Port DefaultPixelBlenders

src/ImageSharp/PixelFormats/PixelBlenders/DefaultPixelBlenders.Generated.tt

@@ -13,6 +13,10 @@
 
 // <auto-generated />
 using System.Numerics;
+using System.Runtime.CompilerServices;
+using System.Runtime.InteropServices;
+using System.Runtime.Intrinsics;
+using System.Runtime.Intrinsics.X86;
 
 namespace SixLabors.ImageSharp.PixelFormats.PixelBlenders;
 
@@ -86,18 +90,79 @@ var blenders = new []{
         protected override void BlendFunction(Span<Vector4> destination, ReadOnlySpan<Vector4> background, ReadOnlySpan<Vector4> source, float amount)
         {
             amount = Numerics.Clamp(amount, 0, 1);
-            for (int i = 0; i < destination.Length; i++)
+
+            if (Avx2.IsSupported && destination.Length >= 2)
+            {
+                // Divide by 2 as 4 elements per Vector4 and 8 per Vector256<float>
+                ref Vector256<float> destinationBase = ref Unsafe.As<Vector4, Vector256<float>>(ref MemoryMarshal.GetReference(destination));
+                ref Vector256<float> destinationLast = ref Unsafe.Add(ref destinationBase, (IntPtr)((uint)destination.Length / 2u));
+
+                ref Vector256<float> backgroundBase = ref Unsafe.As<Vector4, Vector256<float>>(ref MemoryMarshal.GetReference(background));
+                ref Vector256<float> sourceBase = ref Unsafe.As<Vector4, Vector256<float>>(ref MemoryMarshal.GetReference(source));
+                Vector256<float> opacity = Vector256.Create(amount);
+
+                while (Unsafe.IsAddressLessThan(ref destinationBase, ref destinationLast))
+                {
+                    destinationBase = PorterDuffFunctions.<#=blender_composer#>(backgroundBase, sourceBase, opacity);
+                    destinationBase = ref Unsafe.Add(ref destinationBase, 1);
+                    backgroundBase = ref Unsafe.Add(ref backgroundBase, 1);
+                    sourceBase = ref Unsafe.Add(ref sourceBase, 1);
+                }
+
+                if (Numerics.Modulo2(destination.Length) != 0)
+                {
+                    // Vector4 fits neatly in pairs. Any overlap has to be equal to 1.
+                    int i = destination.Length - 1;
+                    destination[i] = PorterDuffFunctions.<#=blender_composer#>(background[i], source[i], amount);
+                }
+            }
+            else
             {
-                destination[i] = PorterDuffFunctions.<#=blender_composer#>(background[i], source[i], amount);
+                for (int i = 0; i < destination.Length; i++)
+                {
+                    destination[i] = PorterDuffFunctions.<#=blender_composer#>(background[i], source[i], amount);
+                }
             }
         }
 
         /// <inheritdoc />
         protected override void BlendFunction(Span<Vector4> destination, ReadOnlySpan<Vector4> background, ReadOnlySpan<Vector4> source, ReadOnlySpan<float> amount)
         {
-            for (int i = 0; i < destination.Length; i++)
+            if (Avx2.IsSupported && destination.Length >= 2)
+            {
+                // Divide by 2 as 4 elements per Vector4 and 8 per Vector256<float>
+                ref Vector256<float> destinationBase = ref Unsafe.As<Vector4, Vector256<float>>(ref MemoryMarshal.GetReference(destination));
+                ref Vector256<float> destinationLast = ref Unsafe.Add(ref destinationBase, (IntPtr)((uint)destination.Length / 2u));
+
+                ref Vector256<float> backgroundBase = ref Unsafe.As<Vector4, Vector256<float>>(ref MemoryMarshal.GetReference(background));
+                ref Vector256<float> sourceBase = ref Unsafe.As<Vector4, Vector256<float>>(ref MemoryMarshal.GetReference(source));
+                ref float amountBase = ref MemoryMarshal.GetReference(amount);
+
+                while (Unsafe.IsAddressLessThan(ref destinationBase, ref destinationLast))
+                {
+                    // TODO: It would be better if we can clamp this outside of the loop using our SIMD methods.
+                    Vector256<float> opacity = Vector256.Create(Numerics.Clamp(amountBase, 0, 1F));
+
+                    destinationBase = PorterDuffFunctions.<#=blender_composer#>(backgroundBase, sourceBase, opacity);
+                    destinationBase = ref Unsafe.Add(ref destinationBase, 1);
+                    backgroundBase = ref Unsafe.Add(ref backgroundBase, 1);
+                    sourceBase = ref Unsafe.Add(ref sourceBase, 1);
+                    amountBase = ref Unsafe.Add(ref amountBase, 1);
+                }
+
+                if (Numerics.Modulo2(destination.Length) != 0)
+                {
+                    // Vector4 fits neatly in pairs. Any overlap has to be equal to 1.
+                    int i = destination.Length - 1;
+                    destination[i] = PorterDuffFunctions.<#=blender_composer#>(background[i], source[i], Numerics.Clamp(amount[i], 0, 1F));
+                }
+            }
+            else
             {
-                destination[i] = PorterDuffFunctions.<#=blender_composer#>(background[i], source[i], Numerics.Clamp(amount[i], 0, 1));
+                for (int i = 0; i < destination.Length; i++)
+                {
+                    destination[i] = PorterDuffFunctions.<#=blender_composer#>(background[i], source[i], Numerics.Clamp(amount[i], 0, 1F));
+                }
             }
         }
     }