Blame view
fs/fs-writeback.c
74.8 KB
457c89965
|
1 |
// SPDX-License-Identifier: GPL-2.0-only |
1da177e4c
|
2 3 4 5 6 7 8 9 10 11 |
/* * fs/fs-writeback.c * * Copyright (C) 2002, Linus Torvalds. * * Contains all the functions related to writing back and waiting * upon dirty inodes against superblocks, and writing back dirty * pages against inodes. ie: data writeback. Writeout of the * inode itself is not handled here. * |
e1f8e8744
|
12 |
* 10Apr2002 Andrew Morton |
1da177e4c
|
13 14 15 16 17 |
* Split out of fs/inode.c * Additions for address_space-based writeback */ #include <linux/kernel.h> |
630d9c472
|
18 |
#include <linux/export.h> |
1da177e4c
|
19 |
#include <linux/spinlock.h> |
5a0e3ad6a
|
20 |
#include <linux/slab.h> |
1da177e4c
|
21 22 23 |
#include <linux/sched.h> #include <linux/fs.h> #include <linux/mm.h> |
bc31b86a5
|
24 |
#include <linux/pagemap.h> |
03ba3782e
|
25 |
#include <linux/kthread.h> |
1da177e4c
|
26 27 28 |
#include <linux/writeback.h> #include <linux/blkdev.h> #include <linux/backing-dev.h> |
455b28646
|
29 |
#include <linux/tracepoint.h> |
719ea2fbb
|
30 |
#include <linux/device.h> |
21c6321fb
|
31 |
#include <linux/memcontrol.h> |
07f3f05c1
|
32 |
#include "internal.h" |
1da177e4c
|
33 |
|
d0bceac74
|
34 |
/* |
bc31b86a5
|
35 36 |
* 4MB minimal write chunk size */ |
09cbfeaf1
|
37 |
#define MIN_WRITEBACK_PAGES (4096UL >> (PAGE_SHIFT - 10)) |
bc31b86a5
|
38 39 |
/* |
c4a77a6c7
|
40 41 |
* Passed into wb_writeback(), essentially a subset of writeback_control */ |
83ba7b071
|
42 |
struct wb_writeback_work { |
c4a77a6c7
|
43 44 45 |
long nr_pages; struct super_block *sb; enum writeback_sync_modes sync_mode; |
6e6938b6d
|
46 |
unsigned int tagged_writepages:1; |
52957fe1c
|
47 48 49 |
unsigned int for_kupdate:1; unsigned int range_cyclic:1; unsigned int for_background:1; |
7747bd4bc
|
50 |
unsigned int for_sync:1; /* sync(2) WB_SYNC_ALL writeback */ |
ac7b19a34
|
51 |
unsigned int auto_free:1; /* free on completion */ |
0e175a183
|
52 |
enum wb_reason reason; /* why was writeback initiated? */ |
c4a77a6c7
|
53 |
|
8010c3b63
|
54 |
struct list_head list; /* pending work list */ |
cc395d7f1
|
55 |
struct wb_completion *done; /* set if the caller waits */ |
03ba3782e
|
56 |
}; |
a2f487069
|
57 58 59 60 61 62 63 64 65 66 67 |
/* * If an inode is constantly having its pages dirtied, but then the * updates stop dirtytime_expire_interval seconds in the past, it's * possible for the worst case time between when an inode has its * timestamps updated and when they finally get written out to be two * dirtytime_expire_intervals. We set the default to 12 hours (in * seconds), which means most of the time inodes will have their * timestamps written to disk after 12 hours, but in the worst case a * few inodes might not their timestamps updated for 24 hours. */ unsigned int dirtytime_expire_interval = 12 * 60 * 60; |
7ccf19a80
|
68 69 |
static inline struct inode *wb_inode(struct list_head *head) { |
c7f540849
|
70 |
return list_entry(head, struct inode, i_io_list); |
7ccf19a80
|
71 |
} |
15eb77a07
|
72 73 74 75 76 77 78 |
/* * Include the creation of the trace points after defining the * wb_writeback_work structure and inline functions so that the definition * remains local to this file. */ #define CREATE_TRACE_POINTS #include <trace/events/writeback.h> |
774016b2d
|
79 |
EXPORT_TRACEPOINT_SYMBOL_GPL(wbc_writepage); |
d6c10f1fc
|
80 81 82 83 84 85 |
static bool wb_io_lists_populated(struct bdi_writeback *wb) { if (wb_has_dirty_io(wb)) { return false; } else { set_bit(WB_has_dirty_io, &wb->state); |
95a46c65e
|
86 |
WARN_ON_ONCE(!wb->avg_write_bandwidth); |
766a9d6e6
|
87 88 |
atomic_long_add(wb->avg_write_bandwidth, &wb->bdi->tot_write_bandwidth); |
d6c10f1fc
|
89 90 91 92 93 94 95 |
return true; } } static void wb_io_lists_depopulated(struct bdi_writeback *wb) { if (wb_has_dirty_io(wb) && list_empty(&wb->b_dirty) && |
766a9d6e6
|
96 |
list_empty(&wb->b_io) && list_empty(&wb->b_more_io)) { |
d6c10f1fc
|
97 |
clear_bit(WB_has_dirty_io, &wb->state); |
95a46c65e
|
98 99 |
WARN_ON_ONCE(atomic_long_sub_return(wb->avg_write_bandwidth, &wb->bdi->tot_write_bandwidth) < 0); |
766a9d6e6
|
100 |
} |
d6c10f1fc
|
101 102 103 |
} /** |
c7f540849
|
104 |
* inode_io_list_move_locked - move an inode onto a bdi_writeback IO list |
d6c10f1fc
|
105 106 |
* @inode: inode to be moved * @wb: target bdi_writeback |
bbbc3c1cf
|
107 |
* @head: one of @wb->b_{dirty|io|more_io|dirty_time} |
d6c10f1fc
|
108 |
* |
c7f540849
|
109 |
* Move @inode->i_io_list to @list of @wb and set %WB_has_dirty_io. |
d6c10f1fc
|
110 111 112 |
* Returns %true if @inode is the first occupant of the !dirty_time IO * lists; otherwise, %false. */ |
c7f540849
|
113 |
static bool inode_io_list_move_locked(struct inode *inode, |
d6c10f1fc
|
114 115 116 117 |
struct bdi_writeback *wb, struct list_head *head) { assert_spin_locked(&wb->list_lock); |
c7f540849
|
118 |
list_move(&inode->i_io_list, head); |
d6c10f1fc
|
119 120 121 122 123 124 125 126 127 128 |
/* dirty_time doesn't count as dirty_io until expiration */ if (head != &wb->b_dirty_time) return wb_io_lists_populated(wb); wb_io_lists_depopulated(wb); return false; } /** |
c7f540849
|
129 |
* inode_io_list_del_locked - remove an inode from its bdi_writeback IO list |
d6c10f1fc
|
130 131 132 133 134 135 |
* @inode: inode to be removed * @wb: bdi_writeback @inode is being removed from * * Remove @inode which may be on one of @wb->b_{dirty|io|more_io} lists and * clear %WB_has_dirty_io if all are empty afterwards. */ |
c7f540849
|
136 |
static void inode_io_list_del_locked(struct inode *inode, |
d6c10f1fc
|
137 138 139 |
struct bdi_writeback *wb) { assert_spin_locked(&wb->list_lock); |
b35250c08
|
140 |
assert_spin_locked(&inode->i_lock); |
d6c10f1fc
|
141 |
|
5afced3bf
|
142 |
inode->i_state &= ~I_SYNC_QUEUED; |
c7f540849
|
143 |
list_del_init(&inode->i_io_list); |
d6c10f1fc
|
144 145 |
wb_io_lists_depopulated(wb); } |
f0054bb1e
|
146 |
static void wb_wakeup(struct bdi_writeback *wb) |
5acda9d12
|
147 |
{ |
f0054bb1e
|
148 149 150 151 |
spin_lock_bh(&wb->work_lock); if (test_bit(WB_registered, &wb->state)) mod_delayed_work(bdi_wq, &wb->dwork, 0); spin_unlock_bh(&wb->work_lock); |
5acda9d12
|
152 |
} |
4a3a485b1
|
153 154 155 156 157 158 159 |
static void finish_writeback_work(struct bdi_writeback *wb, struct wb_writeback_work *work) { struct wb_completion *done = work->done; if (work->auto_free) kfree(work); |
8e00c4e9d
|
160 161 162 163 164 165 166 |
if (done) { wait_queue_head_t *waitq = done->waitq; /* @done can't be accessed after the following dec */ if (atomic_dec_and_test(&done->cnt)) wake_up_all(waitq); } |
4a3a485b1
|
167 |
} |
f0054bb1e
|
168 169 |
static void wb_queue_work(struct bdi_writeback *wb, struct wb_writeback_work *work) |
6585027a5
|
170 |
{ |
5634cc2aa
|
171 |
trace_writeback_queue(wb, work); |
6585027a5
|
172 |
|
cc395d7f1
|
173 174 |
if (work->done) atomic_inc(&work->done->cnt); |
4a3a485b1
|
175 176 177 178 179 180 181 182 |
spin_lock_bh(&wb->work_lock); if (test_bit(WB_registered, &wb->state)) { list_add_tail(&work->list, &wb->work_list); mod_delayed_work(bdi_wq, &wb->dwork, 0); } else finish_writeback_work(wb, work); |
f0054bb1e
|
183 |
spin_unlock_bh(&wb->work_lock); |
1da177e4c
|
184 |
} |
cc395d7f1
|
185 186 |
/** * wb_wait_for_completion - wait for completion of bdi_writeback_works |
cc395d7f1
|
187 188 189 |
* @done: target wb_completion * * Wait for one or more work items issued to @bdi with their ->done field |
5b9cce4c7
|
190 191 192 |
* set to @done, which should have been initialized with * DEFINE_WB_COMPLETION(). This function returns after all such work items * are completed. Work items which are waited upon aren't freed |
cc395d7f1
|
193 194 |
* automatically on completion. */ |
5b9cce4c7
|
195 |
void wb_wait_for_completion(struct wb_completion *done) |
cc395d7f1
|
196 197 |
{ atomic_dec(&done->cnt); /* put down the initial count */ |
5b9cce4c7
|
198 |
wait_event(*done->waitq, !atomic_read(&done->cnt)); |
cc395d7f1
|
199 |
} |
703c27088
|
200 |
#ifdef CONFIG_CGROUP_WRITEBACK |
55a694dff
|
201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 |
/* * Parameters for foreign inode detection, see wbc_detach_inode() to see * how they're used. * * These paramters are inherently heuristical as the detection target * itself is fuzzy. All we want to do is detaching an inode from the * current owner if it's being written to by some other cgroups too much. * * The current cgroup writeback is built on the assumption that multiple * cgroups writing to the same inode concurrently is very rare and a mode * of operation which isn't well supported. As such, the goal is not * taking too long when a different cgroup takes over an inode while * avoiding too aggressive flip-flops from occasional foreign writes. * * We record, very roughly, 2s worth of IO time history and if more than * half of that is foreign, trigger the switch. The recording is quantized * to 16 slots. To avoid tiny writes from swinging the decision too much, * writes smaller than 1/8 of avg size are ignored. */ |
2a8149081
|
220 221 |
#define WB_FRN_TIME_SHIFT 13 /* 1s = 2^13, upto 8 secs w/ 16bit */ #define WB_FRN_TIME_AVG_SHIFT 3 /* avg = avg * 7/8 + new * 1/8 */ |
55a694dff
|
222 |
#define WB_FRN_TIME_CUT_DIV 8 /* ignore rounds < avg / 8 */ |
2a8149081
|
223 224 225 226 227 228 229 230 231 |
#define WB_FRN_TIME_PERIOD (2 * (1 << WB_FRN_TIME_SHIFT)) /* 2s */ #define WB_FRN_HIST_SLOTS 16 /* inode->i_wb_frn_history is 16bit */ #define WB_FRN_HIST_UNIT (WB_FRN_TIME_PERIOD / WB_FRN_HIST_SLOTS) /* each slot's duration is 2s / 16 */ #define WB_FRN_HIST_THR_SLOTS (WB_FRN_HIST_SLOTS / 2) /* if foreign slots >= 8, switch */ #define WB_FRN_HIST_MAX_SLOTS (WB_FRN_HIST_THR_SLOTS / 2 + 1) /* one round can affect upto 5 slots */ |
6444f47eb
|
232 |
#define WB_FRN_MAX_IN_FLIGHT 1024 /* don't queue too many concurrently */ |
2a8149081
|
233 |
|
a1a0e23e4
|
234 235 |
static atomic_t isw_nr_in_flight = ATOMIC_INIT(0); static struct workqueue_struct *isw_wq; |
21c6321fb
|
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 |
void __inode_attach_wb(struct inode *inode, struct page *page) { struct backing_dev_info *bdi = inode_to_bdi(inode); struct bdi_writeback *wb = NULL; if (inode_cgwb_enabled(inode)) { struct cgroup_subsys_state *memcg_css; if (page) { memcg_css = mem_cgroup_css_from_page(page); wb = wb_get_create(bdi, memcg_css, GFP_ATOMIC); } else { /* must pin memcg_css, see wb_get_create() */ memcg_css = task_get_css(current, memory_cgrp_id); wb = wb_get_create(bdi, memcg_css, GFP_ATOMIC); css_put(memcg_css); } } if (!wb) wb = &bdi->wb; /* * There may be multiple instances of this function racing to * update the same inode. Use cmpxchg() to tell the winner. */ if (unlikely(cmpxchg(&inode->i_wb, NULL, wb))) wb_put(wb); } |
9b0eb69b7
|
265 |
EXPORT_SYMBOL_GPL(__inode_attach_wb); |
21c6321fb
|
266 |
|
703c27088
|
267 |
/** |
87e1d789b
|
268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 |
* locked_inode_to_wb_and_lock_list - determine a locked inode's wb and lock it * @inode: inode of interest with i_lock held * * Returns @inode's wb with its list_lock held. @inode->i_lock must be * held on entry and is released on return. The returned wb is guaranteed * to stay @inode's associated wb until its list_lock is released. */ static struct bdi_writeback * locked_inode_to_wb_and_lock_list(struct inode *inode) __releases(&inode->i_lock) __acquires(&wb->list_lock) { while (true) { struct bdi_writeback *wb = inode_to_wb(inode); /* * inode_to_wb() association is protected by both * @inode->i_lock and @wb->list_lock but list_lock nests * outside i_lock. Drop i_lock and verify that the * association hasn't changed after acquiring list_lock. */ wb_get(wb); spin_unlock(&inode->i_lock); spin_lock(&wb->list_lock); |
87e1d789b
|
292 |
|
aaa2cacf8
|
293 |
/* i_wb may have changed inbetween, can't use inode_to_wb() */ |
614a4e377
|
294 295 296 297 |
if (likely(wb == inode->i_wb)) { wb_put(wb); /* @inode already has ref */ return wb; } |
87e1d789b
|
298 299 |
spin_unlock(&wb->list_lock); |
614a4e377
|
300 |
wb_put(wb); |
87e1d789b
|
301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 |
