hash.h (13105B)
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 * 3. ***REMOVED*** - see 17 * ftp://ftp.cs.berkeley.edu/pub/4bsd/README.Impt.License.Change 18 * 4. 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 * @(#)hash.h 8.3 (Berkeley) 5/31/94 35 */ 36 37 /* Operations */ 38 39 #include <stdio.h> 40 #include "mcom_db.h" 41 typedef enum { 42 HASH_GET, 43 HASH_PUT, 44 HASH_PUTNEW, 45 HASH_DELETE, 46 HASH_FIRST, 47 HASH_NEXT 48 } ACTION; 49 50 /* Buffer Management structures */ 51 typedef struct _bufhead BUFHEAD; 52 53 struct _bufhead { 54 BUFHEAD *prev; /* LRU links */ 55 BUFHEAD *next; /* LRU links */ 56 BUFHEAD *ovfl; /* Overflow page buffer header */ 57 uint32 addr; /* Address of this page */ 58 char *page; /* Actual page data */ 59 char is_disk; 60 char flags; 61 #define BUF_MOD 0x0001 62 #define BUF_DISK 0x0002 63 #define BUF_BUCKET 0x0004 64 #define BUF_PIN 0x0008 65 }; 66 67 #define IS_BUCKET(X) ((X)&BUF_BUCKET) 68 69 typedef BUFHEAD **SEGMENT; 70 71 typedef int DBFILE_PTR; 72 #define NO_FILE -1 73 #ifdef macintosh 74 #define DBFILE_OPEN(path, flag, mode) open((path), flag) 75 #define EXISTS(path) 76 #else 77 #define DBFILE_OPEN(path, flag, mode) open((path), (flag), (mode)) 78 #endif 79 /* Hash Table Information */ 80 typedef struct hashhdr { /* Disk resident portion */ 81 int32 magic; /* Magic NO for hash tables */ 82 int32 version; /* Version ID */ 83 uint32 lorder; /* Byte Order */ 84 int32 bsize; /* Bucket/Page Size */ 85 int32 bshift; /* Bucket shift */ 86 int32 dsize; /* Directory Size */ 87 int32 ssize; /* Segment Size */ 88 int32 sshift; /* Segment shift */ 89 int32 ovfl_point; /* Where overflow pages are being 90 * allocated */ 91 int32 last_freed; /* Last overflow page freed */ 92 int32 max_bucket; /* ID of Maximum bucket in use */ 93 int32 high_mask; /* Mask to modulo into entire table */ 94 int32 low_mask; /* Mask to modulo into lower half of 95 * table */ 96 int32 ffactor; /* Fill factor */ 97 int32 nkeys; /* Number of keys in hash table */ 98 int32 hdrpages; /* Size of table header */ 99 uint32 h_charkey; /* value of hash(CHARKEY) */ 100 #define NCACHED 32 /* number of bit maps and spare \ 101 * points */ 102 int32 spares[NCACHED]; /* spare pages for overflow */ 103 uint16 bitmaps[NCACHED]; /* address of overflow page 104 * bitmaps */ 105 } HASHHDR; 106 107 typedef struct htab { /* Memory resident data structure */ 108 HASHHDR hdr; /* Header */ 109 int nsegs; /* Number of allocated segments */ 110 int exsegs; /* Number of extra allocated 111 * segments */ 112 uint32 /* Hash function */ 113 (*hash)(const void *, size_t); 114 int flags; /* Flag values */ 115 DBFILE_PTR fp; /* File pointer */ 116 char *filename; 117 char *tmp_buf; /* Temporary Buffer for BIG data */ 118 char *tmp_key; /* Temporary Buffer for BIG keys */ 119 BUFHEAD *cpage; /* Current page */ 120 int cbucket; /* Current bucket */ 121 int cndx; /* Index of next item on cpage */ 122 int dbmerrno; /* Error Number -- for DBM 123 * compatability */ 124 int new_file; /* Indicates if fd is backing store 125 * or no */ 126 int save_file; /* Indicates whether we need to flush 127 * file at 128 * exit */ 129 uint32 *mapp[NCACHED]; /* Pointers to page maps */ 130 int nmaps; /* Initial number of bitmaps */ 131 int nbufs; /* Number of buffers left to 132 * allocate */ 133 BUFHEAD bufhead; /* Header of buffer lru list */ 134 SEGMENT *dir; /* Hash Bucket directory */ 135 off_t file_size; /* in bytes */ 136 char is_temp; /* unlink file on close */ 137 char updateEOF; /* force EOF update on flush */ 138 } HTAB; 139 140 /* 141 * Constants 142 */ 143 #define DATABASE_CORRUPTED_ERROR -999 /* big ugly abort, delete database */ 144 #define OLD_MAX_BSIZE 65536 /* 2^16 */ 145 #define MAX_BSIZE 32l * 1024l /* 2^15 */ 146 #define MIN_BUFFERS 6 147 #define MINHDRSIZE 512 148 #define DEF_BUFSIZE 65536l /* 64 K */ 149 #define DEF_BUCKET_SIZE 4096 150 #define DEF_BUCKET_SHIFT 12 /* log2(BUCKET) */ 151 #define DEF_SEGSIZE 256 152 #define DEF_SEGSIZE_SHIFT 8 /* log2(SEGSIZE) */ 153 #define DEF_DIRSIZE 256 154 #define DEF_FFACTOR 65536l 155 #define MIN_FFACTOR 4 156 #define SPLTMAX 8 157 #define CHARKEY "%$sniglet^&" 158 #define NUMKEY 1038583l 159 #define BYTE_SHIFT 3 160 #define INT_TO_BYTE 2 161 #define INT_BYTE_SHIFT 5 162 #define ALL_SET ((uint32)0xFFFFFFFF) 163 #define ALL_CLEAR 0 164 165 #define PTROF(X) ((ptrdiff_t)(X) == BUF_DISK ? 0 : (X)) 166 #define ISDISK(X) ((X) ? ((ptrdiff_t)(X) == BUF_DISK ? BUF_DISK \ 167 : (X)->is_disk) \ 168 : 0) 169 170 #define BITS_PER_MAP 32 171 172 /* Given the address of the beginning of a big map, clear/set the nth bit */ 173 #define CLRBIT(A, N) ((A)[(N) / BITS_PER_MAP] &= ~(1 << ((N) % BITS_PER_MAP))) 174 #define SETBIT(A, N) ((A)[(N) / BITS_PER_MAP] |= (1 << ((N) % BITS_PER_MAP))) 175 #define ISSET(A, N) ((A)[(N) / BITS_PER_MAP] & (1 << ((N) % BITS_PER_MAP))) 176 177 /* Overflow management */ 178 /* 179 * Overflow page numbers are allocated per split point. At each doubling of 180 * the table, we can allocate extra pages. So, an overflow page number has 181 * the top 5 bits indicate which split point and the lower 11 bits indicate 182 * which page at that split point is indicated (pages within split points are 183 * numberered starting with 1). 184 */ 185 186 #define SPLITSHIFT 11 187 #define SPLITMASK 0x7FF 188 #define SPLITNUM(N) (((uint32)(N)) >> SPLITSHIFT) 189 #define OPAGENUM(N) ((N)&SPLITMASK) 190 #define OADDR_OF(S, O) ((uint32)((uint32)(S) << SPLITSHIFT) + (O)) 191 192 #define BUCKET_TO_PAGE(B) \ 193 (B) + hashp->HDRPAGES + ((B) ? hashp->SPARES[dbm_log2((uint32)((B) + 1)) - 1] : 0) 194 #define OADDR_TO_PAGE(B) \ 195 BUCKET_TO_PAGE((1 << SPLITNUM((B))) - 1) + OPAGENUM((B)); 196 197 /* 198 * page.h contains a detailed description of the page format. 199 * 200 * Normally, keys and data are accessed from offset tables in the top of 201 * each page which point to the beginning of the key and data. There are 202 * four flag values which may be stored in these offset tables which indicate 203 * the following: 204 * 205 * 206 * OVFLPAGE Rather than a key data pair, this pair contains 207 * the address of an overflow page. The format of 208 * the pair is: 209 * OVERFLOW_PAGE_NUMBER OVFLPAGE 210 * 211 * PARTIAL_KEY This must be the first key/data pair on a page 212 * and implies that page contains only a partial key. 213 * That is, the key is too big to fit on a single page 214 * so it starts on this page and continues on the next. 215 * The format of the page is: 216 * KEY_OFF PARTIAL_KEY OVFL_PAGENO OVFLPAGE 217 * 218 * KEY_OFF -- offset of the beginning of the key 219 * PARTIAL_KEY -- 1 220 * OVFL_PAGENO - page number of the next overflow page 221 * OVFLPAGE -- 0 222 * 223 * FULL_KEY This must be the first key/data pair on the page. It 224 * is used in two cases. 225 * 226 * Case 1: 227 * There is a complete key on the page but no data 228 * (because it wouldn't fit). The next page contains 229 * the data. 230 * 231 * Page format it: 232 * KEY_OFF FULL_KEY OVFL_PAGENO OVFL_PAGE 233 * 234 * KEY_OFF -- offset of the beginning of the key 235 * FULL_KEY -- 2 236 * OVFL_PAGENO - page number of the next overflow page 237 * OVFLPAGE -- 0 238 * 239 * Case 2: 240 * This page contains no key, but part of a large 241 * data field, which is continued on the next page. 242 * 243 * Page format it: 244 * DATA_OFF FULL_KEY OVFL_PAGENO OVFL_PAGE 245 * 246 * KEY_OFF -- offset of the beginning of the data on 247 * this page 248 * FULL_KEY -- 2 249 * OVFL_PAGENO - page number of the next overflow page 250 * OVFLPAGE -- 0 251 * 252 * FULL_KEY_DATA 253 * This must be the first key/data pair on the page. 254 * There are two cases: 255 * 256 * Case 1: 257 * This page contains a key and the beginning of the 258 * data field, but the data field is continued on the 259 * next page. 260 * 261 * Page format is: 262 * KEY_OFF FULL_KEY_DATA OVFL_PAGENO DATA_OFF 263 * 264 * KEY_OFF -- offset of the beginning of the key 265 * FULL_KEY_DATA -- 3 266 * OVFL_PAGENO - page number of the next overflow page 267 * DATA_OFF -- offset of the beginning of the data 268 * 269 * Case 2: 270 * This page contains the last page of a big data pair. 271 * There is no key, only the tail end of the data 272 * on this page. 273 * 274 * Page format is: 275 * DATA_OFF FULL_KEY_DATA <OVFL_PAGENO> <OVFLPAGE> 276 * 277 * DATA_OFF -- offset of the beginning of the data on 278 * this page 279 * FULL_KEY_DATA -- 3 280 * OVFL_PAGENO - page number of the next overflow page 281 * OVFLPAGE -- 0 282 * 283 * OVFL_PAGENO and OVFLPAGE are optional (they are 284 * not present if there is no next page). 285 */ 286 287 #define OVFLPAGE 0 288 #define PARTIAL_KEY 1 289 #define FULL_KEY 2 290 #define FULL_KEY_DATA 3 291 #define REAL_KEY 4 292 293 /* Short hands for accessing structure */ 294 #undef BSIZE 295 #define BSIZE hdr.bsize 296 #undef BSHIFT 297 #define BSHIFT hdr.bshift 298 #define DSIZE hdr.dsize 299 #define SGSIZE hdr.ssize 300 #define SSHIFT hdr.sshift 301 #define LORDER hdr.lorder 302 #define OVFL_POINT hdr.ovfl_point 303 #define LAST_FREED hdr.last_freed 304 #define MAX_BUCKET hdr.max_bucket 305 #define FFACTOR hdr.ffactor 306 #define HIGH_MASK hdr.high_mask 307 #define LOW_MASK hdr.low_mask 308 #define NKEYS hdr.nkeys 309 #define HDRPAGES hdr.hdrpages 310 #define SPARES hdr.spares 311 #define BITMAPS hdr.bitmaps 312 #define VERSION hdr.version 313 #define MAGIC hdr.magic 314 #define NEXT_FREE hdr.next_free 315 #define H_CHARKEY hdr.h_charkey 316 317 extern uint32 (*dbm_default_hash)(const void *, size_t); 318 void dbm_buf_init(HTAB *hashp, int32 nbytes); 319 int dbm_big_delete(HTAB *hashp, BUFHEAD *bufp); 320 BUFHEAD *dbm_get_buf(HTAB *hashp, uint32 addr, BUFHEAD *prev_bp, int newpage); 321 uint32 dbm_call_hash(HTAB *hashp, char *k, size_t len); 322 #include "page.h" 323 extern int dbm_big_split(HTAB *hashp, BUFHEAD *op, BUFHEAD *np, 324 BUFHEAD *big_keyp, uint32 addr, uint32 obucket, SPLIT_RETURN *ret); 325 void dbm_free_ovflpage(HTAB *hashp, BUFHEAD *obufp); 326 BUFHEAD *dbm_add_ovflpage(HTAB *hashp, BUFHEAD *bufp); 327 int dbm_big_insert(HTAB *hashp, BUFHEAD *bufp, const DBT *key, const DBT *val); 328 int dbm_expand_table(HTAB *hashp); 329 uint32 dbm_log2(uint32 num); 330 void dbm_reclaim_buf(HTAB *hashp, BUFHEAD *bp); 331 int dbm_get_page(HTAB *hashp, char *p, uint32 bucket, int is_bucket, int is_disk, int is_bitmap); 332 int dbm_put_page(HTAB *hashp, char *p, uint32 bucket, int is_bucket, int is_bitmap); 333 int dbm_ibitmap(HTAB *hashp, int pnum, int nbits, int ndx); 334 int dbm_buf_free(HTAB *hashp, int do_free, int to_disk); 335 int dbm_find_bigpair(HTAB *hashp, BUFHEAD *bufp, int ndx, char *key, int size); 336 uint16 dbm_find_last_page(HTAB *hashp, BUFHEAD **bpp); 337 int dbm_addel(HTAB *hashp, BUFHEAD *bufp, const DBT *key, const DBT *val); 338 int dbm_big_return(HTAB *hashp, BUFHEAD *bufp, int ndx, DBT *val, int set_current); 339 int dbm_delpair(HTAB *hashp, BUFHEAD *bufp, int ndx); 340 int dbm_big_keydata(HTAB *hashp, BUFHEAD *bufp, DBT *key, DBT *val, int set); 341 int dbm_split_page(HTAB *hashp, uint32 obucket, uint32 nbucket);