liblttng-ust/jhash.h

   1 /*
   2  * Copyright (C) 2011 Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
   3  *
   4  * This library is free software; you can redistribute it and/or
   5  * modify it under the terms of the GNU Lesser General Public
   6  * License as published by the Free Software Foundation;
   7  * version 2.1 of the License.
   8  *
   9  * This library is distributed in the hope that it will be useful,
  10  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  11  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  12  * Lesser General Public License for more details.
  13  *
  14  * You should have received a copy of the GNU Lesser General Public
  15  * License along with this library; if not, write to the Free Software
  16  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301 USA
  17  */
  18
  19 #include <urcu/compiler.h>
  20 #include <lttng/ust-endian.h>
  21
  22 /*
  23  * Hash function
  24  * Source: http://burtleburtle.net/bob/c/lookup3.c
  25  * Originally Public Domain
  26  */
  27
  28 #define rot(x, k) (((x) << (k)) | ((x) >> (32 - (k))))
  29
  30 #define mix(a, b, c) \
  31 do { \
  32         a -= c; a ^= rot(c,  4); c += b; \
  33         b -= a; b ^= rot(a,  6); a += c; \
  34         c -= b; c ^= rot(b,  8); b += a; \
  35         a -= c; a ^= rot(c, 16); c += b; \
  36         b -= a; b ^= rot(a, 19); a += c; \
  37         c -= b; c ^= rot(b,  4); b += a; \
  38 } while (0)
  39
  40 #define final(a, b, c) \
  41 { \
  42         c ^= b; c -= rot(b, 14); \
  43         a ^= c; a -= rot(c, 11); \
  44         b ^= a; b -= rot(a, 25); \
  45         c ^= b; c -= rot(b, 16); \
  46         a ^= c; a -= rot(c,  4);\
  47         b ^= a; b -= rot(a, 14); \
  48         c ^= b; c -= rot(b, 24); \
  49 }
  50
  51 #if (BYTE_ORDER == LITTLE_ENDIAN)
  52 #define HASH_LITTLE_ENDIAN      1
  53 #else
  54 #define HASH_LITTLE_ENDIAN      0
  55 #endif
  56
  57 /*
  58  *
  59  * hashlittle() -- hash a variable-length key into a 32-bit value
  60  *   k       : the key (the unaligned variable-length array of bytes)
  61  *   length  : the length of the key, counting by bytes
  62  *   initval : can be any 4-byte value
  63  * Returns a 32-bit value.  Every bit of the key affects every bit of
  64  * the return value.  Two keys differing by one or two bits will have
  65  * totally different hash values.
  66  *
  67  * The best hash table sizes are powers of 2.  There is no need to do
  68  * mod a prime (mod is sooo slow!).  If you need less than 32 bits,
  69  * use a bitmask.  For example, if you need only 10 bits, do
  70  *   h = (h & hashmask(10));
  71  * In which case, the hash table should have hashsize(10) elements.
  72  *
  73  * If you are hashing n strings (uint8_t **)k, do it like this:
  74  *   for (i = 0, h = 0; i < n; ++i) h = hashlittle(k[i], len[i], h);
  75  *
  76  * By Bob Jenkins, 2006.  bob_jenkins@burtleburtle.net.  You may use this
  77  * code any way you wish, private, educational, or commercial.  It's free.
  78  *
  79  * Use for hash table lookup, or anything where one collision in 2^^32 is
  80  * acceptable.  Do NOT use for cryptographic purposes.
  81  */
  82 static
  83 uint32_t hashlittle(const void *key, size_t length, uint32_t initval)
  84 {
  85         uint32_t a, b, c;       /* internal state */
  86         union {
  87                 const void *ptr;
  88                 size_t i;
  89         } u;
  90
  91         /* Set up the internal state */
  92         a = b = c = 0xdeadbeef + ((uint32_t)length) + initval;
  93
  94         u.ptr = key;
  95         if (HASH_LITTLE_ENDIAN && ((u.i & 0x3) == 0)) {
  96                 const uint32_t *k = (const uint32_t *) key;     /* read 32-bit chunks */
  97
  98                 /*------ all but last block: aligned reads and affect 32 bits of (a,b,c) */
  99                 while (length > 12) {
 100                         a += k[0];
 101                         b += k[1];
 102                         c += k[2];
 103                         mix(a, b, c);
 104                         length -= 12;
 105                         k += 3;
 106                 }
 107
 108                 /*----------------------------- handle the last (probably partial) block */
 109                 /*
 110                  * The original jhash.h reads beyond the end of string, and implements
 111                  * a special code path for VALGRIND. It seems to make ASan unhappy too
 112                  * though, so considering that hashing event names is not a fast-path
 113                  * in lttng-ust, remove the "fast" code entirely and use the slower
 114                  * but verifiable VALGRIND version of the code which does not issue
 115                  * out-of-bound reads.
