Blame view
fs/jbd/transaction.c
65 KB
1da177e4c Linux-2.6.12-rc2 |
1 |
/* |
588626996 fix file specific... |
2 |
* linux/fs/jbd/transaction.c |
ae6ddcc5f [PATCH] ext3 and ... |
3 |
* |
1da177e4c Linux-2.6.12-rc2 |
4 5 6 7 8 9 10 11 12 |
* Written by Stephen C. Tweedie <sct@redhat.com>, 1998 * * Copyright 1998 Red Hat corp --- All Rights Reserved * * This file is part of the Linux kernel and is made available under * the terms of the GNU General Public License, version 2, or at your * option, any later version, incorporated herein by reference. * * Generic filesystem transaction handling code; part of the ext2fs |
ae6ddcc5f [PATCH] ext3 and ... |
13 |
* journaling system. |
1da177e4c Linux-2.6.12-rc2 |
14 15 16 17 18 19 20 21 22 23 24 25 |
* * This file manages transactions (compound commits managed by the * journaling code) and handles (individual atomic operations by the * filesystem). */ #include <linux/time.h> #include <linux/fs.h> #include <linux/jbd.h> #include <linux/errno.h> #include <linux/slab.h> #include <linux/timer.h> |
1da177e4c Linux-2.6.12-rc2 |
26 27 |
#include <linux/mm.h> #include <linux/highmem.h> |
f420d4dc4 jbd: improve fsyn... |
28 |
#include <linux/hrtimer.h> |
05713082a jbd: remove depen... |
29 |
#include <linux/backing-dev.h> |
1da177e4c Linux-2.6.12-rc2 |
30 |
|
d394e122b [PATCH] make fs/j... |
31 |
static void __journal_temp_unlink_buffer(struct journal_head *jh); |
1da177e4c Linux-2.6.12-rc2 |
32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 |
/* * get_transaction: obtain a new transaction_t object. * * Simply allocate and initialise a new transaction. Create it in * RUNNING state and add it to the current journal (which should not * have an existing running transaction: we only make a new transaction * once we have started to commit the old one). * * Preconditions: * The journal MUST be locked. We don't perform atomic mallocs on the * new transaction and we can't block without protecting against other * processes trying to touch the journal while it is in transition. * * Called under j_state_lock */ static transaction_t * get_transaction(journal_t *journal, transaction_t *transaction) { transaction->t_journal = journal; transaction->t_state = T_RUNNING; |
f420d4dc4 jbd: improve fsyn... |
53 |
transaction->t_start_time = ktime_get(); |
1da177e4c Linux-2.6.12-rc2 |
54 55 56 57 58 |
transaction->t_tid = journal->j_transaction_sequence++; transaction->t_expires = jiffies + journal->j_commit_interval; spin_lock_init(&transaction->t_handle_lock); /* Set up the commit timer for the new transaction. */ |
b449fc6fc JBD: round commit... |
59 60 |
journal->j_commit_timer.expires = round_jiffies_up(transaction->t_expires); |
e3df18983 [PATCH] jbd: embe... |
61 |
add_timer(&journal->j_commit_timer); |
1da177e4c Linux-2.6.12-rc2 |
62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 |
J_ASSERT(journal->j_running_transaction == NULL); journal->j_running_transaction = transaction; return transaction; } /* * Handle management. * * A handle_t is an object which represents a single atomic update to a * filesystem, and which tracks all of the modifications which form part * of that one update. */ /* * start_this_handle: Given a handle, deal with any locking or stalling * needed to make sure that there is enough journal space for the handle * to begin. Attach the handle to a transaction and set up the |
ae6ddcc5f [PATCH] ext3 and ... |
81 |
* transaction's buffer credits. |
1da177e4c Linux-2.6.12-rc2 |
82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 |
*/ static int start_this_handle(journal_t *journal, handle_t *handle) { transaction_t *transaction; int needed; int nblocks = handle->h_buffer_credits; transaction_t *new_transaction = NULL; int ret = 0; if (nblocks > journal->j_max_transaction_buffers) { printk(KERN_ERR "JBD: %s wants too many credits (%d > %d) ", current->comm, nblocks, journal->j_max_transaction_buffers); ret = -ENOSPC; goto out; } alloc_transaction: if (!journal->j_running_transaction) { |
05713082a jbd: remove depen... |
103 |
new_transaction = kzalloc(sizeof(*new_transaction), GFP_NOFS); |
1da177e4c Linux-2.6.12-rc2 |
104 |
if (!new_transaction) { |
05713082a jbd: remove depen... |
105 106 |
congestion_wait(BLK_RW_ASYNC, HZ/50); goto alloc_transaction; |
1da177e4c Linux-2.6.12-rc2 |
107 |
} |
1da177e4c Linux-2.6.12-rc2 |
108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 |
} jbd_debug(3, "New handle %p going live. ", handle); repeat: /* * We need to hold j_state_lock until t_updates has been incremented, * for proper journal barrier handling */ spin_lock(&journal->j_state_lock); repeat_locked: if (is_journal_aborted(journal) || (journal->j_errno != 0 && !(journal->j_flags & JFS_ACK_ERR))) { spin_unlock(&journal->j_state_lock); |
ae6ddcc5f [PATCH] ext3 and ... |
124 |
ret = -EROFS; |
1da177e4c Linux-2.6.12-rc2 |
125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 |
goto out; } /* Wait on the journal's transaction barrier if necessary */ if (journal->j_barrier_count) { spin_unlock(&journal->j_state_lock); wait_event(journal->j_wait_transaction_locked, journal->j_barrier_count == 0); goto repeat; } if (!journal->j_running_transaction) { if (!new_transaction) { spin_unlock(&journal->j_state_lock); goto alloc_transaction; } get_transaction(journal, new_transaction); new_transaction = NULL; } transaction = journal->j_running_transaction; /* * If the current transaction is locked down for commit, wait for the * lock to be released. */ if (transaction->t_state == T_LOCKED) { DEFINE_WAIT(wait); prepare_to_wait(&journal->j_wait_transaction_locked, &wait, TASK_UNINTERRUPTIBLE); spin_unlock(&journal->j_state_lock); schedule(); finish_wait(&journal->j_wait_transaction_locked, &wait); goto repeat; } /* * If there is not enough space left in the log to write all potential * buffers requested by this operation, we need to stall pending a log * checkpoint to free some more log space. */ spin_lock(&transaction->t_handle_lock); needed = transaction->t_outstanding_credits + nblocks; if (needed > journal->j_max_transaction_buffers) { /* * If the current transaction is already too large, then start * to commit it: we can then go back and attach this handle to * a new transaction. */ DEFINE_WAIT(wait); jbd_debug(2, "Handle %p starting new commit... ", handle); spin_unlock(&transaction->t_handle_lock); prepare_to_wait(&journal->j_wait_transaction_locked, &wait, TASK_UNINTERRUPTIBLE); __log_start_commit(journal, transaction->t_tid); spin_unlock(&journal->j_state_lock); schedule(); finish_wait(&journal->j_wait_transaction_locked, &wait); goto repeat; } |
ae6ddcc5f [PATCH] ext3 and ... |
189 |
/* |
1da177e4c Linux-2.6.12-rc2 |
190 191 192 193 194 195 196 197 |
* The commit code assumes that it can get enough log space * without forcing a checkpoint. This is *critical* for * correctness: a checkpoint of a buffer which is also * associated with a committing transaction creates a deadlock, * so commit simply cannot force through checkpoints. * * We must therefore ensure the necessary space in the journal * *before* starting to dirty potentially checkpointed buffers |
ae6ddcc5f [PATCH] ext3 and ... |
198 |
* in the new transaction. |
1da177e4c Linux-2.6.12-rc2 |
199 200 201 202 203 204 205 206 207 208 209 210 |
* * The worst part is, any transaction currently committing can * reduce the free space arbitrarily. Be careful to account for * those buffers when checkpointing. */ /* * @@@ AKPM: This seems rather over-defensive. We're giving commit * a _lot_ of headroom: 1/4 of the journal plus the size of * the committing transaction. Really, we only need to give it * committing_transaction->t_outstanding_credits plus "enough" for * the log control blocks. |
a34f0b313 fix comment typos... |
211 |
* Also, this test is inconsistent with the matching one in |
1da177e4c Linux-2.6.12-rc2 |
212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 |
* journal_extend(). */ if (__log_space_left(journal) < jbd_space_needed(journal)) { jbd_debug(2, "Handle %p waiting for checkpoint... ", handle); spin_unlock(&transaction->t_handle_lock); __log_wait_for_space(journal); goto repeat_locked; } /* OK, account for the buffers that this operation expects to * use and add the handle to the running transaction. */ handle->h_transaction = transaction; transaction->t_outstanding_credits += nblocks; transaction->t_updates++; transaction->t_handle_count++; jbd_debug(4, "Handle %p given %d credits (total %d, free %d) ", handle, nblocks, transaction->t_outstanding_credits, __log_space_left(journal)); spin_unlock(&transaction->t_handle_lock); spin_unlock(&journal->j_state_lock); |
3adae9da0 jbd: Annotate tra... |
235 236 |
lock_map_acquire(&handle->h_lockdep_map); |
1da177e4c Linux-2.6.12-rc2 |
237 |
out: |
304c4c841 [PATCH] jbd: avoi... |
238 239 |
if (unlikely(new_transaction)) /* It's usually NULL */ kfree(new_transaction); |
1da177e4c Linux-2.6.12-rc2 |
240 241 |
return ret; } |
34a3d1e83 lockdep: annotate... |
242 |
static struct lock_class_key jbd_handle_key; |
1da177e4c Linux-2.6.12-rc2 |
243 244 245 246 247 248 249 250 251 |
/* Allocate a new handle. This should probably be in a slab... */ static handle_t *new_handle(int nblocks) { handle_t *handle = jbd_alloc_handle(GFP_NOFS); if (!handle) return NULL; memset(handle, 0, sizeof(*handle)); handle->h_buffer_credits = nblocks; handle->h_ref = 1; |
34a3d1e83 lockdep: annotate... |
252 |
lockdep_init_map(&handle->h_lockdep_map, "jbd_handle", &jbd_handle_key, 0); |
1da177e4c Linux-2.6.12-rc2 |
253 254 255 256 |
return handle; } /** |
ae6ddcc5f [PATCH] ext3 and ... |
257 |
* handle_t *journal_start() - Obtain a new handle. |
1da177e4c Linux-2.6.12-rc2 |
258 259 260 261 262 |
* @journal: Journal to start transaction on. * @nblocks: number of block buffer we might modify * * We make sure that the transaction can guarantee at least nblocks of * modified buffers in the log. We block until the log can guarantee |
ae6ddcc5f [PATCH] ext3 and ... |
263 |
* that much space. |
1da177e4c Linux-2.6.12-rc2 |
264 265 266 267 |
* * This function is visible to journal users (like ext3fs), so is not * called with the journal already locked. * |
c2b67735e jbd: Fix comment ... |
268 269 |
