Skip to content

perf: walk slot windows older instead of unbounded re-reads - #13236

Merged
iamwhatever merged 1 commit into
mainfrom
perf/bounded-backward-walk
Oct 1, 2026
Merged

iamwhatever merged 1 commit into
mainfrom
perf/bounded-backward-walk

Conversation

@buluoray

@buluoray buluoray commented Sep 24, 2026 •

Copy link
Copy Markdown
Contributor

Problem / Motivation

Goal: a long chat session stops re-downloading its entire history on every turn end, reconnect and session switch, without losing any scrollback the user paged in.

refreshSlot and switchSlot fall back to an unbounded slot-detail read (limit omitted = the whole chained transcript) whenever a count-matched page cannot be proven to cover the rows the tab already holds:

path condition that forces the unbounded read
refreshSlot view holds more than REFRESH_LIMIT_CEILING (500) rows
refreshSlot the 500-row page is disjoint from the view
switchSlot slotCoverageShortfall > 0 (cached rows the window did not cover)

On a long session every one of those is permanently true, so every chat_done and every WebSocket reconnect re-reads the entire history. Measured on a 12,178-row session: 11,134 rows fetched per turn, switchSlot re-fetching the full transcript on every switch, never converging. The scroll inspector on that session reported h=941519px, a single reprice moving 797,415px.

Why it matters

Anyone running a long session pays for the whole transcript on every chat_done and every reconnect: on the measured 12,178-row session that is 11,134 rows per turn, a switchSlot that never converges, and a scroll surface that reprices by hundreds of thousands of pixels. The cost grows with the session, so the sessions people use most are the ones hit hardest.

Each fallback was added deliberately (#3240, #6839, #6947, #8574) as the cheapest proof of one invariant: a replacing reducer must not delete scrollback the user paged in. Two premises made the price acceptable then -- 2,644 rows = 6.2MB/1s, and rare triggers -- and neither holds on a session past the ceiling, where the trigger is every turn.

Not a goal

What changed (motivation → approach → change)

The invariant is kept; the proof is replaced. A window that misses the held rows is extended older, one bounded page at a time (walkWindowBackTo) until its oldest row anchors a held row -- the exact condition under which olderHeadAbovePage keeps the rows above it. The walk pays for the rows the server gained plus one overlapping page. Every request carries limit <= SLOT_DETAIL_MAX_LIMIT; the reducers are untouched.

  • refreshSlot: REFRESH_LIMIT_CEILING becomes a clamp instead of an unbounded fallback; an empty view asks for the floor instead of the whole transcript; the disjoint-page case walks.
  • switchSlot: an observed coverage shortfall walks instead of refetching unbounded. slotCoverageShortfall itself is unchanged -- it stays the conservative multiset, the walk answers "above the anchor vs. in a hole".
  • walkWindowBackTo reaches on either count the reducers accept: the window's oldest row anchors a held row, or the oldest held row anchors a window row (window is a superset). Both go through idAnchorsOneRow, so a repeated mid declines as the reducer's cut would.
  • The walked result carries the run state, queue and context meter of the NEWEST response it read (normalizeSlotDetail, shared with fetchSlotDetail), so a turn that ends mid-walk is not re-armed by the first page's running: true; rows come from every page and total stays the first page's.
  • Two stops besides the anchor: WINDOW_WALK_MAX_PAGES (8) caps the walk -- the reducer then keeps no head and the rows above the window are one page-back away, the documented cost taken instead of a whole-transcript read; and a cursor that fails to move older ends it, so a handler repeating next_before while still saying has_more cannot keep the loop alive.

warmSlotCache is deliberately left alone: #10026 bounds that path with the same window-plus-verify shape. This PR is the other two halves.

Ported onto the chatSlice split (#14665). The walk now lives in store/chat/windowWalk.ts; refreshSlot (chat/slotRefresh.ts) keeps countMatchedFetchLimit as the rule it shares with the warm, and where that rule declines it clamps to floor/ceiling and only stops trusting spansView, instead of reading unbounded. Four newer test files pinned the old unbounded shape (incl. two assertions in UseWebSocketCoverage.test.tsx, now pinned to the floor page) and were repinned to intent: the two reconnect suites told the active refresh from a pane warm by its missing limit and now tell them apart by slot (plus pin the active slot reads exactly once); abortOlderOnSwitch held only cursor-less legs, so its mock now also holds the switch's signal-less walk legs.

