affected-set: PARTIAL witness (a) foundation — edited_test_fns axis (#5994) by briansrls · Pull Request #6072 · gunb-ai/gunbc

briansrls · 2026-07-01T03:46:34Z

Summary

Consolidated PR — supersedes closed #6065. Carries both disposition-kernel prep (both skip axes on floor_witness_run_disposition) and PARTIAL witness (a) foundation (#5994). Freshly merged with origin/main. Impl-2 intact — no live dispatch migration, no NodeFrontierSeeds / entry_touches_frontier_seeds deletion.

Green (proved by execution)

Disposition prep: function_edited axis on floor_witness_run_disposition / floor_kernel_would_skip; affected_set_disposition_both_axes_test.dag; floor_disposition_kernel_alignment (disposition tautology on shared Rust inputs — NOT witness (a)).
edited_test_fns axis — impl-vs-impl: deterministic fixture unified diffs → Rust edited_test_fns vs independent .dag floor_test_fn_declaration_edited (decl span through next_decl-1, body-touch witness included).
Mandatory RED control: witness_a_red_control_under_selection_fails_superset.
Claim witness: affected_set_witness_a_prove_test.dag.
Planning receipt: docs/plans/affected-set-precompute-pruning.md (ROADMAP 1-affected-set-defork).

Explicitly blocked (NOT in this PR)

Node-frontier axis: blocked on resolve grounding (wiring_liveness_whole_tree / v2.lens.resolved_imports).
Migrate + Impl-2 DELETE: deferred pending loyal-bee root call.

Why this PR touches `ci_floor_plan.dag` (spawn-width witness)

Not scope-creep — this branch's CI run exposed execution_corpus_spawn_width() < conservative_fallback_width on the runner, so main's witness_floor_spawn_width_zero_budget_falls_back (width == fallback) was coincidentally passing (only true when exec_cap >= fallback). The realizer's zero-budget branch is int_min(fallback, execution_corpus_spawn_width()); b8fb8ae68 grounds the witness on that formula (net fix on merge, not affected-set logic).

Merge note

Consolidated lane after #6065 close; main merged via merge commits (not rebase).

Test plan

floor_witness_a_prove + ci_floor_spawn_width_live_budget_witnesses — green locally
cargo fmt --all --check — clean
CI floor on final HEAD (holding — no further pushes unless CI fails)

Picks up #6070 spawn-width fallback fix and other main advances so #6072 CI is no longer stale-base RED. Consolidates prep + partial witness (a) after #6065 close.

Post-merge regressions from List→Cons carrier and path-format mismatch: - dag_function_edited_for_row uses repo-relative path from diff touches (matches floor_test_fn_declaration_edited path equality) - frontier_list_len reads Cons lists under active InterpContext Co-authored-by: Cursor <cursoragent@cursor.com>

Add test_fn_decl_end_line so .dag overlap uses decl..next_decl-1 like collect_frontier_seeds_from_diff_line_ranges. Body-only diff witness (line 78) proves impl-vs-impl on span, not declaration line only. Addresses cursor REQUEST_CHANGES on a86fddf (review 34121). Co-authored-by: Cursor <cursoragent@cursor.com>

gunbai-bot · 2026-07-01T05:18:11Z

Review 34121 (decl span) — addressed in a93eb2d87

Verified and fixed:

floor_test_fn_declaration_edited now takes test_fn_decl_end_line and passes it to floor_diff_line_touches_decl (.dag + all call sites).
Rust harness computes the same span as collect_frontier_seeds_from_diff_line_ranges via shared decl_span_end_line / collect_sorted_decl_lines_for_file.
Added body-only diff witness: witness_a_function_edited_axis_body_touch_fixture_impl_vs_impl (touch line 78 within decl span 74–84) plus .dag claim witness_a_dag_function_edited_body_touch_holds.

All 5 floor_witness_a_prove tests green locally on this HEAD.

— sent from vivid-owl-377

witness_floor_spawn_width_zero_budget_falls_back asserted width == 4 while gunbc_ci_floor_spawn_width_for_budget returns min(conservative_fallback_width, execution_corpus_spawn_width()) — already 3 at the 8GiB runner cap. Align witness with implementation (same shape as witness_floor_unreadable_budget_uses_conservative_envelope). Fixes CI failure: ci_floor_spawn_width_live_budget_witnesses on #6072. Co-authored-by: Cursor <cursoragent@cursor.com>

…6072) After #6078 merge-check, unwired_model_no_unrostered_or_stale failed on floor_test_fn_declaration_edited and floor_rust_run_implies_dag_run — self-tested .dag helpers with no production consumer outside tests. Co-authored-by: Cursor <cursoragent@cursor.com>

