| 1 |
// SPDX-License-Identifier: GPL-2.0-or-later |
| 2 |
/* |
| 3 |
* Copyright (C) International Business Machines Corp., 2000-2005 |
| 4 |
* Portions Copyright (C) Christoph Hellwig, 2001-2002 |
| 5 |
*/ |
| 6 |
|
| 7 |
/* |
| 8 |
* jfs_txnmgr.c: transaction manager |
| 9 |
* |
| 10 |
* notes: |
| 11 |
* transaction starts with txBegin() and ends with txCommit() |
| 12 |
* or txAbort(). |
| 13 |
* |
| 14 |
* tlock is acquired at the time of update; |
| 15 |
* (obviate scan at commit time for xtree and dtree) |
| 16 |
* tlock and mp points to each other; |
| 17 |
* (no hashlist for mp -> tlock). |
| 18 |
* |
| 19 |
* special cases: |
| 20 |
* tlock on in-memory inode: |
| 21 |
* in-place tlock in the in-memory inode itself; |
| 22 |
* converted to page lock by iWrite() at commit time. |
| 23 |
* |
| 24 |
* tlock during write()/mmap() under anonymous transaction (tid = 0): |
| 25 |
* transferred (?) to transaction at commit time. |
| 26 |
* |
| 27 |
* use the page itself to update allocation maps |
| 28 |
* (obviate intermediate replication of allocation/deallocation data) |
| 29 |
* hold on to mp+lock thru update of maps |
| 30 |
*/ |
| 31 |
|
| 32 |
#include <linux/fs.h> |
| 33 |
#include <linux/vmalloc.h> |
| 34 |
#include <linux/completion.h> |
| 35 |
#include <linux/freezer.h> |
| 36 |
#include <linux/module.h> |
| 37 |
#include <linux/moduleparam.h> |
| 38 |
#include <linux/kthread.h> |
| 39 |
#include <linux/seq_file.h> |
| 40 |
#include "jfs_incore.h" |
| 41 |
#include "jfs_inode.h" |
| 42 |
#include "jfs_filsys.h" |
| 43 |
#include "jfs_metapage.h" |
| 44 |
#include "jfs_dinode.h" |
| 45 |
#include "jfs_imap.h" |
| 46 |
#include "jfs_dmap.h" |
| 47 |
#include "jfs_superblock.h" |
| 48 |
#include "jfs_debug.h" |
| 49 |
|
| 50 |
/* |
| 51 |
* transaction management structures |
| 52 |
*/ |
| 53 |
static struct { |
| 54 |
int freetid; /* index of a free tid structure */ |
| 55 |
int freelock; /* index first free lock word */ |
| 56 |
wait_queue_head_t freewait; /* eventlist of free tblock */ |
| 57 |
wait_queue_head_t freelockwait; /* eventlist of free tlock */ |
| 58 |
wait_queue_head_t lowlockwait; /* eventlist of ample tlocks */ |
| 59 |
int tlocksInUse; /* Number of tlocks in use */ |
| 60 |
spinlock_t LazyLock; /* synchronize sync_queue & unlock_queue */ |
| 61 |
/* struct tblock *sync_queue; * Transactions waiting for data sync */ |
| 62 |
struct list_head unlock_queue; /* Txns waiting to be released */ |
| 63 |
struct list_head anon_list; /* inodes having anonymous txns */ |
| 64 |
struct list_head anon_list2; /* inodes having anonymous txns |
| 65 |
that couldn't be sync'ed */ |
| 66 |
} TxAnchor; |
| 67 |
|
| 68 |
int jfs_tlocks_low; /* Indicates low number of available tlocks */ |
| 69 |
|
| 70 |
#ifdef CONFIG_JFS_STATISTICS |
| 71 |
static struct { |
| 72 |
uint txBegin; |
| 73 |
uint txBegin_barrier; |
| 74 |
uint txBegin_lockslow; |
| 75 |
uint txBegin_freetid; |
| 76 |
uint txBeginAnon; |
| 77 |
uint txBeginAnon_barrier; |
| 78 |
uint txBeginAnon_lockslow; |
| 79 |
uint txLockAlloc; |
| 80 |
uint txLockAlloc_freelock; |
| 81 |
} TxStat; |
| 82 |
#endif |
| 83 |
|
| 84 |
static int nTxBlock = -1; /* number of transaction blocks */ |
| 85 |
module_param(nTxBlock, int, 0); |
| 86 |
MODULE_PARM_DESC(nTxBlock, |
| 87 |
"Number of transaction blocks (max:65536)"); |
| 88 |
|
| 89 |
static int nTxLock = -1; /* number of transaction locks */ |
| 90 |
module_param(nTxLock, int, 0); |
| 91 |
MODULE_PARM_DESC(nTxLock, |
| 92 |
"Number of transaction locks (max:65536)"); |
| 93 |
|
| 94 |
struct tblock *TxBlock; /* transaction block table */ |
| 95 |
static int TxLockLWM; /* Low water mark for number of txLocks used */ |
| 96 |
static int TxLockHWM; /* High water mark for number of txLocks used */ |
| 97 |
static int TxLockVHWM; /* Very High water mark */ |
| 98 |
struct tlock *TxLock; /* transaction lock table */ |
| 99 |
|
| 100 |
/* |
| 101 |
* transaction management lock |
| 102 |
*/ |
| 103 |
static DEFINE_SPINLOCK(jfsTxnLock); |
| 104 |
|
| 105 |
#define TXN_LOCK() spin_lock(&jfsTxnLock) |
| 106 |
#define TXN_UNLOCK() spin_unlock(&jfsTxnLock) |
| 107 |
|
| 108 |
#define LAZY_LOCK_INIT() spin_lock_init(&TxAnchor.LazyLock) |
| 109 |
#define LAZY_LOCK(flags) spin_lock_irqsave(&TxAnchor.LazyLock, flags) |
| 110 |
#define LAZY_UNLOCK(flags) spin_unlock_irqrestore(&TxAnchor.LazyLock, flags) |
| 111 |
|
| 112 |
static DECLARE_WAIT_QUEUE_HEAD(jfs_commit_thread_wait); |
| 113 |
static int jfs_commit_thread_waking; |
| 114 |
|
| 115 |
/* |
| 116 |
* Retry logic exist outside these macros to protect from spurrious wakeups. |
| 117 |
*/ |
| 118 |
static inline void TXN_SLEEP_DROP_LOCK(wait_queue_head_t * event) |
| 119 |
{ |
| 120 |
DECLARE_WAITQUEUE(wait, current); |
| 121 |
|
| 122 |
add_wait_queue(event, &wait); |
| 123 |
set_current_state(TASK_UNINTERRUPTIBLE); |
| 124 |
TXN_UNLOCK(); |
| 125 |
io_schedule(); |
| 126 |
remove_wait_queue(event, &wait); |
| 127 |
} |
| 128 |
|
| 129 |
#define TXN_SLEEP(event)\ |
| 130 |
{\ |
| 131 |
TXN_SLEEP_DROP_LOCK(event);\ |
| 132 |
TXN_LOCK();\ |
| 133 |
} |
| 134 |
|
| 135 |
#define TXN_WAKEUP(event) wake_up_all(event) |
| 136 |
|
| 137 |
/* |
| 138 |
* statistics |
| 139 |
*/ |
| 140 |
static struct { |
| 141 |
tid_t maxtid; /* 4: biggest tid ever used */ |
| 142 |
lid_t maxlid; /* 4: biggest lid ever used */ |
| 143 |
int ntid; /* 4: # of transactions performed */ |
| 144 |
int nlid; /* 4: # of tlocks acquired */ |
| 145 |
int waitlock; /* 4: # of tlock wait */ |
| 146 |
} stattx; |
| 147 |
|
| 148 |
/* |
| 149 |
* forward references |
| 150 |
*/ |
| 151 |
static void diLog(struct jfs_log *log, struct tblock *tblk, struct lrd *lrd, |
| 152 |
struct tlock *tlck, struct commit *cd); |
| 153 |
static void dataLog(struct jfs_log *log, struct tblock *tblk, struct lrd *lrd, |
| 154 |
struct tlock *tlck); |
| 155 |
static void dtLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, |
| 156 |
struct tlock * tlck); |
| 157 |
static void mapLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, |
| 158 |
struct tlock * tlck); |
| 159 |
static void txAllocPMap(struct inode *ip, struct maplock * maplock, |
| 160 |
struct tblock * tblk); |
| 161 |
static void txForce(struct tblock * tblk); |
| 162 |
static void txLog(struct jfs_log *log, struct tblock *tblk, |
| 163 |
struct commit *cd); |
| 164 |
static void txUpdateMap(struct tblock * tblk); |
| 165 |
static void txRelease(struct tblock * tblk); |
| 166 |
static void xtLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, |
| 167 |
struct tlock * tlck); |
| 168 |
static void LogSyncRelease(struct metapage * mp); |
| 169 |
|
| 170 |
/* |
| 171 |
* transaction block/lock management |
| 172 |
* --------------------------------- |
| 173 |
*/ |
| 174 |
|
| 175 |
/* |
| 176 |
* Get a transaction lock from the free list. If the number in use is |
| 177 |
* greater than the high water mark, wake up the sync daemon. This should |
| 178 |
* free some anonymous transaction locks. (TXN_LOCK must be held.) |
| 179 |
*/ |
| 180 |
static lid_t txLockAlloc(void) |
| 181 |
{ |
| 182 |
lid_t lid; |
| 183 |
|
| 184 |
INCREMENT(TxStat.txLockAlloc); |
| 185 |
if (!TxAnchor.freelock) { |
| 186 |
INCREMENT(TxStat.txLockAlloc_freelock); |
| 187 |
} |
| 188 |
|
| 189 |
while (!(lid = TxAnchor.freelock)) |
| 190 |
TXN_SLEEP(&TxAnchor.freelockwait); |
| 191 |
TxAnchor.freelock = TxLock[lid].next; |
| 192 |
HIGHWATERMARK(stattx.maxlid, lid); |
| 193 |
if ((++TxAnchor.tlocksInUse > TxLockHWM) && (jfs_tlocks_low == 0)) { |
| 194 |
jfs_info("txLockAlloc tlocks low"); |
| 195 |
jfs_tlocks_low = 1; |
| 196 |
wake_up_process(jfsSyncThread); |
| 197 |
} |
| 198 |
|
| 199 |
return lid; |
| 200 |
} |
| 201 |
|
| 202 |
static void txLockFree(lid_t lid) |
| 203 |
{ |
| 204 |
TxLock[lid].tid = 0; |
| 205 |
TxLock[lid].next = TxAnchor.freelock; |
| 206 |
TxAnchor.freelock = lid; |
| 207 |
TxAnchor.tlocksInUse--; |
| 208 |
if (jfs_tlocks_low && (TxAnchor.tlocksInUse < TxLockLWM)) { |
| 209 |
jfs_info("txLockFree jfs_tlocks_low no more"); |
| 210 |
jfs_tlocks_low = 0; |
| 211 |
TXN_WAKEUP(&TxAnchor.lowlockwait); |
| 212 |
} |
| 213 |
TXN_WAKEUP(&TxAnchor.freelockwait); |
| 214 |
} |
| 215 |
|
| 216 |
/* |
| 217 |
* NAME: txInit() |
| 218 |
* |
| 219 |
* FUNCTION: initialize transaction management structures |
| 220 |
* |
| 221 |
* RETURN: |
| 222 |
* |
| 223 |
* serialization: single thread at jfs_init() |
| 224 |
*/ |
| 225 |
int txInit(void) |
| 226 |
{ |
| 227 |
int k, size; |
| 228 |
struct sysinfo si; |
| 229 |
|
| 230 |
/* Set defaults for nTxLock and nTxBlock if unset */ |
| 231 |
|
| 232 |
if (nTxLock == -1) { |
| 233 |
if (nTxBlock == -1) { |
| 234 |
/* Base default on memory size */ |
| 235 |
si_meminfo(&si); |
| 236 |
if (si.totalram > (256 * 1024)) /* 1 GB */ |
| 237 |
nTxLock = 64 * 1024; |
| 238 |
else |
| 239 |
nTxLock = si.totalram >> 2; |
| 240 |
} else if (nTxBlock > (8 * 1024)) |
| 241 |
nTxLock = 64 * 1024; |
| 242 |
else |
| 243 |
nTxLock = nTxBlock << 3; |
| 244 |
} |
| 245 |
if (nTxBlock == -1) |
| 246 |
nTxBlock = nTxLock >> 3; |
| 247 |
|
| 248 |
/* Verify tunable parameters */ |
| 249 |
if (nTxBlock < 16) |
| 250 |
nTxBlock = 16; /* No one should set it this low */ |
| 251 |
if (nTxBlock > 65536) |
| 252 |
nTxBlock = 65536; |
| 253 |
if (nTxLock < 256) |
| 254 |
nTxLock = 256; /* No one should set it this low */ |
| 255 |
if (nTxLock > 65536) |
| 256 |
nTxLock = 65536; |
| 257 |
|
| 258 |
printk(KERN_INFO "JFS: nTxBlock = %d, nTxLock = %d\n", |
| 259 |
nTxBlock, nTxLock); |
| 260 |
/* |
| 261 |
* initialize transaction block (tblock) table |
| 262 |
* |
| 263 |
* transaction id (tid) = tblock index |
| 264 |
* tid = 0 is reserved. |
| 265 |
*/ |
| 266 |
TxLockLWM = (nTxLock * 4) / 10; |
| 267 |
TxLockHWM = (nTxLock * 7) / 10; |
| 268 |
TxLockVHWM = (nTxLock * 8) / 10; |
| 269 |
|
| 270 |
size = sizeof(struct tblock) * nTxBlock; |
| 271 |
TxBlock = vmalloc(size); |
| 272 |
if (TxBlock == NULL) |
| 273 |
return -ENOMEM; |
| 274 |
|
| 275 |
for (k = 0; k < nTxBlock; k++) { |
| 276 |
init_waitqueue_head(&TxBlock[k].gcwait); |
| 277 |
init_waitqueue_head(&TxBlock[k].waitor); |
| 278 |
} |
| 279 |
|
| 280 |
for (k = 1; k < nTxBlock - 1; k++) { |
| 281 |
TxBlock[k].next = k + 1; |
| 282 |
} |
| 283 |
TxBlock[k].next = 0; |
| 284 |
|
| 285 |
TxAnchor.freetid = 1; |
| 286 |
init_waitqueue_head(&TxAnchor.freewait); |
| 287 |
|
| 288 |
stattx.maxtid = 1; /* statistics */ |
| 289 |
|
| 290 |
/* |
| 291 |
* initialize transaction lock (tlock) table |
| 292 |
* |
| 293 |
* transaction lock id = tlock index |
| 294 |
* tlock id = 0 is reserved. |
| 295 |
*/ |
| 296 |
size = sizeof(struct tlock) * nTxLock; |
| 297 |
TxLock = vmalloc(size); |
| 298 |
if (TxLock == NULL) { |
| 299 |
vfree(TxBlock); |
| 300 |
return -ENOMEM; |
| 301 |
} |
| 302 |
|
| 303 |
/* initialize tlock table */ |
| 304 |
for (k = 1; k < nTxLock - 1; k++) |
| 305 |
TxLock[k].next = k + 1; |
| 306 |
TxLock[k].next = 0; |
| 307 |
init_waitqueue_head(&TxAnchor.freelockwait); |
| 308 |
init_waitqueue_head(&TxAnchor.lowlockwait); |
| 309 |
|
| 310 |
TxAnchor.freelock = 1; |
| 311 |
TxAnchor.tlocksInUse = 0; |
| 312 |
INIT_LIST_HEAD(&TxAnchor.anon_list); |
| 313 |
INIT_LIST_HEAD(&TxAnchor.anon_list2); |
| 314 |
|
| 315 |
LAZY_LOCK_INIT(); |
| 316 |
INIT_LIST_HEAD(&TxAnchor.unlock_queue); |
| 317 |
|
| 318 |
stattx.maxlid = 1; /* statistics */ |
| 319 |
|
| 320 |
return 0; |
| 321 |
} |
| 322 |
|
| 323 |
/* |
| 324 |
* NAME: txExit() |
| 325 |
* |
| 326 |
* FUNCTION: clean up when module is unloaded |
| 327 |
*/ |
| 328 |
void txExit(void) |
| 329 |
{ |
| 330 |
vfree(TxLock); |
| 331 |
TxLock = NULL; |
| 332 |
vfree(TxBlock); |
| 333 |
TxBlock = NULL; |
| 334 |
} |
| 335 |
|
| 336 |
/* |
| 337 |
* NAME: txBegin() |
| 338 |
* |
| 339 |
* FUNCTION: start a transaction. |
| 340 |
* |
| 341 |
* PARAMETER: sb - superblock |
| 342 |
* flag - force for nested tx; |
| 343 |
* |
| 344 |
* RETURN: tid - transaction id |
| 345 |
* |
| 346 |
* note: flag force allows to start tx for nested tx |
| 347 |
* to prevent deadlock on logsync barrier; |
| 348 |
*/ |
| 349 |
tid_t txBegin(struct super_block *sb, int flag) |
| 350 |
{ |
| 351 |
tid_t t; |
| 352 |
struct tblock *tblk; |
| 353 |
struct jfs_log *log; |
| 354 |
|
| 355 |
jfs_info("txBegin: flag = 0x%x", flag); |
| 356 |
log = JFS_SBI(sb)->log; |
| 357 |
|
| 358 |
if (!log) { |
| 359 |
jfs_error(sb, "read-only filesystem\n"); |
| 360 |
return 0; |
| 361 |
} |
| 362 |
|
| 363 |
TXN_LOCK(); |
| 364 |
|
| 365 |
INCREMENT(TxStat.txBegin); |
| 366 |
|
| 367 |
retry: |
| 368 |
