]> git.saurik.com Git - apple/libc.git/blame - db/hash/hash.h
Libc-320.1.3.tar.gz
[apple/libc.git] / db / hash / hash.h
CommitLineData
e9ce8d39 1/*
9385eb3d 2 * Copyright (c) 2003 Apple Computer, Inc. All rights reserved.
e9ce8d39
A
3 *
4 * @APPLE_LICENSE_HEADER_START@
5 *
734aad71
A
6 * This file contains Original Code and/or Modifications of Original Code
7 * as defined in and that are subject to the Apple Public Source License
8 * Version 2.0 (the 'License'). You may not use this file except in
9 * compliance with the License. Please obtain a copy of the License at
10 * http://www.opensource.apple.com/apsl/ and read it before using this
11 * file.
12 *
13 * The Original Code and all software distributed under the License are
14 * distributed on an 'AS IS' basis, WITHOUT WARRANTY OF ANY KIND, EITHER
e9ce8d39
A
15 * EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES,
16 * INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY,
734aad71
A
17 * FITNESS FOR A PARTICULAR PURPOSE, QUIET ENJOYMENT OR NON-INFRINGEMENT.
18 * Please see the License for the specific language governing rights and
19 * limitations under the License.
e9ce8d39
A
20 *
21 * @APPLE_LICENSE_HEADER_END@
22 */
9385eb3d
A
23/*-
24 * Copyright (c) 1990, 1993, 1994
e9ce8d39
A
25 * The Regents of the University of California. All rights reserved.
26 *
27 * This code is derived from software contributed to Berkeley by
28 * Margo Seltzer.
29 *
30 * Redistribution and use in source and binary forms, with or without
31 * modification, are permitted provided that the following conditions
32 * are met:
33 * 1. Redistributions of source code must retain the above copyright
34 * notice, this list of conditions and the following disclaimer.
35 * 2. Redistributions in binary form must reproduce the above copyright
36 * notice, this list of conditions and the following disclaimer in the
37 * documentation and/or other materials provided with the distribution.
38 * 3. All advertising materials mentioning features or use of this software
39 * must display the following acknowledgement:
40 * This product includes software developed by the University of
41 * California, Berkeley and its contributors.
42 * 4. Neither the name of the University nor the names of its contributors
43 * may be used to endorse or promote products derived from this software
44 * without specific prior written permission.
45 *
46 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
47 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
48 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
49 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
50 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
51 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
52 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
53 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
54 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
55 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
56 * SUCH DAMAGE.
9385eb3d
A
57 *
58 * @(#)hash.h 8.3 (Berkeley) 5/31/94
59 * $FreeBSD: src/lib/libc/db/hash/hash.h,v 1.6 2002/03/21 22:46:26 obrien Exp $
e9ce8d39
A
60 */
61
62/* Operations */
63typedef enum {
64 HASH_GET, HASH_PUT, HASH_PUTNEW, HASH_DELETE, HASH_FIRST, HASH_NEXT
65} ACTION;
66
67/* Buffer Management structures */
68typedef struct _bufhead BUFHEAD;
69
70struct _bufhead {
9385eb3d
A
71 BUFHEAD *prev; /* LRU links */
72 BUFHEAD *next; /* LRU links */
73 BUFHEAD *ovfl; /* Overflow page buffer header */
74 u_int32_t addr; /* Address of this page */
75 char *page; /* Actual page data */
76 char flags;
e9ce8d39
A
77#define BUF_MOD 0x0001
78#define BUF_DISK 0x0002
79#define BUF_BUCKET 0x0004
80#define BUF_PIN 0x0008
81};
82
83#define IS_BUCKET(X) ((X) & BUF_BUCKET)
84
85typedef BUFHEAD **SEGMENT;
86
87/* Hash Table Information */
9385eb3d
A
88typedef struct hashhdr { /* Disk resident portion */
89 int magic; /* Magic NO for hash tables */
90 int version; /* Version ID */
91 u_int32_t lorder; /* Byte Order */
92 int bsize; /* Bucket/Page Size */
93 int bshift; /* Bucket shift */
94 int dsize; /* Directory Size */