* Return a pointer to a newly allocated handle, or an ERR_PTR() value * on failure. |
1da177e4c Linux-2.6.12-rc2 |
270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 |
*/ handle_t *journal_start(journal_t *journal, int nblocks) { handle_t *handle = journal_current_handle(); int err; if (!journal) return ERR_PTR(-EROFS); if (handle) { J_ASSERT(handle->h_transaction->t_journal == journal); handle->h_ref++; return handle; } handle = new_handle(nblocks); if (!handle) return ERR_PTR(-ENOMEM); current->journal_info = handle; err = start_this_handle(journal, handle); if (err < 0) { jbd_free_handle(handle); current->journal_info = NULL; handle = ERR_PTR(err); } return handle; } /** * int journal_extend() - extend buffer credits. * @handle: handle to 'extend' * @nblocks: nr blocks to try to extend by. |
ae6ddcc5f [PATCH] ext3 and ... |
304 |
* |
1da177e4c Linux-2.6.12-rc2 |
305 306 307 |
* Some transactions, such as large extends and truncates, can be done * atomically all at once or in several stages. The operation requests * a credit for a number of buffer modications in advance, but can |
ae6ddcc5f [PATCH] ext3 and ... |
308 |
* extend its credit if it needs more. |
1da177e4c Linux-2.6.12-rc2 |
309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 |
* * journal_extend tries to give the running handle more buffer credits. * It does not guarantee that allocation - this is a best-effort only. * The calling process MUST be able to deal cleanly with a failure to * extend here. * * Return 0 on success, non-zero on failure. * * return code < 0 implies an error * return code > 0 implies normal transaction-full status. */ int journal_extend(handle_t *handle, int nblocks) { transaction_t *transaction = handle->h_transaction; journal_t *journal = transaction->t_journal; int result; int wanted; result = -EIO; if (is_handle_aborted(handle)) goto out; result = 1; spin_lock(&journal->j_state_lock); /* Don't extend a locked-down transaction! */ if (handle->h_transaction->t_state != T_RUNNING) { jbd_debug(3, "denied handle %p %d blocks: " "transaction not running ", handle, nblocks); goto error_out; } spin_lock(&transaction->t_handle_lock); wanted = transaction->t_outstanding_credits + nblocks; if (wanted > journal->j_max_transaction_buffers) { jbd_debug(3, "denied handle %p %d blocks: " "transaction too large ", handle, nblocks); goto unlock; } if (wanted > __log_space_left(journal)) { jbd_debug(3, "denied handle %p %d blocks: " "insufficient log space ", handle, nblocks); goto unlock; } handle->h_buffer_credits += nblocks; transaction->t_outstanding_credits += nblocks; result = 0; jbd_debug(3, "extended handle %p by %d ", handle, nblocks); unlock: spin_unlock(&transaction->t_handle_lock); error_out: spin_unlock(&journal->j_state_lock); out: return result; } /** |
78a4a50a8 docbook: fix file... |
376 |
* int journal_restart() - restart a handle. |
1da177e4c Linux-2.6.12-rc2 |
377 378 |
* @handle: handle to restart * @nblocks: nr credits requested |
ae6ddcc5f [PATCH] ext3 and ... |
379 |
* |
1da177e4c Linux-2.6.12-rc2 |
380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 |
* Restart a handle for a multi-transaction filesystem * operation. * * If the journal_extend() call above fails to grant new buffer credits * to a running handle, a call to journal_restart will commit the * handle's transaction so far and reattach the handle to a new * transaction capabable of guaranteeing the requested number of * credits. */ int journal_restart(handle_t *handle, int nblocks) { transaction_t *transaction = handle->h_transaction; journal_t *journal = transaction->t_journal; int ret; /* If we've had an abort of any type, don't even think about * actually doing the restart! */ if (is_handle_aborted(handle)) return 0; /* * First unlink the handle from its current transaction, and start the * commit on that. */ J_ASSERT(transaction->t_updates > 0); J_ASSERT(journal_current_handle() == handle); spin_lock(&journal->j_state_lock); spin_lock(&transaction->t_handle_lock); transaction->t_outstanding_credits -= handle->h_buffer_credits; transaction->t_updates--; if (!transaction->t_updates) wake_up(&journal->j_wait_updates); spin_unlock(&transaction->t_handle_lock); jbd_debug(2, "restarting handle %p ", handle); __log_start_commit(journal, transaction->t_tid); spin_unlock(&journal->j_state_lock); |
3adae9da0 jbd: Annotate tra... |
421 |
lock_map_release(&handle->h_lockdep_map); |
1da177e4c Linux-2.6.12-rc2 |
422 423 424 425 426 427 428 429 430 431 |
handle->h_buffer_credits = nblocks; ret = start_this_handle(journal, handle); return ret; } /** * void journal_lock_updates () - establish a transaction barrier. * @journal: Journal to establish a barrier on. * |
004827855 jbd: Remove j_bar... |
432 433 434 |
* This locks out any further updates from being started, and blocks until all * existing updates have completed, returning only once the journal is in a * quiescent state with no updates running. |
1da177e4c Linux-2.6.12-rc2 |
435 |
* |
004827855 jbd: Remove j_bar... |
436 437 438 439 440 |
* We do not use simple mutex for synchronization as there are syscalls which * want to return with filesystem locked and that trips up lockdep. Also * hibernate needs to lock filesystem but locked mutex then blocks hibernation. * Since locking filesystem is rare operation, we use simple counter and * waitqueue for locking. |
1da177e4c Linux-2.6.12-rc2 |
441 442 443 444 |
*/ void journal_lock_updates(journal_t *journal) { DEFINE_WAIT(wait); |
004827855 jbd: Remove j_bar... |
445 446 447 448 |
wait: /* Wait for previous locked operation to finish */ wait_event(journal->j_wait_transaction_locked, journal->j_barrier_count == 0); |
1da177e4c Linux-2.6.12-rc2 |
449 |
spin_lock(&journal->j_state_lock); |
004827855 jbd: Remove j_bar... |
450 451 452 453 454 455 456 457 |
/* * Check reliably under the lock whether we are the ones winning the race * and locking the journal */ if (journal->j_barrier_count > 0) { spin_unlock(&journal->j_state_lock); goto wait; } |
1da177e4c Linux-2.6.12-rc2 |
458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 |
++journal->j_barrier_count; /* Wait until there are no running updates */ while (1) { transaction_t *transaction = journal->j_running_transaction; if (!transaction) break; spin_lock(&transaction->t_handle_lock); if (!transaction->t_updates) { spin_unlock(&transaction->t_handle_lock); break; } prepare_to_wait(&journal->j_wait_updates, &wait, TASK_UNINTERRUPTIBLE); spin_unlock(&transaction->t_handle_lock); spin_unlock(&journal->j_state_lock); schedule(); finish_wait(&journal->j_wait_updates, &wait); spin_lock(&journal->j_state_lock); } spin_unlock(&journal->j_state_lock); |
1da177e4c Linux-2.6.12-rc2 |
481 482 483 484 485 |
} /** * void journal_unlock_updates (journal_t* journal) - release barrier * @journal: Journal to release the barrier on. |
ae6ddcc5f [PATCH] ext3 and ... |
486 |
* |
1da177e4c Linux-2.6.12-rc2 |
487 |
* Release a transaction barrier obtained with journal_lock_updates(). |
1da177e4c Linux-2.6.12-rc2 |
488 489 490 491 |
*/ void journal_unlock_updates (journal_t *journal) { J_ASSERT(journal->j_barrier_count != 0); |
1da177e4c Linux-2.6.12-rc2 |
492 493 494 495 496 |
spin_lock(&journal->j_state_lock); --journal->j_barrier_count; spin_unlock(&journal->j_state_lock); wake_up(&journal->j_wait_transaction_locked); } |
1e9fd53b7 jbd: Fix a race b... |
497 |
static void warn_dirty_buffer(struct buffer_head *bh) |
1da177e4c Linux-2.6.12-rc2 |
498 |
{ |
1e9fd53b7 jbd: Fix a race b... |
499 |
char b[BDEVNAME_SIZE]; |
4407c2b6b [PATCH] Fix race ... |
500 |
|
1e9fd53b7 jbd: Fix a race b... |
501 502 503 504 505 506 |
printk(KERN_WARNING "JBD: Spotted dirty metadata buffer (dev = %s, blocknr = %llu). " "There's a risk of filesystem corruption in case of system " "crash. ", bdevname(bh->b_bdev, b), (unsigned long long)bh->b_blocknr); |
1da177e4c Linux-2.6.12-rc2 |
507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 |
} /* * If the buffer is already part of the current transaction, then there * is nothing we need to do. If it is already part of a prior * transaction which we are still committing to disk, then we need to * make sure that we do not overwrite the old copy: we do copy-out to * preserve the copy going to disk. We also account the buffer against * the handle's metadata buffer credits (unless the buffer is already * part of the transaction, that is). * */ static int do_get_write_access(handle_t *handle, struct journal_head *jh, int force_copy) { struct buffer_head *bh; transaction_t *transaction; journal_t *journal; int error; char *frozen_buffer = NULL; int need_copy = 0; if (is_handle_aborted(handle)) return -EROFS; transaction = handle->h_transaction; journal = transaction->t_journal; |
b8ea49fa9 jbd: Fix debug me... |
535 536 |
jbd_debug(5, "journal_head %p, force_copy %d ", jh, force_copy); |
1da177e4c Linux-2.6.12-rc2 |
537 538 539 540 541 542 543 544 545 546 547 |
JBUFFER_TRACE(jh, "entry"); repeat: bh = jh2bh(jh); /* @@@ Need to check for errors here at some point. */ lock_buffer(bh); jbd_lock_bh_state(bh); /* We now hold the buffer lock so it is safe to query the buffer |
ae6ddcc5f [PATCH] ext3 and ... |
548 549 |
* state. Is the buffer dirty? * |
1da177e4c Linux-2.6.12-rc2 |
550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 |
* If so, there are two possibilities. The buffer may be * non-journaled, and undergoing a quite legitimate writeback. * Otherwise, it is journaled, and we don't expect dirty buffers * in that state (the buffers should be marked JBD_Dirty * instead.) So either the IO is being done under our own * control and this is a bug, or it's a third party IO such as * dump(8) (which may leave the buffer scheduled for read --- * ie. locked but not dirty) or tune2fs (which may actually have * the buffer dirtied, ugh.) */ if (buffer_dirty(bh)) { /* * First question: is this buffer already part of the current * transaction or the existing committing transaction? */ if (jh->b_transaction) { J_ASSERT_JH(jh, |
ae6ddcc5f [PATCH] ext3 and ... |
567 |
jh->b_transaction == transaction || |
1da177e4c Linux-2.6.12-rc2 |
568 569 570 571 572 |
jh->b_transaction == journal->j_committing_transaction); if (jh->b_next_transaction) J_ASSERT_JH(jh, jh->b_next_transaction == transaction); |
