mlemanczyk/even-perfect-numbers-scanner
1
0 Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Heuristic and small prime factor kernels restored in PrimeOrderAcceleratorTemplate.

Browse Source
This commit is contained in:
mlemanczyk committed 2025-12-28 00:12:39 +01:00
1 parent 6603f65993
commit 75a2017385
1 file changed
+77 -80
Download Patch File Download Diff File Expand all files Collapse all files
PerfectNumbers.Core/Gpu/Accelerators/PrimeOrderCalculatorAccelerator.cs
+77 -80
View File
@@ -16,7 +16,7 @@ public sealed class PrimeOrderCalculatorAccelerator
public static PrimeOrderCalculatorAccelerator Rent(int primeTesterCapacity) public static PrimeOrderCalculatorAccelerator Rent(int primeTesterCapacity)
{ {
var pool = _pool ??= new(PerfectNumberConstants.DefaultThreadPoolCapacity); var pool = _pool ??= new(PerfectNumberConstants.DefaultThreadPoolCapacity);
if (pool.Pop() is { } gpu) if (pool.TryPop(out var gpu))
{ {
gpu.EnsureCapacity(PerfectNumberConstants.DefaultFactorsBuffer, primeTesterCapacity); gpu.EnsureCapacity(PerfectNumberConstants.DefaultFactorsBuffer, primeTesterCapacity);
return gpu; return gpu;
@@ -67,8 +67,8 @@ public sealed class PrimeOrderCalculatorAccelerator
// TODO: Review which tables are needed when final execution path is defined. // TODO: Review which tables are needed when final execution path is defined.
LastDigitGpuTables.WarmUp(i, stream); LastDigitGpuTables.WarmUp(i, stream);
_kernels[i] = new CalculatorKernels(accelerator); _kernels[i] = new CalculatorKernels(accelerator);
// HeuristicCombinedGpuTables.WarmUp(i, stream); HeuristicCombinedGpuTables.WarmUp(i, stream);
// SmallPrimeFactorGpuTables.WarmUp(i, stream); SmallPrimeFactorGpuTables.WarmUp(i, stream);
// SharedHeuristicGpuTables.EnsureStaticTables(accelerator, stream); // SharedHeuristicGpuTables.EnsureStaticTables(accelerator, stream);
// _ = GpuKernelPool.GetOrAddKernels(accelerator, stream, KernelType.None); // _ = GpuKernelPool.GetOrAddKernels(accelerator, stream, KernelType.None);
// KernelContainer kernels = GpuKernelPool.GetOrAddKernels(accelerator, stream); // KernelContainer kernels = GpuKernelPool.GetOrAddKernels(accelerator, stream);
@@ -170,7 +170,7 @@ public sealed class PrimeOrderCalculatorAccelerator
// public readonly Action<AcceleratorStream, Index1D, ArrayView<ulong>, ArrayView<byte>> SharesFactorKernelLauncher = KernelUtil.GetKernel(accelerator.LoadAutoGroupedStreamKernel<Index1D, ArrayView<ulong>, ArrayView<byte>>(PrimeTesterKernels.SharesFactorKernel)).CreateLauncherDelegate<Action<AcceleratorStream, Index1D, ArrayView<ulong>, ArrayView<byte>>>(); // public readonly Action<AcceleratorStream, Index1D, ArrayView<ulong>, ArrayView<byte>> SharesFactorKernelLauncher = KernelUtil.GetKernel(accelerator.LoadAutoGroupedStreamKernel<Index1D, ArrayView<ulong>, ArrayView<byte>>(PrimeTesterKernels.SharesFactorKernel)).CreateLauncherDelegate<Action<AcceleratorStream, Index1D, ArrayView<ulong>, ArrayView<byte>>>();
