1 /*-
2 * SPDX-License-Identifier: BSD-3-Clause
3 *
4 * Copyright (c) 1990, 1993, 1994
5 * The Regents of the University of California. All rights reserved.
6 *
7 * This code is derived from software contributed to Berkeley by
8 * Margo Seltzer.
9 *
10 * Redistribution and use in source and binary forms, with or without
11 * modification, are permitted provided that the following conditions
12 * are met:
13 * 1. Redistributions of source code must retain the above copyright
14 * notice, this list of conditions and the following disclaimer.
15 * 2. Redistributions in binary form must reproduce the above copyright
16 * notice, this list of conditions and the following disclaimer in the
17 * documentation and/or other materials provided with the distribution.
18 * 3. Neither the name of the University nor the names of its contributors
19 * may be used to endorse or promote products derived from this software
20 * without specific prior written permission.
21 *
22 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
23 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
24 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
25 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
26 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
27 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
28 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
29 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
30 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
31 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
32 * SUCH DAMAGE.
33 */
34
35 #if defined(LIBC_SCCS) && !defined(lint)
36 static char sccsid[] = "@(#)hash_page.c 8.7 (Berkeley) 8/16/94";
37 #endif /* LIBC_SCCS and not lint */
38 #include <sys/cdefs.h>
39 __FBSDID("$FreeBSD$");
40
41 /*
42 * PACKAGE: hashing
43 *
44 * DESCRIPTION:
45 * Page manipulation for hashing package.
46 *
47 * ROUTINES:
48 *
49 * External
50 * __get_page
51 * __add_ovflpage
52 * Internal
53 * overflow_page
54 * open_temp
55 */
56
57 #include "namespace.h"
58 #include <sys/param.h>
59
60 #include <errno.h>
61 #include <fcntl.h>
62 #include <signal.h>
63 #include <stdio.h>
64 #include <stdlib.h>
65 #include <string.h>
66 #include <unistd.h>
67 #ifdef DEBUG
68 #include <assert.h>
69 #endif
70 #include "un-namespace.h"
71 #include "libc_private.h"
72
73 #include <db.h>
74 #include "hash.h"
75 #include "page.h"
76 #include "extern.h"
77
78 static u_int32_t *fetch_bitmap(HTAB *, int);
79 static u_int32_t first_free(u_int32_t);
80 static int open_temp(HTAB *);
81 static u_int16_t overflow_page(HTAB *);
82 static void putpair(char *, const DBT *, const DBT *);
83 static void squeeze_key(u_int16_t *, const DBT *, const DBT *);
84 static int ugly_split(HTAB *, u_int32_t, BUFHEAD *, BUFHEAD *, int, int);
85
86 #define PAGE_INIT(P) { \
87 ((u_int16_t *)(P))[0] = 0; \
88 ((u_int16_t *)(P))[1] = hashp->BSIZE - 3 * sizeof(u_int16_t); \
89 ((u_int16_t *)(P))[2] = hashp->BSIZE; \
90 }
91
92 /*
93 * This is called AFTER we have verified that there is room on the page for
94 * the pair (PAIRFITS has returned true) so we go right ahead and start moving
95 * stuff on.
96 */
97 static void
putpair(char * p,const DBT * key,const DBT * val)98 putpair(char *p, const DBT *key, const DBT *val)
99 {
100 u_int16_t *bp, n, off;
101
102 bp = (u_int16_t *)p;
103
104 /* Enter the key first. */
105 n = bp[0];
106
107 off = OFFSET(bp) - key->size;
108 memmove(p + off, key->data, key->size);
109 bp[++n] = off;
110
111 /* Now the data. */
112 off -= val->size;
113 memmove(p + off, val->data, val->size);
114 bp[++n] = off;
115
116 /* Adjust page info. */
117 bp[0] = n;
118 bp[n + 1] = off - ((n + 3) * sizeof(u_int16_t));
119 bp[n + 2] = off;
120 }
121
122 /*
123 * Returns:
124 * 0 OK
125 * -1 error
126 */
127 int
__delpair(HTAB * hashp,BUFHEAD * bufp,int ndx)128 __delpair(HTAB *hashp, BUFHEAD *bufp, int ndx)
129 {
130 u_int16_t *bp, newoff, pairlen;
131 int n;
132
133 bp = (u_int16_t *)bufp->page;
134 n = bp[0];
135
136 if (bp[ndx + 1] < REAL_KEY)
137 return (__big_delete(hashp, bufp));
138 if (ndx != 1)
139 newoff = bp[ndx - 1];
140 else
141 newoff = hashp->BSIZE;
142 pairlen = newoff - bp[ndx + 1];
143
144 if (ndx != (n - 1)) {
145 /* Hard Case -- need to shuffle keys */
146 int i;
147 char *src = bufp->page + (int)OFFSET(bp);
148 char *dst = src + (int)pairlen;
149 memmove(dst, src, bp[ndx + 1] - OFFSET(bp));
150
151 /* Now adjust the pointers */
152 for (i = ndx + 2; i <= n; i += 2) {
153 if (bp[i + 1] == OVFLPAGE) {
154 bp[i - 2] = bp[i];
155 bp[i - 1] = bp[i + 1];
156 } else {
157 bp[i - 2] = bp[i] + pairlen;
158 bp[i - 1] = bp[i + 1] + pairlen;
159 }
160 }
161 if (ndx == hashp->cndx) {
162 /*
163 * We just removed pair we were "pointing" to.
