Skip to content

G87-D5 Band-C and gate-84 handoff classification #2952

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
briansrls wants to merge 1 commit into from
Closed

G87-D5 Band-C and gate-84 handoff classification #2952

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
23 changes: 16 additions & 7 deletions docs/briefs/r3-cementing-discipline-pattern-2026-05-12.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -69,14 +69,23 @@ The rows below are the complete Gate #87 `.dag` `Compiles` placeholder set at th

These are not a frozen dispatch inventory; the authoritative list is `EXPECTED_HAND_AUTHORED_TEST` in `src/v3/compiler/tests/integration/sg0_census_test.rs`. The table below records the disposition the #84 cementing-class worker must check before selecting pilots — predicate class (§2), blocker if any with owning lane, and the SG-0 hand-path census delta on a successful port. Refresh date: 2026-05-13 (G87-D5 handoff alignment). On disposition change, update this table together with the live census comment above the same row.

Census coverage check (G87-D5, refresh 2026-05-13): the four rows below are the complete set of paths under `src/v3/compiler/tests/integration/cementing/` that appear in `EXPECTED_HAND_AUTHORED_TEST` at HEAD. No new cementing-test residual rows have landed since the prior refresh; no row in this table has been ported away from Rust since the prior refresh, so the §3 dispositions stand. This brief makes no census-delta claim — it is docs-only.
**G87-D5 dispatch category** (per `docs/briefs/r3-gate-87-lens-cementing-test-discipline-dispatch-2026-05-13.md` §G87-D5): each `cementing/*.rs` row still listed in `EXPECTED_HAND_AUTHORED_TEST` must land in exactly one of:

| Rust module | Predicate class (§2) | Blocker / owning lane | SG-0 census delta on port |
|---|---|---|---|
| `tests/integration/cementing/cementing_provenance_origin_integration_test.rs` | C-LensOutEq once unblocked; C-HandRustBlocker today. | Expected-carrier authoring for the `Origin` sum (`NoProducer` / `MissingPort` / `MissingBehavior` / `Source(NodeId)` / `Computed(NodeId)` / `Selected(NodeId)` / `Accumulated(NodeId)`) in `.dag` `TestPredicate` expected values. Owning lane: tests-as-data carrier completeness for sum-typed lens outputs (`docs/design-tests-as-data-completeness.md` §C5). Target `.dag`: `tests/dag/t_r3_gate_87_cementing_regen_provenance.dag` already cements the seam projection; this Rust module pins the per-`Behavior` mirror that the `.dag` receipt cannot yet express. | -1 when the per-`Behavior` mirror moves into the existing `.dag`. |
| `tests/integration/cementing/complexity_lens_behavioral_completion.rs` | C-HandRustBlocker today; C-LensOutEq on unblock. | `Gate73_ReportPredicateCarriers` — `.dag` `TestClaim` predicates cannot yet consume the published `ComplexitySummary` report carrier. Owning lane: T-LBP / `docs/r3-program-plan.md` gate #73 report-predicate carrier authoring. No `.dag` target yet (waiting on the carrier). | -1 on Gate-73 unblock + same-PR move into a new `tests/dag/t_r3_*_complexity_*.dag` receipt. |
| `tests/integration/cementing/cost_lens_symbolic_consumer_test.rs` | NOT a Band-C lens-cementing row — do not select for #84 cementing-class bulk-port. | `cost_symbolic` COMPLETE-row cementing already landed as data at `tests/dag/t_r3_gate_87_cementing_regen_cost_symbolic.dag` (PR #2769 dissolved `M1_2_8_STRUCTURAL_SYMBOLIC_COST_DATA`). The residual is gate #78 `per_call_pattern_at` / `symbolic_cost_of` host-wrapper pin — owning lane: gate #78 host-wrapper retirement, NOT cementing. | 0 from this class. Census decrement happens with the gate #78 host-wrapper retirement, not by reclassifying this row as cementing. |
| `tests/integration/cementing/memory_peak_cost_basis_demo.rs` | C-HandRustBlocker today; C-LensOutEq or C-CompilesHelper on unblock depending on demonstration shape. | Parser-level `apply_lens(cost, DeclarationScope, Enforce { budget: SymbolicCost { dimension: Memory, … } })` consumer. Owning lane: T-LAS Slice B lens-fold consumer, `docs/r3-program-plan.md` gate #91; gate #94 (`memory_peak_cost_basis_demonstrated`) is the consumer-side gate this receipt evidences. No `.dag` target yet (waiting on the parser-level consumer). | -1 on gate #91 unblock + same-PR move into a new `tests/dag/t_r3_*_memory_peak_*.dag` receipt that preserves max-dominance composition and `LensEnforcement` orientation semantics. |
1. **Gate-87 residual** — hand-Rust that extends or pairs with a `regen.dag` lens already covered by `t_r3_gate_87_cementing_regen_*.dag` / `r3_gate_87_lens_cementing_regen_receipts_test.rs`; #84 cementing-class workers must **not** treat these as freestanding bulk-port pilots outside §1.
2. **Band-C bulk-port candidate** — hand-Rust cementing discipline that is **not** authoritative for proving gate #87 on the registry corpus (typical Phase-3 #84 slice for enumeration-outside-`regen.dag` work).
3. **Host-wrapper retirement** — receipt belongs to a named host-wrapper / alias-collapse gate, not to lens-cementing-class #84 port work.
4. **T-LAS demonstration scope** — lens-application-surface demonstration or consumer ratchet owned by T-LAS gates, not registry-cementing inventory.

