1 /* Copyright (C) 1995-1998 Eric Young (eay (at) cryptsoft.com) 2 * All rights reserved. 3 * 4 * This package is an SSL implementation written 5 * by Eric Young (eay (at) cryptsoft.com). 6 * The implementation was written so as to conform with Netscapes SSL. 7 * 8 * This library is free for commercial and non-commercial use as long as 9 * the following conditions are aheared to. The following conditions 10 * apply to all code found in this distribution, be it the RC4, RSA, 11 * lhash, DES, etc., code; not just the SSL code. The SSL documentation 12 * included with this distribution is covered by the same copyright terms 13 * except that the holder is Tim Hudson (tjh (at) cryptsoft.com). 14 * 15 * Copyright remains Eric Young's, and as such any Copyright notices in 16 * the code are not to be removed. 17 * If this package is used in a product, Eric Young should be given attribution 18 * as the author of the parts of the library used. 19 * This can be in the form of a textual message at program startup or 20 * in documentation (online or textual) provided with the package. 21 * 22 * Redistribution and use in source and binary forms, with or without 23 * modification, are permitted provided that the following conditions 24 * are met: 25 * 1. Redistributions of source code must retain the copyright 26 * notice, this list of conditions and the following disclaimer. 27 * 2. Redistributions in binary form must reproduce the above copyright 28 * notice, this list of conditions and the following disclaimer in the 29 * documentation and/or other materials provided with the distribution. 30 * 3. All advertising materials mentioning features or use of this software 31 * must display the following acknowledgement: 32 * "This product includes cryptographic software written by 33 * Eric Young (eay (at) cryptsoft.com)" 34 * The word 'cryptographic' can be left out if the rouines from the library 35 * being used are not cryptographic related :-). 36 * 4. If you include any Windows specific code (or a derivative thereof) from 37 * the apps directory (application code) you must include an acknowledgement: 38 * "This product includes software written by Tim Hudson (tjh (at) cryptsoft.com)" 39 * 40 * THIS SOFTWARE IS PROVIDED BY ERIC YOUNG ``AS IS'' AND 41 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 42 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 43 * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE 44 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 45 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 46 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 47 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 48 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 49 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 50 * SUCH DAMAGE. 51 * 52 * The licence and distribution terms for any publically available version or 53 * derivative of this code cannot be changed. i.e. this code cannot simply be 54 * copied and put under another distribution licence 55 * [including the GNU Public Licence.] */ 56 57 #include <openssl/lhash.h> 58 59 #include <assert.h> 60 #include <limits.h> 61 #include <string.h> 62 63 #include <openssl/mem.h> 64 65 #include "../internal.h" 66 67 68 /* kMinNumBuckets is the minimum size of the buckets array in an |_LHASH|. */ 69 static const size_t kMinNumBuckets = 16; 70 71 /* kMaxAverageChainLength contains the maximum, average chain length. When the 72 * average chain length exceeds this value, the hash table will be resized. */ 73 static