Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 1 | /* |
| 2 | * Copyright (C) International Business Machines Corp., 2000-2004 |
| 3 | * |
| 4 | * This program is free software; you can redistribute it and/or modify |
| 5 | * it under the terms of the GNU General Public License as published by |
| 6 | * the Free Software Foundation; either version 2 of the License, or |
| 7 | * (at your option) any later version. |
| 8 | * |
| 9 | * This program is distributed in the hope that it will be useful, |
| 10 | * but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 11 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See |
| 12 | * the GNU General Public License for more details. |
| 13 | * |
| 14 | * You should have received a copy of the GNU General Public License |
| 15 | * along with this program; if not, write to the Free Software |
| 16 | * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA |
| 17 | */ |
| 18 | /* |
| 19 | * jfs_xtree.c: extent allocation descriptor B+-tree manager |
| 20 | */ |
| 21 | |
| 22 | #include <linux/fs.h> |
| 23 | #include <linux/quotaops.h> |
| 24 | #include "jfs_incore.h" |
| 25 | #include "jfs_filsys.h" |
| 26 | #include "jfs_metapage.h" |
| 27 | #include "jfs_dmap.h" |
| 28 | #include "jfs_dinode.h" |
| 29 | #include "jfs_superblock.h" |
| 30 | #include "jfs_debug.h" |
| 31 | |
| 32 | /* |
| 33 | * xtree local flag |
| 34 | */ |
| 35 | #define XT_INSERT 0x00000001 |
| 36 | |
| 37 | /* |
| 38 | * xtree key/entry comparison: extent offset |
| 39 | * |
| 40 | * return: |
| 41 | * -1: k < start of extent |
| 42 | * 0: start_of_extent <= k <= end_of_extent |
| 43 | * 1: k > end_of_extent |
| 44 | */ |
| 45 | #define XT_CMP(CMP, K, X, OFFSET64)\ |
| 46 | {\ |
| 47 | OFFSET64 = offsetXAD(X);\ |
| 48 | (CMP) = ((K) >= OFFSET64 + lengthXAD(X)) ? 1 :\ |
| 49 | ((K) < OFFSET64) ? -1 : 0;\ |
| 50 | } |
| 51 | |
| 52 | /* write a xad entry */ |
| 53 | #define XT_PUTENTRY(XAD, FLAG, OFF, LEN, ADDR)\ |
| 54 | {\ |
| 55 | (XAD)->flag = (FLAG);\ |
| 56 | XADoffset((XAD), (OFF));\ |
| 57 | XADlength((XAD), (LEN));\ |
| 58 | XADaddress((XAD), (ADDR));\ |
| 59 | } |
| 60 | |
| 61 | #define XT_PAGE(IP, MP) BT_PAGE(IP, MP, xtpage_t, i_xtroot) |
| 62 | |
| 63 | /* get page buffer for specified block address */ |
| 64 | /* ToDo: Replace this ugly macro with a function */ |
| 65 | #define XT_GETPAGE(IP, BN, MP, SIZE, P, RC)\ |
| 66 | {\ |
| 67 | BT_GETPAGE(IP, BN, MP, xtpage_t, SIZE, P, RC, i_xtroot)\ |
| 68 | if (!(RC))\ |
| 69 | {\ |
| 70 | if ((le16_to_cpu((P)->header.nextindex) < XTENTRYSTART) ||\ |
| 71 | (le16_to_cpu((P)->header.nextindex) > le16_to_cpu((P)->header.maxentry)) ||\ |
| 72 | (le16_to_cpu((P)->header.maxentry) > (((BN)==0)?XTROOTMAXSLOT:PSIZE>>L2XTSLOTSIZE)))\ |
| 73 | {\ |
| 74 | jfs_error((IP)->i_sb, "XT_GETPAGE: xtree page corrupt");\ |
| 75 | BT_PUTPAGE(MP);\ |
| 76 | MP = NULL;\ |
| 77 | RC = -EIO;\ |
| 78 | }\ |
| 79 | }\ |
| 80 | } |
| 81 | |
| 82 | /* for consistency */ |
| 83 | #define XT_PUTPAGE(MP) BT_PUTPAGE(MP) |
| 84 | |
| 85 | #define XT_GETSEARCH(IP, LEAF, BN, MP, P, INDEX) \ |
| 86 | BT_GETSEARCH(IP, LEAF, BN, MP, xtpage_t, P, INDEX, i_xtroot) |
| 87 | /* xtree entry parameter descriptor */ |
| 88 | struct xtsplit { |
| 89 | struct metapage *mp; |
| 90 | s16 index; |
| 91 | u8 flag; |
| 92 | s64 off; |
| 93 | s64 addr; |
| 94 | int len; |
| 95 | struct pxdlist *pxdlist; |
| 96 | }; |
| 97 | |
| 98 | |
| 99 | /* |
| 100 | * statistics |
| 101 | */ |
| 102 | #ifdef CONFIG_JFS_STATISTICS |
| 103 | static struct { |
| 104 | uint search; |
| 105 | uint fastSearch; |
| 106 | uint split; |
| 107 | } xtStat; |
| 108 | #endif |
| 109 | |
| 110 | |
| 111 | /* |
| 112 | * forward references |
| 113 | */ |
| 114 | static int xtSearch(struct inode *ip, |
| 115 | s64 xoff, int *cmpp, struct btstack * btstack, int flag); |
| 116 | |
| 117 | static int xtSplitUp(tid_t tid, |
| 118 | struct inode *ip, |
| 119 | struct xtsplit * split, struct btstack * btstack); |
| 120 | |
| 121 | static int xtSplitPage(tid_t tid, struct inode *ip, struct xtsplit * split, |
| 122 | struct metapage ** rmpp, s64 * rbnp); |
| 123 | |
| 124 | static int xtSplitRoot(tid_t tid, struct inode *ip, |
| 125 | struct xtsplit * split, struct metapage ** rmpp); |
| 126 | |
| 127 | #ifdef _STILL_TO_PORT |
| 128 | static int xtDeleteUp(tid_t tid, struct inode *ip, struct metapage * fmp, |
| 129 | xtpage_t * fp, struct btstack * btstack); |
| 130 | |
| 131 | static int xtSearchNode(struct inode *ip, |
| 132 | xad_t * xad, |
| 133 | int *cmpp, struct btstack * btstack, int flag); |
| 134 | |
| 135 | static int xtRelink(tid_t tid, struct inode *ip, xtpage_t * fp); |
| 136 | #endif /* _STILL_TO_PORT */ |
| 137 | |
| 138 | /* External references */ |
| 139 | |
| 140 | /* |
| 141 | * debug control |
| 142 | */ |
| 143 | /* #define _JFS_DEBUG_XTREE 1 */ |
| 144 | |
| 145 | |
| 146 | /* |
| 147 | * xtLookup() |
| 148 | * |
| 149 | * function: map a single page into a physical extent; |
| 150 | */ |
| 151 | int xtLookup(struct inode *ip, s64 lstart, |
| 152 | s64 llen, int *pflag, s64 * paddr, s32 * plen, int no_check) |
| 153 | { |
| 154 | int rc = 0; |
| 155 | struct btstack btstack; |
| 156 | int cmp; |
| 157 | s64 bn; |
| 158 | struct metapage *mp; |
| 159 | xtpage_t *p; |
| 160 | int index; |
| 161 | xad_t *xad; |
| 162 | s64 size, xoff, xend; |
| 163 | int xlen; |
| 164 | s64 xaddr; |
| 165 | |
| 166 | *plen = 0; |
| 167 | |
| 168 | if (!no_check) { |
| 169 | /* is lookup offset beyond eof ? */ |
| 170 | size = ((u64) ip->i_size + (JFS_SBI(ip->i_sb)->bsize - 1)) >> |
| 171 | JFS_SBI(ip->i_sb)->l2bsize; |
| 172 | if (lstart >= size) { |
| 173 | jfs_err("xtLookup: lstart (0x%lx) >= size (0x%lx)", |
| 174 | (ulong) lstart, (ulong) size); |
| 175 | return 0; |
| 176 | } |
| 177 | } |
| 178 | |
| 179 | /* |
| 180 | * search for the xad entry covering the logical extent |
| 181 | */ |
| 182 | //search: |
| 183 | if ((rc = xtSearch(ip, lstart, &cmp, &btstack, 0))) { |
| 184 | jfs_err("xtLookup: xtSearch returned %d", rc); |
| 185 | return rc; |
| 186 | } |
| 187 | |
| 188 | /* |
| 189 | * compute the physical extent covering logical extent |
| 190 | * |
| 191 | * N.B. search may have failed (e.g., hole in sparse file), |
| 192 | * and returned the index of the next entry. |
| 193 | */ |
| 194 | /* retrieve search result */ |
| 195 | XT_GETSEARCH(ip, btstack.top, bn, mp, p, index); |
| 196 | |
| 197 | /* is xad found covering start of logical extent ? |
| 198 | * lstart is a page start address, |
| 199 | * i.e., lstart cannot start in a hole; |
| 200 | */ |
| 201 | if (cmp) |
| 202 | goto out; |
| 203 | |
| 204 | /* |
| 205 | * lxd covered by xad |
| 206 | */ |
| 207 | xad = &p->xad[index]; |
| 208 | xoff = offsetXAD(xad); |
| 209 | xlen = lengthXAD(xad); |
| 210 | xend = xoff + xlen; |
| 211 | xaddr = addressXAD(xad); |
| 212 | |
| 213 | /* initialize new pxd */ |
| 214 | *pflag = xad->flag; |
| 215 | *paddr = xaddr + (lstart - xoff); |
| 216 | /* a page must be fully covered by an xad */ |
| 217 | *plen = min(xend - lstart, llen); |
| 218 | |
| 219 | out: |
| 220 | XT_PUTPAGE(mp); |
| 221 | |
| 222 | return rc; |
| 223 | } |
| 224 | |
| 225 | |
| 226 | /* |
| 227 | * xtLookupList() |
| 228 | * |
| 229 | * function: map a single logical extent into a list of physical extent; |
| 230 | * |
| 231 | * parameter: |
| 232 | * struct inode *ip, |
| 233 | * struct lxdlist *lxdlist, lxd list (in) |
| 234 | * struct xadlist *xadlist, xad list (in/out) |
| 235 | * int flag) |
| 236 | * |
| 237 | * coverage of lxd by xad under assumption of |
| 238 | * . lxd's are ordered and disjoint. |
| 239 | * . xad's are ordered and disjoint. |
| 240 | * |
| 241 | * return: |
| 242 | * 0: success |
| 243 | * |
| 244 | * note: a page being written (even a single byte) is backed fully, |
| 245 | * except the last page which is only backed with blocks |
| 246 | * required to cover the last byte; |
| 247 | * the extent backing a page is fully contained within an xad; |
| 248 | */ |
| 249 | int xtLookupList(struct inode *ip, struct lxdlist * lxdlist, |
| 250 | struct xadlist * xadlist, int flag) |
| 251 | { |
| 252 | int rc = 0; |
| 253 | struct btstack btstack; |
| 254 | int cmp; |
| 255 | s64 bn; |
| 256 | struct metapage *mp; |
| 257 | xtpage_t *p; |
| 258 | int index; |
| 259 | lxd_t *lxd; |
| 260 | xad_t *xad, *pxd; |
| 261 | s64 size, lstart, lend, xstart, xend, pstart; |
| 262 | s64 llen, xlen, plen; |
| 263 | s64 xaddr, paddr; |
| 264 | int nlxd, npxd, maxnpxd; |
| 265 | |
| 266 | npxd = xadlist->nxad = 0; |
| 267 | maxnpxd = xadlist->maxnxad; |
| 268 | pxd = xadlist->xad; |
| 269 | |
| 270 | nlxd = lxdlist->nlxd; |
| 271 | lxd = lxdlist->lxd; |
| 272 | |
| 273 | lstart = offsetLXD(lxd); |
| 274 | llen = lengthLXD(lxd); |
| 275 | lend = lstart + llen; |
| 276 | |
| 277 | size = (ip->i_size + (JFS_SBI(ip->i_sb)->bsize - 1)) >> |
| 278 | JFS_SBI(ip->i_sb)->l2bsize; |
| 279 | |
| 280 | /* |
| 281 | * search for the xad entry covering the logical extent |
| 282 | */ |
| 283 | search: |
| 284 | if (lstart >= size) |
| 285 | return 0; |
| 286 | |
| 287 | if ((rc = xtSearch(ip, lstart, &cmp, &btstack, 0))) |
| 288 | return rc; |
| 289 | |
| 290 | /* |
| 291 | * compute the physical extent covering logical extent |
| 292 | * |
| 293 | * N.B. search may have failed (e.g., hole in sparse file), |
| 294 | * and returned the index of the next entry. |
| 295 | */ |
| 296 | //map: |
| 297 | /* retrieve search result */ |
| 298 | XT_GETSEARCH(ip, btstack.top, bn, mp, p, index); |
| 299 | |
| 300 | /* is xad on the next sibling page ? */ |
| 301 | if (index == le16_to_cpu(p->header.nextindex)) { |
| 302 | if (p->header.flag & BT_ROOT) |
| 303 | goto mapend; |
| 304 | |
| 305 | if ((bn = le64_to_cpu(p->header.next)) == 0) |
| 306 | goto mapend; |
| 307 | |
| 308 | XT_PUTPAGE(mp); |
| 309 | |
| 310 | /* get next sibling page */ |
| 311 | XT_GETPAGE(ip, bn, mp, PSIZE, p, rc); |
| 312 | if (rc) |
| 313 | return rc; |
| 314 | |
| 315 | index = XTENTRYSTART; |
| 316 | } |
| 317 | |
| 318 | xad = &p->xad[index]; |
| 319 | |
| 320 | /* |
| 321 | * is lxd covered by xad ? |
| 322 | */ |
| 323 | compare: |
| 324 | xstart = offsetXAD(xad); |
| 325 | xlen = lengthXAD(xad); |
| 326 | xend = xstart + xlen; |
| 327 | xaddr = addressXAD(xad); |
| 328 | |
| 329 | compare1: |
| 330 | if (xstart < lstart) |
| 331 | goto compare2; |
| 332 | |
| 333 | /* (lstart <= xstart) */ |
| 334 | |
| 335 | /* lxd is NOT covered by xad */ |
| 336 | if (lend <= xstart) { |
| 337 | /* |
| 338 | * get next lxd |
| 339 | */ |
| 340 | if (--nlxd == 0) |
| 341 | goto mapend; |
| 342 | lxd++; |
| 343 | |
| 344 | lstart = offsetLXD(lxd); |
| 345 | llen = lengthLXD(lxd); |
| 346 | lend = lstart + llen; |
| 347 | if (lstart >= size) |
| 348 | goto mapend; |
| 349 | |
| 350 | /* compare with the current xad */ |
| 351 | goto compare1; |
| 352 | } |
| 353 | /* lxd is covered by xad */ |
| 354 | else { /* (xstart < lend) */ |
| 355 | |
| 356 | /* initialize new pxd */ |
| 357 | pstart = xstart; |
| 358 | plen = min(lend - xstart, xlen); |
| 359 | paddr = xaddr; |
| 360 | |
| 361 | goto cover; |
| 362 | } |
| 363 | |
| 364 | /* (xstart < lstart) */ |
| 365 | compare2: |
| 366 | /* lxd is covered by xad */ |
| 367 | if (lstart < xend) { |
| 368 | /* initialize new pxd */ |
| 369 | pstart = lstart; |
| 370 | plen = min(xend - lstart, llen); |
| 371 | paddr = xaddr + (lstart - xstart); |
| 372 | |
| 373 | goto cover; |
| 374 | } |
| 375 | /* lxd is NOT covered by xad */ |
| 376 | else { /* (xend <= lstart) */ |
| 377 | |
| 378 | /* |
| 379 | * get next xad |
| 380 | * |
| 381 | * linear search next xad covering lxd on |
| 382 | * the current xad page, and then tree search |
| 383 | */ |
| 384 | if (index == le16_to_cpu(p->header.nextindex) - 1) { |
| 385 | if (p->header.flag & BT_ROOT) |
| 386 | goto mapend; |
| 387 | |
| 388 | XT_PUTPAGE(mp); |
| 389 | goto search; |
| 390 | } else { |
| 391 | index++; |
| 392 | xad++; |
| 393 | |
| 394 | /* compare with new xad */ |
| 395 | goto compare; |
| 396 | } |
| 397 | } |
| 398 | |
| 399 | /* |
| 400 | * lxd is covered by xad and a new pxd has been initialized |
| 401 | * (lstart <= xstart < lend) or (xstart < lstart < xend) |
| 402 | */ |
| 403 | cover: |
| 404 | /* finalize pxd corresponding to current xad */ |
| 405 | XT_PUTENTRY(pxd, xad->flag, pstart, plen, paddr); |
| 406 | |
| 407 | if (++npxd >= maxnpxd) |
| 408 | goto mapend; |
| 409 | pxd++; |
| 410 | |
| 411 | /* |
| 412 | * lxd is fully covered by xad |
| 413 | */ |
| 414 | if (lend <= xend) { |
| 415 | /* |
| 416 | * get next lxd |
| 417 | */ |
| 418 | if (--nlxd == 0) |
| 419 | goto mapend; |
| 420 | lxd++; |
| 421 | |
| 422 | lstart = offsetLXD(lxd); |
| 423 | llen = lengthLXD(lxd); |
| 424 | lend = lstart + llen; |
| 425 | if (lstart >= size) |
| 426 | goto mapend; |
| 427 | |
| 428 | /* |
| 429 | * test for old xad covering new lxd |
| 430 | * (old xstart < new lstart) |
| 431 | */ |
| 432 | goto compare2; |
| 433 | } |
| 434 | /* |
| 435 | * lxd is partially covered by xad |
| 436 | */ |
| 437 | else { /* (xend < lend) */ |
| 438 | |
| 439 | /* |
| 440 | * get next xad |
| 441 | * |
| 442 | * linear search next xad covering lxd on |
| 443 | * the current xad page, and then next xad page search |
| 444 | */ |
| 445 | if (index == le16_to_cpu(p->header.nextindex) - 1) { |
| 446 | if (p->header.flag & BT_ROOT) |
| 447 | goto mapend; |
| 448 | |
| 449 | if ((bn = le64_to_cpu(p->header.next)) == 0) |
| 450 | goto mapend; |
| 451 | |
| 452 | XT_PUTPAGE(mp); |
| 453 | |
| 454 | /* get next sibling page */ |
| 455 | XT_GETPAGE(ip, bn, mp, PSIZE, p, rc); |
| 456 | if (rc) |
| 457 | return rc; |
| 458 | |
| 459 | index = XTENTRYSTART; |
| 460 | xad = &p->xad[index]; |
| 461 | } else { |
| 462 | index++; |
| 463 | xad++; |
| 464 | } |
| 465 | |
| 466 | /* |
| 467 | * test for new xad covering old lxd |
| 468 | * (old lstart < new xstart) |
| 469 | */ |
| 470 | goto compare; |
| 471 | } |
| 472 | |
| 473 | mapend: |
| 474 | xadlist->nxad = npxd; |
| 475 | |
| 476 | //out: |
| 477 | XT_PUTPAGE(mp); |
| 478 | |
| 479 | return rc; |
| 480 | } |
| 481 | |
| 482 | |
| 483 | /* |
| 484 | * xtSearch() |
| 485 | * |
| 486 | * function: search for the xad entry covering specified offset. |
| 487 | * |
| 488 | * parameters: |
| 489 | * ip - file object; |
| 490 | * xoff - extent offset; |
| 491 | * cmpp - comparison result: |
| 492 | * btstack - traverse stack; |
| 493 | * flag - search process flag (XT_INSERT); |
| 494 | * |
| 495 | * returns: |
| 496 | * btstack contains (bn, index) of search path traversed to the entry. |
| 497 | * *cmpp is set to result of comparison with the entry returned. |
| 498 | * the page containing the entry is pinned at exit. |
| 499 | */ |
| 500 | static int xtSearch(struct inode *ip, s64 xoff, /* offset of extent */ |
| 501 | int *cmpp, struct btstack * btstack, int flag) |
| 502 | { |
| 503 | struct jfs_inode_info *jfs_ip = JFS_IP(ip); |
| 504 | int rc = 0; |
| 505 | int cmp = 1; /* init for empty page */ |
| 506 | s64 bn; /* block number */ |
| 507 | struct metapage *mp; /* page buffer */ |
| 508 | xtpage_t *p; /* page */ |
| 509 | xad_t *xad; |
| 510 | int base, index, lim, btindex; |
| 511 | struct btframe *btsp; |
| 512 | int nsplit = 0; /* number of pages to split */ |
| 513 | s64 t64; |
| 514 | |
| 515 | INCREMENT(xtStat.search); |
| 516 | |
| 517 | BT_CLR(btstack); |
| 518 | |
| 519 | btstack->nsplit = 0; |
| 520 | |
| 521 | /* |
| 522 | * search down tree from root: |
| 523 | * |
| 524 | * between two consecutive entries of <Ki, Pi> and <Kj, Pj> of |
| 525 | * internal page, child page Pi contains entry with k, Ki <= K < Kj. |
| 526 | * |
| 527 | * if entry with search key K is not found |
| 528 | * internal page search find the entry with largest key Ki |
| 529 | * less than K which point to the child page to search; |
| 530 | * leaf page search find the entry with smallest key Kj |
| 531 | * greater than K so that the returned index is the position of |
| 532 | * the entry to be shifted right for insertion of new entry. |
| 533 | * for empty tree, search key is greater than any key of the tree. |
| 534 | * |
| 535 | * by convention, root bn = 0. |
| 536 | */ |
| 537 | for (bn = 0;;) { |
| 538 | /* get/pin the page to search */ |
| 539 | XT_GETPAGE(ip, bn, mp, PSIZE, p, rc); |
| 540 | if (rc) |
| 541 | return rc; |
| 542 | |
| 543 | /* try sequential access heuristics with the previous |
| 544 | * access entry in target leaf page: |
| 545 | * once search narrowed down into the target leaf, |
| 546 | * key must either match an entry in the leaf or |
| 547 | * key entry does not exist in the tree; |
| 548 | */ |
| 549 | //fastSearch: |
| 550 | if ((jfs_ip->btorder & BT_SEQUENTIAL) && |
| 551 | (p->header.flag & BT_LEAF) && |
| 552 | (index = jfs_ip->btindex) < |
| 553 | le16_to_cpu(p->header.nextindex)) { |
| 554 | xad = &p->xad[index]; |
| 555 | t64 = offsetXAD(xad); |
| 556 | if (xoff < t64 + lengthXAD(xad)) { |
| 557 | if (xoff >= t64) { |
| 558 | *cmpp = 0; |
| 559 | goto out; |
| 560 | } |
| 561 | |
| 562 | /* stop sequential access heuristics */ |
| 563 | goto binarySearch; |
| 564 | } else { /* (t64 + lengthXAD(xad)) <= xoff */ |
| 565 | |
| 566 | /* try next sequential entry */ |
| 567 | index++; |
| 568 | if (index < |
| 569 | le16_to_cpu(p->header.nextindex)) { |
| 570 | xad++; |
| 571 | t64 = offsetXAD(xad); |
| 572 | if (xoff < t64 + lengthXAD(xad)) { |
| 573 | if (xoff >= t64) { |
| 574 | *cmpp = 0; |
| 575 | goto out; |
| 576 | } |
| 577 | |
| 578 | /* miss: key falls between |
| 579 | * previous and this entry |
| 580 | */ |
| 581 | *cmpp = 1; |
| 582 | goto out; |
| 583 | } |
| 584 | |
| 585 | /* (xoff >= t64 + lengthXAD(xad)); |
| 586 | * matching entry may be further out: |
| 587 | * stop heuristic search |
| 588 | */ |
| 589 | /* stop sequential access heuristics */ |
| 590 | goto binarySearch; |
| 591 | } |
| 592 | |
| 593 | /* (index == p->header.nextindex); |
