| // Copyright 2010 Google Inc. All Rights Reserved. |
| // Authors: gpike@google.com (Geoff Pike), jyrki@google.com (Jyrki Alakuijala) |
| // |
| // This file provides CityHash64() and related functions. |
| // |
| // The externally visible functions follow the naming conventions of |
| // hash.h, where the size of the output is part of the name. For |
| // example, CityHash64 returns a 64-bit hash. The internal helpers do |
| // not have the return type in their name, but instead have names like |
| // HashLenXX or HashLenXXtoYY, where XX and YY are input string lengths. |
| // |
| // Most of the constants and tricks here were copied from murmur.cc or |
| // hash.h, or discovered by trial and error. It's probably possible to further |
| // optimize the code here by writing a program that systematically explores |
| // more of the space of possible hash functions, or by using SIMD instructions. |
| |
| #include "kudu/gutil/hash/city.h" |
| |
| #include <sys/types.h> |
| |
| #include <algorithm> |
| #include <utility> |
| #include <vector> |
| |
| #include <glog/logging.h> |
| |
| #include "kudu/gutil/endian.h" |
| #include "kudu/gutil/hash/hash128to64.h" |
| #include "kudu/gutil/int128.h" |
| #include "kudu/gutil/integral_types.h" |
| #include "kudu/gutil/port.h" |
| |
| using std::copy; |
| using std::make_pair; |
| using std::max; |
| using std::min; |
| using std::pair; |
| using std::reverse; |
| using std::sort; |
| using std::swap; |
| |
| namespace util_hash { |
| |
| // Some primes between 2^63 and 2^64 for various uses. |
| static const uint64 k0 = 0xa5b85c5e198ed849ULL; |
| static const uint64 k1 = 0x8d58ac26afe12e47ULL; |
| static const uint64 k2 = 0xc47b6e9e3a970ed3ULL; |
| static const uint64 k3 = 0xc70f6907e782aa0bULL; |
| |
| // Bitwise right rotate. Normally this will compile to a single |
| // instruction, especially if the shift is a manifest constant. |
| static uint64 Rotate(uint64 val, int shift) { |
| DCHECK_GE(shift, 0); |
| DCHECK_LE(shift, 63); |
| // Avoid shifting by 64: doing so yields an undefined result. |
| return shift == 0 ? val : ((val >> shift) | (val << (64 - shift))); |
| } |
| |
| // Equivalent to Rotate(), but requires the second arg to be non-zero. |
| // On x86-64, and probably others, it's possible for this to compile |
| // to a single instruction if both args are already in registers. |
| static uint64 RotateByAtLeast1(uint64 val, int shift) { |
| DCHECK_GE(shift, 1); |
| DCHECK_LE(shift, 63); |
| return (val >> shift) | (val << (64 - shift)); |
| } |
| |
| static uint64 ShiftMix(uint64 val) { |
| return val ^ (val >> 47); |
| } |
| |
| static uint64 HashLen16(uint64 u, uint64 v) { |
| return Hash128to64(uint128(u, v)); |
| } |
| |
| ATTRIBUTE_NO_SANITIZE_INTEGER |
| static uint64 HashLen0to16(const char *s, size_t len) { |
| DCHECK_GE(len, 0); |
| DCHECK_LE(len, 16); |
| if (len > 8) { |
| uint64 a = LittleEndian::Load64(s); |
| uint64 b = LittleEndian::Load64(s + len - 8); |
| return HashLen16(a, RotateByAtLeast1(b + len, len)) ^ b; |
| } |
| if (len >= 4) { |
| uint64 a = LittleEndian::Load32(s); |
| return HashLen16(len + (a << 3), LittleEndian::Load32(s + len - 4)); |
| } |
| if (len > 0) { |
| uint8 a = s[0]; |
| uint8 b = s[len >> 1]; |
| uint8 c = s[len - 1]; |
| uint32 y = static_cast<uint32>(a) + (static_cast<uint32>(b) << 8); |
| uint32 z = len + (static_cast<uint32>(c) << 2); |
| return ShiftMix(y * k2 ^ z * k3) * k2; |
| } |
| return k2; |
| } |
| |
| // This probably works well for 16-byte strings as well, but it may be overkill |
| // in that case. |
| ATTRIBUTE_NO_SANITIZE_INTEGER |
| static uint64 HashLen17to32(const char *s, size_t len) { |
| DCHECK_GE(len, 17); |
| DCHECK_LE(len, 32); |
| uint64 a = LittleEndian::Load64(s) * k1; |
| uint64 b = LittleEndian::Load64(s + 8); |
| uint64 c = LittleEndian::Load64(s + len - 8) * k2; |
| uint64 d = LittleEndian::Load64(s + len - 16) * k0; |
