import jdk.incubator.vector.FloatVector;
import jdk.incubator.vector.VectorOperators;
import jdk.incubator.vector.VectorSpecies;

/**
 * JEP 537, Vector API (twelfth incubator). The same source and the same class file run on JDK 26
 * (eleventh incubator, JEP 529) and JDK 27, because the public API did not change between them:
 * docs/output/50-vector-api-surface.txt is the javap diff that shows it. For what the API is and
 * how to use it well, see the companion guide to JEP 537 on ankurm.com.
 * Run with --add-modules jdk.incubator.vector.
 */
public class VectorDemo {
    static final VectorSpecies<Float> SPECIES = FloatVector.SPECIES_PREFERRED;

    static float scalarDot(float[] a, float[] b) {
        float sum = 0;
        for (int i = 0; i < a.length; i++) {
            sum += a[i] * b[i];
        }
        return sum;
    }

    static float vectorDot(float[] a, float[] b) {
        int i = 0;
        FloatVector acc = FloatVector.zero(SPECIES);
        int bound = SPECIES.loopBound(a.length);
        for (; i < bound; i += SPECIES.length()) {
            acc = FloatVector.fromArray(SPECIES, a, i).fma(FloatVector.fromArray(SPECIES, b, i), acc);
        }
        float sum = acc.reduceLanes(VectorOperators.ADD);
        for (; i < a.length; i++) {          // the tail the vector loop could not cover
            sum += a[i] * b[i];
        }
        return sum;
    }

    public static void main(String[] args) {
        System.out.println("java.version      = " + System.getProperty("java.version"));
        System.out.println("preferred species = " + SPECIES + "  (" + SPECIES.length() + " float lanes, machine dependent)");
        int n = 1_000_003;                       // deliberately not a multiple of the lane count
        float[] a = new float[n], b = new float[n];
        for (int i = 0; i < n; i++) {
            a[i] = (i % 7) * 0.5f;
            b[i] = (i % 5) * 0.25f;
        }
        float scalar = scalarDot(a, b);
        float vector = vectorDot(a, b);
        System.out.println("scalar dot        = " + scalar);
        System.out.println("vector dot        = " + vector);
        // Floating-point addition is not associative, so the two sums differ in the last bits; compare with a tolerance.
        Check.that(Math.abs(scalar - vector) / Math.abs(scalar) < 1e-3, "vector and scalar dot products agree to within 0.1%");
    }
}