164 * By moving back the cndx we ensure subsequent
165 * hash_seq() calls won't skip over any entries.
166 */
167 hashp->cndx -= 2;
168 }
169 }
170 /* Finally adjust the page data */
171 bp[n] = OFFSET(bp) + pairlen;
172 bp[n - 1] = bp[n + 1] + pairlen + 2 * sizeof(u_int16_t);
173 bp[0] = n - 2;
174 hashp->NKEYS--;
175
176 bufp->flags |= BUF_MOD;
177 return (0);
178 }
179 /*
180 * Returns:
181 * 0 ==> OK
182 * -1 ==> Error
183 */
184 int
__split_page(HTAB * hashp,u_int32_t obucket,u_int32_t nbucket)185 __split_page(HTAB *hashp, u_int32_t obucket, u_int32_t nbucket)
186 {
187 BUFHEAD *new_bufp, *old_bufp;
188 u_int16_t *ino;
189 char *np;
190 DBT key, val;
191 int n, ndx, retval;
192 u_int16_t copyto, diff, off, moved;
193 char *op;
194
195 copyto = (u_int16_t)hashp->BSIZE;
196 off = (u_int16_t)hashp->BSIZE;
197 old_bufp = __get_buf(hashp, obucket, NULL, 0);
198 if (old_bufp == NULL)
199 return (-1);
200 new_bufp = __get_buf(hashp, nbucket, NULL, 0);
201 if (new_bufp == NULL)
202 return (-1);
203
204 old_bufp->flags |= (BUF_MOD | BUF_PIN);
205 new_bufp->flags |= (BUF_MOD | BUF_PIN);
206
207 ino = (u_int16_t *)(op = old_bufp->page);
208 np = new_bufp->page;
209
210 moved = 0;
211
212 for (n = 1, ndx = 1; n < ino[0]; n += 2) {
213 if (ino[n + 1] < REAL_KEY) {
214 retval = ugly_split(hashp, obucket, old_bufp, new_bufp,
215 (int)copyto, (int)moved);
216 old_bufp->flags &= ~BUF_PIN;
217 new_bufp->flags &= ~BUF_PIN;
218 return (retval);
219
220 }
221 key.data = (u_char *)op + ino[n];
222 key.size = off - ino[n];
223
224 if (__call_hash(hashp, key.data, key.size) == obucket) {
225 /* Don't switch page */
226 diff = copyto - off;
227 if (diff) {
228 copyto = ino[n + 1] + diff;
229 memmove(op + copyto, op + ino[n + 1],
230 off - ino[n + 1]);
231 ino[ndx] = copyto + ino[n] - ino[n + 1];
232 ino[ndx + 1] = copyto;
233 } else
234 copyto = ino[n + 1];
235 ndx += 2;
236 } else {
237 /* Switch page */
238 val.data = (u_char *)op + ino[n + 1];
239 val.size = ino[n] - ino[n + 1];
240 putpair(np, &key, &val);
241 moved += 2;
242 }
243
244 off = ino[n + 1];
245 }
246
247 /* Now clean up the page */
248 ino[0] -= moved;
249 FREESPACE(ino) = copyto - sizeof(u_int16_t) * (ino[0] + 3);
250 OFFSET(ino) = copyto;
251
252 #ifdef DEBUG3
253 (void)fprintf(stderr, "split %d/%d\n",
254 ((u_int16_t *)np)[0] / 2,
255 ((u_int16_t *)op)[0] / 2);
256 #endif
257 /* unpin both pages */
258 old_bufp->flags &= ~BUF_PIN;
259 new_bufp->flags &= ~BUF_PIN;
260 return (0);
261 }
262
263 /*
264 * Called when we encounter an overflow or big key/data page during split
265 * handling. This is special cased since we have to begin checking whether
266 * the key/data pairs fit on their respective pages and because we may need
267 * overflow pages for both the old and new pages.