if (!(flag & COMMIT_FORCE)) { |
| 369 |
/* |
| 370 |
* synchronize with logsync barrier |
| 371 |
*/ |
| 372 |
if (test_bit(log_SYNCBARRIER, &log->flag) || |
| 373 |
test_bit(log_QUIESCE, &log->flag)) { |
| 374 |
INCREMENT(TxStat.txBegin_barrier); |
| 375 |
TXN_SLEEP(&log->syncwait); |
| 376 |
goto retry; |
| 377 |
} |
| 378 |
} |
| 379 |
if (flag == 0) { |
| 380 |
/* |
| 381 |
* Don't begin transaction if we're getting starved for tlocks |
| 382 |
* unless COMMIT_FORCE or COMMIT_INODE (which may ultimately |
| 383 |
* free tlocks) |
| 384 |
*/ |
| 385 |
if (TxAnchor.tlocksInUse > TxLockVHWM) { |
| 386 |
INCREMENT(TxStat.txBegin_lockslow); |
| 387 |
TXN_SLEEP(&TxAnchor.lowlockwait); |
| 388 |
goto retry; |
| 389 |
} |
| 390 |
} |
| 391 |
|
| 392 |
/* |
| 393 |
* allocate transaction id/block |
| 394 |
*/ |
| 395 |
if ((t = TxAnchor.freetid) == 0) { |
| 396 |
jfs_info("txBegin: waiting for free tid"); |
| 397 |
INCREMENT(TxStat.txBegin_freetid); |
| 398 |
TXN_SLEEP(&TxAnchor.freewait); |
| 399 |
goto retry; |
| 400 |
} |
| 401 |
|
| 402 |
tblk = tid_to_tblock(t); |
| 403 |
|
| 404 |
if ((tblk->next == 0) && !(flag & COMMIT_FORCE)) { |
| 405 |
/* Don't let a non-forced transaction take the last tblk */ |
| 406 |
jfs_info("txBegin: waiting for free tid"); |
| 407 |
INCREMENT(TxStat.txBegin_freetid); |
| 408 |
TXN_SLEEP(&TxAnchor.freewait); |
| 409 |
goto retry; |
| 410 |
} |
| 411 |
|
| 412 |
TxAnchor.freetid = tblk->next; |
| 413 |
|
| 414 |
/* |
| 415 |
* initialize transaction |
| 416 |
*/ |
| 417 |
|
| 418 |
/* |
| 419 |
* We can't zero the whole thing or we screw up another thread being |
| 420 |
* awakened after sleeping on tblk->waitor |
| 421 |
* |
| 422 |
* memset(tblk, 0, sizeof(struct tblock)); |
| 423 |
*/ |
| 424 |
tblk->next = tblk->last = tblk->xflag = tblk->flag = tblk->lsn = 0; |
| 425 |
|
| 426 |
tblk->sb = sb; |
| 427 |
++log->logtid; |
| 428 |
tblk->logtid = log->logtid; |
| 429 |
|
| 430 |
++log->active; |
| 431 |
|
| 432 |
HIGHWATERMARK(stattx.maxtid, t); /* statistics */ |
| 433 |
INCREMENT(stattx.ntid); /* statistics */ |
| 434 |
|
| 435 |
TXN_UNLOCK(); |
| 436 |
|
| 437 |
jfs_info("txBegin: returning tid = %d", t); |
| 438 |
|
| 439 |
return t; |
| 440 |
} |
| 441 |
|
| 442 |
/* |
| 443 |
* NAME: txBeginAnon() |
| 444 |
* |
| 445 |
* FUNCTION: start an anonymous transaction. |
| 446 |
* Blocks if logsync or available tlocks are low to prevent |
| 447 |
* anonymous tlocks from depleting supply. |
| 448 |
* |
| 449 |
* PARAMETER: sb - superblock |
| 450 |
* |
| 451 |
* RETURN: none |
| 452 |
*/ |
| 453 |
void txBeginAnon(struct super_block *sb) |
| 454 |
{ |
| 455 |
struct jfs_log *log; |
| 456 |
|
| 457 |
log = JFS_SBI(sb)->log; |
| 458 |
|
| 459 |
TXN_LOCK(); |
| 460 |
INCREMENT(TxStat.txBeginAnon); |
| 461 |
|
| 462 |
retry: |
| 463 |
/* |
| 464 |
* synchronize with logsync barrier |
| 465 |
*/ |
| 466 |
if (test_bit(log_SYNCBARRIER, &log->flag) || |
| 467 |
test_bit(log_QUIESCE, &log->flag)) { |
| 468 |
INCREMENT(TxStat.txBeginAnon_barrier); |
| 469 |
TXN_SLEEP(&log->syncwait); |
| 470 |
goto retry; |
| 471 |
} |
| 472 |
|
| 473 |
/* |
| 474 |
* Don't begin transaction if we're getting starved for tlocks |
| 475 |
*/ |
| 476 |
if (TxAnchor.tlocksInUse > TxLockVHWM) { |
| 477 |
INCREMENT(TxStat.txBeginAnon_lockslow); |
| 478 |
TXN_SLEEP(&TxAnchor.lowlockwait); |
| 479 |
goto retry; |
| 480 |
} |
| 481 |
TXN_UNLOCK(); |
| 482 |
} |
| 483 |
|
| 484 |
/* |
| 485 |
* txEnd() |
| 486 |
* |
| 487 |
* function: free specified transaction block. |
| 488 |
* |
| 489 |
* logsync barrier processing: |
| 490 |
* |
| 491 |
* serialization: |
| 492 |
*/ |
| 493 |
void txEnd(tid_t tid) |
| 494 |
{ |
| 495 |
struct tblock *tblk = tid_to_tblock(tid); |
| 496 |
struct jfs_log *log; |
| 497 |
|
| 498 |
jfs_info("txEnd: tid = %d", tid); |
| 499 |
TXN_LOCK(); |
| 500 |
|
| 501 |
/* |
| 502 |
* wakeup transactions waiting on the page locked |
| 503 |
* by the current transaction |
| 504 |
*/ |
| 505 |
TXN_WAKEUP(&tblk->waitor); |
| 506 |
|
| 507 |
log = JFS_SBI(tblk->sb)->log; |
| 508 |
|
| 509 |
/* |
| 510 |
* Lazy commit thread can't free this guy until we mark it UNLOCKED, |
| 511 |
* otherwise, we would be left with a transaction that may have been |
| 512 |
* reused. |
| 513 |
* |
| 514 |
* Lazy commit thread will turn off tblkGC_LAZY before calling this |
| 515 |
* routine. |
| 516 |
*/ |
| 517 |
if (tblk->flag & tblkGC_LAZY) { |
| 518 |
jfs_info("txEnd called w/lazy tid: %d, tblk = 0x%p", tid, tblk); |
| 519 |
TXN_UNLOCK(); |
| 520 |
|
| 521 |
spin_lock_irq(&log->gclock); // LOGGC_LOCK |
| 522 |
tblk->flag |= tblkGC_UNLOCKED; |
| 523 |
spin_unlock_irq(&log->gclock); // LOGGC_UNLOCK |
| 524 |
return; |
| 525 |
} |
| 526 |
|
| 527 |
jfs_info("txEnd: tid: %d, tblk = 0x%p", tid, tblk); |
| 528 |
|
| 529 |
assert(tblk->next == 0); |
| 530 |
|
| 531 |
/* |
| 532 |
* insert tblock back on freelist |
| 533 |
*/ |
| 534 |
tblk->next = TxAnchor.freetid; |
| 535 |
TxAnchor.freetid = tid; |
| 536 |
|
| 537 |
/* |
| 538 |
* mark the tblock not active |
| 539 |
*/ |
| 540 |
if (--log->active == 0) { |
| 541 |
clear_bit(log_FLUSH, &log->flag); |
| 542 |
|
| 543 |
/* |
| 544 |
* synchronize with logsync barrier |
| 545 |
*/ |
| 546 |
if (test_bit(log_SYNCBARRIER, &log->flag)) { |
| 547 |
TXN_UNLOCK(); |
| 548 |
|
| 549 |
/* write dirty metadata & forward log syncpt */ |
| 550 |
jfs_syncpt(log, 1); |
| 551 |
|
| 552 |
jfs_info("log barrier off: 0x%x", log->lsn); |
| 553 |
|
| 554 |
/* enable new transactions start */ |
| 555 |
clear_bit(log_SYNCBARRIER, &log->flag); |
| 556 |
|
| 557 |
/* wakeup all waitors for logsync barrier */ |
| 558 |
TXN_WAKEUP(&log->syncwait); |
| 559 |
|
| 560 |
goto wakeup; |
| 561 |
} |
| 562 |
} |
| 563 |
|
| 564 |
TXN_UNLOCK(); |
| 565 |
wakeup: |
| 566 |
/* |
| 567 |
* wakeup all waitors for a free tblock |
| 568 |
*/ |
| 569 |
TXN_WAKEUP(&TxAnchor.freewait); |
| 570 |
} |
| 571 |
|
| 572 |
/* |
| 573 |
* txLock() |
| 574 |
* |
| 575 |
* function: acquire a transaction lock on the specified <mp> |
| 576 |
* |
| 577 |
* parameter: |
| 578 |
* |
| 579 |
* return: transaction lock id |
| 580 |
* |
| 581 |
* serialization: |
| 582 |
*/ |
| 583 |
struct tlock *txLock(tid_t tid, struct inode *ip, struct metapage * mp, |
| 584 |
int type) |
| 585 |
{ |
| 586 |
struct jfs_inode_info *jfs_ip = JFS_IP(ip); |
| 587 |
int dir_xtree = 0; |
| 588 |
lid_t lid; |
| 589 |
tid_t xtid; |
| 590 |
struct tlock *tlck; |
| 591 |
struct xtlock *xtlck; |
| 592 |
struct linelock *linelock; |
| 593 |
xtpage_t *p; |
| 594 |
struct tblock *tblk; |
| 595 |
|
| 596 |
TXN_LOCK(); |
| 597 |
|
| 598 |
if (S_ISDIR(ip->i_mode) && (type & tlckXTREE) && |
| 599 |
!(mp->xflag & COMMIT_PAGE)) { |
| 600 |
/* |
| 601 |
* Directory inode is special. It can have both an xtree tlock |
| 602 |
* and a dtree tlock associated with it. |
| 603 |
*/ |
| 604 |
dir_xtree = 1; |
| 605 |
lid = jfs_ip->xtlid; |
| 606 |
} else |
| 607 |
lid = mp->lid; |
| 608 |
|
| 609 |
/* is page not locked by a transaction ? */ |
| 610 |
if (lid == 0) |
| 611 |
goto allocateLock; |
| 612 |
|
| 613 |
jfs_info("txLock: tid:%d ip:0x%p mp:0x%p lid:%d", tid, ip, mp, lid); |
| 614 |
|
| 615 |
/* is page locked by the requester transaction ? */ |
| 616 |
tlck = lid_to_tlock(lid); |
| 617 |
if ((xtid = tlck->tid) == tid) { |
| 618 |
TXN_UNLOCK(); |
| 619 |
goto grantLock; |
| 620 |
} |
| 621 |
|
| 622 |
/* |
| 623 |
* is page locked by anonymous transaction/lock ? |
| 624 |
* |
| 625 |
* (page update without transaction (i.e., file write) is |
| 626 |
* locked under anonymous transaction tid = 0: |
| 627 |
* anonymous tlocks maintained on anonymous tlock list of |
| 628 |
* the inode of the page and available to all anonymous |
| 629 |
* transactions until txCommit() time at which point |
| 630 |
* they are transferred to the transaction tlock list of |
| 631 |
* the committing transaction of the inode) |
| 632 |
*/ |
| 633 |
if (xtid == 0) { |
| 634 |
tlck->tid = tid; |
| 635 |
TXN_UNLOCK(); |
| 636 |
tblk = tid_to_tblock(tid); |
| 637 |
/* |
| 638 |
* The order of the tlocks in the transaction is important |
| 639 |
* (during truncate, child xtree pages must be freed before |
| 640 |
* parent's tlocks change the working map). |
| 641 |
* Take tlock off anonymous list and add to tail of |
| 642 |
* transaction list |
| 643 |
* |
| 644 |
* Note: We really need to get rid of the tid & lid and |
| 645 |
* use list_head's. This code is getting UGLY! |
| 646 |
*/ |
| 647 |
if (jfs_ip->atlhead == lid) { |
| 648 |
if (jfs_ip->atltail == lid) { |
| 649 |
/* only anonymous txn. |
| 650 |
* Remove from anon_list |
| 651 |
*/ |
| 652 |
TXN_LOCK(); |
| 653 |
list_del_init(&jfs_ip->anon_inode_list); |
| 654 |
TXN_UNLOCK(); |
| 655 |
} |
| 656 |
jfs_ip->atlhead = tlck->next; |
| 657 |
} else { |
| 658 |
lid_t last; |
| 659 |
for (last = jfs_ip->atlhead; |
| 660 |
lid_to_tlock(last)->next != lid; |
| 661 |
last = lid_to_tlock(last)->next) { |
| 662 |
assert(last); |
| 663 |
} |
| 664 |
lid_to_tlock(last)->next = tlck->next; |
| 665 |
if (jfs_ip->atltail == lid) |
| 666 |
jfs_ip->atltail = last; |
| 667 |
} |
| 668 |
|
| 669 |
/* insert the tlock at tail of transaction tlock list */ |
| 670 |
|
| 671 |
if (tblk->next) |
| 672 |
lid_to_tlock(tblk->last)->next = lid; |
| 673 |
else |
| 674 |
tblk->next = lid; |
| 675 |
tlck->next = 0; |
| 676 |
tblk->last = lid; |
| 677 |
|
| 678 |
goto grantLock; |
| 679 |
} |
| 680 |
|
| 681 |
goto waitLock; |
| 682 |
|
| 683 |
/* |
| 684 |
* allocate a tlock |
| 685 |
*/ |
| 686 |
allocateLock: |
| 687 |
lid = txLockAlloc(); |
| 688 |
tlck = lid_to_tlock(lid); |
| 689 |
|
| 690 |
/* |
| 691 |
* initialize tlock |
| 692 |
*/ |
| 693 |
tlck->tid = tid; |
| 694 |
|
| 695 |
TXN_UNLOCK(); |
| 696 |
|
| 697 |
/* mark tlock for meta-data page */ |
| 698 |
if (mp->xflag & COMMIT_PAGE) { |
| 699 |
|
| 700 |
tlck->flag = tlckPAGELOCK; |
| 701 |
|
| 702 |
/* mark the page dirty and nohomeok */ |
| 703 |
metapage_nohomeok(mp); |
| 704 |
|
| 705 |
jfs_info("locking mp = 0x%p, nohomeok = %d tid = %d tlck = 0x%p", |
| 706 |
mp, mp->nohomeok, tid, tlck); |
| 707 |
|
| 708 |
/* if anonymous transaction, and buffer is on the group |
| 709 |
* commit synclist, mark inode to show this. This will |
| 710 |
* prevent the buffer from being marked nohomeok for too |
| 711 |
* long a time. |
| 712 |
*/ |
| 713 |
if ((tid == 0) && mp->lsn) |
| 714 |
set_cflag(COMMIT_Synclist, ip); |
| 715 |
} |
| 716 |
/* mark tlock for in-memory inode */ |
| 717 |
else |
| 718 |
tlck->flag = tlckINODELOCK; |
| 719 |
|
| 720 |
if (S_ISDIR(ip->i_mode)) |
| 721 |
tlck->flag |= tlckDIRECTORY; |
| 722 |
|
| 723 |
tlck->type = 0; |
| 724 |
|
| 725 |
/* bind the tlock and the page */ |
| 726 |
tlck->ip = ip; |
| 727 |
tlck->mp = mp; |
| 728 |
if (dir_xtree) |
| 729 |
jfs_ip->xtlid = lid; |
| 730 |
else |
| 731 |
mp->lid = lid; |
| 732 |
|
| 733 |
/* |
| 734 |
* enqueue transaction lock to transaction/inode |
| 735 |
*/ |
| 736 |
/* insert the tlock at tail of transaction tlock list */ |
| 737 |
if (tid) { |
| 738 |
tblk = tid_to_tblock(tid); |
| 739 |
if (tblk->next) |
| 740 |
lid_to_tlock(tblk->last)->next = lid; |
| 741 |
else |
| 742 |
tblk->next = lid; |
| 743 |
tlck->next = 0; |
| 744 |
tblk->last = lid; |
| 745 |
} |
| 746 |
/* anonymous transaction: |
| 747 |
* insert the tlock at head of inode anonymous tlock list |
| 748 |
*/ |
| 749 |
else { |
| 750 |
tlck->next = jfs_ip->atlhead; |
| 751 |
jfs_ip->atlhead = lid; |
| 752 |
if (tlck->next == 0) { |
| 753 |
/* This inode's first anonymous transaction */ |
| 754 |
jfs_ip->atltail = lid; |
| 755 |
TXN_LOCK(); |
| 756 |
list_add_tail(&jfs_ip->anon_inode_list, |
| 757 |
&TxAnchor.anon_list); |
| 758 |
TXN_UNLOCK(); |
| 759 |
} |
| 760 |
} |
| 761 |
|
| 762 |
/* initialize type dependent area for linelock */ |
| 763 |
linelock = (struct linelock *) & tlck->lock; |
| 764 |
linelock->next = 0; |
| 765 |
linelock->flag = tlckLINELOCK; |
| 766 |
linelock->maxcnt = TLOCKSHORT; |
| 767 |
linelock->index = 0; |
| 768 |
|
| 769 |
switch (type & tlckTYPE) { |
| 770 |
case tlckDTREE: |
| 771 |
linelock->l2linesize = L2DTSLOTSIZE; |
| 772 |
break; |
| 773 |
|
| 774 |
case tlckXTREE: |
| 775 |
linelock->l2linesize = L2XTSLOTSIZE; |
| 776 |
|
| 777 |
xtlck = (struct xtlock *) linelock; |
| 778 |
xtlck->header.offset = 0; |
| 779 |
xtlck->header.length = 2; |
| 780 |
|
| 781 |
if (type & tlckNEW) { |
| 782 |
xtlck->lwm.offset = XTENTRYSTART; |
| 783 |
} else { |
| 784 |
if (mp->xflag & COMMIT_PAGE) |
| 785 |
p = (xtpage_t *) mp->data; |
| 786 |
else |
| 787 |
p = (xtpage_t *) &jfs_ip->i_xtroot; |