const size_t kMaxAverageChainLength = 2; 74 static const size_t kMinAverageChainLength = 1; 75 76 _LHASH *lh_new(lhash_hash_func hash, lhash_cmp_func comp) { 77 _LHASH *ret = OPENSSL_malloc(sizeof(_LHASH)); 78 if (ret == NULL) { 79 return NULL; 80 } 81 OPENSSL_memset(ret, 0, sizeof(_LHASH)); 82 83 ret->num_buckets = kMinNumBuckets; 84 ret->buckets = OPENSSL_malloc(sizeof(LHASH_ITEM *) * ret->num_buckets); 85 if (ret->buckets == NULL) { 86 OPENSSL_free(ret); 87 return NULL; 88 } 89 OPENSSL_memset(ret->buckets, 0, sizeof(LHASH_ITEM *) * ret->num_buckets); 90 91 ret->comp = comp; 92 ret->hash = hash; 93 return ret; 94 } 95 96 void lh_free(_LHASH *lh) { 97 if (lh == NULL) { 98 return; 99 } 100 101 for (size_t i = 0; i < lh->num_buckets; i++) { 102 LHASH_ITEM *next; 103 for (LHASH_ITEM *n = lh->buckets[i]; n != NULL; n = next) { 104 next = n->next; 105 OPENSSL_free(n); 106 } 107 } 108 109 OPENSSL_free(lh->buckets); 110 OPENSSL_free(lh); 111 } 112 113 size_t lh_num_items(const _LHASH *lh) { return lh->num_items; } 114 115 /* get_next_ptr_and_hash returns a pointer to the pointer that points to the 116 * item equal to |data|. In other words, it searches for an item equal to |data| 117 * and, if it's at the start of a chain, then it returns a pointer to an 118 * element of |lh->buckets|, otherwise it returns a pointer to the |next| 119 * element of the previous item in the chain. If an element equal to |data| is 120 * not found, it returns a pointer that points to a NULL pointer. If |out_hash| 121 * is not NULL, then it also puts the hash value of |data| in |*out_hash|. */ 122 static LHASH_ITEM **get_next_ptr_and_hash(const _LHASH *lh, uint32_t *out_hash, 123 const void *data) { 124 const uint32_t hash = lh->hash(data); 125 LHASH_ITEM *cur, **ret; 126 127 if (out_hash != NULL) { 128 *out_hash = hash; 129 } 130 131 ret = &lh->buckets[hash % lh->num_buckets]; 132 for (cur = *ret; cur != NULL; cur = *ret) { 133 if (lh->comp(cur->data, data) == 0) { 134 break; 135 } 136 ret = &cur->next; 137 } 138 139 return ret; 140 } 141 142 void *lh_retrieve(const _LHASH *lh, const void *data) { 143 LHASH_ITEM **next_ptr; 144 145 next_ptr = get_next_ptr_and_hash(lh, NULL, data); 146 147 if (*next_ptr == NULL) { 148 return NULL; 149 } 150 151 return (*next_ptr)->data; 152 } 153 154 /* lh_rebucket allocates a new array of |new_num_buckets| pointers and 155 * redistributes the existing items into it before making it |lh->buckets| and 156 * freeing the old array. */ 157 static void lh_rebucket(_LHASH *lh, const size_t new_num_buckets) { 158 LHASH_ITEM **new_buckets, *cur, *next; 159 size_t i, alloc_size; 160 161 alloc_size = sizeof(LHASH_ITEM *) * new_num_buckets; 162 if (alloc_size / sizeof(LHASH_ITEM*) != new_num_buckets) { 163 return; 164 } 165 166 new_buckets = OPENSSL_malloc(alloc_size); 167 if (new_buckets == NULL) { 168 return; 169 } 170 OPENSSL_memset(new_buckets, 0, alloc_size); 171 172 for (i = 0; i < lh->num_buckets; i++) { 173 for (cur = lh->buckets[i]; cur != NULL; cur = next) { 174 const size_t new_bucket = cur->hash % new_num_buckets; 175 next = cur->next; 176 cur->next = new_buckets[new_bucket]; 177 new_buckets[new_bucket] = cur; 178 } 179 } 180 181 OPENSSL_free(lh->buckets); 182 183 lh->num_buckets = new_num_buckets; 184 lh->buckets = new_buckets; 185 } 186 187 /* lh_maybe_resize resizes the |buckets| array if needed. */ 188 static void lh_maybe_resize(_LHASH *lh) { 189 size_t avg_chain_length; 190 191 if (lh->callback_depth > 0) { 192 /* Don't resize the hash if we are currently iterating over