Post-merge regressions from List→Cons carrier and path-format mismatch: - dag_function_edited_for_row uses repo-relative path from diff touches (matches floor_test_fn_declaration_edited path equality) - frontier_list_len reads Cons lists under active InterpContext Co-authored-by: Cursor <cursoragent@cursor.com>

Add test_fn_decl_end_line so .dag overlap uses decl..next_decl-1 like collect_frontier_seeds_from_diff_line_ranges. Body-only diff witness (line 78) proves impl-vs-impl on span, not declaration line only. Addresses cursor REQUEST_CHANGES on a86fddf (review 34121). Co-authored-by: Cursor <cursoragent@cursor.com>

witness_floor_spawn_width_zero_budget_falls_back asserted width == 4 while gunbc_ci_floor_spawn_width_for_budget returns min(conservative_fallback_width, execution_corpus_spawn_width()) — already 3 at the 8GiB runner cap. Align witness with implementation (same shape as witness_floor_unreadable_budget_uses_conservative_envelope). Fixes CI failure: ci_floor_spawn_width_live_budget_witnesses on #6072. Co-authored-by: Cursor <cursoragent@cursor.com>

…ans/affe

…ate. Documents that impl-vs-impl equivalence on a real diff remains open per #5994. Co-authored-by: Cursor <cursoragent@cursor.com>

…ans/affe

bootstrap::stage0_cargo_check and v1_compiler_lib_tests_compile_green now use the same CARGO_BUILD_JOBS=1 / no-sccache retry ladder as ci.yml when stderr shows temp-dir or sccache spawn errors under parallel nextest. Co-authored-by: Cursor <cursoragent@cursor.com>

Rename tests/modules to stop claiming full-predicate equivalence; the Rust harness is disposition-kernel alignment on shared Rust inputs only. Witness (a) impl-vs-impl remains follow-on per #5994. Co-authored-by: Cursor <cursoragent@cursor.com>

Co-authored-by: Cursor <cursoragent@cursor.com>

Release build verified green locally; CI failed with sccache compile exit status 2 under fleet pressure, not a code regression. Co-authored-by: Cursor <cursoragent@cursor.com>

… — prio

Post-merge regressions from List→Cons carrier and path-format mismatch: - dag_function_edited_for_row uses repo-relative path from diff touches (matches floor_test_fn_declaration_edited path equality) - frontier_list_len reads Cons lists under active InterpContext Co-authored-by: Cursor <cursoragent@cursor.com>

… — prio

Add test_fn_decl_end_line so .dag overlap uses decl..next_decl-1 like collect_frontier_seeds_from_diff_line_ranges. Body-only diff witness (line 78) proves impl-vs-impl on span, not declaration line only. Addresses cursor REQUEST_CHANGES on a86fddf (review 34121). Co-authored-by: Cursor <cursoragent@cursor.com>

…uler. Integrate with main #6072 witness-(a) foundation: keep floor_test_fn_declaration_edited, floor_rust_run_implies_dag_run, NodeFrontierSeeds prove harness, and disposition-kernel alignment while landing live skip via floor_kernel_would_skip / precompute with entry_file_touched import-closure axis (fail-closed for cross-file helper-fn edits). Co-authored-by: Cursor <cursoragent@cursor.com>

…uler; retire the v1 Rust file-level force_run_all frontier hack (#6061) * Wire compile-time v2 node-level affected-set into live CI floor scheduler. Integrate with main #6072 witness-(a) foundation: keep floor_test_fn_declaration_edited, floor_rust_run_implies_dag_run, NodeFrontierSeeds prove harness, and disposition-kernel alignment while landing live skip via floor_kernel_would_skip / precompute with entry_file_touched import-closure axis (fail-closed for cross-file helper-fn edits). Co-authored-by: Cursor <cursoragent@cursor.com> * fix(test): point func_env semantic oracle at reachable baseline commit. 5722326 is docs-only atop db559b4 (identical dsl/src/v1 tree) but is not reachable from any ref in a fresh clone — git archive failed in rust_tests. Use db559b4 (on main) and resolve git toplevel for worktree checkouts. Co-authored-by: Cursor <cursoragent@cursor.com> * fix: cargo fmt on cli_run.rs (extra blank line after entry_touches_frontier_seeds). Co-authored-by: Cursor <cursoragent@cursor.com> --------- Co-authored-by: Brian Searls <briansrls@gunb.ai> Co-authored-by: Cursor <cursoragent@cursor.com>