1e9fd53b7 jbd: Fix a race b... |
573 |
warn_dirty_buffer(bh); |
4407c2b6b [PATCH] Fix race ... |
574 575 576 577 578 579 |
} /* * In any case we need to clean the dirty flag and we must * do it under the buffer lock to be sure we don't race * with running write-out. */ |
1e9fd53b7 jbd: Fix a race b... |
580 581 582 |
JBUFFER_TRACE(jh, "Journalling dirty buffer"); clear_buffer_dirty(bh); set_buffer_jbddirty(bh); |
e9ad5620b [PATCH] ext3: Mor... |
583 |
} |
1da177e4c Linux-2.6.12-rc2 |
584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 |
unlock_buffer(bh); error = -EROFS; if (is_handle_aborted(handle)) { jbd_unlock_bh_state(bh); goto out; } error = 0; /* * The buffer is already part of this transaction if b_transaction or * b_next_transaction points to it */ if (jh->b_transaction == transaction || jh->b_next_transaction == transaction) goto done; /* |
5bc833fea jbd: fix the way ... |
603 604 605 606 607 608 |
* this is the first time this transaction is touching this buffer, * reset the modified flag */ jh->b_modified = 0; /* |
1da177e4c Linux-2.6.12-rc2 |
609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 |
* If there is already a copy-out version of this buffer, then we don't * need to make another one */ if (jh->b_frozen_data) { JBUFFER_TRACE(jh, "has frozen data"); J_ASSERT_JH(jh, jh->b_next_transaction == NULL); jh->b_next_transaction = transaction; goto done; } /* Is there data here we need to preserve? */ if (jh->b_transaction && jh->b_transaction != transaction) { JBUFFER_TRACE(jh, "owned by older transaction"); J_ASSERT_JH(jh, jh->b_next_transaction == NULL); J_ASSERT_JH(jh, jh->b_transaction == journal->j_committing_transaction); /* There is one case we have to be very careful about. * If the committing transaction is currently writing * this buffer out to disk and has NOT made a copy-out, * then we cannot modify the buffer contents at all * right now. The essence of copy-out is that it is the * extra copy, not the primary copy, which gets * journaled. If the primary copy is already going to * disk then we cannot do copy-out here. */ if (jh->b_jlist == BJ_Shadow) { DEFINE_WAIT_BIT(wait, &bh->b_state, BH_Unshadow); wait_queue_head_t *wqh; wqh = bit_waitqueue(&bh->b_state, BH_Unshadow); JBUFFER_TRACE(jh, "on shadow: sleep"); jbd_unlock_bh_state(bh); /* commit wakes up all shadow buffers after IO */ for ( ; ; ) { prepare_to_wait(wqh, &wait.wait, TASK_UNINTERRUPTIBLE); if (jh->b_jlist != BJ_Shadow) break; schedule(); } finish_wait(wqh, &wait.wait); goto repeat; } /* Only do the copy if the currently-owning transaction * still needs it. If it is on the Forget list, the * committing transaction is past that stage. The * buffer had better remain locked during the kmalloc, * but that should be true --- we hold the journal lock * still and the buffer is already on the BUF_JOURNAL |
ae6ddcc5f [PATCH] ext3 and ... |
662 |
* list so won't be flushed. |
1da177e4c Linux-2.6.12-rc2 |
663 664 665 666 667 668 669 670 671 672 673 674 |
* * Subtle point, though: if this is a get_undo_access, * then we will be relying on the frozen_data to contain * the new value of the committed_data record after the * transaction, so we HAVE to force the frozen_data copy * in that case. */ if (jh->b_jlist != BJ_Forget || force_copy) { JBUFFER_TRACE(jh, "generate frozen data"); if (!frozen_buffer) { JBUFFER_TRACE(jh, "allocate memory for buffer"); jbd_unlock_bh_state(bh); |
ea817398e [PATCH] Manage jb... |
675 |
frozen_buffer = |
c089d490d JBD: JBD slab all... |
676 |
jbd_alloc(jh2bh(jh)->b_size, |
ea817398e [PATCH] Manage jb... |
677 |
GFP_NOFS); |
1da177e4c Linux-2.6.12-rc2 |
678 679 680 681 |
if (!frozen_buffer) { printk(KERN_EMERG "%s: OOM for frozen_buffer ", |
08fc99bfc jbd: replace rema... |
682 |
__func__); |
1da177e4c Linux-2.6.12-rc2 |
683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 |
JBUFFER_TRACE(jh, "oom!"); error = -ENOMEM; jbd_lock_bh_state(bh); goto done; } goto repeat; } jh->b_frozen_data = frozen_buffer; frozen_buffer = NULL; need_copy = 1; } jh->b_next_transaction = transaction; } /* * Finally, if the buffer is not journaled right now, we need to make * sure it doesn't get written to disk before the caller actually * commits the new data */ if (!jh->b_transaction) { JBUFFER_TRACE(jh, "no transaction"); J_ASSERT_JH(jh, !jh->b_next_transaction); |
1da177e4c Linux-2.6.12-rc2 |
706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 |
JBUFFER_TRACE(jh, "file as BJ_Reserved"); spin_lock(&journal->j_list_lock); __journal_file_buffer(jh, transaction, BJ_Reserved); spin_unlock(&journal->j_list_lock); } done: if (need_copy) { struct page *page; int offset; char *source; J_EXPECT_JH(jh, buffer_uptodate(jh2bh(jh)), "Possible IO failure. "); page = jh2bh(jh)->b_page; |
8117f98c0 jbd: Use offset_i... |
722 |
offset = offset_in_page(jh2bh(jh)->b_data); |
1da177e4c Linux-2.6.12-rc2 |
723 724 725 726 727 728 729 730 731 732 733 734 735 |
source = kmap_atomic(page, KM_USER0); memcpy(jh->b_frozen_data, source+offset, jh2bh(jh)->b_size); kunmap_atomic(source, KM_USER0); } jbd_unlock_bh_state(bh); /* * If we are about to journal a buffer, then any revoke pending on it is * no longer valid */ journal_cancel_revoke(handle, jh); out: |
304c4c841 [PATCH] jbd: avoi... |
736 |
if (unlikely(frozen_buffer)) /* It's usually NULL */ |
c089d490d JBD: JBD slab all... |
737 |
jbd_free(frozen_buffer, bh->b_size); |
1da177e4c Linux-2.6.12-rc2 |
738 739 740 741 742 743 744 745 746 |
JBUFFER_TRACE(jh, "exit"); return error; } /** * int journal_get_write_access() - notify intent to modify a buffer for metadata (not data) update. * @handle: transaction to add buffer modifications to * @bh: bh to be used for metadata writes |
1da177e4c Linux-2.6.12-rc2 |
747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 |
* * Returns an error code or 0 on success. * * In full data journalling mode the buffer may be of type BJ_AsyncData, * because we're write()ing a buffer which is also part of a shared mapping. */ int journal_get_write_access(handle_t *handle, struct buffer_head *bh) { struct journal_head *jh = journal_add_journal_head(bh); int rc; /* We do not want to get caught playing with fields which the * log thread also manipulates. Make sure that the buffer * completes any outstanding IO before proceeding. */ rc = do_get_write_access(handle, jh, 0); journal_put_journal_head(jh); return rc; } /* * When the user wants to journal a newly created buffer_head * (ie. getblk() returned a new buffer and we are going to populate it * manually rather than reading off disk), then we need to keep the * buffer_head locked until it has been completely filled with new * data. In this case, we should be able to make the assertion that |
ae6ddcc5f [PATCH] ext3 and ... |
774 775 |
* the bh is not already part of an existing transaction. * |
1da177e4c Linux-2.6.12-rc2 |
776 777 778 779 780 781 782 783 784 785 786 |
* The buffer should already be locked by the caller by this point. * There is no lock ranking violation: it was a newly created, * unlocked buffer beforehand. */ /** * int journal_get_create_access () - notify intent to use newly created bh * @handle: transaction to new buffer to * @bh: new buffer. * * Call this if you create a new bh. */ |
ae6ddcc5f [PATCH] ext3 and ... |
787 |
int journal_get_create_access(handle_t *handle, struct buffer_head *bh) |
1da177e4c Linux-2.6.12-rc2 |
788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 |
{ transaction_t *transaction = handle->h_transaction; journal_t *journal = transaction->t_journal; struct journal_head *jh = journal_add_journal_head(bh); int err; jbd_debug(5, "journal_head %p ", jh); err = -EROFS; if (is_handle_aborted(handle)) goto out; err = 0; JBUFFER_TRACE(jh, "entry"); /* * The buffer may already belong to this transaction due to pre-zeroing * in the filesystem's new_block code. It may also be on the previous, * committing transaction's lists, but it HAS to be in Forget state in * that case: the transaction must have deleted the buffer for it to be * reused here. */ jbd_lock_bh_state(bh); spin_lock(&journal->j_list_lock); J_ASSERT_JH(jh, (jh->b_transaction == transaction || jh->b_transaction == NULL || (jh->b_transaction == journal->j_committing_transaction && jh->b_jlist == BJ_Forget))); J_ASSERT_JH(jh, jh->b_next_transaction == NULL); J_ASSERT_JH(jh, buffer_locked(jh2bh(jh))); if (jh->b_transaction == NULL) { |
1e9fd53b7 jbd: Fix a race b... |
820 821 822 823 824 825 826 827 828 |
/* * Previous journal_forget() could have left the buffer * with jbddirty bit set because it was being committed. When * the commit finished, we've filed the buffer for * checkpointing and marked it dirty. Now we are reallocating * the buffer so the transaction freeing it must have * committed and so it's safe to clear the dirty bit. */ clear_buffer_dirty(jh2bh(jh)); |
5bc833fea jbd: fix the way ... |
829 830 831 |
/* first access by this transaction */ jh->b_modified = 0; |
1da177e4c Linux-2.6.12-rc2 |
832 833 834 |
JBUFFER_TRACE(jh, "file as BJ_Reserved"); __journal_file_buffer(jh, transaction, BJ_Reserved); } else if (jh->b_transaction == journal->j_committing_transaction) { |
5bc833fea jbd: fix the way ... |
835 836 |
/* first access by this transaction */ jh->b_modified = 0; |
1da177e4c Linux-2.6.12-rc2 |
837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 |
JBUFFER_TRACE(jh, "set next transaction"); jh->b_next_transaction = transaction; } spin_unlock(&journal->j_list_lock); jbd_unlock_bh_state(bh); /* * akpm: I added this. ext3_alloc_branch can pick up new indirect * blocks which contain freed but then revoked metadata. We need * to cancel the revoke in case we end up freeing it yet again * and the reallocating as data - this would cause a second revoke, * which hits an assertion error. */ JBUFFER_TRACE(jh, "cancelling revoke"); journal_cancel_revoke(handle, jh); |
1da177e4c Linux-2.6.12-rc2 |
852 |
out: |
bd5c9e185 jbd: fix a bug of... |
853 |
journal_put_journal_head(jh); |
1da177e4c Linux-2.6.12-rc2 |
854 855 856 857 |
return err; } /** |
78a4a50a8 docbook: fix file... |
858 |
* int journal_get_undo_access() - Notify intent to modify metadata with non-rewindable consequences |
