Files
even-perfect-numbers-scanner/PerfectNumbers.Core/Gpu/Accelerators/Pow2MontgomeryAccelerator.cs
T

98 lines
3.7 KiB
C#

using System.Runtime.CompilerServices;
using ILGPU;
using ILGPU.Runtime;
using PerfectNumbers.Core.Gpu.Kernels;
namespace PerfectNumbers.Core.Gpu.Accelerators;
public sealed class Pow2MontgomeryAccelerator
{
#region Pool
[ThreadStatic]
private static Queue<Pow2MontgomeryAccelerator>? _pool;
private static AcceleratorPool _accelerators = new(PerfectNumberConstants.RollingAccelerators);
[MethodImpl(MethodImplOptions.AggressiveInlining)]
public static Pow2MontgomeryAccelerator Rent()
{
var pool = _pool ??= [];
if (!pool.TryDequeue(out var gpu))
{
Accelerator accelerator = _accelerators.Rent();
gpu = new(accelerator, PerfectNumberConstants.DefaultFactorsBuffer);
}
else
{
gpu.Stream = gpu.Accelerator.CreateStream();
}
return gpu;
}
[MethodImpl(MethodImplOptions.AggressiveInlining)]
public static void Return(Pow2MontgomeryAccelerator gpu)
{
gpu.Stream!.Dispose();
gpu.Stream = null;
_pool!.Enqueue(gpu);
}
#endregion
public readonly Accelerator Accelerator;
public readonly MemoryBuffer1D<ulong, Stride1D.Dense> Input;
public readonly MemoryBuffer1D<ulong, Stride1D.Dense> Output;
public MemoryBuffer1D<KeyValuePair<ulong, int>, Stride1D.Dense> Pow2ModEntriesToTestOnDevice;
public Dictionary<ulong, int> Pow2ModEntriesToTestOnHost = [];
public AcceleratorStream? Stream;
public readonly Action<AcceleratorStream, Index1D, ArrayView1D<ulong, Stride1D.Dense>, MontgomeryDivisorData, ArrayView1D<ulong, Stride1D.Dense>> ConvertToStandardKernel;
public readonly Action<AcceleratorStream, Index1D, ArrayView1D<ulong, Stride1D.Dense>, MontgomeryDivisorData, ArrayView1D<ulong, Stride1D.Dense>> KeepMontgomeryKernel;
public readonly Kernel CheckFactorsKernel;
public void EnsureCapacity(int factorsCount)
{
if (factorsCount > Pow2ModEntriesToTestOnDevice.Length)
{
Accelerator accelerator = Accelerator;
Pow2ModEntriesToTestOnDevice.Dispose();
Pow2ModEntriesToTestOnDevice = accelerator.Allocate1D<KeyValuePair<ulong, int>>(factorsCount);
}
}
public Pow2MontgomeryAccelerator(Accelerator accelerator, int factorsCount)
{
Accelerator = accelerator;
Stream = accelerator.CreateStream();
Input = accelerator.Allocate1D<ulong>(1);
Output = accelerator.Allocate1D<ulong>(1);
Pow2ModEntriesToTestOnDevice = accelerator.Allocate1D<KeyValuePair<ulong, int>>(factorsCount);
var keepKernel = accelerator.LoadAutoGroupedStreamKernel<Index1D, ArrayView1D<ulong, Stride1D.Dense>, MontgomeryDivisorData, ArrayView1D<ulong, Stride1D.Dense>>(Pow2MontgomeryKernels.Pow2MontgomeryKernelKeepMontgomery);
var keepLauncher = KernelUtil.GetKernel(keepKernel);
KeepMontgomeryKernel = keepLauncher.CreateLauncherDelegate<Action<AcceleratorStream, Index1D, ArrayView1D<ulong, Stride1D.Dense>, MontgomeryDivisorData, ArrayView1D<ulong, Stride1D.Dense>>>();
var convertKernel = accelerator.LoadAutoGroupedStreamKernel<Index1D, ArrayView1D<ulong, Stride1D.Dense>, MontgomeryDivisorData, ArrayView1D<ulong, Stride1D.Dense>>(Pow2MontgomeryKernels.Pow2MontgomeryKernelConvertToStandard);
var convertLauncher = KernelUtil.GetKernel(convertKernel);
ConvertToStandardKernel = convertLauncher.CreateLauncherDelegate<Action<AcceleratorStream, Index1D, ArrayView1D<ulong, Stride1D.Dense>, MontgomeryDivisorData, ArrayView1D<ulong, Stride1D.Dense>>>();
var checkFactorsKernel = accelerator.LoadStreamKernel<int, ulong, ArrayView1D<KeyValuePair<ulong, int>, Stride1D.Dense>, MontgomeryDivisorData, ArrayView1D<ulong, Stride1D.Dense>>(PrimeOrderGpuHeuristics.CheckFactorsKernel);
CheckFactorsKernel = KernelUtil.GetKernel(checkFactorsKernel);
}
public void Dispose()
{
Input.Dispose();
Output.Dispose();
Stream?.Dispose();
Pow2ModEntriesToTestOnDevice.Dispose();
// Accelerator is shared with other threads
// Accelerator.Dispose();
}
}