1 // SPDX-License-Identifier: GPL-2.0-or-later
2 /*
3 * Copyright (C) International Business Machines Corp., 2000-2005
4 */
5 /*
6 * jfs_xtree.c: extent allocation descriptor B+-tree manager
7 */
8
9 #include <linux/fs.h>
10 #include <linux/module.h>
11 #include <linux/quotaops.h>
12 #include <linux/seq_file.h>
13 #include "jfs_incore.h"
14 #include "jfs_filsys.h"
15 #include "jfs_metapage.h"
16 #include "jfs_dmap.h"
17 #include "jfs_dinode.h"
18 #include "jfs_superblock.h"
19 #include "jfs_debug.h"
20
21 /*
22 * xtree local flag
23 */
24 #define XT_INSERT 0x00000001
25
26 /*
27 * xtree key/entry comparison: extent offset
28 *
29 * return:
30 * -1: k < start of extent
31 * 0: start_of_extent <= k <= end_of_extent
32 * 1: k > end_of_extent
33 */
34 #define XT_CMP(CMP, K, X, OFFSET64)\
35 {\
36 OFFSET64 = offsetXAD(X);\
37 (CMP) = ((K) >= OFFSET64 + lengthXAD(X)) ? 1 :\
38 ((K) < OFFSET64) ? -1 : 0;\
39 }
40
41 /* write a xad entry */
42 #define XT_PUTENTRY(XAD, FLAG, OFF, LEN, ADDR)\
43 {\
44 (XAD)->flag = (FLAG);\
45 XADoffset((XAD), (OFF));\
46 XADlength((XAD), (LEN));\
47 XADaddress((XAD), (ADDR));\
48 }
49
50 #define XT_PAGE(IP, MP) BT_PAGE(IP, MP, xtpage_t, i_xtroot)
51
52 /* get page buffer for specified block address */
53 /* ToDo: Replace this ugly macro with a function */
54 #define XT_GETPAGE(IP, BN, MP, SIZE, P, RC) \
55 do { \
56 BT_GETPAGE(IP, BN, MP, xtpage_t, SIZE, P, RC, i_xtroot); \
57 if (!(RC)) { \
58 if ((le16_to_cpu((P)->header.nextindex) < XTENTRYSTART) || \
59 (le16_to_cpu((P)->header.nextindex) > \
60 le16_to_cpu((P)->header.maxentry)) || \
61 (le16_to_cpu((P)->header.maxentry) > \
62 (((BN) == 0) ? XTROOTMAXSLOT : PSIZE >> L2XTSLOTSIZE))) { \
63 jfs_error((IP)->i_sb, \
64 "XT_GETPAGE: xtree page corrupt\n"); \
65 BT_PUTPAGE(MP); \
66 MP = NULL; \
67 RC = -EIO; \
68 } \
69 } \
70 } while (0)
71
72 /* for consistency */
73 #define XT_PUTPAGE(MP) BT_PUTPAGE(MP)
74
75 #define XT_GETSEARCH(IP, LEAF, BN, MP, P, INDEX) \
76 BT_GETSEARCH(IP, LEAF, BN, MP, xtpage_t, P, INDEX, i_xtroot)
77 /* xtree entry parameter descriptor */
78 struct xtsplit {
79 struct metapage *mp;
80 s16 index;
81 u8 flag;
82 s64 off;
83 s64 addr;
84 int len;
85 struct pxdlist *pxdlist;
86 };
87
88
89 /*
90 * statistics
91 */
92 #ifdef CONFIG_JFS_STATISTICS
93 static struct {
94 uint search;
95 uint fastSearch;
96 uint split;
97 } xtStat;
98 #endif
99
100
101 /*
102 * forward references
103 */
104 static int xtSearch(struct inode *ip, s64 xoff, s64 *next, int *cmpp,
105 struct btstack * btstack, int flag);
106
107 static int xtSplitUp(tid_t tid,
108 struct inode *ip,
109 struct xtsplit * split, struct btstack * btstack);
110
111 static int xtSplitPage(tid_t tid, struct inode *ip, struct xtsplit * split,
112 struct metapage ** rmpp, s64 * rbnp);
113
114 static int xtSplitRoot(tid_t tid, struct inode *ip,
115 struct xtsplit * split, struct metapage ** rmpp);
116
117 /*
118 * xtLookup()
119 *
120 * function: map a single page into a physical extent;
121 */
122 int xtLookup(struct inode *ip, s64 lstart,
123 s64 llen, int *pflag, s64 * paddr, s32 * plen, int no_check)
124 {
125 int rc = 0;
126 struct btstack btstack;
127 int cmp;
128 s64 bn;
129 struct metapage *mp;
130 xtpage_t *p;
131 int index;
132 xad_t *xad;
133 s64 next, size, xoff, xend;
134 int xlen;
135 s64 xaddr;
136
137 *paddr = 0;
138 *plen = llen;
139
140 if (!no_check) {
141 /* is lookup offset beyond eof ? */
142 size = ((u64) ip->i_size + (JFS_SBI(ip->i_sb)->bsize - 1)) >>
143 JFS_SBI(ip->i_sb)->l2bsize;
144 if (lstart >= size)
145 return 0;
146 }
147
148 /*
149 * search for the xad entry covering the logical extent
150 */
151 //search:
152 if ((rc = xtSearch(ip, lstart, &next, &cmp, &btstack, 0))) {
153 jfs_err("xtLookup: xtSearch returned %d", rc);
154 return rc;
155 }
156
157 /*
158 * compute the physical extent covering logical extent
159 *
160 * N.B. search may have failed (e.g., hole in sparse file),
161 * and returned the index of the next entry.
162 */
163 /* retrieve search result */
164 XT_GETSEARCH(ip, btstack.top, bn, mp, p, index);
165
166 /* is xad found covering start of logical extent ?
167 * lstart is a page start address,
168 * i.e., lstart cannot start in a hole;
169 */
170 if (cmp) {
171 if (next)
172 *plen = min(next - lstart, llen);
173 goto out;
174 }
175
176 /*
177 * lxd covered by xad
178 */
179 xad = &p->xad[index];
180 xoff = offsetXAD(xad);
181 xlen = lengthXAD(xad);
182 xend = xoff + xlen;
183 xaddr = addressXAD(xad);
184
185 /* initialize new pxd */
186 *pflag = xad->flag;
187 *paddr = xaddr + (lstart - xoff);
188 /* a page must be fully covered by an xad */
189 *plen = min(xend - lstart, llen);
190
191 out:
192 XT_PUTPAGE(mp);
193
194 return rc;
195 }
196
197 /*
198 * xtSearch()
199 *
200 * function: search for the xad entry covering specified offset.
201 *
202 * parameters:
203 * ip - file object;
204 * xoff - extent offset;
205 * nextp - address of next extent (if any) for search miss
206 * cmpp - comparison result:
207 * btstack - traverse stack;
208 * flag - search process flag (XT_INSERT);
209 *
210 * returns:
211 * btstack contains (bn, index) of search path traversed to the entry.
212 * *cmpp is set to result of comparison with the entry returned.
213 * the page containing the entry is pinned at exit.
214 */
215 static int xtSearch(struct inode *ip, s64 xoff, s64 *nextp,
216 int *cmpp, struct btstack * btstack, int flag)
217 {
218 struct jfs_inode_info *jfs_ip = JFS_IP(ip);
219 int rc = 0;
220 int cmp = 1; /* init for empty page */
221 s64 bn; /* block number */
222 struct metapage *mp; /* page buffer */
223 xtpage_t *p; /* page */
224 xad_t *xad;
225 int base, index, lim, btindex;
226 struct btframe *btsp;
227 int nsplit = 0; /* number of pages to split */
228 s64 t64;
229 s64 next = 0;
230
231 INCREMENT(xtStat.search);
232
233 BT_CLR(btstack);
234
235 btstack->nsplit = 0;
236
237 /*
238 * search down tree from root:
239 *
240 * between two consecutive entries of <Ki, Pi> and <Kj, Pj> of
241 * internal page, child page Pi contains entry with k, Ki <= K < Kj.
242 *
243 * if entry with search key K is not found
244 * internal page search find the entry with largest key Ki
245 * less than K which point to the child page to search;
246 * leaf page search find the entry with smallest key Kj
247 * greater than K so that the returned index is the position of
248 * the entry to be shifted right for insertion of new entry.
249 * for empty tree, search key is greater than any key of the tree.
250 *
251 * by convention, root bn = 0.
252 */
253 for (bn = 0;;) {
254 /* get/pin the page to search */
255 XT_GETPAGE(ip, bn, mp, PSIZE, p, rc);
256 if (rc)
257 return rc;
258
259 /* try sequential access heuristics with the previous
260 * access entry in target leaf page:
261 * once search narrowed down into the target leaf,
262 * key must either match an entry in the leaf or
263 * key entry does not exist in the tree;
264 */
265 //fastSearch:
266 if ((jfs_ip->btorder & BT_SEQUENTIAL) &&
267 (p->header.flag & BT_LEAF) &&
268 (index = jfs_ip->btindex) <
269 le16_to_cpu(p->header.nextindex)) {
270 xad = &p->xad[index];
271 t64 = offsetXAD(xad);
272 if (xoff < t64 + lengthXAD(xad)) {
273 if (xoff >= t64) {
274 *cmpp = 0;
275 goto out;
276 }
277
278 /* stop sequential access heuristics */
279 goto binarySearch;
280 } else { /* (t64 + lengthXAD(xad)) <= xoff */
281
282 /* try next sequential entry */
283 index++;
284 if (index <
285 le16_to_cpu(p->header.nextindex)) {
286 xad++;
287 t64 = offsetXAD(xad);
288 if (xoff < t64 + lengthXAD(xad)) {
289 if (xoff >= t64) {
290 *cmpp = 0;
291 goto out;
292 }
293
294 /* miss: key falls between
295 * previous and this entry
296 */
297 *cmpp = 1;
298 next = t64;
299 goto out;
300 }
301
302 /* (xoff >= t64 + lengthXAD(xad));
303 * matching entry may be further out:
304 * stop heuristic search
305 */
306 /* stop sequential access heuristics */
307 goto binarySearch;
308 }
309
310 /* (index == p->header.nextindex);
311 * miss: key entry does not exist in
312 * the target leaf/tree
313 */
314 *cmpp = 1;
315 goto out;
316 }
317
318 /*
319 * if hit, return index of the entry found, and
320 * if miss, where new entry with search key is
321 * to be inserted;
322 */
323 out:
324 /* compute number of pages to split */
325 if (flag & XT_INSERT) {
326 if (p->header.nextindex == /* little-endian */
327 p->header.maxentry)
328 nsplit++;
329 else
330 nsplit = 0;
331 btstack->nsplit = nsplit;
332 }
333
334 /* save search result */
335 btsp = btstack->top;
336 btsp->bn = bn;
337 btsp->index = index;
338 btsp->mp = mp;
339
340 /* update sequential access heuristics */
341 jfs_ip->btindex = index;
342
343 if (nextp)
344 *nextp = next;
345
346 INCREMENT(xtStat.fastSearch);
347 return 0;
348 }
349
350 /* well, ... full search now */
351 binarySearch:
352 lim = le16_to_cpu(p->header.nextindex) - XTENTRYSTART;
353
354 /*
355 * binary search with search key K on the current page
356 */
357 for (base = XTENTRYSTART; lim; lim >>= 1) {
358 index = base + (lim >> 1);
359
360 XT_CMP(cmp, xoff, &p->xad[index], t64);
361 if (cmp == 0) {
362 /*
363 * search hit
364 */
365 /* search hit - leaf page:
366 * return the entry found
367 */
368 if (p->header.flag & BT_LEAF) {
369 *cmpp = cmp;
370
371 /* compute number of pages to split */
372 if (flag & XT_INSERT) {
373 if (p->header.nextindex ==
374 p->header.maxentry)
375 nsplit++;
376 else
377 nsplit = 0;
378 btstack->nsplit = nsplit;
379 }
380
381 /* save search result */
382 btsp = btstack->top;
383 btsp->bn = bn;
384 btsp->index = index;
385 btsp->mp = mp;
386
387 /* init sequential access heuristics */
388 btindex = jfs_ip->btindex;
389 if (index == btindex ||
390 index == btindex + 1)
391 jfs_ip->btorder = BT_SEQUENTIAL;
392 else
393 jfs_ip->btorder = BT_RANDOM;
394 jfs_ip->btindex = index;
395
396 return 0;
397 }
398 /* search hit - internal page:
399 * descend/search its child page
400 */
401 if (index < le16_to_cpu(p->header.nextindex)-1)
402 next = offsetXAD(&p->xad[index + 1]);
403 goto next;
404 }
405
406 if (cmp > 0) {
407 base = index + 1;
408 --lim;
409 }
410 }
411
412 /*
413 * search miss
414 *
415 * base is the smallest index with key (Kj) greater than
416 * search key (K) and may be zero or maxentry index.