95 int ssize; /* Segment Size */
96 int sshift; /* Segment shift */
97 int ovfl_point; /* Where overflow pages are being
98 * allocated */
99 int last_freed; /* Last overflow page freed */
100 int max_bucket; /* ID of Maximum bucket in use */
101 int high_mask; /* Mask to modulo into entire table */
102 int low_mask; /* Mask to modulo into lower half of
103 * table */
104 int ffactor; /* Fill factor */
105 int nkeys; /* Number of keys in hash table */
106 int hdrpages; /* Size of table header */
107 int h_charkey; /* value of hash(CHARKEY) */
108#define NCACHED 32 /* number of bit maps and spare
109 * points */
110 int spares[NCACHED];/* spare pages for overflow */
111 u_int16_t bitmaps[NCACHED]; /* address of overflow page
112 * bitmaps */
e9ce8d39
A
113} HASHHDR;
114
9385eb3d
A
115typedef struct htab { /* Memory resident data structure */
116 HASHHDR hdr; /* Header */
117 int nsegs; /* Number of allocated segments */
118 int exsegs; /* Number of extra allocated
119 * segments */
120 u_int32_t /* Hash function */
121 (*hash)(const void *, size_t);
122 int flags; /* Flag values */
123 int fp; /* File pointer */
124 char *tmp_buf; /* Temporary Buffer for BIG data */
125 char *tmp_key; /* Temporary Buffer for BIG keys */
126 BUFHEAD *cpage; /* Current page */
127 int cbucket; /* Current bucket */
128 int cndx; /* Index of next item on cpage */
129 int error; /* Error Number -- for DBM
130 * compatibility */
131 int new_file; /* Indicates if fd is backing store
132 * or no */
133 int save_file; /* Indicates whether we need to flush
134 * file at
135 * exit */
136 u_int32_t *mapp[NCACHED]; /* Pointers to page maps */
137 int nmaps; /* Initial number of bitmaps */
138 int nbufs; /* Number of buffers left to
139 * allocate */
140 BUFHEAD bufhead; /* Header of buffer lru list */
141 SEGMENT *dir; /* Hash Bucket directory */
e9ce8d39
A
142} HTAB;
143
144/*
145 * Constants
146 */
147#define MAX_BSIZE 65536 /* 2^16 */
148#define MIN_BUFFERS 6
149#define MINHDRSIZE 512
150#define DEF_BUFSIZE 65536 /* 64 K */
151#define DEF_BUCKET_SIZE 4096
152#define DEF_BUCKET_SHIFT 12 /* log2(BUCKET) */
153#define DEF_SEGSIZE 256
154#define DEF_SEGSIZE_SHIFT 8 /* log2(SEGSIZE) */
155#define DEF_DIRSIZE 256
156#define DEF_FFACTOR 65536
157#define MIN_FFACTOR 4
158#define SPLTMAX 8
159#define CHARKEY "%$sniglet^&"
160#define NUMKEY 1038583
161#define BYTE_SHIFT 3
162#define INT_TO_BYTE 2
163#define INT_BYTE_SHIFT 5
9385eb3d 164#define ALL_SET ((u_int32_t)0xFFFFFFFF)
e9ce8d39
A
165#define ALL_CLEAR 0
166
9385eb3d
A
167#define PTROF(X) ((BUFHEAD *)((ptrdiff_t)(X)&~0x3))
168#define ISMOD(X) ((u_int32_t)(ptrdiff_t)(X)&0x1)
169#define DOMOD(X) ((X) = (char *)((ptrdiff_t)(X)|0x1))
170#define ISDISK(X) ((u_int32_t)(ptrdiff_t)(X)&0x2)
171#define DODISK(X) ((X) = (char *)((ptrdiff_t)(X)|0x2))
e9ce8d39
A
172
173#define BITS_PER_MAP 32
174
175/* Given the address of the beginning of a big map, clear/set the nth bit */
176#define CLRBIT(A, N) ((A)[(N)/BITS_PER_MAP] &= ~(1<<((N)%BITS_PER_MAP)))
177#define SETBIT(A, N) ((A)[(N)/BITS_PER_MAP] |= (1<<((N)%BITS_PER_MAP)))
178#define ISSET(A, N) ((A)[(N)/BITS_PER_MAP] & (1<<((N)%BITS_PER_MAP)))
179
180/* Overflow management */
181/*
182 * Overflow page numbers are allocated per split point. At each doubling of
183 * the table, we can allocate extra pages. So, an overflow page number has
184 * the top 5 bits indicate which split point and the lower 11 bits indicate
185 * which page at that split point is indicated (pages within split points are
186 * numberered starting with 1).
187 */
188
189#define SPLITSHIFT 11
190#define SPLITMASK 0x7FF
9385eb3d 191#define SPLITNUM(N) (((u_int32_t)(N)) >> SPLITSHIFT)
e9ce8d39 192#define OPAGENUM(N) ((N) & SPLITMASK)
9385eb3d 193#define OADDR_OF(S,O) ((u_int32_t)((u_int32_t)(S) << SPLITSHIFT) + (O))
e9ce8d39
A
194
195#define BUCKET_TO_PAGE(B) \
196 (B) + hashp->HDRPAGES + ((B) ? hashp->SPARES[__log2((B)+1)-1] : 0)
197#define OADDR_TO_PAGE(B) \
198 BUCKET_TO_PAGE ( (1 << SPLITNUM((B))) -1 ) + OPAGENUM((B));
199
200/*
201 * page.h contains a detailed description of the page format.
