Replay T385 through release and audit R437 complete-stream COPY

.github/workflows/p112-common-slice.yml

@@ -15,9 +15,9 @@ jobs:
     if: github.repository == 'CrocSwap/integer-mult-bounds'
     runs-on: ubuntu-24.04
     timeout-minutes: 20
-    env:
-      TMPDIR: ${{ runner.temp }}/p112-common-slice
     steps:
+      - name: Initialise P112 work directory
+        run: echo "TMPDIR=$RUNNER_TEMP/p112-common-slice" >> "$GITHUB_ENV"
       - uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1
         with:
           fetch-depth: 0

research/t385-r437-copy-audit/.gitattributes

@@ -0,0 +1,3 @@
+# Preserve byte-authenticated predecessor material across checkout platforms.
+inputs/** -text -whitespace
+R437-* -text -whitespace

research/t385-r437-copy-audit/CALLER-BOUNDARY.md

@@ -0,0 +1,79 @@
+# Where the complete-stream implementation handoff reaches actual code
+
+This is a source inspection, not execution or independent validation of
+PR2809's universal refinement proof. The inspected files are pinned as
+provenance at canonical main `20ecf2891d80b5acf18d980b2164fb44ea7faa90`.
+Jose Manuel / jmbarrancoml supplied the physical lowering of David Arias's
+PR2753 caller. Their notices and the selected primitive/parking lineage remain.
+
+## A concrete wrapper already exists
+
+In `research/selected-bit-physical-caller/macro_code.py`, `compile_node`
+handles a local COPY event by emitting, in order:
+
+1. Copy the current staged center to the permanent `copy_source` role.
+2. Execute one residual on `copy_source`, using that event's request and the
+   current common low class.
+3. Transfer its completed output to the local private role.
+
+At ERASE it emits a clear of that private role at the original guarded
+position. No helper-label projection appears in this wrapper. The source
+`LocalState` validates the event lifecycle during compilation; the wrapper
+is not a free-standing license to clear arbitrary persistent state.
+
+`macro_machine.py` implements COPY by committing the source tape to the named
+destination. Its `commit` clears the destination, appends the entire source,
+and rewinds the destination. It clears the returned source only if that
+source is a temporary `Lease`; a staged permanent source role is preserved.
+MOVE transfers between fixed roles; CLEAR calls that role's tape clear.
+These operations make explicit the complete source clone, owned output and
+linear disposal that a cost audit must retain. They do not support charging
+an unspecified dirty-input projection as if it were the whole source.
+
+The source proof, section 3, describes that same source-clone/residual/private
+sequence. Section 5 inventories scratch inside atomic blocks, and sections 6–7
+supply the parent parking/return argument. This is an existing proposed
+implementation path, not an assertion that nobody has written a caller.
+
+## What the reduced execution actually covers
+
+`macro_local_checks.py` observes the copied source before the residual,
+compares the copied output against an independent address permutation, and
+checks that the original source is unchanged **before** later XOR cancellation.
+The published complete GL2(F3) fixture checks 96 COPY endpoints, 192 MOVE and 384
+IDLE residuals on 31,104 bits. Its source proof explicitly says it has no
+internal child. A separate GL2(F2) fixture tests recursive control; the selected
+junction probe is not a complete selected GL64 payload.
+
+This session did not rerun those fixtures or independently review the whole
+parametric proof. The repository's current physical-contract map also keeps
+PR2809's universal refinement pending despite archival incorporation and
+later ordinary-backend compositions. Reduced fixtures and source archival
+status should not be turned into an R437 implementation/admission certificate.
+
+## Precise next implication
+
+The new R437 checker establishes the exact-function endpoint and guards of
+R437's actual rank-seven requests. The concrete wrapper above is structurally
+compatible with a universal complete-stream residual. The unclosed step is
+an exact-source binding and proof that its rank-seven residual/caller:
+
+- accepts R437's current source/owner/chart data and all arbitrary helper
+  functions, including the newly occurring input 340;
+- preserves its borrowed source and returns the transformed complete output;
+- charges source clone, all routing/adapters, child execution, transfer and
+  complete disposal under the retained volume-normalized tariff;
+- restores child scratch and parking state within the actual full reserve;
+- satisfies the admitted numerical format, exceptional-class and recovery
+  conditions on the required all-size domain.
+
+The available source-level compiler/primitive proof may be the route to those
+claims, but this finite application audit does not independently validate it.
+No second recursive uncompute is forced by the existing wrapper: it transforms
+an owned clone and later clears owned output. Conversely, clear is a real tape
+operation and its linear cost and scratch lifetimes remain part of the bill.
+
+A useful next audit should start with `compile_node`'s COPY branch,
+`Machine.commit`/COPY/CLEAR and the bound residual plan, then connect one
+actual R437 request to the general primitive and lease/return arguments.
+Repeating the GL2 endpoint fixture alone would leave that implication open.

