lfs_subr.c revision 1.86
1/* $NetBSD: lfs_subr.c,v 1.86 2015/10/03 08:28:16 dholland Exp $ */ 2 3/*- 4 * Copyright (c) 1999, 2000, 2001, 2002, 2003 The NetBSD Foundation, Inc. 5 * All rights reserved. 6 * 7 * This code is derived from software contributed to The NetBSD Foundation 8 * by Konrad E. Schroder <perseant@hhhh.org>. 9 * 10 * Redistribution and use in source and binary forms, with or without 11 * modification, are permitted provided that the following conditions 12 * are met: 13 * 1. Redistributions of source code must retain the above copyright 14 * notice, this list of conditions and the following disclaimer. 15 * 2. Redistributions in binary form must reproduce the above copyright 16 * notice, this list of conditions and the following disclaimer in the 17 * documentation and/or other materials provided with the distribution. 18 * 19 * THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS 20 * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED 21 * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR 22 * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS 23 * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR 24 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF 25 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS 26 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN 27 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) 28 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE 29 * POSSIBILITY OF SUCH DAMAGE. 30 */ 31/* 32 * Copyright (c) 1991, 1993 33 * The Regents of the University of California. All rights reserved. 34 * 35 * Redistribution and use in source and binary forms, with or without 36 * modification, are permitted provided that the following conditions 37 * are met: 38 * 1. Redistributions of source code must retain the above copyright 39 * notice, this list of conditions and the following disclaimer. 40 * 2. Redistributions in binary form must reproduce the above copyright 41 * notice, this list of conditions and the following disclaimer in the 42 * documentation and/or other materials provided with the distribution. 43 * 3. Neither the name of the University nor the names of its contributors 44 * may be used to endorse or promote products derived from this software 45 * without specific prior written permission. 46 * 47 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 48 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 49 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 50 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 51 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 52 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 53 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 54 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 55 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 56 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 57 * SUCH DAMAGE. 