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 * 3. 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
32 * @(#)hash.h 8.3 (Berkeley) 5/31/94
33 * $FreeBSD: src/lib/libc/db/hash/hash.h,v 1.4.8.1 2001/03/05 03:50:02 kris Exp $
34 * $DragonFly: src/lib/libc/db/hash/hash.h,v 1.4 2005/09/19 09:20:37 asmodai Exp $
39 HASH_GET, HASH_PUT, HASH_PUTNEW, HASH_DELETE, HASH_FIRST, HASH_NEXT
42 /* Buffer Management structures */
43 typedef struct _bufhead BUFHEAD;
46 BUFHEAD *prev; /* LRU links */
47 BUFHEAD *next; /* LRU links */
48 BUFHEAD *ovfl; /* Overflow page buffer header */
49 u_int32_t addr; /* Address of this page */
50 char *page; /* Actual page data */
52 #define BUF_MOD 0x0001
53 #define BUF_DISK 0x0002
54 #define BUF_BUCKET 0x0004
55 #define BUF_PIN 0x0008
58 #define IS_BUCKET(X) ((X) & BUF_BUCKET)
60 typedef BUFHEAD **SEGMENT;
62 /* Hash Table Information */
63 typedef struct hashhdr { /* Disk resident portion */
64 int magic; /* Magic NO for hash tables */
65 int version; /* Version ID */
66 u_int32_t lorder; /* Byte Order */
67 int bsize; /* Bucket/Page Size */
68 int bshift; /* Bucket shift */
69 int dsize; /* Directory Size */
70 int ssize; /* Segment Size */
71 int sshift; /* Segment shift */
72 int ovfl_point; /* Where overflow pages are being
74 int last_freed; /* Last overflow page freed */
75 int max_bucket; /* ID of Maximum bucket in use */
76 int high_mask; /* Mask to modulo into entire table */
77 int low_mask; /* Mask to modulo into lower half of
79 int ffactor; /* Fill factor */
80 int nkeys; /* Number of keys in hash table */
81 int hdrpages; /* Size of table header */
82 int h_charkey; /* value of hash(CHARKEY) */
83 #define NCACHED 32 /* number of bit maps and spare
85 int spares[NCACHED];/* spare pages for overflow */
86 u_int16_t bitmaps[NCACHED]; /* address of overflow page
90 typedef struct htab { /* Memory resident data structure */
91 HASHHDR hdr; /* Header */
92 int nsegs; /* Number of allocated segments */
93 int exsegs; /* Number of extra allocated
95 u_int32_t /* Hash function */
96 (*hash)(const void *, size_t);
97 int flags; /* Flag values */
98 int fp; /* File pointer */
99 char *tmp_buf; /* Temporary Buffer for BIG data */
100 char *tmp_key; /* Temporary Buffer for BIG keys */
101 BUFHEAD *cpage; /* Current page */
102 int cbucket; /* Current bucket */
103 int cndx; /* Index of next item on cpage */
104 int error; /* Error Number -- for DBM
106 int new_file; /* Indicates if fd is backing store
108 int save_file; /* Indicates whether we need to flush
111 u_int32_t *mapp[NCACHED]; /* Pointers to page maps */
112 int nmaps; /* Initial number of bitmaps */
113 int nbufs; /* Number of buffers left to
115 BUFHEAD bufhead; /* Header of buffer lru list */
116 SEGMENT *dir; /* Hash Bucket directory */
122 #define MAX_BSIZE 65536 /* 2^16 */
123 #define MIN_BUFFERS 6
124 #define MINHDRSIZE 512
125 #define DEF_BUFSIZE 65536 /* 64 K */
126 #define DEF_BUCKET_SIZE 4096
127 #define DEF_BUCKET_SHIFT 12 /* log2(BUCKET) */
128 #define DEF_SEGSIZE 256
129 #define DEF_SEGSIZE_SHIFT 8 /* log2(SEGSIZE) */
130 #define DEF_DIRSIZE 256
131 #define DEF_FFACTOR 65536
132 #define MIN_FFACTOR 4
134 #define CHARKEY "%$sniglet^&"
135 #define NUMKEY 1038583
137 #define INT_TO_BYTE 2
138 #define INT_BYTE_SHIFT 5
139 #define ALL_SET ((u_int32_t)0xFFFFFFFF)
142 #define PTROF(X) ((BUFHEAD *)((ptrdiff_t)(X)&~0x3))
143 #define ISMOD(X) ((u_int32_t)(ptrdiff_t)(X)&0x1)
144 #define DOMOD(X) ((X) = (char *)((ptrdiff_t)(X)|0x1))
145 #define ISDISK(X) ((u_int32_t)(ptrdiff_t)(X)&0x2)
146 #define DODISK(X) ((X) = (char *)((ptrdiff_t)(X)|0x2))
148 #define BITS_PER_MAP 32
150 /* Given the address of the beginning of a big map, clear/set the nth bit */
151 #define CLRBIT(A, N) ((A)[(N)/BITS_PER_MAP] &= ~(1<<((N)%BITS_PER_MAP)))
152 #define SETBIT(A, N) ((A)[(N)/BITS_PER_MAP] |= (1<<((N)%BITS_PER_MAP)))
153 #define ISSET(A, N) ((A)[(N)/BITS_PER_MAP] & (1<<((N)%BITS_PER_MAP)))
155 /* Overflow management */
157 * Overflow page numbers are allocated per split point. At each doubling of
158 * the table, we can allocate extra pages. So, an overflow page number has
159 * the top 5 bits indicate which split point and the lower 11 bits indicate
160 * which page at that split point is indicated (pages within split points are
161 * numberered starting with 1).
