| 1 | /* Copyright (C) 1995-1998 Eric Young (eay@cryptsoft.com) |
| 2 | * All rights reserved. |
| 3 | * |
| 4 | * This package is an SSL implementation written |
| 5 | * by Eric Young (eay@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@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@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@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 | struct lhash_st { |
| 77 | // num_items contains the total number of items in the hash table. |
| 78 | size_t num_items; |
| 79 | // buckets is an array of |num_buckets| pointers. Each points to the head of |
| 80 | // a chain of LHASH_ITEM objects that have the same hash value, mod |
| 81 | // |num_buckets|. |
| 82 | LHASH_ITEM **buckets; |
| 83 | // num_buckets contains the length of |buckets|. This value is always >= |
| 84 | // kMinNumBuckets. |
| 85 | size_t num_buckets; |
| 86 | // callback_depth contains the current depth of |lh_doall| or |lh_doall_arg| |
| 87 | // calls. If non-zero then this suppresses resizing of the |buckets| array, |
| 88 | // which would otherwise disrupt the iteration. |
| 89 | unsigned callback_depth; |
| 90 | |
| 91 | lhash_cmp_func comp; |
| 92 | lhash_hash_func hash; |
| 93 | }; |
| 94 | |
| 95 | _LHASH *lh_new(lhash_hash_func hash, lhash_cmp_func comp) { |
| 96 | _LHASH *ret = OPENSSL_malloc(sizeof(_LHASH)); |
| 97 | if (ret == NULL) { |
| 98 | return NULL; |
| 99 | } |
| 100 | OPENSSL_memset(ret, 0, sizeof(_LHASH)); |
| 101 | |
| 102 | ret->num_buckets = kMinNumBuckets; |
| 103 | ret->buckets = OPENSSL_malloc(sizeof(LHASH_ITEM *) * ret->num_buckets); |
| 104 | if (ret->buckets == NULL) { |
| 105 | OPENSSL_free(ret); |
| 106 | return NULL; |
| 107 | } |
| 108 | OPENSSL_memset(ret->buckets, 0, sizeof(LHASH_ITEM *) * ret->num_buckets); |
| 109 | |
| 110 | ret->comp = comp; |
| 111 | ret->hash = hash; |
| 112 | return ret; |
| 113 | } |
| 114 | |
| 115 | void lh_free(_LHASH *lh) { |
| 116 | if (lh == NULL) { |
| 117 | return; |
| 118 | } |
| 119 | |
| 120 | for (size_t i = 0; i < lh->num_buckets; i++) { |
| 121 | LHASH_ITEM *next; |
| 122 | for (LHASH_ITEM *n = lh->buckets[i]; n != NULL; n = next) { |
| 123 | next = n->next; |
| 124 | OPENSSL_free(n); |
| 125 | } |
| 126 | } |
| 127 | |
| 128 | OPENSSL_free(lh->buckets); |
| 129 | OPENSSL_free(lh); |
| 130 | } |
| 131 | |
| 132 | size_t lh_num_items(const _LHASH *lh) { return lh->num_items; } |
| 133 | |
| 134 | // get_next_ptr_and_hash returns a pointer to the pointer that points to the |
| 135 | // item equal to |data|. In other words, it searches for an item equal to |data| |
| 136 | // and, if it's at the start of a chain, then it returns a pointer to an |
| 137 | // element of |lh->buckets|, otherwise it returns a pointer to the |next| |
| 138 | // element of the previous item in the chain. If an element equal to |data| is |
| 139 | // not found, it returns a pointer that points to a NULL pointer. If |out_hash| |
| 140 | // is not NULL, then it also puts the hash value of |data| in |*out_hash|. |
| 141 | static LHASH_ITEM **get_next_ptr_and_hash(const _LHASH *lh, uint32_t *out_hash, |
| 142 | const void *data, |
| 143 | lhash_hash_func_helper call_hash_func, |
| 144 | lhash_cmp_func_helper call_cmp_func) { |
| 145 | const uint32_t hash = call_hash_func(lh->hash, data); |
| 146 | if (out_hash != NULL) { |
| 147 | *out_hash = hash; |
| 148 | } |