cpu_relax(); spin_lock(&inode->i_lock); } } /** * inode_to_wb_and_lock_list - determine an inode's wb and lock it * @inode: inode of interest * * Same as locked_inode_to_wb_and_lock_list() but @inode->i_lock isn't held * on entry. */ static struct bdi_writeback *inode_to_wb_and_lock_list(struct inode *inode) __acquires(&wb->list_lock) { spin_lock(&inode->i_lock); return locked_inode_to_wb_and_lock_list(inode); } |
682aa8e1a
|
319 320 321 322 323 324 325 |
struct inode_switch_wbs_context { struct inode *inode; struct bdi_writeback *new_wb; struct rcu_head rcu_head; struct work_struct work; }; |
7fc5854f8
|
326 327 328 329 330 331 332 333 334 |
static void bdi_down_write_wb_switch_rwsem(struct backing_dev_info *bdi) { down_write(&bdi->wb_switch_rwsem); } static void bdi_up_write_wb_switch_rwsem(struct backing_dev_info *bdi) { up_write(&bdi->wb_switch_rwsem); } |
682aa8e1a
|
335 336 337 338 339 |
static void inode_switch_wbs_work_fn(struct work_struct *work) { struct inode_switch_wbs_context *isw = container_of(work, struct inode_switch_wbs_context, work); struct inode *inode = isw->inode; |
7fc5854f8
|
340 |
struct backing_dev_info *bdi = inode_to_bdi(inode); |
d10c80955
|
341 342 |
struct address_space *mapping = inode->i_mapping; struct bdi_writeback *old_wb = inode->i_wb; |
682aa8e1a
|
343 |
struct bdi_writeback *new_wb = isw->new_wb; |
04edf02cd
|
344 345 |
XA_STATE(xas, &mapping->i_pages, 0); struct page *page; |
d10c80955
|
346 |
bool switched = false; |
682aa8e1a
|
347 348 |
/* |
7fc5854f8
|
349 350 351 352 353 354 |
* If @inode switches cgwb membership while sync_inodes_sb() is * being issued, sync_inodes_sb() might miss it. Synchronize. */ down_read(&bdi->wb_switch_rwsem); /* |
682aa8e1a
|
355 356 357 |
* By the time control reaches here, RCU grace period has passed * since I_WB_SWITCH assertion and all wb stat update transactions * between unlocked_inode_to_wb_begin/end() are guaranteed to be |
b93b01631
|
358 |
* synchronizing against the i_pages lock. |
d10c80955
|
359 |
* |
b93b01631
|
360 |
* Grabbing old_wb->list_lock, inode->i_lock and the i_pages lock |
d10c80955
|
361 362 |
* gives us exclusion against all wb related operations on @inode * including IO list manipulations and stat updates. |
682aa8e1a
|
363 |
*/ |
d10c80955
|
364 365 366 367 368 369 370 |
if (old_wb < new_wb) { spin_lock(&old_wb->list_lock); spin_lock_nested(&new_wb->list_lock, SINGLE_DEPTH_NESTING); } else { spin_lock(&new_wb->list_lock); spin_lock_nested(&old_wb->list_lock, SINGLE_DEPTH_NESTING); } |
682aa8e1a
|
371 |
spin_lock(&inode->i_lock); |
b93b01631
|
372 |
xa_lock_irq(&mapping->i_pages); |
d10c80955
|
373 374 375 |
/* * Once I_FREEING is visible under i_lock, the eviction path owns |
c7f540849
|
376 |
* the inode and we shouldn't modify ->i_io_list. |
d10c80955
|
377 378 379 |
*/ if (unlikely(inode->i_state & I_FREEING)) goto skip_switch; |
3a8e9ac89
|
380 |
trace_inode_switch_wbs(inode, old_wb, new_wb); |
d10c80955
|
381 382 383 |
/* * Count and transfer stats. Note that PAGECACHE_TAG_DIRTY points * to possibly dirty pages while PAGECACHE_TAG_WRITEBACK points to |
b93b01631
|
384 |
* pages actually under writeback. |
d10c80955
|
385 |
*/ |
04edf02cd
|
386 387 |
xas_for_each_marked(&xas, page, ULONG_MAX, PAGECACHE_TAG_DIRTY) { if (PageDirty(page)) { |
3e8f399da
|
388 389 |
dec_wb_stat(old_wb, WB_RECLAIMABLE); inc_wb_stat(new_wb, WB_RECLAIMABLE); |
d10c80955
|
390 391 |
} } |
04edf02cd
|
392 393 394 395 396 |
xas_set(&xas, 0); xas_for_each_marked(&xas, page, ULONG_MAX, PAGECACHE_TAG_WRITEBACK) { WARN_ON_ONCE(!PageWriteback(page)); dec_wb_stat(old_wb, WB_WRITEBACK); inc_wb_stat(new_wb, WB_WRITEBACK); |
d10c80955
|
397 398 399 400 401 402 403 404 405 406 |
} wb_get(new_wb); /* * Transfer to @new_wb's IO list if necessary. The specific list * @inode was on is ignored and the inode is put on ->b_dirty which * is always correct including from ->b_dirty_time. The transfer * preserves @inode->dirtied_when ordering. */ |
c7f540849
|
407 |
if (!list_empty(&inode->i_io_list)) { |
d10c80955
|
408 |
struct inode *pos; |
c7f540849
|
409 |
inode_io_list_del_locked(inode, old_wb); |
d10c80955
|
410 |
inode->i_wb = new_wb; |
c7f540849
|
411 |
list_for_each_entry(pos, &new_wb->b_dirty, i_io_list) |
d10c80955
|
412 413 414 |
if (time_after_eq(inode->dirtied_when, pos->dirtied_when)) break; |
c7f540849
|
415 |
inode_io_list_move_locked(inode, new_wb, pos->i_io_list.prev); |
d10c80955
|
416 417 418 |
} else { inode->i_wb = new_wb; } |
682aa8e1a
|
419 |
|
d10c80955
|
420 |
/* ->i_wb_frn updates may race wbc_detach_inode() but doesn't matter */ |
682aa8e1a
|
421 422 423 |
inode->i_wb_frn_winner = 0; inode->i_wb_frn_avg_time = 0; inode->i_wb_frn_history = 0; |
d10c80955
|
424 425 |
switched = true; skip_switch: |
682aa8e1a
|
426 427 428 429 430 |
/* * Paired with load_acquire in unlocked_inode_to_wb_begin() and * ensures that the new wb is visible if they see !I_WB_SWITCH. */ smp_store_release(&inode->i_state, inode->i_state & ~I_WB_SWITCH); |
b93b01631
|
431 |
xa_unlock_irq(&mapping->i_pages); |
682aa8e1a
|
432 |
spin_unlock(&inode->i_lock); |
d10c80955
|
433 434 |
spin_unlock(&new_wb->list_lock); spin_unlock(&old_wb->list_lock); |
682aa8e1a
|
435 |
|
7fc5854f8
|
436 |
up_read(&bdi->wb_switch_rwsem); |
d10c80955
|
437 438 439 440 |
if (switched) { wb_wakeup(new_wb); wb_put(old_wb); } |
682aa8e1a
|
441 |
wb_put(new_wb); |
d10c80955
|
442 443 |
iput(inode); |
682aa8e1a
|
444 |
kfree(isw); |
a1a0e23e4
|
445 446 |
atomic_dec(&isw_nr_in_flight); |
682aa8e1a
|
447 448 449 450 451 452 453 454 455 |
} static void inode_switch_wbs_rcu_fn(struct rcu_head *rcu_head) { struct inode_switch_wbs_context *isw = container_of(rcu_head, struct inode_switch_wbs_context, rcu_head); /* needs to grab bh-unsafe locks, bounce to work item */ INIT_WORK(&isw->work, inode_switch_wbs_work_fn); |
a1a0e23e4
|
456 |
queue_work(isw_wq, &isw->work); |
682aa8e1a
|
457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 |
} /** * inode_switch_wbs - change the wb association of an inode * @inode: target inode * @new_wb_id: ID of the new wb * * Switch @inode's wb association to the wb identified by @new_wb_id. The * switching is performed asynchronously and may fail silently. */ static void inode_switch_wbs(struct inode *inode, int new_wb_id) { struct backing_dev_info *bdi = inode_to_bdi(inode); struct cgroup_subsys_state *memcg_css; struct inode_switch_wbs_context *isw; /* noop if seems to be already in progress */ if (inode->i_state & I_WB_SWITCH) return; |
6444f47eb
|
476 477 |
/* avoid queueing a new switch if too many are already in flight */ if (atomic_read(&isw_nr_in_flight) > WB_FRN_MAX_IN_FLIGHT) |
7fc5854f8
|
478 |
return; |
682aa8e1a
|
479 480 |
isw = kzalloc(sizeof(*isw), GFP_ATOMIC); if (!isw) |
6444f47eb
|
481 |
return; |
682aa8e1a
|
482 483 484 485 486 487 488 489 490 491 492 493 |
/* find and pin the new wb */ rcu_read_lock(); memcg_css = css_from_id(new_wb_id, &memory_cgrp_subsys); if (memcg_css) isw->new_wb = wb_get_create(bdi, memcg_css, GFP_ATOMIC); rcu_read_unlock(); if (!isw->new_wb) goto out_free; /* while holding I_WB_SWITCH, no one else can update the association */ spin_lock(&inode->i_lock); |
1751e8a6c
|
494 |
if (!(inode->i_sb->s_flags & SB_ACTIVE) || |
a1a0e23e4
|
495 496 497 498 499 |
inode->i_state & (I_WB_SWITCH | I_FREEING) || inode_to_wb(inode) == isw->new_wb) { spin_unlock(&inode->i_lock); goto out_free; } |
682aa8e1a
|
500 |
inode->i_state |= I_WB_SWITCH; |
745249555
|
501 |
__iget(inode); |
682aa8e1a
|
502 |
spin_unlock(&inode->i_lock); |
682aa8e1a
|
503 504 505 506 |
isw->inode = inode; /* * In addition to synchronizing among switchers, I_WB_SWITCH tells |
b93b01631
|
507 508 |
* the RCU protected stat update paths to grab the i_page * lock so that stat transfer can synchronize against them. |
682aa8e1a
|
509 510 511 |
* Let's continue after I_WB_SWITCH is guaranteed to be visible. */ call_rcu(&isw->rcu_head, inode_switch_wbs_rcu_fn); |
ec084de92
|
512 513 |
atomic_inc(&isw_nr_in_flight); |
6444f47eb
|
514 |
return; |
682aa8e1a
|
515 516 517 518 519 520 |
out_free: if (isw->new_wb) wb_put(isw->new_wb); kfree(isw); } |
87e1d789b
|
521 |
/** |
b16b1deb5
|
522 523 524 525 526 527 528 529 530 531 532 533 |
* wbc_attach_and_unlock_inode - associate wbc with target inode and unlock it * @wbc: writeback_control of interest * @inode: target inode * * @inode is locked and about to be written back under the control of @wbc. * Record @inode's writeback context into @wbc and unlock the i_lock. On * writeback completion, wbc_detach_inode() should be called. This is used * to track the cgroup writeback context. */ void wbc_attach_and_unlock_inode(struct writeback_control *wbc, struct inode *inode) { |
dd73e4b7d
|
534 535 536 537 |
if (!inode_cgwb_enabled(inode)) { spin_unlock(&inode->i_lock); return; } |
b16b1deb5
|
538 |
wbc->wb = inode_to_wb(inode); |
2a8149081
|
539 540 541 542 543 544 545 546 |
wbc->inode = inode; wbc->wb_id = wbc->wb->memcg_css->id; wbc->wb_lcand_id = inode->i_wb_frn_winner; wbc->wb_tcand_id = 0; wbc->wb_bytes = 0; wbc->wb_lcand_bytes = 0; wbc->wb_tcand_bytes = 0; |
b16b1deb5
|
547 548 |
wb_get(wbc->wb); spin_unlock(&inode->i_lock); |
e8a7abf5a
|
549 550 |
/* |
65de03e25
|
551 552 553 554 555 |
* A dying wb indicates that either the blkcg associated with the * memcg changed or the associated memcg is dying. In the first * case, a replacement wb should already be available and we should * refresh the wb immediately. In the second case, trying to * refresh will keep failing. |
e8a7abf5a
|
556 |
*/ |
65de03e25
|
557 |
if (unlikely(wb_dying(wbc->wb) && !css_is_dying(wbc->wb->memcg_css))) |
e8a7abf5a
|
558 |
inode_switch_wbs(inode, wbc->wb_id); |
b16b1deb5
|
559 |
} |
9b0eb69b7
|
560 |
EXPORT_SYMBOL_GPL(wbc_attach_and_unlock_inode); |
b16b1deb5
|
561 562 |
/** |
2a8149081
|
563 564 |
* wbc_detach_inode - disassociate wbc from inode and perform foreign detection * @wbc: writeback_control of the just finished writeback |
b16b1deb5
|
565 566 567 |
* * To be called after a writeback attempt of an inode finishes and undoes * wbc_attach_and_unlock_inode(). Can be called under any context. |
2a8149081
|
568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 |
* * As concurrent write sharing of an inode is expected to be very rare and * memcg only tracks page ownership on first-use basis severely confining * the usefulness of such sharing, cgroup writeback tracks ownership * per-inode. While the support for concurrent write sharing of an inode * is deemed unnecessary, an inode being written to by different cgroups at * different points in time is a lot more common, and, more importantly, * charging only by first-use can too readily lead to grossly incorrect * behaviors (single foreign page can lead to gigabytes of writeback to be * incorrectly attributed). * * To resolve this issue, cgroup writeback detects the majority dirtier of * an inode and transfers the ownership to it. To avoid unnnecessary * oscillation, the detection mechanism keeps track of history and gives * out the switch verdict only if the foreign usage pattern is stable over * a certain amount of time and/or writeback attempts. * * On each writeback attempt, @wbc tries to detect the majority writer * using Boyer-Moore majority vote algorithm. In addition to the byte * count from the majority voting, it also counts the bytes written for the * current wb and the last round's winner wb (max of last round's current * wb, the winner from two rounds ago, and the last round's majority * candidate). Keeping track of the historical winner helps the algorithm * to semi-reliably detect the most active writer even when it's not the * absolute majority. * * Once the winner of the round is determined, whether the winner is * foreign or not and how much IO time the round consumed is recorded in * inode->i_wb_frn_history. If the amount of recorded foreign IO time is * over a certain threshold, the switch verdict is given. |
b16b1deb5
|
598 599 600 |
*/ void wbc_detach_inode(struct writeback_control *wbc) { |
2a8149081
|
601 602 |
struct bdi_writeback *wb = wbc->wb; struct inode *inode = wbc->inode; |
dd73e4b7d
|
603 604 |
unsigned long avg_time, max_bytes, max_time; u16 history; |
2a8149081
|
605 |
int max_id; |
dd73e4b7d
|
606 607 608 609 610 |
if (!wb) return; history = inode->i_wb_frn_history; avg_time = inode->i_wb_frn_avg_time; |