Backwards compatibility

Compatible: client-only change to which limit/before the store sends to the existing slot-detail endpoint; every request shape it now sends is one the handler already accepts (bounded page, and bounded page with a cursor, both used by loadOlderMessages today).

Tests

  • New chatSlice.windowWalk.test.ts (7, incl. a turn ending mid-walk reports idle): stops at the first page that anchors, not the start; spends exactly the cap on a gap wider than it covers; stalled cursor stops after one page; no walk when the page already reaches; switch closes an observed hole by walking and stops where the cache begins; every request bounded when the cache sits deeper than one page.
  • 19 existing assertions pinned the literal undefined (unbounded) limit; each was rewritten to pin its intent (view does not shrink, cursor stays consistent, page never shorter than view) and every "no rows lost" assertion still passes unchanged.
  • Mutation-verified, baseline green first: cap removed -> 1 red; refresh walk -> unbounded -> 16 red; switch walk -> unbounded -> 4 red; ceiling clamp removed -> 1 red; stalled-cursor guard removed -> 1 red.

Manual verification

N/A — store/thunk change; the request sequence and resulting transcript are pinned by the unit suites above against a mocked handler that mirrors the real clamp and cursor semantics.

Ran locally: tsc -b, eslint on the touched files, the 51 test files that hit chatSlotDetail through refresh/switch (1354 passed). Mutation re-checked after the port: refresh walk -> unbounded = 17 red, switch walk -> unbounded = 3 red. Full suite left to CI.

Screenshots / video

Why no screenshot: store/thunk change only, no rendered surface differs.

Related Issues

N/A

Checklist

  • At most two commits (one is the norm), with a Conventional Commits title (feat|fix|docs|style|refactor|perf|test|chore|ci|build|revert: ...)
  • Existing tests pass and new tests added for new functionality
  • Self-review completed; code follows project style guidelines
  • Documentation updated (if applicable)
  • No secrets, credentials, or internal references in the diff

Contribution License Agreement

@buluoray
buluoray requested review from a team and chenmingwei23 September 24, 2026 01:45
@github-actions github-actions Bot added the readiness: checking Automated validation is still running label Sep 24, 2026
@github-actions

github-actions Bot commented Sep 24, 2026 •

Copy link
Copy Markdown
Contributor

UX Review (Fable 5) — ✅ PASS

UX-level review of 0e2e1de2c82c537f27de2ccbe6fdad33a3537172 — updated in place on each push. A BLOCK verdict blocks PR readiness; PASS/CONCERNS are advisory.

UX-Verdict: PASS

Verify the 8-page walk cap tradeoff: on an extreme reconnect gap, paged-in scrollback silently leaves the view until the user pages back.

[UX-REVIEWED] 0e2e1de

Description read

[DESCRIPTION-READ] 04803ffae457a06786f9055e984ab2e9281413c754314da52108a38dad7a91c1

That sha256 names everything this verdict read: the PR description, plus the 3 evidence file(s) the workflow collected from it and handed this review. It covers that evidence because the media strip replaces every attachment URL with the same placeholder, so an edit swapping one attachment for another would leave the description's captured text identical while what this review judged changed. Recompute it with .github/scripts/pr-description-capture.sh over the same inputs; if it differs, the description or that evidence changed after this verdict was formed and any finding drawn from either is unproven.

@github-actions

github-actions Bot commented Sep 24, 2026 •

Copy link
Copy Markdown
Contributor

Design Review (Fable 5) — 🟡 CONCERNS

Design-level review of 0e2e1de2c82c537f27de2ccbe6fdad33a3537172 — updated in place on each push. A BLOCK verdict blocks PR readiness; PASS/CONCERNS are advisory.

Design-Verdict: CONCERNS

A view with no meta.mid can never anchor, so every refresh re-pays a capped 8-page walk and, past ~4,000 rows, drops paged-in scrollback — every turn.