Refresh stale receipts in docs/plans/affected-set-precompute-pruning.md: - Header: PR #6061 (2026-07-01) landed Step 4 consumer-1; update status - Receipt table: add rows for Step 4 consumer-1 GREEN and consumer-2 IN PROGRESS - Line 103: Fix stale claim — function_edited is LIVE (not ABSENT) with three axes (touches_frontier, function_edited, entry_file_touched; landed PR #6065/#6072) Update ROADMAP.md and dag/gunbc/roadmap_authority.dag: - PR #6105 closed 2026-07-02 as measured work-neutral - Successor lane: resolver S2a + skip-before-resolve remaining fix Fix unparseable fixture (src/v2/test/fixture/floor_skip/provenance_fail_closed_contract_test.dag): - .dag grammar has NO comments (// and /* */ are unparseable) - Convert comment block (lines 3-7) to data-string note - Restructure: move data declaration after all imports Refresh ci_floor_plan_witness_test.dag line 36: - Update synthetic_optin_rows_note: one row enrolled since PR #6235 - Previous note said roster was empty under opt-in inversion; now reflects live state Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>

Update affected-set-precompute-pruning.md for consistency: - Line 54-58: Update description of skip decision to reflect all three axes now live (not a gap to fill) - Line 15: Prep step now covers all axes (PR #6065/#6072 + #6234) - Line 18: Consumer-1 receipt clarified — all three axes live, not contradictory 'both' Resolves REQUEST_CHANGES from cursor/composer-2.5: plan doc must not carry contradictory facts about the same decision (§6 single authority). Step 4 consumer-1 is GREEN with all three axes modeled and wired.

) * WIP: Stale-carrier sweep: affected-set plan docs + unparseable fixture * Stale-carrier sweep: affected-set plan docs + unparseable fixture Refresh stale receipts in docs/plans/affected-set-precompute-pruning.md: - Header: PR #6061 (2026-07-01) landed Step 4 consumer-1; update status - Receipt table: add rows for Step 4 consumer-1 GREEN and consumer-2 IN PROGRESS - Line 103: Fix stale claim — function_edited is LIVE (not ABSENT) with three axes (touches_frontier, function_edited, entry_file_touched; landed PR #6065/#6072) Update ROADMAP.md and dag/gunbc/roadmap_authority.dag: - PR #6105 closed 2026-07-02 as measured work-neutral - Successor lane: resolver S2a + skip-before-resolve remaining fix Fix unparseable fixture (src/v2/test/fixture/floor_skip/provenance_fail_closed_contract_test.dag): - .dag grammar has NO comments (// and /* */ are unparseable) - Convert comment block (lines 3-7) to data-string note - Restructure: move data declaration after all imports Refresh ci_floor_plan_witness_test.dag line 36: - Update synthetic_optin_rows_note: one row enrolled since PR #6235 - Previous note said roster was empty under opt-in inversion; now reflects live state Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com> * Fix contradictions in plan doc per review feedback (DESIGN §6) Update affected-set-precompute-pruning.md for consistency: - Line 54-58: Update description of skip decision to reflect all three axes now live (not a gap to fill) - Line 15: Prep step now covers all axes (PR #6065/#6072 + #6234) - Line 18: Consumer-1 receipt clarified — all three axes live, not contradictory 'both' Resolves REQUEST_CHANGES from cursor/composer-2.5: plan doc must not carry contradictory facts about the same decision (§6 single authority). Step 4 consumer-1 is GREEN with all three axes modeled and wired. * Complete contradiction sweep: update all two-axis language to three-axis (DESIGN §6) Fix remaining §6 violations where plan doc carried both 'two axes' and 'three axes' descriptions: - Line 80: Refresh intro prose — all three axes now covered as of Step 4 - Line 95: Acceptance witness (a) — all three axes, not just 'node-frontier + function-edited' - Line 110: Consumer 2 gate — all three conditions required (including touched_entry_files) - Line 150: Acceptance witness (a) spec — full predicate over all three axes - Line 19: Receipt table — consumer 2 gate conditions now include all three All sections now consistently describe the three-axis skip decision: touches_frontier OR function_edited OR entry_file_touched. Plan doc is single authority. * Fix axis naming: third axis is entry_file_touched, not entry_touches_frontier_seeds (DESIGN §3) Corrected predicate names and descriptions per cursor REQUEST_CHANGES: - Line 106: Axis (iii) source is diff_edits.touched_entry_files (not frontier-seeds) - Line 95: Acceptance bar predicate — entry_file_touched (not frontier-seeds proxy) - Line 150: Witness (a) spec — entry_file_touched (not frontier-seeds on non-data-fns) - Line 112: 'Conservative gate' paragraph — all three conditions (not just two) - Line 15: Prep receipt — note that cited witness exercises two axes; full three-axes proof is Step 3 witness (a) Fixes DESIGN §3 single-authority violation where plan doc used wrong name for third axis. * Fix final contradiction: witness (c) acceptance bar requires all three conditions (DESIGN §6) Line 154 acceptance witness (c) was still defining the scoped-diff condition with only two axes (empty frontier AND empty edited_test_fns), contradicting the three-axis gate in floor_precompute_would_skip (src/v2/workflow/affected_set_floor_runner.dag:177-194) and cli_run.rs:4926-4931 which includes touched_entry_files. Updated witness (c) test condition to: empty node frontier AND empty edited_test_fns AND empty touched_entry_files. Single authority now: plan doc matches floor-runner implementation. --------- Co-authored-by: Brian Searls <briansearls1@gmail.com> Co-authored-by: Claude Haiku 4.5 <noreply@anthropic.com>