417 */
418 if (base < le16_to_cpu(p->header.nextindex))
419 next = offsetXAD(&p->xad[base]);
420 /*
421 * search miss - leaf page:
422 *
423 * return location of entry (base) where new entry with
424 * search key K is to be inserted.
425 */
426 if (p->header.flag & BT_LEAF) {
427 *cmpp = cmp;
428
429 /* compute number of pages to split */
430 if (flag & XT_INSERT) {
431 if (p->header.nextindex ==
432 p->header.maxentry)
433 nsplit++;
434 else
435 nsplit = 0;
436 btstack->nsplit = nsplit;
437 }
438
439 /* save search result */
440 btsp = btstack->top;
441 btsp->bn = bn;
442 btsp->index = base;
443 btsp->mp = mp;
444
445 /* init sequential access heuristics */
446 btindex = jfs_ip->btindex;
447 if (base == btindex || base == btindex + 1)
448 jfs_ip->btorder = BT_SEQUENTIAL;
449 else
450 jfs_ip->btorder = BT_RANDOM;
451 jfs_ip->btindex = base;
452
453 if (nextp)
454 *nextp = next;
455
456 return 0;
457 }
458
459 /*
460 * search miss - non-leaf page:
461 *
462 * if base is non-zero, decrement base by one to get the parent
463 * entry of the child page to search.
464 */
465 index = base ? base - 1 : base;
466
467 /*
468 * go down to child page
469 */
470 next:
471 /* update number of pages to split */
472 if (p->header.nextindex == p->header.maxentry)
473 nsplit++;
474 else
475 nsplit = 0;
476
477 /* push (bn, index) of the parent page/entry */
478 if (BT_STACK_FULL(btstack)) {
479 jfs_error(ip->i_sb, "stack overrun!\n");
480 XT_PUTPAGE(mp);
481 return -EIO;
482 }
483 BT_PUSH(btstack, bn, index);
484
485 /* get the child page block number */
486 bn = addressXAD(&p->xad[index]);
487
488 /* unpin the parent page */
489 XT_PUTPAGE(mp);
490 }
491 }
492
493 /*
494 * xtInsert()
495 *
496 * function:
497 *
498 * parameter:
499 * tid - transaction id;
500 * ip - file object;
501 * xflag - extent flag (XAD_NOTRECORDED):
502 * xoff - extent offset;
503 * xlen - extent length;
504 * xaddrp - extent address pointer (in/out):
505 * if (*xaddrp)
506 * caller allocated data extent at *xaddrp;
507 * else
508 * allocate data extent and return its xaddr;
509 * flag -
510 *
511 * return:
512 */
513 int xtInsert(tid_t tid, /* transaction id */
514 struct inode *ip, int xflag, s64 xoff, s32 xlen, s64 * xaddrp,
515 int flag)
516 {
517 int rc = 0;
518 s64 xaddr, hint;
519 struct metapage *mp; /* meta-page buffer */
520 xtpage_t *p; /* base B+-tree index page */
521 s64 bn;
522 int index, nextindex;
523 struct btstack btstack; /* traverse stack */
524 struct xtsplit split; /* split information */
525 xad_t *xad;
526 int cmp;
527 s64 next;
528 struct tlock *tlck;
529 struct xtlock *xtlck;
530
531 jfs_info("xtInsert: nxoff:0x%lx nxlen:0x%x", (ulong) xoff, xlen);
532
533 /*
534 * search for the entry location at which to insert:
535 *
536 * xtFastSearch() and xtSearch() both returns (leaf page
537 * pinned, index at which to insert).
538 * n.b. xtSearch() may return index of maxentry of
539 * the full page.
540 */
541 if ((rc = xtSearch(ip, xoff, &next, &cmp, &btstack, XT_INSERT)))
542 return rc;
543
544 /* retrieve search result */
545 XT_GETSEARCH(ip, btstack.top, bn, mp, p, index);
546
547 /* This test must follow XT_GETSEARCH since mp must be valid if
548 * we branch to out: */
549 if ((cmp == 0) || (next && (xlen > next - xoff))) {
550 rc = -EEXIST;
551 goto out;
552 }
553
554 /*
555 * allocate data extent requested
556 *
557 * allocation hint: last xad
558 */
559 if ((xaddr = *xaddrp) == 0) {
560 if (index > XTENTRYSTART) {
561 xad = &p->xad[index - 1];
562 hint = addressXAD(xad) + lengthXAD(xad) - 1;
563 } else
564 hint = 0;
565 if ((rc = dquot_alloc_block(ip, xlen)))
566 goto out;
567 if ((rc = dbAlloc(ip, hint, (s64) xlen, &xaddr))) {
568 dquot_free_block(ip, xlen);
569 goto out;
570 }
571 }
572
573 /*
574 * insert entry for new extent
575 */
576 xflag |= XAD_NEW;
577
578 /*
579 * if the leaf page is full, split the page and
580 * propagate up the router entry for the new page from split
581 *
582 * The xtSplitUp() will insert the entry and unpin the leaf page.
583 */
584 nextindex = le16_to_cpu(p->header.nextindex);
585 if (nextindex == le16_to_cpu(p->header.maxentry)) {
586 split.mp = mp;
587 split.index = index;
588 split.flag = xflag;
589 split.off = xoff;
590 split.len = xlen;
591 split.addr = xaddr;
592 split.pxdlist = NULL;
593 if ((rc = xtSplitUp(tid, ip, &split, &btstack))) {
594 /* undo data extent allocation */
595 if (*xaddrp == 0) {
596 dbFree(ip, xaddr, (s64) xlen);
597 dquot_free_block(ip, xlen);
598 }
599 return rc;
600 }
601
602 *xaddrp = xaddr;
603 return 0;
604 }
605
606 /*
607 * insert the new entry into the leaf page
608 */
609 /*
610 * acquire a transaction lock on the leaf page;
611 *
612 * action: xad insertion/extension;
613 */
614 BT_MARK_DIRTY(mp, ip);
615
616 /* if insert into middle, shift right remaining entries. */
617 if (index < nextindex)
618 memmove(&p->xad[index + 1], &p->xad[index],
619 (nextindex - index) * sizeof(xad_t));
620
621 /* insert the new entry: mark the entry NEW */
622 xad = &p->xad[index];
623 XT_PUTENTRY(xad, xflag, xoff, xlen, xaddr);
624
625 /* advance next available entry index */
626 le16_add_cpu(&p->header.nextindex, 1);
627
628 /* Don't log it if there are no links to the file */
629 if (!test_cflag(COMMIT_Nolink, ip)) {
630 tlck = txLock(tid, ip, mp, tlckXTREE | tlckGROW);
631 xtlck = (struct xtlock *) & tlck->lock;
632 xtlck->lwm.offset =
633 (xtlck->lwm.offset) ? min(index,
634 (int)xtlck->lwm.offset) : index;
635 xtlck->lwm.length =
636 le16_to_cpu(p->header.nextindex) - xtlck->lwm.offset;
637 }
638
639 *xaddrp = xaddr;
640
641 out:
642 /* unpin the leaf page */
643 XT_PUTPAGE(mp);
644
645 return rc;
646 }
647
648
649 /*
650 * xtSplitUp()
651 *
652 * function:
653 * split full pages as propagating insertion up the tree
654 *
655 * parameter:
656 * tid - transaction id;
657 * ip - file object;
658 * split - entry parameter descriptor;
659 * btstack - traverse stack from xtSearch()
660 *
661 * return:
662 */
663 static int
664 xtSplitUp(tid_t tid,
665 struct inode *ip, struct xtsplit * split, struct btstack * btstack)
666 {
667 int rc = 0;
668 struct metapage *smp;
669 xtpage_t *sp; /* split page */
670 struct metapage *rmp;
671 s64 rbn; /* new right page block number */
672 struct metapage *rcmp;
673 xtpage_t *rcp; /* right child page */
674 s64 rcbn; /* right child page block number */
675 int skip; /* index of entry of insertion */
676 int nextindex; /* next available entry index of p */
677 struct btframe *parent; /* parent page entry on traverse stack */
678 xad_t *xad;
679 s64 xaddr;
680 int xlen;
681 int nsplit; /* number of pages split */
682 struct pxdlist pxdlist;
683 pxd_t *pxd;
684 struct tlock *tlck;
685 struct xtlock *xtlck;
686
687 smp = split->mp;
688 sp = XT_PAGE(ip, smp);
689
690 /* is inode xtree root extension/inline EA area free ? */
691 if ((sp->header.flag & BT_ROOT) && (!S_ISDIR(ip->i_mode)) &&
692 (le16_to_cpu(sp->header.maxentry) < XTROOTMAXSLOT) &&
693 (JFS_IP(ip)->mode2 & INLINEEA)) {
694 sp->header.maxentry = cpu_to_le16(XTROOTMAXSLOT);
695 JFS_IP(ip)->mode2 &= ~INLINEEA;
696
697 BT_MARK_DIRTY(smp, ip);
698 /*
699 * acquire a transaction lock on the leaf page;
700 *
701 * action: xad insertion/extension;
702 */
703
704 /* if insert into middle, shift right remaining entries. */
705 skip = split->index;
706 nextindex = le16_to_cpu(sp->header.nextindex);
707 if (skip < nextindex)
708 memmove(&sp->xad[skip + 1], &sp->xad[skip],
709 (nextindex - skip) * sizeof(xad_t));
710
711 /* insert the new entry: mark the entry NEW */
712 xad = &sp->xad[skip];
713 XT_PUTENTRY(xad, split->flag, split->off, split->len,
714 split->addr);
715
716 /* advance next available entry index */
717 le16_add_cpu(&sp->header.nextindex, 1);
718
719 /* Don't log it if there are no links to the file */
720 if (!test_cflag(COMMIT_Nolink, ip)) {
721 tlck = txLock(tid, ip, smp, tlckXTREE | tlckGROW);
722 xtlck = (struct xtlock *) & tlck->lock;
723 xtlck->lwm.offset = (xtlck->lwm.offset) ?
724 min(skip, (int)xtlck->lwm.offset) : skip;
725 xtlck->lwm.length =
726 le16_to_cpu(sp->header.nextindex) -
727 xtlck->lwm.offset;
728 }
729
730 return 0;
731 }
732
733 /*
734 * allocate new index blocks to cover index page split(s)
735 *
736 * allocation hint: ?
737 */
738 if (split->pxdlist == NULL) {
739 nsplit = btstack->nsplit;
740 split->pxdlist = &pxdlist;
741 pxdlist.maxnpxd = pxdlist.npxd = 0;
742 pxd = &pxdlist.pxd[0];
743 xlen = JFS_SBI(ip->i_sb)->nbperpage;
744 for (; nsplit > 0; nsplit--, pxd++) {
745 if ((rc = dbAlloc(ip, (s64) 0, (s64) xlen, &xaddr))
746 == 0) {
747 PXDaddress(pxd, xaddr);
748 PXDlength(pxd, xlen);
749
750 pxdlist.maxnpxd++;
751
752 continue;
753 }
754
755 /* undo allocation */
756
757 XT_PUTPAGE(smp);
758 return rc;
759 }
760 }
761
762 /*
763 * Split leaf page <sp> into <sp> and a new right page <rp>.
764 *
765 * The split routines insert the new entry into the leaf page,
766 * and acquire txLock as appropriate.
767 * return <rp> pinned and its block number <rpbn>.
768 */
769 rc = (sp->header.flag & BT_ROOT) ?
770 xtSplitRoot(tid, ip, split, &rmp) :
771 xtSplitPage(tid, ip, split, &rmp, &rbn);
772
773 XT_PUTPAGE(smp);
774
775 if (rc)
776 return -EIO;
777 /*
778 * propagate up the router entry for the leaf page just split
779 *
780 * insert a router entry for the new page into the parent page,
781 * propagate the insert/split up the tree by walking back the stack
782 * of (bn of parent page, index of child page entry in parent page)
783 * that were traversed during the search for the page that split.
784 *
785 * the propagation of insert/split up the tree stops if the root
786 * splits or the page inserted into doesn't have to split to hold
787 * the new entry.
