Persist GPU capability snapshot without volatile

PerfectNumbers.Core/Gpu/PrimeOrderGpuHeuristics.cs

@@ -1,7 +1,6 @@
 using System;
 using System.Collections.Concurrent;
 using System.Numerics;
-using System.Threading;
 
 namespace PerfectNumbers.Core.Gpu;
 
@@ -15,23 +14,18 @@ internal enum GpuPow2ModStatus
 internal static class PrimeOrderGpuHeuristics
 {
     private static readonly ConcurrentDictionary<ulong, byte> OverflowedPrimes = new();
-    private static int s_modulusBitLimit = PrimeOrderGpuCapability.Default.ModulusBits;
+    private static PrimeOrderGpuCapability s_capability = PrimeOrderGpuCapability.Default;
-    private static int s_exponentBitLimit = PrimeOrderGpuCapability.Default.ExponentBits;
 
     internal static void OverrideCapabilitiesForTesting(PrimeOrderGpuCapability capability)
     {
-        Volatile.Write(ref s_modulusBitLimit, capability.ModulusBits);
+        s_capability = capability;
-        Volatile.Write(ref s_exponentBitLimit, capability.ExponentBits);
     }
 
     internal static void ResetCapabilitiesForTesting()
     {
-        OverrideCapabilitiesForTesting(PrimeOrderGpuCapability.Default);
+        s_capability = PrimeOrderGpuCapability.Default;
     }
 
-    private static PrimeOrderGpuCapability Capabilities
-        => new(Volatile.Read(ref s_modulusBitLimit), Volatile.Read(ref s_exponentBitLimit));
-
     internal static void MarkOverflow(ulong prime) => OverflowedPrimes[prime] = 0;
 
     internal static void ClearOverflow(ulong prime) => OverflowedPrimes.TryRemove(prime, out _);
@@ -51,7 +45,7 @@ internal static class PrimeOrderGpuHeuristics
             return GpuPow2ModStatus.Overflow;
         }
 
-        PrimeOrderGpuCapability capability = Capabilities;
+        PrimeOrderGpuCapability capability = s_capability;
 
         if (!SupportsPrime(prime, capability))
         {
@@ -92,7 +86,7 @@ internal static class PrimeOrderGpuHeuristics
             return GpuPow2ModStatus.Overflow;
         }
 
-        PrimeOrderGpuCapability capability = Capabilities;
+        PrimeOrderGpuCapability capability = s_capability;
 
         if (!SupportsPrime(prime, capability))
         {

todo.txt

@@ -1,2 +1,4 @@
+- 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.