 Marek Lemańczyk
|
93a82f756f
|
Merge pull request #99 from mlemanczyk/codex/add-benchmarks-for-mulmod64-variations
Extend MulMod64 benchmarks with operand reduction variants
|
2025-09-30 15:31:43 +02:00 |
|
|
Marek Lemańczyk
|
8c877cecb3
|
Add operand reduction benchmark variants for MulMod64
|
2025-09-30 15:31:20 +02:00 |
|
|
Marek Lemańczyk
|
bc5e98450a
|
Merge pull request #98 from mlemanczyk/codex/analyze-recent-performance-improvements
Refine Mul64 benchmarks with GPU-friendly variant
|
2025-09-30 04:24:52 +02:00 |
|
|
Marek Lemańczyk
|
ebe3b306b9
|
Refine Mul64 benchmarks with GPU-friendly variant
|
2025-09-30 04:24:40 +02:00 |
|
|
Marek Lemańczyk
|
daab18504b
|
Performance improvements
|
2025-09-30 03:21:51 +02:00 |
|
|
Marek Lemańczyk
|
28ee92b1b9
|
Performance improvements
|
2025-09-30 03:12:33 +02:00 |
|
|
Marek Lemańczyk
|
e6c8cb35e7
|
Benchmark limits removed to make it accurate
|
2025-09-29 23:29:12 +02:00 |
|
|
Marek Lemańczyk
|
0bdf85730d
|
Tiny optimization
|
2025-09-29 23:28:35 +02:00 |
|
|
Marek Lemańczyk
|
79cb8823ad
|
Merge pull request #97 from mlemanczyk/codex/add-benchmarks-for-ulongextensions.mulhigh
Refine GPU Mul64 helpers for mutation
|
2025-09-29 22:51:07 +02:00 |
|
|
Marek Lemańczyk
|
f63254f0f0
|
Refactor GPU-compatible Mul64 handling
|
2025-09-29 22:35:46 +02:00 |
|
|
Marek Lemańczyk
|
b677e6cdec
|
Merge pull request #96 from mlemanczyk/codex/modify-evenperfectbitscanner-for-mersenne-mode
Extract by-divisor scanner into shared tester
|
2025-09-29 20:49:22 +02:00 |
|
|
Marek Lemańczyk
|
1ba04047c1
|
Extract by-divisor scanner into shared tester
|
2025-09-29 20:49:01 +02:00 |
|
|
Marek Lemańczyk
|
e9270d16f3
|
Merge pull request #95 from mlemanczyk/codex/modify-evenperfectbitscanner-to-skip-candidates
Skip previously excluded by-divisor candidates
|
2025-09-29 14:46:53 +02:00 |
|
|
Marek Lemańczyk
|
4d2c70da8b
|
Skip previously excluded by-divisor candidates
|
2025-09-29 14:46:37 +02:00 |
|
|
Marek Lemańczyk
|
fc74f2b59b
|
Merge pull request #94 from mlemanczyk/codex/check-current-benchmarks-for-montgomery-method
Document cycle computation benchmark results
|
2025-09-25 23:24:11 +02:00 |
|
|
Marek Lemańczyk
|
da4dec2369
|
Document cycle computation benchmark results
|
2025-09-25 23:23:54 +02:00 |
|
|
Marek Lemańczyk
|
06061862db
|
Merge pull request #93 from mlemanczyk/codex/implement-gpu-optimizations-for-evenperfectbitscanner
Clarify GPU divisor batch guard rationale
|
2025-09-25 22:59:10 +02:00 |
|
|
Marek Lemańczyk
|
7ef8a304e9
|
Clarify GPU batch capacity guards
|
2025-09-25 22:58:57 +02:00 |
|
|
Marek Lemańczyk
|
ab3f39ec0d
|
Merge pull request #92 from mlemanczyk/codex/refactor-cycleremainderstepper-and-mersennenumberdivisor
Initialize cycle remainder stepping upfront
|
2025-09-25 21:10:22 +02:00 |
|
|
Marek Lemańczyk
|
9958338820
|
Initialize cycle remainder stepping upfront
|
2025-09-25 21:09:28 +02:00 |
|
|
Marek Lemańczyk
|
699612f4f9
|
Merge pull request #91 from mlemanczyk/codex/refactor-mersenne-bydivisor-method
Trim GPU exponent batches using cached divisor cycles
|
2025-09-25 20:23:41 +02:00 |
|
|
Marek Lemańczyk
|
9b7611fe62
|
Trim GPU exponent batches using cached divisor cycles
|
2025-09-25 20:23:25 +02:00 |
|
|
Marek Lemańczyk
|
35691ff710
|
Merge pull request #90 from mlemanczyk/codex/optimize-memory-usage-in-evenperfectbitscanner
Optimize cycle remainder reuse in by-divisor testers
|
2025-09-25 18:52:06 +02:00 |
|
|
Marek Lemańczyk
|
9b9c6959be
|
Reuse cycle remainder stepper in by-divisor testers
|
2025-09-25 18:51:47 +02:00 |
|
|
Marek Lemańczyk
|
d91507c52e
|
Merge pull request #89 from mlemanczyk/codex/optimize-buildprimebuffer-method
Reduce synchronization when building divisor prime buffer
|
2025-09-25 14:28:06 +02:00 |
|
|
Marek Lemańczyk
|
0f3b354a93
|
Reduce synchronization when building divisor prime buffer
|
2025-09-25 14:27:28 +02:00 |
|
|
Marek Lemańczyk
|
b40c61b527
|
Tiny optimizations
|
2025-09-25 12:43:38 +02:00 |
|
|
Marek Lemańczyk
|
05077f7e90
|
Fixed regression with ref taken only once while it may move since..
