xref: /freebsd/lib/libc/db/hash/hash_bigkey.c (revision f0a75d274af375d15b97b830966b99a02b7db911)
1 /*-
2  * Copyright (c) 1990, 1993, 1994
3  *	The Regents of the University of California.  All rights reserved.
4  *
5  * This code is derived from software contributed to Berkeley by
6  * Margo Seltzer.
7  *
8  * Redistribution and use in source and binary forms, with or without
9  * modification, are permitted provided that the following conditions
10  * are met:
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.
19  *
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
30  * SUCH DAMAGE.
31  */
32 
33 #if defined(LIBC_SCCS) && !defined(lint)
34 static char sccsid[] = "@(#)hash_bigkey.c	8.3 (Berkeley) 5/31/94";
35 #endif /* LIBC_SCCS and not lint */
36 #include <sys/cdefs.h>
37 __FBSDID("$FreeBSD$");
38 
39 /*
40  * PACKAGE: hash
41  * DESCRIPTION:
42  *	Big key/data handling for the hashing package.
43  *
44  * ROUTINES:
45  * External
46  *	__big_keydata
47  *	__big_split
48  *	__big_insert
49  *	__big_return
50  *	__big_delete
51  *	__find_last_page
52  * Internal
53  *	collect_key
54  *	collect_data
55  */
56 
57 #include <sys/param.h>
58 
59 #include <errno.h>
60 #include <stdio.h>
61 #include <stdlib.h>
62 #include <string.h>
63 
64 #ifdef DEBUG
65 #include <assert.h>
66 #endif
67 
68 #include <db.h>
69 #include "hash.h"
70 #include "page.h"
71 #include "extern.h"
72 
73 static int collect_key(HTAB *, BUFHEAD *, int, DBT *, int);
74 static int collect_data(HTAB *, BUFHEAD *, int, int);
75 
76 /*
77  * Big_insert
78  *
79  * You need to do an insert and the key/data pair is too big
80  *
81  * Returns:
82  * 0 ==> OK
83  *-1 ==> ERROR
84  */
85 extern int
86 __big_insert(hashp, bufp, key, val)
87 	HTAB *hashp;
88 	BUFHEAD *bufp;
89 	const DBT *key, *val;
90 {
91 	u_int16_t *p;
92 	int key_size, n, val_size;
93 	u_int16_t space, move_bytes, off;
94 	char *cp, *key_data, *val_data;
95 
96 	cp = bufp->page;		/* Character pointer of p. */
97 	p = (u_int16_t *)cp;
98 
99 	key_data = (char *)key->data;
100 	key_size = key->size;
101 	val_data = (char *)val->data;
102 	val_size = val->size;
103 
104 	/* First move the Key */
105 	for (space = FREESPACE(p) - BIGOVERHEAD; key_size;
106 	    space = FREESPACE(p) - BIGOVERHEAD) {
107 		move_bytes = MIN(space, key_size);
108 		off = OFFSET(p) - move_bytes;
109 		memmove(cp + off, key_data, move_bytes);
110 		key_size -= move_bytes;
111 		key_data += move_bytes;
112 		n = p[0];
113 		p[++n] = off;
114 		p[0] = ++n;
115 		FREESPACE(p) = off - PAGE_META(n);
116 		OFFSET(p) = off;
117 		p[n] = PARTIAL_KEY;
118 		bufp = __add_ovflpage(hashp, bufp);
119 		if (!bufp)
120 			return (-1);
121 		n = p[0];
122 		if (!key_size) {
123 			if (FREESPACE(p)) {
124 				move_bytes = MIN(FREESPACE(p), val_size);
125 				off = OFFSET(p) - move_bytes;
126 				p[n] = off;
127 				memmove(cp + off, val_data, move_bytes);
128 				val_data += move_bytes;
129 				val_size -= move_bytes;
130 				p[n - 2] = FULL_KEY_DATA;
131 				FREESPACE(p) = FREESPACE(p) - move_bytes;
132 				OFFSET(p) = off;
133 			} else
134 				p[n - 2] = FULL_KEY;
135 		}
136 		p = (u_int16_t *)bufp->page;
137 		cp = bufp->page;
138 		bufp->flags |= BUF_MOD;
139 	}
140 
141 	/* Now move the data */
142 	for (space = FREESPACE(p) - BIGOVERHEAD; val_size;
143 	    space = FREESPACE(p) - BIGOVERHEAD) {
144 		move_bytes = MIN(space, val_size);
145 		/*
146 		 * Here's the hack to make sure that if the data ends on the
147 		 * same page as the key ends, FREESPACE is at least one.
