[AdaWorldAPI]

S2 of dr-widefieldmask-dt-classview-v1.md §5: each 256-row band of a slab plane is readable as a borrowed Selection in the same chunk layout as WideFieldMask::words, without copying or allocating. The plane stays one contiguous word slice for mask-RISC.

What changed

• crates/dr256-ternlog-lab/src/lib.rs: MASK_BITS, MASK_WORDS, SLAB_MASKS; Plane::{masks, mask, selections}; Selection<'p> { space, row, words } with has, count, is_empty, same_bits(&WideFieldMask).
• tests/lab.rs: four tests (T1–T4).
• examples/selection_forms.rs: six arms, each checked against a row-at-a-time oracle before timing.

Existing behaviour changed: none (additive).

Evidence

• cargo test -p dr256-ternlog-lab: 19 passed. clippy -D warnings and fmt clean.
• Disable runs, each turned its test red, file restored byte-exact afterwards (anchor asserted to occur once before applying):
  • T1: row * MASK_WORDS + 1; n % 32
  • T2: lo > len; row as u8
  • T3: .min instead of .max in same_bits
  • T4: zeroed space
• Release timing (cargo run --release --example selection_forms), median of 31:

n	contiguous words	fold (count_and3)	borrowed view, per row	owned WideFieldMask, per row	build owned
65 536	259 ns	328 ns	3.51 µs	1.74 µs	32.66 µs
1 000 000	3.27 µs	5.24 µs	28.87 µs	14.50 µs	394.44 µs

Reading of the numbers

Reading the planes row by row is 9–14× slower than one pass over the contiguous words. Owned per-row masks are faster to scan than the borrowed view, but building them costs 120× a contiguous pass. So the per-row Selection is an addressing view (which rows a reader names, compare against a WideFieldMask). Folds keep reading the contiguous plane. Why the borrowed per-row loop is slower than the owned one (three mask() calls with bounds checks per row is the likely cost) is not measured.

Not verified

Only this crate was built and tested. Nothing outside the lab reads Selection yet.

---

Generated by Claude Code

[coderabbitai]

Important

• Trigger review

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

Run configuration

• Source mode: Auto
• Repository: Unavailable
• Branch: Unavailable
• File: No configuration file from this repository applied
• Commit: Unavailable
• Layers: Organization UI
• Review profile: CHILL
• Plan: Advanced
• Run ID: 9d5f1df5-e786-4237-bb9c-fff1494c755d

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

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

• X
• Mastodon
• Reddit
• LinkedIn

[chatgpt-codex-connector]

Codex Review Summary: ✅ Completed

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

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-11-11T19:58:38.484842Z	b1bbc6e	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.