58 * 59 * @(#)lfs_subr.c 8.4 (Berkeley) 5/8/95 60 */ 61 62#include <sys/cdefs.h> 63__KERNEL_RCSID(0, "$NetBSD: lfs_subr.c,v 1.86 2015/10/03 08:28:16 dholland Exp $"); 64 65#include <sys/param.h> 66#include <sys/systm.h> 67#include <sys/namei.h> 68#include <sys/vnode.h> 69#include <sys/buf.h> 70#include <sys/mount.h> 71#include <sys/malloc.h> 72#include <sys/proc.h> 73#include <sys/kauth.h> 74 75#include <ufs/lfs/ulfs_inode.h> 76#include <ufs/lfs/lfs.h> 77#include <ufs/lfs/lfs_accessors.h> 78#include <ufs/lfs/lfs_kernel.h> 79#include <ufs/lfs/lfs_extern.h> 80 81#include <uvm/uvm.h> 82 83#ifdef DEBUG 84const char *lfs_res_names[LFS_NB_COUNT] = { 85 "summary", 86 "superblock", 87 "file block", 88 "cluster", 89 "clean", 90 "blkiov", 91}; 92#endif 93 94int lfs_res_qty[LFS_NB_COUNT] = { 95 LFS_N_SUMMARIES, 96 LFS_N_SBLOCKS, 97 LFS_N_IBLOCKS, 98 LFS_N_CLUSTERS, 99 LFS_N_CLEAN, 100 LFS_N_BLKIOV, 101}; 102 103void 104lfs_setup_resblks(struct lfs *fs) 105{ 106 int i, j; 107 int maxbpp; 108 109 ASSERT_NO_SEGLOCK(fs); 110 fs->lfs_resblk = malloc(LFS_N_TOTAL * sizeof(res_t), M_SEGMENT, 111 M_WAITOK); 112 for (i = 0; i < LFS_N_TOTAL; i++) { 113 fs->lfs_resblk[i].inuse = 0; 114 fs->lfs_resblk[i].p = NULL; 115 } 116 for (i = 0; i < LFS_RESHASH_WIDTH; i++) 117 LIST_INIT(fs->lfs_reshash + i); 118 119 /* 120 * These types of allocations can be larger than a page, 121 * so we can't use the pool subsystem for them. 122 */ 123 for (i = 0, j = 0; j < LFS_N_SUMMARIES; j++, i++) 124 fs->lfs_resblk[i].size = lfs_sb_getsumsize(fs); 125 for (j = 0; j < LFS_N_SBLOCKS; j++, i++) 126 fs->lfs_resblk[i].size = LFS_SBPAD; 127 for (j = 0; j < LFS_N_IBLOCKS; j++, i++) 128 fs->lfs_resblk[i].size = lfs_sb_getbsize(fs); 129 for (j = 0; j < LFS_N_CLUSTERS; j++, i++) 130 fs->lfs_resblk[i].size = MAXPHYS; 131 for (j = 0; j < LFS_N_CLEAN; j++, i++) 132 fs->lfs_resblk[i].size = MAXPHYS; 133 for (j = 0; j < LFS_N_BLKIOV; j++, i++) 134 fs->lfs_resblk[i].size = LFS_MARKV_MAXBLKCNT * sizeof(BLOCK_INFO); 135 136 for (i = 0; i < LFS_N_TOTAL; i++) { 137 fs->lfs_resblk[i].p = malloc(fs->lfs_resblk[i].size, 138 M_SEGMENT, M_WAITOK); 139 } 140 141 /* 142 * Initialize pools for small types (XXX is BPP small?) 143 */ 144 pool_init(&fs->lfs_clpool, sizeof(struct lfs_cluster), 0, 0, 0, 145 "lfsclpl", &pool_allocator_nointr, IPL_NONE); 146 pool_init(&fs->lfs_segpool, sizeof(struct segment), 0, 0, 0, 147 "lfssegpool", &pool_allocator_nointr, IPL_NONE); 148 /* XXX: should this int32 be 32/64? */ 149 maxbpp = ((lfs_sb_getsumsize(fs) - SEGSUM_SIZE(fs)) / sizeof(int32_t) + 2); 150 maxbpp = MIN(maxbpp, lfs_segsize(fs) / lfs_sb_getfsize(fs) + 2); 151 pool_init(&fs->lfs_bpppool, maxbpp * sizeof(struct buf *), 0, 0, 0, 152 "lfsbpppl", &pool_allocator_nointr, IPL_NONE); 153} 154 155void 156lfs_free_resblks(struct lfs *fs) 157{ 158 int i; 159 160 pool_destroy(&fs->lfs_bpppool); 161 pool_destroy(&fs->lfs_segpool); 162 pool_destroy(&fs->lfs_clpool); 163 164 mutex_enter(&lfs_lock); 165 for (i = 0; i < LFS_N_TOTAL; i++) { 166 while (fs->lfs_resblk[i].inuse) 167 mtsleep(&fs->lfs_resblk, PRIBIO + 1, "lfs_free", 0, 168 &lfs_lock); 169 if (fs->lfs_resblk[i].p != NULL) 170 free(fs->lfs_resblk[i].p, M_SEGMENT); 171 } 172 free(fs->lfs_resblk, M_SEGMENT); 173 mutex_exit(&lfs_lock); 174} 175 176static unsigned int 177lfs_mhash(void *vp) 178{ 179 return (unsigned int)(((unsigned long)vp) >> 2) % LFS_RESHASH_WIDTH; 180} 181 182/* 183 * Return memory of the