HashMap hash方法分析

HashMap 中hash table 定位算法：

int hash = hash(key.hashCode());
int i = indexFor(hash, table.length);

        int hash = hash(key.hashCode());
        int i = indexFor(hash, table.length);

其中indexFor和hash源码如下：

/**
 * Applies a supplemental hash function to a given hashCode, which
 * defends against poor quality hash functions.  This is critical
 * because HashMap uses power-of-two length hash tables, that
 * otherwise encounter collisions for hashCodes that do not differ
 * in lower bits. Note: Null keys always map to hash 0, thus index 0.
 */
static int hash(int h) {
    // This function ensures that hashCodes that differ only by
    // constant multiples at each bit position have a bounded
    // number of collisions (approximately 8 at default load factor).
    h ^= (h >>> 20) ^ (h >>> 12);
    return h ^ (h >>> 7) ^ (h >>> 4);
}

/**
 * Returns index for hash code h.
 */
static int indexFor(int h, int length) {
    return h & (length-1);
}

    /**
     * Applies a supplemental hash function to a given hashCode, which
     * defends against poor quality hash functions.  This is critical
     * because HashMap uses power-of-two length hash tables, that
     * otherwise encounter collisions for hashCodes that do not differ
     * in lower bits. Note: Null keys always map to hash 0, thus index 0.
     */
    static int hash(int h) {
        // This function ensures that hashCodes that differ only by
        // constant multiples at each bit position have a bounded
        // number of collisions (approximately 8 at default load factor).
        h ^= (h >>> 20) ^ (h >>> 12);
        return h ^ (h >>> 7) ^ (h >>> 4);
    }

    /**
     * Returns index for hash code h.
     */
    static int indexFor(int h, int length) {
        return h & (length-1);
    }

indexFor这个方法论坛中已有人分析过，这里就不再分析。
现在分析一下hash算法：

h ^= (h >>> 20) ^ (h >>> 12);
return h ^ (h >>> 7) ^ (h >>> 4);

        h ^= (h >>> 20) ^ (h >>> 12);
        return h ^ (h >>> 7) ^ (h >>> 4);

假设key.hashCode()的值为：0x7FFFFFFF，table.length为默认值16。
上面算法执行如下：



得到i=15

其中h^(h>>>7)^(h>>>4) 结果中的位运行标识是把h>>>7 换成 h>>>8来看。

即最后h^(h>>>8)^(h>>>4) 运算后hashCode值每位数值如下：
8=8
7=7^8
6=6^7^8
5=5^8^7^6
4=4^7^6^5^8
3=3^8^6^5^8^4^7
2=2^7^5^4^7^3^8^6
1=1^6^4^3^8^6^2^7^5
结果中的1、2、3三位出现重复位^运算
3=3^8^6^5^8^4^7     ->   3^6^5^4^7
2=2^7^5^4^7^3^8^6   ->   2^5^4^3^8^6
1=1^6^4^3^8^6^2^7^5 ->   1^4^3^8^2^7^5

算法中是采用(h>>>7)而不是(h>>>8)的算法，应该是考虑1、2、3三位出现重复位^运算的情况。使得最低位上原hashCode的8位都参与了^运算，所以在table.length为默认值16的情况下面，hashCode任意位的变化基本都能反应到最终hash table 定位算法中，这种情况下只有原hashCode第3位高1位变化不会反应到结果中，即：0x7FFFF7FF的i=15。