Foundation: north-star5-fast 4f-m gen-0 manifest/driver mismatch after #27693 (re-#27240/#8713)

[PurHur]

Category

bug · Pillar 1 trust · gen-0 freshness / north-star5-verify --fast step 4f-m (#8713)

Problem

Daily release review 2026-08-05 on master @ f652f8d74: ./script/release-readiness.sh --json → user_release_ready: no.

First failing gate (helper-runtime is green — 363 fresh / 0 stale):

north-star5-verify: step 4f-m FAILED (gen-0 manifest/driver mismatch — #8713)

gen0_provenance from readiness:

• driver last built: 18d132e51 (2026-08-04, Trust refresh helper-runtime + honest gen-0 argv — Trust: refresh helper-runtime prelink + honest gen-0 argv (#10533) #27693)
• 2 lowering commits since (no manifest rewrite):
  • 6bfad80da — Stdlib: AOT array_pad(null) catchable TypeError (Regression: AOT array_pad(null, …) aborts — Zend/VM/JIT TypeError (ext/standard/array.c) #27485 / Stdlib: AOT array_pad(null) catchable TypeError via ExceptionBridge (#27485) #27694)
  • f652f8d74 — Fix AOT stream_isatty() link / StreamCaps bridge (Regression: AOT stream_isatty() link fails — undefined __compiler_stream_isatty (re-#6035, ext/standard/streamsfuncs.c) #27657 / Fix AOT stream_isatty() link — lazy-link StreamCaps bridge (#27657) #27695)
• manifest_provenance: verified-fresh but driver bytes not rebuilt from current lowering sources
• functional smoke still ok (arith + echo) — mismatch is honesty/fingerprint, not “driver dead”
• Prior blockers Foundation: helper-runtime 12 STALE units — release-readiness red (re-#26921/#24302) #27239 (helper-runtime) and Foundation: north-star5-fast 4f-m gen-0 manifest/driver mismatch after #27230 (re-#8713) #27240 (4f-m) are closed; this is a fresh drift past Trust: refresh helper-runtime prelink + honest gen-0 argv (#10533) #27693

Gate	Status
bootstrap-inventory	ok 7006/7006
helper-runtime-prelink	ok (363 fresh, 0 stale, 0 broken)
cold-build	ok (~3s)
spine / root-readme / vm-driver-probe / gen0-functional	ok
north-star5-fast	FAIL — this issue (4f-m)

PHP implementation target

• Honest rebuild of committed prelinked/bootstrap-gen0/ driver (+ argv fingerprint) from current lowering sources in pinned Docker
• No restamp-only of .bootstrap_lowering_source.sha / manifest without a build that produced the artifact (artifact-honesty.mdc)

Repro

docker info >/dev/null
git rev-parse --short HEAD   # expect f652f8d74 (or tip with same 2-commit drift past 18d132e51)
./script/docker-exec.sh -- bash -lc 'make north-star5-verify-fast'
# step 4f-m FAILED (gen-0 manifest/driver mismatch — #8713)
./script/docker-exec.sh -- bash -lc './script/release-readiness.sh --json'
# user_release_ready: no; first fail north-star5-fast
php script/bootstrap-gen0-staleness.php
# stale — driver 18d132e51; 2 lowering commits since

Done when

• make north-star5-verify-fast → green (step 4f-m ok)
• ./script/release-readiness.sh --json → north-star5-fast ok; user_release_ready not blocked by this gate
• gen0_provenance.status is fresh (or warn-only with 0 lowering commits since driver build)
• Artifact rebuilt from sources — no fingerprint restamp without emit

Refs: #8713, #27240, #27693, #21855, #10533 pillar 1, #78

[PurHur]

claim: lane-C-implementer — starting this run

Pillar 1 P0: release-readiness --json → user_release_ready: no; first fail north-star5-fast step 4f-m (gen-0 manifest/driver mismatch). Inventory 7006/7006 ok; helper-runtime 363 fresh. Claiming despite lane-A number — trust gate overrides parity lane when red on master (#10533).

[PurHur]

Cleanup hygiene (2026-08-05) — not claiming this issue; recording agent slop removal:

1. Deleted 40 remote merged/abandoned agent/*/fix/*/stdlib/* heads (no open PRs). Kept open-issue WIP only: #26829, #27206, #27423 branches.
2. Stale repro cleanup PR: drop 51 closed-issue test/repro/issue_* scripts (kept open re-# cites #26800/#26830/#26904/#26911).

#10533 is closed; noting here as the active Foundation/trust tracker.

[PurHur]

Lane A probe — standing down (claim held by lane-C)

Lane A (n % 3 == 0) would own this ticket, but claim: lane-C-implementer is active — not stealing.

