| 1 |
/* $MidnightBSD$ */ |
| 2 |
/*- |
| 3 |
* Copyright (c) 1990, 1993, 1994 |
| 4 |
* The Regents of the University of California. All rights reserved. |
| 5 |
* |
| 6 |
* This code is derived from software contributed to Berkeley by |
| 7 |
* Margo Seltzer. |
| 8 |
* |
| 9 |
* Redistribution and use in source and binary forms, with or without |
| 10 |
* modification, are permitted provided that the following conditions |
| 11 |
* are met: |
| 12 |
* 1. Redistributions of source code must retain the above copyright |
| 13 |
* notice, this list of conditions and the following disclaimer. |
| 14 |
* 2. Redistributions in binary form must reproduce the above copyright |
| 15 |
* notice, this list of conditions and the following disclaimer in the |
| 16 |
* documentation and/or other materials provided with the distribution. |
| 17 |
* 4. Neither the name of the University nor the names of its contributors |
| 18 |
* may be used to endorse or promote products derived from this software |
| 19 |
* without specific prior written permission. |
| 20 |
* |
| 21 |
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND |
| 22 |
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE |
| 23 |
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE |
| 24 |
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE |
| 25 |
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL |
| 26 |
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS |
| 27 |
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) |
| 28 |
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT |
| 29 |
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY |
| 30 |
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF |
| 31 |
* SUCH DAMAGE. |
| 32 |
*/ |
| 33 |
|
| 34 |
#if defined(LIBC_SCCS) && !defined(lint) |
| 35 |
static char sccsid[] = "@(#)hash_page.c 8.7 (Berkeley) 8/16/94"; |
| 36 |
#endif /* LIBC_SCCS and not lint */ |
| 37 |
#include <sys/cdefs.h> |
| 38 |
__FBSDID("$FreeBSD: stable/10/lib/libc/db/hash/hash_page.c 313532 2017-02-10 06:34:52Z ngie $"); |
| 39 |
|
| 40 |
/* |
| 41 |
* PACKAGE: hashing |
| 42 |
* |
| 43 |
* DESCRIPTION: |
| 44 |
* Page manipulation for hashing package. |
| 45 |
* |
| 46 |
* ROUTINES: |
| 47 |
* |
| 48 |
* External |
| 49 |
* __get_page |
| 50 |
* __add_ovflpage |
| 51 |
* Internal |
| 52 |
* overflow_page |
| 53 |
* open_temp |
| 54 |
*/ |
| 55 |
|
| 56 |
#include "namespace.h" |
| 57 |
#include <sys/param.h> |
| 58 |
|
| 59 |
#include <errno.h> |
| 60 |
#include <fcntl.h> |
| 61 |
#include <signal.h> |
| 62 |
#include <stdio.h> |
| 63 |
#include <stdlib.h> |
| 64 |
#include <string.h> |
| 65 |
#include <unistd.h> |
| 66 |
#ifdef DEBUG |
| 67 |
#include <assert.h> |
| 68 |
#endif |
| 69 |
#include "un-namespace.h" |
| 70 |
#include "libc_private.h" |
| 71 |
|
| 72 |
#include <db.h> |
| 73 |
#include "hash.h" |
| 74 |
#include "page.h" |
| 75 |
#include "extern.h" |
| 76 |
|
| 77 |
static u_int32_t *fetch_bitmap(HTAB *, int); |
| 78 |
static u_int32_t first_free(u_int32_t); |
| 79 |
static int open_temp(HTAB *); |
| 80 |
static u_int16_t overflow_page(HTAB *); |
| 81 |
static void putpair(char *, const DBT *, const DBT *); |
| 82 |
static void squeeze_key(u_int16_t *, const DBT *, const DBT *); |
| 83 |
static int ugly_split(HTAB *, u_int32_t, BUFHEAD *, BUFHEAD *, int, int); |
| 84 |
|
| 85 |
#define PAGE_INIT(P) { \ |
| 86 |
((u_int16_t *)(P))[0] = 0; \ |
| 87 |
((u_int16_t *)(P))[1] = hashp->BSIZE - 3 * sizeof(u_int16_t); \ |
| 88 |
((u_int16_t *)(P))[2] = hashp->BSIZE; \ |
| 89 |
} |
| 90 |
|
| 91 |
/* |
| 92 |
* This is called AFTER we have verified that there is room on the page for |
| 93 |
* the pair (PAIRFITS has returned true) so we go right ahead and start moving |
| 94 |
* stuff on. |
| 95 |
*/ |
| 96 |
static void |
| 97 |
putpair(char *p, const DBT *key, const DBT *val) |
| 98 |
{ |
| 99 |
