]>
git.saurik.com Git - apple/libc.git/blob - db/btree/FreeBSD/bt_split.c
2f34975018cbca9fd517588934c51f3dc9f2a4f6
2 * Copyright (c) 1990, 1993, 1994
3 * The Regents of the University of California. All rights reserved.
5 * This code is derived from software contributed to Berkeley by
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
11 * 1. Redistributions of source code must retain the above copyright
12 * notice, this list of conditions and the following disclaimer.
13 * 2. Redistributions in binary form must reproduce the above copyright
14 * notice, this list of conditions and the following disclaimer in the
15 * documentation and/or other materials provided with the distribution.
16 * 4. Neither the name of the University nor the names of its contributors
17 * may be used to endorse or promote products derived from this software
18 * without specific prior written permission.
20 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
21 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
22 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
23 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
24 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
25 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
26 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
27 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
28 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
29 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
33 #if defined(LIBC_SCCS) && !defined(lint)
34 static char sccsid
[] = "@(#)bt_split.c 8.9 (Berkeley) 7/26/94";
35 #endif /* LIBC_SCCS and not lint */
36 #include <sys/cdefs.h>
37 __FBSDID("$FreeBSD: src/lib/libc/db/btree/bt_split.c,v 1.12 2009/03/28 05:45:29 delphij Exp $");
39 #include <sys/types.h>
49 static int bt_broot(BTREE
*, PAGE
*, PAGE
*, PAGE
*);
50 static PAGE
*bt_page(BTREE
*, PAGE
*, PAGE
**, PAGE
**, indx_t
*, size_t);
51 static int bt_preserve(BTREE
*, pgno_t
);
52 static PAGE
*bt_psplit(BTREE
*, PAGE
*, PAGE
*, PAGE
*, indx_t
*, size_t);
53 static PAGE
*bt_root(BTREE
*, PAGE
*, PAGE
**, PAGE
**, indx_t
*, size_t);
54 static int bt_rroot(BTREE
*, PAGE
*, PAGE
*, PAGE
*);
55 static recno_t
rec_total(PAGE
*);
58 u_long bt_rootsplit
, bt_split
, bt_sortsplit
, bt_pfxsaved
;
62 * __BT_SPLIT -- Split the tree.
68 * data: data to insert
69 * flags: BIGKEY/BIGDATA flags
71 * skip: index to leave open
74 * RET_ERROR, RET_SUCCESS
77 __bt_split(BTREE
*t
, PAGE
*sp
, const DBT
*key
, const DBT
*data
, int flags
,
78 size_t ilen
, u_int32_t argskip
)
84 PAGE
*h
, *l
, *r
, *lchild
, *rchild
;
87 u_int32_t n
, nbytes
, nksize
;
92 * Split the page into two pages, l and r. The split routines return
93 * a pointer to the page into which the key should be inserted and with
94 * skip set to the offset which should be used. Additionally, l and r
98 h
= sp
->pgno
== P_ROOT
?
99 bt_root(t
, sp
, &l
, &r
, &skip
, ilen
) :
100 bt_page(t
, sp
, &l
, &r
, &skip
, ilen
);
105 * Insert the new key/data pair into the leaf page. (Key inserts
106 * always cause a leaf page to split first.)
108 h
->linp
[skip
] = h
->upper
-= ilen
;
109 dest
= (char *)h
+ h
->upper
;
110 if (F_ISSET(t
, R_RECNO
))
111 WR_RLEAF(dest
, data
, flags
)
113 WR_BLEAF(dest
, key
, data
, flags
)
115 /* If the root page was split, make it look right. */
116 if (sp
->pgno
== P_ROOT
&&
117 (F_ISSET(t
, R_RECNO
) ?
118 bt_rroot(t
, sp
, l
, r
) : bt_broot(t
, sp
, l
, r
)) == RET_ERROR
)
122 * Now we walk the parent page stack -- a LIFO stack of the pages that
123 * were traversed when we searched for the page that split. Each stack
124 * entry is a page number and a page index offset. The offset is for
125 * the page traversed on the search. We've just split a page, so we
126 * have to insert a new key into the parent page.