788 *
789 * the parent entry for the split page remains the same, and
790 * a new entry is inserted at its right with the first key and
791 * block number of the new right page.
792 *
793 * There are a maximum of 3 pages pinned at any time:
794 * right child, left parent and right parent (when the parent splits)
795 * to keep the child page pinned while working on the parent.
796 * make sure that all pins are released at exit.
797 */
798 while ((parent = BT_POP(btstack)) != NULL) {
799 /* parent page specified by stack frame <parent> */
800
801 /* keep current child pages <rcp> pinned */
802 rcmp = rmp;
803 rcbn = rbn;
804 rcp = XT_PAGE(ip, rcmp);
805
806 /*
807 * insert router entry in parent for new right child page <rp>
808 */
809 /* get/pin the parent page <sp> */
810 XT_GETPAGE(ip, parent->bn, smp, PSIZE, sp, rc);
811 if (rc) {
812 XT_PUTPAGE(rcmp);
813 return rc;
814 }
815
816 /*
817 * The new key entry goes ONE AFTER the index of parent entry,
818 * because the split was to the right.
819 */
820 skip = parent->index + 1;
821
822 /*
823 * split or shift right remaining entries of the parent page
824 */
825 nextindex = le16_to_cpu(sp->header.nextindex);
826 /*
827 * parent page is full - split the parent page
828 */
829 if (nextindex == le16_to_cpu(sp->header.maxentry)) {
830 /* init for parent page split */
831 split->mp = smp;
832 split->index = skip; /* index at insert */
833 split->flag = XAD_NEW;
834 split->off = offsetXAD(&rcp->xad[XTENTRYSTART]);
835 split->len = JFS_SBI(ip->i_sb)->nbperpage;
836 split->addr = rcbn;
837
838 /* unpin previous right child page */
839 XT_PUTPAGE(rcmp);
840
841 /* The split routines insert the new entry,
842 * and acquire txLock as appropriate.
843 * return <rp> pinned and its block number <rpbn>.
844 */
845 rc = (sp->header.flag & BT_ROOT) ?
846 xtSplitRoot(tid, ip, split, &rmp) :
847 xtSplitPage(tid, ip, split, &rmp, &rbn);
848 if (rc) {
849 XT_PUTPAGE(smp);
850 return rc;
851 }
852
853 XT_PUTPAGE(smp);
854 /* keep new child page <rp> pinned */
855 }
856 /*
857 * parent page is not full - insert in parent page
858 */
859 else {
860 /*
861 * insert router entry in parent for the right child
862 * page from the first entry of the right child page:
863 */
864 /*
865 * acquire a transaction lock on the parent page;
866 *
867 * action: router xad insertion;
868 */
869 BT_MARK_DIRTY(smp, ip);
870
871 /*
872 * if insert into middle, shift right remaining entries
873 */
874 if (skip < nextindex)
875 memmove(&sp->xad[skip + 1], &sp->xad[skip],
876 (nextindex -
877 skip) << L2XTSLOTSIZE);
878
879 /* insert the router entry */
880 xad = &sp->xad[skip];
881 XT_PUTENTRY(xad, XAD_NEW,
882 offsetXAD(&rcp->xad[XTENTRYSTART]),
883 JFS_SBI(ip->i_sb)->nbperpage, rcbn);
884
885 /* advance next available entry index. */
886 le16_add_cpu(&sp->header.nextindex, 1);
887
888 /* Don't log it if there are no links to the file */
889 if (!test_cflag(COMMIT_Nolink, ip)) {
890 tlck = txLock(tid, ip, smp,
891 tlckXTREE | tlckGROW);
892 xtlck = (struct xtlock *) & tlck->lock;
893 xtlck->lwm.offset = (xtlck->lwm.offset) ?
894 min(skip, (int)xtlck->lwm.offset) : skip;
895 xtlck->lwm.length =
896 le16_to_cpu(sp->header.nextindex) -
897 xtlck->lwm.offset;
898 }
899
900 /* unpin parent page */
901 XT_PUTPAGE(smp);
902
903 /* exit propagate up */
904 break;
905 }
906 }
907
908 /* unpin current right page */
909 XT_PUTPAGE(rmp);
910
911 return 0;
912 }
913
914
915 /*
916 * xtSplitPage()
917 *
918 * function:
919 * split a full non-root page into
920 * original/split/left page and new right page
921 * i.e., the original/split page remains as left page.
922 *
923 * parameter:
924 * int tid,
925 * struct inode *ip,
926 * struct xtsplit *split,
927 * struct metapage **rmpp,
928 * u64 *rbnp,
929 *
930 * return:
931 * Pointer to page in which to insert or NULL on error.
932 */
933 static int
934 xtSplitPage(tid_t tid, struct inode *ip,
935 struct xtsplit * split, struct metapage ** rmpp, s64 * rbnp)
936 {
937 int rc = 0;
938 struct metapage *smp;
939 xtpage_t *sp;
940 struct metapage *rmp;
941 xtpage_t *rp; /* new right page allocated */
942 s64 rbn; /* new right page block number */
943 struct metapage *mp;
944 xtpage_t *p;
945 s64 nextbn;
946 int skip, maxentry, middle, righthalf, n;
947 xad_t *xad;
948 struct pxdlist *pxdlist;
949 pxd_t *pxd;
950 struct tlock *tlck;
951 struct xtlock *sxtlck = NULL, *rxtlck = NULL;
952 int quota_allocation = 0;
953
954 smp = split->mp;
955 sp = XT_PAGE(ip, smp);
956
957 INCREMENT(xtStat.split);
958
959 pxdlist = split->pxdlist;
960 pxd = &pxdlist->pxd[pxdlist->npxd];
961 pxdlist->npxd++;
962 rbn = addressPXD(pxd);
963
964 /* Allocate blocks to quota. */
965 rc = dquot_alloc_block(ip, lengthPXD(pxd));
966 if (rc)
967 goto clean_up;
968
969 quota_allocation += lengthPXD(pxd);
970
971 /*
972 * allocate the new right page for the split
973 */
974 rmp = get_metapage(ip, rbn, PSIZE, 1);
975 if (rmp == NULL) {
976 rc = -EIO;
977 goto clean_up;
978 }
979
980 jfs_info("xtSplitPage: ip:0x%p smp:0x%p rmp:0x%p", ip, smp, rmp);
981
982 BT_MARK_DIRTY(rmp, ip);
983 /*
984 * action: new page;
985 */
986
987 rp = (xtpage_t *) rmp->data;
988 rp->header.self = *pxd;
989 rp->header.flag = sp->header.flag & BT_TYPE;
990 rp->header.maxentry = sp->header.maxentry; /* little-endian */
991 rp->header.nextindex = cpu_to_le16(XTENTRYSTART);
992
993 BT_MARK_DIRTY(smp, ip);
994 /* Don't log it if there are no links to the file */
995 if (!test_cflag(COMMIT_Nolink, ip)) {
996 /*
997 * acquire a transaction lock on the new right page;
998 */
999 tlck = txLock(tid, ip, rmp, tlckXTREE | tlckNEW);
1000 rxtlck = (struct xtlock *) & tlck->lock;
1001 rxtlck->lwm.offset = XTENTRYSTART;
1002 /*
1003 * acquire a transaction lock on the split page
1004 */
1005 tlck = txLock(tid, ip, smp, tlckXTREE | tlckGROW);
1006 sxtlck = (struct xtlock *) & tlck->lock;
1007 }
1008
1009 /*
1010 * initialize/update sibling pointers of <sp> and <rp>
1011 */
1012 nextbn = le64_to_cpu(sp->header.next);
1013 rp->header.next = cpu_to_le64(nextbn);
1014 rp->header.prev = cpu_to_le64(addressPXD(&sp->header.self));
1015 sp->header.next = cpu_to_le64(rbn);
1016
1017 skip = split->index;
1018
1019 /*
1020 * sequential append at tail (after last entry of last page)
1021 *
1022 * if splitting the last page on a level because of appending
1023 * a entry to it (skip is maxentry), it's likely that the access is
1024 * sequential. adding an empty page on the side of the level is less
1025 * work and can push the fill factor much higher than normal.
1026 * if we're wrong it's no big deal - we will do the split the right
1027 * way next time.
1028 * (it may look like it's equally easy to do a similar hack for
1029 * reverse sorted data, that is, split the tree left, but it's not.
1030 * Be my guest.)
1031 */
1032 if (nextbn == 0 && skip == le16_to_cpu(sp->header.maxentry)) {
1033 /*
1034 * acquire a transaction lock on the new/right page;
1035 *
1036 * action: xad insertion;
1037 */
1038 /* insert entry at the first entry of the new right page */
1039 xad = &rp->xad[XTENTRYSTART];
1040 XT_PUTENTRY(xad, split->flag, split->off, split->len,
1041 split->addr);
1042
1043 rp->header.nextindex = cpu_to_le16(XTENTRYSTART + 1);
1044
1045 if (!test_cflag(COMMIT_Nolink, ip)) {
1046 /* rxtlck->lwm.offset = XTENTRYSTART; */
1047 rxtlck->lwm.length = 1;
1048 }
1049
1050 *rmpp = rmp;
1051 *rbnp = rbn;
1052
1053 jfs_info("xtSplitPage: sp:0x%p rp:0x%p", sp, rp);
1054 return 0;
1055 }
1056
1057 /*
1058 * non-sequential insert (at possibly middle page)
1059 */
1060
1061 /*
1062 * update previous pointer of old next/right page of <sp>
1063 */
1064 if (nextbn != 0) {
1065 XT_GETPAGE(ip, nextbn, mp, PSIZE, p, rc);
1066 if (rc) {
1067 XT_PUTPAGE(rmp);
1068 goto clean_up;
1069 }
1070
1071 BT_MARK_DIRTY(mp, ip);
1072 /*
1073 * acquire a transaction lock on the next page;
1074 *
1075 * action:sibling pointer update;
1076 */
1077 if (!test_cflag(COMMIT_Nolink, ip))
1078 tlck = txLock(tid, ip, mp, tlckXTREE | tlckRELINK);
1079
1080 p->header.prev = cpu_to_le64(rbn);
1081
1082 /* sibling page may have been updated previously, or
1083 * it may be updated later;
1084 */
1085
1086 XT_PUTPAGE(mp);
1087 }
1088
1089 /*
1090 * split the data between the split and new/right pages
1091 */
1092 maxentry = le16_to_cpu(sp->header.maxentry);
1093 middle = maxentry >> 1;
1094 righthalf = maxentry - middle;
1095
1096 /*
1097 * skip index in old split/left page - insert into left page:
1098 */
1099 if (skip <= middle) {
1100 /* move right half of split page to the new right page */
1101 memmove(&rp->xad[XTENTRYSTART], &sp->xad[middle],
1102 righthalf << L2XTSLOTSIZE);
1103
1104 /* shift right tail of left half to make room for new entry */
1105 if (skip < middle)
1106 memmove(&sp->xad[skip + 1], &sp->xad[skip],
1107 (middle - skip) << L2XTSLOTSIZE);
1108
1109 /* insert new entry */
1110 xad = &sp->xad[skip];
1111 XT_PUTENTRY(xad, split->flag, split->off, split->len,
1112 split->addr);
1113
1114 /* update page header */
1115 sp->header.nextindex = cpu_to_le16(middle + 1);
1116 if (!test_cflag(COMMIT_Nolink, ip)) {
1117 sxtlck->lwm.offset = (sxtlck->lwm.offset) ?
1118 min(skip, (int)sxtlck->lwm.offset) : skip;
1119 }
1120
1121 rp->header.nextindex =
1122 cpu_to_le16(XTENTRYSTART + righthalf);
1123 }
1124 /*
1125 * skip index in new right page - insert into right page:
1126 */
1127 else {
1128 /* move left head of right half to right page */
1129 n = skip - middle;
1130 memmove(&rp->xad[XTENTRYSTART], &sp->xad[middle],
1131 n << L2XTSLOTSIZE);
1132
1133 /* insert new entry */
1134 n += XTENTRYSTART;
1135 xad = &rp->xad[n];
1136 XT_PUTENTRY(xad, split->flag, split->off, split->len,
1137 split->addr);
1138
1139 /* move right tail of right half to right page */
1140 if (skip < maxentry)
1141 memmove(&rp->xad[n + 1], &sp->xad[skip],
1142 (maxentry - skip) << L2XTSLOTSIZE);
1143
1144 /* update page header */
1145 sp->header.nextindex = cpu_to_le16(middle);
1146 if (!test_cflag(COMMIT_Nolink, ip)) {
1147 sxtlck->lwm.offset = (sxtlck->lwm.offset) ?