268 *
269 * The first page might be a page with regular key/data pairs in which case
270 * we have a regular overflow condition and just need to go on to the next
271 * page or it might be a big key/data pair in which case we need to fix the
272 * big key/data pair.
273 *
274 * Returns:
275 * 0 ==> success
276 * -1 ==> failure
277 */
278 static int
ugly_split(HTAB * hashp,u_int32_t obucket,BUFHEAD * old_bufp,BUFHEAD * new_bufp,int copyto,int moved)279 ugly_split(HTAB *hashp,
280 u_int32_t obucket, /* Same as __split_page. */
281 BUFHEAD *old_bufp,
282 BUFHEAD *new_bufp,
283 int copyto, /* First byte on page which contains key/data values. */
284 int moved) /* Number of pairs moved to new page. */
285 {
286 BUFHEAD *bufp; /* Buffer header for ino */
287 u_int16_t *ino; /* Page keys come off of */
288 u_int16_t *np; /* New page */
289 u_int16_t *op; /* Page keys go on to if they aren't moving */
290
291 BUFHEAD *last_bfp; /* Last buf header OVFL needing to be freed */
292 DBT key, val;
293 SPLIT_RETURN ret;
294 u_int16_t n, off, ov_addr, scopyto;
295 char *cino; /* Character value of ino */
296
297 bufp = old_bufp;
298 ino = (u_int16_t *)old_bufp->page;
299 np = (u_int16_t *)new_bufp->page;
300 op = (u_int16_t *)old_bufp->page;
301 last_bfp = NULL;
302 scopyto = (u_int16_t)copyto; /* ANSI */
303
304 n = ino[0] - 1;
305 while (n < ino[0]) {
306 if (ino[2] < REAL_KEY && ino[2] != OVFLPAGE) {
307 if (__big_split(hashp, old_bufp,
308 new_bufp, bufp, bufp->addr, obucket, &ret))
309 return (-1);
310 old_bufp = ret.oldp;
311 if (!old_bufp)
312 return (-1);
313 op = (u_int16_t *)old_bufp->page;
314 new_bufp = ret.newp;
315 if (!new_bufp)
316 return (-1);
317 np = (u_int16_t *)new_bufp->page;
318 bufp = ret.nextp;
319 if (!bufp)
320 return (0);
321 cino = (char *)bufp->page;
322 ino = (u_int16_t *)cino;
323 last_bfp = ret.nextp;
324 } else if (ino[n + 1] == OVFLPAGE) {
325 ov_addr = ino[n];
326 /*
327 * Fix up the old page -- the extra 2 are the fields
328 * which contained the overflow information.
329 */
330 ino[0] -= (moved + 2);
331 FREESPACE(ino) =
332 scopyto - sizeof(u_int16_t) * (ino[0] + 3);
333 OFFSET(ino) = scopyto;
334
335 bufp = __get_buf(hashp, ov_addr, bufp, 0);
336 if (!bufp)
337 return (-1);
338
339 ino = (u_int16_t *)bufp->page;
340 n = 1;
341 scopyto = hashp->BSIZE;
342 moved = 0;
343
344 if (last_bfp)
345 __free_ovflpage(hashp, last_bfp);
346 last_bfp = bufp;
347 }
348 /* Move regular sized pairs of there are any */
349 off = hashp->BSIZE;
350 for (n = 1; (n < ino[0]) && (ino[n + 1] >= REAL_KEY); n += 2) {
351 cino = (char *)ino;
352 key.data = (u_char *)cino + ino[n];
353 key.size = off - ino[n];
354 val.data = (u_char *)cino + ino[n + 1];
355 val.size = ino[n] - ino[n + 1];
356 off = ino[n + 1];
357
358 if (__call_hash(hashp, key.data, key.size) == obucket) {
359 /* Keep on old page */
360 if (PAIRFITS(op, (&key), (&val)))
361 putpair((char *)op, &key, &val);
362 else {
363 old_bufp =
364 __add_ovflpage(hashp, old_bufp);
365 if (!old_bufp)
366 return (-1);
367 op = (u_int16_t *)old_bufp->page;
368 putpair((char *)op, &key, &val);
369 }
370 old_bufp->flags |= BUF_MOD;
371 } else {
372 /* Move to new page */
373 if (PAIRFITS(np, (&key), (&val)))
374 putpair((char *)np, &key, &val);
375 else {
376 new_bufp =