u_int16_t *bp, n, off; |
| 100 |
|
| 101 |
bp = (u_int16_t *)p; |
| 102 |
|
| 103 |
/* Enter the key first. */ |
| 104 |
n = bp[0]; |
| 105 |
|
| 106 |
off = OFFSET(bp) - key->size; |
| 107 |
memmove(p + off, key->data, key->size); |
| 108 |
bp[++n] = off; |
| 109 |
|
| 110 |
/* Now the data. */ |
| 111 |
off -= val->size; |
| 112 |
memmove(p + off, val->data, val->size); |
| 113 |
bp[++n] = off; |
| 114 |
|
| 115 |
/* Adjust page info. */ |
| 116 |
bp[0] = n; |
| 117 |
bp[n + 1] = off - ((n + 3) * sizeof(u_int16_t)); |
| 118 |
bp[n + 2] = off; |
| 119 |
} |
| 120 |
|
| 121 |
/* |
| 122 |
* Returns: |
| 123 |
* 0 OK |
| 124 |
* -1 error |
| 125 |
*/ |
| 126 |
int |
| 127 |
__delpair(HTAB *hashp, BUFHEAD *bufp, int ndx) |
| 128 |
{ |
| 129 |
u_int16_t *bp, newoff, pairlen; |
| 130 |
int n; |
| 131 |
|
| 132 |
bp = (u_int16_t *)bufp->page; |
| 133 |
n = bp[0]; |
| 134 |
|
| 135 |
if (bp[ndx + 1] < REAL_KEY) |
| 136 |
return (__big_delete(hashp, bufp)); |
| 137 |
if (ndx != 1) |
| 138 |
newoff = bp[ndx - 1]; |
| 139 |
else |
| 140 |
newoff = hashp->BSIZE; |
| 141 |
pairlen = newoff - bp[ndx + 1]; |
| 142 |
|
| 143 |
if (ndx != (n - 1)) { |
| 144 |
/* Hard Case -- need to shuffle keys */ |
| 145 |
int i; |
| 146 |
char *src = bufp->page + (int)OFFSET(bp); |
| 147 |
char *dst = src + (int)pairlen; |
| 148 |
memmove(dst, src, bp[ndx + 1] - OFFSET(bp)); |
| 149 |
|
| 150 |
/* Now adjust the pointers */ |
| 151 |
for (i = ndx + 2; i <= n; i += 2) { |
| 152 |
if (bp[i + 1] == OVFLPAGE) { |
| 153 |
bp[i - 2] = bp[i]; |
| 154 |
bp[i - 1] = bp[i + 1]; |
| 155 |
} else { |
| 156 |
bp[i - 2] = bp[i] + pairlen; |
| 157 |
bp[i - 1] = bp[i + 1] + pairlen; |
| 158 |
} |
| 159 |
} |
| 160 |
if (ndx == hashp->cndx) { |
| 161 |
/* |
| 162 |
* We just removed pair we were "pointing" to. |
| 163 |
* By moving back the cndx we ensure subsequent |
| 164 |
* hash_seq() calls won't skip over any entries. |
| 165 |
*/ |
| 166 |
hashp->cndx -= 2; |
| 167 |
} |
| 168 |
} |
| 169 |
/* Finally adjust the page data */ |
| 170 |
bp[n] = OFFSET(bp) + pairlen; |
| 171 |
bp[n - 1] = bp[n + 1] + pairlen + 2 * sizeof(u_int16_t); |
| 172 |
bp[0] = n - 2; |
| 173 |
hashp->NKEYS--; |
| 174 |
|
| 175 |
bufp->flags |= BUF_MOD; |
| 176 |
return (0); |
| 177 |
} |
| 178 |
/* |
| 179 |
* Returns: |
| 180 |
* 0 ==> OK |
| 181 |
* -1 ==> Error |
| 182 |
*/ |
| 183 |
int |
| 184 |
__split_page(HTAB *hashp, u_int32_t obucket, u_int32_t nbucket) |
| 185 |
{ |
| 186 |
BUFHEAD *new_bufp, *old_bufp; |
| 187 |
u_int16_t *ino; |
| 188 |
char *np; |
| 189 |
DBT key, val; |
| 190 |
int n, ndx, retval; |
| 191 |
u_int16_t copyto, diff, off, moved; |
| 192 |
char *op; |
| 193 |
|
| 194 |
copyto = (u_int16_t)hashp->BSIZE; |
| 195 |
off = (u_int16_t)hashp->BSIZE; |
| 196 |
old_bufp = __get_buf(hashp, obucket, NULL, 0); |
| 197 |
if (old_bufp == NULL) |
| 198 |
return (-1); |
| 199 |
new_bufp = __get_buf(hashp, nbucket, NULL, 0); |
| 200 |
if (new_bufp == NULL) |
| 201 |
return (-1); |
| 202 |
|
| 203 |
old_bufp->flags |= (BUF_MOD | BUF_PIN); |
| 204 |
new_bufp->flags |= (BUF_MOD | BUF_PIN); |
| 205 |
|
| 206 |
ino = (u_int16_t *)(op = old_bufp->page); |
| 207 |
np = new_bufp->page; |
| 208 |
|
| 209 |
moved = 0; |
| 210 |
|
| 211 |
for (n = 1, ndx = 1; n < ino[0]; n += 2) { |
| 212 |
if (ino[n + 1] < REAL_KEY) { |
| 213 |
retval = ugly_split(hashp, obucket, old_bufp, new_bufp, |
| 214 |
(int)copyto, (int)moved); |
| 215 |
old_bufp->flags &= ~BUF_PIN; |
| 216 |
new_bufp->flags &= ~BUF_PIN; |
| 217 |
return (retval); |
| 218 |
|
| 219 |
} |
| 220 |
key.data = (u_char *)op + ino[n]; |
| 221 |
key.size = off - ino[n]; |
| 222 |
|
| 223 |
if (__call_hash(hashp, key.data, key.size) == obucket) { |
| 224 |
/* Don't switch page */ |
| 225 |
diff = copyto - off; |
| 226 |
if (diff) { |
| 227 |
copyto = ino[n + 1] + diff; |
| 228 |
memmove(op + copyto, op + ino[n + 1], |
| 229 |
