All geek questions in one place

Vector C # with SIMD using Vector <T> is slower than classic loop

I saw several articles describing how Vector<T>SIMD is enabled and implemented using JIT functions, so the compiler will correctly output the AVS / SSE / ... commands when using it, allowing you to use the code much faster than the classic, linear contours (example here ) .

I decided to try to rewrite the method that I should see if I can speed things up, but so far I have not succeeded, and the vectorized code works 3 times slower than the original, and I'm not quite sure why. Here are two versions of the method that check if two instances have Span<float>all pairs of elements in the same position that have the same position relative to the threshold value.

// Classic implementation
public static unsafe bool MatchElementwiseThreshold(this Span<float> x1, Span<float> x2, float threshold)
{
    fixed (float* px1 = &x1.DangerousGetPinnableReference(), px2 = &x2.DangerousGetPinnableReference())
        for (int i = 0; i < x1.Length; i++)
            if (px1[i] > threshold != px2[i] > threshold)
                return false;
    return true;
}

// Vectorized
public static unsafe bool MatchElementwiseThresholdSIMD(this Span<float> x1, Span<float> x2, float threshold)
{
    // Setup the test vector
    int l = Vector<float>.Count;
    float* arr = stackalloc float[l];
    for (int i = 0; i < l; i++)
        arr[i] = threshold;
    Vector<float> cmp = Unsafe.Read<Vector<float>>(arr);
    fixed (float* px1 = &x1.DangerousGetPinnableReference(), px2 = &x2.DangerousGetPinnableReference())
    {
        // Iterate in chunks
        int
            div = x1.Length / l,
            mod = x1.Length % l,
            i = 0,
            offset = 0;
        for (; i < div; i += 1, offset += l)
        {
            Vector<float>
                v1 = Unsafe.Read<Vector<float>>(px1 + offset),
                v1cmp = Vector.GreaterThan<float>(v1, cmp),
                v2 = Unsafe.Read<Vector<float>>(px2 + offset),
                v2cmp = Vector.GreaterThan<float>(v2, cmp);
            float*
                pcmp1 = (float*)Unsafe.AsPointer(ref v1cmp),
                pcmp2 = (float*)Unsafe.AsPointer(ref v2cmp);
            for (int j = 0; j < l; j++)
                if (pcmp1[j] == 0 != (pcmp2[j] == 0))
                    return false;
        }

        // Test the remaining items, if any
        if (mod == 0) return true;
        for (i = x1.Length - mod; i < x1.Length; i++)
            if (px1[i] > threshold != px2[i] > threshold)
                return false;
    }
    return true;
}

As I said, I tested both versions using BenchmarkDotNet, and the one that Vector<T>works is 3 times slower than the other. I tried to run tests at intervals of different lengths (from 100 to 2000), but the vector method continues to be much slower than the other.

Did I miss something obvious here?

Thank!

: , , , , , Parallel.For -.

, , , , .

+4
2

** EDIT ** , , ...

public static bool MatchElementwiseThresholdSIMD(ReadOnlySpan<float> x1, ReadOnlySpan<float> x2, float threshold)
{
    if (x1.Length != x2.Length) throw new ArgumentException("x1.Length != x2.Length");

    if (Vector.IsHardwareAccelerated)
    {
        var vx1 = x1.NonPortableCast<float, Vector<float>>();
        var vx2 = x2.NonPortableCast<float, Vector<float>>();

        var vthreshold = new Vector<float>(threshold);
        for (int i = 0; i < vx1.Length; ++i)
        {
            var v1cmp = Vector.GreaterThan(vx1[i], vthreshold);
            var v2cmp = Vector.GreaterThan(vx2[i], vthreshold);
            if (Vector.Xor(v1cmp, v2cmp) != Vector<int>.Zero)
                return false;
        }

        x1 = x1.Slice(Vector<float>.Count * vx1.Length);
        x2 = x2.Slice(Vector<float>.Count * vx2.Length);
    }

    for (var i = 0; i < x1.Length; i++)
        if (x1[i] > threshold != x2[i] > threshold)
            return false;

    return true;
}

, ( , ), , , SIMD...

( ...)

... , , , , , ... , , , (, - , - ...)

    public static bool MatchElementwiseThreshold<T>(ReadOnlySpan<T> x1, ReadOnlySpan<T> x2, T threshold)
        where T : struct
    {
        if (x1.Length != x2.Length)
            throw new ArgumentException("x1.Length != x2.Length");

        if (Vector.IsHardwareAccelerated)
        {
            var vx1 = x1.NonPortableCast<T, Vector<T>>();
            var vx2 = x2.NonPortableCast<T, Vector<T>>();

            var vthreshold = new Vector<T>(threshold);
            for (int i = 0; i < vx1.Length; ++i)
            {
                var v1cmp = Vector.GreaterThan(vx1[i], vthreshold);
                var v2cmp = Vector.GreaterThan(vx2[i], vthreshold);
                if (Vector.AsVectorInt32(Vector.Xor(v1cmp, v2cmp)) != Vector<int>.Zero)
                    return false;
            }

            // slice them to handling remaining elementss
            x1 = x1.Slice(Vector<T>.Count * vx1.Length);
            x2 = x2.Slice(Vector<T>.Count * vx1.Length);
        }

        var comparer = System.Collections.Generic.Comparer<T>.Default;
        for (int i = 0; i < x1.Length; i++)
            if ((comparer.Compare(x1[i], threshold) > 0) != (comparer.Compare(x2[i], threshold) > 0))
                return false;

        return true;
    }
0

Source: https://habr.com/ru/post/1692013/

More articles:


All Articles