| 149 | |
| 150 | LHASH_ITEM **ret = &lh->buckets[hash % lh->num_buckets]; |
| 151 | for (LHASH_ITEM *cur = *ret; cur != NULL; cur = *ret) { |
| 152 | if (call_cmp_func(lh->comp, cur->data, data) == 0) { |
| 153 | break; |
| 154 | } |
| 155 | ret = &cur->next; |
| 156 | } |
| 157 | |
| 158 | return ret; |
| 159 | } |
| 160 | |
| 161 | // get_next_ptr_by_key behaves like |get_next_ptr_and_hash| but takes a key |
| 162 | // which may be a different type from the values stored in |lh|. |
| 163 | static LHASH_ITEM **get_next_ptr_by_key(const _LHASH *lh, const void *key, |
| 164 | uint32_t key_hash, |
| 165 | int (*cmp_key)(const void *key, |
| 166 | const void *value)) { |
| 167 | LHASH_ITEM **ret = &lh->buckets[key_hash % lh->num_buckets]; |
| 168 | for (LHASH_ITEM *cur = *ret; cur != NULL; cur = *ret) { |
| 169 | if (cmp_key(key, cur->data) == 0) { |
| 170 | break; |
| 171 | } |
| 172 | ret = &cur->next; |
| 173 | } |
| 174 | |
| 175 | return ret; |
| 176 | } |
| 177 | |
| 178 | void *lh_retrieve(const _LHASH *lh, const void *data, |
| 179 | lhash_hash_func_helper call_hash_func, |
| 180 | lhash_cmp_func_helper call_cmp_func) { |
| 181 | LHASH_ITEM **next_ptr = |
| 182 | get_next_ptr_and_hash(lh, NULL, data, call_hash_func, call_cmp_func); |
| 183 | return *next_ptr == NULL ? NULL : (*next_ptr)->data; |
| 184 | } |
| 185 | |
| 186 | void *lh_retrieve_key(const _LHASH *lh, const void *key, uint32_t key_hash, |
| 187 | int (*cmp_key)(const void *key, const void *value)) { |
| 188 | LHASH_ITEM **next_ptr = get_next_ptr_by_key(lh, key, key_hash, cmp_key); |
| 189 | return *next_ptr == NULL ? NULL : (*next_ptr)->data; |
| 190 | } |
| 191 | |
| 192 | // lh_rebucket allocates a new array of |new_num_buckets| pointers and |
| 193 | // redistributes the existing items into it before making it |lh->buckets| and |
| 194 | // freeing the old array. |
| 195 | static void lh_rebucket(_LHASH *lh, const size_t new_num_buckets) { |
| 196 | LHASH_ITEM **new_buckets, *cur, *next; |
| 197 | size_t i, alloc_size; |
| 198 | |
| 199 | alloc_size = sizeof(LHASH_ITEM *) * new_num_buckets; |
| 200 | if (alloc_size / sizeof(LHASH_ITEM*) != new_num_buckets) { |
| 201 | return; |
| 202 | } |
| 203 | |
| 204 | new_buckets = OPENSSL_malloc(alloc_size); |
| 205 | if (new_buckets == NULL) { |
| 206 | return; |
| 207 | } |
| 208 | OPENSSL_memset(new_buckets, 0, alloc_size); |
| 209 | |
| 210 | for (i = 0; i < lh->num_buckets; i++) { |
| 211 | for (cur = lh->buckets[i]; cur != NULL; cur = next) { |
| 212 | const size_t new_bucket = cur->hash % new_num_buckets; |
| 213 | next = cur->next; |
| 214 | cur->next = new_buckets[new_bucket]; |
| 215 | new_buckets[new_bucket] = cur; |
| 216 | } |
| 217 | } |
| 218 | |
| 219 | OPENSSL_free(lh->buckets); |
| 220 | |
| 221 | lh->num_buckets = new_num_buckets; |
| 222 | lh->buckets = new_buckets; |
| 223 | } |
| 224 | |
| 225 | // lh_maybe_resize resizes the |buckets| array if needed. |
| 226 | static void lh_maybe_resize(_LHASH *lh) { |
| 227 | size_t avg_chain_length; |
| 228 | |
| 229 | if (lh->callback_depth > 0) { |
| 230 | // Don't resize the hash if we are currently iterating over it. |
| 231 | return; |
| 232 | } |
| 233 | |
| 234 | assert(lh->num_buckets >= kMinNumBuckets); |
| 235 | avg_chain_length = lh->num_items / lh->num_buckets; |
| 236 | |
| 237 | if (avg_chain_length > kMaxAverageChainLength) { |