research/t385-r437-copy-audit/EVIDENCE.json

@@ -0,0 +1,63 @@
+{
+  "files": {
+    "evidence/t385/admission-result.json": "43ff0e5058f242b2783082581b2f29e0b54deae3f2036f3654a0f3c0a198736c",
+    "evidence/t385/discovery-RESULT.json": "13526e1505efdc599096ac611b9b9a782a670ed98a365d471d8bf997dc2b8e24",
+    "evidence/t385/discovery-STAGES.json": "cff1d2a1200525a23790716e98e849188577a29e5680b3d5433e02e3501beb08",
+    "evidence/t385/discovery-assembly.log": "1518262cd2fb428acb41b5e3bb4fbc2a9aac03b3f0f6f3da0f222226bbd6799d",
+    "evidence/t385/discovery-audit.log": "e7403fd02cfb7a18f5eb1098de1a3d9830c209521ded74b1a4fbf0044349fc79",
+    "evidence/t385/discovery-bind.log": "71f679ce5a123fb58397fd6852ba64ce3832fd20e81a096bfda9a1cf3237237a",
+    "evidence/t385/discovery-emit.log": "cb7715dcffa5a5ec8e15559df8ad80620a6b1d1edd41b1f9ef1572711de318c5",
+    "evidence/t385/discovery-freeze.log": "dd6da2aa4a2d630e61a2f77138921c578c0c6409d170d58568a29af4ee56d379",
+    "evidence/t385/discovery-independent.log": "d2c176fb1b9778e4fb8b98b6272d5f713d676ea3fed84da20213eca036bb58fe",
+    "evidence/t385/discovery-invoice.log": "0fa5bbe81eef6ea6e7824bcb686c9207a748697d9d98348bfd4a05e7511e8945",
+    "evidence/t385/native-receipt.json": "7ddf3f80b2fa61d1ab12035da9ef1c95d43f76077df819d7683144d7862a3f93",
+    "evidence/t385/release-RELEASE-RECEIPT.json": "9f49e2e801cff8dd9402121e16c0d63801975740140637c5943f94149b28c803",
+    "evidence/t385/release-outer.log": "670f60ee3ecc7e2d7fc4b23b468d4cdb8cfab9edaf252cd2410c1cabcb0318dc",
+    "evidence/t385/release-strict-RESULT.json": "7babd8a958bff810adf89d560c73a01c052d98210d6852dbea106400ccbdf65d",
+    "evidence/t385/release-strict-STAGES.json": "6a8ae6fd5198ddbb4443060d27917ffcd083ee55bd32bb68314f2fe951c5eee4",
+    "evidence/t385/release-strict-admission.log": "1cee8d732e126c670f531018d65955c23828a7cc0af8d6ac4bb0f6672962a540",
+    "evidence/t385/release-strict-assembly.log": "1518262cd2fb428acb41b5e3bb4fbc2a9aac03b3f0f6f3da0f222226bbd6799d",
+    "evidence/t385/release-strict-audit.log": "e7403fd02cfb7a18f5eb1098de1a3d9830c209521ded74b1a4fbf0044349fc79",
+    "evidence/t385/release-strict-bind.log": "71f679ce5a123fb58397fd6852ba64ce3832fd20e81a096bfda9a1cf3237237a",
+    "evidence/t385/release-strict-bit.log": "d54d260bf95bd8ae4ab8846314082b93b2e3d8f427eb35376691f2ce543a6b67",