1148 min(middle, (int)sxtlck->lwm.offset) : middle;
1149 }
1150
1151 rp->header.nextindex = cpu_to_le16(XTENTRYSTART +
1152 righthalf + 1);
1153 }
1154
1155 if (!test_cflag(COMMIT_Nolink, ip)) {
1156 sxtlck->lwm.length = le16_to_cpu(sp->header.nextindex) -
1157 sxtlck->lwm.offset;
1158
1159 /* rxtlck->lwm.offset = XTENTRYSTART; */
1160 rxtlck->lwm.length = le16_to_cpu(rp->header.nextindex) -
1161 XTENTRYSTART;
1162 }
1163
1164 *rmpp = rmp;
1165 *rbnp = rbn;
1166
1167 jfs_info("xtSplitPage: sp:0x%p rp:0x%p", sp, rp);
1168 return rc;
1169
1170 clean_up:
1171
1172 /* Rollback quota allocation. */
1173 if (quota_allocation)
1174 dquot_free_block(ip, quota_allocation);
1175
1176 return (rc);
1177 }
1178
1179
1180 /*
1181 * xtSplitRoot()
1182 *
1183 * function:
1184 * split the full root page into original/root/split page and new
1185 * right page
1186 * i.e., root remains fixed in tree anchor (inode) and the root is
1187 * copied to a single new right child page since root page <<
1188 * non-root page, and the split root page contains a single entry
1189 * for the new right child page.
1190 *
1191 * parameter:
1192 * int tid,
1193 * struct inode *ip,
1194 * struct xtsplit *split,
1195 * struct metapage **rmpp)
1196 *
1197 * return:
1198 * Pointer to page in which to insert or NULL on error.
1199 */
1200 static int
1201 xtSplitRoot(tid_t tid,
1202 struct inode *ip, struct xtsplit * split, struct metapage ** rmpp)
1203 {
1204 xtpage_t *sp;
1205 struct metapage *rmp;
1206 xtpage_t *rp;
1207 s64 rbn;
1208 int skip, nextindex;
1209 xad_t *xad;
1210 pxd_t *pxd;
1211 struct pxdlist *pxdlist;
1212 struct tlock *tlck;
1213 struct xtlock *xtlck;
1214 int rc;
1215
1216 sp = (xtpage_t *) &JFS_IP(ip)->i_xtroot;
1217
1218 INCREMENT(xtStat.split);
1219
1220 /*
1221 * allocate a single (right) child page
1222 */
1223 pxdlist = split->pxdlist;
1224 pxd = &pxdlist->pxd[pxdlist->npxd];
1225 pxdlist->npxd++;
1226 rbn = addressPXD(pxd);
1227 rmp = get_metapage(ip, rbn, PSIZE, 1);
1228 if (rmp == NULL)
1229 return -EIO;
1230
1231 /* Allocate blocks to quota. */
1232 rc = dquot_alloc_block(ip, lengthPXD(pxd));
1233 if (rc) {
1234 release_metapage(rmp);
1235 return rc;
1236 }
1237
1238 jfs_info("xtSplitRoot: ip:0x%p rmp:0x%p", ip, rmp);
1239
1240 /*
1241 * acquire a transaction lock on the new right page;
1242 *
1243 * action: new page;
1244 */
1245 BT_MARK_DIRTY(rmp, ip);
1246
1247 rp = (xtpage_t *) rmp->data;
1248 rp->header.flag =
1249 (sp->header.flag & BT_LEAF) ? BT_LEAF : BT_INTERNAL;
1250 rp->header.self = *pxd;
1251 rp->header.nextindex = cpu_to_le16(XTENTRYSTART);
1252 rp->header.maxentry = cpu_to_le16(PSIZE >> L2XTSLOTSIZE);
1253
1254 /* initialize sibling pointers */
1255 rp->header.next = 0;
1256 rp->header.prev = 0;
1257
1258 /*
1259 * copy the in-line root page into new right page extent
1260 */
1261 nextindex = le16_to_cpu(sp->header.maxentry);
1262 memmove(&rp->xad[XTENTRYSTART], &sp->xad[XTENTRYSTART],
1263 (nextindex - XTENTRYSTART) << L2XTSLOTSIZE);
1264
1265 /*
1266 * insert the new entry into the new right/child page
1267 * (skip index in the new right page will not change)
1268 */
1269 skip = split->index;
1270 /* if insert into middle, shift right remaining entries */
1271 if (skip != nextindex)
1272 memmove(&rp->xad[skip + 1], &rp->xad[skip],
1273 (nextindex - skip) * sizeof(xad_t));
1274
1275 xad = &rp->xad[skip];
1276 XT_PUTENTRY(xad, split->flag, split->off, split->len, split->addr);
1277
1278 /* update page header */
1279 rp->header.nextindex = cpu_to_le16(nextindex + 1);
1280
1281 if (!test_cflag(COMMIT_Nolink, ip)) {
1282 tlck = txLock(tid, ip, rmp, tlckXTREE | tlckNEW);
1283 xtlck = (struct xtlock *) & tlck->lock;
1284 xtlck->lwm.offset = XTENTRYSTART;
1285 xtlck->lwm.length = le16_to_cpu(rp->header.nextindex) -
1286 XTENTRYSTART;
1287 }
1288
1289 /*
1290 * reset the root
1291 *
1292 * init root with the single entry for the new right page
1293 * set the 1st entry offset to 0, which force the left-most key
1294 * at any level of the tree to be less than any search key.
1295 */
1296 /*
1297 * acquire a transaction lock on the root page (in-memory inode);
1298 *
1299 * action: root split;
1300 */
1301 BT_MARK_DIRTY(split->mp, ip);
1302
1303 xad = &sp->xad[XTENTRYSTART];
1304 XT_PUTENTRY(xad, XAD_NEW, 0, JFS_SBI(ip->i_sb)->nbperpage, rbn);
1305
1306 /* update page header of root */
1307 sp->header.flag &= ~BT_LEAF;
1308 sp->header.flag |= BT_INTERNAL;
1309
1310 sp->header.nextindex = cpu_to_le16(XTENTRYSTART + 1);
1311
1312 if (!test_cflag(COMMIT_Nolink, ip)) {
1313 tlck = txLock(tid, ip, split->mp, tlckXTREE | tlckGROW);
1314 xtlck = (struct xtlock *) & tlck->lock;
1315 xtlck->lwm.offset = XTENTRYSTART;
1316 xtlck->lwm.length = 1;
1317 }
1318
1319 *rmpp = rmp;
1320
1321 jfs_info("xtSplitRoot: sp:0x%p rp:0x%p", sp, rp);
1322 return 0;
1323 }
1324
1325
1326 /*
1327 * xtExtend()
1328 *
1329 * function: extend in-place;
1330 *
1331 * note: existing extent may or may not have been committed.
1332 * caller is responsible for pager buffer cache update, and
1333 * working block allocation map update;
1334 * update pmap: alloc whole extended extent;
1335 */
1336 int xtExtend(tid_t tid, /* transaction id */
1337 struct inode *ip, s64 xoff, /* delta extent offset */
1338 s32 xlen, /* delta extent length */
1339 int flag)
1340 {
1341 int rc = 0;
1342 int cmp;
1343 struct metapage *mp; /* meta-page buffer */
1344 xtpage_t *p; /* base B+-tree index page */
1345 s64 bn;
1346 int index, nextindex, len;
1347 struct btstack btstack; /* traverse stack */
1348 struct xtsplit split; /* split information */
1349 xad_t *xad;
1350 s64 xaddr;
1351 struct tlock *tlck;
1352 struct xtlock *xtlck = NULL;
1353
1354 jfs_info("xtExtend: nxoff:0x%lx nxlen:0x%x", (ulong) xoff, xlen);
1355
1356 /* there must exist extent to be extended */
1357 if ((rc = xtSearch(ip, xoff - 1, NULL, &cmp, &btstack, XT_INSERT)))
1358 return rc;
1359
1360 /* retrieve search result */
1361 XT_GETSEARCH(ip, btstack.top, bn, mp, p, index);
1362
1363 if (cmp != 0) {
1364 XT_PUTPAGE(mp);
1365 jfs_error(ip->i_sb, "xtSearch did not find extent\n");
1366 return -EIO;
1367 }
1368
1369 /* extension must be contiguous */
1370 xad = &p->xad[index];
1371 if ((offsetXAD(xad) + lengthXAD(xad)) != xoff) {
1372 XT_PUTPAGE(mp);
1373 jfs_error(ip->i_sb, "extension is not contiguous\n");
1374 return -EIO;
1375 }
1376
1377 /*
1378 * acquire a transaction lock on the leaf page;
1379 *
1380 * action: xad insertion/extension;
1381 */
1382 BT_MARK_DIRTY(mp, ip);
1383 if (!test_cflag(COMMIT_Nolink, ip)) {
1384 tlck = txLock(tid, ip, mp, tlckXTREE | tlckGROW);
1385 xtlck = (struct xtlock *) & tlck->lock;
1386 }
1387
1388 /* extend will overflow extent ? */
1389 xlen = lengthXAD(xad) + xlen;
1390 if ((len = xlen - MAXXLEN) <= 0)
1391 goto extendOld;
1392
1393 /*
1394 * extent overflow: insert entry for new extent
1395 */
1396 //insertNew:
1397 xoff = offsetXAD(xad) + MAXXLEN;
1398 xaddr = addressXAD(xad) + MAXXLEN;
1399 nextindex = le16_to_cpu(p->header.nextindex);
1400
1401 /*
1402 * if the leaf page is full, insert the new entry and
1403 * propagate up the router entry for the new page from split
1404 *
1405 * The xtSplitUp() will insert the entry and unpin the leaf page.
