Skip to content

D-SPOG-W-0: the Witness slot as a SPOG sub-context under classid-G - #1371

Merged
AdaWorldAPI merged 2 commits into
mainfrom
claude/awesome-volta-xderaa
Oct 6, 2026
Merged

D-SPOG-W-0: the Witness slot as a SPOG sub-context under classid-G#1371
AdaWorldAPI merged 2 commits into
mainfrom
claude/awesome-volta-xderaa

Conversation

@AdaWorldAPI

@AdaWorldAPI AdaWorldAPI commented Oct 6, 2026 •
edited
Loading

Copy link
Copy Markdown
Owner

Probe only. It adds no library code, persists nothing, and moves no CE64 bits.

Claim

A SPOG coordinate is a view over resident coordinates, not a storage layout.

Coordinate Read from
S/P/O the three resident bytes, passed through blind
G graph_of(classid), the existing "g via classid" rule in spog_tenants.rs
sub-context w_slot(), only for a class that declares this reading; slot 0 = no anchor

Witness refines G and never replaces it. Undeclared classes, classes that declare another reading of the slot, and v1/unknown provenance all refuse.

Routing check against the real MailboxSoA::apply_edges

  • Within one graph: the mailbox accepts an edge exactly when its SPOG context matches, on all 64 × 64 slot pairs (64 accepted, 4032 dropped).
  • Across graphs they diverge. Routing reads no classid, so a mailbox accepts an edge from another graph that carries the same slot. This is pinned as a test and recorded as ISS-MAILBOX-ROUTES-WITNESS-WITHOUT-GRAPH. Fixing it is a production change and is not made here.

Tests

  • 7 tests pass; 6 disable runs turn a named test red:
    • G read from the slot
    • G held constant
    • slot 0 treated as a context
    • declaration check skipped
    • provenance check skipped
    • S/P/O permuted
  • 0 allocations. Clippy (test and non-test builds) and fmt are clean.

Board: D-SPOG-W-0 status row and entry, ISSUES.md row; D-GSO-AFF-0 marked Shipped (#1370).

Also in this PR: one unrelated flaky-test fix

r2il-mask-abi-probe/tests/v3_v4_dual_reading.rs kept its heap counter global while the measuring flag was per thread. A parallel test's 364-byte allocation could therefore fail readings_allocate_nothing. The counter is now thread-local. Measured: 20/300 failures before, 0/300 after. This fixes the red test job on this PR; the change is test-only.

🤖 Generated with Claude Code

https://claude.ai/code/session_01T8w7Ucm7NhitfTduFKmhYx

Generated by Claude Code

Generated by Claude Code

@claude
D-SPOG-W-0: the Witness slot as a SPOG sub-context under classid-G
52703d5 
A SPOG coordinate read from (classid, CausalEdge64) with no stored column:
S/P/O are the resident bytes, G is graph_of(classid) (the existing "g via
classid" rule), and the Witness slot is a sub-context inside G for classes
that declare that reading. Slot 0 reads as no anchor. Undeclared classes,
another reading of the slot, and v1/unknown provenance refuse.

Compared with the real MailboxSoA::apply_edges routing: within one graph
it accepts exactly the edges in the same SPOG context on all 64 x 64 slot
pairs; across graphs it diverges, because routing reads no classid.
Recorded as ISS-MAILBOX-ROUTES-WITNESS-WITHOUT-GRAPH; not fixed here.

7 tests, 6 disable runs red, 0 allocations. Board: D-SPOG-W-0 row and
entry, ISSUES row; D-GSO-AFF-0 marked Shipped (#1370).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01T8w7Ucm7NhitfTduFKmhYx
@coderabbitai

coderabbitai Bot commented Oct 6, 2026 •
edited
Loading

Copy link
Copy Markdown

Warning

Review limit reached

Your organization has reached its usage spending cap. Adjust your spending cap in the billing tab.

Next included review available in 21 minutes.

Check out review usage here.

View limit details

Limit details: You’ve used the included review currently available. Your 69 included PR review attempts over the past 7 days set your current allowance at 1 review per hour.

Learn how review limits work.

Review configuration:

⚙️ Run configuration
  • Configuration used: Organization UI
  • Review profile: CHILL
  • Plan: Essentials
  • Run ID: 9de98ccf-da2d-48b1-b8f6-7e9f57313524
📥 Commits

Reviewing files that changed from the base of the PR and between 6f0c739 and 4cf13d0.

📒 Files selected for processing (8)
  • .claude/board/ISSUES.md
  • .claude/board/STATUS_BOARD.md
  • .claude/board/entries/2026-10-06-spog-witness-sub-context.md
  • .claude/board/entries/README.md
  • crates/cognitive-shader-driver/Cargo.toml
  • crates/cognitive-shader-driver/examples/README.md
  • crates/cognitive-shader-driver/examples/spog_witness_probe.rs
  • crates/r2il-mask-abi-probe/tests/v3_v4_dual_reading.rs
  • Autopilot · 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.