| 594 | * miss: key entry does not exist in |
| 595 | * the target leaf/tree |
| 596 | */ |
| 597 | *cmpp = 1; |
| 598 | goto out; |
| 599 | } |
| 600 | |
| 601 | /* |
| 602 | * if hit, return index of the entry found, and |
| 603 | * if miss, where new entry with search key is |
| 604 | * to be inserted; |
| 605 | */ |
| 606 | out: |
| 607 | /* compute number of pages to split */ |
| 608 | if (flag & XT_INSERT) { |
| 609 | if (p->header.nextindex == /* little-endian */ |
| 610 | p->header.maxentry) |
| 611 | nsplit++; |
| 612 | else |
| 613 | nsplit = 0; |
| 614 | btstack->nsplit = nsplit; |
| 615 | } |
| 616 | |
| 617 | /* save search result */ |
| 618 | btsp = btstack->top; |
| 619 | btsp->bn = bn; |
| 620 | btsp->index = index; |
| 621 | btsp->mp = mp; |
| 622 | |
| 623 | /* update sequential access heuristics */ |
| 624 | jfs_ip->btindex = index; |
| 625 | |
| 626 | INCREMENT(xtStat.fastSearch); |
| 627 | return 0; |
| 628 | } |
| 629 | |
| 630 | /* well, ... full search now */ |
| 631 | binarySearch: |
| 632 | lim = le16_to_cpu(p->header.nextindex) - XTENTRYSTART; |
| 633 | |
| 634 | /* |
| 635 | * binary search with search key K on the current page |
| 636 | */ |
| 637 | for (base = XTENTRYSTART; lim; lim >>= 1) { |
| 638 | index = base + (lim >> 1); |
| 639 | |
| 640 | XT_CMP(cmp, xoff, &p->xad[index], t64); |
| 641 | if (cmp == 0) { |
| 642 | /* |
| 643 | * search hit |
| 644 | */ |
| 645 | /* search hit - leaf page: |
| 646 | * return the entry found |
| 647 | */ |
| 648 | if (p->header.flag & BT_LEAF) { |
| 649 | *cmpp = cmp; |
| 650 | |
| 651 | /* compute number of pages to split */ |
| 652 | if (flag & XT_INSERT) { |
| 653 | if (p->header.nextindex == |
| 654 | p->header.maxentry) |
| 655 | nsplit++; |
| 656 | else |
| 657 | nsplit = 0; |
| 658 | btstack->nsplit = nsplit; |
| 659 | } |
| 660 | |
| 661 | /* save search result */ |
| 662 | btsp = btstack->top; |
| 663 | btsp->bn = bn; |
| 664 | btsp->index = index; |
| 665 | btsp->mp = mp; |
| 666 | |
| 667 | /* init sequential access heuristics */ |
| 668 | btindex = jfs_ip->btindex; |
| 669 | if (index == btindex || |
| 670 | index == btindex + 1) |
| 671 | jfs_ip->btorder = BT_SEQUENTIAL; |
| 672 | else |
| 673 | jfs_ip->btorder = BT_RANDOM; |
| 674 | jfs_ip->btindex = index; |
| 675 | |
| 676 | return 0; |
| 677 | } |
| 678 | |
| 679 | /* search hit - internal page: |
| 680 | * descend/search its child page |
| 681 | */ |
| 682 | goto next; |
| 683 | } |
| 684 | |
| 685 | if (cmp > 0) { |
| 686 | base = index + 1; |
| 687 | --lim; |
| 688 | } |
| 689 | } |
| 690 | |
| 691 | /* |
| 692 | * search miss |
| 693 | * |
| 694 | * base is the smallest index with key (Kj) greater than |
| 695 | * search key (K) and may be zero or maxentry index. |
| 696 | */ |
| 697 | /* |
| 698 | * search miss - leaf page: |
| 699 | * |
| 700 | * return location of entry (base) where new entry with |
| 701 | * search key K is to be inserted. |
| 702 | */ |
| 703 | if (p->header.flag & BT_LEAF) { |
| 704 | *cmpp = cmp; |
| 705 | |
| 706 | /* compute number of pages to split */ |
| 707 | if (flag & XT_INSERT) { |
| 708 | if (p->header.nextindex == |
| 709 | p->header.maxentry) |
| 710 | nsplit++; |
| 711 | else |
| 712 | nsplit = 0; |
| 713 | btstack->nsplit = nsplit; |
| 714 | } |
| 715 | |
| 716 | /* save search result */ |
| 717 | btsp = btstack->top; |
| 718 | btsp->bn = bn; |
| 719 | btsp->index = base; |
| 720 | btsp->mp = mp; |
| 721 | |
| 722 | /* init sequential access heuristics */ |
| 723 | btindex = jfs_ip->btindex; |
| 724 | if (base == btindex || base == btindex + 1) |
| 725 | jfs_ip->btorder = BT_SEQUENTIAL; |
| 726 | else |
| 727 | jfs_ip->btorder = BT_RANDOM; |
| 728 | jfs_ip->btindex = base; |
| 729 | |
| 730 | return 0; |
| 731 | } |
| 732 | |
| 733 | /* |
| 734 | * search miss - non-leaf page: |
| 735 | * |
| 736 | * if base is non-zero, decrement base by one to get the parent |
| 737 | * entry of the child page to search. |
| 738 | */ |
| 739 | index = base ? base - 1 : base; |
| 740 | |
| 741 | /* |
| 742 | * go down to child page |
| 743 | */ |
| 744 | next: |
| 745 | /* update number of pages to split */ |
| 746 | if (p->header.nextindex == p->header.maxentry) |
| 747 | nsplit++; |
| 748 | else |
| 749 | nsplit = 0; |
| 750 | |
| 751 | /* push (bn, index) of the parent page/entry */ |
| 752 | BT_PUSH(btstack, bn, index); |
| 753 | |
| 754 | /* get the child page block number */ |
| 755 | bn = addressXAD(&p->xad[index]); |
| 756 | |
| 757 | /* unpin the parent page */ |
| 758 | XT_PUTPAGE(mp); |
| 759 | } |
| 760 | } |
| 761 | |
| 762 | /* |
| 763 | * xtInsert() |
| 764 | * |
| 765 | * function: |
| 766 | * |
| 767 | * parameter: |
| 768 | * tid - transaction id; |
| 769 | * ip - file object; |
| 770 | * xflag - extent flag (XAD_NOTRECORDED): |
| 771 | * xoff - extent offset; |
| 772 | * xlen - extent length; |
| 773 | * xaddrp - extent address pointer (in/out): |
| 774 | * if (*xaddrp) |
| 775 | * caller allocated data extent at *xaddrp; |
| 776 | * else |
| 777 | * allocate data extent and return its xaddr; |
| 778 | * flag - |
| 779 | * |
| 780 | * return: |
| 781 | */ |
| 782 | int xtInsert(tid_t tid, /* transaction id */ |
| 783 | struct inode *ip, int xflag, s64 xoff, s32 xlen, s64 * xaddrp, |
| 784 | int flag) |
| 785 | { |
| 786 | int rc = 0; |
| 787 | s64 xaddr, hint; |
| 788 | struct metapage *mp; /* meta-page buffer */ |
| 789 | xtpage_t *p; /* base B+-tree index page */ |
| 790 | s64 bn; |
| 791 | int index, nextindex; |
| 792 | struct btstack btstack; /* traverse stack */ |
| 793 | struct xtsplit split; /* split information */ |
| 794 | xad_t *xad; |
| 795 | int cmp; |
| 796 | struct tlock *tlck; |
| 797 | struct xtlock *xtlck; |
| 798 | |
| 799 | jfs_info("xtInsert: nxoff:0x%lx nxlen:0x%x", (ulong) xoff, xlen); |
| 800 | |
| 801 | /* |
| 802 | * search for the entry location at which to insert: |
| 803 | * |
| 804 | * xtFastSearch() and xtSearch() both returns (leaf page |
| 805 | * pinned, index at which to insert). |
| 806 | * n.b. xtSearch() may return index of maxentry of |
| 807 | * the full page. |
| 808 | */ |
| 809 | if ((rc = xtSearch(ip, xoff, &cmp, &btstack, XT_INSERT))) |
| 810 | return rc; |
| 811 | |
| 812 | /* retrieve search result */ |
| 813 | XT_GETSEARCH(ip, btstack.top, bn, mp, p, index); |
| 814 | |
| 815 | /* This test must follow XT_GETSEARCH since mp must be valid if |
| 816 | * we branch to out: */ |
| 817 | if (cmp == 0) { |
| 818 | rc = -EEXIST; |
| 819 | goto out; |
| 820 | } |
| 821 | |
| 822 | /* |
| 823 | * allocate data extent requested |
| 824 | * |
| 825 | * allocation hint: last xad |
| 826 | */ |
| 827 | if ((xaddr = *xaddrp) == 0) { |
| 828 | if (index > XTENTRYSTART) { |
| 829 | xad = &p->xad[index - 1]; |
| 830 | hint = addressXAD(xad) + lengthXAD(xad) - 1; |
| 831 | } else |
| 832 | hint = 0; |
| 833 | if ((rc = DQUOT_ALLOC_BLOCK(ip, xlen))) |
| 834 | goto out; |
| 835 | if ((rc = dbAlloc(ip, hint, (s64) xlen, &xaddr))) { |
| 836 | DQUOT_FREE_BLOCK(ip, xlen); |
| 837 | goto out; |
| 838 | } |
| 839 | } |
| 840 | |
| 841 | /* |
| 842 | * insert entry for new extent |
| 843 | */ |
| 844 | xflag |= XAD_NEW; |
| 845 | |
| 846 | /* |
| 847 | * if the leaf page is full, split the page and |
| 848 | * propagate up the router entry for the new page from split |
| 849 | * |
| 850 | * The xtSplitUp() will insert the entry and unpin the leaf page. |
| 851 | */ |
| 852 | nextindex = le16_to_cpu(p->header.nextindex); |
| 853 | if (nextindex == le16_to_cpu(p->header.maxentry)) { |
| 854 | split.mp = mp; |
| 855 | split.index = index; |
| 856 | split.flag = xflag; |
| 857 | split.off = xoff; |
| 858 | split.len = xlen; |
| 859 | split.addr = xaddr; |
| 860 | split.pxdlist = NULL; |
| 861 | if ((rc = xtSplitUp(tid, ip, &split, &btstack))) { |
| 862 | /* undo data extent allocation */ |
| 863 | if (*xaddrp == 0) { |
| 864 | dbFree(ip, xaddr, (s64) xlen); |
| 865 | DQUOT_FREE_BLOCK(ip, xlen); |
| 866 | } |
| 867 | return rc; |
| 868 | } |
| 869 | |
| 870 | *xaddrp = xaddr; |
| 871 | return 0; |
| 872 | } |
| 873 | |
| 874 | /* |
| 875 | * insert the new entry into the leaf page |
| 876 | */ |
| 877 | /* |
| 878 | * acquire a transaction lock on the leaf page; |
| 879 | * |
| 880 | * action: xad insertion/extension; |
| 881 | */ |
| 882 | BT_MARK_DIRTY(mp, ip); |
| 883 | |
| 884 | /* if insert into middle, shift right remaining entries. */ |
| 885 | if (index < nextindex) |
| 886 | memmove(&p->xad[index + 1], &p->xad[index], |
| 887 | (nextindex - index) * sizeof(xad_t)); |
| 888 | |
| 889 | /* insert the new entry: mark the entry NEW */ |
| 890 | xad = &p->xad[index]; |
| 891 | XT_PUTENTRY(xad, xflag, xoff, xlen, xaddr); |
| 892 | |
| 893 | /* advance next available entry index */ |
| 894 | p->header.nextindex = |
| 895 | cpu_to_le16(le16_to_cpu(p->header.nextindex) + 1); |
| 896 | |
| 897 | /* Don't log it if there are no links to the file */ |
| 898 | if (!test_cflag(COMMIT_Nolink, ip)) { |
| 899 | tlck = txLock(tid, ip, mp, tlckXTREE | tlckGROW); |
| 900 | xtlck = (struct xtlock *) & tlck->lock; |
| 901 | xtlck->lwm.offset = |
| 902 | (xtlck->lwm.offset) ? min(index, |
| 903 | (int)xtlck->lwm.offset) : index; |
| 904 | xtlck->lwm.length = |
| 905 | le16_to_cpu(p->header.nextindex) - xtlck->lwm.offset; |
| 906 | } |
| 907 | |
| 908 | *xaddrp = xaddr; |
| 909 | |
| 910 | out: |
| 911 | /* unpin the leaf page */ |
| 912 | XT_PUTPAGE(mp); |
| 913 | |
| 914 | return rc; |
| 915 | } |
| 916 | |
| 917 | |
| 918 | /* |
| 919 | * xtSplitUp() |
| 920 | * |
| 921 | * function: |
| 922 | * split full pages as propagating insertion up the tree |
| 923 | * |
| 924 | * parameter: |
| 925 | * tid - transaction id; |
| 926 | * ip - file object; |
| 927 | * split - entry parameter descriptor; |
| 928 | * btstack - traverse stack from xtSearch() |
| 929 | * |
| 930 | * return: |
| 931 | */ |
| 932 | static int |
| 933 | xtSplitUp(tid_t tid, |
| 934 | struct inode *ip, struct xtsplit * split, struct btstack * btstack) |
| 935 | { |
| 936 | int rc = 0; |
| 937 | struct metapage *smp; |
| 938 | xtpage_t *sp; /* split page */ |
| 939 | struct metapage *rmp; |
| 940 | s64 rbn; /* new right page block number */ |
| 941 | struct metapage *rcmp; |
| 942 | xtpage_t *rcp; /* right child page */ |
| 943 | s64 rcbn; /* right child page block number */ |
| 944 | int skip; /* index of entry of insertion */ |
| 945 | int nextindex; /* next available entry index of p */ |
| 946 | struct btframe *parent; /* parent page entry on traverse stack */ |
| 947 | xad_t *xad; |
| 948 | s64 xaddr; |
| 949 | int xlen; |
| 950 | int nsplit; /* number of pages split */ |
| 951 | struct pxdlist pxdlist; |
| 952 | pxd_t *pxd; |
| 953 | struct tlock *tlck; |
| 954 | struct xtlock *xtlck; |
| 955 | |
| 956 | smp = split->mp; |
| 957 | sp = XT_PAGE(ip, smp); |
| 958 | |
| 959 | /* is inode xtree root extension/inline EA area free ? */ |
| 960 | if ((sp->header.flag & BT_ROOT) && (!S_ISDIR(ip->i_mode)) && |
| 961 | (le16_to_cpu(sp->header.maxentry) < XTROOTMAXSLOT) && |
| 962 | (JFS_IP(ip)->mode2 & INLINEEA)) { |
| 963 | sp->header.maxentry = cpu_to_le16(XTROOTMAXSLOT); |
| 964 | JFS_IP(ip)->mode2 &= ~INLINEEA; |
| 965 | |
| 966 | BT_MARK_DIRTY(smp, ip); |
| 967 | /* |
| 968 | * acquire a transaction lock on the leaf page; |
| 969 | * |
| 970 | * action: xad insertion/extension; |
| 971 | */ |
| 972 | |
| 973 | /* if insert into middle, shift right remaining entries. */ |
| 974 | skip = split->index; |
| 975 | nextindex = le16_to_cpu(sp->header.nextindex); |
| 976 | if (skip < nextindex) |
| 977 | memmove(&sp->xad[skip + 1], &sp->xad[skip], |
| 978 | (nextindex - skip) * sizeof(xad_t)); |
| 979 | |
| 980 | /* insert the new entry: mark the entry NEW */ |
| 981 | xad = &sp->xad[skip]; |
| 982 | XT_PUTENTRY(xad, split->flag, split->off, split->len, |
| 983 | split->addr); |
| 984 | |
| 985 | /* advance next available entry index */ |
| 986 | sp->header.nextindex = |
| 987 | cpu_to_le16(le16_to_cpu(sp->header.nextindex) + 1); |
| 988 | |
| 989 | /* Don't log it if there are no links to the file */ |
| 990 | if (!test_cflag(COMMIT_Nolink, ip)) { |
| 991 | tlck = txLock(tid, ip, smp, tlckXTREE | tlckGROW); |
| 992 | xtlck = (struct xtlock *) & tlck->lock; |
| 993 | xtlck->lwm.offset = (xtlck->lwm.offset) ? |
| 994 | min(skip, (int)xtlck->lwm.offset) : skip; |
| 995 | xtlck->lwm.length = |
| 996 | le16_to_cpu(sp->header.nextindex) - |
| 997 | xtlck->lwm.offset; |
| 998 | } |
| 999 | |
| 1000 | return 0; |
| 1001 | } |
| 1002 | |
| 1003 | /* |
| 1004 | * allocate new index blocks to cover index page split(s) |
| 1005 | * |
| 1006 | * allocation hint: ? |
| 1007 | */ |
| 1008 | if (split->pxdlist == NULL) { |
| 1009 | nsplit = btstack->nsplit; |
| 1010 | split->pxdlist = &pxdlist; |
| 1011 | pxdlist.maxnpxd = pxdlist.npxd = 0; |
| 1012 | pxd = &pxdlist.pxd[0]; |
| 1013 | xlen = JFS_SBI(ip->i_sb)->nbperpage; |
| 1014 | for (; nsplit > 0; nsplit--, pxd++) { |
| 1015 | if ((rc = dbAlloc(ip, (s64) 0, (s64) xlen, &xaddr)) |
| 1016 | == 0) { |
| 1017 | PXDaddress(pxd, xaddr); |
| 1018 | PXDlength(pxd, xlen); |
| 1019 | |
| 1020 | pxdlist.maxnpxd++; |
| 1021 | |
| 1022 | continue; |
| 1023 | } |
| 1024 | |
| 1025 | /* undo allocation */ |
| 1026 | |
| 1027 | XT_PUTPAGE(smp); |
| 1028 | return rc; |
| 1029 | } |
| 1030 | } |
| 1031 | |
| 1032 | /* |
| 1033 | * Split leaf page <sp> into <sp> and a new right page <rp>. |
| 1034 | * |
| 1035 | * The split routines insert the new entry into the leaf page, |
| 1036 | * and acquire txLock as appropriate. |
| 1037 | * return <rp> pinned and its block number <rpbn>. |
| 1038 | */ |
| 1039 | rc = (sp->header.flag & BT_ROOT) ? |
| 1040 | xtSplitRoot(tid, ip, split, &rmp) : |
| 1041 | xtSplitPage(tid, ip, split, &rmp, &rbn); |
| 1042 | |
| 1043 | XT_PUTPAGE(smp); |
| 1044 | |
| 1045 | if (rc) |
| 1046 | return -EIO; |
| 1047 | /* |
| 1048 | * propagate up the router entry for the leaf page just split |
| 1049 | * |
| 1050 | * insert a router entry for the new page into the parent page, |
| 1051 | * propagate the insert/split up the tree by walking back the stack |
| 1052 | * of (bn of parent page, index of child page entry in parent page) |
| 1053 | * that were traversed during the search for the page that split. |
| 1054 | * |
| 1055 | * the propagation of insert/split up the tree stops if the root |
| 1056 | * splits or the page inserted into doesn't have to split to hold |
| 1057 | * the new entry. |
| 1058 | * |
| 1059 | * the parent entry for the split page remains the same, and |
| 1060 | * a new entry is inserted at its right with the first key and |
| 1061 | * block number of the new right page. |
| 1062 | * |
| 1063 | * There are a maximum of 3 pages pinned at any time: |
| 1064 | * right child, left parent and right parent (when the parent splits) |
| 1065 | * to keep the child page pinned while working on the parent. |
| 1066 | * make sure that all pins are released at exit. |
| 1067 | */ |
| 1068 | while ((parent = BT_POP(btstack)) != NULL) { |
| 1069 | /* parent page specified by stack frame <parent> */ |
| 1070 | |
| 1071 | /* keep current child pages <rcp> pinned */ |
| 1072 | rcmp = rmp; |
| 1073 | rcbn = rbn; |
| 1074 | rcp = XT_PAGE(ip, rcmp); |
| 1075 | |
| 1076 | /* |
| 1077 | * insert router entry in parent for new right child page <rp> |
| 1078 | */ |
| 1079 | /* get/pin the parent page <sp> */ |
| 1080 | XT_GETPAGE(ip, parent->bn, smp, PSIZE, sp, rc); |
| 1081 | if (rc) { |
| 1082 | XT_PUTPAGE(rcmp); |
| 1083 | return rc; |
| 1084 | } |
| 1085 | |
| 1086 | /* |
| 1087 | * The new key entry goes ONE AFTER the index of parent entry, |
| 1088 | * because the split was to the right. |
| 1089 | */ |
| 1090 | skip = parent->index + 1; |
| 1091 | |
| 1092 | /* |
| 1093 | * split or shift right remaining entries of the parent page |
| 1094 | */ |
| 1095 | nextindex = le16_to_cpu(sp->header.nextindex); |
| 1096 | /* |
| 1097 | * parent page is full - split the parent page |
| 1098 | */ |
| 1099 | if (nextindex == le16_to_cpu(sp->header.maxentry)) { |
| 1100 | /* init for parent page split */ |
| 1101 | split->mp = smp; |
| 1102 | split->index = skip; /* index at insert */ |
| 1103 | split->flag = XAD_NEW; |
| 1104 | split->off = offsetXAD(&rcp->xad[XTENTRYSTART]); |
| 1105 | split->len = JFS_SBI(ip->i_sb)->nbperpage; |
| 1106 | split->addr = rcbn; |
| 1107 | |
| 1108 | /* unpin previous right child page */ |
| 1109 | XT_PUTPAGE(rcmp); |
| 1110 | |
| 1111 | /* The split routines insert the new entry, |
| 1112 | * and acquire txLock as appropriate. |
| 1113 | * return <rp> pinned and its block number <rpbn>. |
| 1114 | */ |
| 1115 | rc = (sp->header.flag & BT_ROOT) ? |
| 1116 | xtSplitRoot(tid, ip, split, &rmp) : |
| 1117 | xtSplitPage(tid, ip, split, &rmp, &rbn); |
| 1118 | if (rc) { |
| 1119 | XT_PUTPAGE(smp); |
| 1120 | return rc; |
| 1121 | } |
| 1122 | |
| 1123 | XT_PUTPAGE(smp); |
| 1124 | /* keep new child page <rp> pinned */ |
| 1125 | } |
| 1126 | /* |
| 1127 | * parent page is not full - insert in parent page |
| 1128 | */ |
| 1129 | else { |
| 1130 | /* |
| 1131 | * insert router entry in parent for the right child |
| 1132 | * page from the first entry of the right child page: |
| 1133 | */ |
| 1134 | /* |
| 1135 | * acquire a transaction lock on the parent page; |
| 1136 | * |
| 1137 | * action: router xad insertion; |
| 1138 | */ |
| 1139 | BT_MARK_DIRTY(smp, ip); |
| 1140 | |
| 1141 | /* |
| 1142 | * if insert into middle, shift right remaining entries |
| 1143 | */ |
| 1144 | if (skip < nextindex) |
| 1145 | memmove(&sp->xad[skip + 1], &sp->xad[skip], |
| 1146 | (nextindex - |
| 1147 | skip) << L2XTSLOTSIZE); |
| 1148 | |
| 1149 | /* insert the router entry */ |
| 1150 | xad = &sp->xad[skip]; |
| 1151 | XT_PUTENTRY(xad, XAD_NEW, |
| 1152 | offsetXAD(&rcp->xad[XTENTRYSTART]), |
| 1153 | JFS_SBI(ip->i_sb)->nbperpage, rcbn); |
| 1154 | |
| 1155 | /* advance next available entry index. */ |
| 1156 | sp->header.nextindex = |
| 1157 | cpu_to_le16(le16_to_cpu(sp->header.nextindex) + |