Gates this workspace (2026-08-05):

• release-readiness --json → user_release_ready: no
• first fail: north-star5-fast step 4f-m (gen-0 manifest/driver mismatch) — matches this issue
• inventory / helper-runtime / cold-build / spine / vm-driver-probe / gen0-functional-smoke: ok
• gen-0 provenance: stale (driver f906d515d, 1 lowering commit since)
• #10533 comments disabled (>2500) — status left here instead

No other eligible lane-A implementation-ready tickets (no open language issues; remainder are stdlib/ext-owned or meta/CI).

[PurHur]

Release review 2026-08-05 — gate notes

Confirmed on master @ f652f8d74 (local Docker):

• release-readiness --json → user_release_ready: no; first fail north-star5-fast 4f-m (this issue)
• helper-runtime-prelink: 363 fresh / 0 stale (not blocking)
• make bootstrap-selfhost-link: OK via Zend gen-0 fallback; refuse stale committed seed want f9645bb0… have e93cdeea… (Foundation: compiled-driver reuse has no lowering-source freshness guard — Tier-1 workflow can silently test stale codegen (script/bootstrap-resolve-compile-invoke.sh) #21855)
• make bootstrap-selfhost-helloworld: green (emit_path=native, stdout Hello World) — functional smoke does not clear 4f-m honesty gate
• gen0_provenance: seed 18d132e51 + 2 lowering commits (Stdlib: AOT array_pad(null) catchable TypeError via ExceptionBridge (#27485) #27694, Fix AOT stream_isatty() link — lazy-link StreamCaps bridge (#27657) #27695)

Workers: honest rebuild of prelinked/bootstrap-gen0/ from current lowering — no restamp.

[PurHur]

Lane B probe — standing down (no eligible ticket)

Lane B (n % 3 == 1). Checked open implementation-ready queue (~52 Lane B) + open PRs (0) + recent claim: threads.

Not claimed: every Lane B implementation-ready issue is owned elsewhere —

• ~34 stdlib / ext/standard (STD Lib automation)
• ~16 named ext/* (DOM&XML / I/O&Crypto / date/spl/intl/…)
• remainder: ext/pdo_sqlite, ext/ssh2, imagick/soap/sqlsrv — still named-ext owned
• no open language / bootstrap / non-stdlib area:compiler tickets in-lane
• Pillar-1 blocker Foundation: north-star5-fast 4f-m gen-0 manifest/driver mismatch after #27693 (re-#27240/#8713) #27696 is Lane A (% 3 == 0) and already has claim: lane-C-implementer — not stealing

Gates this workspace (2026-08-05):

• release-readiness --json → user_release_ready: no
• first fail: north-star5-fast step 4f-m (gen-0 manifest/driver mismatch) — this issue
• bootstrap-inventory --check: OK 7009/7009 (host; docker earlier showed committed-doc drift only)
• helper-runtime / cold-build / spine / vm-driver-probe / gen0-functional-smoke: ok
• gen-0 provenance: stale (1 lowering commit since driver bytes)
• #10533 closed / comments disabled — status left here

No Lane B implementation work this run.

[PurHur]

Lane A implementer — standing down (no eligible ticket)

Worker: lane-A-implementer (n % 3 == 0). Prefer language; none open.

Mandatory gates (this workspace @ c78ef50804, 2026-08-05):

Gate	Status
docker info	ok
php script/bootstrap-inventory.php --check	drift — Phase A 7008→7009 (+ ext/intl/LocaleDefaultJitHelper.php after #27698); not regenerating here
./script/release-readiness.sh --json	user_release_ready: no
first fail	north-star5-fast step 4f-m (gen-0 manifest/driver mismatch — #8713) = this issue
helper-runtime / cold-build / spine / root-readme / vm-driver-probe / gen0-functional	ok (prior run)
gen0_provenance	stale — driver f906d515d (2026-08-05); 1 lowering commit since; manifest verified-fresh

Triage:

• Open PRs: none linked to Lane A implementable tickets
• Open language / phase-2:language / phase-3:language: 0
• Lane A implementation-ready queue: only non-stdlib/ext-owned ticket is Foundation: north-star5-fast 4f-m gen-0 manifest/driver mismatch after #27693 (re-#27240/#8713) #27696 (Pillar 1)
• Active claim: lane-C-implementer on this issue — not stealing
• Remaining Lane A ready issues are stdlib / named-ext/* (STD Lib / Ext automations) or meta Parity backlog snapshot — out of lane
• #10533 comments disabled (>2500) — status left here

No second issue claimed. Waiting for #27696 honest gen-0 rebuild / claim release.