 Marek Lemańczyk
|
faab60c2cd
|
Reduce host hit detection overhead
|
2025-10-22 23:51:19 +02:00 |
|
|
Marek Lemańczyk
|
4d9158e8f3
|
Make GPU mask compaction branchless with shared staging
|
2025-10-22 22:01:39 +02:00 |
|
|
Marek Lemańczyk
|
ea0770a84a
|
Clarify stream kernel bounds assumptions
|
2025-10-22 21:41:51 +02:00 |
|
|
Marek Lemańczyk
|
586eb8d303
|
Document hot-path guards in GPU mask compaction helpers
|
2025-10-22 18:48:03 +02:00 |
|
|
Marek Lemańczyk
|
add62df697
|
Compact GPU divisor masks on device
|
2025-10-22 18:47:55 +02:00 |
|
|
Marek Lemańczyk
|
bae39ad95e
|
Capture GPU next start for divisor batches
|
2025-10-22 06:43:05 +02:00 |
|
|
Marek Lemańczyk
|
5b973fb513
|
Document and streamline GPU candidate kernel guards
|
2025-10-22 06:43:00 +02:00 |
|
|
Marek Lemańczyk
|
4d8ea9dccd
|
Use GpuUInt128 in GPU divisor candidate generation
|
2025-10-22 06:05:22 +02:00 |
|
|
Marek Lemańczyk
|
bfc5712d81
|
Add GPU kernel for divisor candidate generation
|
2025-10-22 05:24:01 +02:00 |
|
|
Marek Lemańczyk
|
00341fbb8c
|
Simplify GPU by-divisor configuration guards
|
2025-10-22 03:49:04 +02:00 |
|
|
mlemanczyk
|
43cd067e09
|
Formatting
|
2025-10-22 01:47:40 +02:00 |
|
|
Marek Lemańczyk
|
24eaec7a21
|
Refine Montgomery GPU helpers to avoid kernel branching
|
2025-10-22 01:43:10 +02:00 |
|
|
Marek Lemańczyk
|
f12919e220
|
Split Pow2 Montgomery GPU kernels by Montgomery conversion
|
2025-10-22 01:43:02 +02:00 |
|
|
Marek Lemańczyk
|
b08175dadf
|
Simplify Pow2Montgomery GPU executor branching
|
2025-10-22 01:17:35 +02:00 |
|
|
mlemanczyk
|
865a88fac0
|
Switched to GPU for prime testing
|
2025-10-22 00:40:30 +02:00 |
|
|
Marek Lemańczyk
|
3e7606025d
|
Add dedicated GPU context pool for PrimeTester
|
2025-10-21 22:34:30 +02:00 |
|
|
Marek Lemańczyk
|
3afd9399db
|
Isolate residue GPU leases and drop try/finally
|
2025-10-21 22:01:50 +02:00 |
|
|
mlemanczyk
|
40ba538587
|
Switched to range operator from Slice, where applicable.
|
2025-10-21 21:28:55 +02:00 |
|
|
Marek Lemańczyk
|
819cb7b00b
|
Guard residue GPU scanning with execution lock
|
2025-10-21 21:26:53 +02:00 |
|
|
Marek Lemańczyk
|
5dd9eb636d
|
Inline GPU divisor locking adjustments
|
2025-10-21 21:19:57 +02:00 |
|
|
Marek Lemańczyk
|
9bcf1bd9a1
|
Merge pull request #218 from mlemanczyk/codex/analyze-mersenne=bydivisor-path-for-tracking-remainder-bbul2g
Reuse stack spans across divisor processing loops
|
2025-10-21 20:54:45 +02:00 |
|
|
Marek Lemańczyk
|
a226ae10cb
|
Stop clearing pooled buffers inside rent helpers
|
2025-10-21 20:52:21 +02:00 |
|
|
Marek Lemańczyk
|
7baf3b619f
|
Simplify exception guards in CPU remainder paths
|
2025-10-21 20:52:14 +02:00 |
|
|
Marek Lemańczyk
|
6adf78faf6
|
Streamline CPU remainder buffers
|
2025-10-21 20:52:09 +02:00 |
|
|
Marek Lemańczyk
|
b418d79fb3
|
Adopt fixed stack capacities for CPU prime order helpers
|
2025-10-21 19:46:59 +02:00 |
|
|
Marek Lemańczyk