off - ino[n + 1]); |
| 230 |
ino[ndx] = copyto + ino[n] - ino[n + 1]; |
| 231 |
ino[ndx + 1] = copyto; |
| 232 |
} else |
| 233 |
copyto = ino[n + 1]; |
| 234 |
ndx += 2; |
| 235 |
} else { |
| 236 |
/* Switch page */ |
| 237 |
val.data = (u_char *)op + ino[n + 1]; |
| 238 |
val.size = ino[n] - ino[n + 1]; |
| 239 |
putpair(np, &key, &val); |
| 240 |
moved += 2; |
| 241 |
} |
| 242 |
|
| 243 |
off = ino[n + 1]; |
| 244 |
} |
| 245 |
|
| 246 |
/* Now clean up the page */ |
| 247 |
ino[0] -= moved; |
| 248 |
FREESPACE(ino) = copyto - sizeof(u_int16_t) * (ino[0] + 3); |
| 249 |
OFFSET(ino) = copyto; |
| 250 |
|
| 251 |
#ifdef DEBUG3 |
| 252 |
(void)fprintf(stderr, "split %d/%d\n", |
| 253 |
((u_int16_t *)np)[0] / 2, |
| 254 |
((u_int16_t *)op)[0] / 2); |
| 255 |
#endif |
| 256 |
/* unpin both pages */ |
| 257 |
old_bufp->flags &= ~BUF_PIN; |
| 258 |
new_bufp->flags &= ~BUF_PIN; |
| 259 |
return (0); |
| 260 |
} |
| 261 |
|
| 262 |
/* |
| 263 |
* Called when we encounter an overflow or big key/data page during split |
| 264 |
* handling. This is special cased since we have to begin checking whether |
| 265 |
* the key/data pairs fit on their respective pages and because we may need |
| 266 |
* overflow pages for both the old and new pages. |
| 267 |
* |
| 268 |
* The first page might be a page with regular key/data pairs in which case |
| 269 |
* we have a regular overflow condition and just need to go on to the next |
| 270 |
* page or it might be a big key/data pair in which case we need to fix the |
| 271 |
* big key/data pair. |
| 272 |
* |
| 273 |
* Returns: |
| 274 |
* 0 ==> success |
| 275 |
* -1 ==> failure |
| 276 |
*/ |
| 277 |
static int |
| 278 |
ugly_split(HTAB *hashp, |
| 279 |
u_int32_t obucket, /* Same as __split_page. */ |
| 280 |
BUFHEAD *old_bufp, |
| 281 |
BUFHEAD *new_bufp, |
| 282 |
int copyto, /* First byte on page which contains key/data values. */ |
| 283 |
int moved) /* Number of pairs moved to new page. */ |
| 284 |
{ |
| 285 |
BUFHEAD *bufp; /* Buffer header for ino */ |
| 286 |
u_int16_t *ino; /* Page keys come off of */ |
| 287 |
u_int16_t *np; /* New page */ |
| 288 |
u_int16_t *op; /* Page keys go on to if they aren't moving */ |
| 289 |
|
| 290 |
BUFHEAD *last_bfp; /* Last buf header OVFL needing to be freed */ |
| 291 |
DBT key, val; |
| 292 |
SPLIT_RETURN ret; |
| 293 |
u_int16_t n, off, ov_addr, scopyto; |
| 294 |
char *cino; /* Character value of ino */ |
| 295 |
|
| 296 |
bufp = old_bufp; |
| 297 |
ino = (u_int16_t *)old_bufp->page; |
| 298 |
np = (u_int16_t *)new_bufp->page; |
| 299 |
op = (u_int16_t *)old_bufp->page; |
| 300 |
last_bfp = NULL; |
| 301 |
scopyto = (u_int16_t)copyto; /* ANSI */ |
| 302 |
|
| 303 |
n = ino[0] - 1; |
| 304 |
while (n < ino[0]) { |
| 305 |
if (ino[2] < REAL_KEY && ino[2] != OVFLPAGE) { |
| 306 |
if (__big_split(hashp, old_bufp, |
| 307 |
new_bufp, bufp, bufp->addr, obucket, &ret)) |
| 308 |
return (-1); |
| 309 |
old_bufp = ret.oldp; |
| 310 |
if (!old_bufp) |
| 311 |
return (-1); |
| 312 |
op = (u_int16_t *)old_bufp->page; |
| 313 |
new_bufp = ret.newp; |
| 314 |
if (!new_bufp) |
| 315 |
return (-1); |
| 316 |
np = (u_int16_t *)new_bufp->page; |
| 317 |
bufp = ret.nextp; |
| 318 |
if (!bufp) |
| 319 |
return (0); |
| 320 |
cino = (char *)bufp->page; |
| 321 |
ino = (u_int16_t *)cino; |
| 322 |
last_bfp = ret.nextp; |
| 323 |
} else if (ino[n + 1] == OVFLPAGE) { |
| 324 |
ov_addr = ino[n]; |
| 325 |
/* |
| 326 |
* Fix up the old page -- the extra 2 are the fields |
| 327 |
* which contained the overflow information. |
| 328 |
*/ |
| 329 |
ino[0] -= (moved + 2); |
| 330 |
FREESPACE(ino) = |
| 331 |
scopyto - sizeof(u_int16_t) * (ino[0] + 3); |
| 332 |
OFFSET(ino) = scopyto; |
| 333 |
|
| 334 |
bufp = __get_buf(hashp, ov_addr, bufp, 0); |
| 335 |
if (!bufp) |
| 336 |
return (-1); |
| 337 |
|
| 338 |
ino = (u_int16_t *)bufp->page; |
| 339 |
n = 1; |
| 340 |
