1 /*- 2 * Copyright (c) 1990, 1993 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. 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. 19 * 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 30 * SUCH DAMAGE. 31 * 32 * @(#)hash_func.c 8.2 (Berkeley) 2/21/94 33 * $FreeBSD: head/lib/libc/db/hash/hash_func.c 190498 2009-03-28 07:31:02Z delphij $ 34 */ 35 36 #include <sys/types.h> 37 38 #include <db.h> 39 #include "hash.h" 40 #include "page.h" 41 #include "extern.h" 42 43 #ifdef notdef 44 static uint32_t hash1(const void *, size_t) __unused; 45 static uint32_t hash2(const void *, size_t) __unused; 46 static uint32_t hash3(const void *, size_t) __unused; 47 #endif 48 static uint32_t hash4(const void *, size_t); 49 50 /* Default hash function. */ 51 uint32_t (*__default_hash)(const void *, size_t) = hash4; 52 53 #ifdef notdef 54 /* 55 * Assume that we've already split the bucket to which this key hashes, 56 * calculate that bucket, and check that in fact we did already split it. 57 * 58 * EJB's original hsearch hash. 59 */ 60 #define PRIME1 37 61 #define PRIME2 1048583 62 63 uint32_t 64 hash1(const void *key, size_t len) 65 { 66 uint32_t h; 67 uint8_t *k; 68 69 h = 0; 70 k = (uint8_t *)key; 71 /* Convert string to integer */ 72 while (len--) 73 h = h * PRIME1 ^ (*k++ - ' '); 74 h %= PRIME2; 75 return (h); 76 } 77 78 /* 79 * Phong Vo's linear congruential hash 80 */ 81 #define dcharhash(h, c) ((h) = 0x63c63cd9*(h) + 0x9c39c33d + (c)) 82 83 uint32_t 84 hash2(const void *key, size_t len) 85 { 86 uint32_t h; 87 uint8_t *e, c, *k; 88 89 k = (uint8_t *)key; 90 e = k + len; 91 for (h = 0; k != e;) { 92 c = *k++; 93 if (!c && k > e) 94 break; 95 dcharhash(h, c); 96 } 97 return (h); 98 } 99 100 /* 101 * This is INCREDIBLY ugly, but fast. We break the string up into 8 byte 102 * units. On the first time through the loop we get the "leftover bytes" 103 * (strlen % 8). On every other iteration, we perform 8 HASHC's so we handle 104 * all 8 bytes. Essentially, this saves us 7 cmp & branch instructions. If 105 * this routine is heavily used enough, it's worth the ugly coding. 106 * 107 * Ozan Yigit's original sdbm hash. 108 */ 109 uint32_t 110 hash3(const void *key, size_t len) 111 { 112 uint32_t n, loop; 113 uint8_t *k; 114 115 #define HASHC n = *k++ + 65599 * n 116 117 n = 0; 118 k = (uint8_t *)key; 119 if (len > 0) { 120 loop = (len + 8 - 1) >> 3; 121 122 switch (len & (8 - 1)) { 123 case 0: 124 do { /* All fall throughs */ 125 HASHC; 126 case 7: 127 HASHC; 128 case 6: 129 HASHC; 130 case 5: 131 HASHC; 132 case 4: 133 HASHC; 134 case 3: 135 HASHC; 136 case 2: 137 HASHC; 138 case 1: 139 HASHC; 140 } while (--loop); 141 } 142 143 } 144 return (n); 145 } 146 #endif /* notdef */ 147 148 /* Chris Torek's hash function. */ 149 uint32_t 150 hash4(const void *key, size_t len) 151 { 152 uint32_t h, loop; 153 const uint8_t *k; 154 155 #define HASH4a h = (h << 5) - h + *k++; 156 #define HASH4b h = (h << 5) + h + *k++; 157 #define HASH4 HASH4b 158 159 h = 0; 160 k = key; 161 if (len > 0) { 162 loop = (len + 8 - 1) >> 3; 163 164 switch (len & (8 - 1)) { 165 case 0: 166 do { /* All fall throughs */ 167 HASH4; 168 case 7: 169 HASH4; 170 case 6: 171 HASH4; 172 case 5: 173 HASH4; 174 case 4: 175 HASH4; 176 case 3: 177 HASH4; 178 case 2: 179 HASH4; 180 case 1: 181 HASH4; 182 } while (--loop); 183 } 184 185 } 186 return (h); 187 } 188