+    "evidence/t385/release-strict-emit.log": "cb7715dcffa5a5ec8e15559df8ad80620a6b1d1edd41b1f9ef1572711de318c5",
+    "evidence/t385/release-strict-independent.log": "d2c176fb1b9778e4fb8b98b6272d5f713d676ea3fed84da20213eca036bb58fe",
+    "evidence/t385/release-strict-invoice.log": "0fa5bbe81eef6ea6e7824bcb686c9207a748697d9d98348bfd4a05e7511e8945",
+    "evidence/t385/release-verify-RESULT.json": "0372a78fcb546538d781ae54d70424b272a3faff632e0a391cb383326f1aa6ba",
+    "evidence/t385/release-verify-STAGES.json": "8f23bb4af5567f2d8fa2e09d3cf54f6973591cb536bbb35d4daa247a132aa1cb",
+    "evidence/t385/release-verify-audit-local-566-base.log": "c16c2e894f5bb470a8b16bd478f3e311c2dfa397d4f87f0ab36f46c2ee709f78",
+    "evidence/t385/release-verify-audit-local-566-extra.log": "0141a0a056a6c1c31dcb88c9ae613dd76c771dc43ef5f3867ad46f984dba8a01",
+    "evidence/t385/release-verify-declared-edit-review.log": "9812aa8d08d4251d390eb0cea1e5731dc59d22e09e0c914823c77d3e2e233e13",
+    "evidence/t385/release-verify-delayed-release-independent-review.log": "cc8f2e06baec65ca81cb91674bd23ea14320f6b1ceee9a6cfba251387d77c737",
+    "evidence/t385/release-verify-fine-engine-oracle.log": "ea6df7be568d6461ac54e0d56754797de0f70f2f1ecc335114791445b75795d3",
+    "evidence/t385/release-verify-geometry-controls.log": "0a3ecf05cfa2bebb61fffd2fbb4b244c03d742069e2705d02512699bbf4b01f9",
+    "evidence/t385/release-verify-lemmag-labels.log": "91fafdc76824e278546bb51465cf84d60a3513df3f2e8ab9854919c416cd3469",
+    "evidence/t385/release-verify-lift-parent.log": "c5102a5a2c10b2a5e08927636187a9c1df2841a094c58f1a61326a2b11274483",
+    "evidence/t385/release-verify-paired-extension-review.log": "f5928ef4aecde0f00d2bcded7b6cf9577c4bfe756bc190b68a8f7901dc2e1dc7",
+    "evidence/t385/release-verify-preload-move-independent-review.log": "c456d3e39c09e960ceebf80251d5e808fa9299991471fd86f4093d9bc4dd3eda",
+    "evidence/t385/release-verify-rebind-566.log": "4efcc424f53c49777d308af3bc568688d2eb0de9c0236aae9af6ab50190974a7",
+    "evidence/t385/release-verify-rebuild-borrow-words.log": "59dc4ab25e1e91611f2049d9dd9c3f5f210e4fc9b5b394b6a21de60745694f54",
+    "evidence/t385/release-verify-rebuild-words-566.log": "dcbcbeda80423df29ef008bfac58409378f1fd0a3b9b23fc31ab9f03659dd86f",
+    "evidence/t385/release-verify-reconstruct-owners.log": "239d8562814d1627c19e7e6bf75f06ae2ef219b50c948bb612dcfa0f3f21ef1c",
