rizin/librz/hash/algorithms/murmur3/murmur3.c

386 lines
7.3 KiB
C

// SPDX-FileCopyrightText: 2015 Andrey Jivsov <crypto@brainhub.org>
// SPDX-License-Identifier: MIT
/**
* \file MurmurHash3.cpp
* High performance non cryptographic hashing algorithm.
* Implementation of the MurmurHash3 algorithm.
* 3 variants of this algo are provided to hash 32,64 and 128 bit data
* MurmurHash3 was written by Austin Appleby, and is placed in the public domain.
* is engineered for speed and high-quality distribution,
* passing the full SMHasher test suite. This makes it a go-to choice for
* hash tables, bloom filters, and large-scale data de-duplication where
* performance is the primary constraint.
*
* Note - The x86 and x64 versions do _not_ produce the same results, as the
* algorithms are optimized for their respective platforms. You can still
* compile and run any of them on any platform, but your performance with the
* non-native version will be less than optimal.
*
*/
#include "murmur3.h"
#include "rz_types.h"
#include "rz_types_base.h"
#include <string.h>
static RZ_INLINE ut32 rotl32(ut32 x, int8_t r) {
return (x << r) | (x >> (32 - r));
}
static RZ_INLINE ut64 rotl64(ut64 x, int8_t r) {
return (x << r) | (x >> (64 - r));
}
/**
* \brief Block read - if your platform needs to do endian-swapping or can only
* handle aligned reads, do the conversion here
*/
#define getblock(p, i) \
(sizeof(*(p)) == 8 ? rz_read_le64(p + i) : rz_read_le32(p + i))
/*
* \brief Finalization mix - force all bits of a hash block to avalanche
*/
static RZ_INLINE ut32 fmix32(ut32 h) {
h ^= h >> 16;
h *= 0x85ebca6b;
h ^= h >> 13;
h *= 0xc2b2ae35;
h ^= h >> 16;
return h;
}
static RZ_INLINE ut64 fmix64(ut64 k) {
k ^= k >> 33;
k *= BIG_CONSTANT(0xff51afd7ed558ccd);
k ^= k >> 33;
k *= BIG_CONSTANT(0xc4ceb9fe1a85ec53);
k ^= k >> 33;
return k;
}
RZ_IPI void MurmurHash3_x86_32(const void *key, RzRef len, size_t seed, void *out) {
const ut8 *data = (const ut8 *)key;
const RzRef nblocks = len / 4;
RzRef i;
ut32 h1 = seed;
ut32 c1 = 0xcc9e2d51;
ut32 c2 = 0x1b873593;
const ut32 *blocks = (const ut32 *)(data + nblocks * 4);
for (i = -nblocks; i; i++) {
ut32 k1 = getblock(blocks, i);
k1 *= c1;
k1 = ROTL32_Murmur3(k1, 15);
k1 *= c2;
h1 ^= k1;
h1 = ROTL32_Murmur3(h1, 13);
h1 = h1 * 5 + 0xe6546b64;
}
const ut8 *tail = (const ut8 *)(data + nblocks * 4);
ut32 k1 = 0;
switch (len & 3) {
case 3:
k1 ^= tail[2] << 16;
/* fall through */
case 2:
k1 ^= tail[1] << 8;
/* fall through */
case 1:
k1 ^= tail[0];
k1 *= c1;
k1 = ROTL32_Murmur3(k1, 15);
k1 *= c2;
h1 ^= k1;
};
// finalization
h1 ^= len;
h1 = fmix32(h1);
*(ut32 *)out = h1;
}
RZ_IPI void MurmurHash3_x86_128(const void *key, const RzRef len, size_t seed, void *out) {
const ut8 *data = (const ut8 *)key;
const RzRef nblocks = len / 16;
RzRef i;
size_t h1 = seed;
size_t h2 = seed;
size_t h3 = seed;
size_t h4 = seed;
const size_t c1 = 0x239b961b;
const size_t c2 = 0xab0e9789;
const size_t c3 = 0x38b34ae5;
const size_t c4 = 0xa1e38b93;
const ut32 *blocks = (const ut32 *)(data + nblocks * 16);
for (i = -nblocks; i; i++) {
ut32 k1 = getblock(blocks, i * 4 + 0);
ut32 k2 = getblock(blocks, i * 4 + 1);
ut32 k3 = getblock(blocks, i * 4 + 2);
ut32 k4 = getblock(blocks, i * 4 + 3);
k1 *= c1;
k1 = ROTL32_Murmur3(k1, 15);
k1 *= c2;
h1 ^= k1;
h1 = ROTL32_Murmur3(h1, 19);
h1 += h2;
h1 = h1 * 5 + 0x561ccd1b;
k2 *= c2;
k2 = ROTL32_Murmur3(k2, 16);
k2 *= c3;
h2 ^= k2;
h2 = ROTL32_Murmur3(h2, 17);
h2 += h3;
h2 = h2 * 5 + 0x0bcaa747;
k3 *= c3;
k3 = ROTL32_Murmur3(k3, 17);