given size for the given purpose, or use one of a 184 * number of spare last-resort buffers, if malloc returns NULL. 185 */ 186void * 187lfs_malloc(struct lfs *fs, size_t size, int type) 188{ 189 struct lfs_res_blk *re; 190 void *r; 191 int i, s, start; 192 unsigned int h; 193 194 ASSERT_MAYBE_SEGLOCK(fs); 195 r = NULL; 196 197 /* If no mem allocated for this type, it just waits */ 198 if (lfs_res_qty[type] == 0) { 199 r = malloc(size, M_SEGMENT, M_WAITOK); 200 return r; 201 } 202 203 /* Otherwise try a quick malloc, and if it works, great */ 204 if ((r = malloc(size, M_SEGMENT, M_NOWAIT)) != NULL) { 205 return r; 206 } 207 208 /* 209 * If malloc returned NULL, we are forced to use one of our 210 * reserve blocks. We have on hand at least one summary block, 211 * at least one cluster block, at least one superblock, 212 * and several indirect blocks. 213 */ 214 215 mutex_enter(&lfs_lock); 216 /* skip over blocks of other types */ 217 for (i = 0, start = 0; i < type; i++) 218 start += lfs_res_qty[i]; 219 while (r == NULL) { 220 for (i = 0; i < lfs_res_qty[type]; i++) { 221 if (fs->lfs_resblk[start + i].inuse == 0) { 222 re = fs->lfs_resblk + start + i; 223 re->inuse = 1; 224 r = re->p; 225 KASSERT(re->size >= size); 226 h = lfs_mhash(r); 227 s = splbio(); 228 LIST_INSERT_HEAD(&fs->lfs_reshash[h], re, res); 229 splx(s); 230 mutex_exit(&lfs_lock); 231 return r; 232 } 233 } 234 DLOG((DLOG_MALLOC, "sleeping on %s (%d)\n", 235 lfs_res_names[type], lfs_res_qty[type])); 236 mtsleep(&fs->lfs_resblk, PVM, "lfs_malloc", 0, 237 &lfs_lock); 238 DLOG((DLOG_MALLOC, "done sleeping on %s\n", 239 lfs_res_names[type])); 240 } 241 /* NOTREACHED */ 242 mutex_exit(&lfs_lock); 243 return r; 244} 245 246void 247lfs_free(struct lfs *fs, void *p, int type) 248{ 249 int s; 250 unsigned int h; 251 res_t *re; 252#ifdef DEBUG 253 int i; 254#endif 255 256 ASSERT_MAYBE_SEGLOCK(fs); 257 h = lfs_mhash(p); 258 mutex_enter(&lfs_lock); 259 s = splbio(); 260 LIST_FOREACH(re, &fs->lfs_reshash[h], res) { 261 if (re->p == p) { 262 KASSERT(re->inuse == 1); 263 LIST_REMOVE(re, res); 264 re->inuse = 0; 265 wakeup(&fs->lfs_resblk); 266 splx(s); 267 mutex_exit(&lfs_lock); 268 return; 269 } 270 } 271#ifdef DEBUG 272 for (i = 0; i < LFS_N_TOTAL; i++) { 273 if (fs->lfs_resblk[i].p == p) 274 panic("lfs_free: inconsistent reserved block"); 275 } 276#endif 277 splx(s); 278 mutex_exit(&lfs_lock); 279 280 /* 281 * If we didn't find it, free it. 282 */ 283 free(p, M_SEGMENT); 284} 285 286/* 287 * lfs_seglock -- 288 * Single thread the segment writer. 289 */ 290int 291lfs_seglock(struct lfs *fs, unsigned long flags) 292{ 293 struct segment *sp; 294 295 mutex_enter(&lfs_lock); 296 if (fs->lfs_seglock) { 297 if (fs->lfs_lockpid == curproc->p_pid && 298 fs->lfs_locklwp == curlwp->l_lid) { 299 ++fs->lfs_seglock; 300 fs->lfs_sp->seg_flags |= flags; 301 mutex_exit(&lfs_lock); 302 return 0; 303 } else if (flags & SEGM_PAGEDAEMON) { 304 mutex_exit(&lfs_lock); 305 return EWOULDBLOCK; 306 } else { 307 while (fs->lfs_seglock) { 308 (void)mtsleep(&fs->lfs_seglock, PRIBIO + 1, 309 "lfs_seglock", 0, &lfs_lock); 310 } 311 } 312 } 313 314 fs->lfs_seglock = 1; 315 fs->lfs_lockpid = curproc->p_pid; 316 fs->lfs_locklwp = curlwp->l_lid; 317 mutex_exit(&lfs_lock); 318 fs->lfs_cleanind = 0; 319 320#ifdef DEBUG 321 LFS_ENTER_LOG("seglock", __FILE__, __LINE__, 0, flags, curproc->p_pid); 322#endif 323 /* Drain fragment size changes out */ 324 rw_enter(&fs->lfs_fraglock, RW_WRITER); 325 326 sp = fs->lfs_sp = pool_get(&fs->lfs_segpool, PR_WAITOK); 327 sp->bpp = pool_get(&fs->lfs_bpppool, PR_WAITOK); 328 sp->seg_flags = flags; 329 sp->vp = NULL; 330 sp->seg_iocount = 0; 331 (void) lfs_initseg(fs); 332 333 /* 334 * Keep a cumulative count of the outstanding I/O operations. If the 335 * disk drive catches up with us it could go to zero before we finish, 336 * so we artificially increment it by one until we've scheduled all of 337 * the writes we intend to do. 338 */ 339 mutex_enter(&lfs_lock); 340 ++fs->lfs_iocount; 341 fs->lfs_startseg = lfs_sb_getcurseg(fs); 342 mutex_exit(&lfs_lock); 343 return 0; 344} 345 346static void lfs_unmark_dirop(struct lfs *); 347 348static void 349lfs_unmark_dirop(struct lfs *fs) 350{ 351 struct inode *ip, *nip; 352 struct vnode *vp; 353 int doit; 354 355 ASSERT_NO_SEGLOCK(fs); 356 mutex_enter(&lfs_lock); 357 doit = !(fs->lfs_flags & LFS_UNDIROP); 358 if (doit) 359 fs->lfs_flags |= LFS_UNDIROP; 360 if (!doit) { 361 mutex_exit(&lfs_lock); 362 return; 363 } 364 365 for (ip = TAILQ_FIRST(&fs->lfs_dchainhd); ip != NULL; ip = nip) { 366 nip = TAILQ_NEXT(ip, i_lfs_dchain); 367 vp = ITOV(ip); 368 if ((ip->i_flag & (IN_ADIROP | IN_CDIROP)) == IN_CDIROP) { 369 --lfs_dirvcount; 370 --fs->lfs_dirvcount; 371 vp->v_uflag &= ~VU_DIROP; 372 TAILQ_REMOVE(&fs->lfs_dchainhd, ip, i_lfs_dchain); 373 wakeup(&lfs_dirvcount); 374 fs->lfs_unlockvp = vp; 375 mutex_exit(&lfs_lock); 376 vrele(vp); 377 mutex_enter(&lfs_lock); 378 fs->lfs_unlockvp = NULL; 379 ip->i_flag &= ~IN_CDIROP; 380 } 381 } 382 383 fs->lfs_flags &= ~LFS_UNDIROP; 384 wakeup(&fs->lfs_flags); 385 mutex_exit(&lfs_lock); 386} 387 388static void 389lfs_auto_segclean(struct lfs *fs) 390{ 391 int i, error, s, waited; 392 393 ASSERT_SEGLOCK(fs); 394 /* 395 * Now that we've swapped lfs_activesb, but while we still 396 * hold the segment lock, run through the segment list marking 397 * the empty ones clean. 398 * XXX - do we really need to do them all at once? 399 */ 400 waited = 0; 401 for (i = 0; i < lfs_sb_getnseg(fs); i++) { 402 if ((fs->lfs_suflags[0][i] & 403 (SEGUSE_ACTIVE | SEGUSE_DIRTY | SEGUSE_EMPTY)) == 404 (SEGUSE_DIRTY | SEGUSE_EMPTY) && 405 (fs->lfs_suflags[1][i] & 406 (SEGUSE_ACTIVE | SEGUSE_DIRTY | SEGUSE_EMPTY)) == 407 (SEGUSE_DIRTY | SEGUSE_EMPTY)) { 408 409 /* Make sure the sb is written before we clean */ 410 mutex_enter(&lfs_lock); 411 s = splbio(); 412 while (waited == 0 && fs->lfs_sbactive) 413 mtsleep(&fs->lfs_sbactive, PRIBIO+1, "lfs asb", 414 0, &lfs_lock); 415 splx(s); 416 mutex_exit(&lfs_lock); 417 waited = 1; 418 419 if ((error = lfs_do_segclean(fs, i)) != 0) { 420 DLOG((DLOG_CLEAN, "lfs_auto_segclean: lfs_do_segclean returned %d for seg %d\n", error, i)); 421 } 422 } 423 fs->lfs_suflags[1 - fs->lfs_activesb][i] = 424 fs->lfs_suflags[fs->lfs_activesb][i]; 425 } 426} 427 428/* 429 * lfs_segunlock -- 430 * Single thread the segment writer. 