Census coverage check (G87-D5, refresh 2026-05-13, re-checked session `proud-koi-122`): the four rows below remain the complete set of paths under `src/v3/compiler/tests/integration/cementing/` that appear in `EXPECTED_HAND_AUTHORED_TEST` at HEAD (matches on-disk `cementing/*.rs` and `tests/integration.rs` `#[path = …/cementing/…]` wiring). No new cementing-test residual rows have landed since the prior refresh; no row in this table has been ported away from Rust since the prior refresh, so the §3 dispositions stand. This brief makes no census-delta claim — it is docs-only.

**Bulk-port note:** at HEAD, **none** of the four `cementing/` integration modules are classified as **Band-C bulk-port candidate**. Phase-3 #84 cementing-class pilots for work *outside* the `regen.dag` enumeration must be selected from other `EXPECTED_HAND_AUTHORED_TEST` rows (coordinator live inventory), not from this subdirectory.

| Rust module | G87-D5 category | Predicate class (§2) | Blocker / owning lane | SG-0 census delta on port |
|---|---|---|---|---|
| `tests/integration/cementing/cementing_provenance_origin_integration_test.rs` | Gate-87 residual | C-LensOutEq once unblocked; C-HandRustBlocker today. | Expected-carrier authoring for the `Origin` sum (`NoProducer` / `MissingPort` / `MissingBehavior` / `Source(NodeId)` / `Computed(NodeId)` / `Selected(NodeId)` / `Accumulated(NodeId)`) in `.dag` `TestPredicate` expected values. Owning lane: tests-as-data carrier completeness for sum-typed lens outputs (`docs/design-tests-as-data-completeness.md` §C5). Target `.dag`: `tests/dag/t_r3_gate_87_cementing_regen_provenance.dag` already cements the seam projection; this Rust module pins the per-`Behavior` mirror that the `.dag` receipt cannot yet express. | -1 when the per-`Behavior` mirror moves into the existing `.dag`. |
| `tests/integration/cementing/complexity_lens_behavioral_completion.rs` | Gate-87 residual | C-HandRustBlocker today; C-LensOutEq on unblock. | `Gate73_ReportPredicateCarriers` — `.dag` `TestClaim` predicates cannot yet consume the published `ComplexitySummary` report carrier. Owning lane: T-LBP / `docs/r3-program-plan.md` gate #73 report-predicate carrier authoring. No `.dag` target yet (waiting on the carrier). | -1 on Gate-73 unblock + same-PR move into a new `tests/dag/t_r3_*_complexity_*.dag` receipt. |
| `tests/integration/cementing/cost_lens_symbolic_consumer_test.rs` | Host-wrapper retirement | NOT a Band-C lens-cementing row — do not select for #84 cementing-class bulk-port. | `cost_symbolic` COMPLETE-row cementing already landed as data at `tests/dag/t_r3_gate_87_cementing_regen_cost_symbolic.dag` (PR #2769 dissolved `M1_2_8_STRUCTURAL_SYMBOLIC_COST_DATA`). The residual is gate #78 `per_call_pattern_at` / `symbolic_cost_of` host-wrapper pin — owning lane: gate #78 host-wrapper retirement, NOT cementing. | 0 from this class. Census decrement happens with the gate #78 host-wrapper retirement, not by reclassifying this row as cementing. |
| `tests/integration/cementing/memory_peak_cost_basis_demo.rs` | T-LAS demonstration scope | C-HandRustBlocker today; C-LensOutEq or C-CompilesHelper on unblock depending on demonstration shape. | Parser-level `apply_lens(cost, DeclarationScope, Enforce { budget: SymbolicCost { dimension: Memory, … } })` consumer. Owning lane: T-LAS Slice B lens-fold consumer, `docs/r3-program-plan.md` gate #91; gate #94 (`memory_peak_cost_basis_demonstrated`) is the consumer-side gate this receipt evidences. No `.dag` target yet (waiting on the parser-level consumer). | -1 on gate #91 unblock + same-PR move into a new `tests/dag/t_r3_*_memory_peak_*.dag` receipt that preserves max-dominance composition and `LensEnforcement` orientation semantics. |

Rows that remain blocked after this check route to D2 with the named blocker. Rows unblocked by later substrate/compiler work hand back to V2 for same-PR replacement artifact plus census decrement under §4. This table is the only hand-maintained disposition surface for these rows; #84 cementing-class workers must not create a parallel inventory.

Expand Down
Loading