src/util/set.c

   1 /*
   2  * Copyright © 2009-2012 Intel Corporation
   3  * Copyright © 1988-2004 Keith Packard and Bart Massey.
   4  *
   5  * Permission is hereby granted, free of charge, to any person obtaining a
   6  * copy of this software and associated documentation files (the "Software"),
   7  * to deal in the Software without restriction, including without limitation
   8  * the rights to use, copy, modify, merge, publish, distribute, sublicense,
   9  * and/or sell copies of the Software, and to permit persons to whom the
  10  * Software is furnished to do so, subject to the following conditions:
  11  *
  12  * The above copyright notice and this permission notice (including the next
  13  * paragraph) shall be included in all copies or substantial portions of the
  14  * Software.
  15  *
  16  * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
  17  * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
  18  * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
  19  * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
  20  * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
  21  * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
  22  * IN THE SOFTWARE.
  23  *
  24  * Except as contained in this notice, the names of the authors
  25  * or their institutions shall not be used in advertising or
  26  * otherwise to promote the sale, use or other dealings in this
  27  * Software without prior written authorization from the
  28  * authors.
  29  *
  30  * Authors:
  31  *    Eric Anholt <eric@anholt.net>
  32  *    Keith Packard <keithp@keithp.com>
  33  */
  34
  35 #include <stdlib.h>
  36
  37 #include "macros.h"
  38 #include "ralloc.h"
  39 #include "set.h"
  40
  41 /*
  42  * From Knuth -- a good choice for hash/rehash values is p, p-2 where
  43  * p and p-2 are both prime.  These tables are sized to have an extra 10%
  44  * free to avoid exponential performance degradation as the hash table fills
  45  */
  46
  47 uint32_t deleted_key_value;
  48 const void *deleted_key = &deleted_key_value;
  49
  50 static const struct {
  51    uint32_t max_entries, size, rehash;
  52 } hash_sizes[] = {
  53    { 2,            5,            3            },
  54    { 4,            7,            5            },
  55    { 8,            13,           11           },
  56    { 16,           19,           17           },
  57    { 32,           43,           41           },
  58    { 64,           73,           71           },
  59    { 128,          151,          149          },
  60    { 256,          283,          281          },
  61    { 512,          571,          569          },
  62    { 1024,         1153,         1151         },
  63    { 2048,         2269,         2267         },
  64    { 4096,         4519,         4517         },
  65    { 8192,         9013,         9011         },
  66    { 16384,        18043,        18041        },
  67    { 32768,        36109,        36107        },
  68    { 65536,        72091,        72089        },
  69    { 131072,       144409,       144407       },
  70    { 262144,       288361,       288359       },
  71    { 524288,       576883,       576881       },
  72    { 1048576,      1153459,      1153457      },
  73    { 2097152,      2307163,      2307161      },
  74    { 4194304,      4613893,      4613891      },
  75    { 8388608,      9227641,      9227639      },
  76    { 16777216,     18455029,     18455027     },
  77    { 33554432,     36911011,     36911009     },
  78    { 67108864,     73819861,     73819859     },
  79    { 134217728,    147639589,    147639587    },
  80    { 268435456,    295279081,    295279079    },
  81    { 536870912,    590559793,    590559791    },
  82    { 1073741824,   1181116273,   1181116271   },
  83    { 2147483648ul, 2362232233ul, 2362232231ul }
  84 };
  85
  86 static int
  87 entry_is_free(struct set_entry *entry)
  88 {
  89    return entry->key == NULL;
  90 }
  91
  92 static int
  93 entry_is_deleted(struct set_entry *entry)
  94 {
  95    return entry->key == deleted_key;
  96 }
  97
  98 static int
  99 entry_is_present(struct set_entry *entry)
 100 {
 101    return entry->key != NULL && entry->key != deleted_key;
 102 }
 103
 104 struct set *
 105 _mesa_set_create(void *mem_ctx,
 106                  bool (*key_equals_function)(const void *a,
 107                                              const void *b))
 108 {
 109    struct set *ht;
 110
 111    ht = ralloc(mem_ctx, struct set);
 112    if (ht == NULL)
 113       return NULL;
 114
 115    ht->size_index = 0;
 116    ht->size = hash_sizes[ht->size_index].size;
 117    ht->rehash = hash_sizes[ht->size_index].rehash;
 118    ht->max_entries = hash_sizes[ht->size_index].max_entries;
 119    ht->key_equals_function = key_equals_function;
 120    ht->table = rzalloc_array(ht, struct set_entry, ht->size);
 121    ht->entries = 0;
 122    ht->deleted_entries = 0;
 123
 124    if (ht->table == NULL) {
 125       ralloc_free(ht);
 126       return NULL;
 127    }
 128
 129    return ht;
 130 }
 131
 132 /**
 133  * Frees the given set.
 134  *
 135  * If delete_function is passed, it gets called on each entry present before
 136  * freeing.
 137  */
 138 void
 139 _mesa_set_destroy(struct set *ht, void (*delete_function)(struct set_entry *entry))
 140 {
 141    if (!ht)
 142       return;
 143
 144    if (delete_function) {
 145       struct set_entry *entry;
 146
 147       set_foreach (ht, entry) {
 148          delete_function(entry);
 149       }
 150    }
 151    ralloc_free(ht->table);
 152    ralloc_free(ht);
 153 }
 154
 155 /**
 156  * Finds a set entry with the given key and hash of that key.
 157  *
 158  * Returns NULL if no entry is found.