1da177e4c Linux-2.6.12-rc2 |
859 860 |
* @handle: transaction * @bh: buffer to undo |
1da177e4c Linux-2.6.12-rc2 |
861 862 863 864 865 866 867 |
* * Sometimes there is a need to distinguish between metadata which has * been committed to disk and that which has not. The ext3fs code uses * this for freeing and allocating space, we have to make sure that we * do not reuse freed space until the deallocation has been committed, * since if we overwrote that space we would make the delete * un-rewindable in case of a crash. |
ae6ddcc5f [PATCH] ext3 and ... |
868 |
* |
1da177e4c Linux-2.6.12-rc2 |
869 870 871 872 873 |
* To deal with that, journal_get_undo_access requests write access to a * buffer for parts of non-rewindable operations such as delete * operations on the bitmaps. The journaling code must keep a copy of * the buffer's contents prior to the undo_access call until such time * as we know that the buffer has definitely been committed to disk. |
ae6ddcc5f [PATCH] ext3 and ... |
874 |
* |
1da177e4c Linux-2.6.12-rc2 |
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 |
* We never need to know which transaction the committed data is part * of, buffers touched here are guaranteed to be dirtied later and so * will be committed to a new transaction in due course, at which point * we can discard the old committed data pointer. * * Returns error number or 0 on success. */ int journal_get_undo_access(handle_t *handle, struct buffer_head *bh) { int err; struct journal_head *jh = journal_add_journal_head(bh); char *committed_data = NULL; JBUFFER_TRACE(jh, "entry"); /* * Do this first --- it can drop the journal lock, so we want to * make sure that obtaining the committed_data is done * atomically wrt. completion of any outstanding commits. */ err = do_get_write_access(handle, jh, 1); if (err) goto out; repeat: if (!jh->b_committed_data) { |
c089d490d JBD: JBD slab all... |
901 |
committed_data = jbd_alloc(jh2bh(jh)->b_size, GFP_NOFS); |
1da177e4c Linux-2.6.12-rc2 |
902 903 904 |
if (!committed_data) { printk(KERN_EMERG "%s: No memory for committed data ", |
08fc99bfc jbd: replace rema... |
905 |
__func__); |
1da177e4c Linux-2.6.12-rc2 |
906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 |
err = -ENOMEM; goto out; } } jbd_lock_bh_state(bh); if (!jh->b_committed_data) { /* Copy out the current buffer contents into the * preserved, committed copy. */ JBUFFER_TRACE(jh, "generate b_committed data"); if (!committed_data) { jbd_unlock_bh_state(bh); goto repeat; } jh->b_committed_data = committed_data; committed_data = NULL; memcpy(jh->b_committed_data, bh->b_data, bh->b_size); } jbd_unlock_bh_state(bh); out: journal_put_journal_head(jh); |
304c4c841 [PATCH] jbd: avoi... |
928 |
if (unlikely(committed_data)) |
c089d490d JBD: JBD slab all... |
929 |
jbd_free(committed_data, bh->b_size); |
1da177e4c Linux-2.6.12-rc2 |
930 931 |
return err; } |
ae6ddcc5f [PATCH] ext3 and ... |
932 |
/** |
78a4a50a8 docbook: fix file... |
933 |
* int journal_dirty_data() - mark a buffer as containing dirty data to be flushed |
1da177e4c Linux-2.6.12-rc2 |
934 935 |
* @handle: transaction * @bh: bufferhead to mark |
ae6ddcc5f [PATCH] ext3 and ... |
936 |
* |
78a4a50a8 docbook: fix file... |
937 938 939 940 |
* Description: * Mark a buffer as containing dirty data which needs to be flushed before * we can commit the current transaction. * |
1da177e4c Linux-2.6.12-rc2 |
941 942 943 944 945 946 947 948 949 950 951 952 953 |
* The buffer is placed on the transaction's data list and is marked as * belonging to the transaction. * * Returns error number or 0 on success. * * journal_dirty_data() can be called via page_launder->ext3_writepage * by kswapd. */ int journal_dirty_data(handle_t *handle, struct buffer_head *bh) { journal_t *journal = handle->h_transaction->t_journal; int need_brelse = 0; struct journal_head *jh; |
960a22ae6 jbd: ordered data... |
954 |
int ret = 0; |
1da177e4c Linux-2.6.12-rc2 |
955 956 |
if (is_handle_aborted(handle)) |
960a22ae6 jbd: ordered data... |
957 |
return ret; |
1da177e4c Linux-2.6.12-rc2 |
958 959 960 961 962 963 964 965 966 967 968 969 970 |
jh = journal_add_journal_head(bh); JBUFFER_TRACE(jh, "entry"); /* * The buffer could *already* be dirty. Writeout can start * at any time. */ jbd_debug(4, "jh: %p, tid:%d ", jh, handle->h_transaction->t_tid); /* * What if the buffer is already part of a running transaction? |
ae6ddcc5f [PATCH] ext3 and ... |
971 |
* |
1da177e4c Linux-2.6.12-rc2 |
972 973 974 |
* There are two cases: * 1) It is part of the current running transaction. Refile it, * just in case we have allocated it as metadata, deallocated |
ae6ddcc5f [PATCH] ext3 and ... |
975 |
* it, then reallocated it as data. |
1da177e4c Linux-2.6.12-rc2 |
976 977 978 |
* 2) It is part of the previous, still-committing transaction. * If all we want to do is to guarantee that the buffer will be * written to disk before this new transaction commits, then |
ae6ddcc5f [PATCH] ext3 and ... |
979 |
* being sure that the *previous* transaction has this same |
1da177e4c Linux-2.6.12-rc2 |
980 981 982 983 984 985 986 987 988 989 990 991 |
* property is sufficient for us! Just leave it on its old * transaction. * * In case (2), the buffer must not already exist as metadata * --- that would violate write ordering (a transaction is free * to write its data at any point, even before the previous * committing transaction has committed). The caller must * never, ever allow this to happen: there's nothing we can do * about it in this layer. */ jbd_lock_bh_state(bh); spin_lock(&journal->j_list_lock); |
f58a74dca [PATCH] jbd: jour... |
992 993 994 995 996 997 |
/* Now that we have bh_state locked, are we really still mapped? */ if (!buffer_mapped(bh)) { JBUFFER_TRACE(jh, "unmapped buffer, bailing out"); goto no_journal; } |
1da177e4c Linux-2.6.12-rc2 |
998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 |
if (jh->b_transaction) { JBUFFER_TRACE(jh, "has transaction"); if (jh->b_transaction != handle->h_transaction) { JBUFFER_TRACE(jh, "belongs to older transaction"); J_ASSERT_JH(jh, jh->b_transaction == journal->j_committing_transaction); /* @@@ IS THIS TRUE ? */ /* * Not any more. Scenario: someone does a write() * in data=journal mode. The buffer's transaction has * moved into commit. Then someone does another * write() to the file. We do the frozen data copyout * and set b_next_transaction to point to j_running_t. * And while we're in that state, someone does a * writepage() in an attempt to pageout the same area * of the file via a shared mapping. At present that * calls journal_dirty_data(), and we get right here. * It may be too late to journal the data. Simply * falling through to the next test will suffice: the * data will be dirty and wil be checkpointed. The * ordering comments in the next comment block still * apply. */ //J_ASSERT_JH(jh, jh->b_next_transaction == NULL); /* * If we're journalling data, and this buffer was * subject to a write(), it could be metadata, forget * or shadow against the committing transaction. Now, * someone has dirtied the same darn page via a mapping * and it is being writepage()'d. * We *could* just steal the page from commit, with some * fancy locking there. Instead, we just skip it - * don't tie the page's buffers to the new transaction * at all. * Implication: if we crash before the writepage() data * is written into the filesystem, recovery will replay * the write() data. */ if (jh->b_jlist != BJ_None && jh->b_jlist != BJ_SyncData && jh->b_jlist != BJ_Locked) { JBUFFER_TRACE(jh, "Not stealing"); goto no_journal; } /* * This buffer may be undergoing writeout in commit. We * can't return from here and let the caller dirty it * again because that can cause the write-out loop in * commit to never terminate. */ if (buffer_dirty(bh)) { get_bh(bh); spin_unlock(&journal->j_list_lock); jbd_unlock_bh_state(bh); need_brelse = 1; sync_dirty_buffer(bh); jbd_lock_bh_state(bh); spin_lock(&journal->j_list_lock); |
f58a74dca [PATCH] jbd: jour... |
1059 1060 1061 1062 1063 |
/* Since we dropped the lock... */ if (!buffer_mapped(bh)) { JBUFFER_TRACE(jh, "buffer got unmapped"); goto no_journal; } |
1da177e4c Linux-2.6.12-rc2 |
1064 1065 1066 |
/* The buffer may become locked again at any time if it is redirtied */ } |
960a22ae6 jbd: ordered data... |
1067 1068 1069 1070 1071 1072 1073 1074 1075 |
/* * We cannot remove the buffer with io error from the * committing transaction, because otherwise it would * miss the error and the commit would not abort. */ if (unlikely(!buffer_uptodate(bh))) { ret = -EIO; goto no_journal; } |
bb189247f jbd: Fix oops in ... |
1076 1077 1078 |
/* We might have slept so buffer could be refiled now */ if (jh->b_transaction != NULL && jh->b_transaction != handle->h_transaction) { |
1da177e4c Linux-2.6.12-rc2 |
1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 |
JBUFFER_TRACE(jh, "unfile from commit"); __journal_temp_unlink_buffer(jh); /* It still points to the committing * transaction; move it to this one so * that the refile assert checks are * happy. */ jh->b_transaction = handle->h_transaction; } /* The buffer will be refiled below */ } /* * Special case --- the buffer might actually have been * allocated and then immediately deallocated in the previous, * committing transaction, so might still be left on that * transaction's metadata lists. */ if (jh->b_jlist != BJ_SyncData && jh->b_jlist != BJ_Locked) { JBUFFER_TRACE(jh, "not on correct data list: unfile"); J_ASSERT_JH(jh, jh->b_jlist != BJ_Shadow); |
1da177e4c Linux-2.6.12-rc2 |
1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 |
JBUFFER_TRACE(jh, "file as data"); __journal_file_buffer(jh, handle->h_transaction, BJ_SyncData); } } else { JBUFFER_TRACE(jh, "not on a transaction"); __journal_file_buffer(jh, handle->h_transaction, BJ_SyncData); } no_journal: spin_unlock(&journal->j_list_lock); jbd_unlock_bh_state(bh); if (need_brelse) { BUFFER_TRACE(bh, "brelse"); __brelse(bh); } JBUFFER_TRACE(jh, "exit"); journal_put_journal_head(jh); |
960a22ae6 jbd: ordered data... |
1116 |
return ret; |
1da177e4c Linux-2.6.12-rc2 |
1117 |
} |
ae6ddcc5f [PATCH] ext3 and ... |
1118 |
/** |
78a4a50a8 docbook: fix file... |
1119 |
