Repository navigation
board: composite-key issue re-scoped to destination binding; SIMD and dependency scope notes - #1811
Conversation
…dependency scope notes ISS-REPORT-NO-COMPOSITE-KEY-GROUP-FOLD said Report lacked a composite-key group fold. It has one: mask-risc GroupKey::Pair, Quack GroupAddr::Pair and ndarray masked_group_*_pair all landed on 2026-09-23 (869ba24, #1275), the day the issue was filed. lance-graph-report/src/exec.rs has named the real gap since #1331: destination binding / coordinate resolution. Report refuses to lower to dense Pair on purpose. The issue and the D-RPT-5 row now say so, with dated supersession notes. The 32x12 / 4M-row / 25-pass / 177 ms measurement is kept verbatim. Also, on existing rows only (no new debt, per H-12): - TD-SIMD-SWEEP-W2: the ndarray_bridge intrinsics compile only without the default ndarray-hpc feature. - TD-NDARRAY-PATCH-0_16: lance-index 13.0.0 still requires crates.io ndarray 0.16.1 (validated against the published crate); fork path conflicts with E-LANCE-IS-UPSTREAM-AUTHORITATIVE-1. The new entry records CP-1..CP-7, the tests still required (wasm mul_add parity, cross-implementation ternlog parity, bound-ordinal pass count) and the operator decisions (EXO-112, EXO-46, ternlog placement). Docs only. No source, manifest, ABI or contract change. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_015yemQWusMpV2ueqrHkMnkB
|
Important
This repository does not receive automatic reviews because it has fewer than 10 stars.
|
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
Linear updatedAt and Corpus::mark_complete are read receipts, not owner-issued snapshot pins, ACL attestations or complete negative observations; FrameStore refuses stale lab coordinates only and is not proposed for DT64-Q0. Reuse for Q0, #1811 or EXO-46 needs its own attested witness. Docs only, per cross-session review on #1812. Co-Authored-By: Claude <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01XAqKMLnsLW2mFBBbEAT6Ch
Linear updatedAt and Corpus::mark_complete are read receipts, not owner-issued snapshot pins, ACL attestations or complete negative observations; FrameStore refuses stale lab coordinates only and is not proposed for DT64-Q0. Reuse for Q0, #1811 or EXO-46 needs its own attested witness. Docs only, per cross-session review on #1812. Co-Authored-By: Claude <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01XAqKMLnsLW2mFBBbEAT6Ch
Docs only. No source, manifest, ABI or contract change. Do not merge without separate authorization.
What changed
ISS-REPORT-NO-COMPOSITE-KEY-GROUP-FOLD(ISSUES.md) and D-RPT-5 (STATUS_BOARD.md). The issue said Report lacked a composite-key group fold. It has one, verified in source ataa569bc5:GroupKey::Pair { hi, lo, stride }(ir.rs:505), executed for Count / Min / Max / SumSym and power sums (exec.rs:1995-2085);GroupAddr::Pair, lowered atlib.rs:2460and checked against DuckDB intests/duckdb_differential.rs;masked_group_count_u32_pairand its siblings (src/simd.rs:933-957, ndarrayaf4910a1).The fold landed in
869ba24d(#1275) on 2026-09-23, the day the issue was filed. Since #1331,lance-graph-report/src/exec.rs:42-69has named the real gap as destination binding / coordinate resolution, and Report deliberately does not lower to the densePairproduct. The issue and the D-RPT-5 row now say this in dated supersession notes. The original text, including the 32×12 / 4M rows / 25 passes / 177 ms measurement, is kept as history.The open question is restated: can a source-attested semantic destination be bound to a compact, witnessed local coordinate so that Report can use the existing
Local/Viakeyed fold? The issue relates it to D-DT64-Q0, EXO-46, ADMISSION-B and DR64³. None of these provides the binding today.TECH_DEBT.md, existing rows only (no new debt, per H-12):
TD-SIMD-SWEEP-W2: thendarray_bridge.rsraw intrinsics all sit undercfg(not(feature = "ndarray-hpc")), andndarray-hpcis a default feature.TD-NDARRAY-PATCH-0_16: lance-6-era wording refreshed. The publishedlance-index-13.0.0still requires crates.iondarray = "0.16.1", used only invector/bq/builder.rs(EXO-113 confirmed). The fork path conflicts withE-LANCE-IS-UPSTREAM-AUTHORITATIVE-1.New entry
entries/2026-10-27-harvest-integration-composite-key-and-simd-scope.mdrecords the CP-1..CP-7 dispositions, the tests still required and the operator decisions.entries/README.mdis regenerated.Still open
Tests:
mul_addparity, stated per realization and feature set, withrelaxed-simdnamed as in or out of scope. This is separate from thereduce_sumordering difference. Both are ndarray-side.Operator decisions:
TD-NDARRAY-PATCH-0_16.Not in this PR: the stale lance-6.0 comment at
crates/lance-graph/Cargo.toml:27-28. Fixing it is a manifest edit.Checks run locally
All passed:
entries_index.py --check,fruits_index.py --check,epiphany_provenance.py,compact_epiphanies.py --check,append_only_gate.py origin/main(no protected file shrank) andcitation_decay.py --since origin/main(0 new decays).SUPERSESSION-INDEX.mdandFRUITS.mdregenerate byte-identical.Generated by Claude Code