// public readonly Action<AcceleratorStream, Index1D, ulong, uint, ArrayView1D<uint, Stride1D.Dense>, ArrayView1D<ulong, Stride1D.Dense>, int, ArrayView1D<ulong, Stride1D.Dense>, ArrayView1D<int, Stride1D.Dense>, ArrayView1D<int, Stride1D.Dense>, ArrayView1D<ulong, Stride1D.Dense>> SmallPrimeFactorKernelLauncher = KernelUtil.GetKernel(accelerator.LoadAutoGroupedStreamKernel<Index1D, ulong, uint, ArrayView1D<uint, Stride1D.Dense>, ArrayView1D<ulong, Stride1D.Dense>, int, ArrayView1D<ulong, Stride1D.Dense>, ArrayView1D<int, Stride1D.Dense>, ArrayView1D<int, Stride1D.Dense>, ArrayView1D<ulong, Stride1D.Dense>>(SmallPrimeFactorKernels.SmallPrimeFactorKernelScan)).CreateLauncherDelegate<Action<AcceleratorStream, Index1D, ulong, uint, ArrayView1D<uint, Stride1D.Dense>, ArrayView1D<ulong, Stride1D.Dense>, int, ArrayView1D<ulong, Stride1D.Dense>, ArrayView1D<int, Stride1D.Dense>, ArrayView1D<int, Stride1D.Dense>, ArrayView1D<ulong, Stride1D.Dense>>>(); public readonly Action<AcceleratorStream, Index1D, ulong, uint, ArrayView1D<uint, Stride1D.Dense>, ArrayView1D<ulong, Stride1D.Dense>, int, ArrayView1D<ulong, Stride1D.Dense>, ArrayView1D<int, Stride1D.Dense>, ArrayView1D<int, Stride1D.Dense>, ArrayView1D<ulong, Stride1D.Dense>> SmallPrimeFactorKernelLauncher = KernelUtil.GetKernel(accelerator.LoadAutoGroupedStreamKernel<Index1D, ulong, uint, ArrayView1D<uint, Stride1D.Dense>, ArrayView1D<ulong, Stride1D.Dense>, int, ArrayView1D<ulong, Stride1D.Dense>, ArrayView1D<int, Stride1D.Dense>, ArrayView1D<int, Stride1D.Dense>, ArrayView1D<ulong, Stride1D.Dense>>(SmallPrimeFactorKernels.SmallPrimeFactorKernelScan)).CreateLauncherDelegate<Action<AcceleratorStream, Index1D, ulong, uint, ArrayView1D<uint, Stride1D.Dense>, ArrayView1D<ulong, Stride1D.Dense>, int, ArrayView1D<ulong, Stride1D.Dense>, ArrayView1D<int, Stride1D.Dense>, ArrayView1D<int, Stride1D.Dense>, ArrayView1D<ulong, Stride1D.Dense>>>();
// public readonly Action<AcceleratorStream, Index1D, ArrayView<ulong>, ArrayView<uint>, ArrayView<uint>, ArrayView<uint>, ArrayView<uint>, ArrayView<ulong>, ArrayView<ulong>, ArrayView<ulong>, ArrayView<ulong>, ArrayView<byte>> SmallPrimeSieveKernelLauncher = KernelUtil.GetKernel(accelerator.LoadAutoGroupedStreamKernel<Index1D, ArrayView<ulong>, ArrayView<uint>, ArrayView<uint>, ArrayView<uint>, ArrayView<uint>, ArrayView<ulong>, ArrayView<ulong>, ArrayView<ulong>, ArrayView<ulong>, ArrayView<byte>>(PrimeTesterKernels.SmallPrimeSieveKernel)).CreateLauncherDelegate<Action<AcceleratorStream, Index1D, ArrayView<ulong>, ArrayView<uint>, ArrayView<uint>, ArrayView<uint>, ArrayView<uint>, ArrayView<ulong>, ArrayView<ulong>, ArrayView<ulong>, ArrayView<ulong>, ArrayView<byte>>>(); // public readonly Action<AcceleratorStream, Index1D, ArrayView<ulong>, ArrayView<uint>, ArrayView<uint>, ArrayView<uint>, ArrayView<uint>, ArrayView<ulong>, ArrayView<ulong>, ArrayView<ulong>, ArrayView<ulong>, ArrayView<byte>> SmallPrimeSieveKernelLauncher = KernelUtil.GetKernel(accelerator.LoadAutoGroupedStreamKernel<Index1D, ArrayView<ulong>, ArrayView<uint>, ArrayView<uint>, ArrayView<uint>, ArrayView<uint>, ArrayView<ulong>, ArrayView<ulong>, ArrayView<ulong>, ArrayView<ulong>, ArrayView<byte>>(PrimeTesterKernels.SmallPrimeSieveKernel)).CreateLauncherDelegate<Action<AcceleratorStream, Index1D, ArrayView<ulong>, ArrayView<uint>, ArrayView<uint>, ArrayView<uint>, ArrayView<uint>, ArrayView<ulong>, ArrayView<ulong>, ArrayView<ulong>, ArrayView<ulong>, ArrayView<byte>>>();