148 		 */
149 		if (space == val_size && val_size == val->size)
150 			move_bytes--;
151 		off = OFFSET(p) - move_bytes;
152 		memmove(cp + off, val_data, move_bytes);
153 		val_size -= move_bytes;
154 		val_data += move_bytes;
155 		n = p[0];
156 		p[++n] = off;
157 		p[0] = ++n;
158 		FREESPACE(p) = off - PAGE_META(n);
159 		OFFSET(p) = off;
160 		if (val_size) {
161 			p[n] = FULL_KEY;
162 			bufp = __add_ovflpage(hashp, bufp);
163 			if (!bufp)
164 				return (-1);
165 			cp = bufp->page;
166 			p = (u_int16_t *)cp;
167 		} else
168 			p[n] = FULL_KEY_DATA;
169 		bufp->flags |= BUF_MOD;
170 	}
171 	return (0);
172 }
173 
174 /*
175  * Called when bufp's page  contains a partial key (index should be 1)
176  *
177  * All pages in the big key/data pair except bufp are freed.  We cannot
178  * free bufp because the page pointing to it is lost and we can't get rid
179  * of its pointer.
180  *
181  * Returns:
182  * 0 => OK
183  *-1 => ERROR
184  */
185 extern int
186 __big_delete(hashp, bufp)
187 	HTAB *hashp;
188 	BUFHEAD *bufp;
189 {
190 	BUFHEAD *last_bfp, *rbufp;
191 	u_int16_t *bp, pageno;
192 	int key_done, n;
193 
194 	rbufp = bufp;
195 	last_bfp = NULL;
196 	bp = (u_int16_t *)bufp->page;
197 	pageno = 0;
198 	key_done = 0;
199 
200 	while (!key_done || (bp[2] != FULL_KEY_DATA)) {
201 		if (bp[2] == FULL_KEY || bp[2] == FULL_KEY_DATA)
202 			key_done = 1;
203 
204 		/*
205 		 * If there is freespace left on a FULL_KEY_DATA page, then
206 		 * the data is short and fits entirely on this page, and this
207 		 * is the last page.
208 		 */
209 		if (bp[2] == FULL_KEY_DATA && FREESPACE(bp))
210 			break;
211 		pageno = bp[bp[0] - 1];
212 		rbufp->flags |= BUF_MOD;
213 		rbufp = __get_buf(hashp, pageno, rbufp, 0);
214 		if (last_bfp)
215 			__free_ovflpage(hashp, last_bfp);
216 		last_bfp = rbufp;
217 		if (!rbufp)
218 			return (-1);		/* Error. */
219 		bp = (u_int16_t *)rbufp->page;
220 	}
221 
222 	/*
223 	 * If we get here then rbufp points to the last page of the big
224 	 * key/data pair.  Bufp points to the first one -- it should now be
225 	 * empty pointing to the next page after this pair.  Can't free it
226 	 * because we don't have the page pointing to it.