377 __add_ovflpage(hashp, new_bufp);
378 if (!new_bufp)
379 return (-1);
380 np = (u_int16_t *)new_bufp->page;
381 putpair((char *)np, &key, &val);
382 }
383 new_bufp->flags |= BUF_MOD;
384 }
385 }
386 }
387 if (last_bfp)
388 __free_ovflpage(hashp, last_bfp);
389 return (0);
390 }
391
392 /*
393 * Add the given pair to the page
394 *
395 * Returns:
396 * 0 ==> OK
397 * 1 ==> failure
398 */
399 int
__addel(HTAB * hashp,BUFHEAD * bufp,const DBT * key,const DBT * val)400 __addel(HTAB *hashp, BUFHEAD *bufp, const DBT *key, const DBT *val)
401 {
402 u_int16_t *bp, *sop;
403 int do_expand;
404
405 bp = (u_int16_t *)bufp->page;
406 do_expand = 0;
407 while (bp[0] && (bp[2] < REAL_KEY || bp[bp[0]] < REAL_KEY))
408 /* Exception case */
409 if (bp[2] == FULL_KEY_DATA && bp[0] == 2)
410 /* This is the last page of a big key/data pair
411 and we need to add another page */
412 break;
413 else if (bp[2] < REAL_KEY && bp[bp[0]] != OVFLPAGE) {
414 bufp = __get_buf(hashp, bp[bp[0] - 1], bufp, 0);
415 if (!bufp)
416 return (-1);
417 bp = (u_int16_t *)bufp->page;
418 } else if (bp[bp[0]] != OVFLPAGE) {
419 /* Short key/data pairs, no more pages */
420 break;
421 } else {
422 /* Try to squeeze key on this page */
423 if (bp[2] >= REAL_KEY &&
424 FREESPACE(bp) >= PAIRSIZE(key, val)) {
425 squeeze_key(bp, key, val);
426 goto stats;
427 } else {
428 bufp = __get_buf(hashp, bp[bp[0] - 1], bufp, 0);
429 if (!bufp)
430 return (-1);
431 bp = (u_int16_t *)bufp->page;
432 }
433 }
434
435 if (PAIRFITS(bp, key, val))
436 putpair(bufp->page, key, val);
437 else {
438 do_expand = 1;
439 bufp = __add_ovflpage(hashp, bufp);
440 if (!bufp)
441 return (-1);
442 sop = (u_int16_t *)bufp->page;
443
444 if (PAIRFITS(sop, key, val))
445 putpair((char *)sop, key, val);
446 else
447 if (__big_insert(hashp, bufp, key, val))
448 return (-1);
449 }
450 stats:
451 bufp->flags |= BUF_MOD;
452 /*
453 * If the average number of keys per bucket exceeds the fill factor,
454 * expand the table.
455 */
456 hashp->NKEYS++;
457 if (do_expand ||
458 (hashp->NKEYS / (hashp->MAX_BUCKET + 1) > hashp->FFACTOR))
459 return (__expand_table(hashp));
460 return (0);
461 }
462
463 /*
464 *
465 * Returns:
466 * pointer on success
467 * NULL on error
468 */
469 BUFHEAD *
__add_ovflpage(HTAB * hashp,BUFHEAD * bufp)470 __add_ovflpage(HTAB *hashp, BUFHEAD *bufp)
471 {
472 u_int16_t *sp, ndx, ovfl_num;
473 #ifdef DEBUG1
474 int tmp1, tmp2;
475 #endif
476 sp = (u_int16_t *)bufp->page;
477
478 /* Check if we are dynamically determining the fill factor */
479 if (hashp->FFACTOR == DEF_FFACTOR) {
480 hashp->FFACTOR = sp[0] >> 1;
481 if (hashp->FFACTOR < MIN_FFACTOR)
482 hashp->FFACTOR = MIN_FFACTOR;
483 }
484 bufp->flags |= BUF_MOD;
485 ovfl_num = overflow_page(hashp);
486 #ifdef DEBUG1
487 tmp1 = bufp->addr;
488 tmp2 = bufp->ovfl ? bufp->ovfl->addr : 0;
489 #endif
490 if (!ovfl_num || !(bufp->ovfl = __get_buf(hashp, ovfl_num, bufp, 1)))
491 return (NULL);
492 bufp->ovfl->flags |= BUF_MOD;
493 #ifdef DEBUG1
494 (void)fprintf(stderr, "ADDOVFLPAGE: %d->ovfl was %d is now %d\n",
495 tmp1, tmp2, bufp->ovfl->addr);
496 #endif
497 ndx = sp[0];
498 /*
499 * Since a pair is allocated on a page only if there's room to add
500 * an overflow page, we know that the OVFL information will fit on
501 * the page.