Watch

Legacy (mid-less) transcripts never converge: reaches() runs on idAnchorsOneRow, which declines absent ids, so walkWindowBackTo spends WINDOW_WALK_MAX_PAGES plus the newest re-read on every chat_done, and on a mid-less session deeper than the cap the reducer keeps no head — the stated goal ("without losing any scrollback the user paged in") is not met for exactly this class, recurring per turn rather than once. The cap doc names the case but not its every-turn recurrence.
Clears when: the walk short-circuits a view it provably cannot anchor (e.g. skip the walk when anchor holds no identified row, or remember per-slot that anchoring is impossible), or a maintainer accepts the per-turn cap cost for legacy sessions in the PR/doc.

switchSlot's walk has no live-frame decline: refreshSlot samples liveFrameSeq across its up-to-10 awaits and declines, but switchSlot.pending has already made the incoming slot active, so a chunk reduced during its walk is replaced away by switchSlot.fulfilled — the same race the guard was built for, widened ~10× over the old single unbounded await, self-correcting only at end of turn.
Clears when: the switch walk takes the same liveFrameSeq decline (falling back to first rather than null), or a comment/test records why the newest-edge re-read is judged sufficient there.

Suggestions

  • refreshAppliedSeq entries are never pruned on slot deletion; clear a slot's entry where its cache is dropped so the map tracks live slots only.

[DESIGN-REVIEWED] 0e2e1de

Description read

[DESCRIPTION-READ] f3a3e78b091aefc5e5871330a96edfdd64413a8a7a7ef96aebfd7208e67db077

That sha256 names everything this verdict read: the PR description, plus the 1 evidence file(s) the workflow collected from it and handed this review. It covers that evidence because the media strip replaces every attachment URL with the same placeholder, so an edit swapping one attachment for another would leave the description's captured text identical while what this review judged changed. Recompute it with .github/scripts/pr-description-capture.sh over the same inputs; if it differs, the description or that evidence changed after this verdict was formed and any finding drawn from either is unproven.

@github-actions

github-actions Bot commented Sep 24, 2026 •

Copy link
Copy Markdown
Contributor

GPT 5.6 Review — ✅ no blocking findings

GPT 5.6 completed its review of 0e2e1de2c82c537f27de2ccbe6fdad33a3537172 and found no blocking issues.

This comment is updated in place on each push.

Review details

No findings.
[GPT-REVIEWED] 0e2e1de

False positive or not applicable? A repository writer can comment:
/ai-review override gpt 0e2e1de2c82c537f27de2ccbe6fdad33a3537172: <one-sentence reason>

@github-actions github-actions Bot added readiness: action required A blocking check or review needs attention and removed readiness: checking Automated validation is still running labels Sep 24, 2026
@github-actions

github-actions Bot commented Sep 24, 2026 •

Copy link
Copy Markdown
Contributor

First Principles Review (Fable 5) — ✅ PASS

Premise-level review of 0e2e1de2c82c537f27de2ccbe6fdad33a3537172 — why this exists and whether the shipped surface is the smallest honest version. Updated in place on each push. A BLOCK verdict blocks PR readiness; PASS/CONCERNS are advisory.

First-Principles-Verdict: PASS

Verify the cap case: a legacy view whose rows carry no meta.mid and sit deeper than ~4,050 rows now drops its held head on every refresh.

Not justified as shipped

  • Item 7 — undeclared: liveFrameSeq and its five countLiveFrame call sites in applyActiveFrame appear nowhere in the description; the guard is derived (the walk's multi-await window) and tested, so it carries no premise risk.
  • Item 8 — undeclared: refreshAppliedSeq ordering in state.ts / refreshSlot.fulfilled is never mentioned; derived from overlapping walking refreshes and mirrors the existing warmSeq rule (runState.ts:62), not a second spelling of it.

What this change ships

Inventory (8 items) — 6 justified

Intent: stop a long session re-downloading its whole transcript on every turn end, reconnect and slot switch — a FIX, with a measured reproduction (11,134 rows/turn on a 12,178-row session) and the reversed pins' origins (#3240, #6839, #6947, #8574) named with evidence their premises no longer hold.

  1. A refresh whose page misses the view pages older in 500-row steps instead of re-reading everything — justified
  2. A slot switch with an observed coverage hole walks older pages instead of an unbounded re-read — justified
  3. A view past 500 rows refreshes with the handler's 500-row clamp, never unbounded — justified
  4. An empty view refreshes with one floor-sized page instead of the whole transcript — justified
  5. The walk stops after 8 pages; a wider (or anchorless) gap drops held rows out of view, one page-back away — justified
  6. The walk re-reads the newest page once so run status and tail describe the same moment — justified
  7. A refresh declines to land when a live frame streamed in mid-walk (liveFrameSeq) — undeclared
  8. An older refresh payload settling after a newer applied one is dropped (refreshAppliedSeq) — undeclared

Sibling unbounded reads (warmSlotCache, pane hydrate — 2, grepped limit === undefined callers in wire.ts/slotRefresh.ts) are explicitly deferred to #10026 under ## Not a goal.