gunbai-bot Bot marked this pull request as ready for review July 1, 2026 04:01

gunbai-bot Bot mentioned this pull request Jul 1, 2026

Affected-set de-fork — prep: close function_edited axis in .dag disposition kernel + floor witness (witness (a)/migrate/delete deferred) #6065

Closed

gunbai-bot Bot mentioned this pull request Jul 1, 2026

Revert #6078 (Ban unwired models lens) — MAIN-RED: unwired_model_no_unrostered_or_stale returns Bool(false) on live corpus #6084

Closed

gunbai-bot Bot force-pushed the session/vivid-owl-377-witness-a branch from a5d8db4 to 31ed4e0 Compare July 1, 2026 07:23

gunbai-bot Bot mentioned this pull request Jul 1, 2026

Fix main-red: ground spawn-width zero-budget witness on realizer int_min(fallback, exec_cap) #6088

Closed

briansrls and others added 15 commits July 1, 2026 12:44

WIP: Implement affected-set CI de-fork (signed design #5994 / docs/pl…

38821f4

…ans/affe

WIP: Implement affected-set CI de-fork (signed design #5994 / docs/pl…

4495e83

…ans/affe

Clarify witness (a) test is partial disposition coverage, not PROVE g…

4108494

…ate. Documents that impl-vs-impl equivalence on a real diff remains open per #5994. Co-authored-by: Cursor <cursoragent@cursor.com>

WIP: Implement affected-set CI de-fork (signed design #5994 / docs/pl…

ca46117

…ans/affe

cargo fmt: fix cli_run disposition-kernel test formatting.

43a7bf4

Co-authored-by: Cursor <cursoragent@cursor.com>

ci: retrigger rust_tests after sccache fleet flake on c2f8b2f.

316074e

Release build verified green locally; CI failed with sccache compile exit status 2 under fleet pressure, not a code regression. Co-authored-by: Cursor <cursoragent@cursor.com>

WIP: CONTINUE + COMPLETE affected-set CI de-fork (signed design #5994)…

a068a4d

… — prio

WIP: CONTINUE + COMPLETE affected-set CI de-fork (signed design #5994)…

a4a70a4

… — prio

WIP: CONTINUE + COMPLETE affected-set CI de-fork (signed design #5994)…

320324f

… — prio

WIP: CONTINUE + COMPLETE affected-set CI de-fork (signed design #5994)…

cba155d

… — prio

WIP: CONTINUE + COMPLETE affected-set CI de-fork (signed design #5994)…

7084320

… — prio

gunbai-bot Bot force-pushed the session/vivid-owl-377-witness-a branch from cc57034 to 9af2e75 Compare July 1, 2026 12:46

briansrls merged commit 1911eca into main Jul 1, 2026
3 checks passed

briansrls deleted the session/vivid-owl-377-witness-a branch July 1, 2026 13:26

gunbai-bot Bot mentioned this pull request Jul 1, 2026

Merge-admission wiring + roster-trim (#5974); flip HELD pending real CI proof #6080

Merged

2 of 3 tasks

gunbai-bot Bot mentioned this pull request Jul 4, 2026

Stale-carrier sweep: affected-set plan docs + unparseable fixture #6245

Merged

Navigation Menu

Repository navigation

affected-set: PARTIAL witness (a) foundation — edited_test_fns axis (#5994) - #6072

affected-set: PARTIAL witness (a) foundation — edited_test_fns axis (#5994)#6072
briansrls merged 15 commits into
mainfrom
session/vivid-owl-377-witness-a

briansrls commented Jul 1, 2026 •

edited by gunbai-bot Bot

Loading

Uh oh!

gunbai-bot Bot commented Jul 1, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Conversation

briansrls commented Jul 1, 2026 • edited by gunbai-bot Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Green (proved by execution)

Explicitly blocked (NOT in this PR)

Why this PR touches ci_floor_plan.dag (spawn-width witness)

Merge note

Test plan

Uh oh!

gunbai-bot Bot commented Jul 1, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

briansrls commented Jul 1, 2026 •

edited by gunbai-bot Bot

Loading

Why this PR touches `ci_floor_plan.dag` (spawn-width witness)