| 788 |
xtlck->lwm.offset = |
| 789 |
le16_to_cpu(p->header.nextindex); |
| 790 |
} |
| 791 |
xtlck->lwm.length = 0; /* ! */ |
| 792 |
xtlck->twm.offset = 0; |
| 793 |
xtlck->hwm.offset = 0; |
| 794 |
|
| 795 |
xtlck->index = 2; |
| 796 |
break; |
| 797 |
|
| 798 |
case tlckINODE: |
| 799 |
linelock->l2linesize = L2INODESLOTSIZE; |
| 800 |
break; |
| 801 |
|
| 802 |
case tlckDATA: |
| 803 |
linelock->l2linesize = L2DATASLOTSIZE; |
| 804 |
break; |
| 805 |
|
| 806 |
default: |
| 807 |
jfs_err("UFO tlock:0x%p", tlck); |
| 808 |
} |
| 809 |
|
| 810 |
/* |
| 811 |
* update tlock vector |
| 812 |
*/ |
| 813 |
grantLock: |
| 814 |
tlck->type |= type; |
| 815 |
|
| 816 |
return tlck; |
| 817 |
|
| 818 |
/* |
| 819 |
* page is being locked by another transaction: |
| 820 |
*/ |
| 821 |
waitLock: |
| 822 |
/* Only locks on ipimap or ipaimap should reach here */ |
| 823 |
/* assert(jfs_ip->fileset == AGGREGATE_I); */ |
| 824 |
if (jfs_ip->fileset != AGGREGATE_I) { |
| 825 |
printk(KERN_ERR "txLock: trying to lock locked page!"); |
| 826 |
print_hex_dump(KERN_ERR, "ip: ", DUMP_PREFIX_ADDRESS, 16, 4, |
| 827 |
ip, sizeof(*ip), 0); |
| 828 |
print_hex_dump(KERN_ERR, "mp: ", DUMP_PREFIX_ADDRESS, 16, 4, |
| 829 |
mp, sizeof(*mp), 0); |
| 830 |
print_hex_dump(KERN_ERR, "Locker's tblock: ", |
| 831 |
DUMP_PREFIX_ADDRESS, 16, 4, tid_to_tblock(tid), |
| 832 |
sizeof(struct tblock), 0); |
| 833 |
print_hex_dump(KERN_ERR, "Tlock: ", DUMP_PREFIX_ADDRESS, 16, 4, |
| 834 |
tlck, sizeof(*tlck), 0); |
| 835 |
BUG(); |
| 836 |
} |
| 837 |
INCREMENT(stattx.waitlock); /* statistics */ |
| 838 |
TXN_UNLOCK(); |
| 839 |
release_metapage(mp); |
| 840 |
TXN_LOCK(); |
| 841 |
xtid = tlck->tid; /* reacquire after dropping TXN_LOCK */ |
| 842 |
|
| 843 |
jfs_info("txLock: in waitLock, tid = %d, xtid = %d, lid = %d", |
| 844 |
tid, xtid, lid); |
| 845 |
|
| 846 |
/* Recheck everything since dropping TXN_LOCK */ |
| 847 |
if (xtid && (tlck->mp == mp) && (mp->lid == lid)) |
| 848 |
TXN_SLEEP_DROP_LOCK(&tid_to_tblock(xtid)->waitor); |
| 849 |
else |
| 850 |
TXN_UNLOCK(); |
| 851 |
jfs_info("txLock: awakened tid = %d, lid = %d", tid, lid); |
| 852 |
|
| 853 |
return NULL; |
| 854 |
} |
| 855 |
|
| 856 |
/* |
| 857 |
* NAME: txRelease() |
| 858 |
* |
| 859 |
* FUNCTION: Release buffers associated with transaction locks, but don't |
| 860 |
* mark homeok yet. The allows other transactions to modify |
| 861 |
* buffers, but won't let them go to disk until commit record |
| 862 |
* actually gets written. |
| 863 |
* |
| 864 |
* PARAMETER: |
| 865 |
* tblk - |
| 866 |
* |
| 867 |
* RETURN: Errors from subroutines. |
| 868 |
*/ |
| 869 |
static void txRelease(struct tblock * tblk) |
| 870 |
{ |
| 871 |
struct metapage *mp; |
| 872 |
lid_t lid; |
| 873 |
struct tlock *tlck; |
| 874 |
|
| 875 |
TXN_LOCK(); |
| 876 |
|
| 877 |
for (lid = tblk->next; lid; lid = tlck->next) { |
| 878 |
tlck = lid_to_tlock(lid); |
| 879 |
if ((mp = tlck->mp) != NULL && |
| 880 |
(tlck->type & tlckBTROOT) == 0) { |
| 881 |
assert(mp->xflag & COMMIT_PAGE); |
| 882 |
mp->lid = 0; |
| 883 |
} |
| 884 |
} |
| 885 |
|
| 886 |
/* |
| 887 |
* wakeup transactions waiting on a page locked |
| 888 |
* by the current transaction |
| 889 |
*/ |
| 890 |
TXN_WAKEUP(&tblk->waitor); |
| 891 |
|
| 892 |
TXN_UNLOCK(); |
| 893 |
} |
| 894 |
|
| 895 |
/* |
| 896 |
* NAME: txUnlock() |
| 897 |
* |
| 898 |
* FUNCTION: Initiates pageout of pages modified by tid in journalled |
| 899 |
* objects and frees their lockwords. |
| 900 |
*/ |
| 901 |
static void txUnlock(struct tblock * tblk) |
| 902 |
{ |
| 903 |
struct tlock *tlck; |
| 904 |
struct linelock *linelock; |
| 905 |
lid_t lid, next, llid, k; |
| 906 |
struct metapage *mp; |
| 907 |
struct jfs_log *log; |
| 908 |
int difft, diffp; |
| 909 |
unsigned long flags; |
| 910 |
|
| 911 |
jfs_info("txUnlock: tblk = 0x%p", tblk); |
| 912 |
log = JFS_SBI(tblk->sb)->log; |
| 913 |
|
| 914 |
/* |
| 915 |
* mark page under tlock homeok (its log has been written): |
| 916 |
*/ |
| 917 |
for (lid = tblk->next; lid; lid = next) { |
| 918 |
tlck = lid_to_tlock(lid); |
| 919 |
next = tlck->next; |
| 920 |
|
| 921 |
jfs_info("unlocking lid = %d, tlck = 0x%p", lid, tlck); |
| 922 |
|
| 923 |
/* unbind page from tlock */ |
| 924 |
if ((mp = tlck->mp) != NULL && |
| 925 |
(tlck->type & tlckBTROOT) == 0) { |
| 926 |
assert(mp->xflag & COMMIT_PAGE); |
| 927 |
|
| 928 |
/* hold buffer |
| 929 |
*/ |
| 930 |
hold_metapage(mp); |
| 931 |
|
| 932 |
assert(mp->nohomeok > 0); |
| 933 |
_metapage_homeok(mp); |
| 934 |
|
| 935 |
/* inherit younger/larger clsn */ |
| 936 |
LOGSYNC_LOCK(log, flags); |
| 937 |
if (mp->clsn) { |
| 938 |
logdiff(difft, tblk->clsn, log); |
| 939 |
logdiff(diffp, mp->clsn, log); |
| 940 |
if (difft > diffp) |
| 941 |
mp->clsn = tblk->clsn; |
| 942 |
} else |
| 943 |
mp->clsn = tblk->clsn; |
| 944 |
LOGSYNC_UNLOCK(log, flags); |
| 945 |
|
| 946 |
assert(!(tlck->flag & tlckFREEPAGE)); |
| 947 |
|
| 948 |
put_metapage(mp); |
| 949 |
} |
| 950 |
|
| 951 |
/* insert tlock, and linelock(s) of the tlock if any, |
| 952 |
* at head of freelist |
| 953 |
*/ |
| 954 |
TXN_LOCK(); |
| 955 |
|
| 956 |
llid = ((struct linelock *) & tlck->lock)->next; |
| 957 |
while (llid) { |
| 958 |
linelock = (struct linelock *) lid_to_tlock(llid); |
| 959 |
k = linelock->next; |
| 960 |
txLockFree(llid); |
| 961 |
llid = k; |
| 962 |
} |
| 963 |
txLockFree(lid); |
| 964 |
|
| 965 |
TXN_UNLOCK(); |
| 966 |
} |
| 967 |
tblk->next = tblk->last = 0; |
| 968 |
|
| 969 |
/* |
| 970 |
* remove tblock from logsynclist |
| 971 |
* (allocation map pages inherited lsn of tblk and |
| 972 |
* has been inserted in logsync list at txUpdateMap()) |
| 973 |
*/ |
| 974 |
if (tblk->lsn) { |
| 975 |
LOGSYNC_LOCK(log, flags); |
| 976 |
log->count--; |
| 977 |
list_del(&tblk->synclist); |
| 978 |
LOGSYNC_UNLOCK(log, flags); |
| 979 |
} |
| 980 |
} |
| 981 |
|
| 982 |
/* |
| 983 |
* txMaplock() |
| 984 |
* |
| 985 |
* function: allocate a transaction lock for freed page/entry; |
| 986 |
* for freed page, maplock is used as xtlock/dtlock type; |
| 987 |
*/ |
| 988 |
struct tlock *txMaplock(tid_t tid, struct inode *ip, int type) |
| 989 |
{ |
| 990 |
struct jfs_inode_info *jfs_ip = JFS_IP(ip); |
| 991 |
lid_t lid; |
| 992 |
struct tblock *tblk; |
| 993 |
struct tlock *tlck; |
| 994 |
struct maplock *maplock; |
| 995 |
|
| 996 |
TXN_LOCK(); |
| 997 |
|
| 998 |
/* |
| 999 |
* allocate a tlock |
| 1000 |
*/ |
| 1001 |
lid = txLockAlloc(); |
| 1002 |
tlck = lid_to_tlock(lid); |
| 1003 |
|
| 1004 |
/* |
| 1005 |
* initialize tlock |
| 1006 |
*/ |
| 1007 |
tlck->tid = tid; |
| 1008 |
|
| 1009 |
/* bind the tlock and the object */ |
| 1010 |
tlck->flag = tlckINODELOCK; |
| 1011 |
if (S_ISDIR(ip->i_mode)) |
| 1012 |
tlck->flag |= tlckDIRECTORY; |
| 1013 |
tlck->ip = ip; |
| 1014 |
tlck->mp = NULL; |
| 1015 |
|
| 1016 |
tlck->type = type; |
| 1017 |
|
| 1018 |
/* |
| 1019 |
* enqueue transaction lock to transaction/inode |
| 1020 |
*/ |
| 1021 |
/* insert the tlock at tail of transaction tlock list */ |
| 1022 |
if (tid) { |
| 1023 |
tblk = tid_to_tblock(tid); |
| 1024 |
if (tblk->next) |
| 1025 |
lid_to_tlock(tblk->last)->next = lid; |
| 1026 |
else |
| 1027 |
tblk->next = lid; |
| 1028 |
tlck->next = 0; |
| 1029 |
tblk->last = lid; |
| 1030 |
} |
| 1031 |
/* anonymous transaction: |
| 1032 |
* insert the tlock at head of inode anonymous tlock list |
| 1033 |
*/ |
| 1034 |
else { |
| 1035 |
tlck->next = jfs_ip->atlhead; |
| 1036 |
jfs_ip->atlhead = lid; |
| 1037 |
if (tlck->next == 0) { |
| 1038 |
/* This inode's first anonymous transaction */ |
| 1039 |
jfs_ip->atltail = lid; |
| 1040 |
list_add_tail(&jfs_ip->anon_inode_list, |
| 1041 |
&TxAnchor.anon_list); |
| 1042 |
} |
| 1043 |
} |
| 1044 |
|
| 1045 |
TXN_UNLOCK(); |
| 1046 |
|
| 1047 |
/* initialize type dependent area for maplock */ |
| 1048 |
maplock = (struct maplock *) & tlck->lock; |
| 1049 |
maplock->next = 0; |
| 1050 |
maplock->maxcnt = 0; |
| 1051 |
maplock->index = 0; |
| 1052 |
|
| 1053 |
return tlck; |
| 1054 |
} |
| 1055 |
|
| 1056 |
/* |
| 1057 |
* txLinelock() |
| 1058 |
* |
| 1059 |
* function: allocate a transaction lock for log vector list |
| 1060 |
*/ |
| 1061 |
struct linelock *txLinelock(struct linelock * tlock) |
| 1062 |
{ |
| 1063 |
lid_t lid; |
| 1064 |
struct tlock *tlck; |
| 1065 |
struct linelock *linelock; |
| 1066 |
|
| 1067 |
TXN_LOCK(); |
| 1068 |
|
| 1069 |
/* allocate a TxLock structure */ |
| 1070 |
lid = txLockAlloc(); |
| 1071 |
tlck = lid_to_tlock(lid); |
| 1072 |
|
| 1073 |
TXN_UNLOCK(); |
| 1074 |
|
| 1075 |
/* initialize linelock */ |
| 1076 |
linelock = (struct linelock *) tlck; |
| 1077 |
linelock->next = 0; |
| 1078 |
linelock->flag = tlckLINELOCK; |
| 1079 |
linelock->maxcnt = TLOCKLONG; |
| 1080 |
linelock->index = 0; |
| 1081 |
if (tlck->flag & tlckDIRECTORY) |
| 1082 |
linelock->flag |= tlckDIRECTORY; |
| 1083 |
|
| 1084 |
/* append linelock after tlock */ |
| 1085 |
linelock->next = tlock->next; |
| 1086 |
tlock->next = lid; |
| 1087 |
|
| 1088 |
return linelock; |
| 1089 |
} |
| 1090 |
|
| 1091 |
/* |
| 1092 |
* transaction commit management |
| 1093 |
* ----------------------------- |
| 1094 |
*/ |
| 1095 |
|
| 1096 |
/* |
| 1097 |
* NAME: txCommit() |
| 1098 |
* |
| 1099 |
* FUNCTION: commit the changes to the objects specified in |
| 1100 |
* clist. For journalled segments only the |
| 1101 |
* changes of the caller are committed, ie by tid. |
| 1102 |
* for non-journalled segments the data are flushed to |
| 1103 |
* disk and then the change to the disk inode and indirect |
| 1104 |
* blocks committed (so blocks newly allocated to the |
| 1105 |
* segment will be made a part of the segment atomically). |
| 1106 |
* |
| 1107 |
* all of the segments specified in clist must be in |
| 1108 |
* one file system. no more than 6 segments are needed |
| 1109 |
* to handle all unix svcs. |
| 1110 |
* |
| 1111 |
* if the i_nlink field (i.e. disk inode link count) |
| 1112 |
* is zero, and the type of inode is a regular file or |
| 1113 |
* directory, or symbolic link , the inode is truncated |
| 1114 |
* to zero length. the truncation is committed but the |
| 1115 |
* VM resources are unaffected until it is closed (see |
| 1116 |
* iput and iclose). |
| 1117 |
* |
| 1118 |
* PARAMETER: |
| 1119 |
* |
| 1120 |
* RETURN: |
| 1121 |
* |
| 1122 |
* serialization: |
| 1123 |
* on entry the inode lock on each segment is assumed |
| 1124 |
* to be held. |
| 1125 |
* |
| 1126 |
* i/o error: |
| 1127 |
*/ |
| 1128 |
int txCommit(tid_t tid, /* transaction identifier */ |
| 1129 |
int nip, /* number of inodes to commit */ |
| 1130 |
struct inode **iplist, /* list of inode to commit */ |
| 1131 |
int flag) |
| 1132 |
{ |
| 1133 |
int rc = 0; |
| 1134 |
struct commit cd; |
| 1135 |
struct jfs_log *log; |
| 1136 |
struct tblock *tblk; |
| 1137 |
struct lrd *lrd; |
| 1138 |
struct inode *ip; |
| 1139 |
struct jfs_inode_info *jfs_ip; |
| 1140 |
int k, n; |
| 1141 |
ino_t top; |
| 1142 |
struct super_block *sb; |
| 1143 |
|
| 1144 |
jfs_info("txCommit, tid = %d, flag = %d", tid, flag); |
| 1145 |
/* is read-only file system ? */ |
| 1146 |
if (isReadOnly(iplist[0])) { |
| 1147 |
rc = -EROFS; |
| 1148 |
goto TheEnd; |
| 1149 |
} |
| 1150 |
|
| 1151 |
sb = cd.sb = iplist[0]->i_sb; |
| 1152 |
cd.tid = tid; |
| 1153 |
|
| 1154 |
if (tid == 0) |
| 1155 |
tid = txBegin(sb, 0); |
| 1156 |
tblk = tid_to_tblock(tid); |
| 1157 |
|
| 1158 |
/* |
| 1159 |
* initialize commit structure |
| 1160 |
*/ |
| 1161 |
log = JFS_SBI(sb)->log; |
| 1162 |
cd.log = log; |
| 1163 |
|
| 1164 |
/* initialize log record descriptor in commit */ |
| 1165 |
lrd = &cd.lrd; |
| 1166 |
lrd->logtid = cpu_to_le32(tblk->logtid); |
| 1167 |
lrd->backchain = 0; |
| 1168 |
|
| 1169 |
tblk->xflag |= flag; |
| 1170 |
|
| 1171 |
if ((flag & (COMMIT_FORCE | COMMIT_SYNC)) == 0) |
| 1172 |
tblk->xflag |= COMMIT_LAZY; |
| 1173 |
/* |
| 1174 |
* prepare non-journaled objects for commit |
| 1175 |
* |
| 1176 |
* flush data pages of non-journaled file |
| 1177 |
* to prevent the file getting non-initialized disk blocks |
| 1178 |
* in case of crash. |
| 1179 |
* (new blocks - ) |
| 1180 |
*/ |
| 1181 |
cd.iplist = iplist; |
| 1182 |
cd.nip = nip; |
| 1183 |
|
| 1184 |
/* |
| 1185 |
* acquire transaction lock on (on-disk) inodes |
| 1186 |
* |
| 1187 |
* update on-disk inode from in-memory inode |
| 1188 |
* acquiring transaction locks for AFTER records |
| 1189 |
* on the on-disk inode of file object |
| 1190 |
* |
| 1191 |
* sort the inodes array by inode number in descending order |
| 1192 |