@@ -360,98 +360,95 @@ public sealed class PrimeOrderCalculatorAccelerator
Accelerator = accelerator; Accelerator = accelerator;
AcceleratorIndex = acceleratorIndex; AcceleratorIndex = acceleratorIndex;
// lock (accelerator) Input = accelerator.Allocate1D<ulong>(Math.Max(Math.Max(specialMaxFactorCapacity, primeTesterCapacity), PrimeOrderConstants.MaxGpuBatchSize));
// { InputView = Input.View;
Input = accelerator.Allocate1D<ulong>(Math.Max(Math.Max(specialMaxFactorCapacity, primeTesterCapacity), PrimeOrderConstants.MaxGpuBatchSize));
InputView = Input.View;
OutputByte = Accelerator.Allocate1D<byte>(primeTesterCapacity); OutputByte = Accelerator.Allocate1D<byte>(primeTesterCapacity);
OutputByteView = OutputByte.View; OutputByteView = OutputByte.View;
OutputInt = Accelerator.Allocate1D<int>(Math.Max(Math.Max(primeTesterCapacity, smallPrimeFactorSlotCount), PrimeOrderConstants.GpuSmallPrimeFactorSlots)); OutputInt = Accelerator.Allocate1D<int>(Math.Max(Math.Max(primeTesterCapacity, smallPrimeFactorSlotCount), PrimeOrderConstants.GpuSmallPrimeFactorSlots));
OutputIntView = OutputInt.View; OutputIntView = OutputInt.View;
OutputInt2 = accelerator.Allocate1D<int>(PrimeOrderConstants.GpuSmallPrimeFactorSlots); OutputInt2 = accelerator.Allocate1D<int>(PrimeOrderConstants.GpuSmallPrimeFactorSlots);
OutputIntView2 = OutputInt2.View; OutputIntView2 = OutputInt2.View;
OutputUlong = accelerator.Allocate1D<ulong>(Math.Max(Math.Max(Math.Max(specialMaxFactorCapacity, smallPrimeFactorSlotCount), PrimeOrderConstants.MaxGpuBatchSize), PrimeOrderConstants.GpuSmallPrimeFactorSlots)); OutputUlong = accelerator.Allocate1D<ulong>(Math.Max(Math.Max(Math.Max(specialMaxFactorCapacity, smallPrimeFactorSlotCount), PrimeOrderConstants.MaxGpuBatchSize), PrimeOrderConstants.GpuSmallPrimeFactorSlots));
OutputUlongView = OutputUlong.View; OutputUlongView = OutputUlong.View;
OutputUlong2 = accelerator.Allocate1D<ulong>(1); OutputUlong2 = accelerator.Allocate1D<ulong>(1);
OutputUlongView2 = OutputUlong2.View; OutputUlongView2 = OutputUlong2.View;
Pow2ModEntriesToTestOnDevice = accelerator.Allocate1D<KeyValuePair<ulong, int>>(factorsCount); Pow2ModEntriesToTestOnDevice = accelerator.Allocate1D<KeyValuePair<ulong, int>>(factorsCount);
Pow2ModEntriesToTestOnDeviceView = Pow2ModEntriesToTestOnDevice.View; Pow2ModEntriesToTestOnDeviceView = Pow2ModEntriesToTestOnDevice.View;
WorkFactorBuffer = accelerator.Allocate1D<ulong>(PrimeOrderConstants.GpuSmallPrimeFactorSlots); WorkFactorBuffer = accelerator.Allocate1D<ulong>(PrimeOrderConstants.GpuSmallPrimeFactorSlots);
CandidateBuffer = accelerator.Allocate1D<ulong>(PrimeOrderConstants.HeuristicCandidateLimit); CandidateBuffer = accelerator.Allocate1D<ulong>(PrimeOrderConstants.HeuristicCandidateLimit);
StackIndexBuffer = accelerator.Allocate1D<int>(PrimeOrderConstants.HeuristicStackCapacity); StackIndexBuffer = accelerator.Allocate1D<int>(PrimeOrderConstants.HeuristicStackCapacity);