227 	 */
228 
229 	/* This is information from the last page of the pair. */
230 	n = bp[0];
231 	pageno = bp[n - 1];
232 
233 	/* Now, bp is the first page of the pair. */
234 	bp = (u_int16_t *)bufp->page;
235 	if (n > 2) {
236 		/* There is an overflow page. */
237 		bp[1] = pageno;
238 		bp[2] = OVFLPAGE;
239 		bufp->ovfl = rbufp->ovfl;
240 	} else
241 		/* This is the last page. */
242 		bufp->ovfl = NULL;
243 	n -= 2;
244 	bp[0] = n;
245 	FREESPACE(bp) = hashp->BSIZE - PAGE_META(n);
246 	OFFSET(bp) = hashp->BSIZE - 1;
247 
248 	bufp->flags |= BUF_MOD;
249 	if (rbufp)
250 		__free_ovflpage(hashp, rbufp);
251 	if (last_bfp != rbufp)
252 		__free_ovflpage(hashp, last_bfp);
253 
254 	hashp->NKEYS--;
255 	return (0);
256 }
257 /*
258  * Returns:
259  *  0 = key not found
260  * -1 = get next overflow page
261  * -2 means key not found and this is big key/data
262  * -3 error
263  */
264 extern int
265 __find_bigpair(hashp, bufp, ndx, key, size)
266 	HTAB *hashp;
267 	BUFHEAD *bufp;
268 	int ndx;
269 	char *key;
270 	int size;
271 {
272 	u_int16_t *bp;
273 	char *p;
274 	int ksize;
275 	u_int16_t bytes;
276 	char *kkey;
277 
278 	bp = (u_int16_t *)bufp->page;
279 	p = bufp->page;
280 	ksize = size;
281 	kkey = key;
282 
283 	for (bytes = hashp->BSIZE - bp[ndx];
284 	    bytes <= size && bp[ndx + 1] == PARTIAL_KEY;
285 	    bytes = hashp->BSIZE - bp[ndx]) {
286 		if (memcmp(p + bp[ndx], kkey, bytes))
287 			return (-2);
288 		kkey += bytes;
289 		ksize -= bytes;
290 		bufp = __get_buf(hashp, bp[ndx + 2], bufp, 0);
291 		if (!bufp)
292 			return (-3);
293 		p = bufp->page;
294 		bp = (u_int16_t *)p;
295 		ndx = 1;
296 	}
297 
298 	if (bytes != ksize || memcmp(p + bp[ndx], kkey, bytes)) {
299 #ifdef HASH_STATISTICS
300 		++hash_collisions;
301 #endif
302 		return (-2);
303 	} else
304 		return (ndx);
305 }
306 
307 /*
308  * Given the buffer pointer of the first overflow page of a big pair,
309  * find the end of the big pair
310  *
311  * This will set bpp to the buffer header of the last page of the big pair.
312  * It will return the pageno of the overflow page following the last page
313  * of the pair; 0 if there isn't any (i.e. big pair is the last key in the
314  * bucket)
315  */
316 extern u_int16_t
317 __find_last_page(hashp, bpp)
318 	HTAB *hashp;
319 	BUFHEAD **bpp;
320 {
321 	BUFHEAD *bufp;
322 	u_int16_t *bp, pageno;
323 	int n;
324 
325 	bufp = *bpp;
326 	bp = (u_int16_t *)bufp->page;
327 	for (;;) {
328 		n = bp[0];
329 
330 		/*
331 		 * This is the last page if: the tag is FULL_KEY_DATA and
332 		 * either only 2 entries OVFLPAGE marker is explicit there
333 		 * is freespace on the page.
334 		 */
335 		if (bp[2] == FULL_KEY_DATA &&
336 		    ((n == 2) || (bp[n] == OVFLPAGE) || (FREESPACE(bp))))
337 			break;
338 
339 		pageno = bp[n - 1];
340 		bufp = __get_buf(hashp, pageno, bufp, 0);
341 		if (!bufp)
342 			return (0);	/* Need to indicate an error! */
343 		bp = (u_int16_t *)bufp->page;
344 	}
345 
346 	*bpp = bufp;
347 	if (bp[0] > 2)
348 		return (bp[3]);
349 	else
350 		return (0);
351 }
352 
353 /*
354  * Return the data for the key/data pair that begins on this page at this
355  * index (index should always be 1).