1406 */
1407 if (nextindex == le16_to_cpu(p->header.maxentry)) {
1408 /* xtSpliUp() unpins leaf pages */
1409 split.mp = mp;
1410 split.index = index + 1;
1411 split.flag = XAD_NEW;
1412 split.off = xoff; /* split offset */
1413 split.len = len;
1414 split.addr = xaddr;
1415 split.pxdlist = NULL;
1416 if ((rc = xtSplitUp(tid, ip, &split, &btstack)))
1417 return rc;
1418
1419 /* get back old page */
1420 XT_GETPAGE(ip, bn, mp, PSIZE, p, rc);
1421 if (rc)
1422 return rc;
1423 /*
1424 * if leaf root has been split, original root has been
1425 * copied to new child page, i.e., original entry now
1426 * resides on the new child page;
1427 */
1428 if (p->header.flag & BT_INTERNAL) {
1429 ASSERT(p->header.nextindex ==
1430 cpu_to_le16(XTENTRYSTART + 1));
1431 xad = &p->xad[XTENTRYSTART];
1432 bn = addressXAD(xad);
1433 XT_PUTPAGE(mp);
1434
1435 /* get new child page */
1436 XT_GETPAGE(ip, bn, mp, PSIZE, p, rc);
1437 if (rc)
1438 return rc;
1439
1440 BT_MARK_DIRTY(mp, ip);
1441 if (!test_cflag(COMMIT_Nolink, ip)) {
1442 tlck = txLock(tid, ip, mp, tlckXTREE|tlckGROW);
1443 xtlck = (struct xtlock *) & tlck->lock;
1444 }
1445 }
1446 }
1447 /*
1448 * insert the new entry into the leaf page
1449 */
1450 else {
1451 /* insert the new entry: mark the entry NEW */
1452 xad = &p->xad[index + 1];
1453 XT_PUTENTRY(xad, XAD_NEW, xoff, len, xaddr);
1454
1455 /* advance next available entry index */
1456 le16_add_cpu(&p->header.nextindex, 1);
1457 }
1458
1459 /* get back old entry */
1460 xad = &p->xad[index];
1461 xlen = MAXXLEN;
1462
1463 /*
1464 * extend old extent
1465 */
1466 extendOld:
1467 XADlength(xad, xlen);
1468 if (!(xad->flag & XAD_NEW))
1469 xad->flag |= XAD_EXTENDED;
1470
1471 if (!test_cflag(COMMIT_Nolink, ip)) {
1472 xtlck->lwm.offset =
1473 (xtlck->lwm.offset) ? min(index,
1474 (int)xtlck->lwm.offset) : index;
1475 xtlck->lwm.length =
1476 le16_to_cpu(p->header.nextindex) - xtlck->lwm.offset;
1477 }
1478
1479 /* unpin the leaf page */
1480 XT_PUTPAGE(mp);
1481
1482 return rc;
1483 }
1484
1485 /*
1486 * xtUpdate()
1487 *
1488 * function: update XAD;
1489 *
1490 * update extent for allocated_but_not_recorded or
1491 * compressed extent;
1492 *
1493 * parameter:
1494 * nxad - new XAD;
1495 * logical extent of the specified XAD must be completely
1496 * contained by an existing XAD;
1497 */
1498 int xtUpdate(tid_t tid, struct inode *ip, xad_t * nxad)
1499 { /* new XAD */
1500 int rc = 0;
1501 int cmp;
1502 struct metapage *mp; /* meta-page buffer */
1503 xtpage_t *p; /* base B+-tree index page */
1504 s64 bn;
1505 int index0, index, newindex, nextindex;
1506 struct btstack btstack; /* traverse stack */
1507 struct xtsplit split; /* split information */
1508 xad_t *xad, *lxad, *rxad;
1509 int xflag;
1510 s64 nxoff, xoff;
1511 int nxlen, xlen, lxlen, rxlen;
1512 s64 nxaddr, xaddr;
1513 struct tlock *tlck;
1514 struct xtlock *xtlck = NULL;
1515 int newpage = 0;
1516
1517 /* there must exist extent to be tailgated */
1518 nxoff = offsetXAD(nxad);
1519 nxlen = lengthXAD(nxad);
1520 nxaddr = addressXAD(nxad);
1521
1522 if ((rc = xtSearch(ip, nxoff, NULL, &cmp, &btstack, XT_INSERT)))
1523 return rc;
1524
1525 /* retrieve search result */
1526 XT_GETSEARCH(ip, btstack.top, bn, mp, p, index0);
1527
1528 if (cmp != 0) {
1529 XT_PUTPAGE(mp);
1530 jfs_error(ip->i_sb, "Could not find extent\n");
1531 return -EIO;
1532 }
1533
1534 BT_MARK_DIRTY(mp, ip);
1535 /*
1536 * acquire tlock of the leaf page containing original entry
1537 */
1538 if (!test_cflag(COMMIT_Nolink, ip)) {
1539 tlck = txLock(tid, ip, mp, tlckXTREE | tlckGROW);
1540 xtlck = (struct xtlock *) & tlck->lock;
1541 }
1542
1543 xad = &p->xad[index0];
1544 xflag = xad->flag;
1545 xoff = offsetXAD(xad);
1546 xlen = lengthXAD(xad);
1547 xaddr = addressXAD(xad);
1548
1549 /* nXAD must be completely contained within XAD */
1550 if ((xoff > nxoff) ||
1551 (nxoff + nxlen > xoff + xlen)) {
1552 XT_PUTPAGE(mp);
1553 jfs_error(ip->i_sb,
1554 "nXAD in not completely contained within XAD\n");
1555 return -EIO;
1556 }
1557
1558 index = index0;
1559 newindex = index + 1;
1560 nextindex = le16_to_cpu(p->header.nextindex);
1561
1562 if (xoff < nxoff)
1563 goto coalesceRight;
1564
1565 /*
1566 * coalesce with left XAD
1567 */
1568 /* is XAD first entry of page ? */
1569 if (index == XTENTRYSTART)
1570 goto replace;
1571
1572 /* is nXAD logically and physically contiguous with lXAD ? */
1573 lxad = &p->xad[index - 1];
1574 lxlen = lengthXAD(lxad);
1575 if (!(lxad->flag & XAD_NOTRECORDED) &&
1576 (nxoff == offsetXAD(lxad) + lxlen) &&
1577 (nxaddr == addressXAD(lxad) + lxlen) &&
1578 (lxlen + nxlen < MAXXLEN)) {
1579 /* extend right lXAD */
1580 index0 = index - 1;
1581 XADlength(lxad, lxlen + nxlen);
1582
1583 /* If we just merged two extents together, need to make sure the
1584 * right extent gets logged. If the left one is marked XAD_NEW,
1585 * then we know it will be logged. Otherwise, mark as
1586 * XAD_EXTENDED
1587 */
1588 if (!(lxad->flag & XAD_NEW))
1589 lxad->flag |= XAD_EXTENDED;
1590
1591 if (xlen > nxlen) {
1592 /* truncate XAD */
1593 XADoffset(xad, xoff + nxlen);
1594 XADlength(xad, xlen - nxlen);
1595 XADaddress(xad, xaddr + nxlen);
1596 goto out;
1597 } else { /* (xlen == nxlen) */
1598
1599 /* remove XAD */
1600 if (index < nextindex - 1)
1601 memmove(&p->xad[index], &p->xad[index + 1],
1602 (nextindex - index -
1603 1) << L2XTSLOTSIZE);
1604
1605 p->header.nextindex =
1606 cpu_to_le16(le16_to_cpu(p->header.nextindex) -
1607 1);
1608
1609 index = index0;
1610 newindex = index + 1;
1611 nextindex = le16_to_cpu(p->header.nextindex);
1612 xoff = nxoff = offsetXAD(lxad);
1613 xlen = nxlen = lxlen + nxlen;
1614 xaddr = nxaddr = addressXAD(lxad);
1615 goto coalesceRight;
1616 }
1617 }
1618
1619 /*
1620 * replace XAD with nXAD
1621 */
1622 replace: /* (nxoff == xoff) */
1623 if (nxlen == xlen) {
1624 /* replace XAD with nXAD:recorded */
1625 *xad = *nxad;
1626 xad->flag = xflag & ~XAD_NOTRECORDED;
1627
1628 goto coalesceRight;
1629 } else /* (nxlen < xlen) */
1630 goto updateLeft;
1631
1632 /*
1633 * coalesce with right XAD
1634 */
1635 coalesceRight: /* (xoff <= nxoff) */
1636 /* is XAD last entry of page ? */
1637 if (newindex == nextindex) {
1638 if (xoff == nxoff)
1639 goto out;
1640 goto updateRight;
1641 }
1642
1643 /* is nXAD logically and physically contiguous with rXAD ? */
1644 rxad = &p->xad[index + 1];
1645 rxlen = lengthXAD(rxad);
1646 if (!(rxad->flag & XAD_NOTRECORDED) &&
1647 (nxoff + nxlen == offsetXAD(rxad)) &&
1648 (nxaddr + nxlen == addressXAD(rxad)) &&
1649 (rxlen + nxlen < MAXXLEN)) {
1650 /* extend left rXAD */
1651 XADoffset(rxad, nxoff);
1652 XADlength(rxad, rxlen + nxlen);
1653 XADaddress(rxad, nxaddr);
1654
1655 /* If we just merged two extents together, need to make sure
1656 * the left extent gets logged. If the right one is marked
1657 * XAD_NEW, then we know it will be logged. Otherwise, mark as
1658 * XAD_EXTENDED
1659 */
1660 if (!(rxad->flag & XAD_NEW))
1661 rxad->flag |= XAD_EXTENDED;
1662
1663 if (xlen > nxlen)
1664 /* truncate XAD */
1665 XADlength(xad, xlen - nxlen);
1666 else { /* (xlen == nxlen) */
1667
1668 /* remove XAD */
1669 memmove(&p->xad[index], &p->xad[index + 1],
1670 (nextindex - index - 1) << L2XTSLOTSIZE);
1671
1672 p->header.nextindex =
1673 cpu_to_le16(le16_to_cpu(p->header.nextindex) -
1674 1);
1675 }
1676
1677 goto out;
1678 } else if (xoff == nxoff)
1679 goto out;
1680
1681 if (xoff >= nxoff) {
1682 XT_PUTPAGE(mp);
1683 jfs_error(ip->i_sb, "xoff >= nxoff\n");
1684 return -EIO;
1685 }
1686
1687 /*
1688 * split XAD into (lXAD, nXAD):
1689 *
1690 * |---nXAD--->
1691 * --|----------XAD----------|--
1692 * |-lXAD-|
1693 */
1694 updateRight: /* (xoff < nxoff) */
1695 /* truncate old XAD as lXAD:not_recorded */
1696 xad = &p->xad[index];
1697 XADlength(xad, nxoff - xoff);
1698
1699 /* insert nXAD:recorded */
1700 if (nextindex == le16_to_cpu(p->header.maxentry)) {
1701
1702 /* xtSpliUp() unpins leaf pages */
1703 split.mp = mp;
1704 split.index = newindex;
1705 split.flag = xflag & ~XAD_NOTRECORDED;
1706 split.off = nxoff;
1707 split.len = nxlen;
1708 split.addr = nxaddr;
1709 split.pxdlist = NULL;
1710 if ((rc = xtSplitUp(tid, ip, &split, &btstack)))
1711 return rc;
1712
1713 /* get back old page */
1714 XT_GETPAGE(ip, bn, mp, PSIZE, p, rc);
1715 if (rc)
1716 return rc;
1717 /*
1718 * if leaf root has been split, original root has been
1719 * copied to new child page, i.e., original entry now
1720 * resides on the new child page;
1721 */
1722 if (p->header.flag & BT_INTERNAL) {
1723 ASSERT(p->header.nextindex ==
1724 cpu_to_le16(XTENTRYSTART + 1));
1725 xad = &p->xad[XTENTRYSTART];
1726 bn = addressXAD(xad);
1727 XT_PUTPAGE(mp);
1728
1729 /* get new child page */
1730 XT_GETPAGE(ip, bn, mp, PSIZE, p, rc);
1731 if (rc)
1732 return rc;
1733
1734 BT_MARK_DIRTY(mp, ip);
1735 if (!test_cflag(COMMIT_Nolink, ip)) {
1736 tlck = txLock(tid, ip, mp, tlckXTREE|tlckGROW);
1737 xtlck = (struct xtlock *) & tlck->lock;
1738 }
1739 } else {
1740 /* is nXAD on new page ? */
1741 if (newindex >
1742 (le16_to_cpu(p->header.maxentry) >> 1)) {
1743 newindex =
1744 newindex -
1745 le16_to_cpu(p->header.nextindex) +
1746 XTENTRYSTART;
1747 newpage = 1;
1748 }
1749 }
1750 } else {
1751 /* if insert into middle, shift right remaining entries */
1752 if (newindex < nextindex)
1753 memmove(&p->xad[newindex + 1], &p->xad[newindex],
1754 (nextindex - newindex) << L2XTSLOTSIZE);
1755
1756 /* insert the entry */
1757 xad = &p->xad[newindex];
1758 *xad = *nxad;
1759 xad->flag = xflag & ~XAD_NOTRECORDED;
1760
1761 /* advance next available entry index. */
1762 p->header.nextindex =
1763 cpu_to_le16(le16_to_cpu(p->header.nextindex) + 1);
1764 }
1765
1766 /*
1767 * does nXAD force 3-way split ?
1768 *
1769 * |---nXAD--->|
1770 * --|----------XAD-------------|--
1771 * |-lXAD-| |-rXAD -|
1772 */
1773 if (nxoff + nxlen == xoff + xlen)
1774 goto out;
1775
1776 /* reorient nXAD as XAD for further split XAD into (nXAD, rXAD) */
1777 if (newpage) {
1778 /* close out old page */
1779 if (!test_cflag(COMMIT_Nolink, ip)) {
1780 xtlck->lwm.offset = (xtlck->lwm.offset) ?