scopyto = hashp->BSIZE; |
| 341 |
moved = 0; |
| 342 |
|
| 343 |
if (last_bfp) |
| 344 |
__free_ovflpage(hashp, last_bfp); |
| 345 |
last_bfp = bufp; |
| 346 |
} |
| 347 |
/* Move regular sized pairs of there are any */ |
| 348 |
off = hashp->BSIZE; |
| 349 |
for (n = 1; (n < ino[0]) && (ino[n + 1] >= REAL_KEY); n += 2) { |
| 350 |
cino = (char *)ino; |
| 351 |
key.data = (u_char *)cino + ino[n]; |
| 352 |
key.size = off - ino[n]; |
| 353 |
val.data = (u_char *)cino + ino[n + 1]; |
| 354 |
val.size = ino[n] - ino[n + 1]; |
| 355 |
off = ino[n + 1]; |
| 356 |
|
| 357 |
if (__call_hash(hashp, key.data, key.size) == obucket) { |
| 358 |
/* Keep on old page */ |
| 359 |
if (PAIRFITS(op, (&key), (&val))) |
| 360 |
putpair((char *)op, &key, &val); |
| 361 |
else { |
| 362 |
old_bufp = |
| 363 |
__add_ovflpage(hashp, old_bufp); |
| 364 |
if (!old_bufp) |
| 365 |
return (-1); |
| 366 |
op = (u_int16_t *)old_bufp->page; |
| 367 |
putpair((char *)op, &key, &val); |
| 368 |
} |
| 369 |
old_bufp->flags |= BUF_MOD; |
| 370 |
} else { |
| 371 |
/* Move to new page */ |
| 372 |
if (PAIRFITS(np, (&key), (&val))) |
| 373 |
putpair((char *)np, &key, &val); |
| 374 |
else { |
| 375 |
new_bufp = |
| 376 |
__add_ovflpage(hashp, new_bufp); |
| 377 |
if (!new_bufp) |
| 378 |
return (-1); |
| 379 |
np = (u_int16_t *)new_bufp->page; |
| 380 |
putpair((char *)np, &key, &val); |
| 381 |
} |
| 382 |
new_bufp->flags |= BUF_MOD; |
| 383 |
} |
| 384 |
} |
| 385 |
} |
| 386 |
if (last_bfp) |
| 387 |
__free_ovflpage(hashp, last_bfp); |
| 388 |
return (0); |
| 389 |
} |
| 390 |
|
| 391 |
/* |
| 392 |
* Add the given pair to the page |
| 393 |
* |
| 394 |
* Returns: |
| 395 |
* 0 ==> OK |
| 396 |
* 1 ==> failure |
| 397 |
*/ |
| 398 |
int |
| 399 |
__addel(HTAB *hashp, BUFHEAD *bufp, const DBT *key, const DBT *val) |
| 400 |
{ |
| 401 |
u_int16_t *bp, *sop; |
| 402 |
int do_expand; |
| 403 |
|
| 404 |
bp = (u_int16_t *)bufp->page; |
| 405 |
do_expand = 0; |
| 406 |
while (bp[0] && (bp[2] < REAL_KEY || bp[bp[0]] < REAL_KEY)) |
| 407 |
/* Exception case */ |
| 408 |
if (bp[2] == FULL_KEY_DATA && bp[0] == 2) |
| 409 |
/* This is the last page of a big key/data pair |
| 410 |
and we need to add another page */ |
| 411 |
break; |
| 412 |
else if (bp[2] < REAL_KEY && bp[bp[0]] != OVFLPAGE) { |
| 413 |
bufp = __get_buf(hashp, bp[bp[0] - 1], bufp, 0); |
| 414 |
if (!bufp) |
| 415 |
return (-1); |
| 416 |
bp = (u_int16_t *)bufp->page; |
| 417 |
} else if (bp[bp[0]] != OVFLPAGE) { |
| 418 |
/* Short key/data pairs, no more pages */ |
| 419 |
break; |
| 420 |
} else { |
| 421 |
/* Try to squeeze key on this page */ |
| 422 |
if (bp[2] >= REAL_KEY && |
| 423 |
FREESPACE(bp) >= PAIRSIZE(key, val)) { |
| 424 |
squeeze_key(bp, key, val); |
| 425 |
goto stats; |
| 426 |
} else { |
| 427 |
bufp = __get_buf(hashp, bp[bp[0] - 1], bufp, 0); |
| 428 |
if (!bufp) |
| 429 |
return (-1); |
| 430 |
bp = (u_int16_t *)bufp->page; |
| 431 |
} |
| 432 |
} |
| 433 |
|
| 434 |
if (PAIRFITS(bp, key, val)) |
| 435 |
putpair(bufp->page, key, val); |
| 436 |
else { |
| 437 |
do_expand = 1; |
| 438 |
bufp = __add_ovflpage(hashp, bufp); |
| 439 |
if (!bufp) |
| 440 |
return (-1); |
| 441 |
sop = (u_int16_t *)bufp->page; |
| 442 |
|
| 443 |
if (PAIRFITS(sop, key, val)) |
| 444 |
putpair((char *)sop, key, val); |
| 445 |
else |
| 446 |
if (__big_insert(hashp, bufp, key, val)) |
| 447 |
return (-1); |
| 448 |
} |
| 449 |
stats: |
| 450 |
bufp->flags |= BUF_MOD; |
| 451 |
/* |
| 452 |
* If the average number of keys per bucket exceeds the fill factor, |
| 453 |
* expand the table. |
| 454 |
*/ |
| 455 |
hashp->NKEYS++; |
| 456 |
if (do_expand || |
| 457 |
(hashp->NKEYS / (hashp->MAX_BUCKET + 1) > hashp->FFACTOR)) |
| 458 |
return (__expand_table(hashp)); |
| 459 |
return (0); |
| 460 |
} |
| 461 |
|
| 462 |
/* |
| 463 |
* |
| 464 |
* Returns: |
| 465 |
* pointer on success |
| 466 |
