Inline GPU overflow tracking helpers

PerfectNumbers.Core.Tests/Gpu/PrimeOrderGpuHeuristicsTests.cs

@@ -11,7 +11,7 @@ public class PrimeOrderGpuHeuristicsTests
 {
     public PrimeOrderGpuHeuristicsTests()
     {
-        PrimeOrderGpuHeuristics.ClearAllOverflowForTesting();
+        PrimeOrderGpuHeuristics.OverflowRegistry.Clear();
         PrimeOrderGpuHeuristics.ResetCapabilitiesForTesting();
     }
 
@@ -21,7 +21,7 @@ public class PrimeOrderGpuHeuristicsTests
         const ulong prime = 97UL;
         try
         {
-            PrimeOrderGpuHeuristics.MarkOverflow(prime);
+            PrimeOrderGpuHeuristics.OverflowRegistry[prime] = 0;
 
             GpuPow2ModStatus status = PrimeOrderGpuHeuristics.TryPow2Mod(1UL, prime, out ulong remainder);
 
@@ -30,7 +30,7 @@ public class PrimeOrderGpuHeuristicsTests
         }
         finally
         {
-            PrimeOrderGpuHeuristics.ClearOverflow(prime);
+            PrimeOrderGpuHeuristics.OverflowRegistry.TryRemove(prime, out _);
         }
     }
 
@@ -59,7 +59,7 @@ public class PrimeOrderGpuHeuristicsTests
         finally
         {
             PrimeOrderGpuHeuristics.ResetCapabilitiesForTesting();
-            PrimeOrderGpuHeuristics.ClearOverflow(prime);
+            PrimeOrderGpuHeuristics.OverflowRegistry.TryRemove(prime, out _);
         }
     }
 
@@ -91,7 +91,7 @@ public class PrimeOrderGpuHeuristicsTests
         const ulong prime = 193UL;
         try
         {
-            PrimeOrderGpuHeuristics.MarkOverflow(prime);
+            PrimeOrderGpuHeuristics.OverflowRegistry[prime] = 0;
 
             ulong[] remainders = new ulong[2];
             ulong[] exponents = { 1UL, 2UL };
@@ -103,7 +103,7 @@ public class PrimeOrderGpuHeuristicsTests
         }
         finally
         {
-            PrimeOrderGpuHeuristics.ClearOverflow(prime);
+            PrimeOrderGpuHeuristics.OverflowRegistry.TryRemove(prime, out _);
         }
     }
 
@@ -139,7 +139,7 @@ public class PrimeOrderGpuHeuristicsTests
         finally
         {
             PrimeOrderGpuHeuristics.ResetCapabilitiesForTesting();
-            PrimeOrderGpuHeuristics.ClearOverflow(prime);
+            PrimeOrderGpuHeuristics.OverflowRegistry.TryRemove(prime, out _);
         }
     }
 
@@ -182,7 +182,7 @@ public class PrimeOrderGpuHeuristicsTests
         const ulong prime = 7UL;
         try
         {
-            PrimeOrderGpuHeuristics.MarkOverflow(prime);
+            PrimeOrderGpuHeuristics.OverflowRegistry[prime] = 0;
 
             PrimeOrderCalculator.PrimeOrderResult result = PrimeOrderCalculator.Calculate(prime, null, PrimeOrderCalculator.PrimeOrderSearchConfig.HeuristicDefault);
 
@@ -191,7 +191,7 @@ public class PrimeOrderGpuHeuristicsTests
         }
         finally
         {
-            PrimeOrderGpuHeuristics.ClearOverflow(prime);
+            PrimeOrderGpuHeuristics.OverflowRegistry.TryRemove(prime, out _);
         }
     }
 }

PerfectNumbers.Core/Gpu/PrimeOrderGpuHeuristics.cs

@@ -16,6 +16,8 @@ internal static class PrimeOrderGpuHeuristics
     private static readonly ConcurrentDictionary<ulong, byte> OverflowedPrimes = new();
     private static PrimeOrderGpuCapability s_capability = PrimeOrderGpuCapability.Default;
 
+    internal static ConcurrentDictionary<ulong, byte> OverflowRegistry => OverflowedPrimes;
+
     internal static void OverrideCapabilitiesForTesting(PrimeOrderGpuCapability capability)
     {
         s_capability = capability;
@@ -26,12 +28,6 @@ internal static class PrimeOrderGpuHeuristics
         s_capability = PrimeOrderGpuCapability.Default;
     }
 
-    internal static void MarkOverflow(ulong prime) => OverflowedPrimes[prime] = 0;
-
-    internal static void ClearOverflow(ulong prime) => OverflowedPrimes.TryRemove(prime, out _);
-
-    internal static void ClearAllOverflowForTesting() => OverflowedPrimes.Clear();
-
     public static GpuPow2ModStatus TryPow2Mod(ulong exponent, ulong prime, out ulong remainder)
     {
         remainder = 0UL;
@@ -40,7 +36,9 @@ internal static class PrimeOrderGpuHeuristics
             return GpuPow2ModStatus.Unavailable;
         }
 
-        if (OverflowedPrimes.ContainsKey(prime))
+        ConcurrentDictionary<ulong, byte> overflowRegistry = OverflowedPrimes;
+
+        if (overflowRegistry.ContainsKey(prime))
         {
             return GpuPow2ModStatus.Overflow;
         }
@@ -49,7 +47,7 @@ internal static class PrimeOrderGpuHeuristics
 
         if (!SupportsPrime(prime, capability))
         {
-            MarkOverflow(prime);
+            overflowRegistry[prime] = 0;
             return GpuPow2ModStatus.Overflow;
         }
 
@@ -81,7 +79,9 @@ internal static class PrimeOrderGpuHeuristics
             return GpuPow2ModStatus.Unavailable;
         }
 
-        if (OverflowedPrimes.ContainsKey(prime))
+        ConcurrentDictionary<ulong, byte> overflowRegistry = OverflowedPrimes;
+
+        if (overflowRegistry.ContainsKey(prime))
         {
             return GpuPow2ModStatus.Overflow;
         }
@@ -90,7 +90,7 @@ internal static class PrimeOrderGpuHeuristics
 
         if (!SupportsPrime(prime, capability))
         {
-            MarkOverflow(prime);
+            overflowRegistry[prime] = 0;
             return GpuPow2ModStatus.Overflow;
         }
 

todo.txt

@@ -1,4 +1,3 @@
-- Inline the overflow tracking helpers in PrimeOrderGpuHeuristics so call sites can cache the dictionary locally and reduce delegate indirection.
 - Move the ulong bit-length helper into ULongExtensions and update PrimeOrderGpuHeuristics (and any other callers) to reuse it.
 - Implement the GPU heuristics for PrimeOrderCalculator using ILGPU. Reuse the numeric type compatibility probe results to select between EInteger, ERational, GpuUInt128, or any new GPU-specific big-integer representations, guarantee parity with the CPU heuristic algorithm (including 128-bit primes), prefer the shift-first multiply/shift variant to stay within device ranges, and wire the calculator through the accelerator path.
 - Extend the GPU numeric type probe to cover proposed custom types such as GpuUInt256 and GpuUInt512 once their definitions are available so we can plan the wide-order heuristics on the device.