StackExponentBuffer = accelerator.Allocate1D<int>(PrimeOrderConstants.HeuristicStackCapacity); StackExponentBuffer = accelerator.Allocate1D<int>(PrimeOrderConstants.HeuristicStackCapacity);
StackProductBuffer = accelerator.Allocate1D<ulong>(PrimeOrderConstants.HeuristicStackCapacity); StackProductBuffer = accelerator.Allocate1D<ulong>(PrimeOrderConstants.HeuristicStackCapacity);
WorkFactorBufferView = WorkFactorBuffer.View; WorkFactorBufferView = WorkFactorBuffer.View;
CandidateBufferView = CandidateBuffer.View; CandidateBufferView = CandidateBuffer.View;
StackIndexBufferView = StackIndexBuffer.View; StackIndexBufferView = StackIndexBuffer.View;
StackExponentBufferView = StackExponentBuffer.View; StackExponentBufferView = StackExponentBuffer.View;
StackProductBufferView = StackProductBuffer.View; StackProductBufferView = StackProductBuffer.View;
CalculateOrderKernelBuffers = new OrderKernelBuffers( CalculateOrderKernelBuffers = new OrderKernelBuffers(
OutputUlongView, OutputUlongView,
OutputIntView, OutputIntView,
WorkFactorBufferView, WorkFactorBufferView,
OutputIntView2, OutputIntView2,
CandidateBufferView, CandidateBufferView,
StackIndexBufferView, StackIndexBufferView,
StackExponentBufferView, StackExponentBufferView,
StackProductBufferView, StackProductBufferView,
OutputUlongView2, OutputUlongView2,
OutputByteView OutputByteView
); );
CalculateOrderWideExponentBuffer = accelerator.Allocate1D<GpuUInt128>(PrimeOrderConstants.MaxGpuBatchSize); CalculateOrderWideExponentBuffer = accelerator.Allocate1D<GpuUInt128>(PrimeOrderConstants.MaxGpuBatchSize);
CalculateOrderWideRemainderBuffer = accelerator.Allocate1D<GpuUInt128>(PrimeOrderConstants.MaxGpuBatchSize); CalculateOrderWideRemainderBuffer = accelerator.Allocate1D<GpuUInt128>(PrimeOrderConstants.MaxGpuBatchSize);
CalculateOrderWideExponentBufferView = CalculateOrderWideExponentBuffer.View; CalculateOrderWideExponentBufferView = CalculateOrderWideExponentBuffer.View;
CalculateOrderWideRemainderBufferView = CalculateOrderWideRemainderBuffer.View; CalculateOrderWideRemainderBufferView = CalculateOrderWideRemainderBuffer.View;
LastDigitGpuTables lastDigitSharedTables = LastDigitGpuTables.GetStaticTables(acceleratorIndex); LastDigitGpuTables lastDigitSharedTables = LastDigitGpuTables.GetStaticTables(acceleratorIndex);
DevicePrimesLastOne = lastDigitSharedTables.DevicePrimesLastOne.View; DevicePrimesLastOne = lastDigitSharedTables.DevicePrimesLastOne.View;
DevicePrimesLastSeven = lastDigitSharedTables.DevicePrimesLastSeven.View; DevicePrimesLastSeven = lastDigitSharedTables.DevicePrimesLastSeven.View;
DevicePrimesLastThree = lastDigitSharedTables.DevicePrimesLastThree.View; DevicePrimesLastThree = lastDigitSharedTables.DevicePrimesLastThree.View;
DevicePrimesLastNine = lastDigitSharedTables.DevicePrimesLastNine.View; DevicePrimesLastNine = lastDigitSharedTables.DevicePrimesLastNine.View;
DevicePrimesPow2LastOne = lastDigitSharedTables.DevicePrimesPow2LastOne.View; DevicePrimesPow2LastOne = lastDigitSharedTables.DevicePrimesPow2LastOne.View;
