Repository navigation
Compile programmable native stages and coded-thin fusion with exact precision (no new kappa) - #5562
Merged
Mister-Meeseeks merged 1 commit intoOct 25, 2026
Conversation
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
This extends #5466/#5467 with an actual rational-amplitude native event compiler, coded-thin fusion at existing common-frame cuts, and source-bound exact dyadic precision for both orientations. It retains #5464's actual paid/monomial/phase paths and checks all 1,024 independent columns. The fusion adds no Gaussian child, field or phase path. Exact dyadic storage B = g + ceil(25f/2) + 4 suffices for the actual streams; a constructive composition theorem covers an emitted dyadic caller.
No new complete κ is established. The old closing swap still gives one wrong-side-Q output branch. C-coded thin service is not uncoded same-port service. Ideal funding models are excluded from this result, and no arithmetic-only endpoint is promoted as a multiplier bound.
Reproduce with Python 3.12 and the standard library:
Normal and optimized checks, literal chart/column/paid-profile tests, exact noncommuting controls and precision mutation rejection pass. No Lean kernel audit or hosted Linux CI is claimed. No selected record, predecessor evidence or shared tooling changes.
Credit THU New Media Research Center #5464, Ali Romman #5439, Dugongue #4929/#5466/#5467, Dillon Amburgey, Cao Yifan, Cinna Davis, zxv, Jacob Alan Sussman, Chris R/Clear1nsight and the retained native/Clifford/compiler/analytic lineage. Exact source pins and original notices remain. Substantial OpenAI Codex assistance by one coordinator and three agents; Apache-2.0. Open next questions give the precise sufficient global closure still needed.