431 */ 432void 433lfs_segunlock(struct lfs *fs) 434{ 435 struct segment *sp; 436 unsigned long sync, ckp; 437 struct buf *bp; 438 int do_unmark_dirop = 0; 439 440 sp = fs->lfs_sp; 441 442 mutex_enter(&lfs_lock); 443 KASSERT(LFS_SEGLOCK_HELD(fs)); 444 if (fs->lfs_seglock == 1) { 445 if ((sp->seg_flags & (SEGM_PROT | SEGM_CLEAN)) == 0) 446 do_unmark_dirop = 1; 447 mutex_exit(&lfs_lock); 448 sync = sp->seg_flags & SEGM_SYNC; 449 ckp = sp->seg_flags & SEGM_CKP; 450 451 /* We should have a segment summary, and nothing else */ 452 KASSERT(sp->cbpp == sp->bpp + 1); 453 454 /* Free allocated segment summary */ 455 lfs_sb_suboffset(fs, lfs_btofsb(fs, lfs_sb_getsumsize(fs))); 456 bp = *sp->bpp; 457 lfs_freebuf(fs, bp); 458 459 pool_put(&fs->lfs_bpppool, sp->bpp); 460 sp->bpp = NULL; 461 462 /* 463 * If we're not sync, we're done with sp, get rid of it. 464 * Otherwise, we keep a local copy around but free 465 * fs->lfs_sp so another process can use it (we have to 466 * wait but they don't have to wait for us). 467 */ 468 if (!sync) 469 pool_put(&fs->lfs_segpool, sp); 470 fs->lfs_sp = NULL; 471 472 /* 473 * If the I/O count is non-zero, sleep until it reaches zero. 474 * At the moment, the user's process hangs around so we can 475 * sleep. 476 */ 477 mutex_enter(&lfs_lock); 478 if (--fs->lfs_iocount == 0) { 479 LFS_DEBUG_COUNTLOCKED("lfs_segunlock"); 480 } 481 if (fs->lfs_iocount <= 1) 482 wakeup(&fs->lfs_iocount); 483 mutex_exit(&lfs_lock); 484 /* 485 * If we're not checkpointing, we don't have to block 486 * other processes to wait for a synchronous write 487 * to complete. 488 */ 489 if (!ckp) { 490#ifdef DEBUG 491 LFS_ENTER_LOG("segunlock_std", __FILE__, __LINE__, 0, 0, curproc->p_pid); 492#endif 493 mutex_enter(&lfs_lock); 494 --fs->lfs_seglock; 495 fs->lfs_lockpid = 0; 496 fs->lfs_locklwp = 0; 497 mutex_exit(&lfs_lock); 498 wakeup(&fs->lfs_seglock); 499 } 500 /* 501 * We let checkpoints happen asynchronously. That means 502 * that during recovery, we have to roll forward between 503 * the two segments described by the first and second 504 * superblocks to make sure that the checkpoint described 505 * by a superblock completed. 506 */ 507 mutex_enter(&lfs_lock); 508 while (ckp && sync && fs->lfs_iocount) { 509 (void)mtsleep(&fs->lfs_iocount, PRIBIO + 1, 510 "lfs_iocount", 0, &lfs_lock); 511 DLOG((DLOG_SEG, "sleeping on iocount %x == %d\n", fs, fs->lfs_iocount)); 512 } 513 while (sync && sp->seg_iocount) { 514 (void)mtsleep(&sp->seg_iocount, PRIBIO + 1, 515 "seg_iocount", 0, &lfs_lock); 516 DLOG((DLOG_SEG, "sleeping on iocount %x == %d\n", sp, sp->seg_iocount)); 517 } 518 mutex_exit(&lfs_lock); 519 if (sync) 520 pool_put(&fs->lfs_segpool, sp); 521 522 if (ckp) { 523 fs->lfs_nactive = 0; 524 /* If we *know* everything's on disk, write both sbs */ 525 /* XXX should wait for this one */ 526 if (sync) 527 lfs_writesuper(fs, lfs_sb_getsboff(fs, fs->lfs_activesb)); 528 lfs_writesuper(fs, lfs_sb_getsboff(fs, 1 - fs->lfs_activesb)); 529 if (!