* to prevent deadlock when acquiring transaction lock |
| 1193 |
* of on-disk inodes on multiple on-disk inode pages by |
| 1194 |
* multiple concurrent transactions |
| 1195 |
*/ |
| 1196 |
for (k = 0; k < cd.nip; k++) { |
| 1197 |
top = (cd.iplist[k])->i_ino; |
| 1198 |
for (n = k + 1; n < cd.nip; n++) { |
| 1199 |
ip = cd.iplist[n]; |
| 1200 |
if (ip->i_ino > top) { |
| 1201 |
top = ip->i_ino; |
| 1202 |
cd.iplist[n] = cd.iplist[k]; |
| 1203 |
cd.iplist[k] = ip; |
| 1204 |
} |
| 1205 |
} |
| 1206 |
|
| 1207 |
ip = cd.iplist[k]; |
| 1208 |
jfs_ip = JFS_IP(ip); |
| 1209 |
|
| 1210 |
/* |
| 1211 |
* BUGBUG - This code has temporarily been removed. The |
| 1212 |
* intent is to ensure that any file data is written before |
| 1213 |
* the metadata is committed to the journal. This prevents |
| 1214 |
* uninitialized data from appearing in a file after the |
| 1215 |
* journal has been replayed. (The uninitialized data |
| 1216 |
* could be sensitive data removed by another user.) |
| 1217 |
* |
| 1218 |
* The problem now is that we are holding the IWRITELOCK |
| 1219 |
* on the inode, and calling filemap_fdatawrite on an |
| 1220 |
* unmapped page will cause a deadlock in jfs_get_block. |
| 1221 |
* |
| 1222 |
* The long term solution is to pare down the use of |
| 1223 |
* IWRITELOCK. We are currently holding it too long. |
| 1224 |
* We could also be smarter about which data pages need |
| 1225 |
* to be written before the transaction is committed and |
| 1226 |
* when we don't need to worry about it at all. |
| 1227 |
* |
| 1228 |
* if ((!S_ISDIR(ip->i_mode)) |
| 1229 |
* && (tblk->flag & COMMIT_DELETE) == 0) |
| 1230 |
* filemap_write_and_wait(ip->i_mapping); |
| 1231 |
*/ |
| 1232 |
|
| 1233 |
/* |
| 1234 |
* Mark inode as not dirty. It will still be on the dirty |
| 1235 |
* inode list, but we'll know not to commit it again unless |
| 1236 |
* it gets marked dirty again |
| 1237 |
*/ |
| 1238 |
clear_cflag(COMMIT_Dirty, ip); |
| 1239 |
|
| 1240 |
/* inherit anonymous tlock(s) of inode */ |
| 1241 |
if (jfs_ip->atlhead) { |
| 1242 |
lid_to_tlock(jfs_ip->atltail)->next = tblk->next; |
| 1243 |
tblk->next = jfs_ip->atlhead; |
| 1244 |
if (!tblk->last) |
| 1245 |
tblk->last = jfs_ip->atltail; |
| 1246 |
jfs_ip->atlhead = jfs_ip->atltail = 0; |
| 1247 |
TXN_LOCK(); |
| 1248 |
list_del_init(&jfs_ip->anon_inode_list); |
| 1249 |
TXN_UNLOCK(); |
| 1250 |
} |
| 1251 |
|
| 1252 |
/* |
| 1253 |
* acquire transaction lock on on-disk inode page |
| 1254 |
* (become first tlock of the tblk's tlock list) |
| 1255 |
*/ |
| 1256 |
if (((rc = diWrite(tid, ip)))) |
| 1257 |
goto out; |
| 1258 |
} |
| 1259 |
|
| 1260 |
/* |
| 1261 |
* write log records from transaction locks |
| 1262 |
* |
| 1263 |
* txUpdateMap() resets XAD_NEW in XAD. |
| 1264 |
*/ |
| 1265 |
txLog(log, tblk, &cd); |
| 1266 |
|
| 1267 |
/* |
| 1268 |
* Ensure that inode isn't reused before |
| 1269 |
* lazy commit thread finishes processing |
| 1270 |
*/ |
| 1271 |
if (tblk->xflag & COMMIT_DELETE) { |
| 1272 |
ihold(tblk->u.ip); |
| 1273 |
/* |
| 1274 |
* Avoid a rare deadlock |
| 1275 |
* |
| 1276 |
* If the inode is locked, we may be blocked in |
| 1277 |
* jfs_commit_inode. If so, we don't want the |
| 1278 |
* lazy_commit thread doing the last iput() on the inode |
| 1279 |
* since that may block on the locked inode. Instead, |
| 1280 |
* commit the transaction synchronously, so the last iput |
| 1281 |
* will be done by the calling thread (or later) |
| 1282 |
*/ |
| 1283 |
/* |
| 1284 |
* I believe this code is no longer needed. Splitting I_LOCK |
| 1285 |
* into two bits, I_NEW and I_SYNC should prevent this |
| 1286 |
* deadlock as well. But since I don't have a JFS testload |
| 1287 |
* to verify this, only a trivial s/I_LOCK/I_SYNC/ was done. |
| 1288 |
* Joern |
| 1289 |
*/ |
| 1290 |
if (inode_state_read_once(tblk->u.ip) & I_SYNC) |
| 1291 |
tblk->xflag &= ~COMMIT_LAZY; |
| 1292 |
} |
| 1293 |
|
| 1294 |
ASSERT((!(tblk->xflag & COMMIT_DELETE)) || |
| 1295 |
((tblk->u.ip->i_nlink == 0) && |
| 1296 |
!test_cflag(COMMIT_Nolink, tblk->u.ip))); |
| 1297 |
|
| 1298 |
/* |
| 1299 |
* write COMMIT log record |
| 1300 |
*/ |
| 1301 |
lrd->type = cpu_to_le16(LOG_COMMIT); |
| 1302 |
lrd->length = 0; |
| 1303 |
lmLog(log, tblk, lrd, NULL); |
| 1304 |
|
| 1305 |
lmGroupCommit(log, tblk); |
| 1306 |
|
| 1307 |
/* |
| 1308 |
* - transaction is now committed - |
| 1309 |
*/ |
| 1310 |
|
| 1311 |
/* |
| 1312 |
* force pages in careful update |
| 1313 |
* (imap addressing structure update) |
| 1314 |
*/ |
| 1315 |
if (flag & COMMIT_FORCE) |
| 1316 |
txForce(tblk); |
| 1317 |
|
| 1318 |
/* |
| 1319 |
* update allocation map. |
| 1320 |
* |
| 1321 |
* update inode allocation map and inode: |
| 1322 |
* free pager lock on memory object of inode if any. |
| 1323 |
* update block allocation map. |
| 1324 |
* |
| 1325 |
* txUpdateMap() resets XAD_NEW in XAD. |
| 1326 |
*/ |
| 1327 |
if (tblk->xflag & COMMIT_FORCE) |
| 1328 |
txUpdateMap(tblk); |
| 1329 |
|
| 1330 |
/* |
| 1331 |
* free transaction locks and pageout/free pages |
| 1332 |
*/ |
| 1333 |
txRelease(tblk); |
| 1334 |
|
| 1335 |
if ((tblk->flag & tblkGC_LAZY) == 0) |
| 1336 |
txUnlock(tblk); |
| 1337 |
|
| 1338 |
|
| 1339 |
/* |
| 1340 |
* reset in-memory object state |
| 1341 |
*/ |
| 1342 |
for (k = 0; k < cd.nip; k++) { |
| 1343 |
ip = cd.iplist[k]; |
| 1344 |
jfs_ip = JFS_IP(ip); |
| 1345 |
|
| 1346 |
/* |
| 1347 |
* reset in-memory inode state |
| 1348 |
*/ |
| 1349 |
jfs_ip->bxflag = 0; |
| 1350 |
jfs_ip->blid = 0; |
| 1351 |
} |
| 1352 |
|
| 1353 |
out: |
| 1354 |
if (rc != 0) |
| 1355 |
txAbort(tid, 1); |
| 1356 |
|
| 1357 |
TheEnd: |
| 1358 |
jfs_info("txCommit: tid = %d, returning %d", tid, rc); |
| 1359 |
return rc; |
| 1360 |
} |
| 1361 |
|
| 1362 |
/* |
| 1363 |
* NAME: txLog() |
| 1364 |
* |
| 1365 |
* FUNCTION: Writes AFTER log records for all lines modified |
| 1366 |
* by tid for segments specified by inodes in comdata. |
| 1367 |
* Code assumes only WRITELOCKS are recorded in lockwords. |
| 1368 |
* |
| 1369 |
* PARAMETERS: |
| 1370 |
* |
| 1371 |
* RETURN : |
| 1372 |
*/ |
| 1373 |
static void txLog(struct jfs_log *log, struct tblock *tblk, struct commit *cd) |
| 1374 |
{ |
| 1375 |
struct inode *ip; |
| 1376 |
lid_t lid; |
| 1377 |
struct tlock *tlck; |
| 1378 |
struct lrd *lrd = &cd->lrd; |
| 1379 |
|
| 1380 |
/* |
| 1381 |
* write log record(s) for each tlock of transaction, |
| 1382 |
*/ |
| 1383 |
for (lid = tblk->next; lid; lid = tlck->next) { |
| 1384 |
tlck = lid_to_tlock(lid); |
| 1385 |
|
| 1386 |
tlck->flag |= tlckLOG; |
| 1387 |
|
| 1388 |
/* initialize lrd common */ |
| 1389 |
ip = tlck->ip; |
| 1390 |
lrd->aggregate = cpu_to_le32(JFS_SBI(ip->i_sb)->aggregate); |
| 1391 |
lrd->log.redopage.fileset = cpu_to_le32(JFS_IP(ip)->fileset); |
| 1392 |
lrd->log.redopage.inode = cpu_to_le32(ip->i_ino); |
| 1393 |
|
| 1394 |
/* write log record of page from the tlock */ |
| 1395 |
switch (tlck->type & tlckTYPE) { |
| 1396 |
case tlckXTREE: |
| 1397 |
xtLog(log, tblk, lrd, tlck); |
| 1398 |
break; |
| 1399 |
|
| 1400 |
case tlckDTREE: |
| 1401 |
dtLog(log, tblk, lrd, tlck); |
| 1402 |
break; |
| 1403 |
|
| 1404 |
case tlckINODE: |
| 1405 |
diLog(log, tblk, lrd, tlck, cd); |
| 1406 |
break; |
| 1407 |
|
| 1408 |
case tlckMAP: |
| 1409 |
mapLog(log, tblk, lrd, tlck); |
| 1410 |
break; |
| 1411 |
|
| 1412 |
case tlckDATA: |
| 1413 |
dataLog(log, tblk, lrd, tlck); |
| 1414 |
break; |
| 1415 |
|
| 1416 |
default: |
| 1417 |
jfs_err("UFO tlock:0x%p", tlck); |
| 1418 |
} |
| 1419 |
} |
| 1420 |
|
| 1421 |
return; |
| 1422 |
} |
| 1423 |
|
| 1424 |
/* |
| 1425 |
* diLog() |
| 1426 |
* |
| 1427 |
* function: log inode tlock and format maplock to update bmap; |
| 1428 |
*/ |
| 1429 |
static void diLog(struct jfs_log *log, struct tblock *tblk, struct lrd *lrd, |
| 1430 |
struct tlock *tlck, struct commit *cd) |
| 1431 |
{ |
| 1432 |
struct metapage *mp; |
| 1433 |
pxd_t *pxd; |
| 1434 |
struct pxd_lock *pxdlock; |
| 1435 |
|
| 1436 |
mp = tlck->mp; |
| 1437 |
|
| 1438 |
/* initialize as REDOPAGE record format */ |
| 1439 |
lrd->log.redopage.type = cpu_to_le16(LOG_INODE); |
| 1440 |
lrd->log.redopage.l2linesize = cpu_to_le16(L2INODESLOTSIZE); |
| 1441 |
|
| 1442 |
pxd = &lrd->log.redopage.pxd; |
| 1443 |
|
| 1444 |
/* |
| 1445 |
* inode after image |
| 1446 |
*/ |
| 1447 |
if (tlck->type & tlckENTRY) { |
| 1448 |
/* log after-image for logredo(): */ |
| 1449 |
lrd->type = cpu_to_le16(LOG_REDOPAGE); |
| 1450 |
PXDaddress(pxd, mp->index); |
| 1451 |
PXDlength(pxd, |
| 1452 |
mp->logical_size >> tblk->sb->s_blocksize_bits); |
| 1453 |
lrd->backchain = cpu_to_le32(lmLog(log, tblk, lrd, tlck)); |
| 1454 |
|
| 1455 |
/* mark page as homeward bound */ |
| 1456 |
tlck->flag |= tlckWRITEPAGE; |
| 1457 |
} else if (tlck->type & tlckFREE) { |
| 1458 |
/* |
| 1459 |
* free inode extent |
| 1460 |
* |
| 1461 |
* (pages of the freed inode extent have been invalidated and |
| 1462 |
* a maplock for free of the extent has been formatted at |
| 1463 |
* txLock() time); |
| 1464 |
* |
| 1465 |
* the tlock had been acquired on the inode allocation map page |
| 1466 |
* (iag) that specifies the freed extent, even though the map |
| 1467 |
* page is not itself logged, to prevent pageout of the map |
| 1468 |
* page before the log; |
| 1469 |
*/ |
| 1470 |
|
| 1471 |
/* log LOG_NOREDOINOEXT of the freed inode extent for |
| 1472 |
* logredo() to start NoRedoPage filters, and to update |
| 1473 |
* imap and bmap for free of the extent; |
| 1474 |
*/ |
| 1475 |
lrd->type = cpu_to_le16(LOG_NOREDOINOEXT); |
| 1476 |
/* |
| 1477 |
* For the LOG_NOREDOINOEXT record, we need |
| 1478 |
* to pass the IAG number and inode extent |
| 1479 |
* index (within that IAG) from which the |
| 1480 |
* extent is being released. These have been |
| 1481 |
* passed to us in the iplist[1] and iplist[2]. |
| 1482 |
*/ |
| 1483 |
lrd->log.noredoinoext.iagnum = |
| 1484 |
cpu_to_le32((u32) (size_t) cd->iplist[1]); |
| 1485 |
lrd->log.noredoinoext.inoext_idx = |
| 1486 |
cpu_to_le32((u32) (size_t) cd->iplist[2]); |
| 1487 |
|
| 1488 |
pxdlock = (struct pxd_lock *) & tlck->lock; |
| 1489 |
*pxd = pxdlock->pxd; |
| 1490 |
lrd->backchain = cpu_to_le32(lmLog(log, tblk, lrd, NULL)); |
| 1491 |
|
| 1492 |
/* update bmap */ |
| 1493 |
tlck->flag |= tlckUPDATEMAP; |
| 1494 |
|
| 1495 |
/* mark page as homeward bound */ |
| 1496 |
tlck->flag |= tlckWRITEPAGE; |
| 1497 |
} else |
| 1498 |
jfs_err("diLog: UFO type tlck:0x%p", tlck); |
| 1499 |
return; |
| 1500 |
} |
| 1501 |
|
| 1502 |
/* |
| 1503 |
* dataLog() |
| 1504 |
* |
| 1505 |
* function: log data tlock |
| 1506 |
*/ |
| 1507 |
static void dataLog(struct jfs_log *log, struct tblock *tblk, struct lrd *lrd, |
| 1508 |
struct tlock *tlck) |
| 1509 |
{ |
| 1510 |
struct metapage *mp; |
| 1511 |
pxd_t *pxd; |
| 1512 |
|
| 1513 |
mp = tlck->mp; |
| 1514 |
|
| 1515 |
/* initialize as REDOPAGE record format */ |
| 1516 |
lrd->log.redopage.type = cpu_to_le16(LOG_DATA); |
| 1517 |
lrd->log.redopage.l2linesize = cpu_to_le16(L2DATASLOTSIZE); |
| 1518 |
|
| 1519 |
pxd = &lrd->log.redopage.pxd; |
| 1520 |
|
| 1521 |
/* log after-image for logredo(): */ |
| 1522 |
lrd->type = cpu_to_le16(LOG_REDOPAGE); |
| 1523 |
|
| 1524 |
if (jfs_dirtable_inline(tlck->ip)) { |
| 1525 |
/* |
| 1526 |
* The table has been truncated, we've must have deleted |
| 1527 |
* the last entry, so don't bother logging this |
| 1528 |
*/ |
| 1529 |
mp->lid = 0; |
| 1530 |
grab_metapage(mp); |
| 1531 |
metapage_homeok(mp); |
| 1532 |
discard_metapage(mp); |
| 1533 |
tlck->mp = NULL; |
| 1534 |
return; |
| 1535 |
} |
| 1536 |
|
| 1537 |
PXDaddress(pxd, mp->index); |
| 1538 |
PXDlength(pxd, mp->logical_size >> tblk->sb->s_blocksize_bits); |
| 1539 |
|
| 1540 |
lrd->backchain = cpu_to_le32(lmLog(log, tblk, lrd, tlck)); |
| 1541 |
|
| 1542 |
/* mark page as homeward bound */ |
| 1543 |
tlck->flag |= tlckWRITEPAGE; |
| 1544 |
|
| 1545 |
return; |
| 1546 |
} |
| 1547 |
|
| 1548 |
/* |
| 1549 |
* dtLog() |
| 1550 |
* |
| 1551 |
* function: log dtree tlock and format maplock to update bmap; |
| 1552 |
*/ |
| 1553 |
static void dtLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, |
| 1554 |
struct tlock * tlck) |
| 1555 |
{ |
| 1556 |
struct metapage *mp; |
| 1557 |
struct pxd_lock *pxdlock; |
| 1558 |
pxd_t *pxd; |
| 1559 |
|
| 1560 |
mp = tlck->mp; |
| 1561 |
|
| 1562 |
/* initialize as REDOPAGE/NOREDOPAGE record format */ |
| 1563 |
lrd->log.redopage.type = cpu_to_le16(LOG_DTREE); |
| 1564 |
lrd->log.redopage.l2linesize = cpu_to_le16(L2DTSLOTSIZE); |
| 1565 |
|
| 1566 |
