
bc16a4f93 block: reorganize... |
1088 |
goto out; |
af75cd3c6 blk-throttle: Mak... |
1089 1090 1091 1092 1093 1094 1095 1096 |
} } rcu_read_unlock(); /* * Either group has not been allocated yet or it is not an unlimited * IO group */ |
e43473b7f blkio: Core imple... |
1097 1098 |
spin_lock_irq(q->queue_lock); tg = throtl_get_tg(td); |
bc16a4f93 block: reorganize... |
1099 1100 |
if (unlikely(!tg)) goto out_unlock; |
f469a7b4d blk-cgroup: Allow... |
1101 |
|
e43473b7f blkio: Core imple... |
1102 1103 1104 1105 1106 |
if (tg->nr_queued[rw]) { /* * There is already another bio queued in same dir. No * need to update dispatch time. */ |
231d704b4 blk-throttle: pro... |
1107 |
update_disptime = false; |
e43473b7f blkio: Core imple... |
1108 |
goto queue_bio; |
de701c74a blk-throttle: Som... |
1109 |
|
e43473b7f blkio: Core imple... |
1110 1111 1112 1113 1114 |
} /* Bio is with-in rate limit of group */ if (tg_may_dispatch(td, tg, bio, NULL)) { throtl_charge_bio(tg, bio); |
04521db04 blk-throttle: Res... |
1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 |
/* * We need to trim slice even when bios are not being queued * otherwise it might happen that a bio is not queued for * a long time and slice keeps on extending and trim is not * called for a long time. Now if limits are reduced suddenly * we take into account all the IO dispatched so far at new * low rate and * newly queued IO gets a really long dispatch * time. * * So keep on trimming slice even if bio is not queued. */ throtl_trim_slice(td, tg, rw); |
bc16a4f93 block: reorganize... |
1128 |
goto out_unlock; |
e43473b7f blkio: Core imple... |
1129 1130 1131 |
} queue_bio: |
fd16d2631 block: Add __attr... |
1132 |
throtl_log_tg(td, tg, "[%c] bio. bdisp=%llu sz=%u bps=%llu" |
8e89d13f4 blkio: Implementa... |
1133 1134 |
" iodisp=%u iops=%u queued=%d/%d", rw == READ ? 'R' : 'W', |
e43473b7f blkio: Core imple... |
1135 |
tg->bytes_disp[rw], bio->bi_size, tg->bps[rw], |
8e89d13f4 blkio: Implementa... |
1136 |
tg->io_disp[rw], tg->iops[rw], |
e43473b7f blkio: Core imple... |
1137 1138 1139 |
tg->nr_queued[READ], tg->nr_queued[WRITE]); throtl_add_bio_tg(q->td, tg, bio); |
bc16a4f93 block: reorganize... |
1140 |
throttled = true; |
e43473b7f blkio: Core imple... |
1141 1142 1143 1144 1145 |
if (update_disptime) { tg_update_disptime(td, tg); throtl_schedule_next_dispatch(td); } |
bc16a4f93 block: reorganize... |
1146 |
out_unlock: |
e43473b7f blkio: Core imple... |
1147 |
spin_unlock_irq(q->queue_lock); |
bc16a4f93 block: reorganize... |
1148 1149 |
out: return throttled; |
e43473b7f blkio: Core imple... |
1150 |
} |
c9a929dde block: fix reques... |
1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 |
/** * blk_throtl_drain - drain throttled bios * @q: request_queue to drain throttled bios for * * Dispatch all currently throttled bios on @q through ->make_request_fn(). */ void blk_throtl_drain(struct request_queue *q) __releases(q->queue_lock) __acquires(q->queue_lock) { struct throtl_data *td = q->td; struct throtl_rb_root *st = &td->tg_service_tree; struct throtl_grp *tg; struct bio_list bl; struct bio *bio; |
334c2b0b8 blk-throttle: use... |
1165 |
WARN_ON_ONCE(!queue_is_locked(q)); |
c9a929dde block: fix reques... |
1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 |