502 */
503 sp[ndx + 4] = OFFSET(sp);
504 sp[ndx + 3] = FREESPACE(sp) - OVFLSIZE;
505 sp[ndx + 1] = ovfl_num;
506 sp[ndx + 2] = OVFLPAGE;
507 sp[0] = ndx + 2;
508 #ifdef HASH_STATISTICS
509 hash_overflows++;
510 #endif
511 return (bufp->ovfl);
512 }
513
514 /*
515 * Returns:
516 * 0 indicates SUCCESS
517 * -1 indicates FAILURE
518 */
519 int
__get_page(HTAB * hashp,char * p,u_int32_t bucket,int is_bucket,int is_disk,int is_bitmap)520 __get_page(HTAB *hashp, char *p, u_int32_t bucket, int is_bucket, int is_disk,
521 int is_bitmap)
522 {
523 int fd, page, size, rsize;
524 u_int16_t *bp;
525
526 fd = hashp->fp;
527 size = hashp->BSIZE;
528
529 if ((fd == -1) || !is_disk) {
530 PAGE_INIT(p);
531 return (0);
532 }
533 if (is_bucket)
534 page = BUCKET_TO_PAGE(bucket);
535 else
536 page = OADDR_TO_PAGE(bucket);
537 if ((rsize = pread(fd, p, size, (off_t)page << hashp->BSHIFT)) == -1)
538 return (-1);
539 bp = (u_int16_t *)p;
540 if (!rsize)
541 bp[0] = 0; /* We hit the EOF, so initialize a new page */
542 else
543 if (rsize != size) {
544 errno = EFTYPE;
545 return (-1);
546 }
547 if (!is_bitmap && !bp[0]) {
548 PAGE_INIT(p);
549 } else
550 if (hashp->LORDER != BYTE_ORDER) {
551 int i, max;
552
553 if (is_bitmap) {
554 max = hashp->BSIZE >> 2; /* divide by 4 */
555 for (i = 0; i < max; i++)
556 M_32_SWAP(((int *)p)[i]);
557 } else {
558 M_16_SWAP(bp[0]);
559 max = bp[0] + 2;
560 for (i = 1; i <= max; i++)
561 M_16_SWAP(bp[i]);
562 }
563 }
564 return (0);
565 }
566
567 /*
568 * Write page p to disk
569 *
570 * Returns:
571 * 0 ==> OK
572 * -1 ==>failure
573 */
574 int
__put_page(HTAB * hashp,char * p,u_int32_t bucket,int is_bucket,int is_bitmap)575 __put_page(HTAB *hashp, char *p, u_int32_t bucket, int is_bucket, int is_bitmap)
576 {
577 int fd, page, size;
578 ssize_t wsize;
579 char pbuf[MAX_BSIZE];
580
581 size = hashp->BSIZE;
582 if ((hashp->fp == -1) && open_temp(hashp))
583 return (-1);
584 fd = hashp->fp;
585
586 if (hashp->LORDER != BYTE_ORDER) {
587 int i, max;
588
589 memcpy(pbuf, p, size);
590 if (is_bitmap) {
591 max = hashp->BSIZE >> 2; /* divide by 4 */
592 for (i = 0; i < max; i++)
593 M_32_SWAP(((int *)pbuf)[i]);
594 } else {
595 uint16_t *bp = (uint16_t *)(void *)pbuf;
596 max = bp[0] + 2;
597 for (i = 0; i <= max; i++)
598 M_16_SWAP(bp[i]);
599 }
600 p = pbuf;
601 }
602 if (is_bucket)
603 page = BUCKET_TO_PAGE(bucket);
604 else
605 page = OADDR_TO_PAGE(bucket);
606 if ((wsize = pwrite(fd, p, size, (off_t)page << hashp->BSHIFT)) == -1)
607 /* Errno is set */
608 return (-1);
609 if (wsize != size) {
610 errno = EFTYPE;
611 return (-1);
612 }
613 return (0);
614 }
615
616 #define BYTE_MASK ((1 << INT_BYTE_SHIFT) -1)
617 /*
618 * Initialize a new bitmap page. Bitmap pages are left in memory
619 * once they are read in.