2a8149081
|
611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 |
/* pick the winner of this round */ if (wbc->wb_bytes >= wbc->wb_lcand_bytes && wbc->wb_bytes >= wbc->wb_tcand_bytes) { max_id = wbc->wb_id; max_bytes = wbc->wb_bytes; } else if (wbc->wb_lcand_bytes >= wbc->wb_tcand_bytes) { max_id = wbc->wb_lcand_id; max_bytes = wbc->wb_lcand_bytes; } else { max_id = wbc->wb_tcand_id; max_bytes = wbc->wb_tcand_bytes; } /* * Calculate the amount of IO time the winner consumed and fold it * into the running average kept per inode. If the consumed IO * time is lower than avag / WB_FRN_TIME_CUT_DIV, ignore it for * deciding whether to switch or not. This is to prevent one-off * small dirtiers from skewing the verdict. */ max_time = DIV_ROUND_UP((max_bytes >> PAGE_SHIFT) << WB_FRN_TIME_SHIFT, wb->avg_write_bandwidth); if (avg_time) avg_time += (max_time >> WB_FRN_TIME_AVG_SHIFT) - (avg_time >> WB_FRN_TIME_AVG_SHIFT); else avg_time = max_time; /* immediate catch up on first run */ if (max_time >= avg_time / WB_FRN_TIME_CUT_DIV) { int slots; /* * The switch verdict is reached if foreign wb's consume * more than a certain proportion of IO time in a * WB_FRN_TIME_PERIOD. This is loosely tracked by 16 slot * history mask where each bit represents one sixteenth of * the period. Determine the number of slots to shift into * history from @max_time. */ slots = min(DIV_ROUND_UP(max_time, WB_FRN_HIST_UNIT), (unsigned long)WB_FRN_HIST_MAX_SLOTS); history <<= slots; if (wbc->wb_id != max_id) history |= (1U << slots) - 1; |
3a8e9ac89
|
655 656 |
if (history) trace_inode_foreign_history(inode, wbc, history); |
2a8149081
|
657 658 659 660 661 662 663 |
/* * Switch if the current wb isn't the consistent winner. * If there are multiple closely competing dirtiers, the * inode may switch across them repeatedly over time, which * is okay. The main goal is avoiding keeping an inode on * the wrong wb for an extended period of time. */ |
682aa8e1a
|
664 665 |
if (hweight32(history) > WB_FRN_HIST_THR_SLOTS) inode_switch_wbs(inode, max_id); |
2a8149081
|
666 667 668 669 670 671 672 673 674 |
} /* * Multiple instances of this function may race to update the * following fields but we don't mind occassional inaccuracies. */ inode->i_wb_frn_winner = max_id; inode->i_wb_frn_avg_time = min(avg_time, (unsigned long)U16_MAX); inode->i_wb_frn_history = history; |
b16b1deb5
|
675 676 677 |
wb_put(wbc->wb); wbc->wb = NULL; } |
9b0eb69b7
|
678 |
EXPORT_SYMBOL_GPL(wbc_detach_inode); |
b16b1deb5
|
679 680 |
/** |
34e51a5e1
|
681 |
* wbc_account_cgroup_owner - account writeback to update inode cgroup ownership |
2a8149081
|
682 683 684 685 686 687 688 689 |
* @wbc: writeback_control of the writeback in progress * @page: page being written out * @bytes: number of bytes being written out * * @bytes from @page are about to written out during the writeback * controlled by @wbc. Keep the book for foreign inode detection. See * wbc_detach_inode(). */ |
34e51a5e1
|
690 691 |
void wbc_account_cgroup_owner(struct writeback_control *wbc, struct page *page, size_t bytes) |
2a8149081
|
692 |
{ |
663114222
|
693 |
struct cgroup_subsys_state *css; |
2a8149081
|
694 695 696 697 698 699 700 701 |
int id; /* * pageout() path doesn't attach @wbc to the inode being written * out. This is intentional as we don't want the function to block * behind a slow cgroup. Ultimately, we want pageout() to kick off * regular writeback instead of writing things out itself. */ |
27b36d8fa
|
702 |
if (!wbc->wb || wbc->no_cgroup_owner) |
2a8149081
|
703 |
return; |
663114222
|
704 705 706 707 708 709 |
css = mem_cgroup_css_from_page(page); /* dead cgroups shouldn't contribute to inode ownership arbitration */ if (!(css->flags & CSS_ONLINE)) return; id = css->id; |
2a8149081
|
710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 |
if (id == wbc->wb_id) { wbc->wb_bytes += bytes; return; } if (id == wbc->wb_lcand_id) wbc->wb_lcand_bytes += bytes; /* Boyer-Moore majority vote algorithm */ if (!wbc->wb_tcand_bytes) wbc->wb_tcand_id = id; if (id == wbc->wb_tcand_id) wbc->wb_tcand_bytes += bytes; else wbc->wb_tcand_bytes -= min(bytes, wbc->wb_tcand_bytes); } |
34e51a5e1
|
727 |
EXPORT_SYMBOL_GPL(wbc_account_cgroup_owner); |
2a8149081
|
728 729 |
/** |
703c27088
|
730 |
* inode_congested - test whether an inode is congested |
60292bcc1
|
731 |
* @inode: inode to test for congestion (may be NULL) |
703c27088
|
732 733 734 735 736 737 738 739 740 |
* @cong_bits: mask of WB_[a]sync_congested bits to test * * Tests whether @inode is congested. @cong_bits is the mask of congestion * bits to test and the return value is the mask of set bits. * * If cgroup writeback is enabled for @inode, the congestion state is * determined by whether the cgwb (cgroup bdi_writeback) for the blkcg * associated with @inode is congested; otherwise, the root wb's congestion * state is used. |
60292bcc1
|
741 742 743 |
* * @inode is allowed to be NULL as this function is often called on * mapping->host which is NULL for the swapper space. |
703c27088
|
744 745 746 |
*/ int inode_congested(struct inode *inode, int cong_bits) { |
5cb8b8241
|
747 748 749 750 |
/* * Once set, ->i_wb never becomes NULL while the inode is alive. * Start transaction iff ->i_wb is visible. */ |
aaa2cacf8
|
751 |
if (inode && inode_to_wb_is_valid(inode)) { |
5cb8b8241
|
752 |
struct bdi_writeback *wb; |
2e898e4c0
|
753 754 |
struct wb_lock_cookie lock_cookie = {}; bool congested; |
5cb8b8241
|
755 |
|
2e898e4c0
|
756 |
wb = unlocked_inode_to_wb_begin(inode, &lock_cookie); |
5cb8b8241
|
757 |
congested = wb_congested(wb, cong_bits); |
2e898e4c0
|
758 |
unlocked_inode_to_wb_end(inode, &lock_cookie); |
5cb8b8241
|
759 |
return congested; |
703c27088
|
760 761 762 763 764 |
} return wb_congested(&inode_to_bdi(inode)->wb, cong_bits); } EXPORT_SYMBOL_GPL(inode_congested); |
f2b651216
|
765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 |
/** * wb_split_bdi_pages - split nr_pages to write according to bandwidth * @wb: target bdi_writeback to split @nr_pages to * @nr_pages: number of pages to write for the whole bdi * * Split @wb's portion of @nr_pages according to @wb's write bandwidth in * relation to the total write bandwidth of all wb's w/ dirty inodes on * @wb->bdi. */ static long wb_split_bdi_pages(struct bdi_writeback *wb, long nr_pages) { unsigned long this_bw = wb->avg_write_bandwidth; unsigned long tot_bw = atomic_long_read(&wb->bdi->tot_write_bandwidth); if (nr_pages == LONG_MAX) return LONG_MAX; /* * This may be called on clean wb's and proportional distribution * may not make sense, just use the original @nr_pages in those * cases. In general, we wanna err on the side of writing more. */ if (!tot_bw || this_bw >= tot_bw) return nr_pages; else return DIV_ROUND_UP_ULL((u64)nr_pages * this_bw, tot_bw); } |
db1253604
|
792 |
/** |
db1253604
|
793 794 795 796 797 798 799 800 801 802 803 804 805 806 |
* bdi_split_work_to_wbs - split a wb_writeback_work to all wb's of a bdi * @bdi: target backing_dev_info * @base_work: wb_writeback_work to issue * @skip_if_busy: skip wb's which already have writeback in progress * * Split and issue @base_work to all wb's (bdi_writeback's) of @bdi which * have dirty inodes. If @base_work->nr_page isn't %LONG_MAX, it's * distributed to the busy wbs according to each wb's proportion in the * total active write bandwidth of @bdi. */ static void bdi_split_work_to_wbs(struct backing_dev_info *bdi, struct wb_writeback_work *base_work, bool skip_if_busy) { |
b817525a4
|
807 |
struct bdi_writeback *last_wb = NULL; |
b33e18f61
|
808 809 |
struct bdi_writeback *wb = list_entry(&bdi->wb_list, struct bdi_writeback, bdi_node); |
db1253604
|
810 811 |
might_sleep(); |
db1253604
|
812 813 |
restart: rcu_read_lock(); |
b817525a4
|
814 |
list_for_each_entry_continue_rcu(wb, &bdi->wb_list, bdi_node) { |
5b9cce4c7
|
815 |
DEFINE_WB_COMPLETION(fallback_work_done, bdi); |
8a1270cda
|
816 817 818 |
struct wb_writeback_work fallback_work; struct wb_writeback_work *work; long nr_pages; |
b817525a4
|
819 820 821 822 |
if (last_wb) { wb_put(last_wb); last_wb = NULL; } |
006a0973e
|
823 824 825 826 827 828 |
/* SYNC_ALL writes out I_DIRTY_TIME too */ if (!wb_has_dirty_io(wb) && (base_work->sync_mode == WB_SYNC_NONE || list_empty(&wb->b_dirty_time))) continue; if (skip_if_busy && writeback_in_progress(wb)) |
db1253604
|
829 |
continue; |
8a1270cda
|
830 831 832 833 834 835 836 837 838 |
nr_pages = wb_split_bdi_pages(wb, base_work->nr_pages); work = kmalloc(sizeof(*work), GFP_ATOMIC); if (work) { *work = *base_work; work->nr_pages = nr_pages; work->auto_free = 1; wb_queue_work(wb, work); continue; |
db1253604
|
839 |
} |
8a1270cda
|
840 841 842 843 844 845 846 847 848 |
/* alloc failed, execute synchronously using on-stack fallback */ work = &fallback_work; *work = *base_work; work->nr_pages = nr_pages; work->auto_free = 0; work->done = &fallback_work_done; wb_queue_work(wb, work); |
b817525a4
|
849 850 851 852 853 854 855 |
/* * Pin @wb so that it stays on @bdi->wb_list. This allows * continuing iteration from @wb after dropping and * regrabbing rcu read lock. */ wb_get(wb); last_wb = wb; |
8a1270cda
|
856 |
rcu_read_unlock(); |
5b9cce4c7
|
857 |
wb_wait_for_completion(&fallback_work_done); |
8a1270cda
|
858 |
goto restart; |
db1253604
|
859 860 |
} rcu_read_unlock(); |
b817525a4
|
861 862 863 |
if (last_wb) wb_put(last_wb); |
db1253604
|
864 |
} |
a1a0e23e4
|
865 |
/** |
d62241c7a
|
866 867 868 |
* cgroup_writeback_by_id - initiate cgroup writeback from bdi and memcg IDs * @bdi_id: target bdi id * @memcg_id: target memcg css id |
b46ec1da5
|
869 |
* @nr: number of pages to write, 0 for best-effort dirty flushing |
d62241c7a
|
870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 |
* @reason: reason why some writeback work initiated * @done: target wb_completion * * Initiate flush of the bdi_writeback identified by @bdi_id and @memcg_id * with the specified parameters. */ int cgroup_writeback_by_id(u64 bdi_id, int memcg_id, unsigned long nr, enum wb_reason reason, struct wb_completion *done) { struct backing_dev_info *bdi; struct cgroup_subsys_state *memcg_css; struct bdi_writeback *wb; struct wb_writeback_work *work; int ret; /* lookup bdi and memcg */ bdi = bdi_get_by_id(bdi_id); if (!bdi) return -ENOENT; rcu_read_lock(); memcg_css = css_from_id(memcg_id, &memory_cgrp_subsys); if (memcg_css && !css_tryget(memcg_css)) memcg_css = NULL; rcu_read_unlock(); if (!memcg_css) { ret = -ENOENT; goto out_bdi_put; } /* * And find the associated wb. If the wb isn't there already * there's nothing to flush, don't create one. */ wb = wb_get_lookup(bdi, memcg_css); if (!wb) { ret = -ENOENT; goto out_css_put; } /* * If @nr is zero, the caller is attempting to write out most of * the currently dirty pages. Let's take the current dirty page * count and inflate it by 25% which should be large enough to * flush out most dirty pages while avoiding getting livelocked by * concurrent dirtiers. */ if (!nr) { unsigned long filepages, headroom, dirty, writeback; mem_cgroup_wb_stats(wb, &filepages, &headroom, &dirty, &writeback); nr = dirty * 10 / 8; } /* issue the writeback work */ work = kzalloc(sizeof(*work), GFP_NOWAIT | __GFP_NOWARN); if (work) { work->nr_pages = nr; work->sync_mode = WB_SYNC_NONE; work->range_cyclic = 1; work->reason = reason; work->done = done; work->auto_free = 1; wb_queue_work(wb, work); ret = 0; } else { ret = -ENOMEM; } wb_put(wb); out_css_put: css_put(memcg_css); out_bdi_put: bdi_put(bdi); return ret; } /** |
a1a0e23e4
|
949 950 951 952 953 954 955 956 957 958 959 960 |
* cgroup_writeback_umount - flush inode wb switches for umount * * This function is called when a super_block is about to be destroyed and * flushes in-flight inode wb switches. An inode wb switch goes through * RCU and then workqueue, so the two need to be flushed in order to ensure * that all previously scheduled switches are finished. As wb switches are * rare occurrences and synchronize_rcu() can take a while, perform * flushing iff wb switches are in flight. */ void cgroup_writeback_umount(void) { if (atomic_read(&isw_nr_in_flight)) { |
ec084de92
|
961 962 963 964 965 |
/* * Use rcu_barrier() to wait for all pending callbacks to * ensure that all in-flight wb switches are in the workqueue. */ rcu_barrier(); |