pxd = &lrd->log.redopage.pxd; |
| 1567 |
|
| 1568 |
if (tlck->type & tlckBTROOT) |
| 1569 |
lrd->log.redopage.type |= cpu_to_le16(LOG_BTROOT); |
| 1570 |
|
| 1571 |
/* |
| 1572 |
* page extension via relocation: entry insertion; |
| 1573 |
* page extension in-place: entry insertion; |
| 1574 |
* new right page from page split, reinitialized in-line |
| 1575 |
* root from root page split: entry insertion; |
| 1576 |
*/ |
| 1577 |
if (tlck->type & (tlckNEW | tlckEXTEND)) { |
| 1578 |
/* log after-image of the new page for logredo(): |
| 1579 |
* mark log (LOG_NEW) for logredo() to initialize |
| 1580 |
* freelist and update bmap for alloc of the new page; |
| 1581 |
*/ |
| 1582 |
lrd->type = cpu_to_le16(LOG_REDOPAGE); |
| 1583 |
if (tlck->type & tlckEXTEND) |
| 1584 |
lrd->log.redopage.type |= cpu_to_le16(LOG_EXTEND); |
| 1585 |
else |
| 1586 |
lrd->log.redopage.type |= cpu_to_le16(LOG_NEW); |
| 1587 |
PXDaddress(pxd, mp->index); |
| 1588 |
PXDlength(pxd, |
| 1589 |
mp->logical_size >> tblk->sb->s_blocksize_bits); |
| 1590 |
lrd->backchain = cpu_to_le32(lmLog(log, tblk, lrd, tlck)); |
| 1591 |
|
| 1592 |
/* format a maplock for txUpdateMap() to update bPMAP for |
| 1593 |
* alloc of the new page; |
| 1594 |
*/ |
| 1595 |
if (tlck->type & tlckBTROOT) |
| 1596 |
return; |
| 1597 |
tlck->flag |= tlckUPDATEMAP; |
| 1598 |
pxdlock = (struct pxd_lock *) & tlck->lock; |
| 1599 |
pxdlock->flag = mlckALLOCPXD; |
| 1600 |
pxdlock->pxd = *pxd; |
| 1601 |
|
| 1602 |
pxdlock->index = 1; |
| 1603 |
|
| 1604 |
/* mark page as homeward bound */ |
| 1605 |
tlck->flag |= tlckWRITEPAGE; |
| 1606 |
return; |
| 1607 |
} |
| 1608 |
|
| 1609 |
/* |
| 1610 |
* entry insertion/deletion, |
| 1611 |
* sibling page link update (old right page before split); |
| 1612 |
*/ |
| 1613 |
if (tlck->type & (tlckENTRY | tlckRELINK)) { |
| 1614 |
/* log after-image for logredo(): */ |
| 1615 |
lrd->type = cpu_to_le16(LOG_REDOPAGE); |
| 1616 |
PXDaddress(pxd, mp->index); |
| 1617 |
PXDlength(pxd, |
| 1618 |
mp->logical_size >> tblk->sb->s_blocksize_bits); |
| 1619 |
lrd->backchain = cpu_to_le32(lmLog(log, tblk, lrd, tlck)); |
| 1620 |
|
| 1621 |
/* mark page as homeward bound */ |
| 1622 |
tlck->flag |= tlckWRITEPAGE; |
| 1623 |
return; |
| 1624 |
} |
| 1625 |
|
| 1626 |
/* |
| 1627 |
* page deletion: page has been invalidated |
| 1628 |
* page relocation: source extent |
| 1629 |
* |
| 1630 |
* a maplock for free of the page has been formatted |
| 1631 |
* at txLock() time); |
| 1632 |
*/ |
| 1633 |
if (tlck->type & (tlckFREE | tlckRELOCATE)) { |
| 1634 |
/* log LOG_NOREDOPAGE of the deleted page for logredo() |
| 1635 |
* to start NoRedoPage filter and to update bmap for free |
| 1636 |
* of the deletd page |
| 1637 |
*/ |
| 1638 |
lrd->type = cpu_to_le16(LOG_NOREDOPAGE); |
| 1639 |
pxdlock = (struct pxd_lock *) & tlck->lock; |
| 1640 |
*pxd = pxdlock->pxd; |
| 1641 |
lrd->backchain = cpu_to_le32(lmLog(log, tblk, lrd, NULL)); |
| 1642 |
|
| 1643 |
/* a maplock for txUpdateMap() for free of the page |
| 1644 |
* has been formatted at txLock() time; |
| 1645 |
*/ |
| 1646 |
tlck->flag |= tlckUPDATEMAP; |
| 1647 |
} |
| 1648 |
return; |
| 1649 |
} |
| 1650 |
|
| 1651 |
/* |
| 1652 |
* xtLog() |
| 1653 |
* |
| 1654 |
* function: log xtree tlock and format maplock to update bmap; |
| 1655 |
*/ |
| 1656 |
static void xtLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, |
| 1657 |
struct tlock * tlck) |
| 1658 |
{ |
| 1659 |
struct inode *ip; |
| 1660 |
struct metapage *mp; |
| 1661 |
xtpage_t *p; |
| 1662 |
struct xtlock *xtlck; |
| 1663 |
struct maplock *maplock; |
| 1664 |
struct xdlistlock *xadlock; |
| 1665 |
struct pxd_lock *pxdlock; |
| 1666 |
pxd_t *page_pxd; |
| 1667 |
int next, lwm, hwm; |
| 1668 |
|
| 1669 |
ip = tlck->ip; |
| 1670 |
mp = tlck->mp; |
| 1671 |
|
| 1672 |
/* initialize as REDOPAGE/NOREDOPAGE record format */ |
| 1673 |
lrd->log.redopage.type = cpu_to_le16(LOG_XTREE); |
| 1674 |
lrd->log.redopage.l2linesize = cpu_to_le16(L2XTSLOTSIZE); |
| 1675 |
|
| 1676 |
page_pxd = &lrd->log.redopage.pxd; |
| 1677 |
|
| 1678 |
if (tlck->type & tlckBTROOT) { |
| 1679 |
lrd->log.redopage.type |= cpu_to_le16(LOG_BTROOT); |
| 1680 |
p = (xtpage_t *) &JFS_IP(ip)->i_xtroot; |
| 1681 |
if (S_ISDIR(ip->i_mode)) |
| 1682 |
lrd->log.redopage.type |= |
| 1683 |
cpu_to_le16(LOG_DIR_XTREE); |
| 1684 |
} else |
| 1685 |
p = (xtpage_t *) mp->data; |
| 1686 |
next = le16_to_cpu(p->header.nextindex); |
| 1687 |
|
| 1688 |
xtlck = (struct xtlock *) & tlck->lock; |
| 1689 |
|
| 1690 |
maplock = (struct maplock *) & tlck->lock; |
| 1691 |
xadlock = (struct xdlistlock *) maplock; |
| 1692 |
|
| 1693 |
/* |
| 1694 |
* entry insertion/extension; |
| 1695 |
* sibling page link update (old right page before split); |
| 1696 |
*/ |
| 1697 |
if (tlck->type & (tlckNEW | tlckGROW | tlckRELINK)) { |
| 1698 |
/* log after-image for logredo(): |
| 1699 |
* logredo() will update bmap for alloc of new/extended |
| 1700 |
* extents (XAD_NEW|XAD_EXTEND) of XAD[lwm:next) from |
| 1701 |
* after-image of XADlist; |
| 1702 |
* logredo() resets (XAD_NEW|XAD_EXTEND) flag when |
| 1703 |
* applying the after-image to the meta-data page. |
| 1704 |
*/ |
| 1705 |
lrd->type = cpu_to_le16(LOG_REDOPAGE); |
| 1706 |
PXDaddress(page_pxd, mp->index); |
| 1707 |
PXDlength(page_pxd, |
| 1708 |
mp->logical_size >> tblk->sb->s_blocksize_bits); |
| 1709 |
lrd->backchain = cpu_to_le32(lmLog(log, tblk, lrd, tlck)); |
| 1710 |
|
| 1711 |
/* format a maplock for txUpdateMap() to update bPMAP |
| 1712 |
* for alloc of new/extended extents of XAD[lwm:next) |
| 1713 |
* from the page itself; |
| 1714 |
* txUpdateMap() resets (XAD_NEW|XAD_EXTEND) flag. |
| 1715 |
*/ |
| 1716 |
lwm = xtlck->lwm.offset; |
| 1717 |
if (lwm == 0) |
| 1718 |
lwm = XTPAGEMAXSLOT; |
| 1719 |
|
| 1720 |
if (lwm == next) |
| 1721 |
goto out; |
| 1722 |
if (lwm > next) { |
| 1723 |
jfs_err("xtLog: lwm > next"); |
| 1724 |
goto out; |
| 1725 |
} |
| 1726 |
tlck->flag |= tlckUPDATEMAP; |
| 1727 |
xadlock->flag = mlckALLOCXADLIST; |
| 1728 |
xadlock->count = next - lwm; |
| 1729 |
if ((xadlock->count <= 4) && (tblk->xflag & COMMIT_LAZY)) { |
| 1730 |
int i; |
| 1731 |
pxd_t *pxd; |
| 1732 |
/* |
| 1733 |
* Lazy commit may allow xtree to be modified before |
| 1734 |
* txUpdateMap runs. Copy xad into linelock to |
| 1735 |
* preserve correct data. |
| 1736 |
* |
| 1737 |
* We can fit twice as may pxd's as xads in the lock |
| 1738 |
*/ |
| 1739 |
xadlock->flag = mlckALLOCPXDLIST; |
| 1740 |
pxd = xadlock->xdlist = &xtlck->pxdlock; |
| 1741 |
for (i = 0; i < xadlock->count; i++) { |
| 1742 |
PXDaddress(pxd, addressXAD(&p->xad[lwm + i])); |
| 1743 |
PXDlength(pxd, lengthXAD(&p->xad[lwm + i])); |
| 1744 |
p->xad[lwm + i].flag &= |
| 1745 |
~(XAD_NEW | XAD_EXTENDED); |
| 1746 |
pxd++; |
| 1747 |
} |
| 1748 |
} else { |
| 1749 |
/* |
| 1750 |
* xdlist will point to into inode's xtree, ensure |
| 1751 |
* that transaction is not committed lazily. |
| 1752 |
*/ |
| 1753 |
xadlock->flag = mlckALLOCXADLIST; |
| 1754 |
xadlock->xdlist = &p->xad[lwm]; |
| 1755 |
tblk->xflag &= ~COMMIT_LAZY; |
| 1756 |
} |
| 1757 |
jfs_info("xtLog: alloc ip:0x%p mp:0x%p tlck:0x%p lwm:%d count:%d", |
| 1758 |
tlck->ip, mp, tlck, lwm, xadlock->count); |
| 1759 |
|
| 1760 |
maplock->index = 1; |
| 1761 |
|
| 1762 |
out: |
| 1763 |
/* mark page as homeward bound */ |
| 1764 |
tlck->flag |= tlckWRITEPAGE; |
| 1765 |
|
| 1766 |
return; |
| 1767 |
} |
| 1768 |
|
| 1769 |
/* |
| 1770 |
* page deletion: file deletion/truncation (ref. xtTruncate()) |
| 1771 |
* |
| 1772 |
* (page will be invalidated after log is written and bmap |
| 1773 |
* is updated from the page); |
| 1774 |
*/ |
| 1775 |
if (tlck->type & tlckFREE) { |
| 1776 |
/* LOG_NOREDOPAGE log for NoRedoPage filter: |
| 1777 |
* if page free from file delete, NoRedoFile filter from |
| 1778 |
* inode image of zero link count will subsume NoRedoPage |
| 1779 |
* filters for each page; |
| 1780 |
* if page free from file truncattion, write NoRedoPage |
| 1781 |
* filter; |
| 1782 |
* |
| 1783 |
* upadte of block allocation map for the page itself: |
| 1784 |
* if page free from deletion and truncation, LOG_UPDATEMAP |
| 1785 |
* log for the page itself is generated from processing |
| 1786 |
* its parent page xad entries; |
| 1787 |
*/ |
| 1788 |
/* if page free from file truncation, log LOG_NOREDOPAGE |
| 1789 |
* of the deleted page for logredo() to start NoRedoPage |
| 1790 |
* filter for the page; |
| 1791 |
*/ |
| 1792 |
if (tblk->xflag & COMMIT_TRUNCATE) { |
| 1793 |
/* write NOREDOPAGE for the page */ |
| 1794 |
lrd->type = cpu_to_le16(LOG_NOREDOPAGE); |
| 1795 |
PXDaddress(page_pxd, mp->index); |
| 1796 |
PXDlength(page_pxd, |
| 1797 |
mp->logical_size >> tblk->sb-> |
| 1798 |
s_blocksize_bits); |
| 1799 |
lrd->backchain = |
| 1800 |
cpu_to_le32(lmLog(log, tblk, lrd, NULL)); |
| 1801 |
|
| 1802 |
if (tlck->type & tlckBTROOT) { |
| 1803 |
/* Empty xtree must be logged */ |
| 1804 |
lrd->type = cpu_to_le16(LOG_REDOPAGE); |
| 1805 |
lrd->backchain = |
| 1806 |
cpu_to_le32(lmLog(log, tblk, lrd, tlck)); |
| 1807 |
} |
| 1808 |
} |
| 1809 |
|
| 1810 |
/* init LOG_UPDATEMAP of the freed extents |
| 1811 |
* XAD[XTENTRYSTART:hwm) from the deleted page itself |
| 1812 |
* for logredo() to update bmap; |
| 1813 |
*/ |
| 1814 |
lrd->type = cpu_to_le16(LOG_UPDATEMAP); |
| 1815 |
lrd->log.updatemap.type = cpu_to_le16(LOG_FREEXADLIST); |
| 1816 |
xtlck = (struct xtlock *) & tlck->lock; |
| 1817 |
hwm = xtlck->hwm.offset; |
| 1818 |
lrd->log.updatemap.nxd = |
| 1819 |
cpu_to_le16(hwm - XTENTRYSTART + 1); |
| 1820 |
/* reformat linelock for lmLog() */ |
| 1821 |
xtlck->header.offset = XTENTRYSTART; |
| 1822 |
xtlck->header.length = hwm - XTENTRYSTART + 1; |
| 1823 |
xtlck->index = 1; |
| 1824 |
lrd->backchain = cpu_to_le32(lmLog(log, tblk, lrd, tlck)); |
| 1825 |
|
| 1826 |
/* format a maplock for txUpdateMap() to update bmap |
| 1827 |
* to free extents of XAD[XTENTRYSTART:hwm) from the |
| 1828 |
* deleted page itself; |
| 1829 |
*/ |
| 1830 |
tlck->flag |= tlckUPDATEMAP; |
| 1831 |
xadlock->count = hwm - XTENTRYSTART + 1; |
| 1832 |
if ((xadlock->count <= 4) && (tblk->xflag & COMMIT_LAZY)) { |
| 1833 |
int i; |
| 1834 |
pxd_t *pxd; |
| 1835 |
/* |
| 1836 |
* Lazy commit may allow xtree to be modified before |
| 1837 |
* txUpdateMap runs. Copy xad into linelock to |
| 1838 |
* preserve correct data. |
| 1839 |
* |
| 1840 |
* We can fit twice as may pxd's as xads in the lock |
| 1841 |
*/ |
| 1842 |
xadlock->flag = mlckFREEPXDLIST; |
| 1843 |
pxd = xadlock->xdlist = &xtlck->pxdlock; |
| 1844 |
for (i = 0; i < xadlock->count; i++) { |
| 1845 |
PXDaddress(pxd, |
| 1846 |
addressXAD(&p->xad[XTENTRYSTART + i])); |
| 1847 |
PXDlength(pxd, |
| 1848 |
lengthXAD(&p->xad[XTENTRYSTART + i])); |
| 1849 |
pxd++; |
| 1850 |
} |
| 1851 |
} else { |
| 1852 |
/* |
| 1853 |
* xdlist will point to into inode's xtree, ensure |
| 1854 |
* that transaction is not committed lazily. |
| 1855 |
*/ |
| 1856 |
xadlock->flag = mlckFREEXADLIST; |
| 1857 |
xadlock->xdlist = &p->xad[XTENTRYSTART]; |
| 1858 |
tblk->xflag &= ~COMMIT_LAZY; |
| 1859 |
} |
| 1860 |
jfs_info("xtLog: free ip:0x%p mp:0x%p count:%d lwm:2", |
| 1861 |
tlck->ip, mp, xadlock->count); |
| 1862 |
|
| 1863 |
maplock->index = 1; |
| 1864 |
|
| 1865 |
/* mark page as invalid */ |
| 1866 |
if (((tblk->xflag & COMMIT_PWMAP) || S_ISDIR(ip->i_mode)) |
| 1867 |
&& !(tlck->type & tlckBTROOT)) |
| 1868 |
tlck->flag |= tlckFREEPAGE; |
| 1869 |
/* |
| 1870 |
else (tblk->xflag & COMMIT_PMAP) |
| 1871 |
? release the page; |
| 1872 |
*/ |
| 1873 |
return; |
| 1874 |
} |
| 1875 |
|
| 1876 |
/* |
| 1877 |
* page/entry truncation: file truncation (ref. xtTruncate()) |
| 1878 |
* |
| 1879 |
* |----------+------+------+---------------| |
| 1880 |
* | | | |
| 1881 |
* | | hwm - hwm before truncation |
| 1882 |
* | next - truncation point |
| 1883 |
* lwm - lwm before truncation |
| 1884 |
* header ? |
| 1885 |
*/ |
| 1886 |
if (tlck->type & tlckTRUNCATE) { |
| 1887 |
pxd_t pxd; /* truncated extent of xad */ |
| 1888 |
int twm; |
| 1889 |
|
| 1890 |
/* |
| 1891 |
* For truncation the entire linelock may be used, so it would |
| 1892 |
* be difficult to store xad list in linelock itself. |
| 1893 |
* Therefore, we'll just force transaction to be committed |
| 1894 |
* synchronously, so that xtree pages won't be changed before |
| 1895 |
* txUpdateMap runs. |
| 1896 |
*/ |
| 1897 |
tblk->xflag &= ~COMMIT_LAZY; |
| 1898 |
lwm = xtlck->lwm.offset; |
| 1899 |
if (lwm == 0) |
| 1900 |
lwm = XTPAGEMAXSLOT; |
| 1901 |
hwm = xtlck->hwm.offset; |