128 * If the insert into the parent page causes it to split, may have to
129 * continue splitting all the way up the tree. We stop if the root
130 * splits or the page inserted into didn't have to split to hold the
131 * new key. Some algorithms replace the key for the old page as well
132 * as the new page. We don't, as there's no reason to believe that the
133 * first key on the old page is any better than the key we have, and,
134 * in the case of a key being placed at index 0 causing the split, the
135 * key is unavailable.
137 * There are a maximum of 5 pages pinned at any time. We keep the left
138 * and right pages pinned while working on the parent. The 5 are the
139 * two children, left parent and right parent (when the parent splits)
140 * and the root page or the overflow key page when calling bt_preserve.
141 * This code must make sure that all pins are released other than the
142 * root page or overflow page which is unlocked elsewhere.
144 while ((parent
= BT_POP(t
)) != NULL
) {
148 /* Get the parent page. */
149 if ((h
= mpool_get(t
->bt_mp
, parent
->pgno
, 0)) == NULL
)
153 * The new key goes ONE AFTER the index, because the split
156 skip
= parent
->index
+ 1;
159 * Calculate the space needed on the parent page.
161 * Prefix trees: space hack when inserting into BINTERNAL
162 * pages. Retain only what's needed to distinguish between
163 * the new entry and the LAST entry on the page to its left.
164 * If the keys compare equal, retain the entire key. Note,
165 * we don't touch overflow keys, and the entire key must be
166 * retained for the next-to-left most key on the leftmost
167 * page of each level, or the search will fail. Applicable
168 * ONLY to internal pages that have leaf pages as children.
169 * Further reduction of the key between pairs of internal
170 * pages loses too much information.
172 switch (rchild
->flags
& P_TYPE
) {
174 bi
= GETBINTERNAL(rchild
, 0);
175 nbytes
= NBINTERNAL(bi
->ksize
);
178 bl
= GETBLEAF(rchild
, 0);
179 nbytes
= NBINTERNAL(bl
->ksize
);
180 if (t
->bt_pfx
&& !(bl
->flags
& P_BIGKEY
) &&
181 (h
->prevpg
!= P_INVALID
|| skip
> 1)) {
182 tbl
= GETBLEAF(lchild
, NEXTINDEX(lchild
) - 1);
187 nksize
= t
->bt_pfx(&a
, &b
);
188 n
= NBINTERNAL(nksize
);
191 bt_pfxsaved
+= nbytes
- n
;
207 /* Split the parent page if necessary or shift the indices. */
208 if ((u_int32_t
)(h
->upper
- h
->lower
) < nbytes
+ sizeof(indx_t
)) {
210 h
= h
->pgno
== P_ROOT
?
211 bt_root(t
, h
, &l
, &r
, &skip
, nbytes
) :
212 bt_page(t
, h
, &l
, &r
, &skip
, nbytes
);
217 if (skip
< (nxtindex
= NEXTINDEX(h
)))
218 memmove(h
->linp
+ skip
+ 1, h
->linp
+ skip
,
219 (nxtindex
- skip
) * sizeof(indx_t
));
220 h
->lower
+= sizeof(indx_t
);
224 /* Insert the key into the parent page. */
225 switch (rchild
->flags
& P_TYPE
) {
227 h
->linp
[skip
] = h
->upper
-= nbytes
;
228 dest
= (char *)h
+ h
->linp
[skip
];
229 memmove(dest
, bi
, nbytes
);
230 ((BINTERNAL
*)dest
)->pgno
= rchild
->pgno
;
233 h
->linp
[skip
] = h
->upper
-= nbytes
;
234 dest
= (char *)h
+ h
->linp
[skip
];
235 WR_BINTERNAL(dest
, nksize
? nksize
: bl
->ksize
,
236 rchild
->pgno
, bl
->flags
& P_BIGKEY
);
237 memmove(dest
, bl
->bytes
, nksize
? nksize
: bl
->ksize
);
238 if (bl
->flags
& P_BIGKEY
&&
239 bt_preserve(t
, *(pgno_t
*)bl
->bytes
) == RET_ERROR
)
244 * Update the left page count. If split
245 * added at index 0, fix the correct page.