202 *
203 * Normally, keys and data are accessed from offset tables in the top of
204 * each page which point to the beginning of the key and data. There are
205 * four flag values which may be stored in these offset tables which indicate
206 * the following:
207 *
208 *
209 * OVFLPAGE Rather than a key data pair, this pair contains
210 * the address of an overflow page. The format of
211 * the pair is:
212 * OVERFLOW_PAGE_NUMBER OVFLPAGE
213 *
214 * PARTIAL_KEY This must be the first key/data pair on a page
215 * and implies that page contains only a partial key.
216 * That is, the key is too big to fit on a single page
217 * so it starts on this page and continues on the next.
218 * The format of the page is:
219 * KEY_OFF PARTIAL_KEY OVFL_PAGENO OVFLPAGE
9385eb3d 220 *
e9ce8d39
A
221 * KEY_OFF -- offset of the beginning of the key
222 * PARTIAL_KEY -- 1
223 * OVFL_PAGENO - page number of the next overflow page
224 * OVFLPAGE -- 0
225 *
226 * FULL_KEY This must be the first key/data pair on the page. It
227 * is used in two cases.
228 *
229 * Case 1:
230 * There is a complete key on the page but no data
231 * (because it wouldn't fit). The next page contains
232 * the data.
233 *
234 * Page format it:
235 * KEY_OFF FULL_KEY OVFL_PAGENO OVFL_PAGE
236 *
237 * KEY_OFF -- offset of the beginning of the key
238 * FULL_KEY -- 2
239 * OVFL_PAGENO - page number of the next overflow page
240 * OVFLPAGE -- 0
241 *
242 * Case 2:
243 * This page contains no key, but part of a large
244 * data field, which is continued on the next page.
245 *
246 * Page format it:
247 * DATA_OFF FULL_KEY OVFL_PAGENO OVFL_PAGE
248 *
249 * KEY_OFF -- offset of the beginning of the data on
250 * this page
251 * FULL_KEY -- 2
252 * OVFL_PAGENO - page number of the next overflow page
253 * OVFLPAGE -- 0
254 *
9385eb3d 255 * FULL_KEY_DATA
e9ce8d39
A
256 * This must be the first key/data pair on the page.
257 * There are two cases:
258 *
259 * Case 1:
260 * This page contains a key and the beginning of the
261 * data field, but the data field is continued on the
262 * next page.
263 *
264 * Page format is:
265 * KEY_OFF FULL_KEY_DATA OVFL_PAGENO DATA_OFF
266 *
267 * KEY_OFF -- offset of the beginning of the key
268 * FULL_KEY_DATA -- 3
269 * OVFL_PAGENO - page number of the next overflow page
270 * DATA_OFF -- offset of the beginning of the data
271 *
272 * Case 2:
273 * This page contains the last page of a big data pair.
274 * There is no key, only the tail end of the data
275 * on this page.
276 *
277 * Page format is:
278 * DATA_OFF FULL_KEY_DATA <OVFL_PAGENO> <OVFLPAGE>
279 *
280 * DATA_OFF -- offset of the beginning of the data on
281 * this page
282 * FULL_KEY_DATA -- 3
283 * OVFL_PAGENO - page number of the next overflow page
284 * OVFLPAGE -- 0
285 *
286 * OVFL_PAGENO and OVFLPAGE are optional (they are
287 * not present if there is no next page).
288 */
289
290#define OVFLPAGE 0
291#define PARTIAL_KEY 1
292#define FULL_KEY 2
293#define FULL_KEY_DATA 3
294#define REAL_KEY 4
295
296/* Short hands for accessing structure */
297#define BSIZE hdr.bsize
298#define BSHIFT hdr.bshift
299#define DSIZE hdr.dsize
300#define SGSIZE hdr.ssize
301#define SSHIFT hdr.sshift
302#define LORDER hdr.lorder
303#define OVFL_POINT hdr.ovfl_point
304#define LAST_FREED hdr.last_freed
305#define MAX_BUCKET hdr.max_bucket
306#define FFACTOR hdr.ffactor
307#define HIGH_MASK hdr.high_mask
308#define LOW_MASK hdr.low_mask
309#define NKEYS hdr.nkeys
310#define HDRPAGES hdr.hdrpages
311#define SPARES hdr.spares
312#define BITMAPS hdr.bitmaps
313#define VERSION hdr.version
314#define MAGIC hdr.magic
315#define NEXT_FREE hdr.next_free
316#define H_CHARKEY hdr.h_charkey