| 1902 |
twm = xtlck->twm.offset; |
| 1903 |
|
| 1904 |
/* |
| 1905 |
* write log records |
| 1906 |
*/ |
| 1907 |
/* log after-image for logredo(): |
| 1908 |
* |
| 1909 |
* logredo() will update bmap for alloc of new/extended |
| 1910 |
* extents (XAD_NEW|XAD_EXTEND) of XAD[lwm:next) from |
| 1911 |
* after-image of XADlist; |
| 1912 |
* logredo() resets (XAD_NEW|XAD_EXTEND) flag when |
| 1913 |
* applying the after-image to the meta-data page. |
| 1914 |
*/ |
| 1915 |
lrd->type = cpu_to_le16(LOG_REDOPAGE); |
| 1916 |
PXDaddress(page_pxd, mp->index); |
| 1917 |
PXDlength(page_pxd, |
| 1918 |
mp->logical_size >> tblk->sb->s_blocksize_bits); |
| 1919 |
lrd->backchain = cpu_to_le32(lmLog(log, tblk, lrd, tlck)); |
| 1920 |
|
| 1921 |
/* |
| 1922 |
* truncate entry XAD[twm == next - 1]: |
| 1923 |
*/ |
| 1924 |
if (twm == next - 1) { |
| 1925 |
/* init LOG_UPDATEMAP for logredo() to update bmap for |
| 1926 |
* free of truncated delta extent of the truncated |
| 1927 |
* entry XAD[next - 1]: |
| 1928 |
* (xtlck->pxdlock = truncated delta extent); |
| 1929 |
*/ |
| 1930 |
pxdlock = (struct pxd_lock *) & xtlck->pxdlock; |
| 1931 |
/* assert(pxdlock->type & tlckTRUNCATE); */ |
| 1932 |
lrd->type = cpu_to_le16(LOG_UPDATEMAP); |
| 1933 |
lrd->log.updatemap.type = cpu_to_le16(LOG_FREEPXD); |
| 1934 |
lrd->log.updatemap.nxd = cpu_to_le16(1); |
| 1935 |
lrd->log.updatemap.pxd = pxdlock->pxd; |
| 1936 |
pxd = pxdlock->pxd; /* save to format maplock */ |
| 1937 |
lrd->backchain = |
| 1938 |
cpu_to_le32(lmLog(log, tblk, lrd, NULL)); |
| 1939 |
} |
| 1940 |
|
| 1941 |
/* |
| 1942 |
* free entries XAD[next:hwm]: |
| 1943 |
*/ |
| 1944 |
if (hwm >= next) { |
| 1945 |
/* init LOG_UPDATEMAP of the freed extents |
| 1946 |
* XAD[next:hwm] from the deleted page itself |
| 1947 |
* for logredo() to update bmap; |
| 1948 |
*/ |
| 1949 |
lrd->type = cpu_to_le16(LOG_UPDATEMAP); |
| 1950 |
lrd->log.updatemap.type = |
| 1951 |
cpu_to_le16(LOG_FREEXADLIST); |
| 1952 |
xtlck = (struct xtlock *) & tlck->lock; |
| 1953 |
hwm = xtlck->hwm.offset; |
| 1954 |
lrd->log.updatemap.nxd = |
| 1955 |
cpu_to_le16(hwm - next + 1); |
| 1956 |
/* reformat linelock for lmLog() */ |
| 1957 |
xtlck->header.offset = next; |
| 1958 |
xtlck->header.length = hwm - next + 1; |
| 1959 |
xtlck->index = 1; |
| 1960 |
lrd->backchain = |
| 1961 |
cpu_to_le32(lmLog(log, tblk, lrd, tlck)); |
| 1962 |
} |
| 1963 |
|
| 1964 |
/* |
| 1965 |
* format maplock(s) for txUpdateMap() to update bmap |
| 1966 |
*/ |
| 1967 |
maplock->index = 0; |
| 1968 |
|
| 1969 |
/* |
| 1970 |
* allocate entries XAD[lwm:next): |
| 1971 |
*/ |
| 1972 |
if (lwm < next) { |
| 1973 |
/* format a maplock for txUpdateMap() to update bPMAP |
| 1974 |
* for alloc of new/extended extents of XAD[lwm:next) |
| 1975 |
* from the page itself; |
| 1976 |
* txUpdateMap() resets (XAD_NEW|XAD_EXTEND) flag. |
| 1977 |
*/ |
| 1978 |
tlck->flag |= tlckUPDATEMAP; |
| 1979 |
xadlock->flag = mlckALLOCXADLIST; |
| 1980 |
xadlock->count = next - lwm; |
| 1981 |
xadlock->xdlist = &p->xad[lwm]; |
| 1982 |
|
| 1983 |
jfs_info("xtLog: alloc ip:0x%p mp:0x%p count:%d lwm:%d next:%d", |
| 1984 |
tlck->ip, mp, xadlock->count, lwm, next); |
| 1985 |
maplock->index++; |
| 1986 |
xadlock++; |
| 1987 |
} |
| 1988 |
|
| 1989 |
/* |
| 1990 |
* truncate entry XAD[twm == next - 1]: |
| 1991 |
*/ |
| 1992 |
if (twm == next - 1) { |
| 1993 |
/* format a maplock for txUpdateMap() to update bmap |
| 1994 |
* to free truncated delta extent of the truncated |
| 1995 |
* entry XAD[next - 1]; |
| 1996 |
* (xtlck->pxdlock = truncated delta extent); |
| 1997 |
*/ |
| 1998 |
tlck->flag |= tlckUPDATEMAP; |
| 1999 |
pxdlock = (struct pxd_lock *) xadlock; |
| 2000 |
pxdlock->flag = mlckFREEPXD; |
| 2001 |
pxdlock->count = 1; |
| 2002 |
pxdlock->pxd = pxd; |
| 2003 |
|
| 2004 |
jfs_info("xtLog: truncate ip:0x%p mp:0x%p count:%d hwm:%d", |
| 2005 |
ip, mp, pxdlock->count, hwm); |
| 2006 |
maplock->index++; |
| 2007 |
xadlock++; |
| 2008 |
} |
| 2009 |
|
| 2010 |
/* |
| 2011 |
* free entries XAD[next:hwm]: |
| 2012 |
*/ |
| 2013 |
if (hwm >= next) { |
| 2014 |
/* format a maplock for txUpdateMap() to update bmap |
| 2015 |
* to free extents of XAD[next:hwm] from thedeleted |
| 2016 |
* page itself; |
| 2017 |
*/ |
| 2018 |
tlck->flag |= tlckUPDATEMAP; |
| 2019 |
xadlock->flag = mlckFREEXADLIST; |
| 2020 |
xadlock->count = hwm - next + 1; |
| 2021 |
xadlock->xdlist = &p->xad[next]; |
| 2022 |
|
| 2023 |
jfs_info("xtLog: free ip:0x%p mp:0x%p count:%d next:%d hwm:%d", |
| 2024 |
tlck->ip, mp, xadlock->count, next, hwm); |
| 2025 |
maplock->index++; |
| 2026 |
} |
| 2027 |
|
| 2028 |
/* mark page as homeward bound */ |
| 2029 |
tlck->flag |= tlckWRITEPAGE; |
| 2030 |
} |
| 2031 |
return; |
| 2032 |
} |
| 2033 |
|
| 2034 |
/* |
| 2035 |
* mapLog() |
| 2036 |
* |
| 2037 |
* function: log from maplock of freed data extents; |
| 2038 |
*/ |
| 2039 |
static void mapLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, |
| 2040 |
struct tlock * tlck) |
| 2041 |
{ |
| 2042 |
struct pxd_lock *pxdlock; |
| 2043 |
int i, nlock; |
| 2044 |
pxd_t *pxd; |
| 2045 |
|
| 2046 |
/* |
| 2047 |
* page relocation: free the source page extent |
| 2048 |
* |
| 2049 |
* a maplock for txUpdateMap() for free of the page |
| 2050 |
* has been formatted at txLock() time saving the src |
| 2051 |
* relocated page address; |
| 2052 |
*/ |
| 2053 |
if (tlck->type & tlckRELOCATE) { |
| 2054 |
/* log LOG_NOREDOPAGE of the old relocated page |
| 2055 |
* for logredo() to start NoRedoPage filter; |
| 2056 |
*/ |
| 2057 |
lrd->type = cpu_to_le16(LOG_NOREDOPAGE); |
| 2058 |
pxdlock = (struct pxd_lock *) & tlck->lock; |
| 2059 |
pxd = &lrd->log.redopage.pxd; |
| 2060 |
*pxd = pxdlock->pxd; |
| 2061 |
lrd->backchain = cpu_to_le32(lmLog(log, tblk, lrd, NULL)); |
| 2062 |
|
| 2063 |
/* (N.B. currently, logredo() does NOT update bmap |
| 2064 |
* for free of the page itself for (LOG_XTREE|LOG_NOREDOPAGE); |
| 2065 |
* if page free from relocation, LOG_UPDATEMAP log is |
| 2066 |
* specifically generated now for logredo() |
| 2067 |
* to update bmap for free of src relocated page; |
| 2068 |
* (new flag LOG_RELOCATE may be introduced which will |
| 2069 |
* inform logredo() to start NORedoPage filter and also |
| 2070 |
* update block allocation map at the same time, thus |
| 2071 |
* avoiding an extra log write); |
| 2072 |
*/ |
| 2073 |
lrd->type = cpu_to_le16(LOG_UPDATEMAP); |
| 2074 |
lrd->log.updatemap.type = cpu_to_le16(LOG_FREEPXD); |
| 2075 |
lrd->log.updatemap.nxd = cpu_to_le16(1); |
| 2076 |
lrd->log.updatemap.pxd = pxdlock->pxd; |
| 2077 |
lrd->backchain = cpu_to_le32(lmLog(log, tblk, lrd, NULL)); |
| 2078 |
|
| 2079 |
/* a maplock for txUpdateMap() for free of the page |
| 2080 |
* has been formatted at txLock() time; |
| 2081 |
*/ |
| 2082 |
tlck->flag |= tlckUPDATEMAP; |
| 2083 |
return; |
| 2084 |
} |
| 2085 |
/* |
| 2086 |
|
| 2087 |
* Otherwise it's not a relocate request |
| 2088 |
* |
| 2089 |
*/ |
| 2090 |
else { |
| 2091 |
/* log LOG_UPDATEMAP for logredo() to update bmap for |
| 2092 |
* free of truncated/relocated delta extent of the data; |
| 2093 |
* e.g.: external EA extent, relocated/truncated extent |
| 2094 |
* from xtTailgate(); |
| 2095 |
*/ |
| 2096 |
lrd->type = cpu_to_le16(LOG_UPDATEMAP); |
| 2097 |
pxdlock = (struct pxd_lock *) & tlck->lock; |
| 2098 |
nlock = pxdlock->index; |
| 2099 |
for (i = 0; i < nlock; i++, pxdlock++) { |
| 2100 |
if (pxdlock->flag & mlckALLOCPXD) |
| 2101 |
lrd->log.updatemap.type = |
| 2102 |
cpu_to_le16(LOG_ALLOCPXD); |
| 2103 |
else |
| 2104 |
lrd->log.updatemap.type = |
| 2105 |
cpu_to_le16(LOG_FREEPXD); |
| 2106 |
lrd->log.updatemap.nxd = cpu_to_le16(1); |
| 2107 |
lrd->log.updatemap.pxd = pxdlock->pxd; |
| 2108 |
lrd->backchain = |
| 2109 |
cpu_to_le32(lmLog(log, tblk, lrd, NULL)); |
| 2110 |
jfs_info("mapLog: xaddr:0x%lx xlen:0x%x", |
| 2111 |
(ulong) addressPXD(&pxdlock->pxd), |
| 2112 |
lengthPXD(&pxdlock->pxd)); |
| 2113 |
} |
| 2114 |
|
| 2115 |
/* update bmap */ |
| 2116 |
tlck->flag |= tlckUPDATEMAP; |
| 2117 |
} |
| 2118 |
} |
| 2119 |
|
| 2120 |
/* |
| 2121 |
* txEA() |
| 2122 |
* |
| 2123 |
* function: acquire maplock for EA/ACL extents or |
| 2124 |
* set COMMIT_INLINE flag; |
| 2125 |
*/ |
| 2126 |
void txEA(tid_t tid, struct inode *ip, dxd_t * oldea, dxd_t * newea) |
| 2127 |
{ |
| 2128 |
struct tlock *tlck = NULL; |
| 2129 |
struct pxd_lock *maplock = NULL, *pxdlock = NULL; |
| 2130 |
|
| 2131 |
/* |
| 2132 |
* format maplock for alloc of new EA extent |
| 2133 |
*/ |
| 2134 |
if (newea) { |
| 2135 |
/* Since the newea could be a completely zeroed entry we need to |
| 2136 |
* check for the two flags which indicate we should actually |
| 2137 |
* commit new EA data |
| 2138 |
*/ |
| 2139 |
if (newea->flag & DXD_EXTENT) { |
| 2140 |
tlck = txMaplock(tid, ip, tlckMAP); |
| 2141 |
maplock = (struct pxd_lock *) & tlck->lock; |
| 2142 |
pxdlock = (struct pxd_lock *) maplock; |
| 2143 |
pxdlock->flag = mlckALLOCPXD; |
| 2144 |
PXDaddress(&pxdlock->pxd, addressDXD(newea)); |
| 2145 |
PXDlength(&pxdlock->pxd, lengthDXD(newea)); |
| 2146 |
pxdlock++; |
| 2147 |
maplock->index = 1; |
| 2148 |
} else if (newea->flag & DXD_INLINE) { |
| 2149 |
tlck = NULL; |
| 2150 |
|
| 2151 |
set_cflag(COMMIT_Inlineea, ip); |
| 2152 |
} |
| 2153 |
} |
| 2154 |
|
| 2155 |
/* |
| 2156 |
* format maplock for free of old EA extent |
| 2157 |
*/ |
| 2158 |
if (!test_cflag(COMMIT_Nolink, ip) && oldea->flag & DXD_EXTENT) { |
| 2159 |
if (tlck == NULL) { |
| 2160 |
tlck = txMaplock(tid, ip, tlckMAP); |
| 2161 |
maplock = (struct pxd_lock *) & tlck->lock; |
| 2162 |
pxdlock = (struct pxd_lock *) maplock; |
| 2163 |
maplock->index = 0; |
| 2164 |
} |
| 2165 |
pxdlock->flag = mlckFREEPXD; |
| 2166 |
PXDaddress(&pxdlock->pxd, addressDXD(oldea)); |
| 2167 |
PXDlength(&pxdlock->pxd, lengthDXD(oldea)); |
| 2168 |
maplock->index++; |
| 2169 |
} |
| 2170 |
} |
| 2171 |
|
| 2172 |
/* |
| 2173 |
* txForce() |
| 2174 |
* |
| 2175 |
* function: synchronously write pages locked by transaction |
| 2176 |
* after txLog() but before txUpdateMap(); |
| 2177 |
*/ |
| 2178 |
static void txForce(struct tblock * tblk) |
| 2179 |
{ |
| 2180 |
struct tlock *tlck; |
| 2181 |
lid_t lid, next; |
| 2182 |
struct metapage *mp; |
| 2183 |
|
| 2184 |
/* |
| 2185 |
* reverse the order of transaction tlocks in |
| 2186 |
* careful update order of address index pages |
| 2187 |
* (right to left, bottom up) |
| 2188 |
*/ |
| 2189 |
tlck = lid_to_tlock(tblk->next); |
| 2190 |
lid = tlck->next; |
| 2191 |
tlck->next = 0; |
| 2192 |
while (lid) { |
| 2193 |
tlck = lid_to_tlock(lid); |
| 2194 |
next = tlck->next; |
| 2195 |
tlck->next = tblk->next; |
| 2196 |
tblk->next = lid; |
| 2197 |
lid = next; |
| 2198 |
} |
| 2199 |
|
| 2200 |
/* |
| 2201 |
* synchronously write the page, and |
| 2202 |
* hold the page for txUpdateMap(); |
| 2203 |
*/ |
| 2204 |
for (lid = tblk->next; lid; lid = next) { |
| 2205 |
tlck = lid_to_tlock(lid); |
| 2206 |
next = tlck->next; |
| 2207 |
|
| 2208 |
if ((mp = tlck->mp) != NULL && |
| 2209 |
(tlck->type & tlckBTROOT) == 0) { |
| 2210 |
assert(mp->xflag & COMMIT_PAGE); |
| 2211 |
|
| 2212 |
if (tlck->flag & tlckWRITEPAGE) { |
| 2213 |
tlck->flag &= ~tlckWRITEPAGE; |
| 2214 |
|
| 2215 |
/* do not release page to freelist */ |
| 2216 |
force_metapage(mp); |
| 2217 |
#if 0 |
| 2218 |
/* |
| 2219 |
* The "right" thing to do here is to |
| 2220 |
* synchronously write the metadata. |
| 2221 |
* With the current implementation this |
| 2222 |
* is hard since write_metapage requires |
| 2223 |
* us to kunmap & remap the page. If we |
| 2224 |
* have tlocks pointing into the metadata |
| 2225 |
* pages, we don't want to do this. I think |
| 2226 |
* we can get by with synchronously writing |
| 2227 |
* the pages when they are released. |
| 2228 |
*/ |
| 2229 |
assert(mp->nohomeok); |
| 2230 |
set_bit(META_dirty, &mp->flag); |
| 2231 |
set_bit(META_sync, &mp->flag); |
| 2232 |
#endif |
| 2233 |
} |
| 2234 |
} |
| 2235 |
} |
| 2236 |
} |
| 2237 |
|
| 2238 |
/* |
| 2239 |
* txUpdateMap() |
| 2240 |
* |
| 2241 |
* function: update persistent allocation map (and working map |
| 2242 |
* if appropriate); |
| 2243 |
* |
| 2244 |
* parameter: |
| 2245 |
*/ |
| 2246 |
static void txUpdateMap(struct tblock * tblk) |
| 2247 |
{ |
| 2248 |
struct inode *ip; |
| 2249 |
struct inode *ipimap; |
| 2250 |
lid_t lid; |
| 2251 |
struct tlock *tlck; |
| 2252 |
struct maplock *maplock; |
| 2253 |
struct pxd_lock pxdlock; |
| 2254 |
int maptype; |
| 2255 |
int k, nlock; |
| 2256 |
struct metapage *mp = NULL; |