1781 min(index0, (int)xtlck->lwm.offset) : index0;
1782 xtlck->lwm.length =
1783 le16_to_cpu(p->header.nextindex) -
1784 xtlck->lwm.offset;
1785 }
1786
1787 bn = le64_to_cpu(p->header.next);
1788 XT_PUTPAGE(mp);
1789
1790 /* get new right page */
1791 XT_GETPAGE(ip, bn, mp, PSIZE, p, rc);
1792 if (rc)
1793 return rc;
1794
1795 BT_MARK_DIRTY(mp, ip);
1796 if (!test_cflag(COMMIT_Nolink, ip)) {
1797 tlck = txLock(tid, ip, mp, tlckXTREE | tlckGROW);
1798 xtlck = (struct xtlock *) & tlck->lock;
1799 }
1800
1801 index0 = index = newindex;
1802 } else
1803 index++;
1804
1805 newindex = index + 1;
1806 nextindex = le16_to_cpu(p->header.nextindex);
1807 xlen = xlen - (nxoff - xoff);
1808 xoff = nxoff;
1809 xaddr = nxaddr;
1810
1811 /* recompute split pages */
1812 if (nextindex == le16_to_cpu(p->header.maxentry)) {
1813 XT_PUTPAGE(mp);
1814
1815 if ((rc = xtSearch(ip, nxoff, NULL, &cmp, &btstack, XT_INSERT)))
1816 return rc;
1817
1818 /* retrieve search result */
1819 XT_GETSEARCH(ip, btstack.top, bn, mp, p, index0);
1820
1821 if (cmp != 0) {
1822 XT_PUTPAGE(mp);
1823 jfs_error(ip->i_sb, "xtSearch failed\n");
1824 return -EIO;
1825 }
1826
1827 if (index0 != index) {
1828 XT_PUTPAGE(mp);
1829 jfs_error(ip->i_sb, "unexpected value of index\n");
1830 return -EIO;
1831 }
1832 }
1833
1834 /*
1835 * split XAD into (nXAD, rXAD)
1836 *
1837 * ---nXAD---|
1838 * --|----------XAD----------|--
1839 * |-rXAD-|
1840 */
1841 updateLeft: /* (nxoff == xoff) && (nxlen < xlen) */
1842 /* update old XAD with nXAD:recorded */
1843 xad = &p->xad[index];
1844 *xad = *nxad;
1845 xad->flag = xflag & ~XAD_NOTRECORDED;
1846
1847 /* insert rXAD:not_recorded */
1848 xoff = xoff + nxlen;
1849 xlen = xlen - nxlen;
1850 xaddr = xaddr + nxlen;
1851 if (nextindex == le16_to_cpu(p->header.maxentry)) {
1852 /*
1853 printf("xtUpdate.updateLeft.split p:0x%p\n", p);
1854 */
1855 /* xtSpliUp() unpins leaf pages */
1856 split.mp = mp;
1857 split.index = newindex;
1858 split.flag = xflag;
1859 split.off = xoff;
1860 split.len = xlen;
1861 split.addr = xaddr;
1862 split.pxdlist = NULL;
1863 if ((rc = xtSplitUp(tid, ip, &split, &btstack)))
1864 return rc;
1865
1866 /* get back old page */
1867 XT_GETPAGE(ip, bn, mp, PSIZE, p, rc);
1868 if (rc)
1869 return rc;
1870
1871 /*
1872 * if leaf root has been split, original root has been
1873 * copied to new child page, i.e., original entry now
1874 * resides on the new child page;
1875 */
1876 if (p->header.flag & BT_INTERNAL) {
1877 ASSERT(p->header.nextindex ==
1878 cpu_to_le16(XTENTRYSTART + 1));
1879 xad = &p->xad[XTENTRYSTART];
1880 bn = addressXAD(xad);
1881 XT_PUTPAGE(mp);
1882
1883 /* get new child page */
1884 XT_GETPAGE(ip, bn, mp, PSIZE, p, rc);
1885 if (rc)
1886 return rc;
1887
1888 BT_MARK_DIRTY(mp, ip);
1889 if (!test_cflag(COMMIT_Nolink, ip)) {
1890 tlck = txLock(tid, ip, mp, tlckXTREE|tlckGROW);
1891 xtlck = (struct xtlock *) & tlck->lock;
1892 }
1893 }
1894 } else {
1895 /* if insert into middle, shift right remaining entries */
1896 if (newindex < nextindex)
1897 memmove(&p->xad[newindex + 1], &p->xad[newindex],
1898 (nextindex - newindex) << L2XTSLOTSIZE);
1899
1900 /* insert the entry */
1901 xad = &p->xad[newindex];
1902 XT_PUTENTRY(xad, xflag, xoff, xlen, xaddr);
1903
1904 /* advance next available entry index. */
1905 p->header.nextindex =
1906 cpu_to_le16(le16_to_cpu(p->header.nextindex) + 1);
1907 }
1908
1909 out:
1910 if (!test_cflag(COMMIT_Nolink, ip)) {
1911 xtlck->lwm.offset = (xtlck->lwm.offset) ?
1912 min(index0, (int)xtlck->lwm.offset) : index0;
1913 xtlck->lwm.length = le16_to_cpu(p->header.nextindex) -
1914 xtlck->lwm.offset;
1915 }
1916
1917 /* unpin the leaf page */
1918 XT_PUTPAGE(mp);
1919
1920 return rc;
1921 }
1922
1923
1924 /*
1925 * xtAppend()
1926 *
1927 * function: grow in append mode from contiguous region specified ;
1928 *
1929 * parameter:
1930 * tid - transaction id;
1931 * ip - file object;
1932 * xflag - extent flag:
1933 * xoff - extent offset;
1934 * maxblocks - max extent length;
1935 * xlen - extent length (in/out);
1936 * xaddrp - extent address pointer (in/out):
1937 * flag -
1938 *
1939 * return:
1940 */
1941 int xtAppend(tid_t tid, /* transaction id */
1942 struct inode *ip, int xflag, s64 xoff, s32 maxblocks,
1943 s32 * xlenp, /* (in/out) */
1944 s64 * xaddrp, /* (in/out) */
1945 int flag)
1946 {
1947 int rc = 0;
1948 struct metapage *mp; /* meta-page buffer */
1949 xtpage_t *p; /* base B+-tree index page */
1950 s64 bn, xaddr;
1951 int index, nextindex;
1952 struct btstack btstack; /* traverse stack */
1953 struct xtsplit split; /* split information */
1954 xad_t *xad;
1955 int cmp;
1956 struct tlock *tlck;
1957 struct xtlock *xtlck;
1958 int nsplit, nblocks, xlen;
1959 struct pxdlist pxdlist;
1960 pxd_t *pxd;
1961 s64 next;
1962
1963 xaddr = *xaddrp;
1964 xlen = *xlenp;
1965 jfs_info("xtAppend: xoff:0x%lx maxblocks:%d xlen:%d xaddr:0x%lx",
1966 (ulong) xoff, maxblocks, xlen, (ulong) xaddr);
1967
1968 /*
1969 * search for the entry location at which to insert:
1970 *
1971 * xtFastSearch() and xtSearch() both returns (leaf page
1972 * pinned, index at which to insert).
1973 * n.b. xtSearch() may return index of maxentry of
1974 * the full page.
1975 */
1976 if ((rc = xtSearch(ip, xoff, &next, &cmp, &btstack, XT_INSERT)))
1977 return rc;
1978
1979 /* retrieve search result */
1980 XT_GETSEARCH(ip, btstack.top, bn, mp, p, index);
1981
1982 if (cmp == 0) {
1983 rc = -EEXIST;
1984 goto out;
1985 }
1986
1987 if (next)
1988 xlen = min(xlen, (int)(next - xoff));
1989 //insert:
1990 /*
1991 * insert entry for new extent
1992 */
1993 xflag |= XAD_NEW;
1994
1995 /*
1996 * if the leaf page is full, split the page and
1997 * propagate up the router entry for the new page from split
1998 *
1999 * The xtSplitUp() will insert the entry and unpin the leaf page.
2000 */
2001 nextindex = le16_to_cpu(p->header.nextindex);
2002 if (nextindex < le16_to_cpu(p->header.maxentry))
2003 goto insertLeaf;
2004
2005 /*
2006 * allocate new index blocks to cover index page split(s)
2007 */
2008 nsplit = btstack.nsplit;
2009 split.pxdlist = &pxdlist;
2010 pxdlist.maxnpxd = pxdlist.npxd = 0;
2011 pxd = &pxdlist.pxd[0];
2012 nblocks = JFS_SBI(ip->i_sb)->nbperpage;
2013 for (; nsplit > 0; nsplit--, pxd++, xaddr += nblocks, maxblocks -= nblocks) {
2014 if ((rc = dbAllocBottomUp(ip, xaddr, (s64) nblocks)) == 0) {
2015 PXDaddress(pxd, xaddr);
2016 PXDlength(pxd, nblocks);
2017
2018 pxdlist.maxnpxd++;
2019
2020 continue;
2021 }
2022
2023 /* undo allocation */
2024
2025 goto out;
2026 }
2027
2028 xlen = min(xlen, maxblocks);
2029
2030 /*
2031 * allocate data extent requested
2032 */
2033 if ((rc = dbAllocBottomUp(ip, xaddr, (s64) xlen)))
2034 goto out;
2035
2036 split.mp = mp;
2037 split.index = index;
2038 split.flag = xflag;
2039 split.off = xoff;
2040 split.len = xlen;
2041 split.addr = xaddr;
2042 if ((rc = xtSplitUp(tid, ip, &split, &btstack))) {
2043 /* undo data extent allocation */
2044 dbFree(ip, *xaddrp, (s64) * xlenp);
2045
2046 return rc;
2047 }
2048
2049 *xaddrp = xaddr;
2050 *xlenp = xlen;
2051 return 0;
2052
2053 /*
2054 * insert the new entry into the leaf page
2055 */
2056 insertLeaf:
2057 /*
2058 * allocate data extent requested
2059 */
2060 if ((rc = dbAllocBottomUp(ip, xaddr, (s64) xlen)))
2061 goto out;
2062
2063 BT_MARK_DIRTY(mp, ip);
2064 /*
2065 * acquire a transaction lock on the leaf page;
2066 *
2067 * action: xad insertion/extension;
2068 */
2069 tlck = txLock(tid, ip, mp, tlckXTREE | tlckGROW);
2070 xtlck = (struct xtlock *) & tlck->lock;
2071
2072 /* insert the new entry: mark the entry NEW */
2073 xad = &p->xad[index];
2074 XT_PUTENTRY(xad, xflag, xoff, xlen, xaddr);
2075
2076 /* advance next available entry index */
2077 le16_add_cpu(&p->header.nextindex, 1);
2078
2079 xtlck->lwm.offset =
2080 (xtlck->lwm.offset) ? min(index,(int) xtlck->lwm.offset) : index;
2081 xtlck->lwm.length = le16_to_cpu(p->header.nextindex) -
2082 xtlck->lwm.offset;
2083
2084 *xaddrp = xaddr;
2085 *xlenp = xlen;
2086
2087 out:
2088 /* unpin the leaf page */
2089 XT_PUTPAGE(mp);
2090
2091 return rc;
2092 }
2093
2094 /*
2095 * xtInitRoot()
2096 *
2097 * initialize file root (inline in inode)
2098 */
2099 void xtInitRoot(tid_t tid, struct inode *ip)
2100 {
2101 xtroot_t *p;
2102
2103 /*
2104 * acquire a transaction lock on the root
2105 *
2106 * action:
2107 */
2108 txLock(tid, ip, (struct metapage *) &JFS_IP(ip)->bxflag,
2109 tlckXTREE | tlckNEW);
2110 p = &JFS_IP(ip)->i_xtroot;
2111
2112 p->header.flag = DXD_INDEX | BT_ROOT | BT_LEAF;
2113 p->header.nextindex = cpu_to_le16(XTENTRYSTART);
2114
2115 if (S_ISDIR(ip->i_mode))
2116 p->header.maxentry = cpu_to_le16(XTROOTINITSLOT_DIR);
2117 else {
2118 p->header.maxentry = cpu_to_le16(XTROOTINITSLOT);
2119 ip->i_size = 0;
2120 }
2121
2122
2123 return;
2124 }
2125
2126
2127 /*
2128 * We can run into a deadlock truncating a file with a large number of
2129 * xtree pages (large fragmented file). A robust fix would entail a
2130 * reservation system where we would reserve a number of metadata pages
2131 * and tlocks which we would be guaranteed without a deadlock. Without
2132 * this, a partial fix is to limit number of metadata pages we will lock
2133 * in a single transaction. Currently we will truncate the file so that
2134 * no more than 50 leaf pages will be locked. The caller of xtTruncate
2135 * will be responsible for ensuring that the current transaction gets
2136 * committed, and that subsequent transactions are created to truncate
2137 * the file further if needed.
2138 */
2139 #define MAX_TRUNCATE_LEAVES 50
2140
2141 /*
2142 * xtTruncate()
2143 *
2144 * function:
2145 * traverse for truncation logging backward bottom up;
2146 * terminate at the last extent entry at the current subtree
2147 * root page covering new down size.
2148 * truncation may occur within the last extent entry.