k3 *= c4;
h3 ^= k3;
h3 = ROTL32_Murmur3(h3, 15);
h3 += h4;
h3 = h3 * 5 + 0x96cd1c35;
k4 *= c4;
k4 = ROTL32_Murmur3(k4, 18);
k4 *= c1;
h4 ^= k4;
h4 = ROTL32_Murmur3(h4, 13);
h4 += h1;
h4 = h4 * 5 + 0x32ac3b17;
}
const ut8 *tail = (const ut8 *)(data + nblocks * 16);
ut32 k1 = 0;
ut32 k2 = 0;
ut32 k3 = 0;
ut32 k4 = 0;
switch (len & 15) {
case 15:
k4 ^= tail[14] << 16;
/* fall through */
case 14:
k4 ^= tail[13] << 8;
/* fall through */
case 13:
k4 ^= tail[12] << 0;
k4 *= c4;
k4 = ROTL32_Murmur3(k4, 18);
k4 *= c1;
h4 ^= k4;
/* fall through */
case 12:
k3 ^= tail[11] << 24;
/* fall through */
case 11:
k3 ^= tail[10] << 16;
/* fall through */
case 10:
k3 ^= tail[9] << 8;
/* fall through */
case 9:
k3 ^= tail[8] << 0;
k3 *= c3;
k3 = ROTL32_Murmur3(k3, 17);
k3 *= c4;
h3 ^= k3;
/* fall through */
case 8:
k2 ^= tail[7] << 24;
/* fall through */
case 7:
k2 ^= tail[6] << 16;
/* fall through */
case 6:
k2 ^= tail[5] << 8;
/* fall through */
case 5:
k2 ^= tail[4] << 0;
k2 *= c2;
k2 = ROTL32_Murmur3(k2, 16);
k2 *= c3;
h2 ^= k2;
/* fall through */
case 4:
k1 ^= tail[3] << 24;
/* fall through */
case 3:
k1 ^= tail[2] << 16;
/* fall through */
case 2:
k1 ^= tail[1] << 8;
/* fall through */
case 1:
k1 ^= tail[0] << 0;
k1 *= c1;
k1 = ROTL32_Murmur3(k1, 15);
k1 *= c2;
h1 ^= k1;
};
h1 ^= len;
h2 ^= len;
h3 ^= len;
h4 ^= len;
h1 += h2;
h1 += h3;
h1 += h4;
h2 += h1;
h3 += h1;
h4 += h1;
h1 = fmix32(h1);
h2 = fmix32(h2);
h3 = fmix32(h3);
h4 = fmix32(h4);
h1 += h2;
h1 += h3;
h1 += h4;
h2 += h1;
h3 += h1;
h4 += h1;
((ut32 *)out)[0] = h1;
((ut32 *)out)[1] = h2;
((ut32 *)out)[2] = h3;
((ut32 *)out)[3] = h4;
}
RZ_IPI void MurmurHash3_x64_128(const void *key, const RzRef len, const ut64 seed, void *out) {
const ut8 *data = (const ut8 *)key;
const RzRef nblocks = len / 16;
RzRef i;
ut64 h1 = seed;
ut64 h2 = seed;
const ut64 c1 = BIG_CONSTANT(0x87c37b91114253d5);
const ut64 c2 = BIG_CONSTANT(0x4cf5ad432745937f);
const ut64 *blocks = (const ut64 *)(data);
for (i = 0; i < nblocks; i++) {
ut64 k1 = getblock(blocks, i * 2 + 0);
ut64 k2 = getblock(blocks, i * 2 + 1);
k1 *= c1;
k1 = ROTL64_Murmur3(k1, 31);
k1 *= c2;
h1 ^= k1;
h1 = ROTL64_Murmur3(h1, 27);
h1 += h2;
h1 = h1 * 5 + 0x52dce729;
k2 *= c2;
k2 = ROTL64_Murmur3(k2, 33);
k2 *= c1;
h2 ^= k2;
h2 = ROTL64_Murmur3(h2, 31);
h2 += h1;
h2 = h2 * 5 + 0x38495ab5;
}
const ut8 *tail = (const ut8 *)(data + nblocks * 16);
ut64 k1 = 0;
ut64 k2 = 0;
switch (len & 15) {
case 15:
k2 ^= (ut64)(tail[14]) << 48;
/* fall through */
case 14:
k2 ^= (ut64)(tail[13]) << 40;
/* fall through */
case 13:
k2 ^= (ut64)(tail[12]) << 32;
/* fall through */
case 12:
k2 ^= (ut64)(tail[11]) << 24;
/* fall through */
case 11:
k2 ^= (ut64)(tail[10]) << 16;
/* fall through */
case 10:
k2 ^= (ut64)(tail[9]) << 8;
/* fall through */
case 9:
k2 ^= (ut64)(tail[8]) << 0;
k2 *= c2;
k2 = ROTL64_Murmur3(k2, 33);
k2 *= c1;
h2 ^= k2;
/* fall through */
case 8:
k1 ^= (ut64)(tail[7]) << 56;
/* fall through */
case 7:
k1 ^= (ut64)(tail[6]) << 48;
/* fall through */
case 6:
k1 ^= (ut64)(tail[5]) << 40;
/* fall through */
case 5:
k1 ^= (ut64)(tail[4]) << 32;
/* fall through */
case 4:
k1 ^= (ut64)(tail[3]) << 24;
/* fall through */
case 3:
k1 ^= (ut64)(tail[2]) << 16;
/* fall through */
case 2:
k1 ^= (ut64)(tail[1]) << 8;
/* fall through */
case 1:
k1 ^= (ut64)(tail[0]) << 0;
k1 *= c1;
k1 = ROTL64_Murmur3(k1, 31);
k1 *= c2;
h1 ^= k1;
};
h1 ^= len;
h2 ^= len;
h1 += h2;
h2 += h1;
h1 = fmix64(h1);
h2 = fmix64(h2);
h1 += h2;
h2 += h1;
((ut64 *)out)[0] = h1;
((ut64 *)out)[1] = h2;
}