| 2257 |
|
| 2258 |
ipimap = JFS_SBI(tblk->sb)->ipimap; |
| 2259 |
|
| 2260 |
maptype = (tblk->xflag & COMMIT_PMAP) ? COMMIT_PMAP : COMMIT_PWMAP; |
| 2261 |
|
| 2262 |
|
| 2263 |
/* |
| 2264 |
* update block allocation map |
| 2265 |
* |
| 2266 |
* update allocation state in pmap (and wmap) and |
| 2267 |
* update lsn of the pmap page; |
| 2268 |
*/ |
| 2269 |
/* |
| 2270 |
* scan each tlock/page of transaction for block allocation/free: |
| 2271 |
* |
| 2272 |
* for each tlock/page of transaction, update map. |
| 2273 |
* ? are there tlock for pmap and pwmap at the same time ? |
| 2274 |
*/ |
| 2275 |
for (lid = tblk->next; lid; lid = tlck->next) { |
| 2276 |
tlck = lid_to_tlock(lid); |
| 2277 |
|
| 2278 |
if ((tlck->flag & tlckUPDATEMAP) == 0) |
| 2279 |
continue; |
| 2280 |
|
| 2281 |
if (tlck->flag & tlckFREEPAGE) { |
| 2282 |
/* |
| 2283 |
* Another thread may attempt to reuse freed space |
| 2284 |
* immediately, so we want to get rid of the metapage |
| 2285 |
* before anyone else has a chance to get it. |
| 2286 |
* Lock metapage, update maps, then invalidate |
| 2287 |
* the metapage. |
| 2288 |
*/ |
| 2289 |
mp = tlck->mp; |
| 2290 |
ASSERT(mp->xflag & COMMIT_PAGE); |
| 2291 |
grab_metapage(mp); |
| 2292 |
} |
| 2293 |
|
| 2294 |
/* |
| 2295 |
* extent list: |
| 2296 |
* . in-line PXD list: |
| 2297 |
* . out-of-line XAD list: |
| 2298 |
*/ |
| 2299 |
maplock = (struct maplock *) & tlck->lock; |
| 2300 |
nlock = maplock->index; |
| 2301 |
|
| 2302 |
for (k = 0; k < nlock; k++, maplock++) { |
| 2303 |
/* |
| 2304 |
* allocate blocks in persistent map: |
| 2305 |
* |
| 2306 |
* blocks have been allocated from wmap at alloc time; |
| 2307 |
*/ |
| 2308 |
if (maplock->flag & mlckALLOC) { |
| 2309 |
txAllocPMap(ipimap, maplock, tblk); |
| 2310 |
} |
| 2311 |
/* |
| 2312 |
* free blocks in persistent and working map: |
| 2313 |
* blocks will be freed in pmap and then in wmap; |
| 2314 |
* |
| 2315 |
* ? tblock specifies the PMAP/PWMAP based upon |
| 2316 |
* transaction |
| 2317 |
* |
| 2318 |
* free blocks in persistent map: |
| 2319 |
* blocks will be freed from wmap at last reference |
| 2320 |
* release of the object for regular files; |
| 2321 |
* |
| 2322 |
* Alway free blocks from both persistent & working |
| 2323 |
* maps for directories |
| 2324 |
*/ |
| 2325 |
else { /* (maplock->flag & mlckFREE) */ |
| 2326 |
|
| 2327 |
if (tlck->flag & tlckDIRECTORY) |
| 2328 |
txFreeMap(ipimap, maplock, |
| 2329 |
tblk, COMMIT_PWMAP); |
| 2330 |
else |
| 2331 |
txFreeMap(ipimap, maplock, |
| 2332 |
tblk, maptype); |
| 2333 |
} |
| 2334 |
} |
| 2335 |
if (tlck->flag & tlckFREEPAGE) { |
| 2336 |
if (!(tblk->flag & tblkGC_LAZY)) { |
| 2337 |
/* This is equivalent to txRelease */ |
| 2338 |
ASSERT(mp->lid == lid); |
| 2339 |
tlck->mp->lid = 0; |
| 2340 |
} |
| 2341 |
assert(mp->nohomeok == 1); |
| 2342 |
metapage_homeok(mp); |
| 2343 |
discard_metapage(mp); |
| 2344 |
tlck->mp = NULL; |
| 2345 |
} |
| 2346 |
} |
| 2347 |
/* |
| 2348 |
* update inode allocation map |
| 2349 |
* |
| 2350 |
* update allocation state in pmap and |
| 2351 |
* update lsn of the pmap page; |
| 2352 |
* update in-memory inode flag/state |
| 2353 |
* |
| 2354 |
* unlock mapper/write lock |
| 2355 |
*/ |
| 2356 |
if (tblk->xflag & COMMIT_CREATE) { |
| 2357 |
diUpdatePMap(ipimap, tblk->ino, false, tblk); |
| 2358 |
/* update persistent block allocation map |
| 2359 |
* for the allocation of inode extent; |
| 2360 |
*/ |
| 2361 |
pxdlock.flag = mlckALLOCPXD; |
| 2362 |
pxdlock.pxd = tblk->u.ixpxd; |
| 2363 |
pxdlock.index = 1; |
| 2364 |
txAllocPMap(ipimap, (struct maplock *) & pxdlock, tblk); |
| 2365 |
} else if (tblk->xflag & COMMIT_DELETE) { |
| 2366 |
ip = tblk->u.ip; |
| 2367 |
diUpdatePMap(ipimap, ip->i_ino, true, tblk); |
| 2368 |
iput(ip); |
| 2369 |
} |
| 2370 |
} |
| 2371 |
|
| 2372 |
/* |
| 2373 |
* txAllocPMap() |
| 2374 |
* |
| 2375 |
* function: allocate from persistent map; |
| 2376 |
* |
| 2377 |
* parameter: |
| 2378 |
* ipbmap - |
| 2379 |
* malock - |
| 2380 |
* xad list: |
| 2381 |
* pxd: |
| 2382 |
* |
| 2383 |
* maptype - |
| 2384 |
* allocate from persistent map; |
| 2385 |
* free from persistent map; |
| 2386 |
* (e.g., tmp file - free from working map at releae |
| 2387 |
* of last reference); |
| 2388 |
* free from persistent and working map; |
| 2389 |
* |
| 2390 |
* lsn - log sequence number; |
| 2391 |
*/ |
| 2392 |
static void txAllocPMap(struct inode *ip, struct maplock * maplock, |
| 2393 |
struct tblock * tblk) |
| 2394 |
{ |
| 2395 |
struct inode *ipbmap = JFS_SBI(ip->i_sb)->ipbmap; |
| 2396 |
struct xdlistlock *xadlistlock; |
| 2397 |
xad_t *xad; |
| 2398 |
s64 xaddr; |
| 2399 |
int xlen; |
| 2400 |
struct pxd_lock *pxdlock; |
| 2401 |
struct xdlistlock *pxdlistlock; |
| 2402 |
pxd_t *pxd; |
| 2403 |
int n; |
| 2404 |
|
| 2405 |
/* |
| 2406 |
* allocate from persistent map; |
| 2407 |
*/ |
| 2408 |
if (maplock->flag & mlckALLOCXADLIST) { |
| 2409 |
xadlistlock = (struct xdlistlock *) maplock; |
| 2410 |
xad = xadlistlock->xdlist; |
| 2411 |
for (n = 0; n < xadlistlock->count; n++, xad++) { |
| 2412 |
if (xad->flag & (XAD_NEW | XAD_EXTENDED)) { |
| 2413 |
xaddr = addressXAD(xad); |
| 2414 |
xlen = lengthXAD(xad); |
| 2415 |
dbUpdatePMap(ipbmap, false, xaddr, |
| 2416 |
(s64) xlen, tblk); |
| 2417 |
xad->flag &= ~(XAD_NEW | XAD_EXTENDED); |
| 2418 |
jfs_info("allocPMap: xaddr:0x%lx xlen:%d", |
| 2419 |
(ulong) xaddr, xlen); |
| 2420 |
} |
| 2421 |
} |
| 2422 |
} else if (maplock->flag & mlckALLOCPXD) { |
| 2423 |
pxdlock = (struct pxd_lock *) maplock; |
| 2424 |
xaddr = addressPXD(&pxdlock->pxd); |
| 2425 |
xlen = lengthPXD(&pxdlock->pxd); |
| 2426 |
dbUpdatePMap(ipbmap, false, xaddr, (s64) xlen, tblk); |
| 2427 |
jfs_info("allocPMap: xaddr:0x%lx xlen:%d", (ulong) xaddr, xlen); |
| 2428 |
} else { /* (maplock->flag & mlckALLOCPXDLIST) */ |
| 2429 |
|
| 2430 |
pxdlistlock = (struct xdlistlock *) maplock; |
| 2431 |
pxd = pxdlistlock->xdlist; |
| 2432 |
for (n = 0; n < pxdlistlock->count; n++, pxd++) { |
| 2433 |
xaddr = addressPXD(pxd); |
| 2434 |
xlen = lengthPXD(pxd); |
| 2435 |
dbUpdatePMap(ipbmap, false, xaddr, (s64) xlen, |
| 2436 |
tblk); |
| 2437 |
jfs_info("allocPMap: xaddr:0x%lx xlen:%d", |
| 2438 |
(ulong) xaddr, xlen); |
| 2439 |
} |
| 2440 |
} |
| 2441 |
} |
| 2442 |
|
| 2443 |
/* |
| 2444 |
* txFreeMap() |
| 2445 |
* |
| 2446 |
* function: free from persistent and/or working map; |
| 2447 |
* |
| 2448 |
* todo: optimization |
| 2449 |
*/ |
| 2450 |
void txFreeMap(struct inode *ip, |
| 2451 |
struct maplock * maplock, struct tblock * tblk, int maptype) |
| 2452 |
{ |
| 2453 |
struct inode *ipbmap = JFS_SBI(ip->i_sb)->ipbmap; |
| 2454 |
struct xdlistlock *xadlistlock; |
| 2455 |
xad_t *xad; |
| 2456 |
s64 xaddr; |
| 2457 |
int xlen; |
| 2458 |
struct pxd_lock *pxdlock; |
| 2459 |
struct xdlistlock *pxdlistlock; |
| 2460 |
pxd_t *pxd; |
| 2461 |
int n; |
| 2462 |
|
| 2463 |
jfs_info("txFreeMap: tblk:0x%p maplock:0x%p maptype:0x%x", |
| 2464 |
tblk, maplock, maptype); |
| 2465 |
|
| 2466 |
/* |
| 2467 |
* free from persistent map; |
| 2468 |
*/ |
| 2469 |
if (maptype == COMMIT_PMAP || maptype == COMMIT_PWMAP) { |
| 2470 |
if (maplock->flag & mlckFREEXADLIST) { |
| 2471 |
xadlistlock = (struct xdlistlock *) maplock; |
| 2472 |
xad = xadlistlock->xdlist; |
| 2473 |
for (n = 0; n < xadlistlock->count; n++, xad++) { |
| 2474 |
if (!(xad->flag & XAD_NEW)) { |
| 2475 |
xaddr = addressXAD(xad); |
| 2476 |
xlen = lengthXAD(xad); |
| 2477 |
dbUpdatePMap(ipbmap, true, xaddr, |
| 2478 |
(s64) xlen, tblk); |
| 2479 |
jfs_info("freePMap: xaddr:0x%lx xlen:%d", |
| 2480 |
(ulong) xaddr, xlen); |
| 2481 |
} |
| 2482 |
} |
| 2483 |
} else if (maplock->flag & mlckFREEPXD) { |
| 2484 |
pxdlock = (struct pxd_lock *) maplock; |
| 2485 |
xaddr = addressPXD(&pxdlock->pxd); |
| 2486 |
xlen = lengthPXD(&pxdlock->pxd); |
| 2487 |
dbUpdatePMap(ipbmap, true, xaddr, (s64) xlen, |
| 2488 |
tblk); |
| 2489 |
jfs_info("freePMap: xaddr:0x%lx xlen:%d", |
| 2490 |
(ulong) xaddr, xlen); |
| 2491 |
} else { /* (maplock->flag & mlckALLOCPXDLIST) */ |
| 2492 |
|
| 2493 |
pxdlistlock = (struct xdlistlock *) maplock; |
| 2494 |
pxd = pxdlistlock->xdlist; |
| 2495 |
for (n = 0; n < pxdlistlock->count; n++, pxd++) { |
| 2496 |
xaddr = addressPXD(pxd); |
| 2497 |
xlen = lengthPXD(pxd); |
| 2498 |
dbUpdatePMap(ipbmap, true, xaddr, |
| 2499 |
(s64) xlen, tblk); |
| 2500 |
jfs_info("freePMap: xaddr:0x%lx xlen:%d", |
| 2501 |
(ulong) xaddr, xlen); |
| 2502 |
} |
| 2503 |
} |
| 2504 |
} |
| 2505 |
|
| 2506 |
/* |
| 2507 |
* free from working map; |
| 2508 |
*/ |
| 2509 |
if (maptype == COMMIT_PWMAP || maptype == COMMIT_WMAP) { |
| 2510 |
if (maplock->flag & mlckFREEXADLIST) { |
| 2511 |
xadlistlock = (struct xdlistlock *) maplock; |
| 2512 |
xad = xadlistlock->xdlist; |
| 2513 |
for (n = 0; n < xadlistlock->count; n++, xad++) { |
| 2514 |
xaddr = addressXAD(xad); |
| 2515 |
xlen = lengthXAD(xad); |
| 2516 |
dbFree(ip, xaddr, (s64) xlen); |
| 2517 |
xad->flag = 0; |
| 2518 |
jfs_info("freeWMap: xaddr:0x%lx xlen:%d", |
| 2519 |
(ulong) xaddr, xlen); |
| 2520 |
} |
| 2521 |
} else if (maplock->flag & mlckFREEPXD) { |
| 2522 |
pxdlock = (struct pxd_lock *) maplock; |
| 2523 |
xaddr = addressPXD(&pxdlock->pxd); |
| 2524 |
xlen = lengthPXD(&pxdlock->pxd); |
| 2525 |
dbFree(ip, xaddr, (s64) xlen); |
| 2526 |
jfs_info("freeWMap: xaddr:0x%lx xlen:%d", |
| 2527 |
(ulong) xaddr, xlen); |
| 2528 |
} else { /* (maplock->flag & mlckFREEPXDLIST) */ |
| 2529 |
|
| 2530 |
pxdlistlock = (struct xdlistlock *) maplock; |
| 2531 |
pxd = pxdlistlock->xdlist; |
| 2532 |
for (n = 0; n < pxdlistlock->count; n++, pxd++) { |
| 2533 |
xaddr = addressPXD(pxd); |
| 2534 |
xlen = lengthPXD(pxd); |
| 2535 |
dbFree(ip, xaddr, (s64) xlen); |
| 2536 |
jfs_info("freeWMap: xaddr:0x%lx xlen:%d", |
| 2537 |
(ulong) xaddr, xlen); |
| 2538 |
} |
| 2539 |
} |
| 2540 |
} |
| 2541 |
} |
| 2542 |
|
| 2543 |
/* |
| 2544 |
* txFreelock() |
| 2545 |
* |
| 2546 |
* function: remove tlock from inode anonymous locklist |
| 2547 |
*/ |
| 2548 |
void txFreelock(struct inode *ip) |
| 2549 |
{ |
| 2550 |
struct jfs_inode_info *jfs_ip = JFS_IP(ip); |
| 2551 |
struct tlock *xtlck, *tlck; |
| 2552 |
lid_t xlid = 0, lid; |
| 2553 |
|
| 2554 |
if (!jfs_ip->atlhead) |
| 2555 |
return; |
| 2556 |
|
| 2557 |
TXN_LOCK(); |
| 2558 |
xtlck = (struct tlock *) &jfs_ip->atlhead; |
| 2559 |
|
| 2560 |
while ((lid = xtlck->next) != 0) { |
| 2561 |
tlck = lid_to_tlock(lid); |
| 2562 |
if (tlck->flag & tlckFREELOCK) { |
| 2563 |
xtlck->next = tlck->next; |
| 2564 |
txLockFree(lid); |
| 2565 |
} else { |
| 2566 |
xtlck = tlck; |
| 2567 |
xlid = lid; |
| 2568 |
} |
| 2569 |
} |
| 2570 |
|
| 2571 |
if (jfs_ip->atlhead) |
| 2572 |
jfs_ip->atltail = xlid; |
| 2573 |
else { |
| 2574 |
jfs_ip->atltail = 0; |
| 2575 |
/* |
| 2576 |
* If inode was on anon_list, remove it |
| 2577 |
*/ |
| 2578 |
list_del_init(&jfs_ip->anon_inode_list); |
| 2579 |
} |
| 2580 |
TXN_UNLOCK(); |
| 2581 |
} |
| 2582 |
|
| 2583 |
/* |
| 2584 |
* txAbort() |
| 2585 |
* |
| 2586 |
* function: abort tx before commit; |
| 2587 |
* |
| 2588 |
* frees line-locks and segment locks for all |
| 2589 |
* segments in comdata structure. |
| 2590 |
* Optionally sets state of file-system to FM_DIRTY in super-block. |
| 2591 |
* log age of page-frames in memory for which caller has |
| 2592 |
* are reset to 0 (to avoid logwarap). |
| 2593 |
*/ |
| 2594 |
void txAbort(tid_t tid, int dirty) |
| 2595 |
{ |
| 2596 |
lid_t lid, next; |
| 2597 |
struct metapage *mp; |
| 2598 |
struct tblock *tblk = tid_to_tblock(tid); |
| 2599 |
struct tlock *tlck; |
| 2600 |
|
| 2601 |
/* |
| 2602 |
* free tlocks of the transaction |
| 2603 |
*/ |
| 2604 |
for (lid = tblk->next; lid; lid = next) { |
| 2605 |
tlck = lid_to_tlock(lid); |
| 2606 |
next = tlck->next; |
| 2607 |
mp = tlck->mp; |
| 2608 |
JFS_IP(tlck->ip)->xtlid = 0; |
| 2609 |
|
| 2610 |
if (mp) { |
| 2611 |
mp->lid = 0; |
| 2612 |
|
| 2613 |
/* |
| 2614 |
* reset lsn of page to avoid logwarap: |
| 2615 |
* |
| 2616 |
* (page may have been previously committed by another |
| 2617 |
* transaction(s) but has not been paged, i.e., |
| 2618 |
* it may be on logsync list even though it has not |
| 2619 |
* been logged for the current tx.) |
| 2620 |
*/ |
| 2621 |
if (mp->xflag & COMMIT_PAGE && mp->lsn) |
| 2622 |
LogSyncRelease(mp); |
| 2623 |
} |
| 2624 |
/* insert tlock at head of freelist */ |
| 2625 |
TXN_LOCK(); |
| 2626 |
txLockFree(lid); |
| 2627 |
TXN_UNLOCK(); |
| 2628 |
} |
| 2629 |
|
| 2630 |
/* caller will free the transaction block */ |
| 2631 |