|
2025-09-25 10:31:34 +02:00 |
|
|
Marek Lemańczyk
|
5bfc9b1c30
|
Merge pull request #88 from mlemanczyk/codex/fix-divisor-cycle-handling-in-gpu-mode
Filter disallowed divisors when scanning by divisor
|
2025-09-25 10:04:59 +02:00 |
|
|
Marek Lemańczyk
|
c28e36cbe5
|
Filter disallowed divisors when scanning by divisor
|
2025-09-25 10:04:38 +02:00 |
|
|
Marek Lemańczyk
|
5b2995c8a5
|
Divisor cycles added for testing with Codex
|
2025-09-25 09:39:10 +02:00 |
|
|
Marek Lemańczyk
|
0959b471b9
|
Merge branch 'main' of https://github.com/mlemanczyk/even-perfect-numbers-scanner
|
2025-09-25 09:38:14 +02:00 |
|
|
Marek Lemańczyk
|
b792eb90e0
|
Small optimization
|
2025-09-25 09:35:02 +02:00 |
|
|
Marek Lemańczyk
|
756be9a040
|
Merge pull request #87 from mlemanczyk/codex/add-alternative-methods-for-pow2montgomerymod
Move pow2 variants into benchmark harness and add cycle comparison
|
2025-09-25 09:33:30 +02:00 |
|
|
Marek Lemańczyk
|
a225cbd117
|
Move pow2 variants into benchmark harness and add cycle comparison
|
2025-09-25 09:33:16 +02:00 |
|
|
Marek Lemańczyk
|
74747adc7c
|
Performance improvements
|
2025-09-25 08:29:27 +02:00 |
|
|
Marek Lemańczyk
|
845ce41dd8
|
Tiny cleanup and performance improvements
|
2025-09-25 08:28:28 +02:00 |
|
|
Marek Lemańczyk
|
8b95a42f27
|
Divisor cycle cache will now respect --divisor-cycles-batch parameter (renamed from --divisor-cycles-batch-size)
|
2025-09-25 07:02:44 +02:00 |
|
|
Marek Lemańczyk
|
d7f2dbbec6
|
Formatting
|
2025-09-25 06:11:02 +02:00 |
|
|
Marek Lemańczyk
|
8c3e9cb76f
|
Merge pull request #86 from mlemanczyk/codex/implement-todo-for-cpu-prime-residue-calculations
Move cycle filtering and residue prep to CPU
|
2025-09-25 06:10:03 +02:00 |
|
|
Marek Lemańczyk
|
62e4b31f35
|
Adjust divisor cycle cache usage after lease removal
|
2025-09-25 06:09:50 +02:00 |
|
|
Marek Lemańczyk
|
d95195e0c6
|
Move cycle filtering and residue prep to CPU
|
2025-09-25 05:56:56 +02:00 |
|
|
Marek Lemańczyk
|
34681b9738
|
Small optimizations and more TODOs added.
|
2025-09-25 05:49:32 +02:00 |
|
|
Marek Lemańczyk
|
49fcd93747
|
GPU logic updated to use prime % divisor result precomputed on CPU instead of re-calculating it on GPU side. Cycle lease removed.
|
2025-09-25 05:28:23 +02:00 |
|
|
Marek Lemańczyk
|
a171e57353
|
Removed useless status method
|
2025-09-25 04:41:47 +02:00 |
|
|
Marek Lemańczyk
|
26567bbdaf
|
Formatting
|
2025-09-25 04:21:53 +02:00 |
|
|
Marek Lemańczyk
|
3fb0524259
|
Merge pull request #85 from mlemanczyk/codex/implement-remaining-todo-items-in-evenperfectbitscanner
Address review feedback on divisor processing
|
2025-09-25 04:21:19 +02:00 |
|
|
Marek Lemańczyk
|
e876a9f2eb
|
Remove redundant sort and CPU divisor struct
|
2025-09-25 04:21:02 +02:00 |
|
|
Marek Lemańczyk
|
b58e2be2b0
|
GPU steps adjusted to prevent overloading it
|
2025-09-25 03:32:07 +02:00 |
|
|
Marek Lemańczyk
|
45b0d60a71
|
Merge pull request #84 from mlemanczyk/codex/fix-gpu-driver-overload-issue
Chunk GPU divisor cycle generation to avoid CPU fallback
|
2025-09-25 03:17:05 +02:00 |
|