| return HashLen16(Rotate(a - b, 43) + Rotate(c, 30) + d, |
| a + Rotate(b ^ k3, 20) - c + len); |
| } |
| |
| // Return a 16-byte hash for 48 bytes. Quick and dirty. |
| // Callers do best to use "random-looking" values for a and b. |
| // (For more, see the code review discussion of CL 18799087.) |
| ATTRIBUTE_NO_SANITIZE_INTEGER |
| static pair<uint64, uint64> WeakHashLen32WithSeeds( |
| uint64 w, uint64 x, uint64 y, uint64 z, uint64 a, uint64 b) { |
| a += w; |
| b = Rotate(b + a + z, 51); |
| uint64 c = a; |
| a += x; |
| a += y; |
| b += Rotate(a, 23); |
| return make_pair(a + z, b + c); |
| } |
| |
| // Return a 16-byte hash for s[0] ... s[31], a, and b. Quick and dirty. |
| static pair<uint64, uint64> WeakHashLen32WithSeeds( |
| const char* s, uint64 a, uint64 b) { |
| return WeakHashLen32WithSeeds(LittleEndian::Load64(s), |
| LittleEndian::Load64(s + 8), |
| LittleEndian::Load64(s + 16), |
| LittleEndian::Load64(s + 24), |
| a, |
| b); |
| } |
| |
| // Return an 8-byte hash for 33 to 64 bytes. |
| ATTRIBUTE_NO_SANITIZE_INTEGER |
| static uint64 HashLen33to64(const char *s, size_t len) { |
| uint64 z = LittleEndian::Load64(s + 24); |
| uint64 a = LittleEndian::Load64(s) + |
| (len + LittleEndian::Load64(s + len - 16)) * k0; |
| uint64 b = Rotate(a + z, 52); |
| uint64 c = Rotate(a, 37); |
| a += LittleEndian::Load64(s + 8); |
| c += Rotate(a, 7); |
| a += LittleEndian::Load64(s + 16); |
| uint64 vf = a + z; |
| uint64 vs = b + Rotate(a, 31) + c; |
| a = LittleEndian::Load64(s + 16) + LittleEndian::Load64(s + len - 32); |
| z += LittleEndian::Load64(s + len - 8); |
| b = Rotate(a + z, 52); |
| c = Rotate(a, 37); |
| a += LittleEndian::Load64(s + len - 24); |
| c += Rotate(a, 7); |
| a += LittleEndian::Load64(s + len - 16); |
| uint64 wf = a + z; |
| uint64 ws = b + Rotate(a, 31) + c; |
| uint64 r = ShiftMix((vf + ws) * k2 + (wf + vs) * k0); |
| return ShiftMix(r * k0 + vs) * k2; |
| } |
| |
| ATTRIBUTE_NO_SANITIZE_INTEGER |
| uint64 CityHash64(const char *s, size_t len) { |
| if (len <= 32) { |
| if (len <= 16) { |
| return HashLen0to16(s, len); |
| } else { |
| return HashLen17to32(s, len); |
| } |
| } else if (len <= 64) { |
| return HashLen33to64(s, len); |
| } |
| |
| // For strings over 64 bytes we hash the end first, and then as we |
| // loop we keep 56 bytes of state: v, w, x, y, and z. |
| uint64 x = LittleEndian::Load64(s + len - 40); |
| uint64 y = LittleEndian::Load64(s + len - 16) + |
| LittleEndian::Load64(s + len - 56); |
| uint64 z = HashLen16(LittleEndian::Load64(s + len - 48) + len, |
| LittleEndian::Load64(s + len - 24)); |
| pair<uint64, uint64> v = WeakHashLen32WithSeeds(s + len - 64, len, z); |
| pair<uint64, uint64> w = WeakHashLen32WithSeeds(s + len - 32, y + k1, x); |
| x = x * k1 + LittleEndian::Load64(s); |
| |
| // Decrease len to the nearest multiple of 64, and operate on 64-byte chunks. |
| len = (len - 1) & ~static_cast<size_t>(63); |
| DCHECK_GT(len, 0); |
| DCHECK_EQ(len, len / 64 * 64); |
| do { |
| x = Rotate(x + y + v.first + LittleEndian::Load64(s + 8), 37) * k1; |
| y = Rotate(y + v.second + LittleEndian::Load64(s + 48), 42) * k1; |
| x ^= w.second; |
| y += v.first + LittleEndian::Load64(s + 40); |
| z = Rotate(z + w.first, 33) * k1; |
| v = WeakHashLen32WithSeeds(s, v.second * k1, x + w.first); |
| w = WeakHashLen32WithSeeds(s + 32, z + w.second, |
| y + LittleEndian::Load64(s + 16)); |
| std::swap(z, x); |
| s += 64; |
| len -= 64; |
| } while (len != 0); |
| return HashLen16(HashLen16(v.first, w.first) + ShiftMix(y) * k1 + z, |
| HashLen16(v.second, w.second) + x); |
| } |
| |
| uint64 CityHash64WithSeed(const char *s, size_t len, uint64 seed) { |
| return CityHash64WithSeeds(s, len, k2, seed); |
| } |
| |
| uint64 CityHash64WithSeeds(const char *s, size_t len, |
| uint64 seed0, uint64 seed1) { |
| return HashLen16(CityHash64(s, len) - seed0, seed1); |
| } |
| |
| // A subroutine for CityHash128(). Returns a decent 128-bit hash for strings |