* NULL on error |
| 467 |
*/ |
| 468 |
BUFHEAD * |
| 469 |
__add_ovflpage(HTAB *hashp, BUFHEAD *bufp) |
| 470 |
{ |
| 471 |
u_int16_t *sp, ndx, ovfl_num; |
| 472 |
#ifdef DEBUG1 |
| 473 |
int tmp1, tmp2; |
| 474 |
#endif |
| 475 |
sp = (u_int16_t *)bufp->page; |
| 476 |
|
| 477 |
/* Check if we are dynamically determining the fill factor */ |
| 478 |
if (hashp->FFACTOR == DEF_FFACTOR) { |
| 479 |
hashp->FFACTOR = sp[0] >> 1; |
| 480 |
if (hashp->FFACTOR < MIN_FFACTOR) |
| 481 |
hashp->FFACTOR = MIN_FFACTOR; |
| 482 |
} |
| 483 |
bufp->flags |= BUF_MOD; |
| 484 |
ovfl_num = overflow_page(hashp); |
| 485 |
#ifdef DEBUG1 |
| 486 |
tmp1 = bufp->addr; |
| 487 |
tmp2 = bufp->ovfl ? bufp->ovfl->addr : 0; |
| 488 |
#endif |
| 489 |
if (!ovfl_num || !(bufp->ovfl = __get_buf(hashp, ovfl_num, bufp, 1))) |
| 490 |
return (NULL); |
| 491 |
bufp->ovfl->flags |= BUF_MOD; |
| 492 |
#ifdef DEBUG1 |
| 493 |
(void)fprintf(stderr, "ADDOVFLPAGE: %d->ovfl was %d is now %d\n", |
| 494 |
tmp1, tmp2, bufp->ovfl->addr); |
| 495 |
#endif |
| 496 |
ndx = sp[0]; |
| 497 |
/* |
| 498 |
* Since a pair is allocated on a page only if there's room to add |
| 499 |
* an overflow page, we know that the OVFL information will fit on |
| 500 |
* the page. |
| 501 |
*/ |
| 502 |
sp[ndx + 4] = OFFSET(sp); |
| 503 |
sp[ndx + 3] = FREESPACE(sp) - OVFLSIZE; |
| 504 |
sp[ndx + 1] = ovfl_num; |
| 505 |
sp[ndx + 2] = OVFLPAGE; |
| 506 |
sp[0] = ndx + 2; |
| 507 |
#ifdef HASH_STATISTICS |
| 508 |
hash_overflows++; |
| 509 |
#endif |
| 510 |
return (bufp->ovfl); |
| 511 |
} |
| 512 |
|
| 513 |
/* |
| 514 |
* Returns: |
| 515 |
* 0 indicates SUCCESS |
| 516 |
* -1 indicates FAILURE |
| 517 |
*/ |
| 518 |
int |
| 519 |
__get_page(HTAB *hashp, char *p, u_int32_t bucket, int is_bucket, int is_disk, |
| 520 |
int is_bitmap) |
| 521 |
{ |
| 522 |
int fd, page, size, rsize; |
| 523 |
u_int16_t *bp; |
| 524 |
|
| 525 |
fd = hashp->fp; |
| 526 |
size = hashp->BSIZE; |
| 527 |
|
| 528 |
if ((fd == -1) || !is_disk) { |
| 529 |
PAGE_INIT(p); |
| 530 |
return (0); |
| 531 |
} |
| 532 |
if (is_bucket) |
| 533 |
page = BUCKET_TO_PAGE(bucket); |
| 534 |
else |
| 535 |
page = OADDR_TO_PAGE(bucket); |
| 536 |
if ((rsize = pread(fd, p, size, (off_t)page << hashp->BSHIFT)) == -1) |
| 537 |
return (-1); |
| 538 |
bp = (u_int16_t *)p; |
| 539 |
if (!rsize) |
| 540 |
bp[0] = 0; /* We hit the EOF, so initialize a new page */ |
| 541 |
else |
| 542 |
if (rsize != size) { |
| 543 |
errno = EFTYPE; |
| 544 |
return (-1); |
| 545 |
} |
| 546 |
if (!is_bitmap && !bp[0]) { |
| 547 |
PAGE_INIT(p); |
| 548 |
} else |
| 549 |
if (hashp->LORDER != BYTE_ORDER) { |
| 550 |
int i, max; |
| 551 |
|
| 552 |
if (is_bitmap) { |
| 553 |
max = hashp->BSIZE >> 2; /* divide by 4 */ |
| 554 |
for (i = 0; i < max; i++) |
| 555 |
M_32_SWAP(((int *)p)[i]); |
| 556 |
} else { |
| 557 |
M_16_SWAP(bp[0]); |
| 558 |
max = bp[0] + 2; |
| 559 |
for (i = 1; i <= max; i++) |
| 560 |
M_16_SWAP(bp[i]); |
| 561 |
} |
| 562 |
} |
| 563 |
return (0); |
| 564 |
} |
| 565 |
|
| 566 |
/* |
| 567 |
* Write page p to disk |
| 568 |
* |
| 569 |
* Returns: |
| 570 |
* 0 ==> OK |
| 571 |
* -1 ==>failure |
| 572 |
*/ |
| 573 |
int |
| 574 |
__put_page(HTAB *hashp, char *p, u_int32_t bucket, int is_bucket, int is_bitmap) |
| 575 |
{ |
| 576 |
int fd, page, size; |
| 577 |
ssize_t wsize; |
| 578 |
char pbuf[MAX_BSIZE]; |
| 579 |
|
| 580 |
size = hashp->BSIZE; |
| 581 |
if ((hashp->fp == -1) && open_temp(hashp)) |
| 582 |
return (-1); |
| 583 |
fd = hashp->fp; |
| 584 |
|
| 585 |
if (hashp->LORDER != BYTE_ORDER) { |
| 586 |
int i, max; |
| 587 |
|
| 588 |
memcpy(pbuf, p, size); |
| 589 |
if (is_bitmap) { |
| 590 |
max = hashp->BSIZE >> 2; /* divide by 4 */ |
| 591 |
for (i = 0; i < max; i++) |
| 592 |
M_32_SWAP(((int *)pbuf)[i]); |
| 593 |
} else { |
| 594 |
uint16_t *bp = (uint16_t *)(void *)pbuf; |
| 595 |
max = bp[0] + 2; |