 116                  */
 117                 {
 118                         const uint8_t *k8;
 119
 120                         k8 = (const uint8_t *) k;
 121                         switch (length) {
 122                         case 12: c+=k[2]; b+=k[1]; a+=k[0]; break;
 123                         case 11: c+=((uint32_t) k8[10])<<16;    /* fall through */
 124                         case 10: c+=((uint32_t) k8[9])<<8;      /* fall through */
 125                         case 9 : c+=k8[8];                      /* fall through */
 126                         case 8 : b+=k[1]; a+=k[0]; break;
 127                         case 7 : b+=((uint32_t) k8[6])<<16;     /* fall through */
 128                         case 6 : b+=((uint32_t) k8[5])<<8;      /* fall through */
 129                         case 5 : b+=k8[4];                      /* fall through */
 130                         case 4 : a+=k[0]; break;
 131                         case 3 : a+=((uint32_t) k8[2])<<16;     /* fall through */
 132                         case 2 : a+=((uint32_t) k8[1])<<8;      /* fall through */
 133                         case 1 : a+=k8[0]; break;
 134                         case 0 : return c;
 135                         }
 136                 }
 137
 138         } else if (HASH_LITTLE_ENDIAN && ((u.i & 0x1) == 0)) {
 139                 const uint16_t *k = (const uint16_t *) key;     /* read 16-bit chunks */
 140                 const uint8_t *k8;
 141
 142                 /*--------------- all but last block: aligned reads and different mixing */
 143                 while (length > 12)
 144                 {
 145                         a += k[0] + (((uint32_t) k[1])<<16);
 146                         b += k[2] + (((uint32_t) k[3])<<16);
 147                         c += k[4] + (((uint32_t) k[5])<<16);
 148                         mix(a, b, c);
 149                         length -= 12;
 150                         k += 6;
 151                 }
 152
 153                 /*----------------------------- handle the last (probably partial) block */
 154                 k8 = (const uint8_t *) k;
 155                 switch(length)
 156                 {
 157                 case 12: c+=k[4]+(((uint32_t) k[5])<<16);
 158                          b+=k[2]+(((uint32_t) k[3])<<16);
 159                          a+=k[0]+(((uint32_t) k[1])<<16);
 160                          break;
 161                 case 11: c+=((uint32_t) k8[10])<<16;    /* fall through */
 162                 case 10: c+=k[4];
 163                          b+=k[2]+(((uint32_t) k[3])<<16);
 164                          a+=k[0]+(((uint32_t) k[1])<<16);
 165                          break;
 166                 case 9 : c+=k8[8];                      /* fall through */
 167                 case 8 : b+=k[2]+(((uint32_t) k[3])<<16);
 168                          a+=k[0]+(((uint32_t) k[1])<<16);
 169                          break;
 170                 case 7 : b+=((uint32_t) k8[6])<<16;     /* fall through */
 171                 case 6 : b+=k[2];
 172                          a+=k[0]+(((uint32_t) k[1])<<16);
 173                          break;
 174                 case 5 : b+=k8[4];                      /* fall through */
 175                 case 4 : a+=k[0]+(((uint32_t) k[1])<<16);
 176                          break;
 177                 case 3 : a+=((uint32_t) k8[2])<<16;     /* fall through */
 178                 case 2 : a+=k[0];
 179                          break;
 180                 case 1 : a+=k8[0];
 181                          break;
 182                 case 0 : return c;                      /* zero length requires no mixing */
 183                 }
 184
 185         } else {                                        /* need to read the key one byte at a time */
 186                 const uint8_t *k = (const uint8_t *)key;
 187
 188                 /*--------------- all but the last block: affect some 32 bits of (a, b, c) */
 189                 while (length > 12) {
 190                         a += k[0];
 191                         a += ((uint32_t) k[1])<<8;
 192                         a += ((uint32_t) k[2])<<16;
 193                         a += ((uint32_t) k[3])<<24;
 194                         b += k[4];
 195                         b += ((uint32_t) k[5])<<8;
 196                         b += ((uint32_t) k[6])<<16;
 197                         b += ((uint32_t) k[7])<<24;
 198                         c += k[8];
 199                         c += ((uint32_t) k[9])<<8;
 200                         c += ((uint32_t) k[10])<<16;
 201                         c += ((uint32_t) k[11])<<24;
 202                         mix(a,b,c);
 203                         length -= 12;
 204                         k += 12;
 205                 }
 206
 207                 /*-------------------------------- last block: affect all 32 bits of (c) */
 208                 switch (length) {                /* all the case statements fall through */
 209                 case 12: c+=((uint32_t) k[11])<<24;
 210                 case 11: c+=((uint32_t) k[10])<<16;
 211                 case 10: c+=((uint32_t) k[9])<<8;
 212                 case 9 : c+=k[8];
 213                 case 8 : b+=((uint32_t) k[7])<<24;
 214                 case 7 : b+=((uint32_t) k[6])<<16;
 215                 case 6 : b+=((uint32_t) k[5])<<8;
 216                 case 5 : b+=k[4];
 217                 case 4 : a+=((uint32_t) k[3])<<24;
 218                 case 3 : a+=((uint32_t) k[2])<<16;
 219                 case 2 : a+=((uint32_t) k[1])<<8;
 220                 case 1 : a+=k[0];
 221                          break;
 222                 case 0 : return c;
 223                 }
 224         }
 225
 226         final(a, b, c);
 227         return c;
 228 }
 229
 230 static inline
 231 uint32_t jhash(const void *key, size_t length, uint32_t seed)
 232 {
 233         return hashlittle(key, length, seed);
 234 }