| 1158 | 1); |
| 1159 | |
| 1160 | /* Don't log it if there are no links to the file */ |
| 1161 | if (!test_cflag(COMMIT_Nolink, ip)) { |
| 1162 | tlck = txLock(tid, ip, smp, |
| 1163 | tlckXTREE | tlckGROW); |
| 1164 | xtlck = (struct xtlock *) & tlck->lock; |
| 1165 | xtlck->lwm.offset = (xtlck->lwm.offset) ? |
| 1166 | min(skip, (int)xtlck->lwm.offset) : skip; |
| 1167 | xtlck->lwm.length = |
| 1168 | le16_to_cpu(sp->header.nextindex) - |
| 1169 | xtlck->lwm.offset; |
| 1170 | } |
| 1171 | |
| 1172 | /* unpin parent page */ |
| 1173 | XT_PUTPAGE(smp); |
| 1174 | |
| 1175 | /* exit propagate up */ |
| 1176 | break; |
| 1177 | } |
| 1178 | } |
| 1179 | |
| 1180 | /* unpin current right page */ |
| 1181 | XT_PUTPAGE(rmp); |
| 1182 | |
| 1183 | return 0; |
| 1184 | } |
| 1185 | |
| 1186 | |
| 1187 | /* |
| 1188 | * xtSplitPage() |
| 1189 | * |
| 1190 | * function: |
| 1191 | * split a full non-root page into |
| 1192 | * original/split/left page and new right page |
| 1193 | * i.e., the original/split page remains as left page. |
| 1194 | * |
| 1195 | * parameter: |
| 1196 | * int tid, |
| 1197 | * struct inode *ip, |
| 1198 | * struct xtsplit *split, |
| 1199 | * struct metapage **rmpp, |
| 1200 | * u64 *rbnp, |
| 1201 | * |
| 1202 | * return: |
| 1203 | * Pointer to page in which to insert or NULL on error. |
| 1204 | */ |
| 1205 | static int |
| 1206 | xtSplitPage(tid_t tid, struct inode *ip, |
| 1207 | struct xtsplit * split, struct metapage ** rmpp, s64 * rbnp) |
| 1208 | { |
| 1209 | int rc = 0; |
| 1210 | struct metapage *smp; |
| 1211 | xtpage_t *sp; |
| 1212 | struct metapage *rmp; |
| 1213 | xtpage_t *rp; /* new right page allocated */ |
| 1214 | s64 rbn; /* new right page block number */ |
| 1215 | struct metapage *mp; |
| 1216 | xtpage_t *p; |
| 1217 | s64 nextbn; |
| 1218 | int skip, maxentry, middle, righthalf, n; |
| 1219 | xad_t *xad; |
| 1220 | struct pxdlist *pxdlist; |
| 1221 | pxd_t *pxd; |
| 1222 | struct tlock *tlck; |
| 1223 | struct xtlock *sxtlck = NULL, *rxtlck = NULL; |
| 1224 | int quota_allocation = 0; |
| 1225 | |
| 1226 | smp = split->mp; |
| 1227 | sp = XT_PAGE(ip, smp); |
| 1228 | |
| 1229 | INCREMENT(xtStat.split); |
| 1230 | |
| 1231 | pxdlist = split->pxdlist; |
| 1232 | pxd = &pxdlist->pxd[pxdlist->npxd]; |
| 1233 | pxdlist->npxd++; |
| 1234 | rbn = addressPXD(pxd); |
| 1235 | |
| 1236 | /* Allocate blocks to quota. */ |
| 1237 | if (DQUOT_ALLOC_BLOCK(ip, lengthPXD(pxd))) { |
| 1238 | rc = -EDQUOT; |
| 1239 | goto clean_up; |
| 1240 | } |
| 1241 | |
| 1242 | quota_allocation += lengthPXD(pxd); |
| 1243 | |
| 1244 | /* |
| 1245 | * allocate the new right page for the split |
| 1246 | */ |
| 1247 | rmp = get_metapage(ip, rbn, PSIZE, 1); |
| 1248 | if (rmp == NULL) { |
| 1249 | rc = -EIO; |
| 1250 | goto clean_up; |
| 1251 | } |
| 1252 | |
| 1253 | jfs_info("xtSplitPage: ip:0x%p smp:0x%p rmp:0x%p", ip, smp, rmp); |
| 1254 | |
| 1255 | BT_MARK_DIRTY(rmp, ip); |
| 1256 | /* |
| 1257 | * action: new page; |
| 1258 | */ |
| 1259 | |
| 1260 | rp = (xtpage_t *) rmp->data; |
| 1261 | rp->header.self = *pxd; |
| 1262 | rp->header.flag = sp->header.flag & BT_TYPE; |
| 1263 | rp->header.maxentry = sp->header.maxentry; /* little-endian */ |
| 1264 | rp->header.nextindex = cpu_to_le16(XTENTRYSTART); |
| 1265 | |
| 1266 | BT_MARK_DIRTY(smp, ip); |
| 1267 | /* Don't log it if there are no links to the file */ |
| 1268 | if (!test_cflag(COMMIT_Nolink, ip)) { |
| 1269 | /* |
| 1270 | * acquire a transaction lock on the new right page; |
| 1271 | */ |
| 1272 | tlck = txLock(tid, ip, rmp, tlckXTREE | tlckNEW); |
| 1273 | rxtlck = (struct xtlock *) & tlck->lock; |
| 1274 | rxtlck->lwm.offset = XTENTRYSTART; |
| 1275 | /* |
| 1276 | * acquire a transaction lock on the split page |
| 1277 | */ |
| 1278 | tlck = txLock(tid, ip, smp, tlckXTREE | tlckGROW); |
| 1279 | sxtlck = (struct xtlock *) & tlck->lock; |
| 1280 | } |
| 1281 | |
| 1282 | /* |
| 1283 | * initialize/update sibling pointers of <sp> and <rp> |
| 1284 | */ |
| 1285 | nextbn = le64_to_cpu(sp->header.next); |
| 1286 | rp->header.next = cpu_to_le64(nextbn); |
| 1287 | rp->header.prev = cpu_to_le64(addressPXD(&sp->header.self)); |
| 1288 | sp->header.next = cpu_to_le64(rbn); |
| 1289 | |
| 1290 | skip = split->index; |
| 1291 | |
| 1292 | /* |
| 1293 | * sequential append at tail (after last entry of last page) |
| 1294 | * |
| 1295 | * if splitting the last page on a level because of appending |
| 1296 | * a entry to it (skip is maxentry), it's likely that the access is |
| 1297 | * sequential. adding an empty page on the side of the level is less |
| 1298 | * work and can push the fill factor much higher than normal. |
| 1299 | * if we're wrong it's no big deal - we will do the split the right |
| 1300 | * way next time. |
| 1301 | * (it may look like it's equally easy to do a similar hack for |
| 1302 | * reverse sorted data, that is, split the tree left, but it's not. |
| 1303 | * Be my guest.) |
| 1304 | */ |
| 1305 | if (nextbn == 0 && skip == le16_to_cpu(sp->header.maxentry)) { |
| 1306 | /* |
| 1307 | * acquire a transaction lock on the new/right page; |
| 1308 | * |
| 1309 | * action: xad insertion; |
| 1310 | */ |
| 1311 | /* insert entry at the first entry of the new right page */ |
| 1312 | xad = &rp->xad[XTENTRYSTART]; |
| 1313 | XT_PUTENTRY(xad, split->flag, split->off, split->len, |
| 1314 | split->addr); |
| 1315 | |
| 1316 | rp->header.nextindex = cpu_to_le16(XTENTRYSTART + 1); |
| 1317 | |
| 1318 | if (!test_cflag(COMMIT_Nolink, ip)) { |
| 1319 | /* rxtlck->lwm.offset = XTENTRYSTART; */ |
| 1320 | rxtlck->lwm.length = 1; |
| 1321 | } |
| 1322 | |
| 1323 | *rmpp = rmp; |
| 1324 | *rbnp = rbn; |
| 1325 | |
| 1326 | jfs_info("xtSplitPage: sp:0x%p rp:0x%p", sp, rp); |
| 1327 | return 0; |
| 1328 | } |
| 1329 | |
| 1330 | /* |
| 1331 | * non-sequential insert (at possibly middle page) |
| 1332 | */ |
| 1333 | |
| 1334 | /* |
| 1335 | * update previous pointer of old next/right page of <sp> |
| 1336 | */ |
| 1337 | if (nextbn != 0) { |
| 1338 | XT_GETPAGE(ip, nextbn, mp, PSIZE, p, rc); |
| 1339 | if (rc) { |
| 1340 | XT_PUTPAGE(rmp); |
| 1341 | goto clean_up; |
| 1342 | } |
| 1343 | |
| 1344 | BT_MARK_DIRTY(mp, ip); |
| 1345 | /* |
| 1346 | * acquire a transaction lock on the next page; |
| 1347 | * |
| 1348 | * action:sibling pointer update; |
| 1349 | */ |
| 1350 | if (!test_cflag(COMMIT_Nolink, ip)) |
| 1351 | tlck = txLock(tid, ip, mp, tlckXTREE | tlckRELINK); |
| 1352 | |
| 1353 | p->header.prev = cpu_to_le64(rbn); |
| 1354 | |
| 1355 | /* sibling page may have been updated previously, or |
| 1356 | * it may be updated later; |
| 1357 | */ |
| 1358 | |
| 1359 | XT_PUTPAGE(mp); |
| 1360 | } |
| 1361 | |
| 1362 | /* |
| 1363 | * split the data between the split and new/right pages |
| 1364 | */ |
| 1365 | maxentry = le16_to_cpu(sp->header.maxentry); |
| 1366 | middle = maxentry >> 1; |
| 1367 | righthalf = maxentry - middle; |
| 1368 | |
| 1369 | /* |
| 1370 | * skip index in old split/left page - insert into left page: |
| 1371 | */ |
| 1372 | if (skip <= middle) { |
| 1373 | /* move right half of split page to the new right page */ |
| 1374 | memmove(&rp->xad[XTENTRYSTART], &sp->xad[middle], |
| 1375 | righthalf << L2XTSLOTSIZE); |
| 1376 | |
| 1377 | /* shift right tail of left half to make room for new entry */ |
| 1378 | if (skip < middle) |
| 1379 | memmove(&sp->xad[skip + 1], &sp->xad[skip], |
| 1380 | (middle - skip) << L2XTSLOTSIZE); |
| 1381 | |
| 1382 | /* insert new entry */ |
| 1383 | xad = &sp->xad[skip]; |
| 1384 | XT_PUTENTRY(xad, split->flag, split->off, split->len, |
| 1385 | split->addr); |
| 1386 | |
| 1387 | /* update page header */ |
| 1388 | sp->header.nextindex = cpu_to_le16(middle + 1); |
| 1389 | if (!test_cflag(COMMIT_Nolink, ip)) { |
| 1390 | sxtlck->lwm.offset = (sxtlck->lwm.offset) ? |
| 1391 | min(skip, (int)sxtlck->lwm.offset) : skip; |
| 1392 | } |
| 1393 | |
| 1394 | rp->header.nextindex = |
| 1395 | cpu_to_le16(XTENTRYSTART + righthalf); |
| 1396 | } |
| 1397 | /* |
| 1398 | * skip index in new right page - insert into right page: |
| 1399 | */ |
| 1400 | else { |
| 1401 | /* move left head of right half to right page */ |
| 1402 | n = skip - middle; |
| 1403 | memmove(&rp->xad[XTENTRYSTART], &sp->xad[middle], |
| 1404 | n << L2XTSLOTSIZE); |
| 1405 | |
| 1406 | /* insert new entry */ |
| 1407 | n += XTENTRYSTART; |
| 1408 | xad = &rp->xad[n]; |
| 1409 | XT_PUTENTRY(xad, split->flag, split->off, split->len, |
| 1410 | split->addr); |
| 1411 | |
| 1412 | /* move right tail of right half to right page */ |
| 1413 | if (skip < maxentry) |
| 1414 | memmove(&rp->xad[n + 1], &sp->xad[skip], |
| 1415 | (maxentry - skip) << L2XTSLOTSIZE); |
| 1416 | |
| 1417 | /* update page header */ |
| 1418 | sp->header.nextindex = cpu_to_le16(middle); |
| 1419 | if (!test_cflag(COMMIT_Nolink, ip)) { |
| 1420 | sxtlck->lwm.offset = (sxtlck->lwm.offset) ? |
| 1421 | min(middle, (int)sxtlck->lwm.offset) : middle; |
| 1422 | } |
| 1423 | |
| 1424 | rp->header.nextindex = cpu_to_le16(XTENTRYSTART + |
| 1425 | righthalf + 1); |
| 1426 | } |
| 1427 | |
| 1428 | if (!test_cflag(COMMIT_Nolink, ip)) { |
| 1429 | sxtlck->lwm.length = le16_to_cpu(sp->header.nextindex) - |
| 1430 | sxtlck->lwm.offset; |
| 1431 | |
| 1432 | /* rxtlck->lwm.offset = XTENTRYSTART; */ |
| 1433 | rxtlck->lwm.length = le16_to_cpu(rp->header.nextindex) - |
| 1434 | XTENTRYSTART; |
| 1435 | } |
| 1436 | |
| 1437 | *rmpp = rmp; |
| 1438 | *rbnp = rbn; |
| 1439 | |
| 1440 | jfs_info("xtSplitPage: sp:0x%p rp:0x%p", sp, rp); |
| 1441 | return rc; |
| 1442 | |
| 1443 | clean_up: |
| 1444 | |
| 1445 | /* Rollback quota allocation. */ |
| 1446 | if (quota_allocation) |
| 1447 | DQUOT_FREE_BLOCK(ip, quota_allocation); |
| 1448 | |
| 1449 | return (rc); |
| 1450 | } |
| 1451 | |
| 1452 | |
| 1453 | /* |
| 1454 | * xtSplitRoot() |
| 1455 | * |
| 1456 | * function: |
| 1457 | * split the full root page into |
| 1458 | * original/root/split page and new right page |
| 1459 | * i.e., root remains fixed in tree anchor (inode) and |
| 1460 | * the root is copied to a single new right child page |
| 1461 | * since root page << non-root page, and |
| 1462 | * the split root page contains a single entry for the |
| 1463 | * new right child page. |
| 1464 | * |
| 1465 | * parameter: |
| 1466 | * int tid, |
| 1467 | * struct inode *ip, |
| 1468 | * struct xtsplit *split, |
| 1469 | * struct metapage **rmpp) |
| 1470 | * |
| 1471 | * return: |
| 1472 | * Pointer to page in which to insert or NULL on error. |
| 1473 | */ |
| 1474 | static int |
| 1475 | xtSplitRoot(tid_t tid, |
| 1476 | struct inode *ip, struct xtsplit * split, struct metapage ** rmpp) |
| 1477 | { |
| 1478 | xtpage_t *sp; |
| 1479 | struct metapage *rmp; |
| 1480 | xtpage_t *rp; |
| 1481 | s64 rbn; |
| 1482 | int skip, nextindex; |
| 1483 | xad_t *xad; |
| 1484 | pxd_t *pxd; |
| 1485 | struct pxdlist *pxdlist; |
| 1486 | struct tlock *tlck; |
| 1487 | struct xtlock *xtlck; |
| 1488 | |
| 1489 | sp = &JFS_IP(ip)->i_xtroot; |
| 1490 | |
| 1491 | INCREMENT(xtStat.split); |
| 1492 | |
| 1493 | /* |
| 1494 | * allocate a single (right) child page |
| 1495 | */ |
| 1496 | pxdlist = split->pxdlist; |
| 1497 | pxd = &pxdlist->pxd[pxdlist->npxd]; |
| 1498 | pxdlist->npxd++; |
| 1499 | rbn = addressPXD(pxd); |
| 1500 | rmp = get_metapage(ip, rbn, PSIZE, 1); |
| 1501 | if (rmp == NULL) |
| 1502 | return -EIO; |
| 1503 | |
| 1504 | /* Allocate blocks to quota. */ |
| 1505 | if (DQUOT_ALLOC_BLOCK(ip, lengthPXD(pxd))) { |
| 1506 | release_metapage(rmp); |
| 1507 | return -EDQUOT; |
| 1508 | } |
| 1509 | |
| 1510 | jfs_info("xtSplitRoot: ip:0x%p rmp:0x%p", ip, rmp); |
| 1511 | |
| 1512 | /* |
| 1513 | * acquire a transaction lock on the new right page; |
| 1514 | * |
| 1515 | * action: new page; |
| 1516 | */ |
| 1517 | BT_MARK_DIRTY(rmp, ip); |
| 1518 | |
| 1519 | rp = (xtpage_t *) rmp->data; |
| 1520 | rp->header.flag = |
| 1521 | (sp->header.flag & BT_LEAF) ? BT_LEAF : BT_INTERNAL; |
| 1522 | rp->header.self = *pxd; |
| 1523 | rp->header.nextindex = cpu_to_le16(XTENTRYSTART); |
| 1524 | rp->header.maxentry = cpu_to_le16(PSIZE >> L2XTSLOTSIZE); |
| 1525 | |
| 1526 | /* initialize sibling pointers */ |
| 1527 | rp->header.next = 0; |
| 1528 | rp->header.prev = 0; |
| 1529 | |
| 1530 | /* |
| 1531 | * copy the in-line root page into new right page extent |
| 1532 | */ |
| 1533 | nextindex = le16_to_cpu(sp->header.maxentry); |
| 1534 | memmove(&rp->xad[XTENTRYSTART], &sp->xad[XTENTRYSTART], |
| 1535 | (nextindex - XTENTRYSTART) << L2XTSLOTSIZE); |
| 1536 | |
| 1537 | /* |
| 1538 | * insert the new entry into the new right/child page |
| 1539 | * (skip index in the new right page will not change) |
| 1540 | */ |
| 1541 | skip = split->index; |
| 1542 | /* if insert into middle, shift right remaining entries */ |
| 1543 | if (skip != nextindex) |
| 1544 | memmove(&rp->xad[skip + 1], &rp->xad[skip], |
| 1545 | (nextindex - skip) * sizeof(xad_t)); |
| 1546 | |
| 1547 | xad = &rp->xad[skip]; |
| 1548 | XT_PUTENTRY(xad, split->flag, split->off, split->len, split->addr); |
| 1549 | |
| 1550 | /* update page header */ |
| 1551 | rp->header.nextindex = cpu_to_le16(nextindex + 1); |
| 1552 | |
| 1553 | if (!test_cflag(COMMIT_Nolink, ip)) { |
| 1554 | tlck = txLock(tid, ip, rmp, tlckXTREE | tlckNEW); |
| 1555 | xtlck = (struct xtlock *) & tlck->lock; |
| 1556 | xtlck->lwm.offset = XTENTRYSTART; |
| 1557 | xtlck->lwm.length = le16_to_cpu(rp->header.nextindex) - |
| 1558 | XTENTRYSTART; |
| 1559 | } |
| 1560 | |
| 1561 | /* |
| 1562 | * reset the root |
| 1563 | * |
| 1564 | * init root with the single entry for the new right page |
| 1565 | * set the 1st entry offset to 0, which force the left-most key |
| 1566 | * at any level of the tree to be less than any search key. |
| 1567 | */ |
| 1568 | /* |
| 1569 | * acquire a transaction lock on the root page (in-memory inode); |
| 1570 | * |
| 1571 | * action: root split; |
| 1572 | */ |
| 1573 | BT_MARK_DIRTY(split->mp, ip); |
| 1574 | |
| 1575 | xad = &sp->xad[XTENTRYSTART]; |
| 1576 | XT_PUTENTRY(xad, XAD_NEW, 0, JFS_SBI(ip->i_sb)->nbperpage, rbn); |
| 1577 | |
| 1578 | /* update page header of root */ |
| 1579 | sp->header.flag &= ~BT_LEAF; |
| 1580 | sp->header.flag |= BT_INTERNAL; |
| 1581 | |
| 1582 | sp->header.nextindex = cpu_to_le16(XTENTRYSTART + 1); |
| 1583 | |
| 1584 | if (!test_cflag(COMMIT_Nolink, ip)) { |
| 1585 | tlck = txLock(tid, ip, split->mp, tlckXTREE | tlckGROW); |
| 1586 | xtlck = (struct xtlock *) & tlck->lock; |
| 1587 | xtlck->lwm.offset = XTENTRYSTART; |
| 1588 | xtlck->lwm.length = 1; |
| 1589 | } |
| 1590 | |
| 1591 | *rmpp = rmp; |
| 1592 | |
| 1593 | jfs_info("xtSplitRoot: sp:0x%p rp:0x%p", sp, rp); |
| 1594 | return 0; |
| 1595 | } |
| 1596 | |
| 1597 | |
| 1598 | /* |
| 1599 | * xtExtend() |
| 1600 | * |
| 1601 | * function: extend in-place; |
| 1602 | * |
| 1603 | * note: existing extent may or may not have been committed. |
| 1604 | * caller is responsible for pager buffer cache update, and |
| 1605 | * working block allocation map update; |
| 1606 | * update pmap: alloc whole extended extent; |
| 1607 | */ |
| 1608 | int xtExtend(tid_t tid, /* transaction id */ |
| 1609 | struct inode *ip, s64 xoff, /* delta extent offset */ |
| 1610 | s32 xlen, /* delta extent length */ |
| 1611 | int flag) |
| 1612 | { |
| 1613 | int rc = 0; |
| 1614 | int cmp; |
| 1615 | struct metapage *mp; /* meta-page buffer */ |
| 1616 | xtpage_t *p; /* base B+-tree index page */ |
| 1617 | s64 bn; |
| 1618 | int index, nextindex, len; |
| 1619 | struct btstack btstack; /* traverse stack */ |
| 1620 | struct xtsplit split; /* split information */ |
| 1621 | xad_t *xad; |
| 1622 | s64 xaddr; |
| 1623 | struct tlock *tlck; |
| 1624 | struct xtlock *xtlck = NULL; |
| 1625 | |
| 1626 | jfs_info("xtExtend: nxoff:0x%lx nxlen:0x%x", (ulong) xoff, xlen); |
| 1627 | |
| 1628 | /* there must exist extent to be extended */ |
| 1629 | if ((rc = xtSearch(ip, xoff - 1, &cmp, &btstack, XT_INSERT))) |
| 1630 | return rc; |
| 1631 | |
| 1632 | /* retrieve search result */ |
| 1633 | XT_GETSEARCH(ip, btstack.top, bn, mp, p, index); |
| 1634 | |
| 1635 | if (cmp != 0) { |
| 1636 | XT_PUTPAGE(mp); |
| 1637 | jfs_error(ip->i_sb, "xtExtend: xtSearch did not find extent"); |
| 1638 | return -EIO; |
| 1639 | } |
| 1640 | |
| 1641 | /* extension must be contiguous */ |
| 1642 | xad = &p->xad[index]; |
| 1643 | if ((offsetXAD(xad) + lengthXAD(xad)) != xoff) { |
| 1644 | XT_PUTPAGE(mp); |
| 1645 | jfs_error(ip->i_sb, "xtExtend: extension is not contiguous"); |
| 1646 | return -EIO; |
| 1647 | } |
| 1648 | |
| 1649 | /* |
| 1650 | * acquire a transaction lock on the leaf page; |
| 1651 | * |
| 1652 | * action: xad insertion/extension; |
| 1653 | */ |
| 1654 | BT_MARK_DIRTY(mp, ip); |
| 1655 | if (!test_cflag(COMMIT_Nolink, ip)) { |
| 1656 | tlck = txLock(tid, ip, mp, tlckXTREE | tlckGROW); |
| 1657 | xtlck = (struct xtlock *) & tlck->lock; |
| 1658 | } |
| 1659 | |
| 1660 | /* extend will overflow extent ? */ |
| 1661 | xlen = lengthXAD(xad) + xlen; |
| 1662 | if ((len = xlen - MAXXLEN) <= 0) |
| 1663 | goto extendOld; |
| 1664 | |
| 1665 | /* |
| 1666 | * extent overflow: insert entry for new extent |
| 1667 | */ |
| 1668 | //insertNew: |
| 1669 | xoff = offsetXAD(xad) + MAXXLEN; |
| 1670 | xaddr = addressXAD(xad) + MAXXLEN; |
| 1671 | nextindex = le16_to_cpu(p->header.nextindex); |
| 1672 | |
| 1673 | /* |
| 1674 | * if the leaf page is full, insert the new entry and |
| 1675 | * propagate up the router entry for the new page from split |
| 1676 | * |
| 1677 | * The xtSplitUp() will insert the entry and unpin the leaf page. |
| 1678 | */ |
| 1679 | if (nextindex == le16_to_cpu(p->header.maxentry)) { |
| 1680 | /* xtSpliUp() unpins leaf pages */ |
| 1681 | split.mp = mp; |
| 1682 | split.index = index + 1; |
| 1683 | split.flag = XAD_NEW; |
| 1684 | split.off = xoff; /* split offset */ |
| 1685 | split.len = len; |
