include/gmx_hash.h

   1 /*
   2  * This file is part of the GROMACS molecular simulation package.
   3  *
   4  * Copyright (c) 1991-2000, University of Groningen, The Netherlands.
   5  * Copyright (c) 2001-2012, The GROMACS development team,
   6  * check out http://www.gromacs.org for more information.
   7  * Copyright (c) 2012,2013, by the GROMACS development team, led by
   8  * David van der Spoel, Berk Hess, Erik Lindahl, and including many
   9  * others, as listed in the AUTHORS file in the top-level source
  10  * directory and at http://www.gromacs.org.
  11  *
  12  * GROMACS is free software; you can redistribute it and/or
  13  * modify it under the terms of the GNU Lesser General Public License
  14  * as published by the Free Software Foundation; either version 2.1
  15  * of the License, or (at your option) any later version.
  16  *
  17  * GROMACS is distributed in the hope that it will be useful,
  18  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  19  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  20  * Lesser General Public License for more details.
  21  *
  22  * You should have received a copy of the GNU Lesser General Public
  23  * License along with GROMACS; if not, see
  24  * http://www.gnu.org/licenses, or write to the Free Software Foundation,
  25  * Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301  USA.
  26  *
  27  * If you want to redistribute modifications to GROMACS, please
  28  * consider that scientific software is very special. Version
  29  * control is crucial - bugs must be traceable. We will be happy to
  30  * consider code for inclusion in the official distribution, but
  31  * derived work must not be called official GROMACS. Details are found
  32  * in the README & COPYING files - if they are missing, get the
  33  * official version at http://www.gromacs.org.
  34  *
  35  * To help us fund GROMACS development, we humbly ask that you cite
  36  * the research papers on the package. Check out http://www.gromacs.org.
  37  */
  38 #ifndef _gmx_hash_h
  39 #define _gmx_hash_h
  40
  41 #include "typedefs.h"
  42 #include "smalloc.h"
  43
  44 #ifdef __cplusplus
  45 extern "C" {
  46 #endif
  47
  48 /* This include file implements the simplest hash table possible.
  49  * It is limited to integer keys and integer values.
  50  * The purpose is highest efficiency and lowest memory usage possible.
  51  *
  52  * The type definition is placed in types/commrec.h, as it is used there:
  53  * typedef struct gmx_hash *gmx_hash_t
  54  */
  55
  56 typedef struct {
  57     int  key;
  58     int  val;
  59     int  next;
  60 } gmx_hash_e_t;
  61
  62 typedef struct gmx_hash {
  63     int           mod;
  64     int           mask;
  65     int           nalloc;
  66     int          *direct;
  67     gmx_hash_e_t *hash;
  68     int           nkey;
  69     int           start_space_search;
  70 } t_gmx_hash;
  71
  72 /* Clear all the entries in the hash table */
  73 static void gmx_hash_clear(gmx_hash_t hash)
  74 {
  75     int i;
  76
  77     for (i = 0; i < hash->nalloc; i++)
  78     {
  79         hash->hash[i].key  = -1;
  80         hash->hash[i].next = -1;
  81     }
  82     hash->start_space_search = hash->mod;
  83
  84     hash->nkey = 0;
  85 }
  86
  87 static void gmx_hash_realloc(gmx_hash_t hash, int nkey_used_estimate)
  88 {
  89     /* Memory requirements:
  90      * nkey_used_est*(2+1-2(1-e^-1/2))*3 ints
  91      * where nkey_used_est is the local number of keys used.
  92      *
  93      * Make the direct list twice as long as the number of local keys.
  94      * The fraction of entries in the list with:
  95      * 0   size lists: e^-f
  96      * >=1 size lists: 1 - e^-f
  97      * where f is: the #keys / mod
  98      * The fraction of keys not in the direct list is: 1-1/f(1-e^-f).
  99      * The optimal table size is roughly double the number of keys.
 100      */
 101     /* Make the hash table a power of 2 and at least double the number of keys */
 102     hash->mod = 4;
 103     while (2*nkey_used_estimate > hash->mod)
 104     {
 105         hash->mod *= 2;
 106     }
 107     hash->mask   = hash->mod - 1;
 108     hash->nalloc = over_alloc_dd(hash->mod);
 109     srenew(hash->hash, hash->nalloc);
 110
 111     if (debug != NULL)
 112     {
 113         fprintf(debug, "Hash table mod %d nalloc %d\n", hash->mod, hash->nalloc);
 114     }
 115 }
 116
 117 /* Clear all the entries in the hash table.
 118  * With the current number of keys check if the table size is still good,
 119  * if not optimize it with the currenr number of keys.