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

@cursor

cursor Bot commented Oct 6, 2026

Copy link
Copy Markdown

Bugbot couldn't run - usage limit reached

Bugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit.

A user or team admin can review and increase usage limits in the Cursor dashboard.

(requestId: d526f83e-9fa0-4c5a-ae75-a0a0f48ea56c)

@AdaWorldAPI
AdaWorldAPI marked this pull request as ready for review October 6, 2026 19:05
@cursor

cursor Bot commented Oct 6, 2026

Copy link
Copy Markdown

Bugbot couldn't run - usage limit reached

Bugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit.

A user or team admin can review and increase usage limits in the Cursor dashboard.

(requestId: b1aa0b2f-5fab-4c1b-89b6-2bc64fce7b60)

@chatgpt-codex-connector

chatgpt-codex-connector Bot commented Oct 6, 2026 •
edited
Loading

Copy link
Copy Markdown

Codex Review Summary

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

Review Status Commit Review trigger
📝 Code Review ✅ Completed 2026-10-06T19:08:17.211697Z 52703d5 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.

@AdaWorldAPI Claude

AdaWorldAPI commented Oct 6, 2026

Copy link
Copy Markdown
Owner Author

test-with-coverage failed while compiling a third-party crate, not this PR's code. xet-client 1.6.0 does not compile against xet-core-structures 1.7.0 (E0061: serialize_xorb_subset_with_expiry gained a 4th argument).

  • Chain: lance-io 12 → opendal 0.59.4 → opendal-service-hf → hf-xet 1.6.0 → xet-client 1.6.0.
  • Requirements: all are caret 1.6.0. The repo tracks no lockfile, so CI resolved fresh and mixed the two versions, apparently while xet-core's 1.7.0 release was only partly on crates.io.
  • Fresh resolve now: a local cargo update --dry-run moves hf-xet, xet-client, xet-core-structures, xet-data and xet-runtime to 1.7.0 together. A re-run should get a consistent set.
  • This PR: it touches only cognitive-shader-driver examples and board files.

Re-running the failed job once to confirm. If it fails again on the same error, the fix is a dependency pin, which would need its own PR.

Generated by Claude Code

@claude
r2il-mask-abi-probe: per-thread heap counter in v3_v4_dual_reading
4cf13d0 
The heap meter kept MEASURING per thread but BYTES global. The test
harness runs tests on parallel threads, so loco_execution_needs_a_gathered_copy
(which measures its own 364-byte allocation) could land in
readings_allocate_nothing's window and fail it with left: 364.

BYTES is now thread-local like MEASURING. Measured locally: the old
version fails 20/300 runs of the test binary, this one 0/300. The
can-fire half (materialize_indices allocates) still passes.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01T8w7Ucm7NhitfTduFKmhYx
@cursor

cursor Bot commented Oct 6, 2026

Copy link
Copy Markdown

Bugbot couldn't run - usage limit reached

Bugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit.

A user or team admin can review and increase usage limits in the Cursor dashboard.

(requestId: 7eccc70b-e733-46f5-a464-a167bb8cbbf8)

@AdaWorldAPI
AdaWorldAPI merged commit 705509f into main Oct 6, 2026
11 checks passed
AdaWorldAPI pushed a commit that referenced this pull request Oct 6, 2026
@claude
D-ALPHA-G-0 + D-WA-0: WorldG from the Alpha route; Witness x Angle as…
b3c5fda 
… a declared local source

D-ALPHA-G-0 (alpha_world_provenance_probe): an epistemic event is
(Alpha-attended address, CausalEdge64). WorldG = graph_of(addr), the order of
worlds is the SpogTenants route. Same edge bits in two worlds are two
events; replaying the same claims recovers the world sequence with all-zero
edges; changing edge fields never moves the world; partitioned by recovered
world each MailboxSoA holds only its own deliveries through the unchanged
apply_edges; an undeclared world is refused at the tunnel (NoTenant).
5 tests, 5 disable runs red.

D-WA-0 (witness_angle_probe): (class, generation, w_slot, bits 43..45) ->
SourceCoord under a probe-local sparse law. Same bits mean different sources
in another class and another generation; classes reading the bits as the
pathology triad or a cohort WitnessTable refuse, as do undeclared classes and
generations, Witness 0 and v1/unknown provenance. World from the attended
key, never the edge. 0 allocations; 7 tests, 7 disable runs red.

Reframe: ISS-MAILBOX-ROUTES-WITNESS-WITHOUT-GRAPH resolved as a false
positive. The #1371 cross-graph case bypasses the split tunnel;
apply_edges has no production caller; MailboxSoA stays graph-blind and gains
no G check. spog_witness_probe docs corrected in place.

Board: STATUS_BOARD (D-SPOG-W-0 Shipped #1371; D-ALPHA-G-0, D-WA-0 rows),
ISSUES status, entry with the 10-point inventory, entries index.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01T8w7Ucm7NhitfTduFKmhYx
This was referenced Oct 6, 2026
Merged
Merged
Merged
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@AdaWorldAPI @claude