a1a0e23e4
|
966 967 968 969 970 971 972 973 974 975 976 977 |
flush_workqueue(isw_wq); } } static int __init cgroup_writeback_init(void) { isw_wq = alloc_workqueue("inode_switch_wbs", 0, 0); if (!isw_wq) return -ENOMEM; return 0; } fs_initcall(cgroup_writeback_init); |
f2b651216
|
978 |
#else /* CONFIG_CGROUP_WRITEBACK */ |
7fc5854f8
|
979 980 |
static void bdi_down_write_wb_switch_rwsem(struct backing_dev_info *bdi) { } static void bdi_up_write_wb_switch_rwsem(struct backing_dev_info *bdi) { } |
87e1d789b
|
981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 |
static struct bdi_writeback * locked_inode_to_wb_and_lock_list(struct inode *inode) __releases(&inode->i_lock) __acquires(&wb->list_lock) { struct bdi_writeback *wb = inode_to_wb(inode); spin_unlock(&inode->i_lock); spin_lock(&wb->list_lock); return wb; } static struct bdi_writeback *inode_to_wb_and_lock_list(struct inode *inode) __acquires(&wb->list_lock) { struct bdi_writeback *wb = inode_to_wb(inode); spin_lock(&wb->list_lock); return wb; } |
f2b651216
|
1001 1002 1003 1004 |
static long wb_split_bdi_pages(struct bdi_writeback *wb, long nr_pages) { return nr_pages; } |
db1253604
|
1005 1006 1007 1008 1009 |
static void bdi_split_work_to_wbs(struct backing_dev_info *bdi, struct wb_writeback_work *base_work, bool skip_if_busy) { might_sleep(); |
006a0973e
|
1010 |
if (!skip_if_busy || !writeback_in_progress(&bdi->wb)) { |
db1253604
|
1011 |
base_work->auto_free = 0; |
db1253604
|
1012 1013 1014 |
wb_queue_work(&bdi->wb, base_work); } } |
703c27088
|
1015 |
#endif /* CONFIG_CGROUP_WRITEBACK */ |
e8e8a0c6c
|
1016 1017 1018 1019 1020 1021 1022 |
/* * Add in the number of potentially dirty inodes, because each inode * write can dirty pagecache in the underlying blockdev. */ static unsigned long get_nr_dirty_pages(void) { return global_node_page_state(NR_FILE_DIRTY) + |
e8e8a0c6c
|
1023 1024 1025 1026 |
get_nr_dirty_inodes(); } static void wb_start_writeback(struct bdi_writeback *wb, enum wb_reason reason) |
b6e51316d
|
1027 |
{ |
c00ddad39
|
1028 1029 1030 1031 |
if (!wb_has_dirty_io(wb)) return; /* |
aac8d41cd
|
1032 1033 1034 1035 1036 |
* All callers of this function want to start writeback of all * dirty pages. Places like vmscan can call this at a very * high frequency, causing pointless allocations of tons of * work items and keeping the flusher threads busy retrieving * that work. Ensure that we only allow one of them pending and |
85009b4f5
|
1037 |
* inflight at the time. |
aac8d41cd
|
1038 |
*/ |
85009b4f5
|
1039 1040 |
if (test_bit(WB_start_all, &wb->state) || test_and_set_bit(WB_start_all, &wb->state)) |
aac8d41cd
|
1041 |
return; |
85009b4f5
|
1042 1043 |
wb->start_all_reason = reason; wb_wakeup(wb); |
c5444198c
|
1044 |
} |
d3ddec763
|
1045 |
|
c5444198c
|
1046 |
/** |
9ecf4866c
|
1047 1048 |
* wb_start_background_writeback - start background writeback * @wb: bdi_writback to write from |
c5444198c
|
1049 1050 |
* * Description: |
6585027a5
|
1051 |
* This makes sure WB_SYNC_NONE background writeback happens. When |
9ecf4866c
|
1052 |
* this function returns, it is only guaranteed that for given wb |
6585027a5
|
1053 1054 |
* some IO is happening if we are over background dirty threshold. * Caller need not hold sb s_umount semaphore. |
c5444198c
|
1055 |
*/ |
9ecf4866c
|
1056 |
void wb_start_background_writeback(struct bdi_writeback *wb) |
c5444198c
|
1057 |
{ |
6585027a5
|
1058 1059 1060 1061 |
/* * We just wake up the flusher thread. It will perform background * writeback as soon as there is no other work to do. */ |
5634cc2aa
|
1062 |
trace_writeback_wake_background(wb); |
9ecf4866c
|
1063 |
wb_wakeup(wb); |
1da177e4c
|
1064 1065 1066 |
} /* |
a66979aba
|
1067 1068 |
* Remove the inode from the writeback list it is on. */ |
c7f540849
|
1069 |
void inode_io_list_del(struct inode *inode) |
a66979aba
|
1070 |
{ |
87e1d789b
|
1071 |
struct bdi_writeback *wb; |
f758eeabe
|
1072 |
|
87e1d789b
|
1073 |
wb = inode_to_wb_and_lock_list(inode); |
b35250c08
|
1074 |
spin_lock(&inode->i_lock); |
c7f540849
|
1075 |
inode_io_list_del_locked(inode, wb); |
b35250c08
|
1076 |
spin_unlock(&inode->i_lock); |
52ebea749
|
1077 |
spin_unlock(&wb->list_lock); |
a66979aba
|
1078 |
} |
4301efa4c
|
1079 |
EXPORT_SYMBOL(inode_io_list_del); |
a66979aba
|
1080 |
|
a66979aba
|
1081 |
/* |
6c60d2b57
|
1082 1083 1084 1085 1086 1087 1088 1089 1090 |
* mark an inode as under writeback on the sb */ void sb_mark_inode_writeback(struct inode *inode) { struct super_block *sb = inode->i_sb; unsigned long flags; if (list_empty(&inode->i_wb_list)) { spin_lock_irqsave(&sb->s_inode_wblist_lock, flags); |
9a46b04f1
|
1091 |
if (list_empty(&inode->i_wb_list)) { |
6c60d2b57
|
1092 |
list_add_tail(&inode->i_wb_list, &sb->s_inodes_wb); |
9a46b04f1
|
1093 1094 |
trace_sb_mark_inode_writeback(inode); } |
6c60d2b57
|
1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 |
spin_unlock_irqrestore(&sb->s_inode_wblist_lock, flags); } } /* * clear an inode as under writeback on the sb */ void sb_clear_inode_writeback(struct inode *inode) { struct super_block *sb = inode->i_sb; unsigned long flags; if (!list_empty(&inode->i_wb_list)) { spin_lock_irqsave(&sb->s_inode_wblist_lock, flags); |
9a46b04f1
|
1109 1110 1111 1112 |
if (!list_empty(&inode->i_wb_list)) { list_del_init(&inode->i_wb_list); trace_sb_clear_inode_writeback(inode); } |
6c60d2b57
|
1113 1114 1115 1116 1117 |
spin_unlock_irqrestore(&sb->s_inode_wblist_lock, flags); } } /* |
6610a0bc8
|
1118 1119 1120 1121 |
* Redirty an inode: set its when-it-was dirtied timestamp and move it to the * furthest end of its superblock's dirty-inode list. * * Before stamping the inode's ->dirtied_when, we check to see whether it is |
66f3b8e2e
|
1122 |
* already the most-recently-dirtied inode on the b_dirty list. If that is |
6610a0bc8
|
1123 1124 1125 |
* the case then the inode must have been redirtied while it was being written * out and we don't reset its dirtied_when. */ |
b35250c08
|
1126 |
static void redirty_tail_locked(struct inode *inode, struct bdi_writeback *wb) |
6610a0bc8
|
1127 |
{ |
b35250c08
|
1128 |
assert_spin_locked(&inode->i_lock); |
03ba3782e
|
1129 |
if (!list_empty(&wb->b_dirty)) { |
66f3b8e2e
|
1130 |
struct inode *tail; |
6610a0bc8
|
1131 |
|
7ccf19a80
|
1132 |
tail = wb_inode(wb->b_dirty.next); |
66f3b8e2e
|
1133 |
if (time_before(inode->dirtied_when, tail->dirtied_when)) |
6610a0bc8
|
1134 1135 |
inode->dirtied_when = jiffies; } |
c7f540849
|
1136 |
inode_io_list_move_locked(inode, wb, &wb->b_dirty); |
5afced3bf
|
1137 |
inode->i_state &= ~I_SYNC_QUEUED; |
6610a0bc8
|
1138 |
} |
b35250c08
|
1139 1140 1141 1142 1143 1144 |
static void redirty_tail(struct inode *inode, struct bdi_writeback *wb) { spin_lock(&inode->i_lock); redirty_tail_locked(inode, wb); spin_unlock(&inode->i_lock); } |
6610a0bc8
|
1145 |
/* |
66f3b8e2e
|
1146 |
* requeue inode for re-scanning after bdi->b_io list is exhausted. |
c986d1e2a
|
1147 |
*/ |
f758eeabe
|
1148 |
static void requeue_io(struct inode *inode, struct bdi_writeback *wb) |
c986d1e2a
|
1149 |
{ |
c7f540849
|
1150 |
inode_io_list_move_locked(inode, wb, &wb->b_more_io); |
c986d1e2a
|
1151 |
} |
1c0eeaf56
|
1152 1153 |
static void inode_sync_complete(struct inode *inode) { |
365b94ae6
|
1154 |
inode->i_state &= ~I_SYNC; |
4eff96dd5
|
1155 1156 |
/* If inode is clean an unused, put it into LRU now... */ inode_add_lru(inode); |
365b94ae6
|
1157 |
/* Waiters must see I_SYNC cleared before being woken up */ |
1c0eeaf56
|
1158 1159 1160 |
smp_mb(); wake_up_bit(&inode->i_state, __I_SYNC); } |
d2caa3c54
|
1161 1162 1163 1164 1165 1166 1167 1168 |
static bool inode_dirtied_after(struct inode *inode, unsigned long t) { bool ret = time_after(inode->dirtied_when, t); #ifndef CONFIG_64BIT /* * For inodes being constantly redirtied, dirtied_when can get stuck. * It _appears_ to be in the future, but is actually in distant past. * This test is necessary to prevent such wrapped-around relative times |
5b0830cb9
|
1169 |
* from permanently stopping the whole bdi writeback. |
d2caa3c54
|
1170 1171 1172 1173 1174 |
*/ ret = ret && time_before_eq(inode->dirtied_when, jiffies); #endif return ret; } |
0ae45f63d
|
1175 |
#define EXPIRE_DIRTY_ATIME 0x0001 |
c986d1e2a
|
1176 |
/* |
f9cae926f
|
1177 |
* Move expired (dirtied before dirtied_before) dirty inodes from |
697e6fed9
|
1178 |
* @delaying_queue to @dispatch_queue. |
2c1365791
|
1179 |
*/ |
e84d0a4f8
|
1180 |
static int move_expired_inodes(struct list_head *delaying_queue, |
2c1365791
|
1181 |
struct list_head *dispatch_queue, |
5fcd57505
|
1182 |
unsigned long dirtied_before) |
2c1365791
|
1183 |
{ |
5c03449d3
|
1184 1185 |
LIST_HEAD(tmp); struct list_head *pos, *node; |
cf137307c
|
1186 |
struct super_block *sb = NULL; |
5c03449d3
|
1187 |
struct inode *inode; |
cf137307c
|
1188 |
int do_sb_sort = 0; |
e84d0a4f8
|
1189 |
int moved = 0; |
5c03449d3
|
1190 |
|
2c1365791
|
1191 |
while (!list_empty(delaying_queue)) { |
7ccf19a80
|
1192 |
inode = wb_inode(delaying_queue->prev); |
f9cae926f
|
1193 |
if (inode_dirtied_after(inode, dirtied_before)) |
2c1365791
|
1194 |
break; |
c7f540849
|
1195 |
list_move(&inode->i_io_list, &tmp); |
a8855990e
|
1196 |
moved++; |
5afced3bf
|
1197 |
spin_lock(&inode->i_lock); |
5afced3bf
|
1198 1199 |
inode->i_state |= I_SYNC_QUEUED; spin_unlock(&inode->i_lock); |
a8855990e
|
1200 1201 |
if (sb_is_blkdev_sb(inode->i_sb)) continue; |
cf137307c
|
1202 1203 1204 |
if (sb && sb != inode->i_sb) do_sb_sort = 1; sb = inode->i_sb; |
5c03449d3
|
1205 |
} |
cf137307c
|
1206 1207 1208 |
/* just one sb in list, splice to dispatch_queue and we're done */ if (!do_sb_sort) { list_splice(&tmp, dispatch_queue); |
e84d0a4f8
|
1209 |
goto out; |
cf137307c
|
1210 |
} |
5c03449d3
|
1211 1212 |
/* Move inodes from one superblock together */ while (!list_empty(&tmp)) { |
7ccf19a80
|
1213 |
sb = wb_inode(tmp.prev)->i_sb; |
5c03449d3
|
1214 |
list_for_each_prev_safe(pos, node, &tmp) { |
7ccf19a80
|
1215 |
inode = wb_inode(pos); |
5c03449d3
|
1216 |
if (inode->i_sb == sb) |
c7f540849
|
1217 |
list_move(&inode->i_io_list, dispatch_queue); |
5c03449d3
|
1218 |
} |
2c1365791
|
1219 |
} |
e84d0a4f8
|
1220 1221 |
out: return moved; |
2c1365791
|
1222 1223 1224 1225 |
} /* * Queue all expired dirty inodes for io, eldest first. |
4ea879b96
|
1226 1227 1228 1229 1230 1231 1232 1233 |
* Before * newly dirtied b_dirty b_io b_more_io * =============> gf edc BA * After * newly dirtied b_dirty b_io b_more_io * =============> g fBAedc * | * +--> dequeue for IO |
2c1365791
|
1234 |
*/ |
f9cae926f
|
1235 1236 |
static void queue_io(struct bdi_writeback *wb, struct wb_writeback_work *work, unsigned long dirtied_before) |
66f3b8e2e
|
1237 |
{ |
e84d0a4f8
|
1238 |
int moved; |
f9cae926f
|
1239 |
unsigned long time_expire_jif = dirtied_before; |
0ae45f63d
|
1240 |
|
f758eeabe
|
1241 |
assert_spin_locked(&wb->list_lock); |
4ea879b96
|
1242 |
list_splice_init(&wb->b_more_io, &wb->b_io); |
5fcd57505
|
1243 |
moved = move_expired_inodes(&wb->b_dirty, &wb->b_io, dirtied_before); |
f9cae926f
|
1244 1245 |
if (!work->for_sync) time_expire_jif = jiffies - dirtytime_expire_interval * HZ; |
0ae45f63d
|
1246 |
moved += move_expired_inodes(&wb->b_dirty_time, &wb->b_io, |
5fcd57505
|
1247 |
time_expire_jif); |
d6c10f1fc
|
1248 1249 |
if (moved) wb_io_lists_populated(wb); |
f9cae926f
|
1250 |
trace_writeback_queue_io(wb, work, dirtied_before, moved); |
66f3b8e2e
|
1251 |
} |
a9185b41a
|
1252 |
static int write_inode(struct inode *inode, struct writeback_control *wbc) |
08d8e9749
|
1253 |
{ |
9fb0a7da0
|
1254 1255 1256 1257 1258 1259 1260 1261 |
int ret; if (inode->i_sb->s_op->write_inode && !is_bad_inode(inode)) { trace_writeback_write_inode_start(inode, wbc); ret = inode->i_sb->s_op->write_inode(inode, wbc); trace_writeback_write_inode(inode, wbc); return ret; } |
03ba3782e
|
1262 |
return 0; |
08d8e9749
|
1263 |
} |
08d8e9749
|
1264 |
|
2c1365791
|
1265 |
/* |
169ebd901
|
1266 1267 |
* Wait for writeback on an inode to complete. Called with i_lock held. * Caller must make sure inode cannot go away when we drop i_lock. |
01c031945
|
1268 |
*/ |
169ebd901
|
1269 1270 1271 |
static void __inode_wait_for_writeback(struct inode *inode) __releases(inode->i_lock) __acquires(inode->i_lock) |
01c031945
|
1272 1273 1274 1275 1276 |