| // of any length representable in ssize_t. Based on City and Murmur128. |
| static uint128 CityMurmur(const char *s, size_t len, uint128 seed) { |
| uint64 a = Uint128Low64(seed); |
| uint64 b = Uint128High64(seed); |
| uint64 c = 0; |
| uint64 d = 0; |
| ssize_t l = len - 16; |
| if (l <= 0) { // len <= 16 |
| c = b * k1 + HashLen0to16(s, len); |
| d = Rotate(a + (len >= 8 ? LittleEndian::Load64(s) : c), 32); |
| } else { // len > 16 |
| c = HashLen16(LittleEndian::Load64(s + len - 8) + k1, a); |
| d = HashLen16(b + len, c + LittleEndian::Load64(s + len - 16)); |
| a += d; |
| do { |
| a ^= ShiftMix(LittleEndian::Load64(s) * k1) * k1; |
| a *= k1; |
| b ^= a; |
| c ^= ShiftMix(LittleEndian::Load64(s + 8) * k1) * k1; |
| c *= k1; |
| d ^= c; |
| s += 16; |
| l -= 16; |
| } while (l > 0); |
| } |
| a = HashLen16(a, c); |
| b = HashLen16(d, b); |
| return uint128(a ^ b, HashLen16(b, a)); |
| } |
| |
| uint128 CityHash128WithSeed(const char *s, size_t len, uint128 seed) { |
| // TODO(user): As of February 2011, there's a beta of Murmur3 that would |
| // most likely be useful here. E.g., if (len < 900) return Murmur3(...) |
| if (len < 128) { |
| return CityMurmur(s, len, seed); |
| } |
| |
| // We expect len >= 128 to be the common case. Keep 56 bytes of state: |
| // v, w, x, y, and z. |
| pair<uint64, uint64> v, w; |
| uint64 x = Uint128Low64(seed); |
| uint64 y = Uint128High64(seed); |
| uint64 z = len * k1; |
| v.first = Rotate(y ^ k1, 49) * k1 + LittleEndian::Load64(s); |
| v.second = Rotate(v.first, 42) * k1 + LittleEndian::Load64(s + 8); |
| w.first = Rotate(y + z, 35) * k1 + x; |
| w.second = Rotate(x + LittleEndian::Load64(s + 88), 53) * k1; |
| |
| // This is similar to the inner loop of CityHash64(), manually unrolled. |
| do { |
| x = Rotate(x + y + v.first + LittleEndian::Load64(s + 16), 37) * k1; |
| y = Rotate(y + v.second + LittleEndian::Load64(s + 48), 42) * k1; |
| x ^= w.second; |
| y ^= v.first; |
| z = Rotate(z ^ w.first, 33); |
| v = WeakHashLen32WithSeeds(s, v.second * k1, x + w.first); |
| w = WeakHashLen32WithSeeds(s + 32, z + w.second, y); |
| std::swap(z, x); |
| s += 64; |
| x = Rotate(x + y + v.first + LittleEndian::Load64(s + 16), 37) * k1; |
| y = Rotate(y + v.second + LittleEndian::Load64(s + 48), 42) * k1; |
| x ^= w.second; |
| y ^= v.first; |
| z = Rotate(z ^ w.first, 33); |
| v = WeakHashLen32WithSeeds(s, v.second * k1, x + w.first); |
| w = WeakHashLen32WithSeeds(s + 32, z + w.second, y); |
| std::swap(z, x); |
| s += 64; |
| len -= 128; |
| } while (PREDICT_TRUE(len >= 128)); |
| y += Rotate(w.first, 37) * k0 + z; |
| x += Rotate(v.first + z, 49) * k0; |
| // If 0 < len < 128, hash up to 4 chunks of 32 bytes each from the end of s. |
| for (size_t tail_done = 0; tail_done < len; ) { |
| tail_done += 32; |
| y = Rotate(y - x, 42) * k0 + v.second; |
| w.first += LittleEndian::Load64(s + len - tail_done + 16); |
| x = Rotate(x, 49) * k0 + w.first; |
| w.first += v.first; |
| v = WeakHashLen32WithSeeds(s + len - tail_done, v.first, v.second); |
| } |
| // At this point our 48 bytes of state should contain more than |
| // enough information for a strong 128-bit hash. We use two |
| // different 48-byte-to-8-byte hashes to get a 16-byte final result. |
| x = HashLen16(x, v.first); |
| y = HashLen16(y, w.first); |
| return uint128(HashLen16(x + v.second, w.second) + y, |
| HashLen16(x + w.second, y + v.second)); |
| } |
| |
| uint128 CityHash128(const char *s, size_t len) { |
| if (len >= 16) { |
| return CityHash128WithSeed(s + 16, |
| len - 16, |
| uint128(LittleEndian::Load64(s) ^ k3, |
| LittleEndian::Load64(s + 8))); |
| } else if (len >= 8) { |
| return CityHash128WithSeed(nullptr, |
| 0, |
| uint128(LittleEndian::Load64(s) ^ (len * k0), |
| LittleEndian::Load64(s + len - 8) ^ k1)); |
| } else { |
| return CityHash128WithSeed(s, len, uint128(k0, k1)); |
| } |
| } |
| |
| } // namespace util_hash |