* string.c (hash): use Bob Jenkins' hash algorithm.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@11255 b2dd03c8-39d4-4d8f-98ff-823fe69b080e

1 files changed, 114 insertions, 88 deletions

diff --git a/string.c b/string.c
index 529c375b4f..79fa217871 100644
--- a/string.c
+++ b/string.c
@@ -879,102 +879,128 @@ rb_str_concat(VALUE str1, VALUE str2)
     return rb_str_append(str1, str2);
 }
 
-/*
- * hash_32 - 32 bit Fowler/Noll/Vo FNV-1a hash code
- *
- * @(#) $hash_32.Revision: 1.1 $
- * @(#) $hash_32.Id: hash_32a.c,v 1.1 2003/10/03 20:38:53 chongo Exp $
- * @(#) $hash_32.Source: /usr/local/src/cmd/fnv/RCS/hash_32a.c,v $
- *
- ***
- *
- * Fowler/Noll/Vo hash
- *
- * The basis of this hash algorithm was taken from an idea sent
- * as reviewer comments to the IEEE POSIX P1003.2 committee by:
- *
- *      Phong Vo (http://www.research.att.com/info/kpv/)
- *      Glenn Fowler (http://www.research.att.com/~gsf/)
- *
- * In a subsequent ballot round:
- *
- *      Landon Curt Noll (http://www.isthe.com/chongo/)
- *
- * improved on their algorithm.  Some people tried this hash
- * and found that it worked rather well.  In an EMail message
- * to Landon, they named it the ``Fowler/Noll/Vo'' or FNV hash.
- *
- * FNV hashes are designed to be fast while maintaining a low
- * collision rate. The FNV speed allows one to quickly hash lots
- * of data while maintaining a reasonable collision rate.  See:
- *
- *      http://www.isthe.com/chongo/tech/comp/fnv/index.html
- *
- * for more details as well as other forms of the FNV hash.
- ***
- *
- * To use the recommended 32 bit FNV-1a hash, pass FNV1_32A_INIT as the
- * Fnv32_t hashval argument to fnv_32a_buf() or fnv_32a_str().
- *
- ***
- *
- * Please do not copyright this code.  This code is in the public domain.
- *
- * LANDON CURT NOLL DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS SOFTWARE,
- * INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS. IN NO
- * EVENT SHALL LANDON CURT NOLL BE LIABLE FOR ANY SPECIAL, INDIRECT OR
- * CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF
- * USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR
- * OTHER TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR
- * PERFORMANCE OF THIS SOFTWARE.
- *
- * By:
- *	chongo <Landon Curt Noll> /\oo/\
- *      http://www.isthe.com/chongo/
- *
- * Share and Enjoy!	:-)
- */
+typedef  unsigned int  ub4;   /* unsigned 4-byte quantities */
+typedef  unsigned char ub1;   /* unsigned 1-byte quantities */
+
+#define hashsize(n) ((ub4)1<<(n))
+#define hashmask(n) (hashsize(n)-1)
 
 /*
- * 32 bit FNV-1 and FNV-1a non-zero initial basis
- *
- * The FNV-1 initial basis is the FNV-0 hash of the following 32 octets:
- *
- *              chongo <Landon Curt Noll> /\../\
- *
- * NOTE: The \'s above are not back-slashing escape characters.
- * They are literal ASCII  backslash 0x5c characters.
- *
- * NOTE: The FNV-1a initial basis is the same value as FNV-1 by definition.
- */
-#define FNV1_32A_INIT 0x811c9dc5
+--------------------------------------------------------------------
+mix -- mix 3 32-bit values reversibly.
+For every delta with one or two bits set, and the deltas of all three
+  high bits or all three low bits, whether the original value of a,b,c
+  is almost all zero or is uniformly distributed,
+* If mix() is run forward or backward, at least 32 bits in a,b,c
+  have at least 1/4 probability of changing.
+* If mix() is run forward, every bit of c will change between 1/3 and
+  2/3 of the time.  (Well, 22/100 and 78/100 for some 2-bit deltas.)
+mix() was built out of 36 single-cycle latency instructions in a 
+  structure that could supported 2x parallelism, like so:
+      a -= b; 
+      a -= c; x = (c>>13);
+      b -= c; a ^= x;
+      b -= a; x = (a<<8);
+      c -= a; b ^= x;
+      c -= b; x = (b>>13);
+      ...
+  Unfortunately, superscalar Pentiums and Sparcs can't take advantage 
+  of that parallelism.  They've also turned some of those single-cycle
+  latency instructions into multi-cycle latency instructions.  Still,
+  this is the fastest good hash I could find.  There were about 2^^68
+  to choose from.  I only looked at a billion or so.
+--------------------------------------------------------------------
+*/
+#define mix(a,b,c) \
+{ \
+  a -= b; a -= c; a ^= (c>>13); \
+  b -= c; b -= a; b ^= (a<<8); \
+  c -= a; c -= b; c ^= (b>>13); \
+  a -= b; a -= c; a ^= (c>>12);  \
+  b -= c; b -= a; b ^= (a<<16); \
+  c -= a; c -= b; c ^= (b>>5); \
+  a -= b; a -= c; a ^= (c>>3);  \
+  b -= c; b -= a; b ^= (a<<10); \
+  c -= a; c -= b; c ^= (b>>15); \
+}
 
