D-HOLO H14b (withdrawn): positional lane mask M; measurements kept, no layout verdict by AdaWorldAPI · Pull Request #1569 · AdaWorldAPI/lance-graph

AdaWorldAPI · 2026-10-15T21:50:03Z

Withdrawn (2026-10-15, before merge)

M is withdrawn as a candidate layout for thought-first C, and so is any A/B/C/M layout verdict drawn from these numbers. Comparing M against B or C was a category error.

H14 and H14b measured whole-row reconstruction: full-row read, session read, and F3 materialization. C does not work that way. In C a thought program selects its lane tables and reads each one at the shared address, Thought(t)[Table(a)[i] : Table(b)[i]]. No row is assembled unless a named terminal asks for one.
Mask[i] transposes what C's per-(thought, lane) presence planes already hold into a row-oriented index: bit l of Mask[i] is bit i of P_{t,l}. That puts row-first discovery back in front of positional access.
The thought knows which lanes its program requests. It does not need to know which lanes are populated at each row; C answers that with one presence test per requested table.
The measurements, tests and commits stay as a record. They say nothing about C's layout. M stays in the probe only as an experimental structure for rebuilding whole rows.
Follow-up: D-HOLO-H14c (D-HOLO H14c: thought-first positional reads, Hamming presence and Lance masked I/O #1573).

What the branch contains

H14's per-lane tables (arm C) plus one u32 per row per thought naming the lanes that row changed (arm M). The pre-registration was committed before any code (ec392689).

Recorded measurements (v4, `tier: v4 (avx512f)` printed; one run, one host, one class)

These are not a verdict on C.

M's full-row and session read, against B at k <= 4: within 1.25× in all 6 cells at 1 % dirty; above it in 9 cells at 10 % and 100 % dirty (up to 1.63× full-row, 5.85× session).
M's merge against C's: faster in 35 of 36 cells (C/M 0.98–3.69×).
M's bytes to the writer: at most B's, but by construction rather than by measurement (2f07941c, from the review). B was charged a per-row offset that M was not, and counted the same way the two are equal.

Tests

11 tests pass. Both pre-registered disable runs turned red: a merge over every touched table, and a mask that drops a lane.
clippy with -D warnings and fmt are clean.

Board

Entry 2026-10-15-d-holo-h14b-lane-mask.md, with its withdrawal note.
STATUS row D-HOLO-H14b: WITHDRAWN.

🤖 Generated with Claude Code

https://claude.ai/code/session_011BMJnDBAfzcuQh5HeT39Rh

Generated by Claude Code

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_011BMJnDBAfzcuQh5HeT39Rh

…board entry, status Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_011BMJnDBAfzcuQh5HeT39Rh

coderabbitai · 2026-10-15T21:50:13Z

Important

🔍 Trigger review

This repository does not receive automatic reviews because it has fewer than 10 stars.

⚙️ Run configuration

Configuration used: Organization UI
Review profile: CHILL
Plan: Advanced
Run ID: cc831c75-c154-4a8a-bc86-76200571ec63

Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

Warning

Billing warning: we have not been able to collect payment for this subscription for more than 72 hours. Please update the payment method or pay any pending invoices in Billing to avoid service interruption.

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

_{Comment @coderabbitai help to get the list of available commands.}

M's per-row lane mask restates what C's per-lane presence planes already hold, and H14/H14b mostly measured whole-row reconstruction, which is not C's primary operation. The numbers stand; their reading as a verdict on C is withdrawn, as is the density-based layout switch. Follow-up: H14c. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_011BMJnDBAfzcuQh5HeT39Rh

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: d311cb3726

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

chatgpt-codex-connector · 2026-10-15T23:48:38Z

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-10-15T23:48:36.653013Z	`d311cb3`	Draft marked ready

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

ArmB::bytes charges a 4-byte value offset per dirty row that ArmM::bytes omits, though both can derive it from the masks, so bm == bb - 4 * dirty rows and `bm <= bb` cannot fail. The plan, the board entry, the doc comment and the printed label now say so. No numbers change. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_011BMJnDBAfzcuQh5HeT39Rh

Resolves the entries index and STATUS_BOARD conflicts from main, keeping the H14c row. PR_ARC_INVENTORY and LATEST_STATE record #1569 as merged and withdrawn (no layout verdict on C); the H14b STATUS row now says its writer-byte rule held by construction. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_011BMJnDBAfzcuQh5HeT39Rh

claude added 5 commits October 15, 2026 21:36

D-HOLO H14b: pre-register C with a positional lane mask per row

ec39268

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_011BMJnDBAfzcuQh5HeT39Rh

D-HOLO H14b: tests for arm M (red against a stub)

eedb9f1

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_011BMJnDBAfzcuQh5HeT39Rh

D-HOLO H14b: arm M, merge reads only the tables the row's mask names

52d0c74

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_011BMJnDBAfzcuQh5HeT39Rh

D-HOLO H14b: time M against A, B, C in the same run; rule lines

7b8ad98

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_011BMJnDBAfzcuQh5HeT39Rh

D-HOLO H14b: v4 results, prediction falsified above 1 % dirty; plan, …

27c51d5

…board entry, status Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_011BMJnDBAfzcuQh5HeT39Rh

AdaWorldAPI mentioned this pull request Oct 15, 2026

D-HOLO H14c: thought-first positional reads, Hamming presence and Lance masked I/O #1573

Merged

AdaWorldAPI marked this pull request as ready for review October 15, 2026 23:46

chatgpt-codex-connector Bot reviewed Oct 15, 2026

View reviewed changes

Comment thread crates/lance-graph-mask-risc/examples/holo_lane_storage_probe.rs

AdaWorldAPI changed the title ~~D-HOLO H14b: inverted fold + positional lane mask; prediction falsified above 1 % dirty~~ D-HOLO H14b (withdrawn): positional lane mask M; measurements kept, no layout verdict Oct 15, 2026

AdaWorldAPI merged commit 09f3156 into main Oct 16, 2026
18 checks passed

Navigation Menu

Repository navigation

D-HOLO H14b (withdrawn): positional lane mask M; measurements kept, no layout verdict - #1569

D-HOLO H14b (withdrawn): positional lane mask M; measurements kept, no layout verdict#1569
AdaWorldAPI merged 7 commits into
mainfrom
ccr-9f3b8075-74u47l-holo-h14b

AdaWorldAPI commented Oct 15, 2026 •

edited

Loading

Uh oh!

coderabbitai Bot commented Oct 15, 2026 •

edited

Loading

Uh oh!

chatgpt-codex-connector Bot left a comment

Uh oh!

Uh oh!

chatgpt-codex-connector Bot commented Oct 15, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Conversation

AdaWorldAPI commented Oct 15, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Withdrawn (2026-10-15, before merge)

What the branch contains

Recorded measurements (v4, tier: v4 (avx512f) printed; one run, one host, one class)

Tests

Board

Uh oh!

coderabbitai Bot commented Oct 15, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

Uh oh!

chatgpt-codex-connector Bot commented Oct 15, 2026

Codex Review Summary

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

AdaWorldAPI commented Oct 15, 2026 •

edited

Loading

Recorded measurements (v4, `tier: v4 (avx512f)` printed; one run, one host, one class)

coderabbitai Bot commented Oct 15, 2026 •

edited

Loading