{ DEFINE_WAIT_BIT(wq, &inode->i_state, __I_SYNC); wait_queue_head_t *wqh; wqh = bit_waitqueue(&inode->i_state, __I_SYNC); |
250df6ed2
|
1277 1278 |
while (inode->i_state & I_SYNC) { spin_unlock(&inode->i_lock); |
743162013
|
1279 1280 |
__wait_on_bit(wqh, &wq, bit_wait, TASK_UNINTERRUPTIBLE); |
250df6ed2
|
1281 |
spin_lock(&inode->i_lock); |
58a9d3d8d
|
1282 |
} |
01c031945
|
1283 1284 1285 |
} /* |
169ebd901
|
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 |
* Wait for writeback on an inode to complete. Caller must have inode pinned. */ void inode_wait_for_writeback(struct inode *inode) { spin_lock(&inode->i_lock); __inode_wait_for_writeback(inode); spin_unlock(&inode->i_lock); } /* * Sleep until I_SYNC is cleared. This function must be called with i_lock * held and drops it. It is aimed for callers not holding any inode reference * so once i_lock is dropped, inode can go away. */ static void inode_sleep_on_writeback(struct inode *inode) __releases(inode->i_lock) { DEFINE_WAIT(wait); wait_queue_head_t *wqh = bit_waitqueue(&inode->i_state, __I_SYNC); int sleep; prepare_to_wait(wqh, &wait, TASK_UNINTERRUPTIBLE); sleep = inode->i_state & I_SYNC; spin_unlock(&inode->i_lock); if (sleep) schedule(); finish_wait(wqh, &wait); } /* |
ccb26b5a6
|
1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 |
* Find proper writeback list for the inode depending on its current state and * possibly also change of its state while we were doing writeback. Here we * handle things such as livelock prevention or fairness of writeback among * inodes. This function can be called only by flusher thread - noone else * processes all inodes in writeback lists and requeueing inodes behind flusher * thread's back can have unexpected consequences. */ static void requeue_inode(struct inode *inode, struct bdi_writeback *wb, struct writeback_control *wbc) { if (inode->i_state & I_FREEING) return; /* * Sync livelock prevention. Each inode is tagged and synced in one * shot. If still dirty, it will be redirty_tail()'ed below. Update * the dirty time to prevent enqueue and sync it again. */ if ((inode->i_state & I_DIRTY) && (wbc->sync_mode == WB_SYNC_ALL || wbc->tagged_writepages)) inode->dirtied_when = jiffies; |
4f8ad655d
|
1337 1338 1339 1340 1341 |
if (wbc->pages_skipped) { /* * writeback is not making progress due to locked * buffers. Skip this inode for now. */ |
b35250c08
|
1342 |
redirty_tail_locked(inode, wb); |
4f8ad655d
|
1343 1344 |
return; } |
ccb26b5a6
|
1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 |
if (mapping_tagged(inode->i_mapping, PAGECACHE_TAG_DIRTY)) { /* * We didn't write back all the pages. nfs_writepages() * sometimes bales out without doing anything. */ if (wbc->nr_to_write <= 0) { /* Slice used up. Queue for next turn. */ requeue_io(inode, wb); } else { /* * Writeback blocked by something other than * congestion. Delay the inode for some time to * avoid spinning on the CPU (100% iowait) * retrying writeback of the dirty page/inode * that cannot be performed immediately. */ |
b35250c08
|
1361 |
redirty_tail_locked(inode, wb); |
ccb26b5a6
|
1362 1363 1364 1365 1366 1367 1368 |
} } else if (inode->i_state & I_DIRTY) { /* * Filesystems can dirty the inode during writeback operations, * such as delayed allocation during submission or metadata * updates after data IO completion. */ |
b35250c08
|
1369 |
redirty_tail_locked(inode, wb); |
0ae45f63d
|
1370 |
} else if (inode->i_state & I_DIRTY_TIME) { |
a2f487069
|
1371 |
inode->dirtied_when = jiffies; |
c7f540849
|
1372 |
inode_io_list_move_locked(inode, wb, &wb->b_dirty_time); |
5afced3bf
|
1373 |
inode->i_state &= ~I_SYNC_QUEUED; |
ccb26b5a6
|
1374 1375 |
} else { /* The inode is clean. Remove from writeback lists. */ |
c7f540849
|
1376 |
inode_io_list_del_locked(inode, wb); |
ccb26b5a6
|
1377 1378 1379 1380 |
} } /* |
4f8ad655d
|
1381 1382 1383 |
* Write out an inode and its dirty pages. Do not update the writeback list * linkage. That is left to the caller. The caller is also responsible for * setting I_SYNC flag and calling inode_sync_complete() to clear it. |
1da177e4c
|
1384 1385 |
*/ static int |
cd8ed2a45
|
1386 |
__writeback_single_inode(struct inode *inode, struct writeback_control *wbc) |
1da177e4c
|
1387 |
{ |
1da177e4c
|
1388 |
struct address_space *mapping = inode->i_mapping; |
251d6a471
|
1389 |
long nr_to_write = wbc->nr_to_write; |
01c031945
|
1390 |
unsigned dirty; |
1da177e4c
|
1391 |
int ret; |
4f8ad655d
|
1392 |
WARN_ON(!(inode->i_state & I_SYNC)); |
1da177e4c
|
1393 |
|
9fb0a7da0
|
1394 |
trace_writeback_single_inode_start(inode, wbc, nr_to_write); |
1da177e4c
|
1395 |
ret = do_writepages(mapping, wbc); |
26821ed40
|
1396 1397 1398 |
/* * Make sure to wait on the data before writing out the metadata. * This is important for filesystems that modify metadata on data |
7747bd4bc
|
1399 1400 1401 |
* I/O completion. We don't do it for sync(2) writeback because it has a * separate, external IO completion path and ->sync_fs for guaranteeing * inode metadata is written back correctly. |
26821ed40
|
1402 |
*/ |
7747bd4bc
|
1403 |
if (wbc->sync_mode == WB_SYNC_ALL && !wbc->for_sync) { |
26821ed40
|
1404 |
int err = filemap_fdatawait(mapping); |
1da177e4c
|
1405 1406 1407 |
if (ret == 0) ret = err; } |
5547e8aac
|
1408 1409 1410 1411 1412 |
/* * Some filesystems may redirty the inode during the writeback * due to delalloc, clear dirty metadata flags right before * write_inode() */ |
250df6ed2
|
1413 |
spin_lock(&inode->i_lock); |
9c6ac78eb
|
1414 |
|
5547e8aac
|
1415 |
dirty = inode->i_state & I_DIRTY; |
5fcd57505
|
1416 1417 1418 1419 1420 1421 1422 1423 |
if ((inode->i_state & I_DIRTY_TIME) && ((dirty & I_DIRTY_INODE) || wbc->sync_mode == WB_SYNC_ALL || wbc->for_sync || time_after(jiffies, inode->dirtied_time_when + dirtytime_expire_interval * HZ))) { dirty |= I_DIRTY_TIME; trace_writeback_lazytime(inode); } |
0ae45f63d
|
1424 |
inode->i_state &= ~dirty; |
9c6ac78eb
|
1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 |
/* * Paired with smp_mb() in __mark_inode_dirty(). This allows * __mark_inode_dirty() to test i_state without grabbing i_lock - * either they see the I_DIRTY bits cleared or we see the dirtied * inode. * * I_DIRTY_PAGES is always cleared together above even if @mapping * still has dirty pages. The flag is reinstated after smp_mb() if * necessary. This guarantees that either __mark_inode_dirty() * sees clear I_DIRTY_PAGES or we see PAGECACHE_TAG_DIRTY. */ smp_mb(); if (mapping_tagged(mapping, PAGECACHE_TAG_DIRTY)) inode->i_state |= I_DIRTY_PAGES; |
250df6ed2
|
1441 |
spin_unlock(&inode->i_lock); |
9c6ac78eb
|
1442 |
|
0ae45f63d
|
1443 1444 |
if (dirty & I_DIRTY_TIME) mark_inode_dirty_sync(inode); |
26821ed40
|
1445 |
/* Don't write the inode if only I_DIRTY_PAGES was set */ |
0ae45f63d
|
1446 |
if (dirty & ~I_DIRTY_PAGES) { |
a9185b41a
|
1447 |
int err = write_inode(inode, wbc); |
1da177e4c
|
1448 1449 1450 |
if (ret == 0) ret = err; } |
4f8ad655d
|
1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 |
trace_writeback_single_inode(inode, wbc, nr_to_write); return ret; } /* * Write out an inode's dirty pages. Either the caller has an active reference * on the inode or the inode has I_WILL_FREE set. * * This function is designed to be called for writing back one inode which * we go e.g. from filesystem. Flusher thread uses __writeback_single_inode() * and does more profound writeback list handling in writeback_sb_inodes(). */ |
aaf255933
|
1463 1464 |
static int writeback_single_inode(struct inode *inode, struct writeback_control *wbc) |
4f8ad655d
|
1465 |
{ |
aaf255933
|
1466 |
struct bdi_writeback *wb; |
4f8ad655d
|
1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 |
int ret = 0; spin_lock(&inode->i_lock); if (!atomic_read(&inode->i_count)) WARN_ON(!(inode->i_state & (I_WILL_FREE|I_FREEING))); else WARN_ON(inode->i_state & I_WILL_FREE); if (inode->i_state & I_SYNC) { if (wbc->sync_mode != WB_SYNC_ALL) goto out; /* |
169ebd901
|
1479 1480 1481 |
* It's a data-integrity sync. We must wait. Since callers hold * inode reference or inode has I_WILL_FREE set, it cannot go * away under us. |
4f8ad655d
|
1482 |
*/ |
169ebd901
|
1483 |
__inode_wait_for_writeback(inode); |
4f8ad655d
|
1484 1485 1486 |
} WARN_ON(inode->i_state & I_SYNC); /* |
f9b0e058c
|
1487 1488 1489 1490 1491 1492 |
* Skip inode if it is clean and we have no outstanding writeback in * WB_SYNC_ALL mode. We don't want to mess with writeback lists in this * function since flusher thread may be doing for example sync in * parallel and if we move the inode, it could get skipped. So here we * make sure inode is on some writeback list and leave it there unless * we have completely cleaned the inode. |
4f8ad655d
|
1493 |
*/ |
0ae45f63d
|
1494 |
if (!(inode->i_state & I_DIRTY_ALL) && |
f9b0e058c
|
1495 1496 |
(wbc->sync_mode != WB_SYNC_ALL || !mapping_tagged(inode->i_mapping, PAGECACHE_TAG_WRITEBACK))) |
4f8ad655d
|
1497 1498 |
goto out; inode->i_state |= I_SYNC; |
b16b1deb5
|
1499 |
wbc_attach_and_unlock_inode(wbc, inode); |
4f8ad655d
|
1500 |
|
cd8ed2a45
|
1501 |
ret = __writeback_single_inode(inode, wbc); |
1da177e4c
|
1502 |
|
b16b1deb5
|
1503 |
wbc_detach_inode(wbc); |
aaf255933
|
1504 1505 |
wb = inode_to_wb_and_lock_list(inode); |
250df6ed2
|
1506 |
spin_lock(&inode->i_lock); |
4f8ad655d
|
1507 1508 1509 1510 |
/* * If inode is clean, remove it from writeback lists. Otherwise don't * touch it. See comment above for explanation. */ |
0ae45f63d
|
1511 |
if (!(inode->i_state & I_DIRTY_ALL)) |
c7f540849
|
1512 |
inode_io_list_del_locked(inode, wb); |
4f8ad655d
|
1513 |
spin_unlock(&wb->list_lock); |
1c0eeaf56
|
1514 |
inode_sync_complete(inode); |
4f8ad655d
|
1515 1516 |
out: spin_unlock(&inode->i_lock); |
1da177e4c
|
1517 1518 |
return ret; } |
a88a341a7
|
1519 |
static long writeback_chunk_size(struct bdi_writeback *wb, |
1a12d8bd7
|
1520 |
struct wb_writeback_work *work) |
d46db3d58
|
1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 |
{ long pages; /* * WB_SYNC_ALL mode does livelock avoidance by syncing dirty * inodes/pages in one big loop. Setting wbc.nr_to_write=LONG_MAX * here avoids calling into writeback_inodes_wb() more than once. * * The intended call sequence for WB_SYNC_ALL writeback is: * * wb_writeback() * writeback_sb_inodes() <== called only once * write_cache_pages() <== called once for each inode * (quickly) tag currently dirty pages * (maybe slowly) sync all tagged pages */ if (work->sync_mode == WB_SYNC_ALL || work->tagged_writepages) pages = LONG_MAX; |
1a12d8bd7
|
1539 |
else { |
a88a341a7
|
1540 |
pages = min(wb->avg_write_bandwidth / 2, |
dcc25ae76
|
1541 |
global_wb_domain.dirty_limit / DIRTY_SCOPE); |
1a12d8bd7
|
1542 1543 1544 1545 |
pages = min(pages, work->nr_pages); pages = round_down(pages + MIN_WRITEBACK_PAGES, MIN_WRITEBACK_PAGES); } |
d46db3d58
|
1546 1547 1548 |
return pages; } |
03ba3782e
|
1549 |
/* |
f11c9c5c2
|
1550 |
* Write a portion of b_io inodes which belong to @sb. |
edadfb10b
|
1551 |
* |
d46db3d58
|
1552 |
* Return the number of pages and/or inodes written. |
0ba13fd19
|
1553 1554 1555 1556 |
* * NOTE! This is called with wb->list_lock held, and will * unlock and relock that for each inode it ends up doing * IO for. |
f11c9c5c2
|
1557 |
*/ |
d46db3d58
|
1558 1559 1560 |
static long writeback_sb_inodes(struct super_block *sb, struct bdi_writeback *wb, struct wb_writeback_work *work) |
1da177e4c
|
1561 |
{ |
d46db3d58
|
1562 1563 1564 1565 1566 |
struct writeback_control wbc = { .sync_mode = work->sync_mode, .tagged_writepages = work->tagged_writepages, .for_kupdate = work->for_kupdate, .for_background = work->for_background, |
7747bd4bc
|
1567 |
.for_sync = work->for_sync, |
d46db3d58
|
1568 1569 1570 1571 1572 1573 1574 |
.range_cyclic = work->range_cyclic, .range_start = 0, .range_end = LLONG_MAX, }; unsigned long start_time = jiffies; long write_chunk; long wrote = 0; /* count both pages and inodes */ |
03ba3782e
|
1575 |
while (!list_empty(&wb->b_io)) { |
7ccf19a80
|
1576 |
struct inode *inode = wb_inode(wb->b_io.prev); |
aaf255933
|
1577 |
struct bdi_writeback *tmp_wb; |
edadfb10b
|
1578 1579 |
if (inode->i_sb != sb) { |
d46db3d58
|
1580 |
if (work->sb) { |
edadfb10b
|
1581 1582 1583 1584 1585 |