[FIRST-PRINCIPLES-REVIEWED] 0e2e1de

Description read

[DESCRIPTION-READ] 1f6f893b2642e8f3a6da72bd56d373f2e2082c077ee2ae888f671b76d15547a7

That sha256 names the PR description this verdict read. Recompute it with .github/scripts/pr-description-capture.sh; if it differs, the description changed after this verdict was formed and any finding drawn from the description is unproven.

@github-actions

github-actions Bot commented Sep 24, 2026 •

Copy link
Copy Markdown
Contributor

Opus 5 Review — ✅ no blocking findings

Reviewed 0e2e1de2c82c537f27de2ccbe6fdad33a3537172 — this comment is updated in place on each push.

Review details

No findings.

[OPUS-REVIEWED] 0e2e1de

Verdict parsed from the review's SHA-scoped output markers for commit 0e2e1de2c82c537f27de2ccbe6fdad33a3537172.

False positive or not applicable? A repository writer can comment:
/ai-review override fable 0e2e1de2c82c537f27de2ccbe6fdad33a3537172: <one-sentence reason>

@github-actions github-actions Bot added the merge conflict Branch has merge conflicts with its base — author must resolve before merge label Sep 24, 2026
@cruisercohen cruisercohen added the needs-pr-triage PR scanner: awaiting automated triage label Sep 30, 2026
@NicholasRBowers NicholasRBowers added drive-to-green PR claimed by drive-to-green pipeline and removed needs-pr-triage PR scanner: awaiting automated triage labels Sep 30, 2026
@NicholasRBowers

Copy link
Copy Markdown
Contributor

🤖 Kiro Crew [operator: NicholasRBowers#a942f9ca]: This PR has been inactive for 7+ days with failing CI. I've assessed the blockers and they appear resolvable — I'll push fixes directly to this branch as a co-author.

Assessment: GPT, Opus and Design all point at the same fix, so there is no split to arbitrate: the 8-page cap can return a window with no anchor, and the reducer then drops loaded scrollback when the gap exceeds 4000 rows or the rows carry no mid — fall back to the unbounded fetchSlotDetail(key) only when the walk cannot anchor, which keeps the PR's measured win intact. Every frontend red is this PR's own tests still asserting the old unbounded call shape: warmedSlots() in useWebSocketReconnect.splitPanes.test.ts / observedPanes.test.ts counts a bounded call as a warm, UseWebSocketCoverage.test.tsx expects ('slot-a') where the call now carries a limit, and chatSlice.abortOlderOnSwitch.test.ts sees the switch retry land in the releaseOlder queue because it now carries before. The Backend (3.12, 3) red is an EADDRNOTAVAIL port-bind flake a frontend-only diff cannot reach. The conflict is a relocation, not a rethink: #14665 split chatSlice.ts into website/src/store/chat/*, so refreshSlot and the warm move to chat/slotRefresh.ts, switchSlot to chat/slotSwitch.ts, the limits to chat/paging.ts — and the unbounded fallbacks this PR removes are still on main, so it is not obsolete. I'll also take Opus's two advisory fixes (build the result from the last walked page's live state, and stop the walk when activeSlot changes).

If you'd prefer I don't touch this PR, add the pr-no-autofix label.

@buluoray
buluoray force-pushed the perf/bounded-backward-walk branch from a866b42 to 4721d9e Compare October 1, 2026 02:44
@buluoray
buluoray requested a review from a team as a code owner October 1, 2026 02:44
@github-actions github-actions Bot added readiness: checking Automated validation is still running and removed readiness: action required A blocking check or review needs attention merge conflict Branch has merge conflicts with its base — author must resolve before merge labels Oct 1, 2026
@buluoray
buluoray force-pushed the perf/bounded-backward-walk branch from 4721d9e to 8f4a471 Compare October 1, 2026 03:09
@buluoray

buluoray commented Oct 1, 2026

Copy link
Copy Markdown
Contributor Author

Disposition of the GPT 5.6 review on 4721d9e895, now addressed in 8f4a47113e:

F2: the walk returns a stale slot status. Fixed. Every page the walk reads is a full slot-detail response, so the walk now reports running, stopping, queue and context from the newest response it saw instead of the first page. The response mapping moved into a shared normalizeSlotDetail in chat/wire.ts, so a walked page is shaped exactly like a first page. total stays the first page's, because that is the count the baseline is compared in. New test: chatSlice.windowWalk.test.ts › "reports the run state of the newest response it saw, not the first page". The first page reads running: true and the older page reads idle. Reverting to page.running turns that test red.