620 */
621 int
__ibitmap(HTAB * hashp,int pnum,int nbits,int ndx)622 __ibitmap(HTAB *hashp, int pnum, int nbits, int ndx)
623 {
624 u_int32_t *ip;
625 int clearbytes, clearints;
626
627 if ((ip = (u_int32_t *)malloc(hashp->BSIZE)) == NULL)
628 return (1);
629 hashp->nmaps++;
630 clearints = ((nbits - 1) >> INT_BYTE_SHIFT) + 1;
631 clearbytes = clearints << INT_TO_BYTE;
632 (void)memset((char *)ip, 0, clearbytes);
633 (void)memset(((char *)ip) + clearbytes, 0xFF,
634 hashp->BSIZE - clearbytes);
635 ip[clearints - 1] = ALL_SET << (nbits & BYTE_MASK);
636 SETBIT(ip, 0);
637 hashp->BITMAPS[ndx] = (u_int16_t)pnum;
638 hashp->mapp[ndx] = ip;
639 return (0);
640 }
641
642 static u_int32_t
first_free(u_int32_t map)643 first_free(u_int32_t map)
644 {
645 u_int32_t i, mask;
646
647 mask = 0x1;
648 for (i = 0; i < BITS_PER_MAP; i++) {
649 if (!(mask & map))
650 return (i);
651 mask = mask << 1;
652 }
653 return (i);
654 }
655
656 static u_int16_t
overflow_page(HTAB * hashp)657 overflow_page(HTAB *hashp)
658 {
659 u_int32_t *freep;
660 int max_free, offset, splitnum;
661 u_int16_t addr;
662 int bit, first_page, free_bit, free_page, i, in_use_bits, j;
663 #ifdef DEBUG2
664 int tmp1, tmp2;
665 #endif
666 splitnum = hashp->OVFL_POINT;
667 max_free = hashp->SPARES[splitnum];
668
669 free_page = (max_free - 1) >> (hashp->BSHIFT + BYTE_SHIFT);
670 free_bit = (max_free - 1) & ((hashp->BSIZE << BYTE_SHIFT) - 1);
671
672 /* Look through all the free maps to find the first free block */
673 first_page = hashp->LAST_FREED >>(hashp->BSHIFT + BYTE_SHIFT);
674 for ( i = first_page; i <= free_page; i++ ) {
675 if (!(freep = (u_int32_t *)hashp->mapp[i]) &&
676 !(freep = fetch_bitmap(hashp, i)))
677 return (0);
678 if (i == free_page)
679 in_use_bits = free_bit;
680 else
681 in_use_bits = (hashp->BSIZE << BYTE_SHIFT) - 1;
682
683 if (i == first_page) {
684 bit = hashp->LAST_FREED &
685 ((hashp->BSIZE << BYTE_SHIFT) - 1);
686 j = bit / BITS_PER_MAP;
687 bit = rounddown2(bit, BITS_PER_MAP);
688 } else {
689 bit = 0;
690 j = 0;
691 }
692 for (; bit <= in_use_bits; j++, bit += BITS_PER_MAP)
693 if (freep[j] != ALL_SET)
694 goto found;
695 }
696
697 /* No Free Page Found */
698 hashp->LAST_FREED = hashp->SPARES[splitnum];
699 hashp->SPARES[splitnum]++;
700 offset = hashp->SPARES[splitnum] -
701 (splitnum ? hashp->SPARES[splitnum - 1] : 0);
702
703 #define OVMSG "HASH: Out of overflow pages. Increase page size\n"
704 if (offset > SPLITMASK) {
705 if (++splitnum >= NCACHED) {
706 (void)_write(STDERR_FILENO, OVMSG, sizeof(OVMSG) - 1);
707 errno = EFBIG;
708 return (0);
709 }
710 hashp->OVFL_POINT = splitnum;
711 hashp->SPARES[splitnum] = hashp->SPARES[splitnum-1];
712 hashp->SPARES[splitnum-1]--;
713 offset = 1;
714 }
715
716 /* Check if we need to allocate a new bitmap page */
717 if (free_bit == (hashp->BSIZE << BYTE_SHIFT) - 1) {
718 free_page++;
719 if (free_page >= NCACHED) {
720 (void)_write(STDERR_FILENO, OVMSG, sizeof(OVMSG) - 1);
721 errno = EFBIG;
722 return (0);
723 }
724 /*
725 * This is tricky. The 1 indicates that you want the new page
726 * allocated with 1 clear bit. Actually, you are going to
727 * allocate 2 pages from this map. The first is going to be
728 * the map page, the second is the overflow page we were
729 * looking for. The init_bitmap routine automatically, sets
730 * the first bit of itself to indicate that the bitmap itself
731 * is in use. We would explicitly set the second bit, but
732 * don't have to if we tell init_bitmap not to leave it clear
733 * in the first place.
