[zhoubot]

Preserve runtime physical-column geometry through binary elementwise operations, scalar fill and row expansion. Validate bounds using instance dimensions instead of the DYNAMIC sentinel while retaining the original allocation and layout.

Expose PTO_PE_DOMAIN(mask) for compiler-owned participation boundaries and propagate TileArray participant masks through output references, assembly, views and TMOV writers. Reject invalid masks and unsupported selected-slot writers. This supports nested PE0-only Tile/element chains inside the original four-PE QLI driver.

Correct lifetime fixtures to publish complete Local parents and reject OPEN-parent escapes; make the header-defined layout formatter inline. Native test matching accepts the exact old and new TLSU/CUBE header spellings. Host make check, focused API tests and combined native/runtime profiles pass. Full target verification still has the existing SharedTLoad handle-store gap; general pointer-TLOAD partial-parent transport outside marked regions is not claimed scalar-free.

Qualified on frozen LLVM a08e1c228dbe8b5fe45a6c613fec6a78499ab920, API 1f72e918fcbb72569d8e28dda11f7c579a06dab2, benchmark code 693d25349b9e17bad57fe7d281be258140d649ba, and unchanged gfrun 0a237ccfe8404d73d92dd144a6af539b2b0ebc5a. The subsequent benchmark documentation commit records these exact code pins. All runtime cases keep the 120-second gfrun timeout and original resource limits.

• Element suite 64/64 (455 memory segments), core TopK 777 9/9 segments, solution regressions 21/21.
• QLI key48 / histogram72 / norm13 / popn18 / static TopK7 / original5 / dynamic4; additionally original four-PE4 and upstream four-PE2, each comparing 18 memory segments.
• Original low-precision matmul7/7; predicate original and edge cases; MoE boundary driver with 11 lengths and repeated calls; D6 C48 and issue130 C16.
• Nested PE-domain assembly/view/TMOV/element chain: two segments, 16,768 bytes, all 21 B.IOT binders use PE0 mask1000.
• Focused API6 + native adapter42 + PE-domain7 + exact descriptor glob24 + upstream lifetime/header16 = 95/95; three additional late-copy/legacy tests pass. Issue60 remains a separately recorded negative, not feature closure.
• Independent final audit: PASS, 3,795 hash records verified, zero errors/pending. Local report /tmp/pto-element-fusion-20261019/final-v19-audit.json (SHA256 ec21921ce2f554e78131ee12be39e9ebc151d7a62d999f1093a51aa814eed0c4).

This qualifies the declared gfrun profiles. gfsim/timing, forced Tile spill runtime, and lifetime retirement semantics are not qualified. The original application inventory remains 17/21 with four pending profiles, distinct from solution21/21. LLVM still has 28 legacy pragma fixtures outside this closure and a valid guard-correlated issue60 proof gap. Broad MC accounting is 82 PASS, 1 XFAIL, and three historical failures (CUBE_Mamulb, MTC_Tloadstore, pto-tileop-macro-ranges); no all-LLVM-tests-green claim.

Companion PRs: LLVM #131, TileOP API #281, SuperNPUBench #252, gfrun #949.

[VV0003]

Superseded by #292: the multi-PE integration branch merged this PR history and then re-landed the same features with follow-up adjustments (participant-mask coverage fitted to supported ElementTile profiles, standalone Tile headers in PE-domain tests, runtime physical-column fixes). The integrated result is squash-merged into linx as b5fbae192912. This PR head now conflicts with linx and would revert those adjustments, so closing as superseded instead of merging.