Examples:

        for (int i = fromIndex; i < end; i++) {
            result = 31 * result + a[i];
        }

    @Benchmark
    public int nativeSegmentJava() {
        int result = 1;
        for (long i = 0; i < ELEM_SIZE; i++) {
            result = 31 * result + nativeSegment.get(JAVA_BYTE, i);
        }
        return result;
    }


Basic idea, this reformulation:

result = result * 31^N + a[0] * 31^(N-1) + ... a[i] * 32^ (N-1-i) ... + a[N-2] * 31 + a[N-1]

All the values can thus be calculated in parallel.

This work would generallize the intrinsics we already have for strings:
https://bugs.openjdk.org/browse/JDK-8282664
https://github.com/openjdk/jdk/pull/10847

Other previous investigation:
https://inside.java/2021/01/27/extending-c2-autovectorization-capabilities/

A further generalization would to vectorize prefix-sums and more generally scans:
https://eme64.github.io/blog/2023/11/03/C2-AutoVectorizer-Improvement-Ideas.html