| 1686 | split.addr = xaddr; |
| 1687 | split.pxdlist = NULL; |
| 1688 | if ((rc = xtSplitUp(tid, ip, &split, &btstack))) |
| 1689 | return rc; |
| 1690 | |
| 1691 | /* get back old page */ |
| 1692 | XT_GETPAGE(ip, bn, mp, PSIZE, p, rc); |
| 1693 | if (rc) |
| 1694 | return rc; |
| 1695 | /* |
| 1696 | * if leaf root has been split, original root has been |
| 1697 | * copied to new child page, i.e., original entry now |
| 1698 | * resides on the new child page; |
| 1699 | */ |
| 1700 | if (p->header.flag & BT_INTERNAL) { |
| 1701 | ASSERT(p->header.nextindex == |
| 1702 | cpu_to_le16(XTENTRYSTART + 1)); |
| 1703 | xad = &p->xad[XTENTRYSTART]; |
| 1704 | bn = addressXAD(xad); |
| 1705 | XT_PUTPAGE(mp); |
| 1706 | |
| 1707 | /* get new child page */ |
| 1708 | XT_GETPAGE(ip, bn, mp, PSIZE, p, rc); |
| 1709 | if (rc) |
| 1710 | return rc; |
| 1711 | |
| 1712 | BT_MARK_DIRTY(mp, ip); |
| 1713 | if (!test_cflag(COMMIT_Nolink, ip)) { |
| 1714 | tlck = txLock(tid, ip, mp, tlckXTREE|tlckGROW); |
| 1715 | xtlck = (struct xtlock *) & tlck->lock; |
| 1716 | } |
| 1717 | } |
| 1718 | } |
| 1719 | /* |
| 1720 | * insert the new entry into the leaf page |
| 1721 | */ |
| 1722 | else { |
| 1723 | /* insert the new entry: mark the entry NEW */ |
| 1724 | xad = &p->xad[index + 1]; |
| 1725 | XT_PUTENTRY(xad, XAD_NEW, xoff, len, xaddr); |
| 1726 | |
| 1727 | /* advance next available entry index */ |
| 1728 | p->header.nextindex = |
| 1729 | cpu_to_le16(le16_to_cpu(p->header.nextindex) + 1); |
| 1730 | } |
| 1731 | |
| 1732 | /* get back old entry */ |
| 1733 | xad = &p->xad[index]; |
| 1734 | xlen = MAXXLEN; |
| 1735 | |
| 1736 | /* |
| 1737 | * extend old extent |
| 1738 | */ |
| 1739 | extendOld: |
| 1740 | XADlength(xad, xlen); |
| 1741 | if (!(xad->flag & XAD_NEW)) |
| 1742 | xad->flag |= XAD_EXTENDED; |
| 1743 | |
| 1744 | if (!test_cflag(COMMIT_Nolink, ip)) { |
| 1745 | xtlck->lwm.offset = |
| 1746 | (xtlck->lwm.offset) ? min(index, |
| 1747 | (int)xtlck->lwm.offset) : index; |
| 1748 | xtlck->lwm.length = |
| 1749 | le16_to_cpu(p->header.nextindex) - xtlck->lwm.offset; |
| 1750 | } |
| 1751 | |
| 1752 | /* unpin the leaf page */ |
| 1753 | XT_PUTPAGE(mp); |
| 1754 | |
| 1755 | return rc; |
| 1756 | } |
| 1757 | |
| 1758 | #ifdef _NOTYET |
| 1759 | /* |
| 1760 | * xtTailgate() |
| 1761 | * |
| 1762 | * function: split existing 'tail' extent |
| 1763 | * (split offset >= start offset of tail extent), and |
| 1764 | * relocate and extend the split tail half; |
| 1765 | * |
| 1766 | * note: existing extent may or may not have been committed. |
| 1767 | * caller is responsible for pager buffer cache update, and |
| 1768 | * working block allocation map update; |
| 1769 | * update pmap: free old split tail extent, alloc new extent; |
| 1770 | */ |
| 1771 | int xtTailgate(tid_t tid, /* transaction id */ |
| 1772 | struct inode *ip, s64 xoff, /* split/new extent offset */ |
| 1773 | s32 xlen, /* new extent length */ |
| 1774 | s64 xaddr, /* new extent address */ |
| 1775 | int flag) |
| 1776 | { |
| 1777 | int rc = 0; |
| 1778 | int cmp; |
| 1779 | struct metapage *mp; /* meta-page buffer */ |
| 1780 | xtpage_t *p; /* base B+-tree index page */ |
| 1781 | s64 bn; |
| 1782 | int index, nextindex, llen, rlen; |
| 1783 | struct btstack btstack; /* traverse stack */ |
| 1784 | struct xtsplit split; /* split information */ |
| 1785 | xad_t *xad; |
| 1786 | struct tlock *tlck; |
| 1787 | struct xtlock *xtlck = 0; |
| 1788 | struct tlock *mtlck; |
| 1789 | struct maplock *pxdlock; |
| 1790 | |
| 1791 | /* |
| 1792 | printf("xtTailgate: nxoff:0x%lx nxlen:0x%x nxaddr:0x%lx\n", |
| 1793 | (ulong)xoff, xlen, (ulong)xaddr); |
| 1794 | */ |
| 1795 | |
| 1796 | /* there must exist extent to be tailgated */ |
| 1797 | if ((rc = xtSearch(ip, xoff, &cmp, &btstack, XT_INSERT))) |
| 1798 | return rc; |
| 1799 | |
| 1800 | /* retrieve search result */ |
| 1801 | XT_GETSEARCH(ip, btstack.top, bn, mp, p, index); |
| 1802 | |
| 1803 | if (cmp != 0) { |
| 1804 | XT_PUTPAGE(mp); |
| 1805 | jfs_error(ip->i_sb, "xtTailgate: couldn't find extent"); |
| 1806 | return -EIO; |
| 1807 | } |
| 1808 | |
| 1809 | /* entry found must be last entry */ |
| 1810 | nextindex = le16_to_cpu(p->header.nextindex); |
| 1811 | if (index != nextindex - 1) { |
| 1812 | XT_PUTPAGE(mp); |
| 1813 | jfs_error(ip->i_sb, |
| 1814 | "xtTailgate: the entry found is not the last entry"); |
| 1815 | return -EIO; |
| 1816 | } |
| 1817 | |
| 1818 | BT_MARK_DIRTY(mp, ip); |
| 1819 | /* |
| 1820 | * acquire tlock of the leaf page containing original entry |
| 1821 | */ |
| 1822 | if (!test_cflag(COMMIT_Nolink, ip)) { |
| 1823 | tlck = txLock(tid, ip, mp, tlckXTREE | tlckGROW); |
| 1824 | xtlck = (struct xtlock *) & tlck->lock; |
| 1825 | } |
| 1826 | |
| 1827 | /* completely replace extent ? */ |
| 1828 | xad = &p->xad[index]; |
| 1829 | /* |
| 1830 | printf("xtTailgate: xoff:0x%lx xlen:0x%x xaddr:0x%lx\n", |
| 1831 | (ulong)offsetXAD(xad), lengthXAD(xad), (ulong)addressXAD(xad)); |
| 1832 | */ |
| 1833 | if ((llen = xoff - offsetXAD(xad)) == 0) |
| 1834 | goto updateOld; |
| 1835 | |
| 1836 | /* |
| 1837 | * partially replace extent: insert entry for new extent |
| 1838 | */ |
| 1839 | //insertNew: |
| 1840 | /* |
| 1841 | * if the leaf page is full, insert the new entry and |
| 1842 | * propagate up the router entry for the new page from split |
| 1843 | * |
| 1844 | * The xtSplitUp() will insert the entry and unpin the leaf page. |
| 1845 | */ |
| 1846 | if (nextindex == le16_to_cpu(p->header.maxentry)) { |
| 1847 | /* xtSpliUp() unpins leaf pages */ |
| 1848 | split.mp = mp; |
| 1849 | split.index = index + 1; |
| 1850 | split.flag = XAD_NEW; |
| 1851 | split.off = xoff; /* split offset */ |
| 1852 | split.len = xlen; |
| 1853 | split.addr = xaddr; |
| 1854 | split.pxdlist = NULL; |
| 1855 | if ((rc = xtSplitUp(tid, ip, &split, &btstack))) |
| 1856 | return rc; |
| 1857 | |
| 1858 | /* get back old page */ |
| 1859 | XT_GETPAGE(ip, bn, mp, PSIZE, p, rc); |
| 1860 | if (rc) |
| 1861 | return rc; |
| 1862 | /* |
| 1863 | * if leaf root has been split, original root has been |
| 1864 | * copied to new child page, i.e., original entry now |
| 1865 | * resides on the new child page; |
| 1866 | */ |
| 1867 | if (p->header.flag & BT_INTERNAL) { |
| 1868 | ASSERT(p->header.nextindex == |
| 1869 | cpu_to_le16(XTENTRYSTART + 1)); |
| 1870 | xad = &p->xad[XTENTRYSTART]; |
| 1871 | bn = addressXAD(xad); |
| 1872 | XT_PUTPAGE(mp); |
| 1873 | |
| 1874 | /* get new child page */ |
| 1875 | XT_GETPAGE(ip, bn, mp, PSIZE, p, rc); |
| 1876 | if (rc) |
| 1877 | return rc; |
| 1878 | |
| 1879 | BT_MARK_DIRTY(mp, ip); |
| 1880 | if (!test_cflag(COMMIT_Nolink, ip)) { |
| 1881 | tlck = txLock(tid, ip, mp, tlckXTREE|tlckGROW); |
| 1882 | xtlck = (struct xtlock *) & tlck->lock; |
| 1883 | } |
| 1884 | } |
| 1885 | } |
| 1886 | /* |
| 1887 | * insert the new entry into the leaf page |
| 1888 | */ |
| 1889 | else { |
| 1890 | /* insert the new entry: mark the entry NEW */ |
| 1891 | xad = &p->xad[index + 1]; |
| 1892 | XT_PUTENTRY(xad, XAD_NEW, xoff, xlen, xaddr); |
| 1893 | |
| 1894 | /* advance next available entry index */ |
| 1895 | p->header.nextindex = |
| 1896 | cpu_to_le16(le16_to_cpu(p->header.nextindex) + 1); |
| 1897 | } |
| 1898 | |
| 1899 | /* get back old XAD */ |
| 1900 | xad = &p->xad[index]; |
| 1901 | |
| 1902 | /* |
| 1903 | * truncate/relocate old extent at split offset |
| 1904 | */ |
| 1905 | updateOld: |
| 1906 | /* update dmap for old/committed/truncated extent */ |
| 1907 | rlen = lengthXAD(xad) - llen; |
| 1908 | if (!(xad->flag & XAD_NEW)) { |
| 1909 | /* free from PWMAP at commit */ |
| 1910 | if (!test_cflag(COMMIT_Nolink, ip)) { |
| 1911 | mtlck = txMaplock(tid, ip, tlckMAP); |
| 1912 | pxdlock = (struct maplock *) & mtlck->lock; |
| 1913 | pxdlock->flag = mlckFREEPXD; |
| 1914 | PXDaddress(&pxdlock->pxd, addressXAD(xad) + llen); |
| 1915 | PXDlength(&pxdlock->pxd, rlen); |
| 1916 | pxdlock->index = 1; |
| 1917 | } |
| 1918 | } else |
| 1919 | /* free from WMAP */ |
| 1920 | dbFree(ip, addressXAD(xad) + llen, (s64) rlen); |
| 1921 | |
| 1922 | if (llen) |
| 1923 | /* truncate */ |
| 1924 | XADlength(xad, llen); |
| 1925 | else |
| 1926 | /* replace */ |
| 1927 | XT_PUTENTRY(xad, XAD_NEW, xoff, xlen, xaddr); |
| 1928 | |
| 1929 | if (!test_cflag(COMMIT_Nolink, ip)) { |
| 1930 | xtlck->lwm.offset = (xtlck->lwm.offset) ? |
| 1931 | min(index, (int)xtlck->lwm.offset) : index; |
| 1932 | xtlck->lwm.length = le16_to_cpu(p->header.nextindex) - |
| 1933 | xtlck->lwm.offset; |
| 1934 | } |
| 1935 | |
| 1936 | /* unpin the leaf page */ |
| 1937 | XT_PUTPAGE(mp); |
| 1938 | |
| 1939 | return rc; |
| 1940 | } |
| 1941 | #endif /* _NOTYET */ |
| 1942 | |
| 1943 | /* |
| 1944 | * xtUpdate() |
| 1945 | * |
| 1946 | * function: update XAD; |
| 1947 | * |
| 1948 | * update extent for allocated_but_not_recorded or |
| 1949 | * compressed extent; |
| 1950 | * |
| 1951 | * parameter: |
| 1952 | * nxad - new XAD; |
| 1953 | * logical extent of the specified XAD must be completely |
| 1954 | * contained by an existing XAD; |
| 1955 | */ |
| 1956 | int xtUpdate(tid_t tid, struct inode *ip, xad_t * nxad) |
| 1957 | { /* new XAD */ |
| 1958 | int rc = 0; |
| 1959 | int cmp; |
| 1960 | struct metapage *mp; /* meta-page buffer */ |
| 1961 | xtpage_t *p; /* base B+-tree index page */ |
| 1962 | s64 bn; |
| 1963 | int index0, index, newindex, nextindex; |
| 1964 | struct btstack btstack; /* traverse stack */ |
| 1965 | struct xtsplit split; /* split information */ |
| 1966 | xad_t *xad, *lxad, *rxad; |
| 1967 | int xflag; |
| 1968 | s64 nxoff, xoff; |
| 1969 | int nxlen, xlen, lxlen, rxlen; |
| 1970 | s64 nxaddr, xaddr; |
| 1971 | struct tlock *tlck; |
| 1972 | struct xtlock *xtlck = NULL; |
| 1973 | int newpage = 0; |
| 1974 | |
| 1975 | /* there must exist extent to be tailgated */ |
| 1976 | nxoff = offsetXAD(nxad); |
| 1977 | nxlen = lengthXAD(nxad); |
| 1978 | nxaddr = addressXAD(nxad); |
| 1979 | |
| 1980 | if ((rc = xtSearch(ip, nxoff, &cmp, &btstack, XT_INSERT))) |
| 1981 | return rc; |
| 1982 | |
| 1983 | /* retrieve search result */ |
| 1984 | XT_GETSEARCH(ip, btstack.top, bn, mp, p, index0); |
| 1985 | |
| 1986 | if (cmp != 0) { |
| 1987 | XT_PUTPAGE(mp); |
| 1988 | jfs_error(ip->i_sb, "xtUpdate: Could not find extent"); |
| 1989 | return -EIO; |
| 1990 | } |
| 1991 | |
| 1992 | BT_MARK_DIRTY(mp, ip); |
| 1993 | /* |
| 1994 | * acquire tlock of the leaf page containing original entry |
| 1995 | */ |
| 1996 | if (!test_cflag(COMMIT_Nolink, ip)) { |
| 1997 | tlck = txLock(tid, ip, mp, tlckXTREE | tlckGROW); |
| 1998 | xtlck = (struct xtlock *) & tlck->lock; |
| 1999 | } |
| 2000 | |
| 2001 | xad = &p->xad[index0]; |
| 2002 | xflag = xad->flag; |
| 2003 | xoff = offsetXAD(xad); |
| 2004 | xlen = lengthXAD(xad); |
| 2005 | xaddr = addressXAD(xad); |
| 2006 | |
| 2007 | /* nXAD must be completely contained within XAD */ |
| 2008 | if ((xoff > nxoff) || |
| 2009 | (nxoff + nxlen > xoff + xlen)) { |
| 2010 | XT_PUTPAGE(mp); |
| 2011 | jfs_error(ip->i_sb, |
| 2012 | "xtUpdate: nXAD in not completely contained within XAD"); |
| 2013 | return -EIO; |
| 2014 | } |
| 2015 | |
| 2016 | index = index0; |
| 2017 | newindex = index + 1; |
| 2018 | nextindex = le16_to_cpu(p->header.nextindex); |
| 2019 | |
| 2020 | #ifdef _JFS_WIP_NOCOALESCE |
| 2021 | if (xoff < nxoff) |
| 2022 | goto updateRight; |
| 2023 | |
| 2024 | /* |
| 2025 | * replace XAD with nXAD |
| 2026 | */ |
| 2027 | replace: /* (nxoff == xoff) */ |
| 2028 | if (nxlen == xlen) { |
| 2029 | /* replace XAD with nXAD:recorded */ |
| 2030 | *xad = *nxad; |
| 2031 | xad->flag = xflag & ~XAD_NOTRECORDED; |
| 2032 | |
| 2033 | goto out; |
| 2034 | } else /* (nxlen < xlen) */ |
| 2035 | goto updateLeft; |
| 2036 | #endif /* _JFS_WIP_NOCOALESCE */ |
| 2037 | |
| 2038 | /* #ifdef _JFS_WIP_COALESCE */ |
| 2039 | if (xoff < nxoff) |
| 2040 | goto coalesceRight; |
| 2041 | |
| 2042 | /* |
| 2043 | * coalesce with left XAD |
| 2044 | */ |
| 2045 | //coalesceLeft: /* (xoff == nxoff) */ |
| 2046 | /* is XAD first entry of page ? */ |
| 2047 | if (index == XTENTRYSTART) |
| 2048 | goto replace; |
| 2049 | |
| 2050 | /* is nXAD logically and physically contiguous with lXAD ? */ |
| 2051 | lxad = &p->xad[index - 1]; |
| 2052 | lxlen = lengthXAD(lxad); |
| 2053 | if (!(lxad->flag & XAD_NOTRECORDED) && |
| 2054 | (nxoff == offsetXAD(lxad) + lxlen) && |
| 2055 | (nxaddr == addressXAD(lxad) + lxlen) && |
| 2056 | (lxlen + nxlen < MAXXLEN)) { |
| 2057 | /* extend right lXAD */ |
| 2058 | index0 = index - 1; |
| 2059 | XADlength(lxad, lxlen + nxlen); |
| 2060 | |
| 2061 | /* If we just merged two extents together, need to make sure the |
| 2062 | * right extent gets logged. If the left one is marked XAD_NEW, |
| 2063 | * then we know it will be logged. Otherwise, mark as |
| 2064 | * XAD_EXTENDED |
| 2065 | */ |
| 2066 | if (!(lxad->flag & XAD_NEW)) |
| 2067 | lxad->flag |= XAD_EXTENDED; |
| 2068 | |
| 2069 | if (xlen > nxlen) { |
| 2070 | /* truncate XAD */ |
| 2071 | XADoffset(xad, xoff + nxlen); |
| 2072 | XADlength(xad, xlen - nxlen); |
| 2073 | XADaddress(xad, xaddr + nxlen); |
| 2074 | goto out; |
| 2075 | } else { /* (xlen == nxlen) */ |
| 2076 | |
| 2077 | /* remove XAD */ |
| 2078 | if (index < nextindex - 1) |
| 2079 | memmove(&p->xad[index], &p->xad[index + 1], |
| 2080 | (nextindex - index - |
| 2081 | 1) << L2XTSLOTSIZE); |
| 2082 | |
| 2083 | p->header.nextindex = |
| 2084 | cpu_to_le16(le16_to_cpu(p->header.nextindex) - |
| 2085 | 1); |
| 2086 | |
| 2087 | index = index0; |
| 2088 | newindex = index + 1; |
| 2089 | nextindex = le16_to_cpu(p->header.nextindex); |
| 2090 | xoff = nxoff = offsetXAD(lxad); |
| 2091 | xlen = nxlen = lxlen + nxlen; |
| 2092 | xaddr = nxaddr = addressXAD(lxad); |
| 2093 | goto coalesceRight; |
| 2094 | } |
| 2095 | } |
| 2096 | |
| 2097 | /* |
| 2098 | * replace XAD with nXAD |
| 2099 | */ |
| 2100 | replace: /* (nxoff == xoff) */ |
| 2101 | if (nxlen == xlen) { |
| 2102 | /* replace XAD with nXAD:recorded */ |
| 2103 | *xad = *nxad; |
| 2104 | xad->flag = xflag & ~XAD_NOTRECORDED; |
| 2105 | |
| 2106 | goto coalesceRight; |
| 2107 | } else /* (nxlen < xlen) */ |
| 2108 | goto updateLeft; |
| 2109 | |
| 2110 | /* |
| 2111 | * coalesce with right XAD |
| 2112 | */ |
| 2113 | coalesceRight: /* (xoff <= nxoff) */ |
| 2114 | /* is XAD last entry of page ? */ |
| 2115 | if (newindex == nextindex) { |
| 2116 | if (xoff == nxoff) |
| 2117 | goto out; |
| 2118 | goto updateRight; |
| 2119 | } |
| 2120 | |
| 2121 | /* is nXAD logically and physically contiguous with rXAD ? */ |
| 2122 | rxad = &p->xad[index + 1]; |
| 2123 | rxlen = lengthXAD(rxad); |
| 2124 | if (!(rxad->flag & XAD_NOTRECORDED) && |
| 2125 | (nxoff + nxlen == offsetXAD(rxad)) && |
| 2126 | (nxaddr + nxlen == addressXAD(rxad)) && |
| 2127 | (rxlen + nxlen < MAXXLEN)) { |
| 2128 | /* extend left rXAD */ |
| 2129 | XADoffset(rxad, nxoff); |
| 2130 | XADlength(rxad, rxlen + nxlen); |
| 2131 | XADaddress(rxad, nxaddr); |
| 2132 | |
| 2133 | /* If we just merged two extents together, need to make sure |
| 2134 | * the left extent gets logged. If the right one is marked |
| 2135 | * XAD_NEW, then we know it will be logged. Otherwise, mark as |
| 2136 | * XAD_EXTENDED |
| 2137 | */ |
| 2138 | if (!(rxad->flag & XAD_NEW)) |
| 2139 | rxad->flag |= XAD_EXTENDED; |
| 2140 | |
| 2141 | if (xlen > nxlen) |
| 2142 | /* truncate XAD */ |
| 2143 | XADlength(xad, xlen - nxlen); |
| 2144 | else { /* (xlen == nxlen) */ |
| 2145 | |
| 2146 | /* remove XAD */ |
| 2147 | memmove(&p->xad[index], &p->xad[index + 1], |
| 2148 | (nextindex - index - 1) << L2XTSLOTSIZE); |
| 2149 | |
| 2150 | p->header.nextindex = |
| 2151 | cpu_to_le16(le16_to_cpu(p->header.nextindex) - |
| 2152 | 1); |
| 2153 | } |
| 2154 | |
| 2155 | goto out; |
| 2156 | } else if (xoff == nxoff) |
| 2157 | goto out; |
| 2158 | |
| 2159 | if (xoff >= nxoff) { |
| 2160 | XT_PUTPAGE(mp); |
| 2161 | jfs_error(ip->i_sb, "xtUpdate: xoff >= nxoff"); |
| 2162 | return -EIO; |
| 2163 | } |
| 2164 | /* #endif _JFS_WIP_COALESCE */ |
| 2165 | |
| 2166 | /* |
| 2167 | * split XAD into (lXAD, nXAD): |
| 2168 | * |
| 2169 | * |---nXAD---> |
| 2170 | * --|----------XAD----------|-- |
| 2171 | * |-lXAD-| |
| 2172 | */ |
| 2173 | updateRight: /* (xoff < nxoff) */ |
| 2174 | /* truncate old XAD as lXAD:not_recorded */ |
| 2175 | xad = &p->xad[index]; |
| 2176 | XADlength(xad, nxoff - xoff); |
| 2177 | |
| 2178 | /* insert nXAD:recorded */ |
| 2179 | if (nextindex == le16_to_cpu(p->header.maxentry)) { |
| 2180 | |
| 2181 | /* xtSpliUp() unpins leaf pages */ |
| 2182 | split.mp = mp; |
| 2183 | split.index = newindex; |
| 2184 | split.flag = xflag & ~XAD_NOTRECORDED; |
| 2185 | split.off = nxoff; |
| 2186 | split.len = nxlen; |
| 2187 | split.addr = nxaddr; |
| 2188 | split.pxdlist = NULL; |
| 2189 | if ((rc = xtSplitUp(tid, ip, &split, &btstack))) |
| 2190 | return rc; |
| 2191 | |
| 2192 | /* get back old page */ |
| 2193 | XT_GETPAGE(ip, bn, mp, PSIZE, p, rc); |
| 2194 | if (rc) |
| 2195 | return rc; |
| 2196 | /* |
| 2197 | * if leaf root has been split, original root has been |
| 2198 | * copied to new child page, i.e., original entry now |
| 2199 | * resides on the new child page; |
| 2200 | */ |
| 2201 | if (p->header.flag & BT_INTERNAL) { |
| 2202 | ASSERT(p->header.nextindex == |
| 2203 | cpu_to_le16(XTENTRYSTART + 1)); |
| 2204 | xad = &p->xad[XTENTRYSTART]; |
| 2205 | bn = addressXAD(xad); |
| 2206 | XT_PUTPAGE(mp); |
| 2207 | |
| 2208 | /* get new child page */ |
| 2209 | XT_GETPAGE(ip, bn, mp, PSIZE, p, rc); |
| 2210 | if (rc) |
| 2211 | return rc; |
| 2212 | |
| 2213 | BT_MARK_DIRTY(mp, ip); |
| 2214 | if (!test_cflag(COMMIT_Nolink, ip)) { |
| 2215 | tlck = txLock(tid, ip, mp, tlckXTREE|tlckGROW); |
| 2216 | xtlck = (struct xtlock *) & tlck->lock; |
| 2217 | } |
| 2218 | } else { |
| 2219 | /* is nXAD on new page ? */ |
| 2220 | if (newindex > |