|
c15405e3b1
|
Reuse stack buffers when lowering divisor exponents
|
2025-10-21 18:22:56 +02:00 |
|
|
mlemanczyk
|
31a0b8d706
|
Added static IsPrimeGpu
|
2025-10-21 04:09:26 +02:00 |
|
|
mlemanczyk
|
ea11db6d0f
|
Restored the Dictionary parameter instead of arrays
|
2025-10-21 04:08:48 +02:00 |
|
|
mlemanczyk
|
8dc7751084
|
Switched back to PrimeTester.IsPrimeInternal for teseting
|
2025-10-21 04:07:49 +02:00 |
|
|
Marek Lemańczyk
|
8acccce471
|
Add cycle cache hit tracking instrumentation
|
2025-10-20 19:28:06 +02:00 |
|
|
Marek Lemańczyk
|
1a4818c794
|
Add diagnostics for removed factorization cache
|
2025-10-20 19:22:02 +02:00 |
|
|
Marek Lemańczyk
|
28ff9292b8
|
Remove per-thread factorization cache
|
2025-10-20 19:08:44 +02:00 |
|
|
Marek Lemańczyk
|
20f30dc182
|
Merge pull request #215 from mlemanczyk/codex/analyze-cpu-paths-in-evenperfectbitscanner
Adjust CPU pow2 Montgomery helper for small divisors
|
2025-10-20 18:41:38 +02:00 |
|
|
Marek Lemańczyk
|
33f9b51593
|
Clarify residue stepper usage and update TODO tracker
|
2025-10-20 18:37:35 +02:00 |
|
|
Marek Lemańczyk
|
da6ea25729
|
Address windowed pow2 review feedback
|
2025-10-20 18:37:29 +02:00 |
|
|
Marek Lemańczyk
|
481d1010e8
|
Retire outdated windowed pow2 TODOs
|
2025-10-20 18:37:21 +02:00 |
|
|
Marek Lemańczyk
|
360a31bd54
|
Align Lucas-Lehmer GPU helpers with windowed pow2 guidance
|
2025-10-20 18:37:15 +02:00 |
|
|
Marek Lemańczyk
|
82e44e3f3f
|
Confirm windowed pow2 usage across GPU kernels
|
2025-10-20 18:37:08 +02:00 |
|
|
Marek Lemańczyk
|
7cbc27d6c3
|
Use adaptive pow2 stepping in residue stepper
|
2025-10-20 18:36:58 +02:00 |
|
|
Marek Lemańczyk
|
4831fc84d1
|
Clean up resolved windowed pow2 TODOs on CPU paths
|
2025-10-20 18:36:51 +02:00 |
|
|
Marek Lemańczyk
|
798ae455b1
|
Adjust CPU Pow2 Montgomery helper for small divisors
|
2025-10-20 17:03:36 +02:00 |
|
|
mlemanczyk
|
5d5fb28765
|
Cache removed. It was barely every hit.
|
2025-10-20 16:56:49 +02:00 |
|
|
Marek Lemańczyk
|
9540a9935d
|
Restore cycle remainder TODO comments
|
2025-10-20 16:06:46 +02:00 |
|
|
Marek Lemańczyk
|
ecc718bc69
|
Merge pull request #213 from mlemanczyk/codex/check-and-complete-todos-in-mulmod-benchmarks
Inline GpuUInt128 Mul64 helper
|
2025-10-20 13:58:34 +02:00 |
|
|
Marek Lemańczyk
|
de8e769f48
|
Switch GpuUInt128 internals to in-place shifts
|
2025-10-20 13:58:16 +02:00 |
|
|
Marek Lemańczyk
|
50d2e462d7
|
Use in-place shifts in GpuUInt128.MulMod
|
2025-10-20 13:58:10 +02:00 |
|
|
Marek Lemańczyk
|
d0c9389bcc
|
Document GpuUInt128 MulMod shift benchmarking
|
2025-10-20 13:58:04 +02:00 |
|
|
Marek Lemańczyk
|
1809ba95b3
|
Inline GpuUInt128 Mul64 helper
|
2025-10-20 13:30:34 +02:00 |
|
|
Marek Lemańczyk
|
5fbbfd7590
|
Keep TaskThreadPool workers idle until work arrives
|
2025-10-20 12:15:31 +02:00 |
|
|
Marek Lemańczyk
|
ea2476c749
|
Clarify modular helper invariants
|
2025-10-20 11:52:04 +02:00 |
|