| 1 |
/* Copyright (C) 1995-1998 Eric Young ([email protected]) |
| 2 |
* All rights reserved. |
| 3 |
* |
| 4 |
* This package is an SSL implementation written |
| 5 |
* by Eric Young ([email protected]). |
| 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 ([email protected]). |
| 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 ([email protected])" |
| 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 ([email protected])" |
| 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 |
/* kMinNumBuckets is the minimum size of the buckets array in an |_LHASH|. */ |
| 66 |
static const size_t kMinNumBuckets = 16; |
| 67 |
|
| 68 |
/* kMaxAverageChainLength contains the maximum, average chain length. When the |
| 69 |
* average chain length exceeds this value, the hash table will be resized. */ |
| 70 |
static const size_t kMaxAverageChainLength = 2; |
| 71 |
static const size_t kMinAverageChainLength = 1; |
| 72 |
|
| 73 |
_LHASH *lh_new(lhash_hash_func hash, lhash_cmp_func comp) { |
| 74 |
_LHASH *ret; |
| 75 |
|
| 76 |
ret = OPENSSL_malloc(sizeof(_LHASH)); |
| 77 |
if (ret == NULL) { |
| 78 |
return NULL; |
| 79 |
} |
| 80 |
memset(ret, 0, sizeof(_LHASH)); |
| 81 |
|
| 82 |
ret->num_buckets = kMinNumBuckets; |
| 83 |
ret->buckets = OPENSSL_malloc(sizeof(LHASH_ITEM *) * ret->num_buckets); |
| 84 |
if (ret->buckets == NULL) { |
| 85 |
OPENSSL_free(ret); |
| 86 |
return NULL; |
| 87 |
} |
| 88 |
memset(ret->buckets, 0, sizeof(LHASH_ITEM *) * ret->num_buckets); |
| 89 |
|
| 90 |
ret->comp = comp; |
| 91 |
if (ret->comp == NULL) { |
| 92 |
ret->comp = (lhash_cmp_func) strcmp; |
| 93 |
} |
| 94 |
ret->hash = hash; |
| 95 |
if (ret->hash == NULL) { |
| 96 |
ret->hash = (lhash_hash_func) lh_strhash; |
| 97 |
} |
| 98 |
|
| 99 |
return ret; |
| 100 |
} |
| 101 |
|
| 102 |
void lh_free(_LHASH *lh) { |
| 103 |
size_t i; |
| 104 |
LHASH_ITEM *n, *next; |
| 105 |
|
| 106 |
if (lh == NULL) { |
| 107 |
return; |
| 108 |
} |
| 109 |
|
| 110 |
for (i = 0; i < lh->num_buckets; i++) { |
| 111 |
for (n = lh->buckets[i]; n != NULL; n = next) { |
| 112 |
next = n->next; |
| 113 |
OPENSSL_free(n); |
| 114 |
} |
| 115 |
} |
| 116 |
|
| 117 |
OPENSSL_free(lh->buckets); |
| 118 |
OPENSSL_free(lh); |
| 119 |
} |
| 120 |
|
| 121 |
size_t lh_num_items(const _LHASH *lh) { return lh->num_items; } |
| 122 |
|
| 123 |
/* get_next_ptr_and_hash returns a pointer to the pointer that points to the |
| 124 |
* item equal to |data|. In other words, it searches for an item equal to |data| |
| 125 |
* and, if it's at the start of a chain, then it returns a pointer to an |
| 126 |
* element of |lh->buckets|, otherwise it returns a pointer to the |next| |
| 127 |
* element of the previous item in the chain. If an element equal to |data| is |
| 128 |
* not found, it returns a pointer that points to a NULL pointer. If |out_hash| |
| 129 |
* is not NULL, then it also puts the hash value of |data| in |*out_hash|. */ |
| 130 |
static LHASH_ITEM **get_next_ptr_and_hash(const _LHASH *lh, uint32_t *out_hash, |
| 131 |
const void *data) { |
| 132 |
const uint32_t hash = lh->hash(data); |
| 133 |
LHASH_ITEM *cur, **ret; |
| 134 |
|
| 135 |
if (out_hash != NULL) { |
| 136 |
*out_hash = hash; |
| 137 |
} |
| 138 |
|
| 139 |