(fs->lfs_ivnode->v_mount->mnt_iflag & IMNT_UNMOUNT)) { 530 lfs_auto_segclean(fs); 531 /* If sync, we can clean the remainder too */ 532 if (sync) 533 lfs_auto_segclean(fs); 534 } 535 fs->lfs_activesb = 1 - fs->lfs_activesb; 536#ifdef DEBUG 537 LFS_ENTER_LOG("segunlock_ckp", __FILE__, __LINE__, 0, 0, curproc->p_pid); 538#endif 539 mutex_enter(&lfs_lock); 540 --fs->lfs_seglock; 541 fs->lfs_lockpid = 0; 542 fs->lfs_locklwp = 0; 543 mutex_exit(&lfs_lock); 544 wakeup(&fs->lfs_seglock); 545 } 546 /* Reenable fragment size changes */ 547 rw_exit(&fs->lfs_fraglock); 548 if (do_unmark_dirop) 549 lfs_unmark_dirop(fs); 550 } else if (fs->lfs_seglock == 0) { 551 mutex_exit(&lfs_lock); 552 panic ("Seglock not held"); 553 } else { 554 --fs->lfs_seglock; 555 mutex_exit(&lfs_lock); 556 } 557} 558 559/* 560 * Drain dirops and start writer. 561 * 562 * No simple_locks are held when we enter and none are held when we return. 563 */ 564int 565lfs_writer_enter(struct lfs *fs, const char *wmesg) 566{ 567 int error = 0; 568 569 ASSERT_MAYBE_SEGLOCK(fs); 570 mutex_enter(&lfs_lock); 571 572 /* disallow dirops during flush */ 573 fs->lfs_writer++; 574 575 while (fs->lfs_dirops > 0) { 576 ++fs->lfs_diropwait; 577 error = mtsleep(&fs->lfs_writer, PRIBIO+1, wmesg, 0, 578 &lfs_lock); 579 --fs->lfs_diropwait; 580 } 581 582 if (error) 583 fs->lfs_writer--; 584 585 mutex_exit(&lfs_lock); 586 587 return error; 588} 589 590void 591lfs_writer_leave(struct lfs *fs) 592{ 593 bool dowakeup; 594 595 ASSERT_MAYBE_SEGLOCK(fs); 596 mutex_enter(&lfs_lock); 597 dowakeup = !(--fs->lfs_writer); 598 mutex_exit(&lfs_lock); 599 if (dowakeup) 600 wakeup(&fs->lfs_dirops); 601} 602 603/* 604 * Unlock, wait for the cleaner, then relock to where we were before. 605 * To be used only at a fairly high level, to address a paucity of free 606 * segments propagated back from lfs_gop_write(). 607 */ 608void 609lfs_segunlock_relock(struct lfs *fs) 610{ 611 int n = fs->lfs_seglock; 612 u_int16_t seg_flags; 613 CLEANERINFO *cip; 614 struct buf *bp; 615 616 if (n == 0) 617 return; 618 619 /* Write anything we've already gathered to disk */ 620 lfs_writeseg(fs, fs->lfs_sp); 621 622 /* Tell cleaner */ 623 LFS_CLEANERINFO(cip, fs, bp); 624 lfs_ci_setflags(fs, cip, 625 lfs_ci_getflags(fs, cip) | LFS_CLEANER_MUST_CLEAN); 626 LFS_SYNC_CLEANERINFO(cip, fs, bp, 1); 627 628 /* Save segment flags for later */ 629 seg_flags = fs->lfs_sp->seg_flags; 630 631 fs->lfs_sp->seg_flags |= SEGM_PROT; /* Don't unmark dirop nodes */ 632 while(fs->lfs_seglock) 633 lfs_segunlock(fs); 634 635 /* Wait for the cleaner */ 636 lfs_wakeup_cleaner(fs); 637 mutex_enter(&lfs_lock); 638 while (LFS_STARVED_FOR_SEGS(fs)) 639 mtsleep(&fs->lfs_availsleep, PRIBIO, "relock", 0, 640 &lfs_lock); 641 mutex_exit(&lfs_lock); 642 643 /* Put the segment lock back the way it was. */ 644 while(n--) 645 lfs_seglock(fs, seg_flags); 646 647 /* Cleaner can relax now */ 648 LFS_CLEANERINFO(cip, fs, bp); 649 lfs_ci_setflags(fs, cip, 650 lfs_ci_getflags(fs, cip) & ~LFS_CLEANER_MUST_CLEAN); 651 LFS_SYNC_CLEANERINFO(cip, fs, bp, 1); 652 653 return; 654} 655 656/* 657 * Wake up the cleaner, provided that nowrap is not set. 658 */ 659void 660lfs_wakeup_cleaner(struct lfs *fs) 661{ 662 if (fs->lfs_nowrap > 0) 663 return; 664 665 wakeup(&fs->lfs_nextsegsleep); 666 wakeup(&lfs_allclean_wakeup); 667} 668