248 dest
= (char *)h
+ h
->linp
[skip
- 1];
250 dest
= (char *)l
+ l
->linp
[NEXTINDEX(l
) - 1];
251 ((RINTERNAL
*)dest
)->nrecs
= rec_total(lchild
);
252 ((RINTERNAL
*)dest
)->pgno
= lchild
->pgno
;
254 /* Update the right page count. */
255 h
->linp
[skip
] = h
->upper
-= nbytes
;
256 dest
= (char *)h
+ h
->linp
[skip
];
257 ((RINTERNAL
*)dest
)->nrecs
= rec_total(rchild
);
258 ((RINTERNAL
*)dest
)->pgno
= rchild
->pgno
;
262 * Update the left page count. If split
263 * added at index 0, fix the correct page.
266 dest
= (char *)h
+ h
->linp
[skip
- 1];
268 dest
= (char *)l
+ l
->linp
[NEXTINDEX(l
) - 1];
269 ((RINTERNAL
*)dest
)->nrecs
= NEXTINDEX(lchild
);
270 ((RINTERNAL
*)dest
)->pgno
= lchild
->pgno
;
272 /* Update the right page count. */
273 h
->linp
[skip
] = h
->upper
-= nbytes
;
274 dest
= (char *)h
+ h
->linp
[skip
];
275 ((RINTERNAL
*)dest
)->nrecs
= NEXTINDEX(rchild
);
276 ((RINTERNAL
*)dest
)->pgno
= rchild
->pgno
;
282 /* Unpin the held pages. */
284 mpool_put(t
->bt_mp
, h
, MPOOL_DIRTY
);
288 /* If the root page was split, make it look right. */
289 if (sp
->pgno
== P_ROOT
&&
290 (F_ISSET(t
, R_RECNO
) ?
291 bt_rroot(t
, sp
, l
, r
) : bt_broot(t
, sp
, l
, r
)) == RET_ERROR
)
294 mpool_put(t
->bt_mp
, lchild
, MPOOL_DIRTY
);
295 mpool_put(t
->bt_mp
, rchild
, MPOOL_DIRTY
);
298 /* Unpin the held pages. */
299 mpool_put(t
->bt_mp
, l
, MPOOL_DIRTY
);
300 mpool_put(t
->bt_mp
, r
, MPOOL_DIRTY
);
302 /* Clear any pages left on the stack. */
303 return (RET_SUCCESS
);
306 * If something fails in the above loop we were already walking back
307 * up the tree and the tree is now inconsistent. Nothing much we can
308 * do about it but release any memory we're holding.
310 err1
: mpool_put(t
->bt_mp
, lchild
, MPOOL_DIRTY
);
311 mpool_put(t
->bt_mp
, rchild
, MPOOL_DIRTY
);
313 err2
: mpool_put(t
->bt_mp
, l
, 0);
314 mpool_put(t
->bt_mp
, r
, 0);
315 __dbpanic(t
->bt_dbp
);
320 * BT_PAGE -- Split a non-root page of a btree.
325 * lp: pointer to left page pointer
326 * rp: pointer to right page pointer
327 * skip: pointer to index to leave open
328 * ilen: insert length
331 * Pointer to page in which to insert or NULL on error.