* int journal_dirty_metadata() - mark a buffer as containing dirty metadata |
1da177e4c Linux-2.6.12-rc2 |
1120 |
* @handle: transaction to add buffer to. |
ae6ddcc5f [PATCH] ext3 and ... |
1121 1122 |
* @bh: buffer to mark * |
78a4a50a8 docbook: fix file... |
1123 |
* Mark dirty metadata which needs to be journaled as part of the current |
1da177e4c Linux-2.6.12-rc2 |
1124 1125 1126 |
* transaction. * * The buffer is placed on the transaction's metadata list and is marked |
ae6ddcc5f [PATCH] ext3 and ... |
1127 |
* as belonging to the transaction. |
1da177e4c Linux-2.6.12-rc2 |
1128 |
* |
ae6ddcc5f [PATCH] ext3 and ... |
1129 |
* Returns error number or 0 on success. |
1da177e4c Linux-2.6.12-rc2 |
1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 |
* * Special care needs to be taken if the buffer already belongs to the * current committing transaction (in which case we should have frozen * data present for that commit). In that case, we don't relink the * buffer: that only gets done when the old transaction finally * completes its commit. */ int journal_dirty_metadata(handle_t *handle, struct buffer_head *bh) { transaction_t *transaction = handle->h_transaction; journal_t *journal = transaction->t_journal; struct journal_head *jh = bh2jh(bh); jbd_debug(5, "journal_head %p ", jh); JBUFFER_TRACE(jh, "entry"); if (is_handle_aborted(handle)) goto out; jbd_lock_bh_state(bh); if (jh->b_modified == 0) { /* * This buffer's got modified and becoming part * of the transaction. This needs to be done * once a transaction -bzzz */ jh->b_modified = 1; J_ASSERT_JH(jh, handle->h_buffer_credits > 0); handle->h_buffer_credits--; } /* * fastpath, to avoid expensive locking. If this buffer is already * on the running transaction's metadata list there is nothing to do. * Nobody can take it off again because there is a handle open. * I _think_ we're OK here with SMP barriers - a mistaken decision will * result in this test being false, so we go in and take the locks. */ if (jh->b_transaction == transaction && jh->b_jlist == BJ_Metadata) { JBUFFER_TRACE(jh, "fastpath"); J_ASSERT_JH(jh, jh->b_transaction == journal->j_running_transaction); goto out_unlock_bh; } set_buffer_jbddirty(bh); |
ae6ddcc5f [PATCH] ext3 and ... |
1177 |
/* |
1da177e4c Linux-2.6.12-rc2 |
1178 1179 1180 |
* Metadata already on the current transaction list doesn't * need to be filed. Metadata on another transaction's list must * be committing, and will be refiled once the commit completes: |
ae6ddcc5f [PATCH] ext3 and ... |
1181 |
* leave it alone for now. |
1da177e4c Linux-2.6.12-rc2 |
1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 |
*/ if (jh->b_transaction != transaction) { JBUFFER_TRACE(jh, "already on other transaction"); J_ASSERT_JH(jh, jh->b_transaction == journal->j_committing_transaction); J_ASSERT_JH(jh, jh->b_next_transaction == transaction); /* And this case is illegal: we can't reuse another * transaction's data buffer, ever. */ goto out_unlock_bh; } /* That test should have eliminated the following case: */ |
c80544dc0 sparse pointer us... |
1194 |
J_ASSERT_JH(jh, jh->b_frozen_data == NULL); |
1da177e4c Linux-2.6.12-rc2 |
1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 |
JBUFFER_TRACE(jh, "file as BJ_Metadata"); spin_lock(&journal->j_list_lock); __journal_file_buffer(jh, handle->h_transaction, BJ_Metadata); spin_unlock(&journal->j_list_lock); out_unlock_bh: jbd_unlock_bh_state(bh); out: JBUFFER_TRACE(jh, "exit"); return 0; } |
ae6ddcc5f [PATCH] ext3 and ... |
1206 |
/* |
1da177e4c Linux-2.6.12-rc2 |
1207 1208 1209 1210 1211 1212 1213 1214 1215 |
* journal_release_buffer: undo a get_write_access without any buffer * updates, if the update decided in the end that it didn't need access. * */ void journal_release_buffer(handle_t *handle, struct buffer_head *bh) { BUFFER_TRACE(bh, "entry"); } |
ae6ddcc5f [PATCH] ext3 and ... |
1216 |
/** |
1da177e4c Linux-2.6.12-rc2 |
1217 1218 1219 1220 1221 1222 |
* void journal_forget() - bforget() for potentially-journaled buffers. * @handle: transaction handle * @bh: bh to 'forget' * * We can only do the bforget if there are no commits pending against the * buffer. If the buffer is dirty in the current running transaction we |
ae6ddcc5f [PATCH] ext3 and ... |
1223 |
* can safely unlink it. |
1da177e4c Linux-2.6.12-rc2 |
1224 1225 1226 1227 1228 |
* * bh may not be a journalled buffer at all - it may be a non-JBD * buffer which came off the hashtable. Check for this. * * Decrements bh->b_count by one. |
ae6ddcc5f [PATCH] ext3 and ... |
1229 |
* |
1da177e4c Linux-2.6.12-rc2 |
1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 |
* Allow this call even if the handle has aborted --- it may be part of * the caller's cleanup after an abort. */ int journal_forget (handle_t *handle, struct buffer_head *bh) { transaction_t *transaction = handle->h_transaction; journal_t *journal = transaction->t_journal; struct journal_head *jh; int drop_reserve = 0; int err = 0; |
5b9a499d7 jbd: fix possible... |
1240 |
int was_modified = 0; |
1da177e4c Linux-2.6.12-rc2 |
1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 |
BUFFER_TRACE(bh, "entry"); jbd_lock_bh_state(bh); spin_lock(&journal->j_list_lock); if (!buffer_jbd(bh)) goto not_jbd; jh = bh2jh(bh); /* Critical error: attempting to delete a bitmap buffer, maybe? * Don't do any jbd operations, and return an error. */ if (!J_EXPECT_JH(jh, !jh->b_committed_data, "inconsistent data on disk")) { err = -EIO; goto not_jbd; } |
5b9a499d7 jbd: fix possible... |
1258 1259 |
/* keep track of wether or not this transaction modified us */ was_modified = jh->b_modified; |
1da177e4c Linux-2.6.12-rc2 |
1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 |
/* * The buffer's going from the transaction, we must drop * all references -bzzz */ jh->b_modified = 0; if (jh->b_transaction == handle->h_transaction) { J_ASSERT_JH(jh, !jh->b_frozen_data); /* If we are forgetting a buffer which is already part * of this transaction, then we can just drop it from * the transaction immediately. */ clear_buffer_dirty(bh); clear_buffer_jbddirty(bh); JBUFFER_TRACE(jh, "belongs to current transaction: unfile"); |
5b9a499d7 jbd: fix possible... |
1276 1277 1278 1279 1280 1281 |
/* * we only want to drop a reference if this transaction * modified the buffer */ if (was_modified) drop_reserve = 1; |
1da177e4c Linux-2.6.12-rc2 |
1282 |
|
ae6ddcc5f [PATCH] ext3 and ... |
1283 |
/* |
1da177e4c Linux-2.6.12-rc2 |
1284 1285 1286 1287 1288 1289 1290 1291 |
* We are no longer going to journal this buffer. * However, the commit of this transaction is still * important to the buffer: the delete that we are now * processing might obsolete an old log entry, so by * committing, we can satisfy the buffer's checkpoint. * * So, if we have a checkpoint on the buffer, we should * now refile the buffer on our BJ_Forget list so that |
ae6ddcc5f [PATCH] ext3 and ... |
1292 |
* we know to remove the checkpoint after we commit. |
1da177e4c Linux-2.6.12-rc2 |
1293 1294 1295 1296 1297 1298 1299 |
*/ if (jh->b_cp_transaction) { __journal_temp_unlink_buffer(jh); __journal_file_buffer(jh, transaction, BJ_Forget); } else { __journal_unfile_buffer(jh); |
1da177e4c Linux-2.6.12-rc2 |
1300 1301 1302 1303 1304 1305 1306 1307 |
if (!buffer_jbd(bh)) { spin_unlock(&journal->j_list_lock); jbd_unlock_bh_state(bh); __bforget(bh); goto drop; } } } else if (jh->b_transaction) { |
ae6ddcc5f [PATCH] ext3 and ... |
1308 |
J_ASSERT_JH(jh, (jh->b_transaction == |
1da177e4c Linux-2.6.12-rc2 |
1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 |
journal->j_committing_transaction)); /* However, if the buffer is still owned by a prior * (committing) transaction, we can't drop it yet... */ JBUFFER_TRACE(jh, "belongs to older transaction"); /* ... but we CAN drop it from the new transaction if we * have also modified it since the original commit. */ if (jh->b_next_transaction) { J_ASSERT(jh->b_next_transaction == transaction); jh->b_next_transaction = NULL; |
5b9a499d7 jbd: fix possible... |
1319 1320 1321 1322 1323 1324 1325 |
/* * only drop a reference if this transaction modified * the buffer */ if (was_modified) drop_reserve = 1; |
1da177e4c Linux-2.6.12-rc2 |
1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 |
} } not_jbd: spin_unlock(&journal->j_list_lock); jbd_unlock_bh_state(bh); __brelse(bh); drop: if (drop_reserve) { /* no need to reserve log space for this block -bzzz */ handle->h_buffer_credits++; } return err; } /** * int journal_stop() - complete a transaction * @handle: tranaction to complete. |
ae6ddcc5f [PATCH] ext3 and ... |
1344 |
* |
1da177e4c Linux-2.6.12-rc2 |
1345 1346 1347 1348 1349 1350 1351 1352 |
* All done for a particular handle. * * There is not much action needed here. We just return any remaining * buffer credits to the transaction and remove the handle. The only * complication is that we need to start a commit operation if the * filesystem is marked for synchronous update. * * journal_stop itself will not usually return an error, but it may |
ae6ddcc5f [PATCH] ext3 and ... |
1353 |
* do so in unusual circumstances. In particular, expect it to |
1da177e4c Linux-2.6.12-rc2 |
1354 1355 1356 1357 1358 1359 1360 |
* return -EIO if a journal_abort has been executed since the * transaction began. */ int journal_stop(handle_t *handle) { transaction_t *transaction = handle->h_transaction; journal_t *journal = transaction->t_journal; |
f420d4dc4 jbd: improve fsyn... |
1361 |
int err; |
fe1dcbc4f [PATCH] jbd: fix ... |
1362 |
pid_t pid; |
1da177e4c Linux-2.6.12-rc2 |
1363 |
|
1da177e4c Linux-2.6.12-rc2 |
1364 1365 1366 1367 |
J_ASSERT(journal_current_handle() == handle); if (is_handle_aborted(handle)) err = -EIO; |
3e2a532b2 [PATCH] ext3/4: f... |
1368 1369 |
else { J_ASSERT(transaction->t_updates > 0); |
1da177e4c Linux-2.6.12-rc2 |
1370 |
err = 0; |
3e2a532b2 [PATCH] ext3/4: f... |
1371 |
} |
1da177e4c Linux-2.6.12-rc2 |
1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 |