 159  */
 160 struct set_entry *
 161 _mesa_set_search(const struct set *ht, uint32_t hash, const void *key)
 162 {
 163    uint32_t hash_address;
 164
 165    hash_address = hash % ht->size;
 166    do {
 167       uint32_t double_hash;
 168
 169       struct set_entry *entry = ht->table + hash_address;
 170
 171       if (entry_is_free(entry)) {
 172          return NULL;
 173       } else if (entry_is_present(entry) && entry->hash == hash) {
 174          if (ht->key_equals_function(key, entry->key)) {
 175             return entry;
 176          }
 177       }
 178
 179       double_hash = 1 + hash % ht->rehash;
 180
 181       hash_address = (hash_address + double_hash) % ht->size;
 182    } while (hash_address != hash % ht->size);
 183
 184    return NULL;
 185 }
 186
 187 static void
 188 set_rehash(struct set *ht, int new_size_index)
 189 {
 190    struct set old_ht;
 191    struct set_entry *table, *entry;
 192
 193    if (new_size_index >= ARRAY_SIZE(hash_sizes))
 194       return;
 195
 196    table = rzalloc_array(ht, struct set_entry,
 197                          hash_sizes[new_size_index].size);
 198    if (table == NULL)
 199       return;
 200
 201    old_ht = *ht;
 202
 203    ht->table = table;
 204    ht->size_index = new_size_index;
 205    ht->size = hash_sizes[ht->size_index].size;
 206    ht->rehash = hash_sizes[ht->size_index].rehash;
 207    ht->max_entries = hash_sizes[ht->size_index].max_entries;
 208    ht->entries = 0;
 209    ht->deleted_entries = 0;
 210
 211    for (entry = old_ht.table;
 212         entry != old_ht.table + old_ht.size;
 213         entry++) {
 214       if (entry_is_present(entry)) {
 215          _mesa_set_add(ht, entry->hash, entry->key);
 216       }
 217    }
 218
 219    ralloc_free(old_ht.table);
 220 }
 221
 222 /**
 223  * Inserts the key with the given hash into the table.
 224  *
 225  * Note that insertion may rearrange the table on a resize or rehash,
 226  * so previously found hash_entries are no longer valid after this function.
 227  */
 228 struct set_entry *
 229 _mesa_set_add(struct set *ht, uint32_t hash, const void *key)
 230 {
 231    uint32_t hash_address;
 232
 233    if (ht->entries >= ht->max_entries) {
 234       set_rehash(ht, ht->size_index + 1);
 235    } else if (ht->deleted_entries + ht->entries >= ht->max_entries) {
 236       set_rehash(ht, ht->size_index);
 237    }
 238
 239    hash_address = hash % ht->size;
 240    do {
 241       struct set_entry *entry = ht->table + hash_address;
 242       uint32_t double_hash;
 243
 244       if (!entry_is_present(entry)) {
 245          if (entry_is_deleted(entry))
 246             ht->deleted_entries--;
 247          entry->hash = hash;
 248          entry->key = key;
 249          ht->entries++;
 250          return entry;
 251       }
 252
 253       /* Implement replacement when another insert happens
 254        * with a matching key.  This is a relatively common
 255        * feature of hash tables, with the alternative
 256        * generally being "insert the new value as well, and
 257        * return it first when the key is searched for".
 258        *
 259        * Note that the hash table doesn't have a delete callback.
 260        * If freeing of old keys is required to avoid memory leaks,
 261        * perform a search before inserting.
 262        */
 263       if (entry->hash == hash &&
 264           ht->key_equals_function(key, entry->key)) {
 265          entry->key = key;
 266          return entry;
 267       }
 268
 269       double_hash = 1 + hash % ht->rehash;
 270
 271       hash_address = (hash_address + double_hash) % ht->size;
 272    } while (hash_address != hash % ht->size);
 273
 274    /* We could hit here if a required resize failed. An unchecked-malloc
 275     * application could ignore this result.
 276     */
 277    return NULL;
 278 }
 279
 280 /**
 281  * This function deletes the given hash table entry.
 282  *
 283  * Note that deletion doesn't otherwise modify the table, so an iteration over
 284  * the table deleting entries is safe.
 285  */
 286 void
 287 _mesa_set_remove(struct set *ht, struct set_entry *entry)
 288 {
 289    if (!entry)
 290       return;
 291
 292    entry->key = deleted_key;
 293    ht->entries--;
 294    ht->deleted_entries++;
 295 }
 296
 297 /**
 298  * This function is an iterator over the hash table.
 299  *
 300  * Pass in NULL for the first entry, as in the start of a for loop.  Note that
 301  * an iteration over the table is O(table_size) not O(entries).
 302  */
 303 struct set_entry *
 304 _mesa_set_next_entry(const struct set *ht, struct set_entry *entry)
 305 {
 306    if (entry == NULL)
 307       entry = ht->table;
 308    else
 309       entry = entry + 1;
 310
 311    for (; entry != ht->table + ht->size; entry++) {
 312       if (entry_is_present(entry)) {
 313          return entry;
 314       }
 315    }
 316
 317    return NULL;
 318 }
 319
 320 struct set_entry *
 321 _mesa_set_random_entry(struct set *ht,
 322                        int (*predicate)(struct set_entry *entry))
 323 {
 324    struct set_entry *entry;
 325    uint32_t i = rand() % ht->size;
 326
 327    if (ht->entries == 0)
 328       return NULL;
 329
 330    for (entry = ht->table + i; entry != ht->table + ht->size; entry++) {
 331       if (entry_is_present(entry) &&
 332           (!predicate || predicate(entry))) {
 333          return entry;
 334       }
 335    }
 336
 337    for (entry = ht->table; entry != ht->table + i; entry++) {
 338       if (entry_is_present(entry) &&
 339           (!predicate || predicate(entry))) {
 340          return entry;
 341       }
 342    }
 343
 344    return NULL;
 345 }