334 bt_page(BTREE
*t
, PAGE
*h
, PAGE
**lp
, PAGE
**rp
, indx_t
*skip
, size_t ilen
)
342 /* Put the new right page for the split into place. */
343 if ((r
= __bt_new(t
, &npg
)) == NULL
)
346 r
->lower
= BTDATAOFF
;
347 r
->upper
= t
->bt_psize
;
348 r
->nextpg
= h
->nextpg
;
350 r
->flags
= h
->flags
& P_TYPE
;
353 * If we're splitting the last page on a level because we're appending
354 * a key to it (skip is NEXTINDEX()), it's likely that the data is
355 * sorted. Adding an empty page on the side of the level is less work
356 * and can push the fill factor much higher than normal. If we're
357 * wrong it's no big deal, we'll just do the split the right way next
358 * time. It may look like it's equally easy to do a similar hack for
359 * reverse sorted data, that is, split the tree left, but it's not.
362 if (h
->nextpg
== P_INVALID
&& *skip
== NEXTINDEX(h
)) {
367 r
->lower
= BTDATAOFF
+ sizeof(indx_t
);
374 /* Put the new left page for the split into place. */
375 if ((l
= (PAGE
*)calloc(1, t
->bt_psize
)) == NULL
) {
376 mpool_put(t
->bt_mp
, r
, 0);
381 l
->prevpg
= h
->prevpg
;
382 l
->lower
= BTDATAOFF
;
383 l
->upper
= t
->bt_psize
;
384 l
->flags
= h
->flags
& P_TYPE
;
386 /* Fix up the previous pointer of the page after the split page. */
387 if (h
->nextpg
!= P_INVALID
) {
388 if ((tp
= mpool_get(t
->bt_mp
, h
->nextpg
, 0)) == NULL
) {
390 /* XXX mpool_free(t->bt_mp, r->pgno); */
393 tp
->prevpg
= r
->pgno
;
394 mpool_put(t
->bt_mp
, tp
, MPOOL_DIRTY
);
398 * Split right. The key/data pairs aren't sorted in the btree page so
399 * it's simpler to copy the data from the split page onto two new pages
400 * instead of copying half the data to the right page and compacting
401 * the left page in place. Since the left page can't change, we have
402 * to swap the original and the allocated left page after the split.
404 tp
= bt_psplit(t
, h
, l
, r
, skip
, ilen
);
406 /* Move the new left page onto the old left page. */
407 memmove(h
, l
, t
->bt_psize
);
418 * BT_ROOT -- Split the root page of a btree.
423 * lp: pointer to left page pointer
424 * rp: pointer to right page pointer
425 * skip: pointer to index to leave open
426 * ilen: insert length
429 * Pointer to page in which to insert or NULL on error.
432 bt_root(BTREE
*t
, PAGE
*h
, PAGE
**lp
, PAGE
**rp
, indx_t
*skip
, size_t ilen
)
441 /* Put the new left and right pages for the split into place. */
442 if ((l
= __bt_new(t
, &lnpg
)) == NULL
||
443 (r
= __bt_new(t
, &rnpg
)) == NULL
)
449 l
->prevpg
= r
->nextpg
= P_INVALID
;
450 l
->lower
= r
->lower
= BTDATAOFF
;
451 l
->upper
= r
->upper
= t
->bt_psize
;
452 l
->flags
= r
->flags
= h
->flags
& P_TYPE
;
454 /* Split the root page. */
455 tp
= bt_psplit(t
, h
, l
, r
, skip
, ilen
);
463 * BT_RROOT -- Fix up the recno root page after it has been split.
472 * RET_ERROR, RET_SUCCESS
475 bt_rroot(BTREE
*t
, PAGE
*h
, PAGE
*l
, PAGE
*r
)
479 /* Insert the left and right keys, set the header information. */
480 h
->linp
[0] = h
->upper
= t
->bt_psize
- NRINTERNAL
;
481 dest
= (char *)h
+ h
->upper
;
483 l
->flags
& P_RLEAF
? NEXTINDEX(l
) : rec_total(l
), l
->pgno
);
485 h
->linp
[1] = h
->upper
-= NRINTERNAL
;
486 dest
= (char *)h
+ h
->upper
;
488 r
->flags
& P_RLEAF
? NEXTINDEX(r
) : rec_total(r
), r
->pgno
);
490 h
->lower
= BTDATAOFF
+ 2 * sizeof(indx_t
);
492 /* Unpin the root page, set to recno internal page. */
494 h
->flags
|= P_RINTERNAL
;
495 mpool_put(t
->bt_mp
, h
, MPOOL_DIRTY
);
497 return (RET_SUCCESS
);
501 * BT_BROOT -- Fix up the btree root page after it has been split.