if (--handle->h_ref > 0) { jbd_debug(4, "h_ref %d -> %d ", handle->h_ref + 1, handle->h_ref); return err; } jbd_debug(4, "Handle %p going down ", handle); /* * Implement synchronous transaction batching. If the handle * was synchronous, don't force a commit immediately. Let's * yield and let another thread piggyback onto this transaction. * Keep doing that while new threads continue to arrive. * It doesn't cost much - we're about to run a commit and sleep * on IO anyway. Speeds up many-threaded, many-dir operations * by 30x or more... |
fe1dcbc4f [PATCH] jbd: fix ... |
1391 |
* |
f420d4dc4 jbd: improve fsyn... |
1392 |
* We try and optimize the sleep time against what the underlying disk |
25985edce Fix common misspe... |
1393 |
* can do, instead of having a static sleep time. This is useful for |
f420d4dc4 jbd: improve fsyn... |
1394 1395 1396 |
* the case where our storage is so fast that it is more optimal to go * ahead and force a flush and wait for the transaction to be committed * than it is to wait for an arbitrary amount of time for new writers to |
3ad2f3fbb tree-wide: Assort... |
1397 |
* join the transaction. We achieve this by measuring how long it takes |
f420d4dc4 jbd: improve fsyn... |
1398 1399 1400 1401 1402 |
* to commit a transaction, and compare it with how long this * transaction has been running, and if run time < commit time then we * sleep for the delta and commit. This greatly helps super fast disks * that would see slowdowns as more threads started doing fsyncs. * |
fe1dcbc4f [PATCH] jbd: fix ... |
1403 1404 1405 1406 |
* But don't do this if this process was the most recent one to * perform a synchronous write. We do this to detect the case where a * single process is doing a stream of sync writes. No point in waiting * for joiners in that case. |
1da177e4c Linux-2.6.12-rc2 |
1407 |
*/ |
fe1dcbc4f [PATCH] jbd: fix ... |
1408 1409 |
pid = current->pid; if (handle->h_sync && journal->j_last_sync_writer != pid) { |
f420d4dc4 jbd: improve fsyn... |
1410 |
u64 commit_time, trans_time; |
fe1dcbc4f [PATCH] jbd: fix ... |
1411 |
journal->j_last_sync_writer = pid; |
f420d4dc4 jbd: improve fsyn... |
1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 |
spin_lock(&journal->j_state_lock); commit_time = journal->j_average_commit_time; spin_unlock(&journal->j_state_lock); trans_time = ktime_to_ns(ktime_sub(ktime_get(), transaction->t_start_time)); commit_time = min_t(u64, commit_time, 1000*jiffies_to_usecs(1)); if (trans_time < commit_time) { ktime_t expires = ktime_add_ns(ktime_get(), commit_time); set_current_state(TASK_UNINTERRUPTIBLE); schedule_hrtimeout(&expires, HRTIMER_MODE_ABS); } |
1da177e4c Linux-2.6.12-rc2 |
1429 |
} |
512a00438 ext3: Use WRITE_S... |
1430 1431 |
if (handle->h_sync) transaction->t_synchronous_commit = 1; |
1da177e4c Linux-2.6.12-rc2 |
1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 |
current->journal_info = NULL; spin_lock(&journal->j_state_lock); spin_lock(&transaction->t_handle_lock); transaction->t_outstanding_credits -= handle->h_buffer_credits; transaction->t_updates--; if (!transaction->t_updates) { wake_up(&journal->j_wait_updates); if (journal->j_barrier_count) wake_up(&journal->j_wait_transaction_locked); } /* * If the handle is marked SYNC, we need to set another commit * going! We also want to force a commit if the current * transaction is occupying too much of the log, or if the * transaction is too old now. */ if (handle->h_sync || transaction->t_outstanding_credits > journal->j_max_transaction_buffers || |
e9ad5620b [PATCH] ext3: Mor... |
1452 |
time_after_eq(jiffies, transaction->t_expires)) { |
1da177e4c Linux-2.6.12-rc2 |
1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 |
/* Do this even for aborted journals: an abort still * completes the commit thread, it just doesn't write * anything to disk. */ tid_t tid = transaction->t_tid; spin_unlock(&transaction->t_handle_lock); jbd_debug(2, "transaction too old, requesting commit for " "handle %p ", handle); /* This is non-blocking */ __log_start_commit(journal, transaction->t_tid); spin_unlock(&journal->j_state_lock); /* * Special case: JFS_SYNC synchronous updates require us |
ae6ddcc5f [PATCH] ext3 and ... |
1468 |
* to wait for the commit to complete. |
1da177e4c Linux-2.6.12-rc2 |
1469 1470 1471 1472 1473 1474 1475 |
*/ if (handle->h_sync && !(current->flags & PF_MEMALLOC)) err = log_wait_commit(journal, tid); } else { spin_unlock(&transaction->t_handle_lock); spin_unlock(&journal->j_state_lock); } |
3295f0ef9 lockdep: rename m... |
1476 |
lock_map_release(&handle->h_lockdep_map); |
34a3d1e83 lockdep: annotate... |
1477 |
|
1da177e4c Linux-2.6.12-rc2 |
1478 1479 1480 |
jbd_free_handle(handle); return err; } |
0cf01f668 jbd: fix jbd kern... |
1481 1482 |
/** * int journal_force_commit() - force any uncommitted transactions |
1da177e4c Linux-2.6.12-rc2 |
1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 |
* @journal: journal to force * * For synchronous operations: force any uncommitted transactions * to disk. May seem kludgy, but it reuses all the handle batching * code in a very simple manner. */ int journal_force_commit(journal_t *journal) { handle_t *handle; int ret; handle = journal_start(journal, 1); if (IS_ERR(handle)) { ret = PTR_ERR(handle); } else { handle->h_sync = 1; ret = journal_stop(handle); } return ret; } /* * * List management code snippets: various functions for manipulating the * transaction buffer lists. * */ /* * Append a buffer to a transaction list, given the transaction's list head * pointer. * * j_list_lock is held. * * jbd_lock_bh_state(jh2bh(jh)) is held. */ |
ae6ddcc5f [PATCH] ext3 and ... |
1519 |
static inline void |
1da177e4c Linux-2.6.12-rc2 |
1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 |
__blist_add_buffer(struct journal_head **list, struct journal_head *jh) { if (!*list) { jh->b_tnext = jh->b_tprev = jh; *list = jh; } else { /* Insert at the tail of the list to preserve order */ struct journal_head *first = *list, *last = first->b_tprev; jh->b_tprev = last; jh->b_tnext = first; last->b_tnext = first->b_tprev = jh; } } |
ae6ddcc5f [PATCH] ext3 and ... |
1533 |
/* |
1da177e4c Linux-2.6.12-rc2 |
1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 |
* Remove a buffer from a transaction list, given the transaction's list * head pointer. * * Called with j_list_lock held, and the journal may not be locked. * * jbd_lock_bh_state(jh2bh(jh)) is held. */ static inline void __blist_del_buffer(struct journal_head **list, struct journal_head *jh) { if (*list == jh) { *list = jh->b_tnext; if (*list == jh) *list = NULL; } jh->b_tprev->b_tnext = jh->b_tnext; jh->b_tnext->b_tprev = jh->b_tprev; } |
ae6ddcc5f [PATCH] ext3 and ... |
1553 |
/* |
1da177e4c Linux-2.6.12-rc2 |
1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 |
* Remove a buffer from the appropriate transaction list. * * Note that this function can *change* the value of * bh->b_transaction->t_sync_datalist, t_buffers, t_forget, * t_iobuf_list, t_shadow_list, t_log_list or t_reserved_list. If the caller * is holding onto a copy of one of thee pointers, it could go bad. * Generally the caller needs to re-read the pointer from the transaction_t. * * Called under j_list_lock. The journal may not be locked. */ |
d394e122b [PATCH] make fs/j... |
1564 |
static void __journal_temp_unlink_buffer(struct journal_head *jh) |
1da177e4c Linux-2.6.12-rc2 |
1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 |
{ struct journal_head **list = NULL; transaction_t *transaction; struct buffer_head *bh = jh2bh(jh); J_ASSERT_JH(jh, jbd_is_locked_bh_state(bh)); transaction = jh->b_transaction; if (transaction) assert_spin_locked(&transaction->t_journal->j_list_lock); J_ASSERT_JH(jh, jh->b_jlist < BJ_Types); if (jh->b_jlist != BJ_None) |
c80544dc0 sparse pointer us... |
1577 |
J_ASSERT_JH(jh, transaction != NULL); |
1da177e4c Linux-2.6.12-rc2 |
1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 |
switch (jh->b_jlist) { case BJ_None: return; case BJ_SyncData: list = &transaction->t_sync_datalist; break; case BJ_Metadata: transaction->t_nr_buffers--; J_ASSERT_JH(jh, transaction->t_nr_buffers >= 0); list = &transaction->t_buffers; break; case BJ_Forget: list = &transaction->t_forget; break; case BJ_IO: list = &transaction->t_iobuf_list; break; case BJ_Shadow: list = &transaction->t_shadow_list; break; case BJ_LogCtl: list = &transaction->t_log_list; break; case BJ_Reserved: list = &transaction->t_reserved_list; break; case BJ_Locked: list = &transaction->t_locked_list; break; } __blist_del_buffer(list, jh); jh->b_jlist = BJ_None; if (test_clear_buffer_jbddirty(bh)) mark_buffer_dirty(bh); /* Expose it to the VM */ } |
bb189247f jbd: Fix oops in ... |
1615 1616 1617 1618 1619 1620 1621 |
/* * Remove buffer from all transactions. * * Called with bh_state lock and j_list_lock * * jh and bh may be already freed when this function returns. */ |
1da177e4c Linux-2.6.12-rc2 |
1622 1623 1624 1625 |
void __journal_unfile_buffer(struct journal_head *jh) { __journal_temp_unlink_buffer(jh); jh->b_transaction = NULL; |
bb189247f jbd: Fix oops in ... |
1626 |
journal_put_journal_head(jh); |
1da177e4c Linux-2.6.12-rc2 |
1627 1628 1629 1630 |
} void journal_unfile_buffer(journal_t *journal, struct journal_head *jh) { |
bb189247f jbd: Fix oops in ... |
1631 1632 1633 1634 1635 |
struct buffer_head *bh = jh2bh(jh); /* Get reference so that buffer cannot be freed before we unlock it */ get_bh(bh); jbd_lock_bh_state(bh); |
1da177e4c Linux-2.6.12-rc2 |
1636 1637 1638 |
spin_lock(&journal->j_list_lock); __journal_unfile_buffer(jh); spin_unlock(&journal->j_list_lock); |
bb189247f jbd: Fix oops in ... |
1639 1640 |
jbd_unlock_bh_state(bh); __brelse(bh); |
1da177e4c Linux-2.6.12-rc2 |
1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 |
} /* * Called from journal_try_to_free_buffers(). * * Called under jbd_lock_bh_state(bh) */ static void __journal_try_to_free_buffer(journal_t *journal, struct buffer_head *bh) { struct journal_head *jh; jh = bh2jh(bh); if (buffer_locked(bh) || buffer_dirty(bh)) goto out; |
c80544dc0 sparse pointer us... |
1657 |
if (jh->b_next_transaction != NULL) |
1da177e4c Linux-2.6.12-rc2 |
1658 1659 1660 |
goto out; spin_lock(&journal->j_list_lock); |
c80544dc0 sparse pointer us... |
1661 |
if (jh->b_transaction != NULL && jh->b_cp_transaction == NULL) { |
1da177e4c Linux-2.6.12-rc2 |
1662 1663 1664 1665 |