 120  */
 121 static void gmx_hash_clear_and_optimize(gmx_hash_t hash)
 122 {
 123     /* Resize the hash table when the occupation is < 1/4 or > 2/3 */
 124     if (hash->nkey > 0 &&
 125         (4*hash->nkey < hash->mod || 3*hash->nkey > 2*hash->mod))
 126     {
 127         if (debug != NULL)
 128         {
 129             fprintf(debug, "Hash table size %d #key %d: resizing\n",
 130                     hash->mod, hash->nkey);
 131         }
 132         gmx_hash_realloc(hash, hash->nkey);
 133     }
 134
 135     gmx_hash_clear(hash);
 136 }
 137
 138 static gmx_hash_t gmx_hash_init(int nkey_used_estimate)
 139 {
 140     gmx_hash_t hash;
 141
 142     snew(hash, 1);
 143     hash->hash = NULL;
 144
 145     gmx_hash_realloc(hash, nkey_used_estimate);
 146
 147     gmx_hash_clear(hash);
 148
 149     return hash;
 150 }
 151
 152 /* Set the hash entry for global atom a_gl to local atom a_loc and cell. */
 153 static void gmx_hash_set(gmx_hash_t hash, int key, int value)
 154 {
 155     int ind, ind_prev, i;
 156
 157     ind = key & hash->mask;
 158
 159     if (hash->hash[ind].key >= 0)
 160     {
 161         /* Search the last entry in the linked list for this index */
 162         ind_prev = ind;
 163         while (hash->hash[ind_prev].next >= 0)
 164         {
 165             ind_prev = hash->hash[ind_prev].next;
 166         }
 167         /* Search for space in the array */
 168         ind = hash->start_space_search;
 169         while (ind < hash->nalloc && hash->hash[ind].key >= 0)
 170         {
 171             ind++;
 172         }
 173         /* If we are at the end of the list we need to increase the size */
 174         if (ind == hash->nalloc)
 175         {
 176             hash->nalloc = over_alloc_dd(ind+1);
 177             srenew(hash->hash, hash->nalloc);
 178             for (i = ind; i < hash->nalloc; i++)
 179             {
 180                 hash->hash[i].key  = -1;
 181                 hash->hash[i].next = -1;
 182             }
 183         }
 184         hash->hash[ind_prev].next = ind;
 185
 186         hash->start_space_search = ind + 1;
 187     }
 188     hash->hash[ind].key = key;
 189     hash->hash[ind].val = value;
 190
 191     hash->nkey++;
 192 }
 193
 194 /* Delete the hash entry for key */
 195 static void gmx_hash_del(gmx_hash_t hash, int key)
 196 {
 197     int ind, ind_prev;
 198
 199     ind_prev = -1;
 200     ind      = key & hash->mask;
 201     do
 202     {
 203         if (hash->hash[ind].key == key)
 204         {
 205             if (ind_prev >= 0)
 206             {
 207                 hash->hash[ind_prev].next = hash->hash[ind].next;
 208
 209                 /* This index is a linked entry, so we free an entry.
 210                  * Check if we are creating the first empty space.
 211                  */
 212                 if (ind < hash->start_space_search)
 213                 {
 214                     hash->start_space_search = ind;
 215                 }
 216             }
 217             hash->hash[ind].key  = -1;
 218             hash->hash[ind].val  = -1;
 219             hash->hash[ind].next = -1;
 220
 221             hash->nkey--;
 222
 223             return;
 224         }
 225         ind_prev = ind;
 226         ind      = hash->hash[ind].next;
 227     }
 228     while (ind >= 0);
 229
 230     return;
 231 }
 232
 233 /* Change the value for present hash entry for key */
 234 static void gmx_hash_change_value(gmx_hash_t hash, int key, int value)
 235 {
 236     int ind;
 237
 238     ind = key & hash->mask;
 239     do
 240     {
 241         if (hash->hash[ind].key == key)
 242         {
 243             hash->hash[ind].val = value;
 244
 245             return;
 246         }
 247         ind = hash->hash[ind].next;
 248     }
 249     while (ind >= 0);
 250
 251     return;
 252 }
 253
 254 /* Change the hash value if already set, otherwise set the hash value */
 255 static void gmx_hash_change_or_set(gmx_hash_t hash, int key, int value)
 256 {
 257     int ind;
 258
 259     ind = key & hash->mask;
 260     do
 261     {
 262         if (hash->hash[ind].key == key)
 263         {
 264             hash->hash[ind].val = value;
 265
 266             return;
 267         }
 268         ind = hash->hash[ind].next;
 269     }
 270     while (ind >= 0);
 271
 272     gmx_hash_set(hash, key, value);
 273
 274     return;
 275 }
 276
 277 /* Returns if the key is present, if the key is present *value is set */
 278 static gmx_bool gmx_hash_get(const gmx_hash_t hash, int key, int *value)
 279 {
 280     int ind;
 281
 282     ind = key & hash->mask;
 283     do
 284     {
 285         if (hash->hash[ind].key == key)
 286         {
 287             *value = hash->hash[ind].val;
 288
 289             return TRUE;
 290         }
 291         ind = hash->hash[ind].next;
 292     }
 293     while (ind >= 0);
 294
 295     return FALSE;
 296 }
 297
 298 /* Returns the value or -1 if the key is not present */
 299 static int gmx_hash_get_minone(const gmx_hash_t hash, int key)
 300 {
 301     int ind;
 302
 303     ind = key & hash->mask;
 304     do
 305     {
 306         if (hash->hash[ind].key == key)
 307         {
 308             return hash->hash[ind].val;
 309         }
 310         ind = hash->hash[ind].next;
 311     }
 312     while (ind >= 0);
 313
 314     return -1;
 315 }
 316
 317 #ifdef __cplusplus
 318 }
 319 #endif
 320
 321 #endif /* _gmx_hash_h */