| 2221 | (le16_to_cpu(p->header.maxentry) >> 1)) { |
| 2222 | newindex = |
| 2223 | newindex - |
| 2224 | le16_to_cpu(p->header.nextindex) + |
| 2225 | XTENTRYSTART; |
| 2226 | newpage = 1; |
| 2227 | } |
| 2228 | } |
| 2229 | } else { |
| 2230 | /* if insert into middle, shift right remaining entries */ |
| 2231 | if (newindex < nextindex) |
| 2232 | memmove(&p->xad[newindex + 1], &p->xad[newindex], |
| 2233 | (nextindex - newindex) << L2XTSLOTSIZE); |
| 2234 | |
| 2235 | /* insert the entry */ |
| 2236 | xad = &p->xad[newindex]; |
| 2237 | *xad = *nxad; |
| 2238 | xad->flag = xflag & ~XAD_NOTRECORDED; |
| 2239 | |
| 2240 | /* advance next available entry index. */ |
| 2241 | p->header.nextindex = |
| 2242 | cpu_to_le16(le16_to_cpu(p->header.nextindex) + 1); |
| 2243 | } |
| 2244 | |
| 2245 | /* |
| 2246 | * does nXAD force 3-way split ? |
| 2247 | * |
| 2248 | * |---nXAD--->| |
| 2249 | * --|----------XAD-------------|-- |
| 2250 | * |-lXAD-| |-rXAD -| |
| 2251 | */ |
| 2252 | if (nxoff + nxlen == xoff + xlen) |
| 2253 | goto out; |
| 2254 | |
| 2255 | /* reorient nXAD as XAD for further split XAD into (nXAD, rXAD) */ |
| 2256 | if (newpage) { |
| 2257 | /* close out old page */ |
| 2258 | if (!test_cflag(COMMIT_Nolink, ip)) { |
| 2259 | xtlck->lwm.offset = (xtlck->lwm.offset) ? |
| 2260 | min(index0, (int)xtlck->lwm.offset) : index0; |
| 2261 | xtlck->lwm.length = |
| 2262 | le16_to_cpu(p->header.nextindex) - |
| 2263 | xtlck->lwm.offset; |
| 2264 | } |
| 2265 | |
| 2266 | bn = le64_to_cpu(p->header.next); |
| 2267 | XT_PUTPAGE(mp); |
| 2268 | |
| 2269 | /* get new right page */ |
| 2270 | XT_GETPAGE(ip, bn, mp, PSIZE, p, rc); |
| 2271 | if (rc) |
| 2272 | return rc; |
| 2273 | |
| 2274 | BT_MARK_DIRTY(mp, ip); |
| 2275 | if (!test_cflag(COMMIT_Nolink, ip)) { |
| 2276 | tlck = txLock(tid, ip, mp, tlckXTREE | tlckGROW); |
| 2277 | xtlck = (struct xtlock *) & tlck->lock; |
| 2278 | } |
| 2279 | |
| 2280 | index0 = index = newindex; |
| 2281 | } else |
| 2282 | index++; |
| 2283 | |
| 2284 | newindex = index + 1; |
| 2285 | nextindex = le16_to_cpu(p->header.nextindex); |
| 2286 | xlen = xlen - (nxoff - xoff); |
| 2287 | xoff = nxoff; |
| 2288 | xaddr = nxaddr; |
| 2289 | |
| 2290 | /* recompute split pages */ |
| 2291 | if (nextindex == le16_to_cpu(p->header.maxentry)) { |
| 2292 | XT_PUTPAGE(mp); |
| 2293 | |
| 2294 | if ((rc = xtSearch(ip, nxoff, &cmp, &btstack, XT_INSERT))) |
| 2295 | return rc; |
| 2296 | |
| 2297 | /* retrieve search result */ |
| 2298 | XT_GETSEARCH(ip, btstack.top, bn, mp, p, index0); |
| 2299 | |
| 2300 | if (cmp != 0) { |
| 2301 | XT_PUTPAGE(mp); |
| 2302 | jfs_error(ip->i_sb, "xtUpdate: xtSearch failed"); |
| 2303 | return -EIO; |
| 2304 | } |
| 2305 | |
| 2306 | if (index0 != index) { |
| 2307 | XT_PUTPAGE(mp); |
| 2308 | jfs_error(ip->i_sb, |
| 2309 | "xtUpdate: unexpected value of index"); |
| 2310 | return -EIO; |
| 2311 | } |
| 2312 | } |
| 2313 | |
| 2314 | /* |
| 2315 | * split XAD into (nXAD, rXAD) |
| 2316 | * |
| 2317 | * ---nXAD---| |
| 2318 | * --|----------XAD----------|-- |
| 2319 | * |-rXAD-| |
| 2320 | */ |
| 2321 | updateLeft: /* (nxoff == xoff) && (nxlen < xlen) */ |
| 2322 | /* update old XAD with nXAD:recorded */ |
| 2323 | xad = &p->xad[index]; |
| 2324 | *xad = *nxad; |
| 2325 | xad->flag = xflag & ~XAD_NOTRECORDED; |
| 2326 | |
| 2327 | /* insert rXAD:not_recorded */ |
| 2328 | xoff = xoff + nxlen; |
| 2329 | xlen = xlen - nxlen; |
| 2330 | xaddr = xaddr + nxlen; |
| 2331 | if (nextindex == le16_to_cpu(p->header.maxentry)) { |
| 2332 | /* |
| 2333 | printf("xtUpdate.updateLeft.split p:0x%p\n", p); |
| 2334 | */ |
| 2335 | /* xtSpliUp() unpins leaf pages */ |
| 2336 | split.mp = mp; |
| 2337 | split.index = newindex; |
| 2338 | split.flag = xflag; |
| 2339 | split.off = xoff; |
| 2340 | split.len = xlen; |
| 2341 | split.addr = xaddr; |
| 2342 | split.pxdlist = NULL; |
| 2343 | if ((rc = xtSplitUp(tid, ip, &split, &btstack))) |
| 2344 | return rc; |
| 2345 | |
| 2346 | /* get back old page */ |
| 2347 | XT_GETPAGE(ip, bn, mp, PSIZE, p, rc); |
| 2348 | if (rc) |
| 2349 | return rc; |
| 2350 | |
| 2351 | /* |
| 2352 | * if leaf root has been split, original root has been |
| 2353 | * copied to new child page, i.e., original entry now |
| 2354 | * resides on the new child page; |
| 2355 | */ |
| 2356 | if (p->header.flag & BT_INTERNAL) { |
| 2357 | ASSERT(p->header.nextindex == |
| 2358 | cpu_to_le16(XTENTRYSTART + 1)); |
| 2359 | xad = &p->xad[XTENTRYSTART]; |
| 2360 | bn = addressXAD(xad); |
| 2361 | XT_PUTPAGE(mp); |
| 2362 | |
| 2363 | /* get new child page */ |
| 2364 | XT_GETPAGE(ip, bn, mp, PSIZE, p, rc); |
| 2365 | if (rc) |
| 2366 | return rc; |
| 2367 | |
| 2368 | BT_MARK_DIRTY(mp, ip); |
| 2369 | if (!test_cflag(COMMIT_Nolink, ip)) { |
| 2370 | tlck = txLock(tid, ip, mp, tlckXTREE|tlckGROW); |
| 2371 | xtlck = (struct xtlock *) & tlck->lock; |
| 2372 | } |
| 2373 | } |
| 2374 | } else { |
| 2375 | /* if insert into middle, shift right remaining entries */ |
| 2376 | if (newindex < nextindex) |
| 2377 | memmove(&p->xad[newindex + 1], &p->xad[newindex], |
| 2378 | (nextindex - newindex) << L2XTSLOTSIZE); |
| 2379 | |
| 2380 | /* insert the entry */ |
| 2381 | xad = &p->xad[newindex]; |
| 2382 | XT_PUTENTRY(xad, xflag, xoff, xlen, xaddr); |
| 2383 | |
| 2384 | /* advance next available entry index. */ |
| 2385 | p->header.nextindex = |
| 2386 | cpu_to_le16(le16_to_cpu(p->header.nextindex) + 1); |
| 2387 | } |
| 2388 | |
| 2389 | out: |
| 2390 | if (!test_cflag(COMMIT_Nolink, ip)) { |
| 2391 | xtlck->lwm.offset = (xtlck->lwm.offset) ? |
| 2392 | min(index0, (int)xtlck->lwm.offset) : index0; |
| 2393 | xtlck->lwm.length = le16_to_cpu(p->header.nextindex) - |
| 2394 | xtlck->lwm.offset; |
| 2395 | } |
| 2396 | |
| 2397 | /* unpin the leaf page */ |
| 2398 | XT_PUTPAGE(mp); |
| 2399 | |
| 2400 | return rc; |
| 2401 | } |
| 2402 | |
| 2403 | |
| 2404 | /* |
| 2405 | * xtAppend() |
| 2406 | * |
| 2407 | * function: grow in append mode from contiguous region specified ; |
| 2408 | * |
| 2409 | * parameter: |
| 2410 | * tid - transaction id; |
| 2411 | * ip - file object; |
| 2412 | * xflag - extent flag: |
| 2413 | * xoff - extent offset; |
| 2414 | * maxblocks - max extent length; |
| 2415 | * xlen - extent length (in/out); |
| 2416 | * xaddrp - extent address pointer (in/out): |
| 2417 | * flag - |
| 2418 | * |
| 2419 | * return: |
| 2420 | */ |
| 2421 | int xtAppend(tid_t tid, /* transaction id */ |
| 2422 | struct inode *ip, int xflag, s64 xoff, s32 maxblocks, |
| 2423 | s32 * xlenp, /* (in/out) */ |
| 2424 | s64 * xaddrp, /* (in/out) */ |
| 2425 | int flag) |
| 2426 | { |
| 2427 | int rc = 0; |
| 2428 | struct metapage *mp; /* meta-page buffer */ |
| 2429 | xtpage_t *p; /* base B+-tree index page */ |
| 2430 | s64 bn, xaddr; |
| 2431 | int index, nextindex; |
| 2432 | struct btstack btstack; /* traverse stack */ |
| 2433 | struct xtsplit split; /* split information */ |
| 2434 | xad_t *xad; |
| 2435 | int cmp; |
| 2436 | struct tlock *tlck; |
| 2437 | struct xtlock *xtlck; |
| 2438 | int nsplit, nblocks, xlen; |
| 2439 | struct pxdlist pxdlist; |
| 2440 | pxd_t *pxd; |
| 2441 | |
| 2442 | xaddr = *xaddrp; |
| 2443 | xlen = *xlenp; |
| 2444 | jfs_info("xtAppend: xoff:0x%lx maxblocks:%d xlen:%d xaddr:0x%lx", |
| 2445 | (ulong) xoff, maxblocks, xlen, (ulong) xaddr); |
| 2446 | |
| 2447 | /* |
| 2448 | * search for the entry location at which to insert: |
| 2449 | * |
| 2450 | * xtFastSearch() and xtSearch() both returns (leaf page |
| 2451 | * pinned, index at which to insert). |
| 2452 | * n.b. xtSearch() may return index of maxentry of |
| 2453 | * the full page. |
| 2454 | */ |
| 2455 | if ((rc = xtSearch(ip, xoff, &cmp, &btstack, XT_INSERT))) |
| 2456 | return rc; |
| 2457 | |
| 2458 | /* retrieve search result */ |
| 2459 | XT_GETSEARCH(ip, btstack.top, bn, mp, p, index); |
| 2460 | |
| 2461 | if (cmp == 0) { |
| 2462 | rc = -EEXIST; |
| 2463 | goto out; |
| 2464 | } |
| 2465 | //insert: |
| 2466 | /* |
| 2467 | * insert entry for new extent |
| 2468 | */ |
| 2469 | xflag |= XAD_NEW; |
| 2470 | |
| 2471 | /* |
| 2472 | * if the leaf page is full, split the page and |
| 2473 | * propagate up the router entry for the new page from split |
| 2474 | * |
| 2475 | * The xtSplitUp() will insert the entry and unpin the leaf page. |
| 2476 | */ |
| 2477 | nextindex = le16_to_cpu(p->header.nextindex); |
| 2478 | if (nextindex < le16_to_cpu(p->header.maxentry)) |
| 2479 | goto insertLeaf; |
| 2480 | |
| 2481 | /* |
| 2482 | * allocate new index blocks to cover index page split(s) |
| 2483 | */ |
| 2484 | nsplit = btstack.nsplit; |
| 2485 | split.pxdlist = &pxdlist; |
| 2486 | pxdlist.maxnpxd = pxdlist.npxd = 0; |
| 2487 | pxd = &pxdlist.pxd[0]; |
| 2488 | nblocks = JFS_SBI(ip->i_sb)->nbperpage; |
| 2489 | for (; nsplit > 0; nsplit--, pxd++, xaddr += nblocks, maxblocks -= nblocks) { |
| 2490 | if ((rc = dbAllocBottomUp(ip, xaddr, (s64) nblocks)) == 0) { |
| 2491 | PXDaddress(pxd, xaddr); |
| 2492 | PXDlength(pxd, nblocks); |
| 2493 | |
| 2494 | pxdlist.maxnpxd++; |
| 2495 | |
| 2496 | continue; |
| 2497 | } |
| 2498 | |
| 2499 | /* undo allocation */ |
| 2500 | |
| 2501 | goto out; |
| 2502 | } |
| 2503 | |
| 2504 | xlen = min(xlen, maxblocks); |
| 2505 | |
| 2506 | /* |
| 2507 | * allocate data extent requested |
| 2508 | */ |
| 2509 | if ((rc = dbAllocBottomUp(ip, xaddr, (s64) xlen))) |
| 2510 | goto out; |
| 2511 | |
| 2512 | split.mp = mp; |
| 2513 | split.index = index; |
| 2514 | split.flag = xflag; |
| 2515 | split.off = xoff; |
| 2516 | split.len = xlen; |
| 2517 | split.addr = xaddr; |
| 2518 | if ((rc = xtSplitUp(tid, ip, &split, &btstack))) { |
| 2519 | /* undo data extent allocation */ |
| 2520 | dbFree(ip, *xaddrp, (s64) * xlenp); |
| 2521 | |
| 2522 | return rc; |
| 2523 | } |
| 2524 | |
| 2525 | *xaddrp = xaddr; |
| 2526 | *xlenp = xlen; |
| 2527 | return 0; |
| 2528 | |
| 2529 | /* |
| 2530 | * insert the new entry into the leaf page |
| 2531 | */ |
| 2532 | insertLeaf: |
| 2533 | /* |
| 2534 | * allocate data extent requested |
| 2535 | */ |
| 2536 | if ((rc = dbAllocBottomUp(ip, xaddr, (s64) xlen))) |
| 2537 | goto out; |
| 2538 | |
| 2539 | BT_MARK_DIRTY(mp, ip); |
| 2540 | /* |
| 2541 | * acquire a transaction lock on the leaf page; |
| 2542 | * |
| 2543 | * action: xad insertion/extension; |
| 2544 | */ |
| 2545 | tlck = txLock(tid, ip, mp, tlckXTREE | tlckGROW); |
| 2546 | xtlck = (struct xtlock *) & tlck->lock; |
| 2547 | |
| 2548 | /* insert the new entry: mark the entry NEW */ |
| 2549 | xad = &p->xad[index]; |
| 2550 | XT_PUTENTRY(xad, xflag, xoff, xlen, xaddr); |
| 2551 | |
| 2552 | /* advance next available entry index */ |
| 2553 | p->header.nextindex = |
| 2554 | cpu_to_le16(le16_to_cpu(p->header.nextindex) + 1); |
| 2555 | |
| 2556 | xtlck->lwm.offset = |
| 2557 | (xtlck->lwm.offset) ? min(index,(int) xtlck->lwm.offset) : index; |
| 2558 | xtlck->lwm.length = le16_to_cpu(p->header.nextindex) - |
| 2559 | xtlck->lwm.offset; |
| 2560 | |
| 2561 | *xaddrp = xaddr; |
| 2562 | *xlenp = xlen; |
| 2563 | |
| 2564 | out: |
| 2565 | /* unpin the leaf page */ |
| 2566 | XT_PUTPAGE(mp); |
| 2567 | |
| 2568 | return rc; |
| 2569 | } |
| 2570 | #ifdef _STILL_TO_PORT |
| 2571 | |
| 2572 | /* - TBD for defragmentaion/reorganization - |
| 2573 | * |
| 2574 | * xtDelete() |
| 2575 | * |
| 2576 | * function: |
| 2577 | * delete the entry with the specified key. |
| 2578 | * |
| 2579 | * N.B.: whole extent of the entry is assumed to be deleted. |
| 2580 | * |
| 2581 | * parameter: |
| 2582 | * |
| 2583 | * return: |
| 2584 | * ENOENT: if the entry is not found. |
| 2585 | * |
| 2586 | * exception: |
| 2587 | */ |
| 2588 | int xtDelete(tid_t tid, struct inode *ip, s64 xoff, s32 xlen, int flag) |
| 2589 | { |
| 2590 | int rc = 0; |
| 2591 | struct btstack btstack; |
| 2592 | int cmp; |
| 2593 | s64 bn; |
| 2594 | struct metapage *mp; |
| 2595 | xtpage_t *p; |
| 2596 | int index, nextindex; |
| 2597 | struct tlock *tlck; |
| 2598 | struct xtlock *xtlck; |
| 2599 | |
| 2600 | /* |
| 2601 | * find the matching entry; xtSearch() pins the page |
| 2602 | */ |
| 2603 | if ((rc = xtSearch(ip, xoff, &cmp, &btstack, 0))) |
| 2604 | return rc; |
| 2605 | |
| 2606 | XT_GETSEARCH(ip, btstack.top, bn, mp, p, index); |
| 2607 | if (cmp) { |
| 2608 | /* unpin the leaf page */ |
| 2609 | XT_PUTPAGE(mp); |
| 2610 | return -ENOENT; |
| 2611 | } |
| 2612 | |
| 2613 | /* |
| 2614 | * delete the entry from the leaf page |
| 2615 | */ |
| 2616 | nextindex = le16_to_cpu(p->header.nextindex); |
| 2617 | p->header.nextindex = |
| 2618 | cpu_to_le16(le16_to_cpu(p->header.nextindex) - 1); |
| 2619 | |
| 2620 | /* |
| 2621 | * if the leaf page bocome empty, free the page |
| 2622 | */ |
| 2623 | if (p->header.nextindex == cpu_to_le16(XTENTRYSTART)) |
| 2624 | return (xtDeleteUp(tid, ip, mp, p, &btstack)); |
| 2625 | |
| 2626 | BT_MARK_DIRTY(mp, ip); |
| 2627 | /* |
| 2628 | * acquire a transaction lock on the leaf page; |
| 2629 | * |
| 2630 | * action:xad deletion; |
| 2631 | */ |
| 2632 | tlck = txLock(tid, ip, mp, tlckXTREE); |
| 2633 | xtlck = (struct xtlock *) & tlck->lock; |
| 2634 | xtlck->lwm.offset = |
| 2635 | (xtlck->lwm.offset) ? min(index, xtlck->lwm.offset) : index; |
| 2636 | |
| 2637 | /* if delete from middle, shift left/compact the remaining entries */ |
| 2638 | if (index < nextindex - 1) |
| 2639 | memmove(&p->xad[index], &p->xad[index + 1], |
| 2640 | (nextindex - index - 1) * sizeof(xad_t)); |
| 2641 | |
| 2642 | XT_PUTPAGE(mp); |
| 2643 | |
| 2644 | return 0; |
| 2645 | } |
| 2646 | |
| 2647 | |
| 2648 | /* - TBD for defragmentaion/reorganization - |
| 2649 | * |
| 2650 | * xtDeleteUp() |
| 2651 | * |
| 2652 | * function: |
| 2653 | * free empty pages as propagating deletion up the tree |
| 2654 | * |
| 2655 | * parameter: |
| 2656 | * |
| 2657 | * return: |
| 2658 | */ |
| 2659 | static int |
| 2660 | xtDeleteUp(tid_t tid, struct inode *ip, |
| 2661 | struct metapage * fmp, xtpage_t * fp, struct btstack * btstack) |
| 2662 | { |
| 2663 | int rc = 0; |
| 2664 | struct metapage *mp; |
| 2665 | xtpage_t *p; |
| 2666 | int index, nextindex; |
| 2667 | s64 xaddr; |
| 2668 | int xlen; |
| 2669 | struct btframe *parent; |
| 2670 | struct tlock *tlck; |
| 2671 | struct xtlock *xtlck; |
| 2672 | |
| 2673 | /* |
| 2674 | * keep root leaf page which has become empty |
| 2675 | */ |
| 2676 | if (fp->header.flag & BT_ROOT) { |
| 2677 | /* keep the root page */ |
| 2678 | fp->header.flag &= ~BT_INTERNAL; |
| 2679 | fp->header.flag |= BT_LEAF; |
| 2680 | fp->header.nextindex = cpu_to_le16(XTENTRYSTART); |
| 2681 | |
| 2682 | /* XT_PUTPAGE(fmp); */ |
| 2683 | |
| 2684 | return 0; |
| 2685 | } |
| 2686 | |
| 2687 | /* |
| 2688 | * free non-root leaf page |
| 2689 | */ |
| 2690 | if ((rc = xtRelink(tid, ip, fp))) { |
| 2691 | XT_PUTPAGE(fmp); |
| 2692 | return rc; |
| 2693 | } |
| 2694 | |
| 2695 | xaddr = addressPXD(&fp->header.self); |
| 2696 | xlen = lengthPXD(&fp->header.self); |
| 2697 | /* free the page extent */ |
| 2698 | dbFree(ip, xaddr, (s64) xlen); |
| 2699 | |
| 2700 | /* free the buffer page */ |
| 2701 | discard_metapage(fmp); |
| 2702 | |
| 2703 | /* |
| 2704 | * propagate page deletion up the index tree |
| 2705 | * |
| 2706 | * If the delete from the parent page makes it empty, |
| 2707 | * continue all the way up the tree. |
| 2708 | * stop if the root page is reached (which is never deleted) or |
| 2709 | * if the entry deletion does not empty the page. |
| 2710 | */ |
| 2711 | while ((parent = BT_POP(btstack)) != NULL) { |
| 2712 | /* get/pin the parent page <sp> */ |
| 2713 | XT_GETPAGE(ip, parent->bn, mp, PSIZE, p, rc); |
| 2714 | if (rc) |
| 2715 | return rc; |
| 2716 | |
| 2717 | index = parent->index; |
| 2718 | |
| 2719 | /* delete the entry for the freed child page from parent. |
| 2720 | */ |
| 2721 | nextindex = le16_to_cpu(p->header.nextindex); |
| 2722 | |
| 2723 | /* |
| 2724 | * the parent has the single entry being deleted: |
| 2725 | * free the parent page which has become empty. |
| 2726 | */ |
| 2727 | if (nextindex == 1) { |
| 2728 | if (p->header.flag & BT_ROOT) { |
| 2729 | /* keep the root page */ |
| 2730 | p->header.flag &= ~BT_INTERNAL; |
| 2731 | p->header.flag |= BT_LEAF; |
| 2732 | p->header.nextindex = |
| 2733 | cpu_to_le16(XTENTRYSTART); |
| 2734 | |
| 2735 | /* XT_PUTPAGE(mp); */ |
| 2736 | |
| 2737 | break; |
| 2738 | } else { |
| 2739 | /* free the parent page */ |
| 2740 | if ((rc = xtRelink(tid, ip, p))) |
| 2741 | return rc; |
| 2742 | |
| 2743 | xaddr = addressPXD(&p->header.self); |
| 2744 | /* free the page extent */ |
| 2745 | dbFree(ip, xaddr, |
| 2746 | (s64) JFS_SBI(ip->i_sb)->nbperpage); |
| 2747 | |
| 2748 | /* unpin/free the buffer page */ |
| 2749 | discard_metapage(mp); |
| 2750 | |
| 2751 | /* propagate up */ |
| 2752 | continue; |
| 2753 | } |
| 2754 | } |
| 2755 | /* |
| 2756 | * the parent has other entries remaining: |
| 2757 | * delete the router entry from the parent page. |
| 2758 | */ |
| 2759 | else { |
| 2760 | BT_MARK_DIRTY(mp, ip); |
| 2761 | /* |
| 2762 | * acquire a transaction lock on the leaf page; |
| 2763 | * |
| 2764 | * action:xad deletion; |
| 2765 | */ |
| 2766 | tlck = txLock(tid, ip, mp, tlckXTREE); |
| 2767 | xtlck = (struct xtlock *) & tlck->lock; |
| 2768 | xtlck->lwm.offset = |
| 2769 | (xtlck->lwm.offset) ? min(index, |
| 2770 | xtlck->lwm. |
| 2771 | offset) : index; |
| 2772 | |
| 2773 | /* if delete from middle, |
| 2774 | * shift left/compact the remaining entries in the page |
| 2775 | */ |
| 2776 | if (index < nextindex - 1) |
| 2777 | memmove(&p->xad[index], &p->xad[index + 1], |
| 2778 | (nextindex - index - |
| 2779 | 1) << L2XTSLOTSIZE); |
| 2780 | |
| 2781 | p->header.nextindex = |
| 2782 | cpu_to_le16(le16_to_cpu(p->header.nextindex) - |
| 2783 | 1); |
| 2784 | jfs_info("xtDeleteUp(entry): 0x%lx[%d]", |
| 2785 | (ulong) parent->bn, index); |
| 2786 | } |
| 2787 | |
| 2788 | /* unpin the parent page */ |