it. */ 193 return; 194 } 195 196 assert(lh->num_buckets >= kMinNumBuckets); 197 avg_chain_length = lh->num_items / lh->num_buckets; 198 199 if (avg_chain_length > kMaxAverageChainLength) { 200 const size_t new_num_buckets = lh->num_buckets * 2; 201 202 if (new_num_buckets > lh->num_buckets) { 203 lh_rebucket(lh, new_num_buckets); 204 } 205 } else if (avg_chain_length < kMinAverageChainLength && 206 lh->num_buckets > kMinNumBuckets) { 207 size_t new_num_buckets = lh->num_buckets / 2; 208 209 if (new_num_buckets < kMinNumBuckets) { 210 new_num_buckets = kMinNumBuckets; 211 } 212 213 lh_rebucket(lh, new_num_buckets); 214 } 215 } 216 217 int lh_insert(_LHASH *lh, void **old_data, void *data) { 218 uint32_t hash; 219 LHASH_ITEM **next_ptr, *item; 220 221 *old_data = NULL; 222 next_ptr = get_next_ptr_and_hash(lh, &hash, data); 223 224 225 if (*next_ptr != NULL) { 226 /* An element equal to |data| already exists in the hash table. It will be 227 * replaced. */ 228 *old_data = (*next_ptr)->data; 229 (*next_ptr)->data = data; 230 return 1; 231 } 232 233 /* An element equal to |data| doesn't exist in the hash table yet. */ 234 item = OPENSSL_malloc(sizeof(LHASH_ITEM)); 235 if (item == NULL) { 236 return 0; 237 } 238 239 item->data = data; 240 item->hash = hash; 241 item->next = NULL; 242 *next_ptr = item; 243 lh->num_items++; 244 lh_maybe_resize(lh); 245 246 return 1; 247 } 248 249 void *lh_delete(_LHASH *lh, const void *data) { 250 LHASH_ITEM **next_ptr, *item, *ret; 251 252 next_ptr = get_next_ptr_and_hash(lh, NULL, data); 253 254 if (*next_ptr == NULL) { 255 /* No such element. */ 256 return NULL; 257 } 258 259 item = *next_ptr; 260 *next_ptr = item->next; 261 ret = item->data; 262 OPENSSL_free(item); 263 264 lh->num_items--; 265 lh_maybe_resize(lh); 266 267 return ret; 268 } 269 270 static void lh_doall_internal(_LHASH *lh, void (*no_arg_func)(void *), 271 void (*arg_func)(void *, void *), void *arg) { 272 if (lh == NULL) { 273 return; 274 } 275 276 if (lh->callback_depth < UINT_MAX) { 277 /* |callback_depth| is a saturating counter. */ 278 lh->callback_depth++; 279 } 280 281 for (size_t i = 0; i < lh->num_buckets; i++) { 282 LHASH_ITEM *next; 283 for (LHASH_ITEM *cur = lh->buckets[i]; cur != NULL; cur = next) { 284 next = cur->next; 285 if (arg_func) { 286 arg_func(cur->data, arg); 287 } else { 288 no_arg_func(cur->data); 289 } 290 } 291 } 292 293 if (lh->callback_depth < UINT_MAX) { 294 lh->callback_depth--; 295 } 296 297 /* The callback may have added or removed elements and the non-zero value of 298 * |callback_depth| will have suppressed any resizing. Thus any needed 299 * resizing is done here. */ 300 lh_maybe_resize(lh); 301 } 302 303 void lh_doall(_LHASH *lh, void (*func)(void *)) { 304 lh_doall_internal(lh, func, NULL, NULL); 305 } 306 307 void lh_doall_arg(_LHASH *lh, void (*func)(void *, void *), void *arg) { 308 lh_doall_internal(lh, NULL, func, arg); 309 } 310 311 uint32_t lh_strhash(const char *c) { 312 /* The following hash seems to work very well on normal text strings 313 * no collisions on /usr/dict/words and it distributes on %2^n quite 314 * well, not as good as MD5, but still good. */ 315 unsigned long ret = 0; 316 long n; 317 unsigned long v; 318 int r; 319 320 if ((c == NULL) || (*c == '\0')) { 321 return (ret); 322 } 323 324 n = 0x100; 325 while (*c) { 326 v = n | (*c); 327 n += 0x100; 328 r = (int)((v >> 2) ^ v) & 0x0f; 329 ret = (ret << r) | (ret >> (32 - r)); 330 ret &= 0xFFFFFFFFL; 331 ret ^= v * v; 332 c++; 333 } 334 335 return ((ret >> 16) ^ ret); 336 } 337