+    "evidence/t385/release-verify-shared-kernel-independent-review.log": "3dc6eb8940cf8a4b993f80129c98b6c4846ad346ccf773a78af225cff3339cc3",
+    "evidence/t385/release-verify-staged3605-review.log": "de2fa433a259e21f69f10db63667ba6034e0cd35f102a989208e865d042c31c8",
+    "evidence/t385/release-verify-staged544-independent-word-review.log": "2b3e72ac971d0d7eacd8277defcd85ff3d24771285681d1b99fb6fdc4c3cdbf0",
+    "evidence/t385/release-verify-strict-t74.log": "5e17e55fd26dcde3a43516c4654f55b572a05f61bb8381e8883f876e6a291dcb",
+    "evidence/t385/release-verify-two-sided-helper408-review.log": "a44341620c98e985c522107cdec7def37b715267c605976b25c33fe8a9f3cefb",
+    "evidence/t385/release-verify.log": "f8db339ba756aa4054dc9329d8f059ed700f72d273cff73422ea820229f47a24",
+    "evidence/t385/runtime-patches.json": "def1cdb3a467a9763f283ebbecbbfba979e915012f5f690d4aaab620a2b21498",
+    "evidence/t385/stage-discovery.log": "0698d7bbe8ffa568e2c0cea6bdd2fea81cb7fe17df8a4ed58e0f10e6bf14a421",
+    "evidence/t385/stage-outer-check_ledger.py.log": "b66b80965a59ac5aec812a16e8cbcddf6c97c02b3e6dd35b6be8ce02ff9174c4",
+    "evidence/t385/stage-outer-independent_check.py.log": "639d5cdd1ad11190a6198809aa3c5b00245d905dabf6d7671ccbf1a587202276",
+    "evidence/t385/stage-owners.log": "239d8562814d1627c19e7e6bf75f06ae2ef219b50c948bb612dcfa0f3f21ef1c",
+    "evidence/t385/stage-parent-lift.log": "c5102a5a2c10b2a5e08927636187a9c1df2841a094c58f1a61326a2b11274483",
+    "evidence/t385/stage-rebind.log": "4efcc424f53c49777d308af3bc568688d2eb0de9c0236aae9af6ab50190974a7",
+    "evidence/t385/stage-release.log": "acdf066ba832393ac990976f6705748d72a1cc0d521b394711516c874e7a4796",
+    "evidence/t385/supervisor.json": "57443cd242961ee84a72e8ee0e18044923b31727ee8c54b3f9350acd3d3b28c8",
+    "evidence/t385/timeout-extension.json": "a2a59b018b656929d63f67b7f4d10cfa1b6256a0690a17d9cdb333b38dfa1c88",
+    "evidence/t385/wrapper-result.json": "70789b05fc62ce1fb8b53332816231dabaa5109a2a8ab24660fe22b79e816dcb",
+    "evidence/t385/wrapper.log": "8289a527218d2bd3b3f76510e0fcbbd05ddc99ec4342590d791733f4265c0337",
+    "T385-EXECUTION.json": "3adbbafd78f2b2c7e1ba9dc8f26ecec0eda3508fe2baa12442e6b0b263064914"
+  },
+  "scope": "Authenticate recorded execution evidence; not an execution or mathematical acceptance claim"
+}