 /*
- * 32 bit magic FNV-1a prime
- */
-#define FNV_32_PRIME 0x01000193
+--------------------------------------------------------------------
+hash() -- hash a variable-length key into a 32-bit value
+  k       : the key (the unaligned variable-length array of bytes)
+  len     : the length of the key, counting by bytes
+  initval : can be any 4-byte value
+Returns a 32-bit value.  Every bit of the key affects every bit of
+the return value.  Every 1-bit and 2-bit delta achieves avalanche.
+About 6*len+35 instructions.
+
+The best hash table sizes are powers of 2.  There is no need to do
+mod a prime (mod is sooo slow!).  If you need less than 32 bits,
+use a bitmask.  For example, if you need only 10 bits, do
+  h = (h & hashmask(10));
+In which case, the hash table should have hashsize(10) elements.
+
+If you are hashing n strings (ub1 **)k, do it like this:
+  for (i=0, h=0; i<n; ++i) h = hash( k[i], len[i], h);
+
+By Bob Jenkins, 1996.  bob_jenkins@burtleburtle.net.  You may use this
+code any way you wish, private, educational, or commercial.  It's free.
+
+See http://burtleburtle.net/bob/hash/evahash.html
+Use for hash table lookup, or anything where one collision in 2^^32 is
+acceptable.  Do NOT use for cryptographic purposes.
+--------------------------------------------------------------------
+*/
+
+static ub4
+hash(const ub1 *k, ub4 length, ub4 initval)
+    /* k: the key */
+    /* length: the length of the key */
+    /* initval: the previous hash, or an arbitrary value */
+{
+    register ub4 a,b,c,len;
+
+    /* Set up the internal state */
+    len = length;
+    a = b = 0x9e3779b9;  /* the golden ratio; an arbitrary value */
+    c = initval;         /* the previous hash value */
+
+   /*---------------------------------------- handle most of the key */
+    while (len >= 12) {
+	a += (k[0] +((ub4)k[1]<<8) +((ub4)k[2]<<16) +((ub4)k[3]<<24));
+	b += (k[4] +((ub4)k[5]<<8) +((ub4)k[6]<<16) +((ub4)k[7]<<24));
+	c += (k[8] +((ub4)k[9]<<8) +((ub4)k[10]<<16)+((ub4)k[11]<<24));
+	mix(a,b,c);
+	k += 12; len -= 12;
+    }
+
+    /*------------------------------------- handle the last 11 bytes */
+    c += length;
+    switch(len)              /* all the case statements fall through */
+    {
+      case 11: c+=((ub4)k[10]<<24);
+      case 10: c+=((ub4)k[9]<<16);
+      case 9 : c+=((ub4)k[8]<<8);
+	/* the first byte of c is reserved for the length */
+      case 8 : b+=((ub4)k[7]<<24);
+      case 7 : b+=((ub4)k[6]<<16);
+      case 6 : b+=((ub4)k[5]<<8);
+      case 5 : b+=k[4];
+      case 4 : a+=((ub4)k[3]<<24);
+      case 3 : a+=((ub4)k[2]<<16);
+      case 2 : a+=((ub4)k[1]<<8);
+      case 1 : a+=k[0];
+	/* case 0: nothing left to add */
+    }
+    mix(a,b,c);
+    /*-------------------------------------------- report the result */
+    return c;
+}
 
 int
 rb_memhash(const void *ptr, long len)
 {
-    register const unsigned char *p = ptr;
-    register unsigned int hval = FNV1_32A_INIT;
-
-    /*
-     * FNV-1a hash each octet in the buffer
-     */
-    while (len--) {
-	/* xor the bottom with the current octet */
-	hval ^= (unsigned int)*p++;
-
-	/* multiply by the 32 bit FNV magic prime mod 2^32 */
-#if defined(FNV_GCC_OPTIMIZATION)
-	hval += (hval<<1) + (hval<<4) + (hval<<7) + (hval<<8) + (hval<<24);
-#else
-	hval *= FNV_32_PRIME;
-#endif
-    }
-    return hval;
+    return hash(ptr, len, 0);
 }
 
 int