/* * We only want to write back data for this * superblock, move all inodes not belonging * to it back onto the dirty list. */ |
f758eeabe
|
1586 |
redirty_tail(inode, wb); |
edadfb10b
|
1587 1588 1589 1590 1591 1592 1593 1594 |
continue; } /* * The inode belongs to a different superblock. * Bounce back to the caller to unpin this and * pin the next superblock. */ |
d46db3d58
|
1595 |
break; |
edadfb10b
|
1596 |
} |
9843b76aa
|
1597 |
/* |
331cbdeed
|
1598 1599 |
* Don't bother with new inodes or inodes being freed, first * kind does not need periodic writeout yet, and for the latter |
9843b76aa
|
1600 1601 |
* kind writeout is handled by the freer. */ |
250df6ed2
|
1602 |
spin_lock(&inode->i_lock); |
9843b76aa
|
1603 |
if (inode->i_state & (I_NEW | I_FREEING | I_WILL_FREE)) { |
b35250c08
|
1604 |
redirty_tail_locked(inode, wb); |
250df6ed2
|
1605 |
spin_unlock(&inode->i_lock); |
7ef0d7377
|
1606 1607 |
continue; } |
cc1676d91
|
1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 |
if ((inode->i_state & I_SYNC) && wbc.sync_mode != WB_SYNC_ALL) { /* * If this inode is locked for writeback and we are not * doing writeback-for-data-integrity, move it to * b_more_io so that writeback can proceed with the * other inodes on s_io. * * We'll have another go at writing back this inode * when we completed a full scan of b_io. */ spin_unlock(&inode->i_lock); requeue_io(inode, wb); trace_writeback_sb_inodes_requeue(inode); continue; } |
f0d07b7ff
|
1623 |
spin_unlock(&wb->list_lock); |
4f8ad655d
|
1624 1625 1626 1627 1628 |
/* * We already requeued the inode if it had I_SYNC set and we * are doing WB_SYNC_NONE writeback. So this catches only the * WB_SYNC_ALL case. */ |
169ebd901
|
1629 1630 1631 1632 |
if (inode->i_state & I_SYNC) { /* Wait for I_SYNC. This function drops i_lock... */ inode_sleep_on_writeback(inode); /* Inode may be gone, start again */ |
ead188f9f
|
1633 |
spin_lock(&wb->list_lock); |
169ebd901
|
1634 1635 |
continue; } |
4f8ad655d
|
1636 |
inode->i_state |= I_SYNC; |
b16b1deb5
|
1637 |
wbc_attach_and_unlock_inode(&wbc, inode); |
169ebd901
|
1638 |
|
a88a341a7
|
1639 |
write_chunk = writeback_chunk_size(wb, work); |
d46db3d58
|
1640 1641 |
wbc.nr_to_write = write_chunk; wbc.pages_skipped = 0; |
250df6ed2
|
1642 |
|
169ebd901
|
1643 1644 1645 1646 |
/* * We use I_SYNC to pin the inode in memory. While it is set * evict_inode() will wait so the inode cannot be freed. */ |
cd8ed2a45
|
1647 |
__writeback_single_inode(inode, &wbc); |
250df6ed2
|
1648 |
|
b16b1deb5
|
1649 |
wbc_detach_inode(&wbc); |
d46db3d58
|
1650 1651 |
work->nr_pages -= write_chunk - wbc.nr_to_write; wrote += write_chunk - wbc.nr_to_write; |
590dca3a7
|
1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 |
if (need_resched()) { /* * We're trying to balance between building up a nice * long list of IOs to improve our merge rate, and * getting those IOs out quickly for anyone throttling * in balance_dirty_pages(). cond_resched() doesn't * unplug, so get our IOs out the door before we * give up the CPU. */ blk_flush_plug(current); cond_resched(); } |
aaf255933
|
1665 1666 1667 1668 1669 |
/* * Requeue @inode if still dirty. Be careful as @inode may * have been switched to another wb in the meantime. */ tmp_wb = inode_to_wb_and_lock_list(inode); |
4f8ad655d
|
1670 |
spin_lock(&inode->i_lock); |
0ae45f63d
|
1671 |
if (!(inode->i_state & I_DIRTY_ALL)) |
d46db3d58
|
1672 |
wrote++; |
aaf255933
|
1673 |
requeue_inode(inode, tmp_wb, &wbc); |
4f8ad655d
|
1674 |
inode_sync_complete(inode); |
0f1b1fd86
|
1675 |
spin_unlock(&inode->i_lock); |
590dca3a7
|
1676 |
|
aaf255933
|
1677 1678 1679 1680 |
if (unlikely(tmp_wb != wb)) { spin_unlock(&tmp_wb->list_lock); spin_lock(&wb->list_lock); } |
d46db3d58
|
1681 1682 1683 1684 1685 1686 1687 1688 1689 |
/* * bail out to wb_writeback() often enough to check * background threshold and other termination conditions. */ if (wrote) { if (time_is_before_jiffies(start_time + HZ / 10UL)) break; if (work->nr_pages <= 0) break; |
8bc3be275
|
1690 |
} |
1da177e4c
|
1691 |
} |
d46db3d58
|
1692 |
return wrote; |
f11c9c5c2
|
1693 |
} |
d46db3d58
|
1694 1695 |
static long __writeback_inodes_wb(struct bdi_writeback *wb, struct wb_writeback_work *work) |
f11c9c5c2
|
1696 |
{ |
d46db3d58
|
1697 1698 |
unsigned long start_time = jiffies; long wrote = 0; |
38f219776
|
1699 |
|
f11c9c5c2
|
1700 |
while (!list_empty(&wb->b_io)) { |
7ccf19a80
|
1701 |
struct inode *inode = wb_inode(wb->b_io.prev); |
f11c9c5c2
|
1702 |
struct super_block *sb = inode->i_sb; |
9ecc2738a
|
1703 |
|
eb6ef3df4
|
1704 |
if (!trylock_super(sb)) { |
0e995816f
|
1705 |
/* |
eb6ef3df4
|
1706 |
* trylock_super() may fail consistently due to |
0e995816f
|
1707 1708 1709 1710 |
* s_umount being grabbed by someone else. Don't use * requeue_io() to avoid busy retrying the inode/sb. */ redirty_tail(inode, wb); |
edadfb10b
|
1711 |
continue; |
f11c9c5c2
|
1712 |
} |
d46db3d58
|
1713 |
wrote += writeback_sb_inodes(sb, wb, work); |
eb6ef3df4
|
1714 |
up_read(&sb->s_umount); |
f11c9c5c2
|
1715 |
|
d46db3d58
|
1716 1717 1718 1719 1720 1721 1722 |
/* refer to the same tests at the end of writeback_sb_inodes */ if (wrote) { if (time_is_before_jiffies(start_time + HZ / 10UL)) break; if (work->nr_pages <= 0) break; } |
f11c9c5c2
|
1723 |
} |
66f3b8e2e
|
1724 |
/* Leave any unwritten inodes on b_io */ |
d46db3d58
|
1725 |
return wrote; |
66f3b8e2e
|
1726 |
} |
7d9f073b8
|
1727 |
static long writeback_inodes_wb(struct bdi_writeback *wb, long nr_pages, |
0e175a183
|
1728 |
enum wb_reason reason) |
edadfb10b
|
1729 |
{ |
d46db3d58
|
1730 1731 1732 1733 |
struct wb_writeback_work work = { .nr_pages = nr_pages, .sync_mode = WB_SYNC_NONE, .range_cyclic = 1, |
0e175a183
|
1734 |
.reason = reason, |
d46db3d58
|
1735 |
}; |
505a666ee
|
1736 |
struct blk_plug plug; |
edadfb10b
|
1737 |
|
505a666ee
|
1738 |
blk_start_plug(&plug); |
f758eeabe
|
1739 |
spin_lock(&wb->list_lock); |
424b351fe
|
1740 |
if (list_empty(&wb->b_io)) |
f9cae926f
|
1741 |
queue_io(wb, &work, jiffies); |
d46db3d58
|
1742 |
__writeback_inodes_wb(wb, &work); |
f758eeabe
|
1743 |
spin_unlock(&wb->list_lock); |
505a666ee
|
1744 |
blk_finish_plug(&plug); |
edadfb10b
|
1745 |
|
d46db3d58
|
1746 1747 |
return nr_pages - work.nr_pages; } |
03ba3782e
|
1748 |
|
03ba3782e
|
1749 1750 |
/* * Explicit flushing or periodic writeback of "old" data. |
66f3b8e2e
|
1751 |
* |
03ba3782e
|
1752 1753 1754 1755 |
* Define "old": the first time one of an inode's pages is dirtied, we mark the * dirtying-time in the inode's address_space. So this periodic writeback code * just walks the superblock inode list, writing back any inodes which are * older than a specific point in time. |
66f3b8e2e
|
1756 |
* |
03ba3782e
|
1757 1758 1759 |
* Try to run once per dirty_writeback_interval. But if a writeback event * takes longer than a dirty_writeback_interval interval, then leave a * one-second gap. |
66f3b8e2e
|
1760 |
* |
f9cae926f
|
1761 |
* dirtied_before takes precedence over nr_to_write. So we'll only write back |
03ba3782e
|
1762 |
* all dirty pages if they are all attached to "old" mappings. |
66f3b8e2e
|
1763 |
*/ |
c4a77a6c7
|
1764 |
static long wb_writeback(struct bdi_writeback *wb, |
83ba7b071
|
1765 |
struct wb_writeback_work *work) |
66f3b8e2e
|
1766 |
{ |
e98be2d59
|
1767 |
unsigned long wb_start = jiffies; |
d46db3d58
|
1768 |
long nr_pages = work->nr_pages; |
f9cae926f
|
1769 |
unsigned long dirtied_before = jiffies; |
a5989bdc9
|
1770 |
struct inode *inode; |
d46db3d58
|
1771 |
long progress; |
505a666ee
|
1772 |
struct blk_plug plug; |
66f3b8e2e
|
1773 |
|
505a666ee
|
1774 |
blk_start_plug(&plug); |
e8dfc3058
|
1775 |
spin_lock(&wb->list_lock); |
03ba3782e
|
1776 1777 |
for (;;) { /* |
d3ddec763
|
1778 |
* Stop writeback when nr_pages has been consumed |
03ba3782e
|
1779 |
*/ |
83ba7b071
|
1780 |
if (work->nr_pages <= 0) |
03ba3782e
|
1781 |
break; |
66f3b8e2e
|
1782 |
|
38f219776
|
1783 |
/* |
aa373cf55
|
1784 1785 1786 1787 1788 1789 |
* Background writeout and kupdate-style writeback may * run forever. Stop them if there is other work to do * so that e.g. sync can proceed. They'll be restarted * after the other works are all done. */ if ((work->for_background || work->for_kupdate) && |
f0054bb1e
|
1790 |
!list_empty(&wb->work_list)) |
aa373cf55
|
1791 1792 1793 |
break; /* |
d3ddec763
|
1794 1795 |
* For background writeout, stop when we are below the * background dirty threshold |
38f219776
|
1796 |
*/ |
aa661bbe1
|
1797 |
if (work->for_background && !wb_over_bg_thresh(wb)) |
03ba3782e
|
1798 |
break; |
38f219776
|
1799 |
|
1bc36b642
|
1800 1801 1802 1803 1804 1805 |
/* * Kupdate and background works are special and we want to * include all inodes that need writing. Livelock avoidance is * handled by these works yielding to any other work so we are * safe. */ |
ba9aa8399
|
1806 |
if (work->for_kupdate) { |
f9cae926f
|
1807 |
dirtied_before = jiffies - |
ba9aa8399
|
1808 |
msecs_to_jiffies(dirty_expire_interval * 10); |
1bc36b642
|
1809 |
} else if (work->for_background) |
f9cae926f
|
1810 |
dirtied_before = jiffies; |
028c2dd18
|
1811 |
|
5634cc2aa
|
1812 |
trace_writeback_start(wb, work); |
e8dfc3058
|
1813 |
if (list_empty(&wb->b_io)) |
f9cae926f
|
1814 |
queue_io(wb, work, dirtied_before); |
83ba7b071
|
1815 |
if (work->sb) |
d46db3d58
|
1816 |
progress = writeback_sb_inodes(work->sb, wb, work); |
edadfb10b
|
1817 |
else |
d46db3d58
|
1818 |
progress = __writeback_inodes_wb(wb, work); |
5634cc2aa
|
1819 |
trace_writeback_written(wb, work); |
028c2dd18
|
1820 |
|
e98be2d59
|
1821 |
wb_update_bandwidth(wb, wb_start); |
03ba3782e
|
1822 1823 |
/* |
e6fb6da2e
|
1824 1825 1826 1827 1828 1829 |
* Did we write something? Try for more * * Dirty inodes are moved to b_io for writeback in batches. * The completion of the current batch does not necessarily * mean the overall work is done. So we keep looping as long * as made some progress on cleaning pages or inodes. |
03ba3782e
|
1830 |
*/ |
d46db3d58
|
1831 |
if (progress) |
71fd05a88
|
1832 1833 |
continue; /* |
e6fb6da2e
|
1834 |
* No more inodes for IO, bail |
71fd05a88
|
1835 |
*/ |
b7a2441f9
|
1836 |
if (list_empty(&wb->b_more_io)) |
03ba3782e
|
1837 |
break; |
71fd05a88
|
1838 |
/* |
71fd05a88
|
1839 1840 1841 1842 |
* Nothing written. Wait for some inode to * become available for writeback. Otherwise * we'll just busyloop. */ |
bace92481
|
1843 1844 1845 1846 1847 1848 1849 |
trace_writeback_wait(wb, work); inode = wb_inode(wb->b_more_io.prev); spin_lock(&inode->i_lock); spin_unlock(&wb->list_lock); /* This function drops i_lock... */ inode_sleep_on_writeback(inode); spin_lock(&wb->list_lock); |
03ba3782e
|
1850 |
} |
e8dfc3058
|
1851 |
spin_unlock(&wb->list_lock); |
505a666ee
|
1852 |
blk_finish_plug(&plug); |
03ba3782e
|
1853 |
|
d46db3d58
|
1854 |
return nr_pages - work->nr_pages; |
03ba3782e
|
1855 1856 1857 |
} /* |
83ba7b071
|
1858 |
* Return the next wb_writeback_work struct that hasn't been processed yet. |
03ba3782e
|
1859 |
*/ |
f0054bb1e
|
1860 |
static struct wb_writeback_work *get_next_work_item(struct bdi_writeback *wb) |
03ba3782e
|
1861 |
{ |
83ba7b071
|
1862 |
struct wb_writeback_work *work = NULL; |
03ba3782e
|
1863 |
|
f0054bb1e
|
1864 1865 1866 |
spin_lock_bh(&wb->work_lock); if (!list_empty(&wb->work_list)) { work = list_entry(wb->work_list.next, |
83ba7b071
|
1867 1868 |
struct wb_writeback_work, list); list_del_init(&work->list); |
03ba3782e
|
1869 |
} |
f0054bb1e
|
1870 |
spin_unlock_bh(&wb->work_lock); |
83ba7b071
|
1871 |
return work; |
03ba3782e
|
1872 |
} |
6585027a5
|
1873 1874 |
static long wb_check_background_flush(struct bdi_writeback *wb) { |
aa661bbe1
|
1875 |
if (wb_over_bg_thresh(wb)) { |
6585027a5
|
1876 1877 1878 1879 1880 1881 |
struct wb_writeback_work work = { .nr_pages = LONG_MAX, .sync_mode = WB_SYNC_NONE, .for_background = 1, .range_cyclic = 1, |
0e175a183
|
1882 |
.reason = WB_REASON_BACKGROUND, |
6585027a5
|
1883 1884 1885 1886 1887 1888 1889 |
}; return wb_writeback(wb, &work); } return 0; } |
03ba3782e
|
1890 1891 1892 1893 |
static long wb_check_old_data_flush(struct bdi_writeback *wb) { unsigned long expired; long nr_pages; |
69b62d01e
|
1894 1895 1896 1897 1898 |
/* * When set to zero, disable periodic writeback */ if (!dirty_writeback_interval) return 0; |
03ba3782e
|
1899 1900 1901 1902 1903 1904 |
