 Marek Lemańczyk
|
01bb731220
|
Optimizations. Unnecessary variables removed.
|
2025-09-24 01:41:54 +02:00 |
|
|
Marek Lemańczyk
|
1978f7dfcb
|
Fix Mod10 shortcut and restore Mod10_8_5_3Steps accuracy
|
2025-09-24 00:32:15 +02:00 |
|
|
Marek Lemańczyk
|
63f3d71ce4
|
Regression issue fixed
|
2025-09-24 00:07:44 +02:00 |
|
|
Marek Lemańczyk
|
773848c577
|
Optimizations
|
2025-09-24 00:01:08 +02:00 |
|
|
Marek Lemańczyk
|
7079d108e7
|
Cleanup
|
2025-09-23 23:29:44 +02:00 |
|
|
Marek Lemańczyk
|
5ce67fa1c5
|
Reduce span slicing in GPU divisor checks
|
2025-09-23 23:28:32 +02:00 |
|
|
Marek Lemańczyk
|
dc22545adb
|
Tiny cleanup
|
2025-09-23 23:22:20 +02:00 |
|
|
Marek Lemańczyk
|
6e99ee39e3
|
Extension methods benchmarked. Slower alternatives replaced.
|
2025-09-23 22:01:09 +02:00 |
|
|
Marek Lemańczyk
|
9becb8df67
|
using cleanup.
|
2025-09-23 22:00:31 +02:00 |
|
|
Marek Lemańczyk
|
7d112c347f
|
Accelerate Mersenne modular multiplication paths
|
2025-09-22 23:24:19 +02:00 |
|
|
Marek Lemańczyk
|
3598f88aaf
|
Replace small-modulo uses with optimized helpers
|
2025-09-22 19:54:50 +02:00 |
|
|
Marek Lemańczyk
|
ca866ffc5b
|
Reapply modulo helpers and inline fast remainders
|
2025-09-22 19:26:07 +02:00 |
|
|
Marek Lemańczyk
|
378d6a2fd8
|
Make by-divisor divisor cycle filtering optional
|
2025-09-22 16:42:27 +02:00 |
|
|
Marek Lemańczyk
|
66d79f7166
|
Add Montgomery multiplication benchmark and optimize implementation
|
2025-09-22 14:04:50 +02:00 |
|
|
Marek Lemańczyk
|
a6571c2487
|
Accelerate by-divisor GPU tester with Montgomery math
|
2025-09-22 13:21:48 +02:00 |
|
|
Marek Lemańczyk
|
d117921c3a
|
Additional checks will be skipped in bydivisor mode to speed up calculations. Console output extended.
|
2025-09-21 14:23:56 +02:00 |
|
|
Marek Lemańczyk
|
198b31f380
|
Console logging extended
|
2025-09-21 13:36:43 +02:00 |
|
|
Marek Lemańczyk
|
458cb77227
|
We'll output to console not only primes, but also composites. Output to console standardized on ConsoleInterval
|
2025-09-21 12:56:30 +02:00 |
|
|
Marek Lemańczyk
|
f44e51c85e
|
Stream by-divisor GPU checks without retaining divisors
|
2025-09-21 12:14:08 +02:00 |
|
|
Marek Lemańczyk
|
3fd39c3d34
|
Wire by-divisor GPU batching to scanner configuration
|
2025-09-21 11:41:51 +02:00 |
|
|
Marek Lemańczyk
|
eb0a5b2655
|
Log by-divisor progress updates
|
2025-09-21 11:11:56 +02:00 |
|
|
Marek Lemańczyk
|
5ce0cf74fc
|
Require filtered primes for by-divisor scanning
|
2025-09-21 11:11:51 +02:00 |
|
|
Marek Lemańczyk
|
b705488b35
|
Document by-divisor option in CLI help
|
2025-09-21 10:59:24 +02:00 |
|
|
Marek Lemańczyk
|
b8b34766ae
|
Merge pull request #53 from mlemanczyk/codex/add-mod3,-mod5,-mod7,-mod8,-mod10-methods
Add optimized modulo helpers and update GPU residue steps
|
2025-09-21 10:14:40 +02:00 |
|
|
Marek Lemańczyk
|
95a0b030f6
|
Add optimized modulo helpers for unsigned integers
|
2025-09-21 10:14:26 +02:00 |
|
|
Marek Lemańczyk
|
860d64baba
|
Tiny cleanup
|
2025-09-21 09:51:05 +02:00 |
|
|
Marek Lemańczyk
|
412ce4fd50
|
% 10 operations replaces with UInt128/ulong.Mod10 methods
|
2025-09-21 07:42:25 +02:00 |
|
|
Marek Lemańczyk
|
53554ffd3d
|
Restored faster version of the calculations
|
2025-09-21 05:40:13 +02:00 |
|
|
Marek Lemańczyk
|
5ab69f8c01
|
Restore struct residue args and add residue benchmark project
|
2025-09-21 05:29:13 +02:00 |
|
|
Marek Lemańczyk
|
4fe701425e
|
Micro-optimizations
|
2025-09-21 04:38:01 +02:00 |
|
|
Marek Lemańczyk
|
f53853f6ff
|
Micro optimization to take ref to orders once and re-use it, until it changes for the last block.
