Optimizations and tiny cleanup

EvenPerfectBitScanner.Benchmarks/GpuUInt128BinaryGcdBenchmarks.cs

@@ -1,4 +1,4 @@
-using System.Collections.Generic;
+using System.Numerics;
 using System.Runtime.CompilerServices;
 using BenchmarkDotNet.Attributes;
 using BenchmarkDotNet.Jobs;
@@ -29,17 +29,17 @@ public class GpuUInt128BinaryGcdBenchmarks
     [Benchmark(Baseline = true)]
     public GpuUInt128 ReusedVariables()
     {
-        return GpuUInt128.BinaryGcd(Input.Left, Input.Right);
+        return BinaryGcdWithReusedVariables(Input.Left, Input.Right);
     }
 
     [Benchmark]
     public GpuUInt128 TemporaryStructPerIteration()
     {
-        return BinaryGcdWithTemporary(Input.Left, Input.Right);
+        return GpuUInt128.BinaryGcd(Input.Left, Input.Right);
     }
 
     [MethodImpl(MethodImplOptions.AggressiveInlining)]
-    private static GpuUInt128 BinaryGcdWithTemporary(GpuUInt128 u, GpuUInt128 v)
+    private static GpuUInt128 BinaryGcdWithReusedVariables(GpuUInt128 u, GpuUInt128 v)
     {
         if (u.IsZero)
         {
@@ -51,7 +51,12 @@ public class GpuUInt128BinaryGcdBenchmarks
             return u;
         }
 
-        int shift = GpuUInt128.TrailingZeroCount(new GpuUInt128(u.High | v.High, u.Low | v.Low));
+        ulong combinedLow = u.Low | v.Low;
+        ulong combinedHigh = u.High | v.High;
+        int shift = combinedLow != 0UL
+            ? BitOperations.TrailingZeroCount(combinedLow)
+            : 64 + BitOperations.TrailingZeroCount(combinedHigh);
+
         int zu = GpuUInt128.TrailingZeroCount(u);
         u >>= zu;
 

EvenPerfectBitScanner.Benchmarks/GpuUInt128Mul64Benchmarks.cs

@@ -24,7 +24,7 @@ public class GpuUInt128Mul64Benchmarks
     public static IEnumerable<Mul64Input> GetInputs() => Inputs;
 
     [Benchmark(Baseline = true)]
-    public GpuUInt128 HighProductMaterialisedInLocal()
+    public GpuUInt128 HighProductMaterializedInLocal()
     {
         GpuUInt128 value = Input.Multiplicand;
         value.Mul64(Input.Multiplier);

EvenPerfectBitScanner.Benchmarks/GpuUInt128MulModBenchmarks.cs

@@ -10,8 +10,8 @@ namespace EvenPerfectBitScanner.Benchmarks;
 [MemoryDiagnoser]
 public class GpuUInt128MulModBenchmarks
 {
-    private static readonly MulModInput[] Inputs = new MulModInput[]
-    {
+    private static readonly MulModInput[] Inputs =
+    [
         new(
             new GpuUInt128(0UL, 5UL),
             new GpuUInt128(0UL, 7UL),
@@ -32,7 +32,7 @@ public class GpuUInt128MulModBenchmarks
             new GpuUInt128(0x7FFF_FFFF_FFFF_FFFFUL, 0x8000_0000_0000_0000UL),
             new GpuUInt128(0x7FFF_FFFF_FFFF_FFFFUL, 0xFFFF_FFFF_FFFF_FFF1UL),
             "MixedMagnitude"),
-    };
+    ];
 
     [ParamsSource(nameof(GetInputs))]
     public MulModInput Input { get; set; }
@@ -42,6 +42,7 @@ public class GpuUInt128MulModBenchmarks
     [Benchmark(Baseline = true)]
     public GpuUInt128 InPlaceMulMod()
     {
+        // TODO: Switch the extension method to allocate per iteration, as it's faster. Keep the benchmarks.
         GpuUInt128 value = Input.Left;
         value.MulMod(Input.Right, Input.Modulus);
         return value;
@@ -56,6 +57,7 @@ public class GpuUInt128MulModBenchmarks
     [MethodImpl(MethodImplOptions.AggressiveInlining)]
     private static GpuUInt128 MulModInline(GpuUInt128 left, GpuUInt128 right, GpuUInt128 modulus)
     {
+        // Reuse the variables to new instances and lower registry pressure, if possible.
         GpuUInt128 a = left;
         GpuUInt128 b = right;
         GpuUInt128 result = new();

EvenPerfectBitScanner.Benchmarks/GpuUInt128MulModByLimbBenchmarks.cs

@@ -42,6 +42,7 @@ public class GpuUInt128MulModByLimbBenchmarks
     [Benchmark]
     public GpuUInt128 InPlaceReduction()
     {
+        // TODO: Switch the extension method back to legacy method, as it's faster. Keep the benchmarks.
         GpuUInt128 value = Input.Left;
         value.MulModByLimb(Input.Right, Input.Modulus);
         return value;