 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
|
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
|
07efd5d2a1
|
Accelerator locks removed
|
2025-11-23 14:59:29 +01:00 |
|
|
mlemanczyk
|
801b1be3e3
|
Cleanup
|
2025-11-18 04:22:39 +01:00 |
|
|
mlemanczyk
|
0540411cad
|
Accelerator index added for dictionary-less use-cases
|
2025-11-15 02:21:45 +01:00 |
|
|
mlemanczyk
|
d0b1639a5d
|
GPU stabilization improvements to prevent calculation crashes
|
2025-11-09 21:52:16 +01:00 |
|
|
mlemanczyk
|
f4e365194d
|
GpuContextPool removed, because we share context and accelerator between threads, using streams to separate.
|
2025-11-07 12:52:20 +01:00 |
|
|
mlemanczyk
|
de2a78b27b
|
Streams added to all kernels and GPU operations to separate threads from each other.
|
2025-11-06 18:53:55 +01:00 |
|
|
mlemanczyk
|
64d615c402
|
Various optimizations to separate non-static data between threads and resolve read/write access issue.
|
2025-11-04 20:45:56 +01:00 |
|
|
mlemanczyk
|
07be671726
|
Various small optimizations
|
2025-11-02 11:24:18 +01:00 |
|
|
mlemanczyk
|
2281d9535e
|
Revert "Merge pull request #251 from mlemanczyk/codex/refactor-prime-checking-on-gpu-n5t2pm"
This reverts commit 936ce239db, reversing
changes made to e14b77e504.
|
2025-11-02 11:22:14 +01:00 |
|
|
mlemanczyk
|
22f20bfe98
|
Various small optimizations
|
2025-11-01 03:42:36 +01:00 |
|
|
mlemanczyk
|
0d7df2e45b
|
Unnecessary memory zeroing commented out. TODOs added for the remainders.
|
2025-10-23 19:42:03 +02:00 |
|
|
Marek Lemańczyk
|
6217ea7888
|
Document GPU kernel audit progress
|
2025-10-18 17:42:33 +02:00 |
|
|
Marek Lemańczyk
|
6918bd28ff
|
Make by-divisor GPU kernels reuse read-only operands
|
2025-10-08 18:47:16 +02:00 |
|
|
Marek Lemańczyk
|
20120f5bc0
|
Fix odd power table updates in GPU heuristics
|
2025-10-05 16:55:27 +02:00 |
|
|
Marek Lemańczyk
|
b8aa845c5c
|
Clarify cycle stepper reset inlining
|
2025-10-02 13:30:56 +02:00 |
|
|
Marek Lemańczyk
|
36d316d19c
|
Add TODOs for divisor metadata caching
|
2025-10-01 22:52:21 +02:00 |
|
|
Marek Lemańczyk
|
9b9c6959be
|
Reuse cycle remainder stepper in by-divisor testers
|
2025-09-25 18:51:47 +02:00 |
|
|
Marek Lemańczyk
|
d4dcd0a534
|
Implement multi-tier divisor cycle caching and per-thread reuse
|
2025-09-24 19:04:07 +02:00 |
|
|
Marek Lemańczyk
|
9becb8df67
|
using cleanup.
|
2025-09-23 22:00:31 +02:00 |
|
|
Marek Lemańczyk
|
860d64baba
|
Tiny cleanup
|
2025-09-21 09:51:05 +02:00 |
|
|
Marek Lemańczyk
|
8e686fbe6e
|
Harden GPU divisor scan buffer handling
|
2025-09-20 21:29:22 +02:00 |
|
|
Marek Lemańczyk
|
9f2cf12ce1
|
Guard GPU residue and divisor scans against overflow
|
2025-09-20 03:21:49 +02:00 |
|
|
Marek Lemańczyk
|
13e048c23d
|
Allow exhaustive divisor scans with 128-bit limits
|
2025-09-15 14:15:35 +02:00 |
|
|
Marek Lemańczyk
|
58eb65ca9e
|
Refine GPU residue utilities and add positive divisor tests
|
2025-09-13 16:47:38 +02:00 |
|
|
Marek Lemańczyk
|
dd6d9a71cf
|
Stack overflow fixed
|
2025-09-12 10:59:58 +02:00 |
|
|
Marek Lemańczyk
|
0535ad2b7b
|
BuildDivisorCandidates moved to MersenneNumberDivisorGpuTester. Micro-optimizations.
|
2025-09-11 21:21:42 +02:00 |
|
|
Marek Lemańczyk
|
11bd89d8d0
|
Unnecessary local variables removed. We'll now reuse local variables instead of declaring new one, where possible. resultBuffer.Dispose() called explicitly before releasing gpu context.
|
2025-09-11 18:57:49 +02:00 |
|
|
Marek Lemańczyk
|
ab856c4995
|
Make GpuUInt128 mutable and update callers
|
2025-09-11 18:38:42 +02:00 |
|
|
Marek Lemańczyk
|
f17f83e326
|
No. of divisor cycle iterations made configurable
|
2025-09-11 01:20:15 +02:00 |
|
|
Marek Lemańczyk
|
596bb7e38f
|
Add GPU divisor tester
|
2025-09-10 22:03:17 +02:00 |
|