Repository navigation
Audit P104 relay composition and later-read departure floor - #2614
Merged
Merged
Conversation
This was referenced Oct 16, 2026
This was referenced Oct 16, 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The P104 relay discussion proposes a 208-helper departure floor while identifying 336 demands accessible through later source frames. This package proves the missing implication for the stated frozen-path, X-only monotone network class: after removing all potentially later-readable demands, each source's forced target hyperplanes still intersect in its initial line. Two departures per distinct source line are therefore necessary, even with rational mixing and cancellation. The numerical 208 bound and the 336 entries are credited predecessor results; this does not strengthen the numerical floor.
It also independently composes Philippe Van Velzen's pinned 1,031-helper relay-v2 with the complete original P104 channel. Separate rational forward and exact dyadic reverse interpreters check all 1,718,721 coefficients of the 1,311-input endpoint, all 21,682 chronological events, dirty-helper restoration and the 9,279 added / 11,647 combined boundary rank mass. The older 2,185-helper word is retained and fully replayed as historical evidence. PR #2594 is credited for its correction-only endpoint and logical-growth audit; no growth, arrival-shape or price-cap result is claimed here.
Validation: optimized CPython 3.12.7 replay and registered launcher passed all three test groups; a separate agent reviewed the proof and implementations and independently replayed both words on optimized CPython 3.12.14. Eight rejecting fault controls and one later-read countercontrol pass. Source hashes, Git blobs, exact receipts and Git-free export checks are included.
Base: current main
0fdd7d82de03e43c9e5a1000dfbcd380d1b687d2, priority revision 4. Picker seed296d172895dc11178a7b95219224afb7, selectedp104-side-correction; its first baseline command passed. TASK.json records the live handoff, update to v2, excluded active scope and final complete overlap cache of 798 open PRs and 1,252 Discussion records. Main was refreshed unchanged before submission.Limits: logical/frame audit and a scoped all-network proof, not native/global supplier admission. The original eight nonzero private-copy erasures are explicit; new native transports, full private implementation, hosting, precision and all-size admission remain open. Credit the pinned relay author and P104 lineage. Prepared by ywz65 with substantial OpenAI Codex assistance and separate agent review; no independent human or proof-assistant review is claimed.