| 596 |
for (i = 0; i <= max; i++) |
| 597 |
M_16_SWAP(bp[i]); |
| 598 |
} |
| 599 |
p = pbuf; |
| 600 |
} |
| 601 |
if (is_bucket) |
| 602 |
page = BUCKET_TO_PAGE(bucket); |
| 603 |
else |
| 604 |
page = OADDR_TO_PAGE(bucket); |
| 605 |
if ((wsize = pwrite(fd, p, size, (off_t)page << hashp->BSHIFT)) == -1) |
| 606 |
/* Errno is set */ |
| 607 |
return (-1); |
| 608 |
if (wsize != size) { |
| 609 |
errno = EFTYPE; |
| 610 |
return (-1); |
| 611 |
} |
| 612 |
return (0); |
| 613 |
} |
| 614 |
|
| 615 |
#define BYTE_MASK ((1 << INT_BYTE_SHIFT) -1) |
| 616 |
/* |
| 617 |
* Initialize a new bitmap page. Bitmap pages are left in memory |
| 618 |
* once they are read in. |
| 619 |
*/ |
| 620 |
int |
| 621 |
__ibitmap(HTAB *hashp, int pnum, int nbits, int ndx) |
| 622 |
{ |
| 623 |
u_int32_t *ip; |
| 624 |
int clearbytes, clearints; |
| 625 |
|
| 626 |
if ((ip = (u_int32_t *)malloc(hashp->BSIZE)) == NULL) |
| 627 |
return (1); |
| 628 |
hashp->nmaps++; |
| 629 |
clearints = ((nbits - 1) >> INT_BYTE_SHIFT) + 1; |
| 630 |
clearbytes = clearints << INT_TO_BYTE; |
| 631 |
(void)memset((char *)ip, 0, clearbytes); |
| 632 |
(void)memset(((char *)ip) + clearbytes, 0xFF, |
| 633 |
hashp->BSIZE - clearbytes); |
| 634 |
ip[clearints - 1] = ALL_SET << (nbits & BYTE_MASK); |
| 635 |
SETBIT(ip, 0); |
| 636 |
hashp->BITMAPS[ndx] = (u_int16_t)pnum; |
| 637 |
hashp->mapp[ndx] = ip; |
| 638 |
return (0); |
| 639 |
} |
| 640 |
|
| 641 |
static u_int32_t |
| 642 |
first_free(u_int32_t map) |
| 643 |
{ |
| 644 |
u_int32_t i, mask; |
| 645 |
|
| 646 |
mask = 0x1; |
| 647 |
for (i = 0; i < BITS_PER_MAP; i++) { |
| 648 |
if (!(mask & map)) |
| 649 |
return (i); |
| 650 |
mask = mask << 1; |
| 651 |
} |
| 652 |
return (i); |
| 653 |
} |
| 654 |
|
| 655 |
static u_int16_t |
| 656 |
overflow_page(HTAB *hashp) |
| 657 |
{ |
| 658 |
u_int32_t *freep; |
| 659 |
int max_free, offset, splitnum; |
| 660 |
u_int16_t addr; |
| 661 |
int bit, first_page, free_bit, free_page, i, in_use_bits, j; |
| 662 |
#ifdef DEBUG2 |
| 663 |
int tmp1, tmp2; |
| 664 |
#endif |
| 665 |
splitnum = hashp->OVFL_POINT; |
| 666 |
max_free = hashp->SPARES[splitnum]; |
| 667 |
|
| 668 |
free_page = (max_free - 1) >> (hashp->BSHIFT + BYTE_SHIFT); |
| 669 |
free_bit = (max_free - 1) & ((hashp->BSIZE << BYTE_SHIFT) - 1); |
| 670 |
|
| 671 |
/* Look through all the free maps to find the first free block */ |
| 672 |
first_page = hashp->LAST_FREED >>(hashp->BSHIFT + BYTE_SHIFT); |
| 673 |
for ( i = first_page; i <= free_page; i++ ) { |
| 674 |
if (!(freep = (u_int32_t *)hashp->mapp[i]) && |
| 675 |
!(freep = fetch_bitmap(hashp, i))) |
| 676 |
return (0); |
| 677 |
if (i == free_page) |
| 678 |
in_use_bits = free_bit; |
| 679 |
else |
| 680 |
in_use_bits = (hashp->BSIZE << BYTE_SHIFT) - 1; |
| 681 |
|
| 682 |
if (i == first_page) { |
| 683 |
bit = hashp->LAST_FREED & |
| 684 |
((hashp->BSIZE << BYTE_SHIFT) - 1); |
| 685 |
j = bit / BITS_PER_MAP; |
| 686 |
bit = bit & ~(BITS_PER_MAP - 1); |
| 687 |
} else { |
| 688 |
bit = 0; |
| 689 |
j = 0; |
| 690 |
} |
| 691 |
for (; bit <= in_use_bits; j++, bit += BITS_PER_MAP) |
| 692 |
if (freep[j] != ALL_SET) |
| 693 |
goto found; |
| 694 |
} |
| 695 |
|
| 696 |
/* No Free Page Found */ |
| 697 |
hashp->LAST_FREED = hashp->SPARES[splitnum]; |
| 698 |
hashp->SPARES[splitnum]++; |
| 699 |
offset = hashp->SPARES[splitnum] - |
| 700 |
(splitnum ? hashp->SPARES[splitnum - 1] : 0); |
| 701 |
|
| 702 |
#define OVMSG "HASH: Out of overflow pages. Increase page size\n" |
| 703 |
if (offset > SPLITMASK) { |
| 704 |
if (++splitnum >= NCACHED) { |
| 705 |
(void)_write(STDERR_FILENO, OVMSG, sizeof(OVMSG) - 1); |
| 706 |
errno = EFBIG; |
| 707 |
return (0); |
| 708 |
} |
| 709 |
hashp->OVFL_POINT = splitnum; |
| 710 |
hashp->SPARES[splitnum] = hashp->SPARES[splitnum-1]; |
| 711 |
hashp->SPARES[splitnum-1]--; |
| 712 |
offset = 1; |
| 713 |
} |
| 714 |
