lj_str.diff

diff --git a/src/lj_str.c b/src/lj_str.c
index 6548ee4..128211f 100644
--- a/src/lj_str.c
+++ b/src/lj_str.c
@@ -90,6 +90,41 @@ void lj_str_resize(lua_State *L, MSize newmask)
   g->strhash = newhash;
 }

+static LJ_AINLINE MSize murmur3_h ( const void * key, MSize len, uintptr_t seed)
+{
+  const uint8_t * data = (const uint8_t*)key;
+  const uint8_t * tail = (const uint8_t*)data+(len&(~3));
+  const MSize nblocks = len / 4;
+  MSize i;
+  uint32_t h1 = (uint32_t)seed & (uint32_t)(seed >> 16);
+  const uint32_t c1 = 0xcc9e2d51;
+  const uint32_t c2 = 0x1b873593;
+  uint32_t k1;
+  for(i = 0; i < nblocks; i++)
+  {
+    k1 = lj_getu32(data + i*4);
+    k1 *= c1; k1 = lj_rol(k1,15); k1 *= c2;
+    h1 ^= k1; h1 = lj_rol(h1,13); h1 = h1*5+0xe6546b64;
+  }
+
+  k1 = 0;
+  switch(len & 3)
+  {
+  case 3: k1 ^= tail[2] << 16;
+  case 2: k1 ^= tail[1] << 8;
+  case 1: k1 ^= tail[0];
+      k1 *= c1; k1 = lj_rol(k1,15); k1 *= c2;
+      h1 ^= k1;
+  };
+
+
+  h1 ^= len;
+  h1 ^= h1 >> 16; h1 *= 0x85ebca6b;
+  h1 ^= h1 >> 13; h1 *= 0xc2b2ae35;
+  h1 ^= h1 >> 16;
+  return h1;
+}
+
 /* Intern a string and return string object. */
 GCstr *lj_str_new(lua_State *L, const char *str, size_t lenx)
 {
@@ -97,28 +132,16 @@ GCstr *lj_str_new(lua_State *L, const char *str, size_t lenx)
   GCstr *s;
   GCobj *o;
   MSize len = (MSize)lenx;
-  MSize a, b, h = len;
+  MSize h;
   if (lenx >= LJ_MAX_STR)
     lj_err_msg(L, LJ_ERR_STROV);
   g = G(L);
   /* Compute string hash. Constants taken from lookup3 hash by Bob Jenkins. */
-  if (len >= 4) {  /* Caveat: unaligned access! */
-    a = lj_getu32(str);
-    h ^= lj_getu32(str+len-4);
-    b = lj_getu32(str+(len>>1)-2);
-    h ^= b; h -= lj_rol(b, 14);
-    b += lj_getu32(str+(len>>2)-1);
-  } else if (len > 0) {
-    a = *(const uint8_t *)str;
-    h ^= *(const uint8_t *)(str+len-1);
-    b = *(const uint8_t *)(str+(len>>1));
-    h ^= b; h -= lj_rol(b, 14);
+  if (len > 0) {
+    h = murmur3_h(str, len, (uintptr_t)g);
   } else {
     return &g->strempty;
   }
-  a ^= h; a -= lj_rol(h, 11);
-  b ^= a; b -= lj_rol(a, 25);
-  h ^= b; h -= lj_rol(b, 16);
   /* Check if the string has already been interned. */
   o = gcref(g->strhash[h & g->strmask]);
   if (LJ_LIKELY((((uintptr_t)str+len-1) & (LJ_PAGESIZE-1)) <= LJ_PAGESIZE-4)) {