|
| 2632 |
tblk->next = tblk->last = 0; |
| 2633 |
|
| 2634 |
/* |
| 2635 |
* mark filesystem dirty |
| 2636 |
*/ |
| 2637 |
if (dirty) |
| 2638 |
jfs_error(tblk->sb, "\n"); |
| 2639 |
|
| 2640 |
return; |
| 2641 |
} |
| 2642 |
|
| 2643 |
/* |
| 2644 |
* txLazyCommit(void) |
| 2645 |
* |
| 2646 |
* All transactions except those changing ipimap (COMMIT_FORCE) are |
| 2647 |
* processed by this routine. This insures that the inode and block |
| 2648 |
* allocation maps are updated in order. For synchronous transactions, |
| 2649 |
* let the user thread finish processing after txUpdateMap() is called. |
| 2650 |
*/ |
| 2651 |
static void txLazyCommit(struct tblock * tblk) |
| 2652 |
{ |
| 2653 |
struct jfs_log *log; |
| 2654 |
|
| 2655 |
while (((tblk->flag & tblkGC_READY) == 0) && |
| 2656 |
((tblk->flag & tblkGC_UNLOCKED) == 0)) { |
| 2657 |
/* We must have gotten ahead of the user thread |
| 2658 |
*/ |
| 2659 |
jfs_info("jfs_lazycommit: tblk 0x%p not unlocked", tblk); |
| 2660 |
yield(); |
| 2661 |
} |
| 2662 |
|
| 2663 |
jfs_info("txLazyCommit: processing tblk 0x%p", tblk); |
| 2664 |
|
| 2665 |
txUpdateMap(tblk); |
| 2666 |
|
| 2667 |
log = (struct jfs_log *) JFS_SBI(tblk->sb)->log; |
| 2668 |
|
| 2669 |
spin_lock_irq(&log->gclock); // LOGGC_LOCK |
| 2670 |
|
| 2671 |
tblk->flag |= tblkGC_COMMITTED; |
| 2672 |
|
| 2673 |
if (tblk->flag & tblkGC_READY) |
| 2674 |
log->gcrtc--; |
| 2675 |
|
| 2676 |
wake_up_all(&tblk->gcwait); // LOGGC_WAKEUP |
| 2677 |
|
| 2678 |
/* |
| 2679 |
* Can't release log->gclock until we've tested tblk->flag |
| 2680 |
*/ |
| 2681 |
if (tblk->flag & tblkGC_LAZY) { |
| 2682 |
spin_unlock_irq(&log->gclock); // LOGGC_UNLOCK |
| 2683 |
txUnlock(tblk); |
| 2684 |
tblk->flag &= ~tblkGC_LAZY; |
| 2685 |
txEnd(tblk - TxBlock); /* Convert back to tid */ |
| 2686 |
} else |
| 2687 |
spin_unlock_irq(&log->gclock); // LOGGC_UNLOCK |
| 2688 |
|
| 2689 |
jfs_info("txLazyCommit: done: tblk = 0x%p", tblk); |
| 2690 |
} |
| 2691 |
|
| 2692 |
/* |
| 2693 |
* jfs_lazycommit(void) |
| 2694 |
* |
| 2695 |
* To be run as a kernel daemon. If lbmIODone is called in an interrupt |
| 2696 |
* context, or where blocking is not wanted, this routine will process |
| 2697 |
* committed transactions from the unlock queue. |
| 2698 |
*/ |
| 2699 |
int jfs_lazycommit(void *arg) |
| 2700 |
{ |
| 2701 |
int WorkDone; |
| 2702 |
struct tblock *tblk; |
| 2703 |
unsigned long flags; |
| 2704 |
struct jfs_sb_info *sbi; |
| 2705 |
|
| 2706 |
set_freezable(); |
| 2707 |
do { |
| 2708 |
LAZY_LOCK(flags); |
| 2709 |
jfs_commit_thread_waking = 0; /* OK to wake another thread */ |
| 2710 |
while (!list_empty(&TxAnchor.unlock_queue)) { |
| 2711 |
WorkDone = 0; |
| 2712 |
list_for_each_entry(tblk, &TxAnchor.unlock_queue, |
| 2713 |
cqueue) { |
| 2714 |
|
| 2715 |
sbi = JFS_SBI(tblk->sb); |
| 2716 |
/* |
| 2717 |
* For each volume, the transactions must be |
| 2718 |
* handled in order. If another commit thread |
| 2719 |
* is handling a tblk for this superblock, |
| 2720 |
* skip it |
| 2721 |
*/ |
| 2722 |
if (sbi->commit_state & IN_LAZYCOMMIT) |
| 2723 |
continue; |
| 2724 |
|
| 2725 |
sbi->commit_state |= IN_LAZYCOMMIT; |
| 2726 |
WorkDone = 1; |
| 2727 |
|
| 2728 |
/* |
| 2729 |
* Remove transaction from queue |
| 2730 |
*/ |
| 2731 |
list_del(&tblk->cqueue); |
| 2732 |
|
| 2733 |
LAZY_UNLOCK(flags); |
| 2734 |
txLazyCommit(tblk); |
| 2735 |
LAZY_LOCK(flags); |
| 2736 |
|
| 2737 |
sbi->commit_state &= ~IN_LAZYCOMMIT; |
| 2738 |
/* |
| 2739 |
* Don't continue in the for loop. (We can't |
| 2740 |
* anyway, it's unsafe!) We want to go back to |
| 2741 |
* the beginning of the list. |
| 2742 |
*/ |
| 2743 |
break; |
| 2744 |
} |
| 2745 |
|
| 2746 |
/* If there was nothing to do, don't continue */ |
| 2747 |
if (!WorkDone) |
| 2748 |
break; |
| 2749 |
} |
| 2750 |
/* In case a wakeup came while all threads were active */ |
| 2751 |
jfs_commit_thread_waking = 0; |
| 2752 |
|
| 2753 |
if (freezing(current)) { |
| 2754 |
LAZY_UNLOCK(flags); |
| 2755 |
try_to_freeze(); |
| 2756 |
} else { |
| 2757 |
DECLARE_WAITQUEUE(wq, current); |
| 2758 |
|
| 2759 |
add_wait_queue(&jfs_commit_thread_wait, &wq); |
| 2760 |
set_current_state(TASK_INTERRUPTIBLE); |
| 2761 |
LAZY_UNLOCK(flags); |
| 2762 |
schedule(); |
| 2763 |
remove_wait_queue(&jfs_commit_thread_wait, &wq); |
| 2764 |
} |
| 2765 |
} while (!kthread_should_stop()); |
| 2766 |
|
| 2767 |
if (!list_empty(&TxAnchor.unlock_queue)) |
| 2768 |
jfs_err("jfs_lazycommit being killed w/pending transactions!"); |
| 2769 |
else |
| 2770 |
jfs_info("jfs_lazycommit being killed"); |
| 2771 |
return 0; |
| 2772 |
} |
| 2773 |
|
| 2774 |
void txLazyUnlock(struct tblock * tblk) |
| 2775 |
{ |
| 2776 |
unsigned long flags; |
| 2777 |
|
| 2778 |
LAZY_LOCK(flags); |
| 2779 |
|
| 2780 |
list_add_tail(&tblk->cqueue, &TxAnchor.unlock_queue); |
| 2781 |
/* |
| 2782 |
* Don't wake up a commit thread if there is already one servicing |
| 2783 |
* this superblock, or if the last one we woke up hasn't started yet. |
| 2784 |
*/ |
| 2785 |
if (!(JFS_SBI(tblk->sb)->commit_state & IN_LAZYCOMMIT) && |
| 2786 |
!jfs_commit_thread_waking) { |
| 2787 |
jfs_commit_thread_waking = 1; |
| 2788 |
wake_up(&jfs_commit_thread_wait); |
| 2789 |
} |
| 2790 |
LAZY_UNLOCK(flags); |
| 2791 |
} |
| 2792 |
|
| 2793 |
static void LogSyncRelease(struct metapage * mp) |
| 2794 |
{ |
| 2795 |
struct jfs_log *log = mp->log; |
| 2796 |
|
| 2797 |
assert(mp->nohomeok); |
| 2798 |
assert(log); |
| 2799 |
metapage_homeok(mp); |
| 2800 |
} |
| 2801 |
|
| 2802 |
/* |
| 2803 |
* txQuiesce |
| 2804 |
* |
| 2805 |
* Block all new transactions and push anonymous transactions to |
| 2806 |
* completion |
| 2807 |
* |
| 2808 |
* This does almost the same thing as jfs_sync below. We don't |
| 2809 |
* worry about deadlocking when jfs_tlocks_low is set, since we would |
| 2810 |
* expect jfs_sync to get us out of that jam. |
| 2811 |
*/ |
| 2812 |
void txQuiesce(struct super_block *sb) |
| 2813 |
{ |
| 2814 |
struct inode *ip; |
| 2815 |
struct jfs_inode_info *jfs_ip; |
| 2816 |
struct jfs_log *log = JFS_SBI(sb)->log; |
| 2817 |
tid_t tid; |
| 2818 |
|
| 2819 |
set_bit(log_QUIESCE, &log->flag); |
| 2820 |
|
| 2821 |
TXN_LOCK(); |
| 2822 |
restart: |
| 2823 |
while (!list_empty(&TxAnchor.anon_list)) { |
| 2824 |
jfs_ip = list_entry(TxAnchor.anon_list.next, |
| 2825 |
struct jfs_inode_info, |
| 2826 |
anon_inode_list); |
| 2827 |
ip = &jfs_ip->vfs_inode; |
| 2828 |
|
| 2829 |
/* |
| 2830 |
* inode will be removed from anonymous list |
| 2831 |
* when it is committed |
| 2832 |
*/ |
| 2833 |
TXN_UNLOCK(); |
| 2834 |
tid = txBegin(ip->i_sb, COMMIT_INODE | COMMIT_FORCE); |
| 2835 |
mutex_lock(&jfs_ip->commit_mutex); |
| 2836 |
txCommit(tid, 1, &ip, 0); |
| 2837 |
txEnd(tid); |
| 2838 |
mutex_unlock(&jfs_ip->commit_mutex); |
| 2839 |
/* |
| 2840 |
* Just to be safe. I don't know how |
| 2841 |
* long we can run without blocking |
| 2842 |
*/ |
| 2843 |
cond_resched(); |
| 2844 |
TXN_LOCK(); |
| 2845 |
} |
| 2846 |
|
| 2847 |
/* |
| 2848 |
* If jfs_sync is running in parallel, there could be some inodes |
| 2849 |
* on anon_list2. Let's check. |
| 2850 |
*/ |
| 2851 |
if (!list_empty(&TxAnchor.anon_list2)) { |
| 2852 |
list_splice_init(&TxAnchor.anon_list2, &TxAnchor.anon_list); |
| 2853 |
goto restart; |
| 2854 |
} |
| 2855 |
TXN_UNLOCK(); |
| 2856 |
|
| 2857 |
/* |
| 2858 |
* We may need to kick off the group commit |
| 2859 |
*/ |
| 2860 |
jfs_flush_journal(log, 0); |
| 2861 |
} |
| 2862 |
|
| 2863 |
/* |
| 2864 |
* txResume() |
| 2865 |
* |
| 2866 |
* Allows transactions to start again following txQuiesce |
| 2867 |
*/ |
| 2868 |
void txResume(struct super_block *sb) |
| 2869 |
{ |
| 2870 |
struct jfs_log *log = JFS_SBI(sb)->log; |
| 2871 |
|
| 2872 |
clear_bit(log_QUIESCE, &log->flag); |
| 2873 |
TXN_WAKEUP(&log->syncwait); |
| 2874 |
} |
| 2875 |
|
| 2876 |
/* |
| 2877 |
* jfs_sync(void) |
| 2878 |
* |
| 2879 |
* To be run as a kernel daemon. This is awakened when tlocks run low. |
| 2880 |
* We write any inodes that have anonymous tlocks so they will become |
| 2881 |
* available. |
| 2882 |
*/ |
| 2883 |
int jfs_sync(void *arg) |
| 2884 |
{ |
| 2885 |
struct inode *ip; |
| 2886 |
struct jfs_inode_info *jfs_ip; |
| 2887 |
tid_t tid; |
| 2888 |
|
| 2889 |
set_freezable(); |
| 2890 |
do { |
| 2891 |
/* |
| 2892 |
* write each inode on the anonymous inode list |
| 2893 |
*/ |
| 2894 |
TXN_LOCK(); |
| 2895 |
while (jfs_tlocks_low && !list_empty(&TxAnchor.anon_list)) { |
| 2896 |
jfs_ip = list_entry(TxAnchor.anon_list.next, |
| 2897 |
struct jfs_inode_info, |
| 2898 |
anon_inode_list); |
| 2899 |
ip = &jfs_ip->vfs_inode; |
| 2900 |
|
| 2901 |
if (! igrab(ip)) { |
| 2902 |
/* |
| 2903 |
* Inode is being freed |
| 2904 |
*/ |
| 2905 |
list_del_init(&jfs_ip->anon_inode_list); |
| 2906 |
} else if (mutex_trylock(&jfs_ip->commit_mutex)) { |
| 2907 |
/* |
| 2908 |
* inode will be removed from anonymous list |
| 2909 |
* when it is committed |
| 2910 |
*/ |
| 2911 |
TXN_UNLOCK(); |
| 2912 |
tid = txBegin(ip->i_sb, COMMIT_INODE); |
| 2913 |
txCommit(tid, 1, &ip, 0); |
| 2914 |
txEnd(tid); |
| 2915 |
mutex_unlock(&jfs_ip->commit_mutex); |
| 2916 |
|
| 2917 |
iput(ip); |
| 2918 |
/* |
| 2919 |
* Just to be safe. I don't know how |
| 2920 |
* long we can run without blocking |
| 2921 |
*/ |
| 2922 |
cond_resched(); |
| 2923 |
TXN_LOCK(); |
| 2924 |
} else { |
| 2925 |
/* We can't get the commit mutex. It may |
| 2926 |
* be held by a thread waiting for tlock's |
| 2927 |
* so let's not block here. Save it to |
| 2928 |
* put back on the anon_list. |
| 2929 |
*/ |
| 2930 |
|
| 2931 |
/* Move from anon_list to anon_list2 */ |
| 2932 |
list_move(&jfs_ip->anon_inode_list, |
| 2933 |
&TxAnchor.anon_list2); |
| 2934 |
|
| 2935 |
TXN_UNLOCK(); |
| 2936 |
iput(ip); |
| 2937 |
TXN_LOCK(); |
| 2938 |
} |
| 2939 |
} |
| 2940 |
/* Add anon_list2 back to anon_list */ |
| 2941 |
list_splice_init(&TxAnchor.anon_list2, &TxAnchor.anon_list); |
| 2942 |
|
| 2943 |
if (freezing(current)) { |
| 2944 |
TXN_UNLOCK(); |
| 2945 |
try_to_freeze(); |
| 2946 |
} else { |
| 2947 |
set_current_state(TASK_INTERRUPTIBLE); |
| 2948 |
TXN_UNLOCK(); |
| 2949 |
schedule(); |
| 2950 |
} |
| 2951 |
} while (!kthread_should_stop()); |
| 2952 |
|
| 2953 |
jfs_info("jfs_sync being killed"); |
| 2954 |
return 0; |
| 2955 |
} |
| 2956 |
|
| 2957 |
#if defined(CONFIG_PROC_FS) && defined(CONFIG_JFS_DEBUG) |
| 2958 |
int jfs_txanchor_proc_show(struct seq_file *m, void *v) |
| 2959 |
{ |
| 2960 |
char *freewait; |
| 2961 |
char *freelockwait; |
| 2962 |
char *lowlockwait; |
| 2963 |
|
| 2964 |
freewait = |
| 2965 |
waitqueue_active(&TxAnchor.freewait) ? "active" : "empty"; |
| 2966 |
freelockwait = |
| 2967 |
waitqueue_active(&TxAnchor.freelockwait) ? "active" : "empty"; |
| 2968 |
lowlockwait = |
| 2969 |
waitqueue_active(&TxAnchor.lowlockwait) ? "active" : "empty"; |
| 2970 |
|
| 2971 |
seq_printf(m, |
| 2972 |
"JFS TxAnchor\n" |
| 2973 |
"============\n" |
| 2974 |
"freetid = %d\n" |
| 2975 |
"freewait = %s\n" |
| 2976 |
"freelock = %d\n" |
| 2977 |
"freelockwait = %s\n" |
| 2978 |
"lowlockwait = %s\n" |
| 2979 |
"tlocksInUse = %d\n" |
| 2980 |
"jfs_tlocks_low = %d\n" |
| 2981 |
"unlock_queue is %sempty\n", |
| 2982 |
TxAnchor.freetid, |
| 2983 |
freewait, |
| 2984 |
TxAnchor.freelock, |
| 2985 |
freelockwait, |
| 2986 |
lowlockwait, |
| 2987 |
TxAnchor.tlocksInUse, |
| 2988 |
jfs_tlocks_low, |
| 2989 |
list_empty(&TxAnchor.unlock_queue) ? "" : "not "); |
| 2990 |
return 0; |
| 2991 |
} |
| 2992 |
#endif |
| 2993 |
|
| 2994 |
#if defined(CONFIG_PROC_FS) && defined(CONFIG_JFS_STATISTICS) |
| 2995 |
int jfs_txstats_proc_show(struct seq_file *m, void *v) |
| 2996 |
{ |
| 2997 |
seq_printf(m, |
| 2998 |
"JFS TxStats\n" |
| 2999 |
"===========\n" |
| 3000 |
"calls to txBegin = %d\n" |
| 3001 |
"txBegin blocked by sync barrier = %d\n" |
| 3002 |
"txBegin blocked by tlocks low = %d\n" |
| 3003 |
"txBegin blocked by no free tid = %d\n" |
| 3004 |
"calls to txBeginAnon = %d\n" |
| 3005 |
"txBeginAnon blocked by sync barrier = %d\n" |
| 3006 |
"txBeginAnon blocked by tlocks low = %d\n" |
| 3007 |
"calls to txLockAlloc = %d\n" |
| 3008 |
"tLockAlloc blocked by no free lock = %d\n", |
| 3009 |
TxStat.txBegin, |
| 3010 |
TxStat.txBegin_barrier, |
| 3011 |
TxStat.txBegin_lockslow, |
| 3012 |
TxStat.txBegin_freetid, |
| 3013 |
TxStat.txBeginAnon, |
| 3014 |
TxStat.txBeginAnon_barrier, |
| 3015 |
TxStat.txBeginAnon_lockslow, |
| 3016 |
TxStat.txLockAlloc, |
| 3017 |
TxStat.txLockAlloc_freelock); |
| 3018 |
return 0; |
| 3019 |
} |
| 3020 |
#endif |