ret = &lh->buckets[hash % lh->num_buckets]; |
| 140 |
for (cur = *ret; cur != NULL; cur = *ret) { |
| 141 |
if (lh->comp(cur->data, data) == 0) { |
| 142 |
break; |
| 143 |
} |
| 144 |
ret = &cur->next; |
| 145 |
} |
| 146 |
|
| 147 |
return ret; |
| 148 |
} |
| 149 |
|
| 150 |
void *lh_retrieve(const _LHASH *lh, const void *data) { |
| 151 |
LHASH_ITEM **next_ptr; |
| 152 |
|
| 153 |
next_ptr = get_next_ptr_and_hash(lh, NULL, data); |
| 154 |
|
| 155 |
if (*next_ptr == NULL) { |
| 156 |
return NULL; |
| 157 |
} |
| 158 |
|
| 159 |
return (*next_ptr)->data; |
| 160 |
} |
| 161 |
|
| 162 |
/* lh_rebucket allocates a new array of |new_num_buckets| pointers and |
| 163 |
* redistributes the existing items into it before making it |lh->buckets| and |
| 164 |
* freeing the old array. */ |
| 165 |
static void lh_rebucket(_LHASH *lh, const size_t new_num_buckets) { |
| 166 |
LHASH_ITEM **new_buckets, *cur, *next; |
| 167 |
size_t i, alloc_size; |
| 168 |
|
| 169 |
alloc_size = sizeof(LHASH_ITEM *) * new_num_buckets; |
| 170 |
if (alloc_size / sizeof(LHASH_ITEM*) != new_num_buckets) { |
| 171 |
return; |
| 172 |
} |
| 173 |
|
| 174 |
new_buckets = OPENSSL_malloc(alloc_size); |
| 175 |
if (new_buckets == NULL) { |
| 176 |
return; |
| 177 |
} |
| 178 |
memset(new_buckets, 0, alloc_size); |
| 179 |
|
| 180 |
for (i = 0; i < lh->num_buckets; i++) { |
| 181 |
for (cur = lh->buckets[i]; cur != NULL; cur = next) { |
| 182 |
const size_t new_bucket = cur->hash % new_num_buckets; |
| 183 |
next = cur->next; |
| 184 |
cur->next = new_buckets[new_bucket]; |
| 185 |
new_buckets[new_bucket] = cur; |
| 186 |
} |
| 187 |
} |
| 188 |
|
| 189 |
OPENSSL_free(lh->buckets); |
| 190 |
|
| 191 |
lh->num_buckets = new_num_buckets; |
| 192 |
lh->buckets = new_buckets; |
| 193 |
} |
| 194 |
|
| 195 |
/* lh_maybe_resize resizes the |buckets| array if needed. */ |
| 196 |
static void lh_maybe_resize(_LHASH *lh) { |
| 197 |
size_t avg_chain_length; |
| 198 |
|
| 199 |
if (lh->callback_depth > 0) { |
| 200 |
/* Don't resize the hash if we are currently iterating over it. */ |
| 201 |
return; |
| 202 |
} |
| 203 |
|
| 204 |
assert(lh->num_buckets >= kMinNumBuckets); |
| 205 |
avg_chain_length = lh->num_items / lh->num_buckets; |
| 206 |
|
| 207 |
if (avg_chain_length > kMaxAverageChainLength) { |
| 208 |
const size_t new_num_buckets = lh->num_buckets * 2; |
| 209 |
|
| 210 |
if (new_num_buckets > lh->num_buckets) { |
| 211 |
lh_rebucket(lh, new_num_buckets); |
| 212 |
} |
| 213 |
} else if (avg_chain_length < kMinAverageChainLength && |
| 214 |
lh->num_buckets > kMinNumBuckets) { |
| 215 |
size_t new_num_buckets = lh->num_buckets / 2; |
| 216 |
|
| 217 |
if (new_num_buckets < kMinNumBuckets) { |
| 218 |
new_num_buckets = kMinNumBuckets; |
| 219 |
} |
| 220 |
|
| 221 |
lh_rebucket(lh, new_num_buckets); |
| 222 |
} |
| 223 |
} |
| 224 |
|
| 225 |
int lh_insert(_LHASH *lh, void **old_data, void *data) { |
| 226 |
uint32_t hash; |
| 227 |
LHASH_ITEM **next_ptr, *item; |
| 228 |
|
| 229 |
*old_data = NULL; |
| 230 |
next_ptr = get_next_ptr_and_hash(lh, &hash, data); |
| 231 |
|
| 232 |
|
| 233 |
if (*next_ptr != NULL) { |
| 234 |
/* An element equal to |data| already exists in the hash table. It will be |
| 235 |
* replaced. */ |