|
| 715 |
/* Check if we need to allocate a new bitmap page */ |
| 716 |
if (free_bit == (hashp->BSIZE << BYTE_SHIFT) - 1) { |
| 717 |
free_page++; |
| 718 |
if (free_page >= NCACHED) { |
| 719 |
(void)_write(STDERR_FILENO, OVMSG, sizeof(OVMSG) - 1); |
| 720 |
errno = EFBIG; |
| 721 |
return (0); |
| 722 |
} |
| 723 |
/* |
| 724 |
* This is tricky. The 1 indicates that you want the new page |
| 725 |
* allocated with 1 clear bit. Actually, you are going to |
| 726 |
* allocate 2 pages from this map. The first is going to be |
| 727 |
* the map page, the second is the overflow page we were |
| 728 |
* looking for. The init_bitmap routine automatically, sets |
| 729 |
* the first bit of itself to indicate that the bitmap itself |
| 730 |
* is in use. We would explicitly set the second bit, but |
| 731 |
* don't have to if we tell init_bitmap not to leave it clear |
| 732 |
* in the first place. |
| 733 |
*/ |
| 734 |
if (__ibitmap(hashp, |
| 735 |
(int)OADDR_OF(splitnum, offset), 1, free_page)) |
| 736 |
return (0); |
| 737 |
hashp->SPARES[splitnum]++; |
| 738 |
#ifdef DEBUG2 |
| 739 |
free_bit = 2; |
| 740 |
#endif |
| 741 |
offset++; |
| 742 |
if (offset > SPLITMASK) { |
| 743 |
if (++splitnum >= NCACHED) { |
| 744 |
(void)_write(STDERR_FILENO, OVMSG, |
| 745 |
sizeof(OVMSG) - 1); |
| 746 |
errno = EFBIG; |
| 747 |
return (0); |
| 748 |
} |
| 749 |
hashp->OVFL_POINT = splitnum; |
| 750 |
hashp->SPARES[splitnum] = hashp->SPARES[splitnum-1]; |
| 751 |
hashp->SPARES[splitnum-1]--; |
| 752 |
offset = 0; |
| 753 |
} |
| 754 |
} else { |
| 755 |
/* |
| 756 |
* Free_bit addresses the last used bit. Bump it to address |
| 757 |
* the first available bit. |
| 758 |
*/ |
| 759 |
free_bit++; |
| 760 |
SETBIT(freep, free_bit); |
| 761 |
} |
| 762 |
|
| 763 |
/* Calculate address of the new overflow page */ |
| 764 |
addr = OADDR_OF(splitnum, offset); |
| 765 |
#ifdef DEBUG2 |
| 766 |
(void)fprintf(stderr, "OVERFLOW_PAGE: ADDR: %d BIT: %d PAGE %d\n", |
| 767 |
addr, free_bit, free_page); |
| 768 |
#endif |
| 769 |
return (addr); |
| 770 |
|
| 771 |
found: |
| 772 |
bit = bit + first_free(freep[j]); |
| 773 |
SETBIT(freep, bit); |
| 774 |
#ifdef DEBUG2 |
| 775 |
tmp1 = bit; |
| 776 |
tmp2 = i; |
| 777 |
#endif |
| 778 |
/* |
| 779 |
* Bits are addressed starting with 0, but overflow pages are addressed |
| 780 |
* beginning at 1. Bit is a bit addressnumber, so we need to increment |
| 781 |
* it to convert it to a page number. |
| 782 |
*/ |
| 783 |
bit = 1 + bit + (i * (hashp->BSIZE << BYTE_SHIFT)); |
| 784 |
if (bit >= hashp->LAST_FREED) |
| 785 |
hashp->LAST_FREED = bit - 1; |
| 786 |
|
| 787 |
/* Calculate the split number for this page */ |
| 788 |
for (i = 0; (i < splitnum) && (bit > hashp->SPARES[i]); i++); |
| 789 |
offset = (i ? bit - hashp->SPARES[i - 1] : bit); |
| 790 |
if (offset >= SPLITMASK) { |
| 791 |
(void)_write(STDERR_FILENO, OVMSG, sizeof(OVMSG) - 1); |
| 792 |
errno = EFBIG; |
| 793 |
return (0); /* Out of overflow pages */ |
| 794 |
} |
| 795 |
addr = OADDR_OF(i, offset); |
| 796 |
#ifdef DEBUG2 |
| 797 |
(void)fprintf(stderr, "OVERFLOW_PAGE: ADDR: %d BIT: %d PAGE %d\n", |
| 798 |
addr, tmp1, tmp2); |
| 799 |
#endif |
| 800 |
|
| 801 |
/* Allocate and return the overflow page */ |
| 802 |
return (addr); |
| 803 |
} |
| 804 |
|
| 805 |
/* |
| 806 |
* Mark this overflow page as free. |
| 807 |
*/ |
| 808 |
void |
| 809 |
__free_ovflpage(HTAB *hashp, BUFHEAD *obufp) |
| 810 |
{ |
| 811 |
u_int16_t addr; |
| 812 |
u_int32_t *freep; |
| 813 |
int bit_address, free_page, free_bit; |
| 814 |
u_int16_t ndx; |
| 815 |
|
| 816 |
addr = obufp->addr; |
| 817 |
#ifdef DEBUG1 |
| 818 |
(void)fprintf(stderr, "Freeing %d\n", addr); |
| 819 |
#endif |
| 820 |
ndx = (((u_int16_t)addr) >> SPLITSHIFT); |
| 821 |
bit_address = |
| 822 |
(ndx ? hashp->SPARES[ndx - 1] : 0) + (addr & SPLITMASK) - 1; |
| 823 |