356  */
357 extern int
358 __big_return(hashp, bufp, ndx, val, set_current)
359 	HTAB *hashp;
360 	BUFHEAD *bufp;
361 	int ndx;
362 	DBT *val;
363 	int set_current;
364 {
365 	BUFHEAD *save_p;
366 	u_int16_t *bp, len, off, save_addr;
367 	char *tp;
368 
369 	bp = (u_int16_t *)bufp->page;
370 	while (bp[ndx + 1] == PARTIAL_KEY) {
371 		bufp = __get_buf(hashp, bp[bp[0] - 1], bufp, 0);
372 		if (!bufp)
373 			return (-1);
374 		bp = (u_int16_t *)bufp->page;
375 		ndx = 1;
376 	}
377 
378 	if (bp[ndx + 1] == FULL_KEY) {
379 		bufp = __get_buf(hashp, bp[bp[0] - 1], bufp, 0);
380 		if (!bufp)
381 			return (-1);
382 		bp = (u_int16_t *)bufp->page;
383 		save_p = bufp;
384 		save_addr = save_p->addr;
385 		off = bp[1];
386 		len = 0;
387 	} else
388 		if (!FREESPACE(bp)) {
389 			/*
390 			 * This is a hack.  We can't distinguish between
391 			 * FULL_KEY_DATA that contains complete data or
392 			 * incomplete data, so we require that if the data
393 			 * is complete, there is at least 1 byte of free
394 			 * space left.
395 			 */
396 			off = bp[bp[0]];
397 			len = bp[1] - off;
398 			save_p = bufp;
399 			save_addr = bufp->addr;
400 			bufp = __get_buf(hashp, bp[bp[0] - 1], bufp, 0);
401 			if (!bufp)
402 				return (-1);
403 			bp = (u_int16_t *)bufp->page;
404 		} else {
405 			/* The data is all on one page. */
406 			tp = (char *)bp;
407 			off = bp[bp[0]];
408 			val->data = (u_char *)tp + off;
409 			val->size = bp[1] - off;
410 			if (set_current) {
411 				if (bp[0] == 2) {	/* No more buckets in
412 							 * chain */
413 					hashp->cpage = NULL;
414 					hashp->cbucket++;
415 					hashp->cndx = 1;
416 				} else {
417 					hashp->cpage = __get_buf(hashp,
418 					    bp[bp[0] - 1], bufp, 0);
419 					if (!hashp->cpage)
420 						return (-1);
421 					hashp->cndx = 1;
422 					if (!((u_int16_t *)
423 					    hashp->cpage->page)[0]) {
424 						hashp->cbucket++;
425 						hashp->cpage = NULL;
426 					}
427 				}
428 			}
429 			return (0);
430 		}
431 
432 	val->size = collect_data(hashp, bufp, (int)len, set_current);
433 	if (val->size == -1)
434 		return (-1);
435 	if (save_p->addr != save_addr) {
436 		/* We are pretty short on buffers. */
437 		errno = EINVAL;			/* OUT OF BUFFERS */
438 		return (-1);
439 	}
440 	memmove(hashp->tmp_buf, (save_p->page) + off, len);
441 	val->data = (u_char *)hashp->tmp_buf;
442 	return (0);
443 }
444 /*
445  * Count how big the total datasize is by recursing through the pages.  Then
446  * allocate a buffer and copy the data as you recurse up.