164 #define SPLITSHIFT 11
165 #define SPLITMASK 0x7FF
166 #define SPLITNUM(N) (((u_int32_t)(N)) >> SPLITSHIFT)
167 #define OPAGENUM(N) ((N) & SPLITMASK)
168 #define OADDR_OF(S,O) ((u_int32_t)((u_int32_t)(S) << SPLITSHIFT) + (O))
170 #define BUCKET_TO_PAGE(B) \
171 (B) + hashp->HDRPAGES + ((B) ? hashp->SPARES[__log2((B)+1)-1] : 0)
172 #define OADDR_TO_PAGE(B) \
173 BUCKET_TO_PAGE ( (1 << SPLITNUM((B))) -1 ) + OPAGENUM((B));
176 * page.h contains a detailed description of the page format.
178 * Normally, keys and data are accessed from offset tables in the top of
179 * each page which point to the beginning of the key and data. There are
180 * four flag values which may be stored in these offset tables which indicate
184 * OVFLPAGE Rather than a key data pair, this pair contains
185 * the address of an overflow page. The format of
187 * OVERFLOW_PAGE_NUMBER OVFLPAGE
189 * PARTIAL_KEY This must be the first key/data pair on a page
190 * and implies that page contains only a partial key.
191 * That is, the key is too big to fit on a single page
192 * so it starts on this page and continues on the next.
193 * The format of the page is:
194 * KEY_OFF PARTIAL_KEY OVFL_PAGENO OVFLPAGE
196 * KEY_OFF -- offset of the beginning of the key
198 * OVFL_PAGENO - page number of the next overflow page
201 * FULL_KEY This must be the first key/data pair on the page. It
202 * is used in two cases.
205 * There is a complete key on the page but no data
206 * (because it wouldn't fit). The next page contains
210 * KEY_OFF FULL_KEY OVFL_PAGENO OVFL_PAGE
212 * KEY_OFF -- offset of the beginning of the key
214 * OVFL_PAGENO - page number of the next overflow page
218 * This page contains no key, but part of a large
219 * data field, which is continued on the next page.
222 * DATA_OFF FULL_KEY OVFL_PAGENO OVFL_PAGE
224 * KEY_OFF -- offset of the beginning of the data on
227 * OVFL_PAGENO - page number of the next overflow page
231 * This must be the first key/data pair on the page.
232 * There are two cases:
235 * This page contains a key and the beginning of the
236 * data field, but the data field is continued on the
240 * KEY_OFF FULL_KEY_DATA OVFL_PAGENO DATA_OFF
242 * KEY_OFF -- offset of the beginning of the key
244 * OVFL_PAGENO - page number of the next overflow page
245 * DATA_OFF -- offset of the beginning of the data
248 * This page contains the last page of a big data pair.
249 * There is no key, only the tail end of the data
253 * DATA_OFF FULL_KEY_DATA <OVFL_PAGENO> <OVFLPAGE>
255 * DATA_OFF -- offset of the beginning of the data on
258 * OVFL_PAGENO - page number of the next overflow page
261 * OVFL_PAGENO and OVFLPAGE are optional (they are
262 * not present if there is no next page).
266 #define PARTIAL_KEY 1
268 #define FULL_KEY_DATA 3
271 /* Short hands for accessing structure */
272 #define BSIZE hdr.bsize
273 #define BSHIFT hdr.bshift
274 #define DSIZE hdr.dsize
275 #define SGSIZE hdr.ssize
276 #define SSHIFT hdr.sshift
277 #define LORDER hdr.lorder
278 #define OVFL_POINT hdr.ovfl_point
279 #define LAST_FREED hdr.last_freed
280 #define MAX_BUCKET hdr.max_bucket
281 #define FFACTOR hdr.ffactor
282 #define HIGH_MASK hdr.high_mask
283 #define LOW_MASK hdr.low_mask
284 #define NKEYS hdr.nkeys
285 #define HDRPAGES hdr.hdrpages
286 #define SPARES hdr.spares
287 #define BITMAPS hdr.bitmaps
288 #define VERSION hdr.version
289 #define MAGIC hdr.magic
290 #define NEXT_FREE hdr.next_free
291 #define H_CHARKEY hdr.h_charkey