DevicePrimesPow2LastSeven = lastDigitSharedTables.DevicePrimesPow2LastSeven.View; DevicePrimesPow2LastSeven = lastDigitSharedTables.DevicePrimesPow2LastSeven.View;
DevicePrimesPow2LastThree = lastDigitSharedTables.DevicePrimesPow2LastThree.View; DevicePrimesPow2LastThree = lastDigitSharedTables.DevicePrimesPow2LastThree.View;
DevicePrimesPow2LastNine = lastDigitSharedTables.DevicePrimesPow2LastNine.View; DevicePrimesPow2LastNine = lastDigitSharedTables.DevicePrimesPow2LastNine.View;
// DivisorTables = HeuristicCombinedGpuTables DivisorTables = HeuristicCombinedGpuTables
// .GetStaticTables(acceleratorIndex) .GetStaticTables(acceleratorIndex)
// .CreateViews(); .CreateViews();
var smallPrimeFactorTables = SmallPrimeFactorGpuTables.GetStaticTables(acceleratorIndex); var smallPrimeFactorTables = SmallPrimeFactorGpuTables.GetStaticTables(acceleratorIndex);
if (smallPrimeFactorTables.HasValue) if (smallPrimeFactorTables.HasValue)
{ {
SmallPrimeFactorGpuTables tables = smallPrimeFactorTables.Value; SmallPrimeFactorGpuTables tables = smallPrimeFactorTables.Value;
SmallPrimeFactorPrimes = tables.Primes.View; SmallPrimeFactorPrimes = tables.Primes.View;
SmallPrimeFactorSquares = tables.Squares.View; SmallPrimeFactorSquares = tables.Squares.View;
} }
var kernels = _kernels[acceleratorIndex]; var kernels = _kernels[acceleratorIndex];
// CalculateOrderKernelLauncher = kernels.CalculateOrderKernelLauncher; // CalculateOrderKernelLauncher = kernels.CalculateOrderKernelLauncher;
CheckFactorsKernel = kernels.CheckFactorsKernel; CheckFactorsKernel = kernels.CheckFactorsKernel;
ConvertToStandardKernelLauncher = kernels.ConvertToStandardKernelLauncher; ConvertToStandardKernelLauncher = kernels.ConvertToStandardKernelLauncher;
HeuristicCombinedTrialDivisionKernelLauncher = kernels.HeuristicCombinedTrialDivisionKernelLauncher; HeuristicCombinedTrialDivisionKernelLauncher = kernels.HeuristicCombinedTrialDivisionKernelLauncher;
// KeepMontgomeryKernelLauncher = kernels.KeepMontgomeryKernelLauncher; // KeepMontgomeryKernelLauncher = kernels.KeepMontgomeryKernelLauncher;
// PartialFactorKernelLauncher = kernels.PartialFactorKernelLauncher; // PartialFactorKernelLauncher = kernels.PartialFactorKernelLauncher;
// PollardRhoKernel = kernels.PollardRhoKernel; // PollardRhoKernel = kernels.PollardRhoKernel;
Pow2ModKernelLauncher = kernels.Pow2ModKernelLauncher; Pow2ModKernelLauncher = kernels.Pow2ModKernelLauncher;
Pow2ModWideKernelLauncher = kernels.Pow2ModWideKernelLauncher; Pow2ModWideKernelLauncher = kernels.Pow2ModWideKernelLauncher;
// SharesFactorKernelLauncher = kernels.SharesFactorKernelLauncher; // SharesFactorKernelLauncher = kernels.SharesFactorKernelLauncher;
// SmallPrimeFactorKernelLauncher = kernels.SmallPrimeFactorKernelLauncher; SmallPrimeFactorKernelLauncher = kernels.SmallPrimeFactorKernelLauncher;
// SmallPrimeSieveKernelLauncher = kernels.SmallPrimeSieveKernelLauncher; // SmallPrimeSieveKernelLauncher = kernels.SmallPrimeSieveKernelLauncher;
SpecialMaxKernelLauncher = kernels.SpecialMaxKernelLauncher; SpecialMaxKernelLauncher = kernels.SpecialMaxKernelLauncher;
// }
} }
public void Dispose() public void Dispose()