Conformance: promote the 10 optimizing-JIT PR-249 cases

[chrisbbreuer]

Parent: #143
Depends on: #146, #132, #133
Roadmap: #134

Ten pinned PR-249 executables exercise optimized-tier trap polling, code/call-link artifact lifetime, retirement/jettison, stale-base invalidation, Class-A GC stops, value-profile publication, compile counters, and IC dictionary transitions. They remain blocked until real optimizing-tier behavior exists; shell stubs or unconditional counters are not promotion.

Exact scope and owners

• checktraps-invalidation.js — optimizer polls and invalidation points (JIT: add optimizing profiles, IR, and tier publication #431, JIT: implement OSR exits, deoptimization, and precise GC #432, JIT: make trap polling and optimized artifact lifetime concurrency-safe #433).
• cve/mc-aint-poll-resume-stale-elided.js — no-GIL poll/park, jettison, and deopt resume (JIT: implement OSR exits, deoptimization, and precise GC #432, JIT: make trap polling and optimized artifact lifetime concurrency-safe #433).
• cve/mc-code-calllink-writer-writer.js — atomic call-link publication and epoch retirement (JIT: make trap polling and optimized artifact lifetime concurrency-safe #433).
• cve/mc-dos-retired-artifact-churn.js — bounded handler/code retirement and GC accounting (JIT: make trap polling and optimized artifact lifetime concurrency-safe #433).
• cve/mc-jit-stale-base-grow-oob.js — array base/length guards across polls (JIT: lower verified packed-array regions into the native tier #133, JIT: implement OSR exits, deoptimization, and precise GC #432, JIT: make trap polling and optimized artifact lifetime concurrency-safe #433).
• cve/mc-safe-gcwait-vs-classa-stop{,-noropevariant}.js — Class-A invalidation ordered against shared GC (JIT: implement OSR exits, deoptimization, and precise GC #432, JIT: make trap polling and optimized artifact lifetime concurrency-safe #433).
• cve/mc-val-fire-vs-link.js — advisory profiles plus link-time watchpoint revalidation (JIT: add optimizing profiles, IR, and tier publication #431, JIT: make trap polling and optimized artifact lifetime concurrency-safe #433).
• jit/foreign-reify-getbyid-converges.js — guarded property lowering, side-exit convergence, and real compile counts (JIT: lower guarded property regions into the native tier #132, JIT: add optimizing profiles, IR, and tier publication #431, JIT: implement OSR exits, deoptimization, and precise GC #432).
• jit/ic-publish-reset-loops.js — atomic property-IC publication/reset and dictionary transitions (JIT: lower guarded property regions into the native tier #132, JIT: make trap polling and optimized artifact lifetime concurrency-safe #433).

Acceptance

• Pin the exact 10-file list and map every hook/premise to a concrete JIT: build the optimizing tier with deoptimization and precise GC #146 subsystem and test owner.
• Implement trap polling/invalidation and resumable optimized execution used by the corpus.
• Implement code artifact, call-link, IC, retirement, jettison, and compile-count semantics through the real optimizing tier.
• Preserve GC stack maps, deoptimization, concurrent invalidation, exception edges, and interpreter/baseline fallback.
• Replace JSC-only option spellings with explicit profile mappings only where semantics are equivalent.
• Promote each case only after default/no-GIL differential and sanitizer evidence passes.
• Update the PR-249 inventory and exact runtime totals without hidden skips.

Current checkpoint

The verifier still reports 243 promoted, 10 optimizer-blocked, and 6 terminal executables out of 259. The real optimizer now has exact terminal side exits, invalidation-polled loop OSR, precise live relocation (#437 complete), region-scoped conditional-exit chains, sequential shared joins, shared/independent latches, and conditional continue chains with three or more backedges through 48bfe60/b80e81f7/3d78eaf2, including general reducible and fused nested-loop regions. Exact termination checkpoints, statement-hook fallback, and fail-closed irreducible control are covered. The latest combined optimizer TSan gate passes 42/42 lowering and 80/80 runtime/context tests.

The corpus remains blocked on #132/#133 native property/array regions, #433 call-link/IC retirement and Class-A stop semantics, backend evidence, and explicit hook/profile promotion. No fake compile counters or shell stubs are exposed. Benchmark scores are unchanged because the published workloads do not use this new CFG shape.

[chrisbbreuer]

Owner #431 has begun in ec65181 with profiles and deterministic CFG planning, but no optimizer plan executes yet. Therefore all ten PR-249 optimizer files remain blocked and the promoted/terminal totals are unchanged; no case is being relabeled from foundation-only evidence.

[chrisbbreuer]

74fd22d adds the first executable optimizer artifact, but only for exact constant SSA returns. None of the ten PR-249 optimizer cases qualify yet, so the blocked/promoted/terminal totals remain unchanged.

[chrisbbreuer]

Prerequisite progress landed in 951c5c5: hot guarded Number-parameter SSA now executes through the real optimizer artifact, with install-only compile counts and exact pre-accounting fallback. This is foundation for the profile/counter requirements here, but it does not yet implement the trap polling, property ICs, call-link lifetime, deoptimization, or retirement semantics exercised by the pinned files.

Accordingly, no corpus case is promoted and the inventory remains exactly 243 promoted, 10 optimizer-blocked, and 6 terminal executables.

[chrisbbreuer]

#431 is now complete at e3f411a, including real guarded numeric/control optimizer execution and install-only counters. The ten pinned cases still require #432/#132/#133/#433 semantics, so none are promoted yet and the inventory remains 243 promoted, 10 optimizer-blocked, and 6 terminal executables.

[chrisbbreuer]

Optimizer prerequisite progress: 561a516 removes the equal-cost premise from the first inner loop diamond and preserves exact path-specific steps, budget, checkpoint distance, and backedge recovery in both longer-arm directions. The 10 PR-249 executables remain optimizer-blocked; this is real tier behavior but does not yet satisfy artifact/call-link/IC retirement, live in-native relocation, broader CFG, or explicit corpus hook/profile evidence.