2149 *
2150 * parameter:
2151 * int tid,
2152 * struct inode *ip,
2153 * s64 newsize,
2154 * int type) {PWMAP, PMAP, WMAP; DELETE, TRUNCATE}
2155 *
2156 * return:
2157 *
2158 * note:
2159 * PWMAP:
2160 * 1. truncate (non-COMMIT_NOLINK file)
2161 * by jfs_truncate() or jfs_open(O_TRUNC):
2162 * xtree is updated;
2163 * 2. truncate index table of directory when last entry removed
2164 * map update via tlock at commit time;
2165 * PMAP:
2166 * Call xtTruncate_pmap instead
2167 * WMAP:
2168 * 1. remove (free zero link count) on last reference release
2169 * (pmap has been freed at commit zero link count);
2170 * 2. truncate (COMMIT_NOLINK file, i.e., tmp file):
2171 * xtree is updated;
2172 * map update directly at truncation time;
2173 *
2174 * if (DELETE)
2175 * no LOG_NOREDOPAGE is required (NOREDOFILE is sufficient);
2176 * else if (TRUNCATE)
2177 * must write LOG_NOREDOPAGE for deleted index page;
2178 *
2179 * pages may already have been tlocked by anonymous transactions
2180 * during file growth (i.e., write) before truncation;
2181 *
2182 * except last truncated entry, deleted entries remains as is
2183 * in the page (nextindex is updated) for other use
2184 * (e.g., log/update allocation map): this avoid copying the page
2185 * info but delay free of pages;
2186 *
2187 */
2188 s64 xtTruncate(tid_t tid, struct inode *ip, s64 newsize, int flag)
2189 {
2190 int rc = 0;
2191 s64 teof;
2192 struct metapage *mp;
2193 xtpage_t *p;
2194 s64 bn;
2195 int index, nextindex;
2196 xad_t *xad;
2197 s64 xoff, xaddr;
2198 int xlen, len, freexlen;
2199 struct btstack btstack;
2200 struct btframe *parent;
2201 struct tblock *tblk = NULL;
2202 struct tlock *tlck = NULL;
2203 struct xtlock *xtlck = NULL;
2204 struct xdlistlock xadlock; /* maplock for COMMIT_WMAP */
2205 struct pxd_lock *pxdlock; /* maplock for COMMIT_WMAP */
2206 s64 nfreed;
2207 int freed, log;
2208 int locked_leaves = 0;
2209
2210 /* save object truncation type */
2211 if (tid) {
2212 tblk = tid_to_tblock(tid);
2213 tblk->xflag |= flag;
2214 }
2215
2216 nfreed = 0;
2217
2218 flag &= COMMIT_MAP;
2219 assert(flag != COMMIT_PMAP);
2220
2221 if (flag == COMMIT_PWMAP)
2222 log = 1;
2223 else {
2224 log = 0;
2225 xadlock.flag = mlckFREEXADLIST;
2226 xadlock.index = 1;
2227 }
2228
2229 /*
2230 * if the newsize is not an integral number of pages,
2231 * the file between newsize and next page boundary will
2232 * be cleared.
2233 * if truncating into a file hole, it will cause
2234 * a full block to be allocated for the logical block.
2235 */
2236
2237 /*
2238 * release page blocks of truncated region <teof, eof>
2239 *
2240 * free the data blocks from the leaf index blocks.
2241 * delete the parent index entries corresponding to
2242 * the freed child data/index blocks.
2243 * free the index blocks themselves which aren't needed
2244 * in new sized file.
2245 *
2246 * index blocks are updated only if the blocks are to be
2247 * retained in the new sized file.
2248 * if type is PMAP, the data and index pages are NOT
2249 * freed, and the data and index blocks are NOT freed
2250 * from working map.
2251 * (this will allow continued access of data/index of
2252 * temporary file (zerolink count file truncated to zero-length)).
2253 */
2254 teof = (newsize + (JFS_SBI(ip->i_sb)->bsize - 1)) >>
2255 JFS_SBI(ip->i_sb)->l2bsize;
2256
2257 /* clear stack */
2258 BT_CLR(&btstack);
2259
2260 /*
2261 * start with root
2262 *
2263 * root resides in the inode
2264 */
2265 bn = 0;
2266
2267 /*
2268 * first access of each page:
2269 */
2270 getPage:
2271 XT_GETPAGE(ip, bn, mp, PSIZE, p, rc);
2272 if (rc)
2273 return rc;
2274
2275 /* process entries backward from last index */
2276 index = le16_to_cpu(p->header.nextindex) - 1;
2277
2278
2279 /* Since this is the rightmost page at this level, and we may have
2280 * already freed a page that was formerly to the right, let's make
2281 * sure that the next pointer is zero.
2282 */
2283 if (p->header.next) {
2284 if (log)
2285 /*
2286 * Make sure this change to the header is logged.
2287 * If we really truncate this leaf, the flag
2288 * will be changed to tlckTRUNCATE
2289 */
2290 tlck = txLock(tid, ip, mp, tlckXTREE|tlckGROW);
2291 BT_MARK_DIRTY(mp, ip);
2292 p->header.next = 0;
2293 }
2294
2295 if (p->header.flag & BT_INTERNAL)
2296 goto getChild;
2297
2298 /*
2299 * leaf page
2300 */
2301 freed = 0;
2302
2303 /* does region covered by leaf page precede Teof ? */
2304 xad = &p->xad[index];
2305 xoff = offsetXAD(xad);
2306 xlen = lengthXAD(xad);
2307 if (teof >= xoff + xlen) {
2308 XT_PUTPAGE(mp);
2309 goto getParent;
2310 }
2311
2312 /* (re)acquire tlock of the leaf page */
2313 if (log) {
2314 if (++locked_leaves > MAX_TRUNCATE_LEAVES) {
2315 /*
2316 * We need to limit the size of the transaction
2317 * to avoid exhausting pagecache & tlocks
2318 */
2319 XT_PUTPAGE(mp);
2320 newsize = (xoff + xlen) << JFS_SBI(ip->i_sb)->l2bsize;
2321 goto getParent;
2322 }
2323 tlck = txLock(tid, ip, mp, tlckXTREE);
2324 tlck->type = tlckXTREE | tlckTRUNCATE;
2325 xtlck = (struct xtlock *) & tlck->lock;
2326 xtlck->hwm.offset = le16_to_cpu(p->header.nextindex) - 1;
2327 }
2328 BT_MARK_DIRTY(mp, ip);
2329
2330 /*
2331 * scan backward leaf page entries
2332 */
2333 for (; index >= XTENTRYSTART; index--) {
2334 xad = &p->xad[index];
2335 xoff = offsetXAD(xad);
2336 xlen = lengthXAD(xad);
2337 xaddr = addressXAD(xad);
2338
2339 /*
2340 * The "data" for a directory is indexed by the block
2341 * device's address space. This metadata must be invalidated
2342 * here
2343 */
2344 if (S_ISDIR(ip->i_mode) && (teof == 0))
2345 invalidate_xad_metapages(ip, *xad);
2346 /*
2347 * entry beyond eof: continue scan of current page
2348 * xad
2349 * ---|---=======------->
2350 * eof
2351 */
2352 if (teof < xoff) {
2353 nfreed += xlen;
2354 continue;
2355 }
2356
2357 /*
2358 * (xoff <= teof): last entry to be deleted from page;
2359 * If other entries remain in page: keep and update the page.
2360 */
2361
2362 /*
2363 * eof == entry_start: delete the entry
2364 * xad
2365 * -------|=======------->
2366 * eof
2367 *
2368 */
2369 if (teof == xoff) {
2370 nfreed += xlen;
2371
2372 if (index == XTENTRYSTART)
2373 break;
2374
2375 nextindex = index;
2376 }
2377 /*
2378 * eof within the entry: truncate the entry.
2379 * xad
2380 * -------===|===------->
2381 * eof
2382 */
2383 else if (teof < xoff + xlen) {
2384 /* update truncated entry */
2385 len = teof - xoff;
2386 freexlen = xlen - len;
2387 XADlength(xad, len);
2388
2389 /* save pxd of truncated extent in tlck */
2390 xaddr += len;
2391 if (log) { /* COMMIT_PWMAP */
2392 xtlck->lwm.offset = (xtlck->lwm.offset) ?
2393 min(index, (int)xtlck->lwm.offset) : index;
2394 xtlck->lwm.length = index + 1 -
2395 xtlck->lwm.offset;
2396 xtlck->twm.offset = index;
2397 pxdlock = (struct pxd_lock *) & xtlck->pxdlock;
2398 pxdlock->flag = mlckFREEPXD;
2399 PXDaddress(&pxdlock->pxd, xaddr);
2400 PXDlength(&pxdlock->pxd, freexlen);
2401 }
2402 /* free truncated extent */
2403 else { /* COMMIT_WMAP */
2404
2405 pxdlock = (struct pxd_lock *) & xadlock;
2406 pxdlock->flag = mlckFREEPXD;
2407 PXDaddress(&pxdlock->pxd, xaddr);
2408 PXDlength(&pxdlock->pxd, freexlen);
2409 txFreeMap(ip, pxdlock, NULL, COMMIT_WMAP);
2410
2411 /* reset map lock */
2412 xadlock.flag = mlckFREEXADLIST;
2413 }
2414
2415 /* current entry is new last entry; */
2416 nextindex = index + 1;
2417
2418 nfreed += freexlen;
2419 }
2420 /*
2421 * eof beyond the entry:
2422 * xad
2423 * -------=======---|--->
2424 * eof
2425 */
2426 else { /* (xoff + xlen < teof) */
2427
2428 nextindex = index + 1;
2429 }
2430
2431 if (nextindex < le16_to_cpu(p->header.nextindex)) {
2432 if (!log) { /* COMMIT_WAMP */
2433 xadlock.xdlist = &p->xad[nextindex];
2434 xadlock.count =
2435 le16_to_cpu(p->header.nextindex) -
2436 nextindex;
2437 txFreeMap(ip, (struct maplock *) & xadlock,
2438 NULL, COMMIT_WMAP);
2439 }
2440 p->header.nextindex = cpu_to_le16(nextindex);
2441 }
2442
2443 XT_PUTPAGE(mp);
2444
2445 /* assert(freed == 0); */
2446 goto getParent;
2447 } /* end scan of leaf page entries */
2448
2449 freed = 1;
2450
2451 /*
2452 * leaf page become empty: free the page if type != PMAP
2453 */
2454 if (log) { /* COMMIT_PWMAP */
2455 /* txCommit() with tlckFREE:
2456 * free data extents covered by leaf [XTENTRYSTART:hwm);
2457 * invalidate leaf if COMMIT_PWMAP;
2458 * if (TRUNCATE), will write LOG_NOREDOPAGE;
2459 */
2460 tlck->type = tlckXTREE | tlckFREE;
2461 } else { /* COMMIT_WAMP */
2462
2463 /* free data extents covered by leaf */
2464 xadlock.xdlist = &p->xad[XTENTRYSTART];
2465 xadlock.count =
2466 le16_to_cpu(p->header.nextindex) - XTENTRYSTART;
2467 txFreeMap(ip, (struct maplock *) & xadlock, NULL, COMMIT_WMAP);
2468 }
2469
2470 if (p->header.flag & BT_ROOT) {
2471 p->header.flag &= ~BT_INTERNAL;
2472 p->header.flag |= BT_LEAF;
2473 p->header.nextindex = cpu_to_le16(XTENTRYSTART);
2474
2475 XT_PUTPAGE(mp); /* debug */
2476 goto out;
2477 } else {
2478 if (log) { /* COMMIT_PWMAP */
2479 /* page will be invalidated at tx completion
2480 */
2481 XT_PUTPAGE(mp);
2482 } else { /* COMMIT_WMAP */
2483
2484 if (mp->lid)
2485 lid_to_tlock(mp->lid)->flag |= tlckFREELOCK;
2486
2487 /* invalidate empty leaf page */
2488 discard_metapage(mp);
2489 }
2490 }
2491
2492 /*
2493 * the leaf page become empty: delete the parent entry
2494 * for the leaf page if the parent page is to be kept
2495 * in the new sized file.
2496 */
2497
2498 /*
2499 * go back up to the parent page
2500 */
2501 getParent:
2502 /* pop/restore parent entry for the current child page */
2503 if ((parent = BT_POP(&btstack)) == NULL)
2504 /* current page must have been root */
2505 goto out;
2506
2507 /* get back the parent page */
2508 bn = parent->bn;
2509 XT_GETPAGE(ip, bn, mp, PSIZE, p, rc);
2510 if (rc)
2511 return rc;
2512
2513 index = parent->index;
2514
2515 /*
2516 * child page was not empty:
2517 */
2518 if (freed == 0) {
2519 /* has any entry deleted from parent ? */
2520 if (index < le16_to_cpu(p->header.nextindex) - 1) {
2521 /* (re)acquire tlock on the parent page */
2522 if (log) { /* COMMIT_PWMAP */
2523 /* txCommit() with tlckTRUNCATE:
2524 * free child extents covered by parent [);
2525 */
2526 tlck = txLock(tid, ip, mp, tlckXTREE);
2527 xtlck = (struct xtlock *) & tlck->lock;
2528 if (!(tlck->type & tlckTRUNCATE)) {
2529 xtlck->hwm.offset =
2530 le16_to_cpu(p->header.