|
2025-09-21 04:27:51 +02:00 |
|
|
Marek Lemańczyk
|
23e8d306b1
|
Breaking change partially reverted
|
2025-09-21 02:48:51 +02:00 |
|
|
Marek Lemańczyk
|
8a6fc0c6a5
|
Revert "Merge pull request #24 from mlemanczyk/codex/modify-evenperfectbitscanner-for-residue-mode"
This reverts commit 2d024cc708, reversing
changes made to b1c29ad16e.
|
2025-09-21 02:25:18 +02:00 |
|
|
Marek Lemańczyk
|
f9b4f912e2
|
Fix GPU residue tester iteration and add regression tests
|
2025-09-21 00:14:33 +02:00 |
|
|
Marek Lemańczyk
|
0ce17d19ae
|
Guard residue order kernels against stale cycle cache entries
|
2025-09-20 23:34:47 +02:00 |
|
|
Marek Lemańczyk
|
7449f146d9
|
Ensure GPU residue copies are synchronized and cover high divisor sets
|
2025-09-20 22:49:37 +02:00 |
|
|
Marek Lemańczyk
|
8e686fbe6e
|
Harden GPU divisor scan buffer handling
|
2025-09-20 21:29:22 +02:00 |
|
|
Marek Lemańczyk
|
7ccac87df3
|
Zero GPU residue result buffer before scanning
|
2025-09-20 19:26:45 +02:00 |
|
|
Marek Lemańczyk
|
abe695788b
|
Serialize GPU leases with per-accelerator locks
|
2025-09-20 18:19:21 +02:00 |
|
|
Marek Lemańczyk
|
5e6e762556
|
Reapply "Merge pull request #34 from mlemanczyk/codex/fix-residue-method-iteration-issue"
This reverts commit 9a199dc258.
|
2025-09-20 10:45:01 +02:00 |
|
|
Marek Lemańczyk
|
b676d89293
|
Merge branch 'main' of https://github.com/mlemanczyk/even-perfect-numbers-scanner
|
2025-09-20 09:41:10 +02:00 |
|
|
Marek Lemańczyk
|
9a199dc258
|
Revert "Merge pull request #34 from mlemanczyk/codex/fix-residue-method-iteration-issue"
This reverts commit 66ac013abc, reversing
changes made to 52a27dfe61.
|
2025-09-20 09:35:10 +02:00 |
|
|
Marek Lemańczyk
|
662cfa8cc7
|
Reset residue toggle after default-limit residue test
|
2025-09-20 09:14:29 +02:00 |
|
|
Marek Lemańczyk
|
ee9c151eaa
|
Fix residue kernel modulus reduction and add regression test
|
2025-09-20 04:24:04 +02:00 |
|
|
Marek Lemańczyk
|
9f2cf12ce1
|
Guard GPU residue and divisor scans against overflow
|
2025-09-20 03:21:49 +02:00 |
|
|
Marek Lemańczyk
|
bc1ec6a3aa
|
Allow residue GPU scanning to cover UInt128 divisors
|
2025-09-20 02:49:20 +02:00 |
|
|
Marek Lemańczyk
|
9a7382fd80
|
Tiny cleanup
|
2025-09-17 19:08:43 +02:00 |
|
|
Marek Lemańczyk
|
1b28584168
|
Add divisor-mode detailed check coverage for factors
|
2025-09-17 19:02:21 +02:00 |
|
|
Marek Lemańczyk
|
34cc0b0fdd
|
Refine GPU residue prime tables
|
2025-09-17 17:35:26 +02:00 |
|
|
Marek Lemańczyk
|
13e048c23d
|
Allow exhaustive divisor scans with 128-bit limits
|
2025-09-15 14:15:35 +02:00 |
|