447  */
448 static int
449 collect_data(hashp, bufp, len, set)
450 	HTAB *hashp;
451 	BUFHEAD *bufp;
452 	int len, set;
453 {
454 	u_int16_t *bp;
455 	char *p;
456 	BUFHEAD *xbp;
457 	u_int16_t save_addr;
458 	int mylen, totlen;
459 
460 	p = bufp->page;
461 	bp = (u_int16_t *)p;
462 	mylen = hashp->BSIZE - bp[1];
463 	save_addr = bufp->addr;
464 
465 	if (bp[2] == FULL_KEY_DATA) {		/* End of Data */
466 		totlen = len + mylen;
467 		if (hashp->tmp_buf)
468 			free(hashp->tmp_buf);
469 		if ((hashp->tmp_buf = (char *)malloc(totlen)) == NULL)
470 			return (-1);
471 		if (set) {
472 			hashp->cndx = 1;
473 			if (bp[0] == 2) {	/* No more buckets in chain */
474 				hashp->cpage = NULL;
475 				hashp->cbucket++;
476 			} else {
477 				hashp->cpage =
478 				    __get_buf(hashp, bp[bp[0] - 1], bufp, 0);
479 				if (!hashp->cpage)
480 					return (-1);
481 				else if (!((u_int16_t *)hashp->cpage->page)[0]) {
482 					hashp->cbucket++;
483 					hashp->cpage = NULL;
484 				}
485 			}
486 		}
487 	} else {
488 		xbp = __get_buf(hashp, bp[bp[0] - 1], bufp, 0);
489 		if (!xbp || ((totlen =
490 		    collect_data(hashp, xbp, len + mylen, set)) < 1))
491 			return (-1);
492 	}
493 	if (bufp->addr != save_addr) {
494 		errno = EINVAL;			/* Out of buffers. */
495 		return (-1);
496 	}
497 	memmove(&hashp->tmp_buf[len], (bufp->page) + bp[1], mylen);
498 	return (totlen);
499 }
500 
501 /*
502  * Fill in the key and data for this big pair.
503  */
504 extern int
505 __big_keydata(hashp, bufp, key, val, set)
506 	HTAB *hashp;
507 	BUFHEAD *bufp;
508 	DBT *key, *val;
509 	int set;
510 {
511 	key->size = collect_key(hashp, bufp, 0, val, set);
512 	if (key->size == -1)
513 		return (-1);
514 	key->data = (u_char *)hashp->tmp_key;
515 	return (0);
516 }
517 
518 /*
519  * Count how big the total key size is by recursing through the pages.  Then
520  * collect the data, allocate a buffer and copy the key as you recurse up.
521  */
522 static int
523 collect_key(hashp, bufp, len, val, set)
524 	HTAB *hashp;
525 	BUFHEAD *bufp;
526 	int len;
527 	DBT *val;
528 	int set;
529 {
530 	BUFHEAD *xbp;
531 	char *p;
532 	int mylen, totlen;
533 	u_int16_t *bp, save_addr;
534 
535 	p = bufp->page;
536 	bp = (u_int16_t *)p;
537 	mylen = hashp->BSIZE - bp[1];
538 
539 	save_addr = bufp->addr;
540 	totlen = len + mylen;
541 	if (bp[2] == FULL_KEY || bp[2] == FULL_KEY_DATA) {    /* End of Key. */
542 		if (hashp->tmp_key != NULL)
543 			free(hashp->tmp_key);
544 		if ((hashp->tmp_key = (char *)malloc(totlen)) == NULL)
545 			return (-1);
546 		if (__big_return(hashp, bufp, 1, val, set))
547 			return (-1);
548 	} else {
549 		xbp = __get_buf(hashp, bp[bp[0] - 1], bufp, 0);
550 		if (!xbp || ((totlen =
551 		    collect_key(hashp, xbp, totlen, val, set)) < 1))
552 			return (-1);
553 	}
554 	if (bufp->addr != save_addr) {
555 		errno = EINVAL;		/* MIS -- OUT OF BUFFERS */
556 		return (-1);
557 	}
558 	memmove(&hashp->tmp_key[len], (bufp->page) + bp[1], mylen);
559 	return (totlen);