2531 nextindex) - 1;
2532 tlck->type =
2533 tlckXTREE | tlckTRUNCATE;
2534 }
2535 } else { /* COMMIT_WMAP */
2536
2537 /* free child extents covered by parent */
2538 xadlock.xdlist = &p->xad[index + 1];
2539 xadlock.count =
2540 le16_to_cpu(p->header.nextindex) -
2541 index - 1;
2542 txFreeMap(ip, (struct maplock *) & xadlock,
2543 NULL, COMMIT_WMAP);
2544 }
2545 BT_MARK_DIRTY(mp, ip);
2546
2547 p->header.nextindex = cpu_to_le16(index + 1);
2548 }
2549 XT_PUTPAGE(mp);
2550 goto getParent;
2551 }
2552
2553 /*
2554 * child page was empty:
2555 */
2556 nfreed += lengthXAD(&p->xad[index]);
2557
2558 /*
2559 * During working map update, child page's tlock must be handled
2560 * before parent's. This is because the parent's tlock will cause
2561 * the child's disk space to be marked available in the wmap, so
2562 * it's important that the child page be released by that time.
2563 *
2564 * ToDo: tlocks should be on doubly-linked list, so we can
2565 * quickly remove it and add it to the end.
2566 */
2567
2568 /*
2569 * Move parent page's tlock to the end of the tid's tlock list
2570 */
2571 if (log && mp->lid && (tblk->last != mp->lid) &&
2572 lid_to_tlock(mp->lid)->tid) {
2573 lid_t lid = mp->lid;
2574 struct tlock *prev;
2575
2576 tlck = lid_to_tlock(lid);
2577
2578 if (tblk->next == lid)
2579 tblk->next = tlck->next;
2580 else {
2581 for (prev = lid_to_tlock(tblk->next);
2582 prev->next != lid;
2583 prev = lid_to_tlock(prev->next)) {
2584 assert(prev->next);
2585 }
2586 prev->next = tlck->next;
2587 }
2588 lid_to_tlock(tblk->last)->next = lid;
2589 tlck->next = 0;
2590 tblk->last = lid;
2591 }
2592
2593 /*
2594 * parent page become empty: free the page
2595 */
2596 if (index == XTENTRYSTART) {
2597 if (log) { /* COMMIT_PWMAP */
2598 /* txCommit() with tlckFREE:
2599 * free child extents covered by parent;
2600 * invalidate parent if COMMIT_PWMAP;
2601 */
2602 tlck = txLock(tid, ip, mp, tlckXTREE);
2603 xtlck = (struct xtlock *) & tlck->lock;
2604 xtlck->hwm.offset =
2605 le16_to_cpu(p->header.nextindex) - 1;
2606 tlck->type = tlckXTREE | tlckFREE;
2607 } else { /* COMMIT_WMAP */
2608
2609 /* free child extents covered by parent */
2610 xadlock.xdlist = &p->xad[XTENTRYSTART];
2611 xadlock.count =
2612 le16_to_cpu(p->header.nextindex) -
2613 XTENTRYSTART;
2614 txFreeMap(ip, (struct maplock *) & xadlock, NULL,
2615 COMMIT_WMAP);
2616 }
2617 BT_MARK_DIRTY(mp, ip);
2618
2619 if (p->header.flag & BT_ROOT) {
2620 p->header.flag &= ~BT_INTERNAL;
2621 p->header.flag |= BT_LEAF;
2622 p->header.nextindex = cpu_to_le16(XTENTRYSTART);
2623 if (le16_to_cpu(p->header.maxentry) == XTROOTMAXSLOT) {
2624 /*
2625 * Shrink root down to allow inline
2626 * EA (otherwise fsck complains)
2627 */
2628 p->header.maxentry =
2629 cpu_to_le16(XTROOTINITSLOT);
2630 JFS_IP(ip)->mode2 |= INLINEEA;
2631 }
2632
2633 XT_PUTPAGE(mp); /* debug */
2634 goto out;
2635 } else {
2636 if (log) { /* COMMIT_PWMAP */
2637 /* page will be invalidated at tx completion
2638 */
2639 XT_PUTPAGE(mp);
2640 } else { /* COMMIT_WMAP */
2641
2642 if (mp->lid)
2643 lid_to_tlock(mp->lid)->flag |=
2644 tlckFREELOCK;
2645
2646 /* invalidate parent page */
2647 discard_metapage(mp);
2648 }
2649
2650 /* parent has become empty and freed:
2651 * go back up to its parent page
2652 */
2653 /* freed = 1; */
2654 goto getParent;
2655 }
2656 }
2657 /*
2658 * parent page still has entries for front region;
2659 */
2660 else {
2661 /* try truncate region covered by preceding entry
2662 * (process backward)
2663 */
2664 index--;
2665
2666 /* go back down to the child page corresponding
2667 * to the entry
2668 */
2669 goto getChild;
2670 }
2671
2672 /*
2673 * internal page: go down to child page of current entry
2674 */
2675 getChild:
2676 /* save current parent entry for the child page */
2677 if (BT_STACK_FULL(&btstack)) {
2678 jfs_error(ip->i_sb, "stack overrun!\n");
2679 XT_PUTPAGE(mp);
2680 return -EIO;
2681 }
2682 BT_PUSH(&btstack, bn, index);
2683
2684 /* get child page */
2685 xad = &p->xad[index];
2686 bn = addressXAD(xad);
2687
2688 /*
2689 * first access of each internal entry:
2690 */
2691 /* release parent page */
2692 XT_PUTPAGE(mp);
2693
2694 /* process the child page */
2695 goto getPage;
2696
2697 out:
2698 /*
2699 * update file resource stat
2700 */
2701 /* set size
2702 */
2703 if (S_ISDIR(ip->i_mode) && !newsize)
2704 ip->i_size = 1; /* fsck hates zero-length directories */
2705 else
2706 ip->i_size = newsize;
2707
2708 /* update quota allocation to reflect freed blocks */
2709 dquot_free_block(ip, nfreed);
2710
2711 /*
2712 * free tlock of invalidated pages
2713 */
2714 if (flag == COMMIT_WMAP)
2715 txFreelock(ip);
2716
2717 return newsize;
2718 }
2719
2720
2721 /*
2722 * xtTruncate_pmap()
2723 *
2724 * function:
2725 * Perform truncate to zero length for deleted file, leaving the
2726 * xtree and working map untouched. This allows the file to
2727 * be accessed via open file handles, while the delete of the file
2728 * is committed to disk.
2729 *
2730 * parameter:
2731 * tid_t tid,
2732 * struct inode *ip,
2733 * s64 committed_size)
2734 *
2735 * return: new committed size
2736 *
2737 * note:
2738 *
2739 * To avoid deadlock by holding too many transaction locks, the
2740 * truncation may be broken up into multiple transactions.
2741 * The committed_size keeps track of part of the file has been
2742 * freed from the pmaps.
2743 */
2744 s64 xtTruncate_pmap(tid_t tid, struct inode *ip, s64 committed_size)
2745 {
2746 s64 bn;
2747 struct btstack btstack;
2748 int cmp;
2749 int index;
2750 int locked_leaves = 0;
2751 struct metapage *mp;
2752 xtpage_t *p;
2753 struct btframe *parent;
2754 int rc;
2755 struct tblock *tblk;
2756 struct tlock *tlck = NULL;
2757 xad_t *xad;
2758 int xlen;
2759 s64 xoff;
2760 struct xtlock *xtlck = NULL;
2761
2762 /* save object truncation type */
2763 tblk = tid_to_tblock(tid);
2764 tblk->xflag |= COMMIT_PMAP;
2765
2766 /* clear stack */
2767 BT_CLR(&btstack);
2768
2769 if (committed_size) {
2770 xoff = (committed_size >> JFS_SBI(ip->i_sb)->l2bsize) - 1;
2771 rc = xtSearch(ip, xoff, NULL, &cmp, &btstack, 0);
2772 if (rc)
2773 return rc;
2774
2775 XT_GETSEARCH(ip, btstack.top, bn, mp, p, index);
2776
2777 if (cmp != 0) {
2778 XT_PUTPAGE(mp);
2779 jfs_error(ip->i_sb, "did not find extent\n");
2780 return -EIO;
2781 }
2782 } else {
2783 /*
2784 * start with root
2785 *
2786 * root resides in the inode
2787 */
2788 bn = 0;
2789
2790 /*
2791 * first access of each page:
2792 */
2793 getPage:
2794 XT_GETPAGE(ip, bn, mp, PSIZE, p, rc);
2795 if (rc)
2796 return rc;
2797
2798 /* process entries backward from last index */
2799 index = le16_to_cpu(p->header.nextindex) - 1;
2800
2801 if (p->header.flag & BT_INTERNAL)
2802 goto getChild;
2803 }
2804
2805 /*
2806 * leaf page
2807 */
2808
2809 if (++locked_leaves > MAX_TRUNCATE_LEAVES) {
2810 /*
2811 * We need to limit the size of the transaction
2812 * to avoid exhausting pagecache & tlocks
2813 */
2814 xad = &p->xad[index];
2815 xoff = offsetXAD(xad);
2816 xlen = lengthXAD(xad);
2817 XT_PUTPAGE(mp);
2818 return (xoff + xlen) << JFS_SBI(ip->i_sb)->l2bsize;
2819 }
2820 tlck = txLock(tid, ip, mp, tlckXTREE);
2821 tlck->type = tlckXTREE | tlckFREE;
2822 xtlck = (struct xtlock *) & tlck->lock;
2823 xtlck->hwm.offset = index;
2824
2825
2826 XT_PUTPAGE(mp);
2827
2828 /*
2829 * go back up to the parent page
2830 */
2831 getParent:
2832 /* pop/restore parent entry for the current child page */
2833 if ((parent = BT_POP(&btstack)) == NULL)
2834 /* current page must have been root */
2835 goto out;
2836
2837 /* get back the parent page */
2838 bn = parent->bn;
2839 XT_GETPAGE(ip, bn, mp, PSIZE, p, rc);
2840 if (rc)
2841 return rc;
2842
2843 index = parent->index;
2844
2845 /*
2846 * parent page become empty: free the page
2847 */
2848 if (index == XTENTRYSTART) {
2849 /* txCommit() with tlckFREE:
2850 * free child extents covered by parent;
2851 * invalidate parent if COMMIT_PWMAP;
2852 */
2853 tlck = txLock(tid, ip, mp, tlckXTREE);
2854 xtlck = (struct xtlock *) & tlck->lock;
2855 xtlck->hwm.offset = le16_to_cpu(p->header.nextindex) - 1;
2856 tlck->type = tlckXTREE | tlckFREE;
2857
2858 XT_PUTPAGE(mp);
2859
2860 if (p->header.flag & BT_ROOT) {
2861
2862 goto out;
2863 } else {
2864 goto getParent;
2865 }
2866 }
2867 /*
2868 * parent page still has entries for front region;
2869 */
2870 else
2871 index--;
2872 /*
2873 * internal page: go down to child page of current entry
2874 */
2875 getChild:
2876 /* save current parent entry for the child page */
2877 if (BT_STACK_FULL(&btstack)) {
2878 jfs_error(ip->i_sb, "stack overrun!\n");
2879 XT_PUTPAGE(mp);
2880 return -EIO;
2881 }
2882 BT_PUSH(&btstack, bn, index);
2883
2884 /* get child page */
2885 xad = &p->xad[index];
2886 bn = addressXAD(xad);
2887
2888 /*
2889 * first access of each internal entry:
2890 */
2891 /* release parent page */
2892 XT_PUTPAGE(mp);
2893
2894 /* process the child page */
2895 goto getPage;
2896
2897 out:
2898
2899 return 0;
2900 }
2901
2902 #ifdef CONFIG_JFS_STATISTICS
2903 int jfs_xtstat_proc_show(struct seq_file *m, void *v)
2904 {
2905 seq_printf(m,
2906 "JFS Xtree statistics\n"
2907 "====================\n"
2908 "searches = %d\n"
2909 "fast searches = %d\n"
2910 "splits = %d\n",
2911 xtStat.search,
2912 xtStat.fastSearch,
2913 xtStat.split);
2914 return 0;
2915 }
2916 #endif
2917
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.