510 * RET_ERROR, RET_SUCCESS
513 bt_broot(BTREE
*t
, PAGE
*h
, PAGE
*l
, PAGE
*r
)
521 * If the root page was a leaf page, change it into an internal page.
522 * We copy the key we split on (but not the key's data, in the case of
523 * a leaf page) to the new root page.
525 * The btree comparison code guarantees that the left-most key on any
526 * level of the tree is never used, so it doesn't need to be filled in.
528 nbytes
= NBINTERNAL(0);
529 h
->linp
[0] = h
->upper
= t
->bt_psize
- nbytes
;
530 dest
= (char *)h
+ h
->upper
;
531 WR_BINTERNAL(dest
, 0, l
->pgno
, 0);
533 switch (h
->flags
& P_TYPE
) {
536 nbytes
= NBINTERNAL(bl
->ksize
);
537 h
->linp
[1] = h
->upper
-= nbytes
;
538 dest
= (char *)h
+ h
->upper
;
539 WR_BINTERNAL(dest
, bl
->ksize
, r
->pgno
, 0);
540 memmove(dest
, bl
->bytes
, bl
->ksize
);
543 * If the key is on an overflow page, mark the overflow chain
544 * so it isn't deleted when the leaf copy of the key is deleted.
546 if (bl
->flags
& P_BIGKEY
&&
547 bt_preserve(t
, *(pgno_t
*)bl
->bytes
) == RET_ERROR
)
551 bi
= GETBINTERNAL(r
, 0);
552 nbytes
= NBINTERNAL(bi
->ksize
);
553 h
->linp
[1] = h
->upper
-= nbytes
;
554 dest
= (char *)h
+ h
->upper
;
555 memmove(dest
, bi
, nbytes
);
556 ((BINTERNAL
*)dest
)->pgno
= r
->pgno
;
562 /* There are two keys on the page. */
563 h
->lower
= BTDATAOFF
+ 2 * sizeof(indx_t
);
565 /* Unpin the root page, set to btree internal page. */
567 h
->flags
|= P_BINTERNAL
;
568 mpool_put(t
->bt_mp
, h
, MPOOL_DIRTY
);
570 return (RET_SUCCESS
);
574 * BT_PSPLIT -- Do the real work of splitting the page.
578 * h: page to be split
579 * l: page to put lower half of data
580 * r: page to put upper half of data
581 * pskip: pointer to index to leave open
582 * ilen: insert length
585 * Pointer to page in which to insert.
588 bt_psplit(BTREE
*t
, PAGE
*h
, PAGE
*l
, PAGE
*r
, indx_t
*pskip
, size_t ilen
)
596 indx_t full
, half
, nxt
, off
, skip
, top
, used
;
598 int bigkeycnt
, isbigkey
;
601 * Split the data to the left and right pages. Leave the skip index
602 * open. Additionally, make some effort not to split on an overflow
603 * key. This makes internal page processing faster and can save
604 * space as overflow keys used by internal pages are never deleted.
608 full
= t
->bt_psize
- BTDATAOFF
;
611 for (nxt
= off
= 0, top
= NEXTINDEX(h
); nxt
< top
; ++off
) {
614 isbigkey
= 0; /* XXX: not really known. */
616 switch (h
->flags
& P_TYPE
) {
618 src
= bi
= GETBINTERNAL(h
, nxt
);
619 nbytes
= NBINTERNAL(bi
->ksize
);
620 isbigkey
= bi
->flags
& P_BIGKEY
;
623 src
= bl
= GETBLEAF(h
, nxt
);
625 isbigkey
= bl
->flags
& P_BIGKEY
;
628 src
= GETRINTERNAL(h
, nxt
);
633 src
= rl
= GETRLEAF(h
, nxt
);
642 * If the key/data pairs are substantial fractions of the max
643 * possible size for the page, it's possible to get situations
644 * where we decide to try and copy too much onto the left page.