560 }
561 
562 /*
563  * Returns:
564  *  0 => OK
565  * -1 => error
566  */
567 extern int
568 __big_split(hashp, op, np, big_keyp, addr, obucket, ret)
569 	HTAB *hashp;
570 	BUFHEAD *op;	/* Pointer to where to put keys that go in old bucket */
571 	BUFHEAD *np;	/* Pointer to new bucket page */
572 			/* Pointer to first page containing the big key/data */
573 	BUFHEAD *big_keyp;
574 	int addr;	/* Address of big_keyp */
575 	u_int32_t   obucket;/* Old Bucket */
576 	SPLIT_RETURN *ret;
577 {
578 	BUFHEAD *tmpp;
579 	u_int16_t *tp;
580 	BUFHEAD *bp;
581 	DBT key, val;
582 	u_int32_t change;
583 	u_int16_t free_space, n, off;
584 
585 	bp = big_keyp;
586 
587 	/* Now figure out where the big key/data goes */
588 	if (__big_keydata(hashp, big_keyp, &key, &val, 0))
589 		return (-1);
590 	change = (__call_hash(hashp, key.data, key.size) != obucket);
591 
592 	if ( (ret->next_addr = __find_last_page(hashp, &big_keyp)) ) {
593 		if (!(ret->nextp =
594 		    __get_buf(hashp, ret->next_addr, big_keyp, 0)))
595 			return (-1);;
596 	} else
597 		ret->nextp = NULL;
598 
599 	/* Now make one of np/op point to the big key/data pair */
600 #ifdef DEBUG
601 	assert(np->ovfl == NULL);
602 #endif
603 	if (change)
604 		tmpp = np;
605 	else
606 		tmpp = op;
607 
608 	tmpp->flags |= BUF_MOD;
609 #ifdef DEBUG1
610 	(void)fprintf(stderr,
611 	    "BIG_SPLIT: %d->ovfl was %d is now %d\n", tmpp->addr,
612 	    (tmpp->ovfl ? tmpp->ovfl->addr : 0), (bp ? bp->addr : 0));
613 #endif
614 	tmpp->ovfl = bp;	/* one of op/np point to big_keyp */
615 	tp = (u_int16_t *)tmpp->page;
616 #ifdef DEBUG
617 	assert(FREESPACE(tp) >= OVFLSIZE);
618 #endif
619 	n = tp[0];
620 	off = OFFSET(tp);
621 	free_space = FREESPACE(tp);
622 	tp[++n] = (u_int16_t)addr;
623 	tp[++n] = OVFLPAGE;
624 	tp[0] = n;
625 	OFFSET(tp) = off;
626 	FREESPACE(tp) = free_space - OVFLSIZE;
627 
628 	/*
629 	 * Finally, set the new and old return values. BIG_KEYP contains a
630 	 * pointer to the last page of the big key_data pair. Make sure that
631 	 * big_keyp has no following page (2 elements) or create an empty
632 	 * following page.
633 	 */
634 
635 	ret->newp = np;
636 	ret->oldp = op;
637 
638 	tp = (u_int16_t *)big_keyp->page;
639 	big_keyp->flags |= BUF_MOD;
640 	if (tp[0] > 2) {
641 		/*
642 		 * There may be either one or two offsets on this page.  If
643 		 * there is one, then the overflow page is linked on normally
644 		 * and tp[4] is OVFLPAGE.  If there are two, tp[4] contains
645 		 * the second offset and needs to get stuffed in after the
646 		 * next overflow page is added.
647 		 */
648 		n = tp[4];
649 		free_space = FREESPACE(tp);
650 		off = OFFSET(tp);
651 		tp[0] -= 2;
652 		FREESPACE(tp) = free_space + OVFLSIZE;
653 		OFFSET(tp) = off;
654 		tmpp = __add_ovflpage(hashp, big_keyp);
655 		if (!tmpp)
656 			return (-1);
657 		tp[4] = n;
658 	} else
659 		tmpp = big_keyp;
660 
661 	if (change)
662 		ret->newp = tmpp;
663 	else
664 		ret->oldp = tmpp;
665 	return (0);
666 }
667