bio_list_init(&bl); while ((tg = throtl_rb_first(st))) { throtl_dequeue_tg(td, tg); while ((bio = bio_list_peek(&tg->bio_lists[READ]))) tg_dispatch_one_bio(td, tg, bio_data_dir(bio), &bl); while ((bio = bio_list_peek(&tg->bio_lists[WRITE]))) tg_dispatch_one_bio(td, tg, bio_data_dir(bio), &bl); } spin_unlock_irq(q->queue_lock); while ((bio = bio_list_pop(&bl))) generic_make_request(bio); spin_lock_irq(q->queue_lock); } |
e43473b7f blkio: Core imple... |
1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 |
int blk_throtl_init(struct request_queue *q) { struct throtl_data *td; struct throtl_grp *tg; td = kzalloc_node(sizeof(*td), GFP_KERNEL, q->node); if (!td) return -ENOMEM; INIT_HLIST_HEAD(&td->tg_list); td->tg_service_tree = THROTL_RB_ROOT; |
de701c74a blk-throttle: Som... |
1195 |
td->limits_changed = false; |
a29a171e7 blk-throttle: Do ... |
1196 |
INIT_DELAYED_WORK(&td->throtl_work, blk_throtl_work); |
e43473b7f blkio: Core imple... |
1197 |
|
29b125892 blk-throttle: Dyn... |
1198 1199 1200 |
/* alloc and Init root group. */ td->queue = q; tg = throtl_alloc_tg(td); |
02977e4af blkio: Add root g... |
1201 |
|
29b125892 blk-throttle: Dyn... |
1202 1203 1204 1205 1206 1207 |
if (!tg) { kfree(td); return -ENOMEM; } td->root_tg = tg; |
e43473b7f blkio: Core imple... |
1208 1209 |
rcu_read_lock(); |
5617cbef7 blk-throttle: Use... |
1210 |
throtl_init_add_tg_lists(td, tg, &blkio_root_cgroup); |
e43473b7f blkio: Core imple... |
1211 1212 1213 |
rcu_read_unlock(); /* Attach throtl data to request queue */ |
e43473b7f blkio: Core imple... |
1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 |
q->td = td; return 0; } void blk_throtl_exit(struct request_queue *q) { struct throtl_data *td = q->td; bool wait = false; BUG_ON(!td); |
da5277700 block: Move blk_t... |
1224 |
throtl_shutdown_wq(q); |
e43473b7f blkio: Core imple... |
1225 1226 1227 |
spin_lock_irq(q->queue_lock); throtl_release_tgs(td); |
e43473b7f blkio: Core imple... |
1228 1229 |
/* If there are other groups */ |
02977e4af blkio: Add root g... |
1230 |
if (td->nr_undestroyed_grps > 0) |
e43473b7f blkio: Core imple... |
1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 |
wait = true; spin_unlock_irq(q->queue_lock); /* * Wait for tg->blkg->key accessors to exit their grace periods. * Do this wait only if there are other undestroyed groups out * there (other than root group). This can happen if cgroup deletion * path claimed the responsibility of cleaning up a group before * queue cleanup code get to the group. * * Do not call synchronize_rcu() unconditionally as there are drivers * which create/delete request queue hundreds of times during scan/boot * and synchronize_rcu() can take significant time and slow down boot. */ if (wait) synchronize_rcu(); |
fe0714377 blkio: Recalculat... |
1248 1249 1250 1251 1252 1253 |
/* * Just being safe to make sure after previous flush if some body did * update limits through cgroup and another work got queued, cancel * it. */ |
da5277700 block: Move blk_t... |
1254 |
throtl_shutdown_wq(q); |
c9a929dde block: fix reques... |
1255 1256 1257 1258 1259 |
} void blk_throtl_release(struct request_queue *q) { kfree(q->td); |
e43473b7f blkio: Core imple... |
1260 1261 1262 1263 |
} static int __init throtl_init(void) { |
450adcbe5 blk-throttle: Do ... |
1264 1265 1266 1267 |
kthrotld_workqueue = alloc_workqueue("kthrotld", WQ_MEM_RECLAIM, 0); if (!kthrotld_workqueue) panic("Failed to create kthrotld "); |
e43473b7f blkio: Core imple... |
1268 1269 1270 1271 1272 |
blkio_policy_register(&blkio_policy_throtl); return 0; } module_init(throtl_init); |