645 * Make sure that doesn't happen.
647 if ((skip
<= off
&& used
+ nbytes
+ sizeof(indx_t
) >= full
)
653 /* Copy the key/data pair, if not the skipped index. */
657 l
->linp
[off
] = l
->upper
-= nbytes
;
658 memmove((char *)l
+ l
->upper
, src
, nbytes
);
661 used
+= nbytes
+ sizeof(indx_t
);
663 if (!isbigkey
|| bigkeycnt
== 3)
671 * Off is the last offset that's valid for the left page.
672 * Nxt is the first offset to be placed on the right page.
674 l
->lower
+= (off
+ 1) * sizeof(indx_t
);
677 * If splitting the page that the cursor was on, the cursor has to be
678 * adjusted to point to the same record as before the split. If the
679 * cursor is at or past the skipped slot, the cursor is incremented by
680 * one. If the cursor is on the right page, it is decremented by the
681 * number of records split to the left page.
684 if (F_ISSET(c
, CURS_INIT
) && c
->pg
.pgno
== h
->pgno
) {
685 if (c
->pg
.index
>= skip
)
687 if (c
->pg
.index
< nxt
) /* Left page. */
688 c
->pg
.pgno
= l
->pgno
;
689 else { /* Right page. */
690 c
->pg
.pgno
= r
->pgno
;
696 * If the skipped index was on the left page, just return that page.
697 * Otherwise, adjust the skip index to reflect the new position on
701 skip
= MAX_PAGE_OFFSET
;
708 for (off
= 0; nxt
< top
; ++off
) {
711 skip
= MAX_PAGE_OFFSET
;
713 switch (h
->flags
& P_TYPE
) {
715 src
= bi
= GETBINTERNAL(h
, nxt
);
716 nbytes
= NBINTERNAL(bi
->ksize
);
719 src
= bl
= GETBLEAF(h
, nxt
);
723 src
= GETRINTERNAL(h
, nxt
);
727 src
= rl
= GETRLEAF(h
, nxt
);
734 r
->linp
[off
] = r
->upper
-= nbytes
;
735 memmove((char *)r
+ r
->upper
, src
, nbytes
);
737 r
->lower
+= off
* sizeof(indx_t
);
739 /* If the key is being appended to the page, adjust the index. */
741 r
->lower
+= sizeof(indx_t
);
747 * BT_PRESERVE -- Mark a chain of pages as used by an internal node.
749 * Chains of indirect blocks pointed to by leaf nodes get reclaimed when the
750 * record that references them gets deleted. Chains pointed to by internal
751 * pages never get deleted. This routine marks a chain as pointed to by an
756 * pg: page number of first page in the chain.
759 * RET_SUCCESS, RET_ERROR.
762 bt_preserve(BTREE
*t
, pgno_t pg
)
766 if ((h
= mpool_get(t
->bt_mp
, pg
, 0)) == NULL
)
768 h
->flags
|= P_PRESERVE
;
769 mpool_put(t
->bt_mp
, h
, MPOOL_DIRTY
);
770 return (RET_SUCCESS
);
774 * REC_TOTAL -- Return the number of recno entries below a page.
780 * The number of recno entries below a page.
783 * These values could be set by the bt_psplit routine. The problem is that the
784 * entry has to be popped off of the stack etc. or the values have to be passed
785 * all the way back to bt_split/bt_rroot and it's not very clean.
793 for (recs
= 0, nxt
= 0, top
= NEXTINDEX(h
); nxt
< top
; ++nxt
)
794 recs
+= GETRINTERNAL(h
, nxt
)->nrecs
;