| 2789 | XT_PUTPAGE(mp); |
| 2790 | |
| 2791 | /* exit propagation up */ |
| 2792 | break; |
| 2793 | } |
| 2794 | |
| 2795 | return 0; |
| 2796 | } |
| 2797 | |
| 2798 | |
| 2799 | /* |
| 2800 | * NAME: xtRelocate() |
| 2801 | * |
| 2802 | * FUNCTION: relocate xtpage or data extent of regular file; |
| 2803 | * This function is mainly used by defragfs utility. |
| 2804 | * |
| 2805 | * NOTE: This routine does not have the logic to handle |
| 2806 | * uncommitted allocated extent. The caller should call |
| 2807 | * txCommit() to commit all the allocation before call |
| 2808 | * this routine. |
| 2809 | */ |
| 2810 | int |
| 2811 | xtRelocate(tid_t tid, struct inode * ip, xad_t * oxad, /* old XAD */ |
| 2812 | s64 nxaddr, /* new xaddr */ |
| 2813 | int xtype) |
| 2814 | { /* extent type: XTPAGE or DATAEXT */ |
| 2815 | int rc = 0; |
| 2816 | struct tblock *tblk; |
| 2817 | struct tlock *tlck; |
| 2818 | struct xtlock *xtlck; |
| 2819 | struct metapage *mp, *pmp, *lmp, *rmp; /* meta-page buffer */ |
| 2820 | xtpage_t *p, *pp, *rp, *lp; /* base B+-tree index page */ |
| 2821 | xad_t *xad; |
| 2822 | pxd_t *pxd; |
| 2823 | s64 xoff, xsize; |
| 2824 | int xlen; |
| 2825 | s64 oxaddr, sxaddr, dxaddr, nextbn, prevbn; |
| 2826 | cbuf_t *cp; |
| 2827 | s64 offset, nbytes, nbrd, pno; |
| 2828 | int nb, npages, nblks; |
| 2829 | s64 bn; |
| 2830 | int cmp; |
| 2831 | int index; |
| 2832 | struct pxd_lock *pxdlock; |
| 2833 | struct btstack btstack; /* traverse stack */ |
| 2834 | |
| 2835 | xtype = xtype & EXTENT_TYPE; |
| 2836 | |
| 2837 | xoff = offsetXAD(oxad); |
| 2838 | oxaddr = addressXAD(oxad); |
| 2839 | xlen = lengthXAD(oxad); |
| 2840 | |
| 2841 | /* validate extent offset */ |
| 2842 | offset = xoff << JFS_SBI(ip->i_sb)->l2bsize; |
| 2843 | if (offset >= ip->i_size) |
| 2844 | return -ESTALE; /* stale extent */ |
| 2845 | |
| 2846 | jfs_info("xtRelocate: xtype:%d xoff:0x%lx xlen:0x%x xaddr:0x%lx:0x%lx", |
| 2847 | xtype, (ulong) xoff, xlen, (ulong) oxaddr, (ulong) nxaddr); |
| 2848 | |
| 2849 | /* |
| 2850 | * 1. get and validate the parent xtpage/xad entry |
| 2851 | * covering the source extent to be relocated; |
| 2852 | */ |
| 2853 | if (xtype == DATAEXT) { |
| 2854 | /* search in leaf entry */ |
| 2855 | rc = xtSearch(ip, xoff, &cmp, &btstack, 0); |
| 2856 | if (rc) |
| 2857 | return rc; |
| 2858 | |
| 2859 | /* retrieve search result */ |
| 2860 | XT_GETSEARCH(ip, btstack.top, bn, pmp, pp, index); |
| 2861 | |
| 2862 | if (cmp) { |
| 2863 | XT_PUTPAGE(pmp); |
| 2864 | return -ESTALE; |
| 2865 | } |
| 2866 | |
| 2867 | /* validate for exact match with a single entry */ |
| 2868 | xad = &pp->xad[index]; |
| 2869 | if (addressXAD(xad) != oxaddr || lengthXAD(xad) != xlen) { |
| 2870 | XT_PUTPAGE(pmp); |
| 2871 | return -ESTALE; |
| 2872 | } |
| 2873 | } else { /* (xtype == XTPAGE) */ |
| 2874 | |
| 2875 | /* search in internal entry */ |
| 2876 | rc = xtSearchNode(ip, oxad, &cmp, &btstack, 0); |
| 2877 | if (rc) |
| 2878 | return rc; |
| 2879 | |
| 2880 | /* retrieve search result */ |
| 2881 | XT_GETSEARCH(ip, btstack.top, bn, pmp, pp, index); |
| 2882 | |
| 2883 | if (cmp) { |
| 2884 | XT_PUTPAGE(pmp); |
| 2885 | return -ESTALE; |
| 2886 | } |
| 2887 | |
| 2888 | /* xtSearchNode() validated for exact match with a single entry |
| 2889 | */ |
| 2890 | xad = &pp->xad[index]; |
| 2891 | } |
| 2892 | jfs_info("xtRelocate: parent xad entry validated."); |
| 2893 | |
| 2894 | /* |
| 2895 | * 2. relocate the extent |
| 2896 | */ |
| 2897 | if (xtype == DATAEXT) { |
| 2898 | /* if the extent is allocated-but-not-recorded |
| 2899 | * there is no real data to be moved in this extent, |
| 2900 | */ |
| 2901 | if (xad->flag & XAD_NOTRECORDED) |
| 2902 | goto out; |
| 2903 | else |
| 2904 | /* release xtpage for cmRead()/xtLookup() */ |
| 2905 | XT_PUTPAGE(pmp); |
| 2906 | |
| 2907 | /* |
| 2908 | * cmRelocate() |
| 2909 | * |
| 2910 | * copy target data pages to be relocated; |
| 2911 | * |
| 2912 | * data extent must start at page boundary and |
| 2913 | * multiple of page size (except the last data extent); |
| 2914 | * read in each page of the source data extent into cbuf, |
| 2915 | * update the cbuf extent descriptor of the page to be |
| 2916 | * homeward bound to new dst data extent |
| 2917 | * copy the data from the old extent to new extent. |
| 2918 | * copy is essential for compressed files to avoid problems |
| 2919 | * that can arise if there was a change in compression |
| 2920 | * algorithms. |
| 2921 | * it is a good strategy because it may disrupt cache |
| 2922 | * policy to keep the pages in memory afterwards. |
| 2923 | */ |
| 2924 | offset = xoff << JFS_SBI(ip->i_sb)->l2bsize; |
| 2925 | assert((offset & CM_OFFSET) == 0); |
| 2926 | nbytes = xlen << JFS_SBI(ip->i_sb)->l2bsize; |
| 2927 | pno = offset >> CM_L2BSIZE; |
| 2928 | npages = (nbytes + (CM_BSIZE - 1)) >> CM_L2BSIZE; |
| 2929 | /* |
| 2930 | npages = ((offset + nbytes - 1) >> CM_L2BSIZE) - |
| 2931 | (offset >> CM_L2BSIZE) + 1; |
| 2932 | */ |
| 2933 | sxaddr = oxaddr; |
| 2934 | dxaddr = nxaddr; |
| 2935 | |
| 2936 | /* process the request one cache buffer at a time */ |
| 2937 | for (nbrd = 0; nbrd < nbytes; nbrd += nb, |
| 2938 | offset += nb, pno++, npages--) { |
| 2939 | /* compute page size */ |
| 2940 | nb = min(nbytes - nbrd, CM_BSIZE); |
| 2941 | |
| 2942 | /* get the cache buffer of the page */ |
| 2943 | if (rc = cmRead(ip, offset, npages, &cp)) |
| 2944 | break; |
| 2945 | |
| 2946 | assert(addressPXD(&cp->cm_pxd) == sxaddr); |
| 2947 | assert(!cp->cm_modified); |
| 2948 | |
| 2949 | /* bind buffer with the new extent address */ |
| 2950 | nblks = nb >> JFS_IP(ip->i_sb)->l2bsize; |
| 2951 | cmSetXD(ip, cp, pno, dxaddr, nblks); |
| 2952 | |
| 2953 | /* release the cbuf, mark it as modified */ |
| 2954 | cmPut(cp, TRUE); |
| 2955 | |
| 2956 | dxaddr += nblks; |
| 2957 | sxaddr += nblks; |
| 2958 | } |
| 2959 | |
| 2960 | /* get back parent page */ |
| 2961 | if ((rc = xtSearch(ip, xoff, &cmp, &btstack, 0))) |
| 2962 | return rc; |
| 2963 | |
| 2964 | XT_GETSEARCH(ip, btstack.top, bn, pmp, pp, index); |
| 2965 | jfs_info("xtRelocate: target data extent relocated."); |
| 2966 | } else { /* (xtype == XTPAGE) */ |
| 2967 | |
| 2968 | /* |
| 2969 | * read in the target xtpage from the source extent; |
| 2970 | */ |
| 2971 | XT_GETPAGE(ip, oxaddr, mp, PSIZE, p, rc); |
| 2972 | if (rc) { |
| 2973 | XT_PUTPAGE(pmp); |
| 2974 | return rc; |
| 2975 | } |
| 2976 | |
| 2977 | /* |
| 2978 | * read in sibling pages if any to update sibling pointers; |
| 2979 | */ |
| 2980 | rmp = NULL; |
| 2981 | if (p->header.next) { |
| 2982 | nextbn = le64_to_cpu(p->header.next); |
| 2983 | XT_GETPAGE(ip, nextbn, rmp, PSIZE, rp, rc); |
| 2984 | if (rc) { |
| 2985 | XT_PUTPAGE(pmp); |
| 2986 | XT_PUTPAGE(mp); |
| 2987 | return (rc); |
| 2988 | } |
| 2989 | } |
| 2990 | |
| 2991 | lmp = NULL; |
| 2992 | if (p->header.prev) { |
| 2993 | prevbn = le64_to_cpu(p->header.prev); |
| 2994 | XT_GETPAGE(ip, prevbn, lmp, PSIZE, lp, rc); |
| 2995 | if (rc) { |
| 2996 | XT_PUTPAGE(pmp); |
| 2997 | XT_PUTPAGE(mp); |
| 2998 | if (rmp) |
| 2999 | XT_PUTPAGE(rmp); |
| 3000 | return (rc); |
| 3001 | } |
| 3002 | } |
| 3003 | |
| 3004 | /* at this point, all xtpages to be updated are in memory */ |
| 3005 | |
| 3006 | /* |
| 3007 | * update sibling pointers of sibling xtpages if any; |
| 3008 | */ |
| 3009 | if (lmp) { |
| 3010 | BT_MARK_DIRTY(lmp, ip); |
| 3011 | tlck = |
| 3012 | txLock(tid, ip, lmp, tlckXTREE | tlckRELINK); |
| 3013 | lp->header.next = cpu_to_le64(nxaddr); |
| 3014 | XT_PUTPAGE(lmp); |
| 3015 | } |
| 3016 | |
| 3017 | if (rmp) { |
| 3018 | BT_MARK_DIRTY(rmp, ip); |
| 3019 | tlck = |
| 3020 | txLock(tid, ip, rmp, tlckXTREE | tlckRELINK); |
| 3021 | rp->header.prev = cpu_to_le64(nxaddr); |
| 3022 | XT_PUTPAGE(rmp); |
| 3023 | } |
| 3024 | |
| 3025 | /* |
| 3026 | * update the target xtpage to be relocated |
| 3027 | * |
| 3028 | * update the self address of the target page |
| 3029 | * and write to destination extent; |
| 3030 | * redo image covers the whole xtpage since it is new page |
| 3031 | * to the destination extent; |
| 3032 | * update of bmap for the free of source extent |
| 3033 | * of the target xtpage itself: |
| 3034 | * update of bmap for the allocation of destination extent |
| 3035 | * of the target xtpage itself: |
| 3036 | * update of bmap for the extents covered by xad entries in |
| 3037 | * the target xtpage is not necessary since they are not |
| 3038 | * updated; |
| 3039 | * if not committed before this relocation, |
| 3040 | * target page may contain XAD_NEW entries which must |
| 3041 | * be scanned for bmap update (logredo() always |
| 3042 | * scan xtpage REDOPAGE image for bmap update); |
| 3043 | * if committed before this relocation (tlckRELOCATE), |
| 3044 | * scan may be skipped by commit() and logredo(); |
| 3045 | */ |
| 3046 | BT_MARK_DIRTY(mp, ip); |
| 3047 | /* tlckNEW init xtlck->lwm.offset = XTENTRYSTART; */ |
| 3048 | tlck = txLock(tid, ip, mp, tlckXTREE | tlckNEW); |
| 3049 | xtlck = (struct xtlock *) & tlck->lock; |
| 3050 | |
| 3051 | /* update the self address in the xtpage header */ |
| 3052 | pxd = &p->header.self; |
| 3053 | PXDaddress(pxd, nxaddr); |
| 3054 | |
| 3055 | /* linelock for the after image of the whole page */ |
| 3056 | xtlck->lwm.length = |
| 3057 | le16_to_cpu(p->header.nextindex) - xtlck->lwm.offset; |
| 3058 | |
| 3059 | /* update the buffer extent descriptor of target xtpage */ |
| 3060 | xsize = xlen << JFS_SBI(ip->i_sb)->l2bsize; |
| 3061 | bmSetXD(mp, nxaddr, xsize); |
| 3062 | |
| 3063 | /* unpin the target page to new homeward bound */ |
| 3064 | XT_PUTPAGE(mp); |
| 3065 | jfs_info("xtRelocate: target xtpage relocated."); |
| 3066 | } |
| 3067 | |
| 3068 | /* |
| 3069 | * 3. acquire maplock for the source extent to be freed; |
| 3070 | * |
| 3071 | * acquire a maplock saving the src relocated extent address; |
| 3072 | * to free of the extent at commit time; |
| 3073 | */ |
| 3074 | out: |
| 3075 | /* if DATAEXT relocation, write a LOG_UPDATEMAP record for |
| 3076 | * free PXD of the source data extent (logredo() will update |
| 3077 | * bmap for free of source data extent), and update bmap for |
| 3078 | * free of the source data extent; |
| 3079 | */ |
| 3080 | if (xtype == DATAEXT) |
| 3081 | tlck = txMaplock(tid, ip, tlckMAP); |
| 3082 | /* if XTPAGE relocation, write a LOG_NOREDOPAGE record |
| 3083 | * for the source xtpage (logredo() will init NoRedoPage |
| 3084 | * filter and will also update bmap for free of the source |
| 3085 | * xtpage), and update bmap for free of the source xtpage; |
| 3086 | * N.B. We use tlckMAP instead of tlkcXTREE because there |
| 3087 | * is no buffer associated with this lock since the buffer |
| 3088 | * has been redirected to the target location. |
| 3089 | */ |
| 3090 | else /* (xtype == XTPAGE) */ |
| 3091 | tlck = txMaplock(tid, ip, tlckMAP | tlckRELOCATE); |
| 3092 | |
| 3093 | pxdlock = (struct pxd_lock *) & tlck->lock; |
| 3094 | pxdlock->flag = mlckFREEPXD; |
| 3095 | PXDaddress(&pxdlock->pxd, oxaddr); |
| 3096 | PXDlength(&pxdlock->pxd, xlen); |
| 3097 | pxdlock->index = 1; |
| 3098 | |
| 3099 | /* |
| 3100 | * 4. update the parent xad entry for relocation; |
| 3101 | * |
| 3102 | * acquire tlck for the parent entry with XAD_NEW as entry |
| 3103 | * update which will write LOG_REDOPAGE and update bmap for |
| 3104 | * allocation of XAD_NEW destination extent; |
| 3105 | */ |
| 3106 | jfs_info("xtRelocate: update parent xad entry."); |
| 3107 | BT_MARK_DIRTY(pmp, ip); |
| 3108 | tlck = txLock(tid, ip, pmp, tlckXTREE | tlckGROW); |
| 3109 | xtlck = (struct xtlock *) & tlck->lock; |
| 3110 | |
| 3111 | /* update the XAD with the new destination extent; */ |
| 3112 | xad = &pp->xad[index]; |
| 3113 | xad->flag |= XAD_NEW; |
| 3114 | XADaddress(xad, nxaddr); |
| 3115 | |
| 3116 | xtlck->lwm.offset = min(index, xtlck->lwm.offset); |
| 3117 | xtlck->lwm.length = le16_to_cpu(pp->header.nextindex) - |
| 3118 | xtlck->lwm.offset; |
| 3119 | |
| 3120 | /* unpin the parent xtpage */ |
| 3121 | XT_PUTPAGE(pmp); |
| 3122 | |
| 3123 | return rc; |
| 3124 | } |
| 3125 | |
| 3126 | |
| 3127 | /* |
| 3128 | * xtSearchNode() |
| 3129 | * |
| 3130 | * function: search for the internal xad entry covering specified extent. |
| 3131 | * This function is mainly used by defragfs utility. |
| 3132 | * |
| 3133 | * parameters: |
| 3134 | * ip - file object; |
| 3135 | * xad - extent to find; |
| 3136 | * cmpp - comparison result: |
| 3137 | * btstack - traverse stack; |
| 3138 | * flag - search process flag; |
| 3139 | * |
| 3140 | * returns: |
| 3141 | * btstack contains (bn, index) of search path traversed to the entry. |
| 3142 | * *cmpp is set to result of comparison with the entry returned. |
| 3143 | * the page containing the entry is pinned at exit. |
| 3144 | */ |
| 3145 | static int xtSearchNode(struct inode *ip, xad_t * xad, /* required XAD entry */ |
| 3146 | int *cmpp, struct btstack * btstack, int flag) |
| 3147 | { |
| 3148 | int rc = 0; |
| 3149 | s64 xoff, xaddr; |
| 3150 | int xlen; |
| 3151 | int cmp = 1; /* init for empty page */ |
| 3152 | s64 bn; /* block number */ |
| 3153 | struct metapage *mp; /* meta-page buffer */ |
| 3154 | xtpage_t *p; /* page */ |
| 3155 | int base, index, lim; |
| 3156 | struct btframe *btsp; |
| 3157 | s64 t64; |
| 3158 | |
| 3159 | BT_CLR(btstack); |
| 3160 | |
| 3161 | xoff = offsetXAD(xad); |
| 3162 | xlen = lengthXAD(xad); |
| 3163 | xaddr = addressXAD(xad); |
| 3164 | |
| 3165 | /* |
| 3166 | * search down tree from root: |
| 3167 | * |
| 3168 | * between two consecutive entries of <Ki, Pi> and <Kj, Pj> of |
| 3169 | * internal page, child page Pi contains entry with k, Ki <= K < Kj. |
| 3170 | * |
| 3171 | * if entry with search key K is not found |
| 3172 | * internal page search find the entry with largest key Ki |
| 3173 | * less than K which point to the child page to search; |
| 3174 | * leaf page search find the entry with smallest key Kj |
| 3175 | * greater than K so that the returned index is the position of |
| 3176 | * the entry to be shifted right for insertion of new entry. |
| 3177 | * for empty tree, search key is greater than any key of the tree. |
| 3178 | * |
| 3179 | * by convention, root bn = 0. |
| 3180 | */ |
| 3181 | for (bn = 0;;) { |
| 3182 | /* get/pin the page to search */ |
| 3183 | XT_GETPAGE(ip, bn, mp, PSIZE, p, rc); |
| 3184 | if (rc) |
| 3185 | return rc; |
| 3186 | if (p->header.flag & BT_LEAF) { |
| 3187 | XT_PUTPAGE(mp); |
| 3188 | return -ESTALE; |
| 3189 | } |
| 3190 | |
| 3191 | lim = le16_to_cpu(p->header.nextindex) - XTENTRYSTART; |
| 3192 | |
| 3193 | /* |
| 3194 | * binary search with search key K on the current page |
| 3195 | */ |
| 3196 | for (base = XTENTRYSTART; lim; lim >>= 1) { |
| 3197 | index = base + (lim >> 1); |
| 3198 | |
| 3199 | XT_CMP(cmp, xoff, &p->xad[index], t64); |
| 3200 | if (cmp == 0) { |
| 3201 | /* |
| 3202 | * search hit |
| 3203 | * |
| 3204 | * verify for exact match; |
| 3205 | */ |
| 3206 | if (xaddr == addressXAD(&p->xad[index]) && |
| 3207 | xoff == offsetXAD(&p->xad[index])) { |
| 3208 | *cmpp = cmp; |
| 3209 | |
| 3210 | /* save search result */ |
| 3211 | btsp = btstack->top; |
| 3212 | btsp->bn = bn; |
| 3213 | btsp->index = index; |
| 3214 | btsp->mp = mp; |
| 3215 | |
| 3216 | return 0; |
| 3217 | } |
| 3218 | |
| 3219 | /* descend/search its child page */ |
| 3220 | goto next; |
| 3221 | } |
| 3222 | |
| 3223 | if (cmp > 0) { |
| 3224 | base = index + 1; |
| 3225 | --lim; |
| 3226 | } |
| 3227 | } |
| 3228 | |
| 3229 | /* |
| 3230 | * search miss - non-leaf page: |
| 3231 | * |
| 3232 | * base is the smallest index with key (Kj) greater than |
| 3233 | * search key (K) and may be zero or maxentry index. |
| 3234 | * if base is non-zero, decrement base by one to get the parent |
| 3235 | * entry of the child page to search. |
| 3236 | */ |
| 3237 | index = base ? base - 1 : base; |
| 3238 | |
| 3239 | /* |
| 3240 | * go down to child page |
| 3241 | */ |
| 3242 | next: |
| 3243 | /* get the child page block number */ |
| 3244 | bn = addressXAD(&p->xad[index]); |
| 3245 | |
| 3246 | /* unpin the parent page */ |
| 3247 | XT_PUTPAGE(mp); |
| 3248 | } |
| 3249 | } |
| 3250 | |
| 3251 | |
| 3252 | /* |
| 3253 | * xtRelink() |
| 3254 | * |
| 3255 | * function: |
| 3256 | * link around a freed page. |
| 3257 | * |
| 3258 | * Parameter: |
| 3259 | * int tid, |
| 3260 | * struct inode *ip, |
| 3261 | * xtpage_t *p) |
| 3262 | * |
| 3263 | * returns: |
| 3264 | */ |
| 3265 | static int xtRelink(tid_t tid, struct inode *ip, xtpage_t * p) |
| 3266 | { |
| 3267 | int rc = 0; |
| 3268 | struct metapage *mp; |
| 3269 | s64 nextbn, prevbn; |
| 3270 | struct tlock *tlck; |
| 3271 | |
| 3272 | nextbn = le64_to_cpu(p->header.next); |
| 3273 | prevbn = le64_to_cpu(p->header.prev); |
| 3274 | |
| 3275 | /* update prev pointer of the next page */ |
| 3276 | if (nextbn != 0) { |
| 3277 | XT_GETPAGE(ip, nextbn, mp, PSIZE, p, rc); |
| 3278 | if (rc) |
| 3279 | return rc; |
| 3280 | |
| 3281 | /* |
| 3282 | * acquire a transaction lock on the page; |
| 3283 | * |
| 3284 | * action: update prev pointer; |
| 3285 | */ |
| 3286 | BT_MARK_DIRTY(mp, ip); |
| 3287 | tlck = txLock(tid, ip, mp, tlckXTREE | tlckRELINK); |
| 3288 | |
| 3289 | /* the page may already have been tlock'd */ |
| 3290 | |
| 3291 | p->header.prev = cpu_to_le64(prevbn); |
| 3292 | |
| 3293 | XT_PUTPAGE(mp); |
| 3294 | } |
| 3295 | |
| 3296 | /* update next pointer of the previous page */ |
| 3297 | if (prevbn != 0) { |
| 3298 | XT_GETPAGE(ip, prevbn, mp, PSIZE, p, rc); |
| 3299 | if (rc) |
| 3300 | return rc; |
| 3301 | |
| 3302 | /* |
| 3303 | * acquire a transaction lock on the page; |
| 3304 | * |
| 3305 | * action: update next pointer; |
| 3306 | */ |
| 3307 | BT_MARK_DIRTY(mp, ip); |
| 3308 | tlck = txLock(tid, ip, mp, tlckXTREE | tlckRELINK); |
| 3309 | |
| 3310 | /* the page may already have been tlock'd */ |
| 3311 | |