| 238 | const size_t new_num_buckets = lh->num_buckets * 2; |
| 239 | |
| 240 | if (new_num_buckets > lh->num_buckets) { |
| 241 | lh_rebucket(lh, new_num_buckets); |
| 242 | } |
| 243 | } else if (avg_chain_length < kMinAverageChainLength && |
| 244 | lh->num_buckets > kMinNumBuckets) { |
| 245 | size_t new_num_buckets = lh->num_buckets / 2; |
| 246 | |
| 247 | if (new_num_buckets < kMinNumBuckets) { |
| 248 | new_num_buckets = kMinNumBuckets; |
| 249 | } |
| 250 | |
| 251 | lh_rebucket(lh, new_num_buckets); |
| 252 | } |
| 253 | } |
| 254 | |
| 255 | int lh_insert(_LHASH *lh, void **old_data, void *data, |
| 256 | lhash_hash_func_helper call_hash_func, |
| 257 | lhash_cmp_func_helper call_cmp_func) { |
| 258 | uint32_t hash; |
| 259 | LHASH_ITEM **next_ptr, *item; |
| 260 | |
| 261 | *old_data = NULL; |
| 262 | next_ptr = |
| 263 | get_next_ptr_and_hash(lh, &hash, data, call_hash_func, call_cmp_func); |
| 264 | |
| 265 | |
| 266 | if (*next_ptr != NULL) { |
| 267 | // An element equal to |data| already exists in the hash table. It will be |
| 268 | // replaced. |
| 269 | *old_data = (*next_ptr)->data; |
| 270 | (*next_ptr)->data = data; |
| 271 | return 1; |
| 272 | } |
| 273 | |
| 274 | // An element equal to |data| doesn't exist in the hash table yet. |
| 275 | item = OPENSSL_malloc(sizeof(LHASH_ITEM)); |
| 276 | if (item == NULL) { |
| 277 | return 0; |
| 278 | } |
| 279 | |
| 280 | item->data = data; |
| 281 | item->hash = hash; |
| 282 | item->next = NULL; |
| 283 | *next_ptr = item; |
| 284 | lh->num_items++; |
| 285 | lh_maybe_resize(lh); |
| 286 | |
| 287 | return 1; |
| 288 | } |
| 289 | |
| 290 | void *lh_delete(_LHASH *lh, const void *data, |
| 291 | lhash_hash_func_helper call_hash_func, |
| 292 | lhash_cmp_func_helper call_cmp_func) { |
| 293 | LHASH_ITEM **next_ptr, *item, *ret; |
| 294 | |
| 295 | next_ptr = |
| 296 | get_next_ptr_and_hash(lh, NULL, data, call_hash_func, call_cmp_func); |
| 297 | |
| 298 | if (*next_ptr == NULL) { |
| 299 | // No such element. |
| 300 | return NULL; |
| 301 | } |
| 302 | |
| 303 | item = *next_ptr; |
| 304 | *next_ptr = item->next; |
| 305 | ret = item->data; |
| 306 | OPENSSL_free(item); |
| 307 | |
| 308 | lh->num_items--; |
| 309 | lh_maybe_resize(lh); |
| 310 | |
| 311 | return ret; |
| 312 | } |
| 313 | |
| 314 | void lh_doall_arg(_LHASH *lh, void (*func)(void *, void *), void *arg) { |
| 315 | if (lh == NULL) { |
| 316 | return; |
| 317 | } |
| 318 | |
| 319 | if (lh->callback_depth < UINT_MAX) { |
| 320 | // |callback_depth| is a saturating counter. |
| 321 | lh->callback_depth++; |
| 322 | } |
| 323 | |
| 324 | for (size_t i = 0; i < lh->num_buckets; i++) { |
| 325 | LHASH_ITEM *next; |
| 326 | for (LHASH_ITEM *cur = lh->buckets[i]; cur != NULL; cur = next) { |
| 327 | next = cur->next; |
| 328 | func(cur->data, arg); |
| 329 | } |
| 330 | } |
| 331 | |
| 332 | if (lh->callback_depth < UINT_MAX) { |
| 333 | lh->callback_depth--; |
| 334 | } |
| 335 | |
| 336 | // The callback may have added or removed elements and the non-zero value of |
| 337 | // |callback_depth| will have suppressed any resizing. Thus any needed |
| 338 | // resizing is done here. |
| 339 | lh_maybe_resize(lh); |
| 340 | } |
| 341 | |
| 342 | uint32_t lh_strhash(const char *c) { |
| 343 | if (c == NULL) { |
| 344 | return 0; |
| 345 | } |
| 346 | |
| 347 | return OPENSSL_hash32(c, strlen(c)); |
| 348 | } |
| 349 | |