if (jh->b_jlist == BJ_SyncData || jh->b_jlist == BJ_Locked) { /* A written-back ordered data buffer */ JBUFFER_TRACE(jh, "release data"); __journal_unfile_buffer(jh); |
1da177e4c Linux-2.6.12-rc2 |
1666 |
} |
c80544dc0 sparse pointer us... |
1667 |
} else if (jh->b_cp_transaction != NULL && jh->b_transaction == NULL) { |
1da177e4c Linux-2.6.12-rc2 |
1668 1669 1670 1671 |
/* written-back checkpointed metadata buffer */ if (jh->b_jlist == BJ_None) { JBUFFER_TRACE(jh, "remove from checkpoint list"); __journal_remove_checkpoint(jh); |
1da177e4c Linux-2.6.12-rc2 |
1672 1673 1674 1675 1676 1677 |
} } spin_unlock(&journal->j_list_lock); out: return; } |
ae6ddcc5f [PATCH] ext3 and ... |
1678 |
/** |
1da177e4c Linux-2.6.12-rc2 |
1679 1680 1681 |
* int journal_try_to_free_buffers() - try to free page buffers. * @journal: journal for operation * @page: to try and free |
3f31fddfa jbd: fix race bet... |
1682 1683 1684 |
* @gfp_mask: we use the mask to detect how hard should we try to release * buffers. If __GFP_WAIT and __GFP_FS is set, we wait for commit code to * release the buffers. |
1da177e4c Linux-2.6.12-rc2 |
1685 |
* |
ae6ddcc5f [PATCH] ext3 and ... |
1686 |
* |
1da177e4c Linux-2.6.12-rc2 |
1687 1688 1689 |
* For all the buffers on this page, * if they are fully written out ordered data, move them onto BUF_CLEAN * so try_to_free_buffers() can reap them. |
ae6ddcc5f [PATCH] ext3 and ... |
1690 |
* |
1da177e4c Linux-2.6.12-rc2 |
1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 |
* This function returns non-zero if we wish try_to_free_buffers() * to be called. We do this if the page is releasable by try_to_free_buffers(). * We also do it if the page has locked or dirty buffers and the caller wants * us to perform sync or async writeout. * * This complicates JBD locking somewhat. We aren't protected by the * BKL here. We wish to remove the buffer from its committing or * running transaction's ->t_datalist via __journal_unfile_buffer. * * This may *change* the value of transaction_t->t_datalist, so anyone * who looks at t_datalist needs to lock against this function. * * Even worse, someone may be doing a journal_dirty_data on this * buffer. So we need to lock against that. journal_dirty_data() * will come out of the lock with the buffer dirty, which makes it * ineligible for release here. * * Who else is affected by this? hmm... Really the only contender * is do_get_write_access() - it could be looking at the buffer while * journal_try_to_free_buffer() is changing its state. But that * cannot happen because we never reallocate freed data as metadata * while the data is part of a transaction. Yes? |
3f31fddfa jbd: fix race bet... |
1713 1714 |
* * Return 0 on failure, 1 on success |
1da177e4c Linux-2.6.12-rc2 |
1715 |
*/ |
ae6ddcc5f [PATCH] ext3 and ... |
1716 |
int journal_try_to_free_buffers(journal_t *journal, |
3f31fddfa jbd: fix race bet... |
1717 |
struct page *page, gfp_t gfp_mask) |
1da177e4c Linux-2.6.12-rc2 |
1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 |
{ struct buffer_head *head; struct buffer_head *bh; int ret = 0; J_ASSERT(PageLocked(page)); head = page_buffers(page); bh = head; do { struct journal_head *jh; /* * We take our own ref against the journal_head here to avoid * having to add tons of locking around each instance of |
bb189247f jbd: Fix oops in ... |
1733 |
* journal_put_journal_head(). |
1da177e4c Linux-2.6.12-rc2 |
1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 |
*/ jh = journal_grab_journal_head(bh); if (!jh) continue; jbd_lock_bh_state(bh); __journal_try_to_free_buffer(journal, bh); journal_put_journal_head(jh); jbd_unlock_bh_state(bh); if (buffer_jbd(bh)) goto busy; } while ((bh = bh->b_this_page) != head); |
3f31fddfa jbd: fix race bet... |
1746 |
|
1da177e4c Linux-2.6.12-rc2 |
1747 |
ret = try_to_free_buffers(page); |
3f31fddfa jbd: fix race bet... |
1748 |
|
1da177e4c Linux-2.6.12-rc2 |
1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 |
busy: return ret; } /* * This buffer is no longer needed. If it is on an older transaction's * checkpoint list we need to record it on this transaction's forget list * to pin this buffer (and hence its checkpointing transaction) down until * this transaction commits. If the buffer isn't on a checkpoint list, we * release it. * Returns non-zero if JBD no longer has an interest in the buffer. * * Called under j_list_lock. * * Called under jbd_lock_bh_state(bh). */ static int __dispose_buffer(struct journal_head *jh, transaction_t *transaction) { int may_free = 1; struct buffer_head *bh = jh2bh(jh); |
1da177e4c Linux-2.6.12-rc2 |
1769 1770 |
if (jh->b_cp_transaction) { JBUFFER_TRACE(jh, "on running+cp transaction"); |
bb189247f jbd: Fix oops in ... |
1771 |
__journal_temp_unlink_buffer(jh); |
1e9fd53b7 jbd: Fix a race b... |
1772 1773 1774 1775 1776 1777 |
/* * We don't want to write the buffer anymore, clear the * bit so that we don't confuse checks in * __journal_file_buffer */ clear_buffer_dirty(bh); |
1da177e4c Linux-2.6.12-rc2 |
1778 |
__journal_file_buffer(jh, transaction, BJ_Forget); |
1da177e4c Linux-2.6.12-rc2 |
1779 1780 1781 |
may_free = 0; } else { JBUFFER_TRACE(jh, "on running transaction"); |
bb189247f jbd: Fix oops in ... |
1782 |
__journal_unfile_buffer(jh); |
1da177e4c Linux-2.6.12-rc2 |
1783 1784 1785 1786 1787 |
} return may_free; } /* |
ae6ddcc5f [PATCH] ext3 and ... |
1788 |
* journal_invalidatepage |
1da177e4c Linux-2.6.12-rc2 |
1789 1790 1791 1792 1793 1794 1795 |
* * This code is tricky. It has a number of cases to deal with. * * There are two invariants which this code relies on: * * i_size must be updated on disk before we start calling invalidatepage on the * data. |
ae6ddcc5f [PATCH] ext3 and ... |
1796 |
* |
1da177e4c Linux-2.6.12-rc2 |
1797 1798 1799 1800 1801 |
* This is done in ext3 by defining an ext3_setattr method which * updates i_size before truncate gets going. By maintaining this * invariant, we can be sure that it is safe to throw away any buffers * attached to the current transaction: once the transaction commits, * we know that the data will not be needed. |
ae6ddcc5f [PATCH] ext3 and ... |
1802 |
* |
1da177e4c Linux-2.6.12-rc2 |
1803 |
* Note however that we can *not* throw away data belonging to the |
ae6ddcc5f [PATCH] ext3 and ... |
1804 |
* previous, committing transaction! |
1da177e4c Linux-2.6.12-rc2 |
1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 |
* * Any disk blocks which *are* part of the previous, committing * transaction (and which therefore cannot be discarded immediately) are * not going to be reused in the new running transaction * * The bitmap committed_data images guarantee this: any block which is * allocated in one transaction and removed in the next will be marked * as in-use in the committed_data bitmap, so cannot be reused until * the next transaction to delete the block commits. This means that * leaving committing buffers dirty is quite safe: the disk blocks * cannot be reallocated to a different file and so buffer aliasing is * not possible. * * * The above applies mainly to ordered data mode. In writeback mode we * don't make guarantees about the order in which data hits disk --- in * particular we don't guarantee that new dirty data is flushed before * transaction commit --- so it is always safe just to discard data |
ae6ddcc5f [PATCH] ext3 and ... |
1823 |
* immediately in that mode. --sct |
1da177e4c Linux-2.6.12-rc2 |
1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 |
*/ /* * The journal_unmap_buffer helper function returns zero if the buffer * concerned remains pinned as an anonymous buffer belonging to an older * transaction. * * We're outside-transaction here. Either or both of j_running_transaction * and j_committing_transaction may be NULL. */ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh) { transaction_t *transaction; struct journal_head *jh; int may_free = 1; int ret; BUFFER_TRACE(bh, "entry"); /* * It is safe to proceed here without the j_list_lock because the * buffers cannot be stolen by try_to_free_buffers as long as we are * holding the page lock. --sct */ if (!buffer_jbd(bh)) goto zap_buffer_unlocked; spin_lock(&journal->j_state_lock); jbd_lock_bh_state(bh); spin_lock(&journal->j_list_lock); jh = journal_grab_journal_head(bh); if (!jh) goto zap_buffer_no_jh; |
869639189 jbd: Delay discar... |
1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 |
/* * We cannot remove the buffer from checkpoint lists until the * transaction adding inode to orphan list (let's call it T) * is committed. Otherwise if the transaction changing the * buffer would be cleaned from the journal before T is * committed, a crash will cause that the correct contents of * the buffer will be lost. On the other hand we have to * clear the buffer dirty bit at latest at the moment when the * transaction marking the buffer as freed in the filesystem * structures is committed because from that moment on the * buffer can be reallocated and used by a different page. * Since the block hasn't been freed yet but the inode has * already been added to orphan list, it is safe for us to add * the buffer to BJ_Forget list of the newest transaction. */ |
1da177e4c Linux-2.6.12-rc2 |
1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 |
transaction = jh->b_transaction; if (transaction == NULL) { /* First case: not on any transaction. If it * has no checkpoint link, then we can zap it: * it's a writeback-mode buffer so we don't care * if it hits disk safely. */ if (!jh->b_cp_transaction) { JBUFFER_TRACE(jh, "not on any transaction: zap"); goto zap_buffer; } if (!buffer_dirty(bh)) { /* bdflush has written it. We can drop it now */ goto zap_buffer; } /* OK, it must be in the journal but still not * written fully to disk: it's metadata or * journaled data... */ if (journal->j_running_transaction) { /* ... and once the current transaction has * committed, the buffer won't be needed any * longer. */ JBUFFER_TRACE(jh, "checkpointed: add to BJ_Forget"); ret = __dispose_buffer(jh, journal->j_running_transaction); journal_put_journal_head(jh); spin_unlock(&journal->j_list_lock); jbd_unlock_bh_state(bh); spin_unlock(&journal->j_state_lock); return ret; } else { /* There is no currently-running transaction. So the * orphan record which we wrote for this file must have * passed into commit. We must attach this buffer to * the committing transaction, if it exists. */ if (journal->j_committing_transaction) { JBUFFER_TRACE(jh, "give to committing trans"); ret = __dispose_buffer(jh, journal->j_committing_transaction); journal_put_journal_head(jh); spin_unlock(&journal->j_list_lock); jbd_unlock_bh_state(bh); spin_unlock(&journal->j_state_lock); return ret; } else { /* The orphan record's transaction has * committed. We can cleanse this buffer */ clear_buffer_jbddirty(bh); goto zap_buffer; } } } else if (transaction == journal->j_committing_transaction) { |
