Repository navigation
Conversation
This was referenced Nov 13, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Scoped written proof with an exact finite exclusion certificate, produced by Nick Petalas. The ordinary four-cycle
(x0,x1,x2,x3) -> (x1,x2,x3,x0)costs exactly ten scalar primitives, with any finite number of arbitrary dirty donors restored individually.The alphabet is distinct-register ADD/SUB, NEG, DOUBLE/HALVE, COPY and CLEAR, with no free signs, renamings, clean allocation or discarded coordinates. The new step is the ordinary-four-cycle endpoint exclusion and its unbounded-donor consequence. Credit Philipp / mightyOil #1303 for the cycle-sensitive reduction and two metric-ball engines, and Ross Richardson #1216/#1145 for the restored-donor model, specialization, modular method and elementary templates. Their original permutation results are cited, not replayed. The package retains all predecessor credit and changes no predecessor source.
Evidence
Each of the two source-credited native representations enumerates 2,299,917 matrices and minimum distances and checks all 151,794,522 middle transitions for the new endpoint
123045. Both find zero direct and zero middle hits. Full canonical matrix/distance streams agree byte for byte. The ten-operation rational word and its inverse agree on all six independent input columns; all ten single-operation omissions are rejected.An independently represented dense 48-state control ball agrees with both engines at three small endpoints. Both engines reject two malformed native domains for their declared reasons, and six damaged endpoint reports are rejected. Warning-strict normal and optimized replays produce identical receipts bound to all five package files and four live sources.
Local checks passed on CPython 3.12.15 and Clang 21.0.0, build clang-2100.4.1.1, targeting arm64-apple-darwin27.0.0. Each registered command takes about 22 seconds locally; the combined registered replay took 45.07 seconds. Exact reproduction commands from the repository root:
A C++17 compiler named
c++withunsigned __int128support is required. Both native engines compile with-Wall -Wextra -Werror. Package metadata/pins, exact committed bytes/modes and declared source ancestry pass. The official nine-file export passes integrity and declared-input coverage checks. An optimized replay of the read-only Git-free export also passes.Generated exact-byte receipt
{ "canonical_ball_sha256": "a943ba6bc12d9da2d57ea31edb27455804adaea1280c1fd226af4c4915f210d0", "claim_type": "scoped proof with an exact finite exclusion certificate", "control_endpoints_per_engine": 3, "dense_control_states": 48, "finite_exclusion": { "direct_hits": 0, "generators": 66, "layers": [ 1, 66, 2655, 83270, 2213925 ], "middle_hits": 0, "minimum_through_nine": null, "n": 6, "permutation": "123045", "states": 2299917, "transitions": 151794522 }, "invalid_native_domains_per_engine": 2, "package_sha256": { "README.md": "7ce2f1150d825f0db9b4d13f10a80246f1c208ce281cb0bbada565dfc7a0f2ab", "SOURCES.json": "5dd609c6429e3026df465a0cb8159f429a3d19052740701a21ff647decd59642", "TASK.json": "da62a398a7b6f276dedc4389aa89612080e358512de6038c881c506aa10613db", "VERIFY.json": "5c2afc80baf6aa0d42e19494f83de852db15259b7111c8d0a19fc7df58d7d4c7", "verify.py": "18ec1768272c89ddea680c01d59e7089507990f473e6243656d052d216aa477b" }, "report_mutations_rejected": 6, "scalar_minimum": 10, "source_sha256": { "research/complex-restored-borrowed-boundaries/PROOF.md": "7808f30e28508e327f68fafdc0a1be58c779d892482d3713c767fd7c0d21e4b0", "research/restored-cycle-gap/README.md": "583bb8453cdd1aaf78307dd9db20d2e16f11db7638109464bcc94f6792327ccd", "research/restored-cycle-gap/packed.cpp": "8ca8361756b87eabbe79bb438324ab20941b8c4ff2967be4f7a444feb9ea9232", "research/restored-cycle-gap/ternary.cpp": "f6f64206c2269a21f571cc2de64c64fb73b5cee9af89de9347d5e38b0dcaa41f" }, "status": "PASS_RESTORED_UNSIGNED_FOUR_CYCLE", "upper_word": { "independent_input_columns": 6, "inverse_checked": true, "length": 10, "omissions_rejected": 10 } }Limitations and unrun obligations
This is an exact scalar boundary price. No current native four-cycle caller, paid recursive improvement, physical frame/tape layout, supplier admission or new kappa is claimed. Selected #7000 is unchanged. Promised inputs, clean/disposable workspace, arbitrary scalar multipliers, free monomial maps, joint complex-component arithmetic, nonlinear gates and changed composite boundaries are outside this model. Full supplier, precision, hosting, compiler and outer-theorem obligations were not run. Independent human and formal proof-kernel review are unclaimed.
Open next questions
Related open work
No other open PR identified for this exact ordinary arithmetic four-cycle endpoint. The source #1303 is merged.
CI notes
CI did not execute or certify this package. At unchanged head
66e05cf6c6058cff8a41a9b3ebe5945f5e36e9be, run 43522890842, plan job 132221300900, failed inOnboarding helper tests. The log reports 760 tests, exactly one failure and 24 skipped. The sole failing test istest_research_current_status.CurrentCiNavigationTests.test_current_ci_section_matches_selected_record, at line 58. Its AssertionError says that 'make frontier7000-verify' was not found in the CI verification document.The repository-level mismatch is present in base main
e54523517d0f03e58352e647c403a5af3e525a57:certificates/selected-result.jsonnamesmake frontier7000-verify, while the current section ofdocs/ci-verification.mddescribes #6030 andmake frontier6030-verify. Those files andtests/test_research_current_status.pyare unchanged by this PR. The dependent replay job 132223031757 was skipped. The local evidence above is therefore not hosted CI certification.I independently inspected the logs of rohanarun's run 42905094314, plan job 130140894651 and dillona's run 42903313790, plan job 130134720799. Both show the same sole failing test, assertion, 760-test count and 24 skips; both dependent replay jobs are skipped. No claim is made here about other runs whose logs were not inspected.
No code fix, retrigger commit or job rerun was made for this repository-level failure.
Merge danger
Two-way door. Blast radius: isolated research package. Only five new package files are added; shared code, frozen sources and the selected frontier are unchanged.