| 3312 | p->header.next = le64_to_cpu(nextbn); |
| 3313 | |
| 3314 | XT_PUTPAGE(mp); |
| 3315 | } |
| 3316 | |
| 3317 | return 0; |
| 3318 | } |
| 3319 | #endif /* _STILL_TO_PORT */ |
| 3320 | |
| 3321 | |
| 3322 | /* |
| 3323 | * xtInitRoot() |
| 3324 | * |
| 3325 | * initialize file root (inline in inode) |
| 3326 | */ |
| 3327 | void xtInitRoot(tid_t tid, struct inode *ip) |
| 3328 | { |
| 3329 | xtpage_t *p; |
| 3330 | |
| 3331 | /* |
| 3332 | * acquire a transaction lock on the root |
| 3333 | * |
| 3334 | * action: |
| 3335 | */ |
| 3336 | txLock(tid, ip, (struct metapage *) &JFS_IP(ip)->bxflag, |
| 3337 | tlckXTREE | tlckNEW); |
| 3338 | p = &JFS_IP(ip)->i_xtroot; |
| 3339 | |
| 3340 | p->header.flag = DXD_INDEX | BT_ROOT | BT_LEAF; |
| 3341 | p->header.nextindex = cpu_to_le16(XTENTRYSTART); |
| 3342 | |
| 3343 | if (S_ISDIR(ip->i_mode)) |
| 3344 | p->header.maxentry = cpu_to_le16(XTROOTINITSLOT_DIR); |
| 3345 | else { |
| 3346 | p->header.maxentry = cpu_to_le16(XTROOTINITSLOT); |
| 3347 | ip->i_size = 0; |
| 3348 | } |
| 3349 | |
| 3350 | |
| 3351 | return; |
| 3352 | } |
| 3353 | |
| 3354 | |
| 3355 | /* |
| 3356 | * We can run into a deadlock truncating a file with a large number of |
| 3357 | * xtree pages (large fragmented file). A robust fix would entail a |
| 3358 | * reservation system where we would reserve a number of metadata pages |
| 3359 | * and tlocks which we would be guaranteed without a deadlock. Without |
| 3360 | * this, a partial fix is to limit number of metadata pages we will lock |
| 3361 | * in a single transaction. Currently we will truncate the file so that |
| 3362 | * no more than 50 leaf pages will be locked. The caller of xtTruncate |
| 3363 | * will be responsible for ensuring that the current transaction gets |
| 3364 | * committed, and that subsequent transactions are created to truncate |
| 3365 | * the file further if needed. |
| 3366 | */ |
| 3367 | #define MAX_TRUNCATE_LEAVES 50 |
| 3368 | |
| 3369 | /* |
| 3370 | * xtTruncate() |
| 3371 | * |
| 3372 | * function: |
| 3373 | * traverse for truncation logging backward bottom up; |
| 3374 | * terminate at the last extent entry at the current subtree |
| 3375 | * root page covering new down size. |
| 3376 | * truncation may occur within the last extent entry. |
| 3377 | * |
| 3378 | * parameter: |
| 3379 | * int tid, |
| 3380 | * struct inode *ip, |
| 3381 | * s64 newsize, |
| 3382 | * int type) {PWMAP, PMAP, WMAP; DELETE, TRUNCATE} |
| 3383 | * |
| 3384 | * return: |
| 3385 | * |
| 3386 | * note: |
| 3387 | * PWMAP: |
| 3388 | * 1. truncate (non-COMMIT_NOLINK file) |
| 3389 | * by jfs_truncate() or jfs_open(O_TRUNC): |
| 3390 | * xtree is updated; |
| 3391 | * 2. truncate index table of directory when last entry removed |
| 3392 | * map update via tlock at commit time; |
| 3393 | * PMAP: |
| 3394 | * Call xtTruncate_pmap instead |
| 3395 | * WMAP: |
| 3396 | * 1. remove (free zero link count) on last reference release |
| 3397 | * (pmap has been freed at commit zero link count); |
| 3398 | * 2. truncate (COMMIT_NOLINK file, i.e., tmp file): |
| 3399 | * xtree is updated; |
| 3400 | * map update directly at truncation time; |
| 3401 | * |
| 3402 | * if (DELETE) |
| 3403 | * no LOG_NOREDOPAGE is required (NOREDOFILE is sufficient); |
| 3404 | * else if (TRUNCATE) |
| 3405 | * must write LOG_NOREDOPAGE for deleted index page; |
| 3406 | * |
| 3407 | * pages may already have been tlocked by anonymous transactions |
| 3408 | * during file growth (i.e., write) before truncation; |
| 3409 | * |
| 3410 | * except last truncated entry, deleted entries remains as is |
| 3411 | * in the page (nextindex is updated) for other use |
| 3412 | * (e.g., log/update allocation map): this avoid copying the page |
| 3413 | * info but delay free of pages; |
| 3414 | * |
| 3415 | */ |
| 3416 | s64 xtTruncate(tid_t tid, struct inode *ip, s64 newsize, int flag) |
| 3417 | { |
| 3418 | int rc = 0; |
| 3419 | s64 teof; |
| 3420 | struct metapage *mp; |
| 3421 | xtpage_t *p; |
| 3422 | s64 bn; |
| 3423 | int index, nextindex; |
| 3424 | xad_t *xad; |
| 3425 | s64 xoff, xaddr; |
| 3426 | int xlen, len, freexlen; |
| 3427 | struct btstack btstack; |
| 3428 | struct btframe *parent; |
| 3429 | struct tblock *tblk = NULL; |
| 3430 | struct tlock *tlck = NULL; |
| 3431 | struct xtlock *xtlck = NULL; |
| 3432 | struct xdlistlock xadlock; /* maplock for COMMIT_WMAP */ |
| 3433 | struct pxd_lock *pxdlock; /* maplock for COMMIT_WMAP */ |
| 3434 | s64 nfreed; |
| 3435 | int freed, log; |
| 3436 | int locked_leaves = 0; |
| 3437 | |
| 3438 | /* save object truncation type */ |
| 3439 | if (tid) { |
| 3440 | tblk = tid_to_tblock(tid); |
| 3441 | tblk->xflag |= flag; |
| 3442 | } |
| 3443 | |
| 3444 | nfreed = 0; |
| 3445 | |
| 3446 | flag &= COMMIT_MAP; |
| 3447 | assert(flag != COMMIT_PMAP); |
| 3448 | |
| 3449 | if (flag == COMMIT_PWMAP) |
| 3450 | log = 1; |
| 3451 | else { |
| 3452 | log = 0; |
| 3453 | xadlock.flag = mlckFREEXADLIST; |
| 3454 | xadlock.index = 1; |
| 3455 | } |
| 3456 | |
| 3457 | /* |
| 3458 | * if the newsize is not an integral number of pages, |
| 3459 | * the file between newsize and next page boundary will |
| 3460 | * be cleared. |
| 3461 | * if truncating into a file hole, it will cause |
| 3462 | * a full block to be allocated for the logical block. |
| 3463 | */ |
| 3464 | |
| 3465 | /* |
| 3466 | * release page blocks of truncated region <teof, eof> |
| 3467 | * |
| 3468 | * free the data blocks from the leaf index blocks. |
| 3469 | * delete the parent index entries corresponding to |
| 3470 | * the freed child data/index blocks. |
| 3471 | * free the index blocks themselves which aren't needed |
| 3472 | * in new sized file. |
| 3473 | * |
| 3474 | * index blocks are updated only if the blocks are to be |
| 3475 | * retained in the new sized file. |
| 3476 | * if type is PMAP, the data and index pages are NOT |
| 3477 | * freed, and the data and index blocks are NOT freed |
| 3478 | * from working map. |
| 3479 | * (this will allow continued access of data/index of |
| 3480 | * temporary file (zerolink count file truncated to zero-length)). |
| 3481 | */ |
| 3482 | teof = (newsize + (JFS_SBI(ip->i_sb)->bsize - 1)) >> |
| 3483 | JFS_SBI(ip->i_sb)->l2bsize; |
| 3484 | |
| 3485 | /* clear stack */ |
| 3486 | BT_CLR(&btstack); |
| 3487 | |
| 3488 | /* |
| 3489 | * start with root |
| 3490 | * |
| 3491 | * root resides in the inode |
| 3492 | */ |
| 3493 | bn = 0; |
| 3494 | |
| 3495 | /* |
| 3496 | * first access of each page: |
| 3497 | */ |
| 3498 | getPage: |
| 3499 | XT_GETPAGE(ip, bn, mp, PSIZE, p, rc); |
| 3500 | if (rc) |
| 3501 | return rc; |
| 3502 | |
| 3503 | /* process entries backward from last index */ |
| 3504 | index = le16_to_cpu(p->header.nextindex) - 1; |
| 3505 | |
| 3506 | if (p->header.flag & BT_INTERNAL) |
| 3507 | goto getChild; |
| 3508 | |
| 3509 | /* |
| 3510 | * leaf page |
| 3511 | */ |
| 3512 | |
| 3513 | /* Since this is the rightmost leaf, and we may have already freed |
| 3514 | * a page that was formerly to the right, let's make sure that the |
| 3515 | * next pointer is zero. |
| 3516 | */ |
| 3517 | if (p->header.next) { |
| 3518 | if (log) |
| 3519 | /* |
| 3520 | * Make sure this change to the header is logged. |
| 3521 | * If we really truncate this leaf, the flag |
| 3522 | * will be changed to tlckTRUNCATE |
| 3523 | */ |
| 3524 | tlck = txLock(tid, ip, mp, tlckXTREE|tlckGROW); |
| 3525 | BT_MARK_DIRTY(mp, ip); |
| 3526 | p->header.next = 0; |
| 3527 | } |
| 3528 | |
| 3529 | freed = 0; |
| 3530 | |
| 3531 | /* does region covered by leaf page precede Teof ? */ |
| 3532 | xad = &p->xad[index]; |
| 3533 | xoff = offsetXAD(xad); |
| 3534 | xlen = lengthXAD(xad); |
| 3535 | if (teof >= xoff + xlen) { |
| 3536 | XT_PUTPAGE(mp); |
| 3537 | goto getParent; |
| 3538 | } |
| 3539 | |
| 3540 | /* (re)acquire tlock of the leaf page */ |
| 3541 | if (log) { |
| 3542 | if (++locked_leaves > MAX_TRUNCATE_LEAVES) { |
| 3543 | /* |
| 3544 | * We need to limit the size of the transaction |
| 3545 | * to avoid exhausting pagecache & tlocks |
| 3546 | */ |
| 3547 | XT_PUTPAGE(mp); |
| 3548 | newsize = (xoff + xlen) << JFS_SBI(ip->i_sb)->l2bsize; |
| 3549 | goto getParent; |
| 3550 | } |
| 3551 | tlck = txLock(tid, ip, mp, tlckXTREE); |
| 3552 | tlck->type = tlckXTREE | tlckTRUNCATE; |
| 3553 | xtlck = (struct xtlock *) & tlck->lock; |
| 3554 | xtlck->hwm.offset = le16_to_cpu(p->header.nextindex) - 1; |
| 3555 | } |
| 3556 | BT_MARK_DIRTY(mp, ip); |
| 3557 | |
| 3558 | /* |
| 3559 | * scan backward leaf page entries |
| 3560 | */ |
| 3561 | for (; index >= XTENTRYSTART; index--) { |
| 3562 | xad = &p->xad[index]; |
| 3563 | xoff = offsetXAD(xad); |
| 3564 | xlen = lengthXAD(xad); |
| 3565 | xaddr = addressXAD(xad); |
| 3566 | |
| 3567 | /* |
| 3568 | * The "data" for a directory is indexed by the block |
| 3569 | * device's address space. This metadata must be invalidated |
| 3570 | * here |
| 3571 | */ |
| 3572 | if (S_ISDIR(ip->i_mode) && (teof == 0)) |
| 3573 | invalidate_xad_metapages(ip, *xad); |
| 3574 | /* |
| 3575 | * entry beyond eof: continue scan of current page |
| 3576 | * xad |
| 3577 | * ---|---=======-------> |
| 3578 | * eof |
| 3579 | */ |
| 3580 | if (teof < xoff) { |
| 3581 | nfreed += xlen; |
| 3582 | continue; |
| 3583 | } |
| 3584 | |
| 3585 | /* |
| 3586 | * (xoff <= teof): last entry to be deleted from page; |
| 3587 | * If other entries remain in page: keep and update the page. |
| 3588 | */ |
| 3589 | |
| 3590 | /* |
| 3591 | * eof == entry_start: delete the entry |
| 3592 | * xad |
| 3593 | * -------|=======-------> |
| 3594 | * eof |
| 3595 | * |
| 3596 | */ |
| 3597 | if (teof == xoff) { |
| 3598 | nfreed += xlen; |
| 3599 | |
| 3600 | if (index == XTENTRYSTART) |
| 3601 | break; |
| 3602 | |
| 3603 | nextindex = index; |
| 3604 | } |
| 3605 | /* |
| 3606 | * eof within the entry: truncate the entry. |
| 3607 | * xad |
| 3608 | * -------===|===-------> |
| 3609 | * eof |
| 3610 | */ |
| 3611 | else if (teof < xoff + xlen) { |
| 3612 | /* update truncated entry */ |
| 3613 | len = teof - xoff; |
| 3614 | freexlen = xlen - len; |
| 3615 | XADlength(xad, len); |
| 3616 | |
| 3617 | /* save pxd of truncated extent in tlck */ |
| 3618 | xaddr += len; |
| 3619 | if (log) { /* COMMIT_PWMAP */ |
| 3620 | xtlck->lwm.offset = (xtlck->lwm.offset) ? |
| 3621 | min(index, (int)xtlck->lwm.offset) : index; |
| 3622 | xtlck->lwm.length = index + 1 - |
| 3623 | xtlck->lwm.offset; |
| 3624 | xtlck->twm.offset = index; |
| 3625 | pxdlock = (struct pxd_lock *) & xtlck->pxdlock; |
| 3626 | pxdlock->flag = mlckFREEPXD; |
| 3627 | PXDaddress(&pxdlock->pxd, xaddr); |
| 3628 | PXDlength(&pxdlock->pxd, freexlen); |
| 3629 | } |
| 3630 | /* free truncated extent */ |
| 3631 | else { /* COMMIT_WMAP */ |
| 3632 | |
| 3633 | pxdlock = (struct pxd_lock *) & xadlock; |
| 3634 | pxdlock->flag = mlckFREEPXD; |
| 3635 | PXDaddress(&pxdlock->pxd, xaddr); |
| 3636 | PXDlength(&pxdlock->pxd, freexlen); |
| 3637 | txFreeMap(ip, pxdlock, NULL, COMMIT_WMAP); |
| 3638 | |
| 3639 | /* reset map lock */ |
| 3640 | xadlock.flag = mlckFREEXADLIST; |
| 3641 | } |
| 3642 | |
| 3643 | /* current entry is new last entry; */ |
| 3644 | nextindex = index + 1; |
| 3645 | |
| 3646 | nfreed += freexlen; |
| 3647 | } |
| 3648 | /* |
| 3649 | * eof beyond the entry: |
| 3650 | * xad |
| 3651 | * -------=======---|---> |
| 3652 | * eof |
| 3653 | */ |
| 3654 | else { /* (xoff + xlen < teof) */ |
| 3655 | |
| 3656 | nextindex = index + 1; |
| 3657 | } |
| 3658 | |
| 3659 | if (nextindex < le16_to_cpu(p->header.nextindex)) { |
| 3660 | if (!log) { /* COMMIT_WAMP */ |
| 3661 | xadlock.xdlist = &p->xad[nextindex]; |
| 3662 | xadlock.count = |
| 3663 | le16_to_cpu(p->header.nextindex) - |
| 3664 | nextindex; |
| 3665 | txFreeMap(ip, (struct maplock *) & xadlock, |
| 3666 | NULL, COMMIT_WMAP); |
| 3667 | } |
| 3668 | p->header.nextindex = cpu_to_le16(nextindex); |
| 3669 | } |
| 3670 | |
| 3671 | XT_PUTPAGE(mp); |
| 3672 | |
| 3673 | /* assert(freed == 0); */ |
| 3674 | goto getParent; |
| 3675 | } /* end scan of leaf page entries */ |
| 3676 | |
| 3677 | freed = 1; |
| 3678 | |
| 3679 | /* |
| 3680 | * leaf page become empty: free the page if type != PMAP |
| 3681 | */ |
| 3682 | if (log) { /* COMMIT_PWMAP */ |
| 3683 | /* txCommit() with tlckFREE: |
| 3684 | * free data extents covered by leaf [XTENTRYSTART:hwm); |
| 3685 | * invalidate leaf if COMMIT_PWMAP; |
| 3686 | * if (TRUNCATE), will write LOG_NOREDOPAGE; |
| 3687 | */ |
| 3688 | tlck->type = tlckXTREE | tlckFREE; |
| 3689 | } else { /* COMMIT_WAMP */ |
| 3690 | |
| 3691 | /* free data extents covered by leaf */ |
| 3692 | xadlock.xdlist = &p->xad[XTENTRYSTART]; |
| 3693 | xadlock.count = |
| 3694 | le16_to_cpu(p->header.nextindex) - XTENTRYSTART; |
| 3695 | txFreeMap(ip, (struct maplock *) & xadlock, NULL, COMMIT_WMAP); |
| 3696 | } |
| 3697 | |
| 3698 | if (p->header.flag & BT_ROOT) { |
| 3699 | p->header.flag &= ~BT_INTERNAL; |
| 3700 | p->header.flag |= BT_LEAF; |
| 3701 | p->header.nextindex = cpu_to_le16(XTENTRYSTART); |
| 3702 | |
| 3703 | XT_PUTPAGE(mp); /* debug */ |
| 3704 | goto out; |
| 3705 | } else { |
| 3706 | if (log) { /* COMMIT_PWMAP */ |
| 3707 | /* page will be invalidated at tx completion |
| 3708 | */ |
| 3709 | XT_PUTPAGE(mp); |
| 3710 | } else { /* COMMIT_WMAP */ |
| 3711 | |
| 3712 | if (mp->lid) |
| 3713 | lid_to_tlock(mp->lid)->flag |= tlckFREELOCK; |
| 3714 | |
| 3715 | /* invalidate empty leaf page */ |
| 3716 | discard_metapage(mp); |
| 3717 | } |
| 3718 | } |
| 3719 | |
| 3720 | /* |
| 3721 | * the leaf page become empty: delete the parent entry |
| 3722 | * for the leaf page if the parent page is to be kept |
| 3723 | * in the new sized file. |
| 3724 | */ |
| 3725 | |
| 3726 | /* |
| 3727 | * go back up to the parent page |
| 3728 | */ |
| 3729 | getParent: |
| 3730 | /* pop/restore parent entry for the current child page */ |
| 3731 | if ((parent = BT_POP(&btstack)) == NULL) |
| 3732 | /* current page must have been root */ |
| 3733 | goto out; |
| 3734 | |
| 3735 | /* get back the parent page */ |
| 3736 | bn = parent->bn; |
| 3737 | XT_GETPAGE(ip, bn, mp, PSIZE, p, rc); |
| 3738 | if (rc) |
| 3739 | return rc; |
| 3740 | |
| 3741 | index = parent->index; |
| 3742 | |
| 3743 | /* |
| 3744 | * child page was not empty: |
| 3745 | */ |
| 3746 | if (freed == 0) { |
| 3747 | /* has any entry deleted from parent ? */ |
| 3748 | if (index < le16_to_cpu(p->header.nextindex) - 1) { |
| 3749 | /* (re)acquire tlock on the parent page */ |
| 3750 | if (log) { /* COMMIT_PWMAP */ |
| 3751 | /* txCommit() with tlckTRUNCATE: |
| 3752 | * free child extents covered by parent [); |
| 3753 | */ |
| 3754 | tlck = txLock(tid, ip, mp, tlckXTREE); |
| 3755 | xtlck = (struct xtlock *) & tlck->lock; |
| 3756 | if (!(tlck->type & tlckTRUNCATE)) { |
| 3757 | xtlck->hwm.offset = |
| 3758 | le16_to_cpu(p->header. |
| 3759 | nextindex) - 1; |
| 3760 | tlck->type = |
| 3761 | tlckXTREE | tlckTRUNCATE; |
| 3762 | } |
| 3763 | } else { /* COMMIT_WMAP */ |
| 3764 | |
| 3765 | /* free child extents covered by parent */ |
| 3766 | xadlock.xdlist = &p->xad[index + 1]; |
| 3767 | xadlock.count = |
| 3768 | le16_to_cpu(p->header.nextindex) - |
| 3769 | index - 1; |
| 3770 | txFreeMap(ip, (struct maplock *) & xadlock, |
| 3771 | NULL, COMMIT_WMAP); |
| 3772 | } |
| 3773 | BT_MARK_DIRTY(mp, ip); |
| 3774 | |
| 3775 | p->header.nextindex = cpu_to_le16(index + 1); |
| 3776 | } |
| 3777 | XT_PUTPAGE(mp); |
| 3778 | goto getParent; |
| 3779 | } |
| 3780 | |
| 3781 | /* |
| 3782 | * child page was empty: |
| 3783 | */ |
| 3784 | nfreed += lengthXAD(&p->xad[index]); |
| 3785 | |
| 3786 | /* |
| 3787 | * During working map update, child page's tlock must be handled |
| 3788 | * before parent's. This is because the parent's tlock will cause |
| 3789 | * the child's disk space to be marked available in the wmap, so |
| 3790 | * it's important that the child page be released by that time. |
| 3791 | * |
| 3792 | * ToDo: tlocks should be on doubly-linked list, so we can |
| 3793 | * quickly remove it and add it to the end. |
| 3794 | */ |
| 3795 | |
| 3796 | /* |
| 3797 | * Move parent page's tlock to the end of the tid's tlock list |
| 3798 | */ |
| 3799 | if (log && mp->lid && (tblk->last != mp->lid) && |
| 3800 | lid_to_tlock(mp->lid)->tid) { |
| 3801 | lid_t lid = mp->lid; |
| 3802 | struct tlock *prev; |
| 3803 | |
| 3804 | tlck = lid_to_tlock(lid); |
| 3805 | |
| 3806 | if (tblk->next == lid) |
| 3807 | tblk->next = tlck->next; |
| 3808 | else { |
| 3809 | for (prev = lid_to_tlock(tblk->next); |
| 3810 | prev->next != lid; |
| 3811 | prev = lid_to_tlock(prev->next)) { |
| 3812 | assert(prev->next); |
| 3813 | } |
| 3814 | prev->next = tlck->next; |
| 3815 | } |
| 3816 | lid_to_tlock(tblk->last)->next = lid; |
| 3817 | tlck->next = 0; |
| 3818 | tblk->last = lid; |
| 3819 | } |
| 3820 | |
| 3821 | /* |
| 3822 | * parent page become empty: free the page |
| 3823 | */ |
| 3824 | if (index == XTENTRYSTART) { |
| 3825 | if (log) { /* COMMIT_PWMAP */ |
| 3826 | /* txCommit() with tlckFREE: |
| 3827 | * free child extents covered by parent; |
| 3828 | * invalidate parent if COMMIT_PWMAP; |
| 3829 | */ |
| 3830 | tlck = txLock(tid, ip, mp, tlckXTREE); |
| 3831 | xtlck = (struct xtlock *) & tlck->lock; |
| 3832 | xtlck->hwm.offset = |
| 3833 | le16_to_cpu(p->header.nextindex) - 1; |
| 3834 | tlck->type = tlckXTREE | tlckFREE; |
| 3835 | } else { /* COMMIT_WMAP */ |
| 3836 | |
| 3837 | /* free child extents covered by parent */ |
| 3838 | xadlock.xdlist = &p->xad[XTENTRYSTART]; |
| 3839 | xadlock.count = |
| 3840 | le16_to_cpu(p->header.nextindex) - |
| 3841 | XTENTRYSTART; |
| 3842 | txFreeMap(ip, (struct maplock *) & xadlock, NULL, |
| 3843 | COMMIT_WMAP); |
| 3844 | } |
| 3845 | BT_MARK_DIRTY(mp, ip); |
| 3846 | |
| 3847 | if (p->header.flag & BT_ROOT) { |
| 3848 | p->header.flag &= ~BT_INTERNAL; |
| 3849 | p->header.flag |= BT_LEAF; |