734 */
735 if (__ibitmap(hashp,
736 (int)OADDR_OF(splitnum, offset), 1, free_page))
737 return (0);
738 hashp->SPARES[splitnum]++;
739 #ifdef DEBUG2
740 free_bit = 2;
741 #endif
742 offset++;
743 if (offset > SPLITMASK) {
744 if (++splitnum >= NCACHED) {
745 (void)_write(STDERR_FILENO, OVMSG,
746 sizeof(OVMSG) - 1);
747 errno = EFBIG;
748 return (0);
749 }
750 hashp->OVFL_POINT = splitnum;
751 hashp->SPARES[splitnum] = hashp->SPARES[splitnum-1];
752 hashp->SPARES[splitnum-1]--;
753 offset = 0;
754 }
755 } else {
756 /*
757 * Free_bit addresses the last used bit. Bump it to address
758 * the first available bit.
759 */
760 free_bit++;
761 SETBIT(freep, free_bit);
762 }
763
764 /* Calculate address of the new overflow page */
765 addr = OADDR_OF(splitnum, offset);
766 #ifdef DEBUG2
767 (void)fprintf(stderr, "OVERFLOW_PAGE: ADDR: %d BIT: %d PAGE %d\n",
768 addr, free_bit, free_page);
769 #endif
770 return (addr);
771
772 found:
773 bit = bit + first_free(freep[j]);
774 SETBIT(freep, bit);
775 #ifdef DEBUG2
776 tmp1 = bit;
777 tmp2 = i;
778 #endif
779 /*
780 * Bits are addressed starting with 0, but overflow pages are addressed
781 * beginning at 1. Bit is a bit addressnumber, so we need to increment
782 * it to convert it to a page number.
783 */
784 bit = 1 + bit + (i * (hashp->BSIZE << BYTE_SHIFT));
785 if (bit >= hashp->LAST_FREED)
786 hashp->LAST_FREED = bit - 1;
787
788 /* Calculate the split number for this page */
789 for (i = 0; (i < splitnum) && (bit > hashp->SPARES[i]); i++);
790 offset = (i ? bit - hashp->SPARES[i - 1] : bit);
791 if (offset >= SPLITMASK) {
792 (void)_write(STDERR_FILENO, OVMSG, sizeof(OVMSG) - 1);
793 errno = EFBIG;
794 return (0); /* Out of overflow pages */
795 }
796 addr = OADDR_OF(i, offset);
797 #ifdef DEBUG2
798 (void)fprintf(stderr, "OVERFLOW_PAGE: ADDR: %d BIT: %d PAGE %d\n",
799 addr, tmp1, tmp2);
800 #endif
801
802 /* Allocate and return the overflow page */
803 return (addr);
804 }
805
806 /*
807 * Mark this overflow page as free.
808 */
809 void
__free_ovflpage(HTAB * hashp,BUFHEAD * obufp)810 __free_ovflpage(HTAB *hashp, BUFHEAD *obufp)
811 {
812 u_int16_t addr;
813 u_int32_t *freep;
814 int bit_address, free_page, free_bit;
815 u_int16_t ndx;
816
817 addr = obufp->addr;
818 #ifdef DEBUG1
819 (void)fprintf(stderr, "Freeing %d\n", addr);
820 #endif
821 ndx = (((u_int16_t)addr) >> SPLITSHIFT);
822 bit_address =
823 (ndx ? hashp->SPARES[ndx - 1] : 0) + (addr & SPLITMASK) - 1;
824 if (bit_address < hashp->LAST_FREED)
825 hashp->LAST_FREED = bit_address;
826 free_page = (bit_address >> (hashp->BSHIFT + BYTE_SHIFT));
827 free_bit = bit_address & ((hashp->BSIZE << BYTE_SHIFT) - 1);
828
829 if (!(freep = hashp->mapp[free_page]))
830 freep = fetch_bitmap(hashp, free_page);
831 #ifdef DEBUG
832 /*
833 * This had better never happen. It means we tried to read a bitmap
834 * that has already had overflow pages allocated off it, and we
835 * failed to read it from the file.