expired = wb->last_old_flush + msecs_to_jiffies(dirty_writeback_interval * 10); if (time_before(jiffies, expired)) return 0; wb->last_old_flush = jiffies; |
cdf01dd54
|
1905 |
nr_pages = get_nr_dirty_pages(); |
03ba3782e
|
1906 |
|
c4a77a6c7
|
1907 |
if (nr_pages) { |
83ba7b071
|
1908 |
struct wb_writeback_work work = { |
c4a77a6c7
|
1909 1910 1911 1912 |
.nr_pages = nr_pages, .sync_mode = WB_SYNC_NONE, .for_kupdate = 1, .range_cyclic = 1, |
0e175a183
|
1913 |
.reason = WB_REASON_PERIODIC, |
c4a77a6c7
|
1914 |
}; |
83ba7b071
|
1915 |
return wb_writeback(wb, &work); |
c4a77a6c7
|
1916 |
} |
03ba3782e
|
1917 1918 1919 |
return 0; } |
85009b4f5
|
1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 |
static long wb_check_start_all(struct bdi_writeback *wb) { long nr_pages; if (!test_bit(WB_start_all, &wb->state)) return 0; nr_pages = get_nr_dirty_pages(); if (nr_pages) { struct wb_writeback_work work = { .nr_pages = wb_split_bdi_pages(wb, nr_pages), .sync_mode = WB_SYNC_NONE, .range_cyclic = 1, .reason = wb->start_all_reason, }; nr_pages = wb_writeback(wb, &work); } clear_bit(WB_start_all, &wb->state); return nr_pages; } |
03ba3782e
|
1942 1943 1944 |
/* * Retrieve work items and do the writeback they describe */ |
25d130ba2
|
1945 |
static long wb_do_writeback(struct bdi_writeback *wb) |
03ba3782e
|
1946 |
{ |
83ba7b071
|
1947 |
struct wb_writeback_work *work; |
c4a77a6c7
|
1948 |
long wrote = 0; |
03ba3782e
|
1949 |
|
4452226ea
|
1950 |
set_bit(WB_writeback_running, &wb->state); |
f0054bb1e
|
1951 |
while ((work = get_next_work_item(wb)) != NULL) { |
5634cc2aa
|
1952 |
trace_writeback_exec(wb, work); |
83ba7b071
|
1953 |
wrote += wb_writeback(wb, work); |
4a3a485b1
|
1954 |
finish_writeback_work(wb, work); |
03ba3782e
|
1955 1956 1957 |
} /* |
85009b4f5
|
1958 1959 1960 1961 1962 |
* Check for a flush-everything request */ wrote += wb_check_start_all(wb); /* |
03ba3782e
|
1963 1964 1965 |
* Check for periodic writeback, kupdated() style */ wrote += wb_check_old_data_flush(wb); |
6585027a5
|
1966 |
wrote += wb_check_background_flush(wb); |
4452226ea
|
1967 |
clear_bit(WB_writeback_running, &wb->state); |
03ba3782e
|
1968 1969 1970 1971 1972 1973 |
return wrote; } /* * Handle writeback of dirty data for the device backed by this bdi. Also |
839a8e866
|
1974 |
* reschedules periodically and does kupdated style flushing. |
03ba3782e
|
1975 |
*/ |
f0054bb1e
|
1976 |
void wb_workfn(struct work_struct *work) |
03ba3782e
|
1977 |
{ |
839a8e866
|
1978 1979 |
struct bdi_writeback *wb = container_of(to_delayed_work(work), struct bdi_writeback, dwork); |
03ba3782e
|
1980 |
long pages_written; |
68f23b890
|
1981 |
set_worker_desc("flush-%s", bdi_dev_name(wb->bdi)); |
766f91641
|
1982 |
current->flags |= PF_SWAPWRITE; |
455b28646
|
1983 |
|
839a8e866
|
1984 |
if (likely(!current_is_workqueue_rescuer() || |
4452226ea
|
1985 |
!test_bit(WB_registered, &wb->state))) { |
6467716a3
|
1986 |
/* |
f0054bb1e
|
1987 |
* The normal path. Keep writing back @wb until its |
839a8e866
|
1988 |
* work_list is empty. Note that this path is also taken |
f0054bb1e
|
1989 |
* if @wb is shutting down even when we're running off the |
839a8e866
|
1990 |
* rescuer as work_list needs to be drained. |
6467716a3
|
1991 |
*/ |
839a8e866
|
1992 |
do { |
25d130ba2
|
1993 |
pages_written = wb_do_writeback(wb); |
839a8e866
|
1994 |
trace_writeback_pages_written(pages_written); |
f0054bb1e
|
1995 |
} while (!list_empty(&wb->work_list)); |
839a8e866
|
1996 1997 1998 1999 2000 2001 |
} else { /* * bdi_wq can't get enough workers and we're running off * the emergency worker. Don't hog it. Hopefully, 1024 is * enough for efficient IO. */ |
f0054bb1e
|
2002 |
pages_written = writeback_inodes_wb(wb, 1024, |
839a8e866
|
2003 |
WB_REASON_FORKER_THREAD); |
455b28646
|
2004 |
trace_writeback_pages_written(pages_written); |
03ba3782e
|
2005 |
} |
f0054bb1e
|
2006 |
if (!list_empty(&wb->work_list)) |
b8b784958
|
2007 |
wb_wakeup(wb); |
6ca738d60
|
2008 |
else if (wb_has_dirty_io(wb) && dirty_writeback_interval) |
f0054bb1e
|
2009 |
wb_wakeup_delayed(wb); |
455b28646
|
2010 |
|
839a8e866
|
2011 |
current->flags &= ~PF_SWAPWRITE; |
03ba3782e
|
2012 2013 2014 |
} /* |
595043e5f
|
2015 2016 2017 2018 |
* Start writeback of `nr_pages' pages on this bdi. If `nr_pages' is zero, * write back the whole world. */ static void __wakeup_flusher_threads_bdi(struct backing_dev_info *bdi, |
e8e8a0c6c
|
2019 |
enum wb_reason reason) |
595043e5f
|
2020 2021 2022 2023 2024 2025 2026 |
{ struct bdi_writeback *wb; if (!bdi_has_dirty_io(bdi)) return; list_for_each_entry_rcu(wb, &bdi->wb_list, bdi_node) |
e8e8a0c6c
|
2027 |
wb_start_writeback(wb, reason); |
595043e5f
|
2028 2029 2030 2031 2032 |
} void wakeup_flusher_threads_bdi(struct backing_dev_info *bdi, enum wb_reason reason) { |
595043e5f
|
2033 |
rcu_read_lock(); |
e8e8a0c6c
|
2034 |
__wakeup_flusher_threads_bdi(bdi, reason); |
595043e5f
|
2035 2036 2037 2038 |
rcu_read_unlock(); } /* |
9ba4b2dfa
|
2039 |
* Wakeup the flusher threads to start writeback of all currently dirty pages |
03ba3782e
|
2040 |
*/ |
9ba4b2dfa
|
2041 |
void wakeup_flusher_threads(enum wb_reason reason) |
03ba3782e
|
2042 |
{ |
b8c2f3474
|
2043 |
struct backing_dev_info *bdi; |
03ba3782e
|
2044 |
|
51350ea0d
|
2045 2046 2047 2048 2049 |
/* * If we are expecting writeback progress we must submit plugged IO. */ if (blk_needs_flush_plug(current)) blk_schedule_flush_plug(current); |
b8c2f3474
|
2050 |
rcu_read_lock(); |
595043e5f
|
2051 |
list_for_each_entry_rcu(bdi, &bdi_list, bdi_list) |
e8e8a0c6c
|
2052 |
__wakeup_flusher_threads_bdi(bdi, reason); |
cfc4ba536
|
2053 |
rcu_read_unlock(); |
1da177e4c
|
2054 |
} |
a2f487069
|
2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 |
/* * Wake up bdi's periodically to make sure dirtytime inodes gets * written back periodically. We deliberately do *not* check the * b_dirtytime list in wb_has_dirty_io(), since this would cause the * kernel to be constantly waking up once there are any dirtytime * inodes on the system. So instead we define a separate delayed work * function which gets called much more rarely. (By default, only * once every 12 hours.) * * If there is any other write activity going on in the file system, * this function won't be necessary. But if the only thing that has * happened on the file system is a dirtytime inode caused by an atime * update, we need this infrastructure below to make sure that inode * eventually gets pushed out to disk. */ static void wakeup_dirtytime_writeback(struct work_struct *w); static DECLARE_DELAYED_WORK(dirtytime_work, wakeup_dirtytime_writeback); static void wakeup_dirtytime_writeback(struct work_struct *w) { struct backing_dev_info *bdi; rcu_read_lock(); list_for_each_entry_rcu(bdi, &bdi_list, bdi_list) { |
001fe6f61
|
2079 |
struct bdi_writeback *wb; |
001fe6f61
|
2080 |
|
b817525a4
|
2081 |
list_for_each_entry_rcu(wb, &bdi->wb_list, bdi_node) |
6fdf860f1
|
2082 2083 |
if (!list_empty(&wb->b_dirty_time)) wb_wakeup(wb); |
a2f487069
|
2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 |
} rcu_read_unlock(); schedule_delayed_work(&dirtytime_work, dirtytime_expire_interval * HZ); } static int __init start_dirtytime_writeback(void) { schedule_delayed_work(&dirtytime_work, dirtytime_expire_interval * HZ); return 0; } __initcall(start_dirtytime_writeback); |
1efff914a
|
2095 |
int dirtytime_interval_handler(struct ctl_table *table, int write, |
9ca48e20e
|
2096 |
void *buffer, size_t *lenp, loff_t *ppos) |
1efff914a
|
2097 2098 2099 2100 2101 2102 2103 2104 |
{ int ret; ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos); if (ret == 0 && write) mod_delayed_work(system_wq, &dirtytime_work, 0); return ret; } |
03ba3782e
|
2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 |
static noinline void block_dump___mark_inode_dirty(struct inode *inode) { if (inode->i_ino || strcmp(inode->i_sb->s_id, "bdev")) { struct dentry *dentry; const char *name = "?"; dentry = d_find_alias(inode); if (dentry) { spin_lock(&dentry->d_lock); name = (const char *) dentry->d_name.name; } printk(KERN_DEBUG "%s(%d): dirtied inode %lu (%s) on %s ", current->comm, task_pid_nr(current), inode->i_ino, name, inode->i_sb->s_id); if (dentry) { spin_unlock(&dentry->d_lock); dput(dentry); } } } /** |
0117d4272
|
2129 2130 2131 2132 2133 2134 2135 |
* __mark_inode_dirty - internal function * * @inode: inode to mark * @flags: what kind of dirty (i.e. I_DIRTY_SYNC) * * Mark an inode as dirty. Callers should use mark_inode_dirty or * mark_inode_dirty_sync. |
1da177e4c
|
2136 |
* |
03ba3782e
|
2137 2138 2139 2140 2141 2142 2143 2144 2145 |
* Put the inode on the super block's dirty list. * * CAREFUL! We mark it dirty unconditionally, but move it onto the * dirty list only if it is hashed or if it refers to a blockdev. * If it was not hashed, it will never be added to the dirty list * even if it is later hashed, as it will have been marked dirty already. * * In short, make sure you hash any inodes _before_ you start marking * them dirty. |
1da177e4c
|
2146 |
* |
03ba3782e
|
2147 2148 2149 2150 2151 2152 |
* Note that for blockdevs, inode->dirtied_when represents the dirtying time of * the block-special inode (/dev/hda1) itself. And the ->dirtied_when field of * the kernel-internal blockdev inode represents the dirtying time of the * blockdev's pages. This is why for I_DIRTY_PAGES we always use * page->mapping->host, so the page-dirtying time is recorded in the internal * blockdev inode. |
1da177e4c
|
2153 |
*/ |
03ba3782e
|
2154 |
void __mark_inode_dirty(struct inode *inode, int flags) |
1da177e4c
|
2155 |
{ |
03ba3782e
|
2156 |
struct super_block *sb = inode->i_sb; |
0ae45f63d
|
2157 2158 2159 |
int dirtytime; trace_writeback_mark_inode_dirty(inode, flags); |
1da177e4c
|
2160 |
|
03ba3782e
|
2161 2162 2163 2164 |
/* * Don't do this for I_DIRTY_PAGES - that doesn't actually * dirty the inode itself */ |
0e11f6443
|
2165 |
if (flags & (I_DIRTY_INODE | I_DIRTY_TIME)) { |
9fb0a7da0
|
2166 |
trace_writeback_dirty_inode_start(inode, flags); |
03ba3782e
|
2167 |
if (sb->s_op->dirty_inode) |
aa3857295
|
2168 |
sb->s_op->dirty_inode(inode, flags); |
9fb0a7da0
|
2169 2170 |
trace_writeback_dirty_inode(inode, flags); |
03ba3782e
|
2171 |
} |
0ae45f63d
|
2172 2173 2174 |
if (flags & I_DIRTY_INODE) flags &= ~I_DIRTY_TIME; dirtytime = flags & I_DIRTY_TIME; |
03ba3782e
|
2175 2176 |
/* |
9c6ac78eb
|
2177 2178 |
* Paired with smp_mb() in __writeback_single_inode() for the * following lockless i_state test. See there for details. |
03ba3782e
|
2179 2180 |
*/ smp_mb(); |
0ae45f63d
|
2181 2182 |
if (((inode->i_state & flags) == flags) || (dirtytime && (inode->i_state & I_DIRTY_INODE))) |
03ba3782e
|
2183 2184 2185 2186 |
return; if (unlikely(block_dump)) block_dump___mark_inode_dirty(inode); |
250df6ed2
|
2187 |
spin_lock(&inode->i_lock); |
0ae45f63d
|
2188 2189 |
if (dirtytime && (inode->i_state & I_DIRTY_INODE)) goto out_unlock_inode; |
03ba3782e
|
2190 2191 |
if ((inode->i_state & flags) != flags) { const int was_dirty = inode->i_state & I_DIRTY; |
52ebea749
|
2192 |
inode_attach_wb(inode, NULL); |
0ae45f63d
|
2193 2194 |
if (flags & I_DIRTY_INODE) inode->i_state &= ~I_DIRTY_TIME; |
03ba3782e
|
2195 2196 2197 |
inode->i_state |= flags; /* |
5afced3bf
|
2198 2199 2200 2201 |
* If the inode is queued for writeback by flush worker, just * update its dirty state. Once the flush worker is done with * the inode it will place it on the appropriate superblock * list, based upon its state. |
03ba3782e
|
2202 |
*/ |
5afced3bf
|
2203 |
if (inode->i_state & I_SYNC_QUEUED) |
250df6ed2
|
2204 |
goto out_unlock_inode; |
03ba3782e
|
2205 2206 2207 2208 2209 2210 |
/* * Only add valid (hashed) inodes to the superblock's * dirty list. Add blockdev inodes as well. */ if (!S_ISBLK(inode->i_mode)) { |
1d3382cbf
|
2211 |
if (inode_unhashed(inode)) |
250df6ed2
|
2212 |
goto out_unlock_inode; |
03ba3782e
|
2213 |
} |
a4ffdde6e
|
2214 |
if (inode->i_state & I_FREEING) |
250df6ed2
|
2215 |
goto out_unlock_inode; |
03ba3782e
|
2216 2217 2218 2219 2220 2221 |
/* * If the inode was already on b_dirty/b_io/b_more_io, don't * reposition it (that would break b_dirty time-ordering). */ if (!was_dirty) { |
87e1d789b
|
2222 |
struct bdi_writeback *wb; |
d6c10f1fc
|
2223 |
struct list_head *dirty_list; |