research/t385-r437-copy-audit/OVERLAP.json

@@ -0,0 +1,99 @@
+{
+  "checked_utc": "2026-10-19 04:59:01 UTC",
+  "records": 638,
+  "unique_prs": 638,
+  "coverage": "All seven open-PR title/body pages including drafts; non-atomic snapshot. Closed history not fully enumerated. Discussion handoff refreshed manually; broader Discussion coverage partial.",
+  "matches": [
+    {
+      "number": 3883,
+      "title": "Independent audit of R437 OQ4 transition and staged544 admission replay",
+      "head": "0ce706e210d133ee36b65144f87f0a74a74bec6d",
+      "url": "https://github.com/CrocSwap/integer-mult-bounds/pull/3883"
+    },
+    {
+      "number": 3890,
+      "title": "Certify the complete staged544 joint width-price region",
+      "head": "772953c9bda6d41db7b89e4cd6bae3dc05d7636e",
+      "url": "https://github.com/CrocSwap/integer-mult-bounds/pull/3890"
+    },
+    {
+      "number": 3880,
+      "title": "Compile dirty-buffer XOR corrections with explicit space–time bounds",
+      "head": "eef2760be4ed4b08f9c59b037e3ad2380a8bf958",
+      "url": "https://github.com/CrocSwap/integer-mult-bounds/pull/3880"
+    },
+    {
+      "number": 3875,
+      "title": "Refine the certified staged544 conditional kappa endpoint to 80 digits",
+      "head": "a65afd67e0a43e02554b58b5593d253581207b04",
+      "url": "https://github.com/CrocSwap/integer-mult-bounds/pull/3875"
+    },
+    {
+      "number": 3857,
+      "title": "Kernel-check the staged supplier moment and refine conditional κ",
+      "head": "ef4315b8caa605e47a2515004845eb2d2ed71a86",
+      "url": "https://github.com/CrocSwap/integer-mult-bounds/pull/3857"
+    },
+    {
+      "number": 3647,
+      "title": "Improve conditional BIT c to 0.001736646042 with composed R437 and credited slot coalescing",
+      "head": "a24f9ff95c12fd0b6792e3f710569237efd72db2",
+      "url": "https://github.com/CrocSwap/integer-mult-bounds/pull/3647"
+    },
+    {
+      "number": 3802,
+      "title": "Improve conditional kappa beyond T488 with a T385 combined-clock allocation",
+      "head": "de2cd5c58009205210e3bf328b0ef5cadf3903af",
+      "url": "https://github.com/CrocSwap/integer-mult-bounds/pull/3802"
+    },
+    {
+      "number": 3815,
+      "title": "Research: certify the staged544 packing ceiling and reach it on PR3734's T360 clock (no new κ; same value as #3802)",
+      "head": "77ffbf287b6ca198b4a31e3835db3bae60718b93",
+      "url": "https://github.com/CrocSwap/integer-mult-bounds/pull/3815"
+    },
+    {
+      "number": 3724,
+      "title": "Independent verification (no κ): arithmetic layer of PR3562's ordinary-rate theorem holds; line-158 exposition gap, 4283584 stands",
+      "head": "946d03dd6b9cc794190876fe3f0f89b0ebc01c59",
+      "url": "https://github.com/CrocSwap/integer-mult-bounds/pull/3724"
+    },
+    {
+      "number": 3700,
+      "title": "Independent verification (no κ): PR3647's composed R437 bit word, hosting and price pass an exact audit against single-change controls (c = 0.001736628253427115)",
+      "head": "aa0bd9fbc8700506b1fafcc55bca9bd32aced7d9",
+      "url": "https://github.com/CrocSwap/integer-mult-bounds/pull/3700"
+    },
+    {
+      "number": 3690,
+      "title": "Research: source-bind R437 ordinary coupling at c=1/576 (no κ claim)",
+      "head": "03c141ef1fd372cf42951488767537e55675c072",
+      "url": "https://github.com/CrocSwap/integer-mult-bounds/pull/3690"
+    },
+    {
+      "number": 3668,
+      "title": "Add the published R437 histogram and compare its coefficient with the selected kappa",
+      "head": "a69d1f76cbfbc5441bccf52de01672308d5447e1",
+      "url": "https://github.com/CrocSwap/integer-mult-bounds/pull/3668"
+    },
+    {
+      "number": 2745,
+      "title": "Conditional κ = 0.001716140425997441: #2696's lender assignment with #2679's pre-life hosting and a window-optimal fill",
+      "head": "4f485de05544884fa0f2edf6063bb9259decad30",
+      "url": "https://github.com/CrocSwap/integer-mult-bounds/pull/2745"
+    },
+    {
+      "number": 1724,
+      "title": "Conditional κ = 0.0016769381110224387711...: stacked pivot 566 + 554 kernels with 600-digit grid and 600-digit endpoints over PR1692",
+      "head": "d7d89428fe5219d56b115371d5074d5e518fd0c7",
+      "url": "https://github.com/CrocSwap/integer-mult-bounds/pull/1724"
+    },
+    {
+      "number": 1692,
+      "title": "Conditional κ = 0.0016769381110224387711...: stacked pivot 566 + 554 kernels with 470-digit grid and 450-digit endpoints over PR1672",
+      "head": "c7b46d54328214469c47423a3bcc98699bd01f76",
+      "url": "https://github.com/CrocSwap/integer-mult-bounds/pull/1692"
+    }
+  ],
+  "claim": "https://github.com/CrocSwap/integer-mult-bounds/discussions/1166#discussioncomment-19030424"
+}