 mlemanczyk
|
e8cca74f68
|
DeeterministicRandom types optimized.
|
2025-12-13 16:00:28 +01:00 |
|
|
mlemanczyk
|
15bcde1776
|
SemaphoreSlim replaced with PollingSemaphore.
|
2025-12-12 03:41:13 +01:00 |
|
|
mlemanczyk
|
49e47b161d
|
Tiny cleanup. Namespace updated.
|
2025-12-12 00:56:33 +01:00 |
|
|
mlemanczyk
|
43376745f5
|
Benchmarks to compare Interlocked with Atomic added. Atomic replaced with Interlocked on CPU path as better performer.
|
2025-12-11 23:45:01 +01:00 |
|
|
mlemanczyk
|
260aa19e44
|
Pools renamed to FixedCapacityPools. Other fixed capacity collections added. Queues, concurrent queues and standard stacks replaced with fixed capacity stacks. MethodImplOptions.AggressiveInlining and .AggressiveOptimization added to simple methods and getters / setters. Variables optimized in few places.
|
2025-12-11 21:10:19 +01:00 |
|
|
mlemanczyk
|
5df2cc7f9b
|
using cleaned up.
|
2025-12-10 23:59:04 +01:00 |
|
|
mlemanczyk
|
05d3061c47
|
MethodImplOptions.AggressiveOptimization & MethodImplOptions.AggressiveInlining optimized. Small optimizations.
|
2025-12-10 19:59:51 +01:00 |
|
|
mlemanczyk
|
7a242b3bfa
|
PrimeOrderCalculator.PartialFactorCpu simplified to work only with spans and throw exception if the number of factors is exceeded. PrimeOrderMode removed in favor of bool flag.
|
2025-12-10 12:47:25 +01:00 |
|
|
mlemanczyk
|
6e68252279
|
Status update improvement.
|
2025-12-08 02:53:27 +01:00 |
|
|
mlemanczyk
|
6887111128
|
BigInteger-friendly steppers added. MersenneResidueStepper converted to struct. Manual residue tracking replaced with steppers. Modulo operations replaced with steppers.
|
2025-12-07 19:16:45 +01:00 |
|
|
mlemanczyk
|
86f20ed083
|
Support for q > 2^64 added in --mersenne=bydivisor mode
|
2025-12-05 23:33:21 +01:00 |
|
|
mlemanczyk
|
928e847f1f
|
AscendingdivisorRemainderStepper used in PrimeTesterByLastDigit
|
2025-12-05 14:17:24 +01:00 |
|
|
mlemanczyk
|
86192a71b0
|
Divisor checking optimized to only flag if divisor was found or not without returning complete array of hits. We'll now also track residues for 13, 17 & 19 as the most frequent ones.
|
2025-12-03 00:57:25 +01:00 |
|
|
mlemanczyk
|
85a90c70bf
|
--gpu-ration command-line argument added. AscendingDivisorRemainderStepper implemented in PrimeTester.IsPrimeCpu. PrimeOrderCalculator.RunOnCpu adjusted for low GPU ratios to prevent infinite increase of the counter.
|
2025-12-02 23:36:18 +01:00 |
|
|
mlemanczyk
|
f43c67c86f
|
Tasks will now save calculations progress and resume calculation upon restart.
|
2025-12-02 00:42:38 +01:00 |
|
|
mlemanczyk
|
941733905b
|
Hybrid device mode implemented as the current --mersenne=bydivisor CPU execution path. CPU path is now limited to CPU calculations, only.
|
2025-12-01 03:45:39 +01:00 |
|
|
mlemanczyk
|
c311b27dcc
|
Prime testing using divisor tables depending on the last digit separated to its own class. PrimeTester switched back to standard endings.
|
2025-11-27 16:45:02 +01:00 |
|
|
mlemanczyk
|
a296e62d4d
|
--min-k command-line argument added. Partial factoring switched back to CPU.
|
2025-11-26 22:03:52 +01:00 |
|
|
mlemanczyk
|
13498dc460
|
Most device buffers integrated into PrimeOrderCalculatorAccelerator. PrimeOrderCalculator switched to partial factoring on GPU.
|
2025-11-26 03:16:39 +01:00 |
|
|
mlemanczyk
|
bacb2a1d18
|
Cleanup. Unnecessary GC collection removed. It doesn't limit RAM usage.
|
2025-11-25 01:30:59 +01:00 |
|
|
mlemanczyk
|
b9ca95bf8c
|
PrimeOrderCalculatorAccelerator.Rent is now created outside of the loops and re-used for all calculations.
|
2025-11-24 22:51:53 +01:00 |
|
|
mlemanczyk
|
57a50e2a37
|
Pow2Mod kernel specific methods moved to Pow2MontgomeryKernels.
