mlemanczyk/even-perfect-numbers-scanner
1
0 Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
Files
42447c4e43050df1ddc4433f1b3676e61691ef16
even-perfect-numbers-scanner/PerfectNumbers.Core/Gpu
T
History
mlemanczyk 42447c4e43 BinaryGcd switched to the best performing variant according to benchmarks.
2025-11-14 11:43:16 +01:00
..
Accelerators
BinaryGcd switched to the best performing variant according to benchmarks.
2025-11-14 11:43:16 +01:00
Kernels
BinaryGcd switched to the best performing variant according to benchmarks.
2025-11-14 11:43:16 +01:00
AcceleratorPool.cs
PrimeTesterByLastDigitAccelerator integrated into Pow2MontgomeryGpuCalculator for better performance.
2025-11-11 13:36:21 +01:00
AcceleratorReferenceComparer.cs
Functionality and classes separated. Kernel initialization made dependent on a flag. Cleanup.
2025-11-10 18:35:50 +01:00
DeterministicRandomGpu.cs
Initial PollardRho implementation for GPU. Added pool for accelerator streams shared across the threads. GPU processing performance and stabilization improvements. GpuUInt128.BinaryGcd switched to the version which outperformed other in benchmarks.
2025-11-12 21:18:30 +01:00
ExponentRemainderStepperGpu.cs
GPU stabilization improvements to prevent calculation crashes
2025-11-09 21:52:16 +01:00
GpuConstants.cs
Clarify cycle stepper reset inlining
2025-10-02 13:30:56 +02:00
GpuDivisorPartialData.cs
Use bit masks for divisor heuristics
2025-10-24 19:50:27 +02:00
GpuKernelPool.cs
GpuPrimeWorkLimiter.Acquire() / .Release() removed to save time on the hot-path with billions of calls. We'll rely on AcceleratorStreamPool limits and synchronization.
2025-11-13 17:52:13 +01:00
GpuPrimeWorkLimiter.cs
GpuScratchBuffer / -Pool naively integrated with PriemOrderCalculatorAccelerator (formerly Pow2MontgomeryAccelerator). FactorEntry replaced with 2 arrays - unused field removed. Small cleanup.
2025-11-13 14:41:27 +01:00
GpuRational.cs
Using cleanup
2025-10-17 02:40:51 +02:00
GpuStaticTableInitializer.cs
Functionality and classes separated. Kernel initialization made dependent on a flag. Cleanup.
2025-11-10 18:35:50 +01:00
GpuUInt128.cs
Initial PollardRho implementation for GPU. Added pool for accelerator streams shared across the threads. GPU processing performance and stabilization improvements. GpuUInt128.BinaryGcd switched to the version which outperformed other in benchmarks.
2025-11-12 21:18:30 +01:00
GpuWorkLimiter.cs
Remove temporary by-divisor review documents
2025-10-28 02:01:28 +01:00
HeuristicGpuDivisorTables.cs
PrimeTesterByLastDigitAccelerator integrated into Pow2MontgomeryGpuCalculator for better performance.
2025-11-11 13:36:21 +01:00
InputOutputScratchBuffers.cs
GPU stabilization improvements to prevent calculation crashes
2025-11-09 21:52:16 +01:00
KernelAuditProgress.md
Document GPU kernel audit progress
2025-10-18 17:42:33 +02:00
KernelContainer.cs
Performance improvements
2025-11-13 00:51:32 +01:00
KernelType.cs
Functionality and classes separated. Kernel initialization made dependent on a flag. Cleanup.
2025-11-10 18:35:50 +01:00
MersenneNumberDivisorByDivisorGpuTester.cs
GpuPrimeWorkLimiter.Acquire() / .Release() removed to save time on the hot-path with billions of calls. We'll rely on AcceleratorStreamPool limits and synchronization.
2025-11-13 17:52:13 +01:00
MersenneNumberDivisorGpuTester.cs
GPU stabilization improvements to prevent calculation crashes
2025-11-09 21:52:16 +01:00
MersenneNumberIncrementalGpuTester.cs
GpuPrimeWorkLimiter.Acquire() / .Release() removed to save time on the hot-path with billions of calls. We'll rely on AcceleratorStreamPool limits and synchronization.
2025-11-13 17:52:13 +01:00
MersenneNumberLucasLehmerGpuTester.cs
GpuPrimeWorkLimiter.Acquire() / .Release() removed to save time on the hot-path with billions of calls. We'll rely on AcceleratorStreamPool limits and synchronization.
2025-11-13 17:52:13 +01:00
MersenneNumberOrderGpuTester.cs
GpuPrimeWorkLimiter.Acquire() / .Release() removed to save time on the hot-path with billions of calls. We'll rely on AcceleratorStreamPool limits and synchronization.
2025-11-13 17:52:13 +01:00
MersenneNumberResidueGpuTester.cs
GpuPrimeWorkLimiter.Acquire() / .Release() removed to save time on the hot-path with billions of calls. We'll rely on AcceleratorStreamPool limits and synchronization.
2025-11-13 17:52:13 +01:00
NttGpuMath.cs
GPU stabilization improvements to prevent calculation crashes
2025-11-09 21:52:16 +01:00
Pow2MontgomeryGpuCalculator.cs
GpuPrimeWorkLimiter.Acquire() / .Release() removed to save time on the hot-path with billions of calls. We'll rely on AcceleratorStreamPool limits and synchronization.
2025-11-13 17:52:13 +01:00
Pow2OddPowerTable.cs
Various micro- and medium- optimizations
2025-10-07 22:14:34 +02:00
PrimeOrderGpuHeuristics.cs
BinaryGcd switched to the best performing variant according to benchmarks.
2025-11-14 11:43:16 +01:00
ReadOnlyGpuUInt128.cs
Using cleanup
2025-10-17 02:40:51 +02:00
ResidueAutomatonArgs.cs
Cleanup. Classes & structs extracted to their own files.
2025-11-07 10:34:31 +01:00
ResiduePrimeViews.cs
Cleanup. Classes & structs extracted to their own files.
2025-11-07 10:34:31 +01:00
SharedGpuContext.cs
GPU stabilization improvements to prevent calculation crashes
2025-11-09 21:52:16 +01:00
SmallPrimeFactorViews.cs
GpuContextPool removed, because we share context and accelerator between threads, using streams to separate.
2025-11-07 12:52:20 +01:00