Disable log_ring_keeps_the_owners_slots behind its filed defect

[Japabu]

Summary

• Disables log_ring_keeps_the_owners_slots in src/redlist.rs behind its filed defect, issues/kernel/a-log-rings-owner-is-named-only-when-logd-reads-its-registration.md.
• That issue already recorded 2 red of 13 on origin/main, each red the same signature. The orchestrator's Fast-tier run of PR The kernel boots on QEMU's stock edk2, network included: the direct map ends at the last memory, every kernel root slot exists before the first user space, and a BAR's free run is a gap inside its width's space #563 at d6716fc7 hit it again: /log carries no "===TEST_END test_rs_log_flood exit=0===": the child's flood took the slots its parent's line needed (1917 flood lines in /log). The project rule is that a flaky test is disabled at once, never re-run.
• The issue's status moves from open to expected-red (required by src/redlist.rs's own gate) and gets one added line recording this new sighting and the row it now backs.
• Review round 2: the issue names its owner (toyos/src/log/region.rs's Ring::push; held by the orchestrator, matching the sibling expected-red issues' form) and its Exit line now names the two mutations — let keep = 0; in Ring::push (toyos/src/log/region.rs:202) and a deleted ring.own(pid) (userland/logd/src/origin.rs:176) — that the exit's test must turn red, since no test today exercises that owner decision. Drops a scratchpad log name, a redundant date and a sentence the row and status: expected-red already say.

Gates (host only — this agent never runs QEMU)

• cargo test --lib: exit 0 (384 passed; 0 failed; 1 ignored) — includes the redlist-row check against the issue file.
• cargo test --test toyos-build -- --list: exit 0 — lists log_ring_keeps_the_owners_slots and confirms it is disabled.
• cargo run -- --known-red log_ring_keeps_the_owners_slots: exit 0, answered YES, disabled — it does not run.

Not run here

No guest test, Fast tier or cargo run launch — this is a redlist/issue-only change with no code under test; the orchestrator's own Fast-tier run of #563 is the evidence cited above, not reproduced by this agent.

🤖 Generated with Claude Code

[Japabu]

Review of #569 at 27a6c823, round 1.

Gates at this head: CI run 36392812754 (host), conclusion success, headSha 27a6c823. The reviewer re-ran the host gates because the brief asked for them. cargo test --lib exited 0 (384 passed, 0 failed, 1 ignored). cargo test --test toyos-build -- --list exited 0 and printed [toyos] disabled: log_ring_keeps_the_owners_slots — issues/kernel/a-log-rings-owner-is-named-only-when-logd-reads-its-registration.md. cargo run -- --known-red log_ring_keeps_the_owners_slots exited 0 and answered YES, disabled. The cited 563r4-fast.log has the failure at lines 1161 and 3365, with 400 passed, 1 failed.

Size: +6 −1, all of it a redlist row and issue lines. No production code and no tests.

Checked with no finding:

• The row's shape is right: its multi-line form matches lan_swap's and it sits in order between latency_wake and partition_claim_departure.
• status: expected-red is the status that check() requires.
• "(fast tier)" still holds: tests/toyos.rs:775 still registers the test as Tier::Fast.
• The exit still takes the row off. Deleting the issue file makes check() refuse the row ("is not a file in this tree").

BLOCKER
(none)

NOTE

• toyos/src/log/region.rs:202 — with this row in place, no test exercises Ring::push's owner/pid decision or logd's ring.own(pid) (userland/logd/src/origin.rs:176). slots_left_to_others_are_theirs (toyos/src/log/proof.rs:208) passes keep in by hand on a Heap. log_program_flood (nightly) counts lines and checks no owner line. Neither covers the decision. let keep = 0; at region.rs:202, or deleting ring.own(pid), now passes every test in the tree. This is for the issue's owner: the exit's test must turn that mutation red.
• issues/kernel/a-log-rings-owner-is-named-only-when-logd-reads-its-registration.md — the disable is a compromise, and its record names no holder. The sibling expected-red issues each name one (partition-claim-departure-…: "## Owner … held by the orchestrator"; a-swaps-redial-…: "Owner: … held by the orchestrator"). Add one owner line in that form.

REMOVE

• issues/kernel/a-log-rings-owner-is-named-only-when-logd-reads-its-registration.md:31 — "563r4-fast.log" — a scratchpad name that resolves nowhere on main. bf936c2 cut the same kind of name.
• issues/kernel/a-log-rings-owner-is-named-only-when-logd-reads-its-registration.md:31 — ", 2026-09-28" — the commit d6716fc7 already pins the sighting.
• issues/kernel/a-log-rings-owner-is-named-only-when-logd-reads-its-registration.md:31 — "Disabled for this reason in src/redlist.rs." — status: expected-red and the row already say it.
• PR body — "(log: 563r4-fast.log)" — the body becomes main's merge record, and this path resolves nowhere.

LAND AFTER NAMED CHANGES