|
2025-11-24 18:04:07 +01:00 |
|
|
mlemanczyk
|
070e2a14fd
|
MontgomeryDivisorData converted to a pooled class to limit GC pressure. MontgomeryDivisorDataGpu added for GPU operations. PrimeOrderCalculator kernels precompiled during warmup.
|
2025-11-24 15:14:31 +01:00 |
|
|
mlemanczyk
|
07efd5d2a1
|
Accelerator locks removed
|
2025-11-23 14:59:29 +01:00 |
|
|
mlemanczyk
|
6d1ff1372e
|
Factoring switched to PrimeTester.IsPrime for the best performance. PrimeOrderCalculator.PartialFactor switched to GPU factorization. Every so often it will send work to GPU instead of CPU. BatchedResources renamed to MersenneNumberDivisorByDivisorAccelerator for consistency and moved to its own file. It now uses array instead of dictionary for lookup free performance.
|
2025-11-23 00:07:03 +01:00 |
|
|
mlemanczyk
|
497b1fcf89
|
Kernels replaced with pre-compiled actions for better performance
|
2025-11-20 03:00:11 +01:00 |
|
|
mlemanczyk
|
9939c5d562
|
Divisor will now be saved to CSV file
|
2025-11-19 22:14:28 +01:00 |
|
|
mlemanczyk
|
5cac51479e
|
GPU memory allocations protected with lock to prevent shared memory access issues. GPU usage limited for the best performance.
|
2025-11-19 16:49:01 +01:00 |
|
|
mlemanczyk
|
a5696ca17f
|
App crash fix related to unassigned value
|
2025-11-18 06:02:22 +01:00 |
|
|
mlemanczyk
|
56be12ab95
|
Cleanup & compilation fixes.
|
2025-11-18 04:45:18 +01:00 |
|
|
mlemanczyk
|
bc51962f9d
|
Cleanup
|
2025-11-18 04:37:11 +01:00 |
|
|
mlemanczyk
|
801b1be3e3
|
Cleanup
|
2025-11-18 04:22:39 +01:00 |
|
|
mlemanczyk
|
66ca969929
|
Naming improvements and small cleanup
|
2025-11-18 03:43:52 +01:00 |
|
|
mlemanczyk
|
bf7c4bab39
|
Accelerator buffers integrated
|
2025-11-18 01:13:12 +01:00 |
|
|
mlemanczyk
|
342eff2726
|
GpuKernelPool.GetOrAddKernels removed in --mersenne=bydivisor mode. PartialFactorResult.WithAdditionalPrime will now modify the current instance instead of creating new copy for better performance. SmallPrimeFactorViews renamed to SmallPrimeFactorGpuTables for consistency with other static tables.
|
2025-11-17 23:33:59 +01:00 |
|
|
mlemanczyk
|
2fc74cf98d
|
Cleanup
|
2025-11-17 20:12:00 +01:00 |
|
|
mlemanczyk
|
f630bfb598
|
HeuristicCombinedPrimeTesterAccelerator integrated with PrimeOrderCalculatorAccelerator.
|
2025-11-17 17:38:23 +01:00 |
|
|
mlemanczyk
|
183b147135
|
Heuristic tables split into 2 groups
|
2025-11-17 16:04:27 +01:00 |
|
|
mlemanczyk
|
c413ee1162
|
Revert "Stream handling simplified to avoid dictionary lookups in favor of array lookups with known indexes."
This reverts commit a84445ef2a.
|
2025-11-16 17:14:48 +01:00 |
|
|
mlemanczyk
|
a84445ef2a
|
Stream handling simplified to avoid dictionary lookups in favor of array lookups with known indexes.
|
2025-11-16 15:01:51 +01:00 |
|
|
mlemanczyk
|
cd7636ddbe
|
Manual accelerator stream creations replaced with pool
|
2025-11-15 13:45:24 +01:00 |
|
|
mlemanczyk
|
0540411cad
|
Accelerator index added for dictionary-less use-cases
|
2025-11-15 02:21:45 +01:00 |
|
|
mlemanczyk
|
6cb63b6e73
|
Internal accelerator array removed
|
2025-11-14 12:03:31 +01:00 |
|
|
mlemanczyk
|
42447c4e43
|
BinaryGcd switched to the best performing variant according to benchmarks.
|
2025-11-14 11:43:16 +01:00 |
|
|
mlemanczyk
|
be435b7760
|
Kernels made part of the accelerators for better performance.
|
2025-11-13 18:52:46 +01:00 |
|
|
mlemanczyk
|
0d2a88c45f
|
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 |
|
|
mlemanczyk
|
c53aa517f7
|
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 |
|
|
mlemanczyk
|
43863d03fb
|
Performance improvements
|
2025-11-13 00:51:32 +01:00 |
|
|
mlemanczyk
|
5bc65464d0
|
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 |
|
|
mlemanczyk
|
fdc2da7dae
|
Accelerator leak fixed.
|
2025-11-11 16:40:44 +01:00 |
|