836 */
837 if (!freep)
838 assert(0);
839 #endif
840 CLRBIT(freep, free_bit);
841 #ifdef DEBUG2
842 (void)fprintf(stderr, "FREE_OVFLPAGE: ADDR: %d BIT: %d PAGE %d\n",
843 obufp->addr, free_bit, free_page);
844 #endif
845 __reclaim_buf(hashp, obufp);
846 }
847
848 /*
849 * Returns:
850 * 0 success
851 * -1 failure
852 */
853 static int
open_temp(HTAB * hashp)854 open_temp(HTAB *hashp)
855 {
856 sigset_t set, oset;
857 int len;
858 char *envtmp = NULL;
859 char path[MAXPATHLEN];
860
861 if (issetugid() == 0)
862 envtmp = getenv("TMPDIR");
863 len = snprintf(path,
864 sizeof(path), "%s/_hash.XXXXXX", envtmp ? envtmp : "/tmp");
865 if (len < 0 || len >= (int)sizeof(path)) {
866 errno = ENAMETOOLONG;
867 return (-1);
868 }
869
870 /* Block signals; make sure file goes away at process exit. */
871 (void)sigfillset(&set);
872 (void)__libc_sigprocmask(SIG_BLOCK, &set, &oset);
873 if ((hashp->fp = mkostemp(path, O_CLOEXEC)) != -1)
874 (void)unlink(path);
875 (void)__libc_sigprocmask(SIG_SETMASK, &oset, (sigset_t *)NULL);
876 return (hashp->fp != -1 ? 0 : -1);
877 }
878
879 /*
880 * We have to know that the key will fit, but the last entry on the page is
881 * an overflow pair, so we need to shift things.
882 */
883 static void
squeeze_key(u_int16_t * sp,const DBT * key,const DBT * val)884 squeeze_key(u_int16_t *sp, const DBT *key, const DBT *val)
885 {
886 char *p;
887 u_int16_t free_space, n, off, pageno;
888
889 p = (char *)sp;
890 n = sp[0];
891 free_space = FREESPACE(sp);
892 off = OFFSET(sp);
893
894 pageno = sp[n - 1];
895 off -= key->size;
896 sp[n - 1] = off;
897 memmove(p + off, key->data, key->size);
898 off -= val->size;
899 sp[n] = off;
900 memmove(p + off, val->data, val->size);
901 sp[0] = n + 2;
902 sp[n + 1] = pageno;
903 sp[n + 2] = OVFLPAGE;
904 FREESPACE(sp) = free_space - PAIRSIZE(key, val);
905 OFFSET(sp) = off;
906 }
907
908 static u_int32_t *
fetch_bitmap(HTAB * hashp,int ndx)909 fetch_bitmap(HTAB *hashp, int ndx)
910 {
911 if (ndx >= hashp->nmaps)
912 return (NULL);
913 if ((hashp->mapp[ndx] = (u_int32_t *)malloc(hashp->BSIZE)) == NULL)
914 return (NULL);
915 if (__get_page(hashp,
916 (char *)hashp->mapp[ndx], hashp->BITMAPS[ndx], 0, 1, 1)) {
917 free(hashp->mapp[ndx]);
918 return (NULL);
919 }
920 return (hashp->mapp[ndx]);
921 }
922
923 #ifdef DEBUG4
924 int
print_chain(int addr)925 print_chain(int addr)
926 {
927 BUFHEAD *bufp;
928 short *bp, oaddr;
929
930 (void)fprintf(stderr, "%d ", addr);
931 bufp = __get_buf(hashp, addr, NULL, 0);
932 bp = (short *)bufp->page;
933 while (bp[0] && ((bp[bp[0]] == OVFLPAGE) ||
934 ((bp[0] > 2) && bp[2] < REAL_KEY))) {
935 oaddr = bp[bp[0] - 1];
936 (void)fprintf(stderr, "%d ", (int)oaddr);
937 bufp = __get_buf(hashp, (int)oaddr, bufp, 0);
938 bp = (short *)bufp->page;
939 }
940 (void)fprintf(stderr, "\n");
941 }
942 #endif
943