if (bit_address < hashp->LAST_FREED) |
| 824 |
hashp->LAST_FREED = bit_address; |
| 825 |
free_page = (bit_address >> (hashp->BSHIFT + BYTE_SHIFT)); |
| 826 |
free_bit = bit_address & ((hashp->BSIZE << BYTE_SHIFT) - 1); |
| 827 |
|
| 828 |
if (!(freep = hashp->mapp[free_page])) |
| 829 |
freep = fetch_bitmap(hashp, free_page); |
| 830 |
#ifdef DEBUG |
| 831 |
/* |
| 832 |
* This had better never happen. It means we tried to read a bitmap |
| 833 |
* that has already had overflow pages allocated off it, and we |
| 834 |
* failed to read it from the file. |
| 835 |
*/ |
| 836 |
if (!freep) |
| 837 |
assert(0); |
| 838 |
#endif |
| 839 |
CLRBIT(freep, free_bit); |
| 840 |
#ifdef DEBUG2 |
| 841 |
(void)fprintf(stderr, "FREE_OVFLPAGE: ADDR: %d BIT: %d PAGE %d\n", |
| 842 |
obufp->addr, free_bit, free_page); |
| 843 |
#endif |
| 844 |
__reclaim_buf(hashp, obufp); |
| 845 |
} |
| 846 |
|
| 847 |
/* |
| 848 |
* Returns: |
| 849 |
* 0 success |
| 850 |
* -1 failure |
| 851 |
*/ |
| 852 |
static int |
| 853 |
open_temp(HTAB *hashp) |
| 854 |
{ |
| 855 |
sigset_t set, oset; |
| 856 |
int len; |
| 857 |
char *envtmp = NULL; |
| 858 |
char path[MAXPATHLEN]; |
| 859 |
|
| 860 |
if (issetugid() == 0) |
| 861 |
envtmp = getenv("TMPDIR"); |
| 862 |
len = snprintf(path, |
| 863 |
sizeof(path), "%s/_hash.XXXXXX", envtmp ? envtmp : "/tmp"); |
| 864 |
if (len < 0 || len >= (int)sizeof(path)) { |
| 865 |
errno = ENAMETOOLONG; |
| 866 |
return (-1); |
| 867 |
} |
| 868 |
|
| 869 |
/* Block signals; make sure file goes away at process exit. */ |
| 870 |
(void)sigfillset(&set); |
| 871 |
(void)__libc_sigprocmask(SIG_BLOCK, &set, &oset); |
| 872 |
if ((hashp->fp = mkostemp(path, O_CLOEXEC)) != -1) |
| 873 |
(void)unlink(path); |
| 874 |
(void)__libc_sigprocmask(SIG_SETMASK, &oset, (sigset_t *)NULL); |
| 875 |
return (hashp->fp != -1 ? 0 : -1); |
| 876 |
} |
| 877 |
|
| 878 |
/* |
| 879 |
* We have to know that the key will fit, but the last entry on the page is |
| 880 |
* an overflow pair, so we need to shift things. |
| 881 |
*/ |
| 882 |
static void |
| 883 |
squeeze_key(u_int16_t *sp, const DBT *key, const DBT *val) |
| 884 |
{ |
| 885 |
char *p; |
| 886 |
u_int16_t free_space, n, off, pageno; |
| 887 |
|
| 888 |
p = (char *)sp; |
| 889 |
n = sp[0]; |
| 890 |
free_space = FREESPACE(sp); |
| 891 |
off = OFFSET(sp); |
| 892 |
|
| 893 |
pageno = sp[n - 1]; |
| 894 |
off -= key->size; |
| 895 |
sp[n - 1] = off; |
| 896 |
memmove(p + off, key->data, key->size); |
| 897 |
off -= val->size; |
| 898 |
sp[n] = off; |
| 899 |
memmove(p + off, val->data, val->size); |
| 900 |
sp[0] = n + 2; |
| 901 |
sp[n + 1] = pageno; |
| 902 |
sp[n + 2] = OVFLPAGE; |
| 903 |
FREESPACE(sp) = free_space - PAIRSIZE(key, val); |
| 904 |
OFFSET(sp) = off; |
| 905 |
} |
| 906 |
|
| 907 |
static u_int32_t * |
| 908 |
fetch_bitmap(HTAB *hashp, int ndx) |
| 909 |
{ |
| 910 |
if (ndx >= hashp->nmaps) |
| 911 |
return (NULL); |
| 912 |
if ((hashp->mapp[ndx] = (u_int32_t *)malloc(hashp->BSIZE)) == NULL) |
| 913 |
return (NULL); |
| 914 |
if (__get_page(hashp, |
| 915 |
(char *)hashp->mapp[ndx], hashp->BITMAPS[ndx], 0, 1, 1)) { |
| 916 |
free(hashp->mapp[ndx]); |
| 917 |
return (NULL); |
| 918 |
} |
| 919 |
return (hashp->mapp[ndx]); |
| 920 |
} |
| 921 |
|
| 922 |
#ifdef DEBUG4 |
| 923 |
int |
| 924 |
print_chain(int addr) |
| 925 |
{ |
| 926 |
BUFHEAD *bufp; |
| 927 |
short *bp, oaddr; |
| 928 |
|
| 929 |
(void)fprintf(stderr, "%d ", addr); |
| 930 |
bufp = __get_buf(hashp, addr, NULL, 0); |
| 931 |
bp = (short *)bufp->page; |
| 932 |
while (bp[0] && ((bp[bp[0]] == OVFLPAGE) || |
| 933 |
((bp[0] > 2) && bp[2] < REAL_KEY))) { |
| 934 |
oaddr = bp[bp[0] - 1]; |
| 935 |
(void)fprintf(stderr, "%d ", (int)oaddr); |
| 936 |
bufp = __get_buf(hashp, (int)oaddr, bufp, 0); |
| 937 |
bp = (short *)bufp->page; |
| 938 |
} |
| 939 |
(void)fprintf(stderr, "\n"); |
| 940 |
} |
| 941 |
#endif |