f58a74dca [PATCH] jbd: jour... |
1928 |
JBUFFER_TRACE(jh, "on committing transaction"); |
d13df84ff [PATCH] jbd dirty... |
1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 |
if (jh->b_jlist == BJ_Locked) { /* * The buffer is on the committing transaction's locked * list. We have the buffer locked, so I/O has * completed. So we can nail the buffer now. */ may_free = __dispose_buffer(jh, transaction); goto zap_buffer; } /* |
869639189 jbd: Delay discar... |
1939 1940 1941 1942 1943 1944 |
* The buffer is committing, we simply cannot touch * it. So we just set j_next_transaction to the * running transaction (if there is one) and mark * buffer as freed so that commit code knows it should * clear dirty bits when it is done with the buffer. */ |
1da177e4c Linux-2.6.12-rc2 |
1945 |
set_buffer_freed(bh); |
869639189 jbd: Delay discar... |
1946 1947 |
if (journal->j_running_transaction && buffer_jbddirty(bh)) jh->b_next_transaction = journal->j_running_transaction; |
1da177e4c Linux-2.6.12-rc2 |
1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 |
journal_put_journal_head(jh); spin_unlock(&journal->j_list_lock); jbd_unlock_bh_state(bh); spin_unlock(&journal->j_state_lock); return 0; } else { /* Good, the buffer belongs to the running transaction. * We are writing our own transaction's data, not any * previous one's, so it is safe to throw it away * (remember that we expect the filesystem to have set * i_size already for this truncate so recovery will not * expose the disk blocks we are discarding here.) */ J_ASSERT_JH(jh, transaction == journal->j_running_transaction); |
f58a74dca [PATCH] jbd: jour... |
1961 |
JBUFFER_TRACE(jh, "on running transaction"); |
1da177e4c Linux-2.6.12-rc2 |
1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 |
may_free = __dispose_buffer(jh, transaction); } zap_buffer: journal_put_journal_head(jh); zap_buffer_no_jh: spin_unlock(&journal->j_list_lock); jbd_unlock_bh_state(bh); spin_unlock(&journal->j_state_lock); zap_buffer_unlocked: clear_buffer_dirty(bh); J_ASSERT_BH(bh, !buffer_jbddirty(bh)); clear_buffer_mapped(bh); clear_buffer_req(bh); clear_buffer_new(bh); bh->b_bdev = NULL; return may_free; } |
ae6ddcc5f [PATCH] ext3 and ... |
1980 |
/** |
a6b91919e fs: fix kernel-do... |
1981 1982 |
* void journal_invalidatepage() - invalidate a journal page * @journal: journal to use for flush |
1da177e4c Linux-2.6.12-rc2 |
1983 1984 1985 1986 |
* @page: page to flush * @offset: length of page to invalidate. * * Reap page buffers containing data after offset in page. |
1da177e4c Linux-2.6.12-rc2 |
1987 |
*/ |
2ff28e22b [PATCH] Make addr... |
1988 |
void journal_invalidatepage(journal_t *journal, |
ae6ddcc5f [PATCH] ext3 and ... |
1989 |
struct page *page, |
1da177e4c Linux-2.6.12-rc2 |
1990 1991 1992 1993 1994 1995 1996 1997 1998 |
unsigned long offset) { struct buffer_head *head, *bh, *next; unsigned int curr_off = 0; int may_free = 1; if (!PageLocked(page)) BUG(); if (!page_has_buffers(page)) |
2ff28e22b [PATCH] Make addr... |
1999 |
return; |
1da177e4c Linux-2.6.12-rc2 |
2000 2001 2002 2003 2004 2005 2006 2007 2008 |
/* We will potentially be playing with lists other than just the * data lists (especially for journaled data mode), so be * cautious in our locking. */ head = bh = page_buffers(page); do { unsigned int next_off = curr_off + bh->b_size; next = bh->b_this_page; |
1da177e4c Linux-2.6.12-rc2 |
2009 |
if (offset <= curr_off) { |
e9ad5620b [PATCH] ext3: Mor... |
2010 |
/* This block is wholly outside the truncation point */ |
1da177e4c Linux-2.6.12-rc2 |
2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 |
lock_buffer(bh); may_free &= journal_unmap_buffer(journal, bh); unlock_buffer(bh); } curr_off = next_off; bh = next; } while (bh != head); if (!offset) { |
2ff28e22b [PATCH] Make addr... |
2021 2022 |
if (may_free && try_to_free_buffers(page)) J_ASSERT(!page_has_buffers(page)); |
1da177e4c Linux-2.6.12-rc2 |
2023 |
} |
1da177e4c Linux-2.6.12-rc2 |
2024 |
} |
ae6ddcc5f [PATCH] ext3 and ... |
2025 2026 |
/* * File a buffer on the given transaction list. |
1da177e4c Linux-2.6.12-rc2 |
2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 |
*/ void __journal_file_buffer(struct journal_head *jh, transaction_t *transaction, int jlist) { struct journal_head **list = NULL; int was_dirty = 0; struct buffer_head *bh = jh2bh(jh); J_ASSERT_JH(jh, jbd_is_locked_bh_state(bh)); assert_spin_locked(&transaction->t_journal->j_list_lock); J_ASSERT_JH(jh, jh->b_jlist < BJ_Types); J_ASSERT_JH(jh, jh->b_transaction == transaction || |
c80544dc0 sparse pointer us... |
2040 |
jh->b_transaction == NULL); |
1da177e4c Linux-2.6.12-rc2 |
2041 2042 2043 |
if (jh->b_transaction && jh->b_jlist == jlist) return; |
ae6ddcc5f [PATCH] ext3 and ... |
2044 |
if (jlist == BJ_Metadata || jlist == BJ_Reserved || |
1da177e4c Linux-2.6.12-rc2 |
2045 |
jlist == BJ_Shadow || jlist == BJ_Forget) { |
1e9fd53b7 jbd: Fix a race b... |
2046 2047 2048 2049 2050 2051 2052 2053 2054 |
/* * For metadata buffers, we track dirty bit in buffer_jbddirty * instead of buffer_dirty. We should not see a dirty bit set * here because we clear it in do_get_write_access but e.g. * tune2fs can modify the sb and set the dirty bit at any time * so we try to gracefully handle that. */ if (buffer_dirty(bh)) warn_dirty_buffer(bh); |
1da177e4c Linux-2.6.12-rc2 |
2055 2056 2057 2058 2059 2060 2061 |
if (test_clear_buffer_dirty(bh) || test_clear_buffer_jbddirty(bh)) was_dirty = 1; } if (jh->b_transaction) __journal_temp_unlink_buffer(jh); |
bb189247f jbd: Fix oops in ... |
2062 2063 |
else journal_grab_journal_head(bh); |
1da177e4c Linux-2.6.12-rc2 |
2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 |
jh->b_transaction = transaction; switch (jlist) { case BJ_None: J_ASSERT_JH(jh, !jh->b_committed_data); J_ASSERT_JH(jh, !jh->b_frozen_data); return; case BJ_SyncData: list = &transaction->t_sync_datalist; break; case BJ_Metadata: transaction->t_nr_buffers++; list = &transaction->t_buffers; break; case BJ_Forget: list = &transaction->t_forget; break; case BJ_IO: list = &transaction->t_iobuf_list; break; case BJ_Shadow: list = &transaction->t_shadow_list; break; case BJ_LogCtl: list = &transaction->t_log_list; break; case BJ_Reserved: list = &transaction->t_reserved_list; break; case BJ_Locked: list = &transaction->t_locked_list; break; } __blist_add_buffer(list, jh); jh->b_jlist = jlist; if (was_dirty) set_buffer_jbddirty(bh); } void journal_file_buffer(struct journal_head *jh, transaction_t *transaction, int jlist) { jbd_lock_bh_state(jh2bh(jh)); spin_lock(&transaction->t_journal->j_list_lock); __journal_file_buffer(jh, transaction, jlist); spin_unlock(&transaction->t_journal->j_list_lock); jbd_unlock_bh_state(jh2bh(jh)); } |
ae6ddcc5f [PATCH] ext3 and ... |
2114 |
/* |
1da177e4c Linux-2.6.12-rc2 |
2115 2116 2117 2118 2119 |
* Remove a buffer from its current buffer list in preparation for * dropping it from its current transaction entirely. If the buffer has * already started to be used by a subsequent transaction, refile the * buffer on that transaction's metadata list. * |
bb189247f jbd: Fix oops in ... |
2120 |
* Called under j_list_lock |
1da177e4c Linux-2.6.12-rc2 |
2121 |
* Called under jbd_lock_bh_state(jh2bh(jh)) |
bb189247f jbd: Fix oops in ... |
2122 2123 |
* * jh and bh may be already free when this function returns |
1da177e4c Linux-2.6.12-rc2 |
2124 2125 2126 |
*/ void __journal_refile_buffer(struct journal_head *jh) { |
869639189 jbd: Delay discar... |
2127 |
int was_dirty, jlist; |
1da177e4c Linux-2.6.12-rc2 |
2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 |
struct buffer_head *bh = jh2bh(jh); J_ASSERT_JH(jh, jbd_is_locked_bh_state(bh)); if (jh->b_transaction) assert_spin_locked(&jh->b_transaction->t_journal->j_list_lock); /* If the buffer is now unused, just drop it. */ if (jh->b_next_transaction == NULL) { __journal_unfile_buffer(jh); return; } /* * It has been modified by a later transaction: add it to the new * transaction's metadata list. */ was_dirty = test_clear_buffer_jbddirty(bh); __journal_temp_unlink_buffer(jh); |
bb189247f jbd: Fix oops in ... |
2147 2148 2149 2150 2151 |
/* * We set b_transaction here because b_next_transaction will inherit * our jh reference and thus __journal_file_buffer() must not take a * new one. */ |
1da177e4c Linux-2.6.12-rc2 |
2152 2153 |
jh->b_transaction = jh->b_next_transaction; jh->b_next_transaction = NULL; |
869639189 jbd: Delay discar... |
2154 2155 2156 2157 2158 2159 2160 |
if (buffer_freed(bh)) jlist = BJ_Forget; else if (jh->b_modified) jlist = BJ_Metadata; else jlist = BJ_Reserved; __journal_file_buffer(jh, jh->b_transaction, jlist); |
1da177e4c Linux-2.6.12-rc2 |
2161 2162 2163 2164 2165 2166 2167 |
J_ASSERT_JH(jh, jh->b_transaction->t_state == T_RUNNING); if (was_dirty) set_buffer_jbddirty(bh); } /* |
bb189247f jbd: Fix oops in ... |
2168 2169 2170 2171 |
* __journal_refile_buffer() with necessary locking added. We take our bh * reference so that we can safely unlock bh. * * The jh and bh may be freed by this call. |
1da177e4c Linux-2.6.12-rc2 |
2172 2173 2174 2175 |
*/ void journal_refile_buffer(journal_t *journal, struct journal_head *jh) { struct buffer_head *bh = jh2bh(jh); |
bb189247f jbd: Fix oops in ... |
2176 2177 |
/* Get reference so that buffer cannot be freed before we unlock it */ get_bh(bh); |
1da177e4c Linux-2.6.12-rc2 |
2178 2179 |
jbd_lock_bh_state(bh); spin_lock(&journal->j_list_lock); |
1da177e4c Linux-2.6.12-rc2 |
2180 2181 |
__journal_refile_buffer(jh); jbd_unlock_bh_state(bh); |
1da177e4c Linux-2.6.12-rc2 |
2182 2183 2184 |
spin_unlock(&journal->j_list_lock); __brelse(bh); } |