| 236 |
*old_data = (*next_ptr)->data; |
| 237 |
(*next_ptr)->data = data; |
| 238 |
return 1; |
| 239 |
} |
| 240 |
|
| 241 |
/* An element equal to |data| doesn't exist in the hash table yet. */ |
| 242 |
item = OPENSSL_malloc(sizeof(LHASH_ITEM)); |
| 243 |
if (item == NULL) { |
| 244 |
return 0; |
| 245 |
} |
| 246 |
|
| 247 |
item->data = data; |
| 248 |
item->hash = hash; |
| 249 |
item->next = NULL; |
| 250 |
*next_ptr = item; |
| 251 |
lh->num_items++; |
| 252 |
lh_maybe_resize(lh); |
| 253 |
|
| 254 |
return 1; |
| 255 |
} |
| 256 |
|
| 257 |
void *lh_delete(_LHASH *lh, const void *data) { |
| 258 |
LHASH_ITEM **next_ptr, *item, *ret; |
| 259 |
|
| 260 |
next_ptr = get_next_ptr_and_hash(lh, NULL, data); |
| 261 |
|
| 262 |
if (*next_ptr == NULL) { |
| 263 |
/* No such element. */ |
| 264 |
return NULL; |
| 265 |
} |
| 266 |
|
| 267 |
item = *next_ptr; |
| 268 |
*next_ptr = item->next; |
| 269 |
ret = item->data; |
| 270 |
OPENSSL_free(item); |
| 271 |
|
| 272 |
lh->num_items--; |
| 273 |
lh_maybe_resize(lh); |
| 274 |
|
| 275 |
return ret; |
| 276 |
} |
| 277 |
|
| 278 |
static void lh_doall_internal(_LHASH *lh, void (*no_arg_func)(void *), |
| 279 |
void (*arg_func)(void *, void *), void *arg) { |
| 280 |
size_t i; |
| 281 |
LHASH_ITEM *cur, *next; |
| 282 |
|
| 283 |
if (lh == NULL) { |
| 284 |
return; |
| 285 |
} |
| 286 |
|
| 287 |
if (lh->callback_depth < UINT_MAX) { |
| 288 |
/* |callback_depth| is a saturating counter. */ |
| 289 |
lh->callback_depth++; |
| 290 |
} |
| 291 |
|
| 292 |
for (i = 0; i < lh->num_buckets; i++) { |
| 293 |
for (cur = lh->buckets[i]; cur != NULL; cur = next) { |
| 294 |
next = cur->next; |
| 295 |
if (arg_func) { |
| 296 |
arg_func(cur->data, arg); |
| 297 |
} else { |
| 298 |
no_arg_func(cur->data); |
| 299 |
} |
| 300 |
} |
| 301 |
} |
| 302 |
|
| 303 |
if (lh->callback_depth < UINT_MAX) { |
| 304 |
lh->callback_depth--; |
| 305 |
} |
| 306 |
|
| 307 |
/* The callback may have added or removed elements and the non-zero value of |
| 308 |
* |callback_depth| will have suppressed any resizing. Thus any needed |
| 309 |
* resizing is done here. */ |
| 310 |
lh_maybe_resize(lh); |
| 311 |
} |
| 312 |
|
| 313 |
void lh_doall(_LHASH *lh, void (*func)(void *)) { |
| 314 |
lh_doall_internal(lh, func, NULL, NULL); |
| 315 |
} |
| 316 |
|
| 317 |
void lh_doall_arg(_LHASH *lh, void (*func)(void *, void *), void *arg) { |
| 318 |
lh_doall_internal(lh, NULL, func, arg); |
| 319 |
} |
| 320 |
|
| 321 |
uint32_t lh_strhash(const char *c) { |
| 322 |
/* The following hash seems to work very well on normal text strings |
| 323 |
* no collisions on /usr/dict/words and it distributes on %2^n quite |
| 324 |
* well, not as good as MD5, but still good. */ |
| 325 |
unsigned long ret = 0; |
| 326 |
long n; |
| 327 |
unsigned long v; |
| 328 |
int r; |
| 329 |
|
| 330 |
if ((c == NULL) || (*c == '\0')) { |
| 331 |
return (ret); |
| 332 |
} |
| 333 |
|
| 334 |
n = 0x100; |
| 335 |
while (*c) { |
| 336 |
v = n | (*c); |
| 337 |
n += 0x100; |
| 338 |
r = (int)((v >> 2) ^ v) & 0x0f; |
| 339 |
ret = (ret << r) | (ret >> (32 - r)); |
| 340 |
ret &= 0xFFFFFFFFL; |
| 341 |
ret ^= v * v; |
| 342 |
c++; |
| 343 |
} |
| 344 |
|
| 345 |
return ((ret >> 16) ^ ret); |
| 346 |
} |