| 3850 | p->header.nextindex = cpu_to_le16(XTENTRYSTART); |
| 3851 | if (le16_to_cpu(p->header.maxentry) == XTROOTMAXSLOT) { |
| 3852 | /* |
| 3853 | * Shrink root down to allow inline |
| 3854 | * EA (otherwise fsck complains) |
| 3855 | */ |
| 3856 | p->header.maxentry = |
| 3857 | cpu_to_le16(XTROOTINITSLOT); |
| 3858 | JFS_IP(ip)->mode2 |= INLINEEA; |
| 3859 | } |
| 3860 | |
| 3861 | XT_PUTPAGE(mp); /* debug */ |
| 3862 | goto out; |
| 3863 | } else { |
| 3864 | if (log) { /* COMMIT_PWMAP */ |
| 3865 | /* page will be invalidated at tx completion |
| 3866 | */ |
| 3867 | XT_PUTPAGE(mp); |
| 3868 | } else { /* COMMIT_WMAP */ |
| 3869 | |
| 3870 | if (mp->lid) |
| 3871 | lid_to_tlock(mp->lid)->flag |= |
| 3872 | tlckFREELOCK; |
| 3873 | |
| 3874 | /* invalidate parent page */ |
| 3875 | discard_metapage(mp); |
| 3876 | } |
| 3877 | |
| 3878 | /* parent has become empty and freed: |
| 3879 | * go back up to its parent page |
| 3880 | */ |
| 3881 | /* freed = 1; */ |
| 3882 | goto getParent; |
| 3883 | } |
| 3884 | } |
| 3885 | /* |
| 3886 | * parent page still has entries for front region; |
| 3887 | */ |
| 3888 | else { |
| 3889 | /* try truncate region covered by preceding entry |
| 3890 | * (process backward) |
| 3891 | */ |
| 3892 | index--; |
| 3893 | |
| 3894 | /* go back down to the child page corresponding |
| 3895 | * to the entry |
| 3896 | */ |
| 3897 | goto getChild; |
| 3898 | } |
| 3899 | |
| 3900 | /* |
| 3901 | * internal page: go down to child page of current entry |
| 3902 | */ |
| 3903 | getChild: |
| 3904 | /* save current parent entry for the child page */ |
| 3905 | BT_PUSH(&btstack, bn, index); |
| 3906 | |
| 3907 | /* get child page */ |
| 3908 | xad = &p->xad[index]; |
| 3909 | bn = addressXAD(xad); |
| 3910 | |
| 3911 | /* |
| 3912 | * first access of each internal entry: |
| 3913 | */ |
| 3914 | /* release parent page */ |
| 3915 | XT_PUTPAGE(mp); |
| 3916 | |
| 3917 | /* process the child page */ |
| 3918 | goto getPage; |
| 3919 | |
| 3920 | out: |
| 3921 | /* |
| 3922 | * update file resource stat |
| 3923 | */ |
| 3924 | /* set size |
| 3925 | */ |
| 3926 | if (S_ISDIR(ip->i_mode) && !newsize) |
| 3927 | ip->i_size = 1; /* fsck hates zero-length directories */ |
| 3928 | else |
| 3929 | ip->i_size = newsize; |
| 3930 | |
| 3931 | /* update quota allocation to reflect freed blocks */ |
| 3932 | DQUOT_FREE_BLOCK(ip, nfreed); |
| 3933 | |
| 3934 | /* |
| 3935 | * free tlock of invalidated pages |
| 3936 | */ |
| 3937 | if (flag == COMMIT_WMAP) |
| 3938 | txFreelock(ip); |
| 3939 | |
| 3940 | return newsize; |
| 3941 | } |
| 3942 | |
| 3943 | |
| 3944 | /* |
| 3945 | * xtTruncate_pmap() |
| 3946 | * |
| 3947 | * function: |
| 3948 | * Perform truncate to zero lenghth for deleted file, leaving the |
| 3949 | * the xtree and working map untouched. This allows the file to |
| 3950 | * be accessed via open file handles, while the delete of the file |
| 3951 | * is committed to disk. |
| 3952 | * |
| 3953 | * parameter: |
| 3954 | * tid_t tid, |
| 3955 | * struct inode *ip, |
| 3956 | * s64 committed_size) |
| 3957 | * |
| 3958 | * return: new committed size |
| 3959 | * |
| 3960 | * note: |
| 3961 | * |
| 3962 | * To avoid deadlock by holding too many transaction locks, the |
| 3963 | * truncation may be broken up into multiple transactions. |
| 3964 | * The committed_size keeps track of part of the file has been |
| 3965 | * freed from the pmaps. |
| 3966 | */ |
| 3967 | s64 xtTruncate_pmap(tid_t tid, struct inode *ip, s64 committed_size) |
| 3968 | { |
| 3969 | s64 bn; |
| 3970 | struct btstack btstack; |
| 3971 | int cmp; |
| 3972 | int index; |
| 3973 | int locked_leaves = 0; |
| 3974 | struct metapage *mp; |
| 3975 | xtpage_t *p; |
| 3976 | struct btframe *parent; |
| 3977 | int rc; |
| 3978 | struct tblock *tblk; |
| 3979 | struct tlock *tlck = NULL; |
| 3980 | xad_t *xad; |
| 3981 | int xlen; |
| 3982 | s64 xoff; |
| 3983 | struct xtlock *xtlck = NULL; |
| 3984 | |
| 3985 | /* save object truncation type */ |
| 3986 | tblk = tid_to_tblock(tid); |
| 3987 | tblk->xflag |= COMMIT_PMAP; |
| 3988 | |
| 3989 | /* clear stack */ |
| 3990 | BT_CLR(&btstack); |
| 3991 | |
| 3992 | if (committed_size) { |
| 3993 | xoff = (committed_size >> JFS_SBI(ip->i_sb)->l2bsize) - 1; |
| 3994 | rc = xtSearch(ip, xoff, &cmp, &btstack, 0); |
| 3995 | if (rc) |
| 3996 | return rc; |
| 3997 | |
| 3998 | XT_GETSEARCH(ip, btstack.top, bn, mp, p, index); |
| 3999 | |
| 4000 | if (cmp != 0) { |
| 4001 | XT_PUTPAGE(mp); |
| 4002 | jfs_error(ip->i_sb, |
| 4003 | "xtTruncate_pmap: did not find extent"); |
| 4004 | return -EIO; |
| 4005 | } |
| 4006 | } else { |
| 4007 | /* |
| 4008 | * start with root |
| 4009 | * |
| 4010 | * root resides in the inode |
| 4011 | */ |
| 4012 | bn = 0; |
| 4013 | |
| 4014 | /* |
| 4015 | * first access of each page: |
| 4016 | */ |
| 4017 | getPage: |
| 4018 | XT_GETPAGE(ip, bn, mp, PSIZE, p, rc); |
| 4019 | if (rc) |
| 4020 | return rc; |
| 4021 | |
| 4022 | /* process entries backward from last index */ |
| 4023 | index = le16_to_cpu(p->header.nextindex) - 1; |
| 4024 | |
| 4025 | if (p->header.flag & BT_INTERNAL) |
| 4026 | goto getChild; |
| 4027 | } |
| 4028 | |
| 4029 | /* |
| 4030 | * leaf page |
| 4031 | */ |
| 4032 | |
| 4033 | if (++locked_leaves > MAX_TRUNCATE_LEAVES) { |
| 4034 | /* |
| 4035 | * We need to limit the size of the transaction |
| 4036 | * to avoid exhausting pagecache & tlocks |
| 4037 | */ |
| 4038 | xad = &p->xad[index]; |
| 4039 | xoff = offsetXAD(xad); |
| 4040 | xlen = lengthXAD(xad); |
| 4041 | XT_PUTPAGE(mp); |
| 4042 | return (xoff + xlen) << JFS_SBI(ip->i_sb)->l2bsize; |
| 4043 | } |
| 4044 | tlck = txLock(tid, ip, mp, tlckXTREE); |
| 4045 | tlck->type = tlckXTREE | tlckFREE; |
| 4046 | xtlck = (struct xtlock *) & tlck->lock; |
| 4047 | xtlck->hwm.offset = index; |
| 4048 | |
| 4049 | |
| 4050 | XT_PUTPAGE(mp); |
| 4051 | |
| 4052 | /* |
| 4053 | * go back up to the parent page |
| 4054 | */ |
| 4055 | getParent: |
| 4056 | /* pop/restore parent entry for the current child page */ |
| 4057 | if ((parent = BT_POP(&btstack)) == NULL) |
| 4058 | /* current page must have been root */ |
| 4059 | goto out; |
| 4060 | |
| 4061 | /* get back the parent page */ |
| 4062 | bn = parent->bn; |
| 4063 | XT_GETPAGE(ip, bn, mp, PSIZE, p, rc); |
| 4064 | if (rc) |
| 4065 | return rc; |
| 4066 | |
| 4067 | index = parent->index; |
| 4068 | |
| 4069 | /* |
| 4070 | * parent page become empty: free the page |
| 4071 | */ |
| 4072 | if (index == XTENTRYSTART) { |
| 4073 | /* txCommit() with tlckFREE: |
| 4074 | * free child extents covered by parent; |
| 4075 | * invalidate parent if COMMIT_PWMAP; |
| 4076 | */ |
| 4077 | tlck = txLock(tid, ip, mp, tlckXTREE); |
| 4078 | xtlck = (struct xtlock *) & tlck->lock; |
| 4079 | xtlck->hwm.offset = |
| 4080 | le16_to_cpu(p->header.nextindex) - 1; |
| 4081 | tlck->type = tlckXTREE | tlckFREE; |
| 4082 | |
| 4083 | XT_PUTPAGE(mp); |
| 4084 | |
| 4085 | if (p->header.flag & BT_ROOT) { |
| 4086 | |
| 4087 | goto out; |
| 4088 | } else { |
| 4089 | goto getParent; |
| 4090 | } |
| 4091 | } |
| 4092 | /* |
| 4093 | * parent page still has entries for front region; |
| 4094 | */ |
| 4095 | else |
| 4096 | index--; |
| 4097 | /* |
| 4098 | * internal page: go down to child page of current entry |
| 4099 | */ |
| 4100 | getChild: |
| 4101 | /* save current parent entry for the child page */ |
| 4102 | BT_PUSH(&btstack, bn, index); |
| 4103 | |
| 4104 | /* get child page */ |
| 4105 | xad = &p->xad[index]; |
| 4106 | bn = addressXAD(xad); |
| 4107 | |
| 4108 | /* |
| 4109 | * first access of each internal entry: |
| 4110 | */ |
| 4111 | /* release parent page */ |
| 4112 | XT_PUTPAGE(mp); |
| 4113 | |
| 4114 | /* process the child page */ |
| 4115 | goto getPage; |
| 4116 | |
| 4117 | out: |
| 4118 | |
| 4119 | return 0; |
| 4120 | } |
| 4121 | |
| 4122 | |
| 4123 | #ifdef _JFS_DEBUG_XTREE |
| 4124 | /* |
| 4125 | * xtDisplayTree() |
| 4126 | * |
| 4127 | * function: traverse forward |
| 4128 | */ |
| 4129 | int xtDisplayTree(struct inode *ip) |
| 4130 | { |
| 4131 | int rc = 0; |
| 4132 | struct metapage *mp; |
| 4133 | xtpage_t *p; |
| 4134 | s64 bn, pbn; |
| 4135 | int index, lastindex, v, h; |
| 4136 | xad_t *xad; |
| 4137 | struct btstack btstack; |
| 4138 | struct btframe *btsp; |
| 4139 | struct btframe *parent; |
| 4140 | |
| 4141 | printk("display B+-tree.\n"); |
| 4142 | |
| 4143 | /* clear stack */ |
| 4144 | btsp = btstack.stack; |
| 4145 | |
| 4146 | /* |
| 4147 | * start with root |
| 4148 | * |
| 4149 | * root resides in the inode |
| 4150 | */ |
| 4151 | bn = 0; |
| 4152 | v = h = 0; |
| 4153 | |
| 4154 | /* |
| 4155 | * first access of each page: |
| 4156 | */ |
| 4157 | getPage: |
| 4158 | XT_GETPAGE(ip, bn, mp, PSIZE, p, rc); |
| 4159 | if (rc) |
| 4160 | return rc; |
| 4161 | |
| 4162 | /* process entries forward from first index */ |
| 4163 | index = XTENTRYSTART; |
| 4164 | lastindex = le16_to_cpu(p->header.nextindex) - 1; |
| 4165 | |
| 4166 | if (p->header.flag & BT_INTERNAL) { |
| 4167 | /* |
| 4168 | * first access of each internal page |
| 4169 | */ |
| 4170 | goto getChild; |
| 4171 | } else { /* (p->header.flag & BT_LEAF) */ |
| 4172 | |
| 4173 | /* |
| 4174 | * first access of each leaf page |
| 4175 | */ |
| 4176 | printf("leaf page "); |
| 4177 | xtDisplayPage(ip, bn, p); |
| 4178 | |
| 4179 | /* unpin the leaf page */ |
| 4180 | XT_PUTPAGE(mp); |
| 4181 | } |
| 4182 | |
| 4183 | /* |
| 4184 | * go back up to the parent page |
| 4185 | */ |
| 4186 | getParent: |
| 4187 | /* pop/restore parent entry for the current child page */ |
| 4188 | if ((parent = (btsp == btstack.stack ? NULL : --btsp)) == NULL) |
| 4189 | /* current page must have been root */ |
| 4190 | return; |
| 4191 | |
| 4192 | /* |
| 4193 | * parent page scan completed |
| 4194 | */ |
| 4195 | if ((index = parent->index) == (lastindex = parent->lastindex)) { |
| 4196 | /* go back up to the parent page */ |
| 4197 | goto getParent; |
| 4198 | } |
| 4199 | |
| 4200 | /* |
| 4201 | * parent page has entries remaining |
| 4202 | */ |
| 4203 | /* get back the parent page */ |
| 4204 | bn = parent->bn; |
| 4205 | /* v = parent->level; */ |
| 4206 | XT_GETPAGE(ip, bn, mp, PSIZE, p, rc); |
| 4207 | if (rc) |
| 4208 | return rc; |
| 4209 | |
| 4210 | /* get next parent entry */ |
| 4211 | index++; |
| 4212 | |
| 4213 | /* |
| 4214 | * internal page: go down to child page of current entry |
| 4215 | */ |
| 4216 | getChild: |
| 4217 | /* push/save current parent entry for the child page */ |
| 4218 | btsp->bn = pbn = bn; |
| 4219 | btsp->index = index; |
| 4220 | btsp->lastindex = lastindex; |
| 4221 | /* btsp->level = v; */ |
| 4222 | /* btsp->node = h; */ |
| 4223 | ++btsp; |
| 4224 | |
| 4225 | /* get child page */ |
| 4226 | xad = &p->xad[index]; |
| 4227 | bn = addressXAD(xad); |
| 4228 | |
| 4229 | /* |
| 4230 | * first access of each internal entry: |
| 4231 | */ |
| 4232 | /* release parent page */ |
| 4233 | XT_PUTPAGE(mp); |
| 4234 | |
| 4235 | printk("traverse down 0x%lx[%d]->0x%lx\n", (ulong) pbn, index, |
| 4236 | (ulong) bn); |
| 4237 | v++; |
| 4238 | h = index; |
| 4239 | |
| 4240 | /* process the child page */ |
| 4241 | goto getPage; |
| 4242 | } |
| 4243 | |
| 4244 | |
| 4245 | /* |
| 4246 | * xtDisplayPage() |
| 4247 | * |
| 4248 | * function: display page |
| 4249 | */ |
| 4250 | int xtDisplayPage(struct inode *ip, s64 bn, xtpage_t * p) |
| 4251 | { |
| 4252 | int rc = 0; |
| 4253 | xad_t *xad; |
| 4254 | s64 xaddr, xoff; |
| 4255 | int xlen, i, j; |
| 4256 | |
| 4257 | /* display page control */ |
| 4258 | printf("bn:0x%lx flag:0x%x nextindex:%d\n", |
| 4259 | (ulong) bn, p->header.flag, |
| 4260 | le16_to_cpu(p->header.nextindex)); |
| 4261 | |
| 4262 | /* display entries */ |
| 4263 | xad = &p->xad[XTENTRYSTART]; |
| 4264 | for (i = XTENTRYSTART, j = 1; i < le16_to_cpu(p->header.nextindex); |
| 4265 | i++, xad++, j++) { |
| 4266 | xoff = offsetXAD(xad); |
| 4267 | xaddr = addressXAD(xad); |
| 4268 | xlen = lengthXAD(xad); |
| 4269 | printf("\t[%d] 0x%lx:0x%lx(0x%x)", i, (ulong) xoff, |
| 4270 | (ulong) xaddr, xlen); |
| 4271 | |
| 4272 | if (j == 4) { |
| 4273 | printf("\n"); |
| 4274 | j = 0; |
| 4275 | } |
| 4276 | } |
| 4277 | |
| 4278 | printf("\n"); |
| 4279 | } |
| 4280 | #endif /* _JFS_DEBUG_XTREE */ |
| 4281 | |
| 4282 | |
| 4283 | #ifdef _JFS_WIP |
| 4284 | /* |
| 4285 | * xtGather() |
| 4286 | * |
| 4287 | * function: |
| 4288 | * traverse for allocation acquiring tlock at commit time |
| 4289 | * (vs at the time of update) logging backward top down |
| 4290 | * |
| 4291 | * note: |
| 4292 | * problem - establishing that all new allocation have been |
| 4293 | * processed both for append and random write in sparse file |
| 4294 | * at the current entry at the current subtree root page |
| 4295 | * |
| 4296 | */ |
| 4297 | int xtGather(btree_t *t) |
| 4298 | { |
| 4299 | int rc = 0; |
| 4300 | xtpage_t *p; |
| 4301 | u64 bn; |
| 4302 | int index; |
| 4303 | btentry_t *e; |
| 4304 | struct btstack btstack; |
| 4305 | struct btsf *parent; |
| 4306 | |
| 4307 | /* clear stack */ |
| 4308 | BT_CLR(&btstack); |
| 4309 | |
| 4310 | /* |
| 4311 | * start with root |
| 4312 | * |
| 4313 | * root resides in the inode |
| 4314 | */ |
| 4315 | bn = 0; |
| 4316 | XT_GETPAGE(ip, bn, mp, PSIZE, p, rc); |
| 4317 | if (rc) |
| 4318 | return rc; |
| 4319 | |
| 4320 | /* new root is NOT pointed by a new entry |
| 4321 | if (p->header.flag & NEW) |
| 4322 | allocate new page lock; |
| 4323 | write a NEWPAGE log; |
| 4324 | */ |
| 4325 | |
| 4326 | dopage: |
| 4327 | /* |
| 4328 | * first access of each page: |
| 4329 | */ |
| 4330 | /* process entries backward from last index */ |
| 4331 | index = le16_to_cpu(p->header.nextindex) - 1; |
| 4332 | |
| 4333 | if (p->header.flag & BT_LEAF) { |
| 4334 | /* |
| 4335 | * first access of each leaf page |
| 4336 | */ |
| 4337 | /* process leaf page entries backward */ |
| 4338 | for (; index >= XTENTRYSTART; index--) { |
| 4339 | e = &p->xad[index]; |
| 4340 | /* |
| 4341 | * if newpage, log NEWPAGE. |
| 4342 | * |
| 4343 | if (e->flag & XAD_NEW) { |
| 4344 | nfound =+ entry->length; |
| 4345 | update current page lock for the entry; |
| 4346 | newpage(entry); |
| 4347 | * |
| 4348 | * if moved, log move. |
| 4349 | * |
| 4350 | } else if (e->flag & XAD_MOVED) { |
| 4351 | reset flag; |
| 4352 | update current page lock for the entry; |
| 4353 | } |
| 4354 | */ |
| 4355 | } |
| 4356 | |
| 4357 | /* unpin the leaf page */ |
| 4358 | XT_PUTPAGE(mp); |
| 4359 | |
| 4360 | /* |
| 4361 | * go back up to the parent page |
| 4362 | */ |
| 4363 | getParent: |
| 4364 | /* restore parent entry for the current child page */ |
| 4365 | if ((parent = BT_POP(&btstack)) == NULL) |
| 4366 | /* current page must have been root */ |
| 4367 | return 0; |
| 4368 | |
| 4369 | if ((index = parent->index) == XTENTRYSTART) { |
| 4370 | /* |
| 4371 | * parent page scan completed |
| 4372 | */ |
| 4373 | /* go back up to the parent page */ |
| 4374 | goto getParent; |
| 4375 | } else { |
| 4376 | /* |
| 4377 | * parent page has entries remaining |
| 4378 | */ |
| 4379 | /* get back the parent page */ |
| 4380 | bn = parent->bn; |
| 4381 | XT_GETPAGE(ip, bn, mp, PSIZE, p, rc); |
| 4382 | if (rc) |
| 4383 | return -EIO; |
| 4384 | |
| 4385 | /* first subroot page which |
| 4386 | * covers all new allocated blocks |
| 4387 | * itself not new/modified. |
| 4388 | * (if modified from split of descendent, |
| 4389 | * go down path of split page) |
| 4390 | |
| 4391 | if (nfound == nnew && |
| 4392 | !(p->header.flag & (NEW | MOD))) |
| 4393 | exit scan; |
| 4394 | */ |
| 4395 | |
| 4396 | /* process parent page entries backward */ |
| 4397 | index--; |
| 4398 | } |
| 4399 | } else { |
| 4400 | /* |
| 4401 | * first access of each internal page |
| 4402 | */ |
| 4403 | } |
| 4404 | |
| 4405 | /* |
| 4406 | * internal page: go down to child page of current entry |
| 4407 | */ |
| 4408 | |
| 4409 | /* save current parent entry for the child page */ |
| 4410 | BT_PUSH(&btstack, bn, index); |
| 4411 | |
| 4412 | /* get current entry for the child page */ |
| 4413 | e = &p->xad[index]; |
| 4414 | |
| 4415 | /* |
| 4416 | * first access of each internal entry: |
| 4417 | */ |
| 4418 | /* |
| 4419 | * if new entry, log btree_tnewentry. |
| 4420 | * |
| 4421 | if (e->flag & XAD_NEW) |
| 4422 | update parent page lock for the entry; |
| 4423 | */ |
| 4424 | |
| 4425 | /* release parent page */ |
| 4426 | XT_PUTPAGE(mp); |
| 4427 | |
| 4428 | /* get child page */ |
| 4429 | bn = e->bn; |
| 4430 | XT_GETPAGE(ip, bn, mp, PSIZE, p, rc); |
| 4431 | if (rc) |
| 4432 | return rc; |
| 4433 | |
| 4434 | /* |
| 4435 | * first access of each non-root page: |
| 4436 | */ |
| 4437 | /* |
| 4438 | * if new, log btree_newpage. |
| 4439 | * |
| 4440 | if (p->header.flag & NEW) |
| 4441 | allocate new page lock; |
| 4442 | write a NEWPAGE log (next, prev); |
| 4443 | */ |
| 4444 | |
| 4445 | /* process the child page */ |
| 4446 | goto dopage; |
| 4447 | |
| 4448 | out: |
| 4449 | return 0; |
| 4450 | } |
| 4451 | #endif /* _JFS_WIP */ |
| 4452 | |
| 4453 | |
| 4454 | #ifdef CONFIG_JFS_STATISTICS |
| 4455 | int jfs_xtstat_read(char *buffer, char **start, off_t offset, int length, |
| 4456 | int *eof, void *data) |
| 4457 | { |
| 4458 | int len = 0; |
| 4459 | off_t begin; |
| 4460 | |
| 4461 | len += sprintf(buffer, |
| 4462 | "JFS Xtree statistics\n" |
| 4463 | "====================\n" |
| 4464 | "searches = %d\n" |
| 4465 | "fast searches = %d\n" |
| 4466 | "splits = %d\n", |
| 4467 | xtStat.search, |
| 4468 | xtStat.fastSearch, |
| 4469 | xtStat.split); |
| 4470 | |
| 4471 | begin = offset; |
| 4472 | *start = buffer + begin; |
| 4473 | len -= begin; |
| 4474 | |
| 4475 | if (len > length) |
| 4476 | len = length; |
| 4477 | else |
| 4478 | *eof = 1; |
| 4479 | |
| 4480 | if (len < 0) |
| 4481 | len = 0; |
| 4482 | |
| 4483 | return len; |
| 4484 | } |
| 4485 | #endif |