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 * $DragonFly: src/lib/libc/db/hash/hash_func.c,v 1.8 2005/11/19 20:46:32 swildner Exp $ 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 static u_int32_t hash1 (const void *, size_t) __unused; 44 static u_int32_t hash2 (const void *, size_t) __unused; 45 static u_int32_t hash3 (const void *, size_t) __unused; 46 static u_int32_t hash4 (const void *, size_t); 47 48 /* Global default hash function */ 49 u_int32_t (*__default_hash) (const void *, size_t) = hash4; 50 51 /* 52 * HASH FUNCTIONS 53 * 54 * Assume that we've already split the bucket to which this key hashes, 55 * calculate that bucket, and check that in fact we did already split it. 56 * 57 * This came from ejb's hsearch. 58 */ 59 60 #define PRIME1 37 61 #define PRIME2 1048583 62 63 static u_int32_t 64 hash1(const void *keyarg, size_t len) 65 { 66 const u_char *key; 67 u_int32_t h; 68 69 /* Convert string to integer */ 70 for (key = keyarg, h = 0; len--;) 71 h = h * PRIME1 ^ (*key++ - ' '); 72 h %= PRIME2; 73 return (h); 74 } 75 76 /* 77 * Phong's linear congruential hash 78 */ 79 #define dcharhash(h, c) ((h) = 0x63c63cd9*(h) + 0x9c39c33d + (c)) 80 81 static u_int32_t 82 hash2(const void *keyarg, size_t len) 83 { 84 const u_char *e, *key; 85 u_int32_t h; 86 u_char c; 87 88 key = keyarg; 89 e = key + len; 90 for (h = 0; key != e;) { 91 c = *key++; 92 if (!c && key > e) 93 break; 94 dcharhash(h, c); 95 } 96 return (h); 97 } 98 99 /* 100 * This is INCREDIBLY ugly, but fast. We break the string up into 8 byte 101 * units. On the first time through the loop we get the "leftover bytes" 102 * (strlen % 8). On every other iteration, we perform 8 HASHC's so we handle 103 * all 8 bytes. Essentially, this saves us 7 cmp & branch instructions. If 104 * this routine is heavily used enough, it's worth the ugly coding. 105 * 106 * OZ's original sdbm hash 107 */ 108 static u_int32_t 109 hash3(const void *keyarg, size_t len) 110 { 111 const u_char *key; 112 size_t loop; 113 u_int32_t h; 114 115 #define HASHC h = *key++ + 65599 * h 116 117 h = 0; 118 key = keyarg; 119 if (len > 0) { 120 loop = (len + 8 - 1) >> 3; 121 122 switch (len & (8 - 1)) { 123 case 0: 124 do { 125 HASHC; 126 /* FALLTHROUGH */ 127 case 7: 128 HASHC; 129 /* FALLTHROUGH */ 130 case 6: 131 HASHC; 132 /* FALLTHROUGH */ 133 case 5: 134 HASHC; 135 /* FALLTHROUGH */ 136 case 4: 137 HASHC; 138 /* FALLTHROUGH */ 139 case 3: 140 HASHC; 141 /* FALLTHROUGH */ 142 case 2: 143 HASHC; 144 /* FALLTHROUGH */ 145 case 1: 146 HASHC; 147 } while (--loop); 148 } 149 } 150 return (h); 151 } 152 153 /* Hash function from Chris Torek. */ 154 static u_int32_t 155 hash4(const void *keyarg, size_t len) 156 { 157 const u_char *key; 158 size_t loop; 159 u_int32_t h; 160 161 #define HASH4a h = (h << 5) - h + *key++; 162 #define HASH4b h = (h << 5) + h + *key++; 163 #define HASH4 HASH4b 164 165 h = 0; 166 key = keyarg; 167 if (len > 0) { 168 loop = (len + 8 - 1) >> 3; 169 170 switch (len & (8 - 1)) { 171 case 0: 172 do { 173 HASH4; 174 /* FALLTHROUGH */ 175 case 7: 176 HASH4; 177 /* FALLTHROUGH */ 178 case 6: 179 HASH4; 180 /* FALLTHROUGH */ 181 case 5: 182 HASH4; 183 /* FALLTHROUGH */ 184 case 4: 185 HASH4; 186 /* FALLTHROUGH */ 187 case 3: 188 HASH4; 189 /* FALLTHROUGH */ 190 case 2: 191 HASH4; 192 /* FALLTHROUGH */ 193 case 1: 194 HASH4; 195 } while (--loop); 196 } 197 } 198 return (h); 199 } 200