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

Add GPU simplified mul-mod helper for host callers

Browse Source
This commit is contained in:
Marek Lemańczyk committed 2025-10-25 02:20:02 +02:00
1 parent 3e081141e0
commit 03194650c9
2 files changed
+11 -2
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

PerfectNumbers.Core/ULongExtensions.Gpu.cs
+9 -1
View File
@@ -19,7 +19,7 @@ public static partial class ULongExtensions
}
[MethodImpl(MethodImplOptions.AggressiveInlining)]
[Obsolete("Deprecated: use ULongExtensions.MulMod64 for CPU code and GpuUInt128.MulMod inside GPU kernels.")]
[Obsolete("Use ULongExtensions.GpuCompatibleMulModSimplifiedExtension for GPU-compatible host code or GpuUInt128.MulMod inside kernels.")]
public static ulong MulMod64Gpu(this ulong a, ulong b, ulong modulus)
{
// TODO: Remove this GPU-compatible shim from production once callers migrate to MulMod64,
@@ -29,6 +29,7 @@ public static partial class ULongExtensions
}
[MethodImpl(MethodImplOptions.AggressiveInlining)]
[Obsolete("Use ULongExtensions.GpuCompatibleMulModSimplifiedExtension for GPU-compatible host code or GpuUInt128.MulMod inside kernels.")]
public static ulong MulMod64GpuDeferred(this ulong a, ulong b, ulong modulus)
{
// TODO: Move this deferred helper to the benchmark suite; the baseline MulMod64 avoids the
@@ -37,6 +38,13 @@ public static partial class ULongExtensions
return state.MulModWithNativeModulo(b, modulus);
}
[MethodImpl(MethodImplOptions.AggressiveInlining)]
public static ulong GpuCompatibleMulModSimplifiedExtension(this ulong left, ulong right, ulong modulus)
{
GpuUInt128 state = new(left);
return state.MulModSimplified(right, modulus);
}
[MethodImpl(MethodImplOptions.AggressiveInlining)]
public static ulong Pow2ModWindowedGpu(this ulong exponent, ulong modulus)
{
docs/gpu-bydivisor-mulmod-audit.md
+2 -1
View File
@@ -6,7 +6,8 @@ This note tracks every `MulMod` call site in the GPU-only `--mersenne=bydivisor`
| Location | Notes |
| --- | --- |
| `ULongExtensions.MulMod64Gpu` | Marked as deprecated to steer new code toward `MulMod64` on the CPU and `GpuUInt128.MulMod` inside GPU kernels. The shim remains only for legacy callers that require the GPU-compatible arithmetic façade.【F:PerfectNumbers.Core/ULongExtensions.Gpu.cs†L1-L33】 |
| `ULongExtensions.GpuCompatibleMulModSimplifiedExtension` | Wraps `GpuUInt128.MulModSimplified` so host code mirrors the benchmark winner without reimplementing GPU logic.【F:PerfectNumbers.Core/ULongExtensions.Gpu.cs†L23-L31】 |
| `ULongExtensions.MulMod64Gpu` | Marked obsolete; migrate to `ULongExtensions.GpuCompatibleMulModSimplifiedExtension` for GPU-compatible host code or `GpuUInt128.MulMod` when running inside kernels.【F:PerfectNumbers.Core/ULongExtensions.Gpu.cs†L17-L24】 |
| `ULongExtensions.Pow2ModWindowedGpu` | Migrated to call `MulMod64` for every square and multiply, eliminating the 6–18× slowdown that the shim introduced while executing on the host.【F:PerfectNumbers.Core/ULongExtensions.Gpu.cs†L39-L109】 |
| `ULongExtensions.InitializeStandardOddPowers` | Populates the odd-power table through `MulMod64`, so the host setup costs now match the CPU benchmark baseline.【F:PerfectNumbers.Core/ULongExtensions.Gpu.cs†L286-L309】 |