a66979aba
|
2224 |
bool wakeup_bdi = false; |
253c34e9b
|
2225 |
|
87e1d789b
|
2226 |
wb = locked_inode_to_wb_and_lock_list(inode); |
253c34e9b
|
2227 |
|
f56753ac2
|
2228 |
WARN((wb->bdi->capabilities & BDI_CAP_WRITEBACK) && |
0747259d1
|
2229 |
!test_bit(WB_registered, &wb->state), |
1cd925d58
|
2230 2231 |
"bdi-%s not registered ", bdi_dev_name(wb->bdi)); |
03ba3782e
|
2232 2233 |
inode->dirtied_when = jiffies; |
a2f487069
|
2234 2235 |
if (dirtytime) inode->dirtied_time_when = jiffies; |
d6c10f1fc
|
2236 |
|
0e11f6443
|
2237 |
if (inode->i_state & I_DIRTY) |
0747259d1
|
2238 |
dirty_list = &wb->b_dirty; |
a2f487069
|
2239 |
else |
0747259d1
|
2240 |
dirty_list = &wb->b_dirty_time; |
d6c10f1fc
|
2241 |
|
c7f540849
|
2242 |
wakeup_bdi = inode_io_list_move_locked(inode, wb, |
d6c10f1fc
|
2243 |
dirty_list); |
0747259d1
|
2244 |
spin_unlock(&wb->list_lock); |
0ae45f63d
|
2245 |
trace_writeback_dirty_inode_enqueue(inode); |
a66979aba
|
2246 |
|
d6c10f1fc
|
2247 2248 2249 2250 2251 2252 |
/* * If this is the first dirty inode for this bdi, * we have to wake-up the corresponding bdi thread * to make sure background write-back happens * later. */ |
f56753ac2
|
2253 2254 |
if (wakeup_bdi && (wb->bdi->capabilities & BDI_CAP_WRITEBACK)) |
0747259d1
|
2255 |
wb_wakeup_delayed(wb); |
a66979aba
|
2256 |
return; |
1da177e4c
|
2257 |
} |
1da177e4c
|
2258 |
} |
250df6ed2
|
2259 2260 |
out_unlock_inode: spin_unlock(&inode->i_lock); |
03ba3782e
|
2261 2262 |
} EXPORT_SYMBOL(__mark_inode_dirty); |
e97fedb9e
|
2263 2264 2265 2266 2267 2268 2269 2270 2271 |
/* * The @s_sync_lock is used to serialise concurrent sync operations * to avoid lock contention problems with concurrent wait_sb_inodes() calls. * Concurrent callers will block on the s_sync_lock rather than doing contending * walks. The queueing maintains sync(2) required behaviour as all the IO that * has been issued up to the time this function is enter is guaranteed to be * completed by the time we have gained the lock and waited for all IO that is * in progress regardless of the order callers are granted the lock. */ |
b6e51316d
|
2272 |
static void wait_sb_inodes(struct super_block *sb) |
03ba3782e
|
2273 |
{ |
6c60d2b57
|
2274 |
LIST_HEAD(sync_list); |
03ba3782e
|
2275 2276 2277 2278 2279 |
/* * We need to be protected against the filesystem going from * r/o to r/w or vice versa. */ |
b6e51316d
|
2280 |
WARN_ON(!rwsem_is_locked(&sb->s_umount)); |
03ba3782e
|
2281 |
|
e97fedb9e
|
2282 |
mutex_lock(&sb->s_sync_lock); |
03ba3782e
|
2283 2284 |
/* |
6c60d2b57
|
2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 |
* Splice the writeback list onto a temporary list to avoid waiting on * inodes that have started writeback after this point. * * Use rcu_read_lock() to keep the inodes around until we have a * reference. s_inode_wblist_lock protects sb->s_inodes_wb as well as * the local list because inodes can be dropped from either by writeback * completion. */ rcu_read_lock(); spin_lock_irq(&sb->s_inode_wblist_lock); list_splice_init(&sb->s_inodes_wb, &sync_list); /* * Data integrity sync. Must wait for all pages under writeback, because * there may have been pages dirtied before our sync call, but which had * writeout started before we write it out. In which case, the inode * may not be on the dirty list, but we still have to wait for that * writeout. |
03ba3782e
|
2303 |
*/ |
6c60d2b57
|
2304 2305 2306 |
while (!list_empty(&sync_list)) { struct inode *inode = list_first_entry(&sync_list, struct inode, i_wb_list); |
250df6ed2
|
2307 |
struct address_space *mapping = inode->i_mapping; |
03ba3782e
|
2308 |
|
6c60d2b57
|
2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 |
/* * Move each inode back to the wb list before we drop the lock * to preserve consistency between i_wb_list and the mapping * writeback tag. Writeback completion is responsible to remove * the inode from either list once the writeback tag is cleared. */ list_move_tail(&inode->i_wb_list, &sb->s_inodes_wb); /* * The mapping can appear untagged while still on-list since we * do not have the mapping lock. Skip it here, wb completion * will remove it. */ if (!mapping_tagged(mapping, PAGECACHE_TAG_WRITEBACK)) continue; spin_unlock_irq(&sb->s_inode_wblist_lock); |
250df6ed2
|
2326 |
spin_lock(&inode->i_lock); |
6c60d2b57
|
2327 |
if (inode->i_state & (I_FREEING|I_WILL_FREE|I_NEW)) { |
250df6ed2
|
2328 |
spin_unlock(&inode->i_lock); |
6c60d2b57
|
2329 2330 |
spin_lock_irq(&sb->s_inode_wblist_lock); |
03ba3782e
|
2331 |
continue; |
250df6ed2
|
2332 |
} |
03ba3782e
|
2333 |
__iget(inode); |
250df6ed2
|
2334 |
spin_unlock(&inode->i_lock); |
6c60d2b57
|
2335 |
rcu_read_unlock(); |
03ba3782e
|
2336 |
|
aa750fd71
|
2337 2338 2339 2340 2341 2342 |
/* * We keep the error status of individual mapping so that * applications can catch the writeback error using fsync(2). * See filemap_fdatawait_keep_errors() for details. */ filemap_fdatawait_keep_errors(mapping); |
03ba3782e
|
2343 2344 |
cond_resched(); |
6c60d2b57
|
2345 2346 2347 2348 |
iput(inode); rcu_read_lock(); spin_lock_irq(&sb->s_inode_wblist_lock); |
03ba3782e
|
2349 |
} |
6c60d2b57
|
2350 2351 |
spin_unlock_irq(&sb->s_inode_wblist_lock); rcu_read_unlock(); |
e97fedb9e
|
2352 |
mutex_unlock(&sb->s_sync_lock); |
1da177e4c
|
2353 |
} |
f30a7d0cc
|
2354 2355 |
static void __writeback_inodes_sb_nr(struct super_block *sb, unsigned long nr, enum wb_reason reason, bool skip_if_busy) |
1da177e4c
|
2356 |
{ |
5b9cce4c7
|
2357 2358 |
struct backing_dev_info *bdi = sb->s_bdi; DEFINE_WB_COMPLETION(done, bdi); |
83ba7b071
|
2359 |
struct wb_writeback_work work = { |
6e6938b6d
|
2360 2361 2362 2363 2364 |
.sb = sb, .sync_mode = WB_SYNC_NONE, .tagged_writepages = 1, .done = &done, .nr_pages = nr, |
0e175a183
|
2365 |
.reason = reason, |
3c4d71653
|
2366 |
}; |
d8a8559cd
|
2367 |
|
e79729123
|
2368 |
if (!bdi_has_dirty_io(bdi) || bdi == &noop_backing_dev_info) |
6eedc7015
|
2369 |
return; |
cf37e9724
|
2370 |
WARN_ON(!rwsem_is_locked(&sb->s_umount)); |
f30a7d0cc
|
2371 |
|
db1253604
|
2372 |
bdi_split_work_to_wbs(sb->s_bdi, &work, skip_if_busy); |
5b9cce4c7
|
2373 |
wb_wait_for_completion(&done); |
e913fc825
|
2374 |
} |
f30a7d0cc
|
2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 |
/** * writeback_inodes_sb_nr - writeback dirty inodes from given super_block * @sb: the superblock * @nr: the number of pages to write * @reason: reason why some writeback work initiated * * Start writeback on some inodes on this super_block. No guarantees are made * on how many (if any) will be written, and this function does not wait * for IO completion of submitted IO. */ void writeback_inodes_sb_nr(struct super_block *sb, unsigned long nr, enum wb_reason reason) { __writeback_inodes_sb_nr(sb, nr, reason, false); } |
3259f8bed
|
2392 2393 2394 2395 2396 |
EXPORT_SYMBOL(writeback_inodes_sb_nr); /** * writeback_inodes_sb - writeback dirty inodes from given super_block * @sb: the superblock |
786228ab3
|
2397 |
* @reason: reason why some writeback work was initiated |
3259f8bed
|
2398 2399 2400 2401 2402 |
* * Start writeback on some inodes on this super_block. No guarantees are made * on how many (if any) will be written, and this function does not wait * for IO completion of submitted IO. */ |
0e175a183
|
2403 |
void writeback_inodes_sb(struct super_block *sb, enum wb_reason reason) |
3259f8bed
|
2404 |
{ |
0e175a183
|
2405 |
return writeback_inodes_sb_nr(sb, get_nr_dirty_pages(), reason); |
3259f8bed
|
2406 |
} |
0e3c9a228
|
2407 |
EXPORT_SYMBOL(writeback_inodes_sb); |
e913fc825
|
2408 2409 |
/** |
8264c3214
|
2410 |
* try_to_writeback_inodes_sb - try to start writeback if none underway |
17bd55d03
|
2411 |
* @sb: the superblock |
8264c3214
|
2412 |
* @reason: reason why some writeback work was initiated |
17bd55d03
|
2413 |
* |
8264c3214
|
2414 |
* Invoke __writeback_inodes_sb_nr if no writeback is currently underway. |
17bd55d03
|
2415 |
*/ |
8264c3214
|
2416 |
void try_to_writeback_inodes_sb(struct super_block *sb, enum wb_reason reason) |
17bd55d03
|
2417 |
{ |
10ee27a06
|
2418 |
if (!down_read_trylock(&sb->s_umount)) |
8264c3214
|
2419 |
return; |
10ee27a06
|
2420 |
|
8264c3214
|
2421 |
__writeback_inodes_sb_nr(sb, get_nr_dirty_pages(), reason, true); |
10ee27a06
|
2422 |
up_read(&sb->s_umount); |
3259f8bed
|
2423 |
} |
10ee27a06
|
2424 |
EXPORT_SYMBOL(try_to_writeback_inodes_sb); |
3259f8bed
|
2425 2426 |
/** |
d8a8559cd
|
2427 |
* sync_inodes_sb - sync sb inode pages |
0dc83bd30
|
2428 |
* @sb: the superblock |
d8a8559cd
|
2429 2430 |
* * This function writes and waits on any dirty inode belonging to this |
0dc83bd30
|
2431 |
* super_block. |
d8a8559cd
|
2432 |
*/ |
0dc83bd30
|
2433 |
void sync_inodes_sb(struct super_block *sb) |
d8a8559cd
|
2434 |
{ |
5b9cce4c7
|
2435 2436 |
struct backing_dev_info *bdi = sb->s_bdi; DEFINE_WB_COMPLETION(done, bdi); |
83ba7b071
|
2437 |
struct wb_writeback_work work = { |
3c4d71653
|
2438 2439 2440 2441 |
.sb = sb, .sync_mode = WB_SYNC_ALL, .nr_pages = LONG_MAX, .range_cyclic = 0, |
83ba7b071
|
2442 |
.done = &done, |
0e175a183
|
2443 |
.reason = WB_REASON_SYNC, |
7747bd4bc
|
2444 |
.for_sync = 1, |
3c4d71653
|
2445 |
}; |
006a0973e
|
2446 2447 2448 2449 2450 2451 |
/* * Can't skip on !bdi_has_dirty() because we should wait for !dirty * inodes under writeback and I_DIRTY_TIME inodes ignored by * bdi_has_dirty() need to be written out too. */ if (bdi == &noop_backing_dev_info) |
6eedc7015
|
2452 |
return; |
cf37e9724
|
2453 |
WARN_ON(!rwsem_is_locked(&sb->s_umount)); |
7fc5854f8
|
2454 2455 |
/* protect against inode wb switch, see inode_switch_wbs_work_fn() */ bdi_down_write_wb_switch_rwsem(bdi); |
db1253604
|
2456 |
bdi_split_work_to_wbs(bdi, &work, false); |
5b9cce4c7
|
2457 |
wb_wait_for_completion(&done); |
7fc5854f8
|
2458 |
bdi_up_write_wb_switch_rwsem(bdi); |
83ba7b071
|
2459 |
|
b6e51316d
|
2460 |
wait_sb_inodes(sb); |
1da177e4c
|
2461 |
} |
d8a8559cd
|
2462 |
EXPORT_SYMBOL(sync_inodes_sb); |
1da177e4c
|
2463 |
|
1da177e4c
|
2464 |
/** |
7f04c26d7
|
2465 2466 2467 2468 2469 2470 |
* write_inode_now - write an inode to disk * @inode: inode to write to disk * @sync: whether the write should be synchronous or not * * This function commits an inode to disk immediately if it is dirty. This is * primarily needed by knfsd. |
1da177e4c
|
2471 |
* |
7f04c26d7
|
2472 |
* The caller must either have a ref on the inode or must have set I_WILL_FREE. |
1da177e4c
|
2473 |
*/ |
1da177e4c
|
2474 2475 |
int write_inode_now(struct inode *inode, int sync) { |
1da177e4c
|
2476 2477 |
struct writeback_control wbc = { .nr_to_write = LONG_MAX, |
18914b188
|
2478 |
.sync_mode = sync ? WB_SYNC_ALL : WB_SYNC_NONE, |
111ebb6e6
|
2479 2480 |
.range_start = 0, .range_end = LLONG_MAX, |
1da177e4c
|
2481 |
}; |
f56753ac2
|
2482 |
if (!mapping_can_writeback(inode->i_mapping)) |
49364ce25
|
2483 |
wbc.nr_to_write = 0; |
1da177e4c
|
2484 2485 |
might_sleep(); |
aaf255933
|
2486 |
return writeback_single_inode(inode, &wbc); |
1da177e4c
|
2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 |
} EXPORT_SYMBOL(write_inode_now); /** * sync_inode - write an inode and its pages to disk. * @inode: the inode to sync * @wbc: controls the writeback mode * * sync_inode() will write an inode and its pages to disk. It will also * correctly update the inode on its superblock's dirty inode lists and will * update inode->i_state. * * The caller must have a ref on the inode. */ int sync_inode(struct inode *inode, struct writeback_control *wbc) { |
aaf255933
|
2503 |
return writeback_single_inode(inode, wbc); |
1da177e4c
|
2504 2505 |
} EXPORT_SYMBOL(sync_inode); |
c37650161
|
2506 2507 |
/** |
c691b9d98
|
2508 |
* sync_inode_metadata - write an inode to disk |
c37650161
|
2509 2510 2511 |
* @inode: the inode to sync * @wait: wait for I/O to complete. * |
c691b9d98
|
2512 |
* Write an inode to disk and adjust its dirty state after completion. |
c37650161
|
2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 |
* * Note: only writes the actual inode, no associated data or other metadata. */ int sync_inode_metadata(struct inode *inode, int wait) { struct writeback_control wbc = { .sync_mode = wait ? WB_SYNC_ALL : WB_SYNC_NONE, .nr_to_write = 0, /* metadata-only */ }; return sync_inode(inode, &wbc); } EXPORT_SYMBOL(sync_inode_metadata); |