F1: a capped walk discards loaded scrollback. No code change; this is the cost the PR chose deliberately, and the adjudicator flagged it the same way. Reaching the cap takes a gap of more than 4,000 rows in one absence, or held rows that carry no mid. Even then the rows are not lost. olderHeadAbovePage declines to an empty head, and pagingCursorAfterKeptHead keeps the walked window's own cursor (paging.ts). The window that gets published is contiguous with no hole, and the rows above it are one "load older" away. The suggested fix ("keep paging until an anchor or the start") is the whole-transcript read this PR exists to remove. The cap and its cost are documented on WINDOW_WALK_MAX_PAGES in chat/windowWalk.ts.

Also in this push: UseWebSocketCoverage.test.tsx had two assertions pinning the old full-transcript read on an empty view. They now pin the floor-sized page.

@buluoray

buluoray commented Oct 1, 2026

Copy link
Copy Markdown
Contributor Author

GPT 5.6 on 8f4a47113e: F2 (the stale run state) no longer appears. F1 (the page cap) is the same finding as before, and my earlier disposition stands: the cap is the documented cost of not re-reading the whole transcript, and the adjudicator FLAGged it again as acceptable. The window it publishes is contiguous with no hole, the paging cursor is kept, and scrolling up re-fetches the rows. Because the lane fences this finding as security-class, clearing it needs a maintainer /ai-review override gpt 8f4a47113e5fecb5da44912aad13a095ca9859a7: ..., not a code change.

@github-actions github-actions Bot added readiness: action required A blocking check or review needs attention and removed readiness: checking Automated validation is still running labels Oct 1, 2026
@buluoray

buluoray commented Oct 1, 2026

Copy link
Copy Markdown
Contributor Author

/ai-review override gpt 8f4a471: page cap is the documented cost of not re-reading the whole transcript; window stays contiguous and re-pageable

@github-actions

github-actions Bot commented Oct 1, 2026

Copy link
Copy Markdown
Contributor

Human judgment recorded

@buluoray marked the gpt AI finding as false positive, not applicable, or explicitly accepted for 8f4a47113e5fecb5da44912aad13a095ca9859a7.

page cap is the documented cost of not re-reading the whole transcript; window stays contiguous and re-pageable

This decision applies only to this commit. A new push requires a new judgment.

@buluoray
buluoray force-pushed the perf/bounded-backward-walk branch from 8f4a471 to a866b42 Compare October 1, 2026 06:49
@buluoray

buluoray commented Oct 1, 2026 •

Copy link
Copy Markdown
Contributor Author

Opus 5 on 8f4a471: the blocking finding (loosened ratchet in chatSlice.boundedRefetchShrink.test.ts) was real and is fixed in 19c7b1c, which is 8f4a471 plus this one test change. The assertion is now pinned to literal values: messages.length === 100 and { hasMore: true, paneHasMore: true }. Checked by mutation: dropping OLDER_PAGE_LIMIT to 10 makes this test fail. (An intermediate push briefly pointed the branch at a stale pre-port commit, a866b42; ignore that head.) This push voids the GPT override on 8f4a471.

@github-actions github-actions Bot removed the readiness: passed Eligible automated validation passed for the current revision label Oct 1, 2026
billygerhard added a commit to billygerhard/KiroCrew that referenced this pull request Oct 1, 2026
Rebased onto kirodotdev#13236, which already bounds refreshSlot and switchSlot with a
clamp plus an older-walking window. This keeps that walk and adds only what a
kept head needs on top of it: a marked view or cache (a serving change) reads
whole instead of walking, since the walk keeps rows above its anchor verbatim,
and a walked or overlapping head still offering a sign-in link reads whole.
The refreshSlotBound over-ceiling test is kirodotdev#13236's own.
billygerhard added a commit to billygerhard/KiroCrew that referenced this pull request Oct 1, 2026
Rebased onto kirodotdev#13236, which already bounds refreshSlot and switchSlot with a
clamp plus an older-walking window. This keeps that walk and adds only what a
kept head needs on top of it: a marked view or cache (a serving change) reads
whole instead of walking, since the walk keeps rows above its anchor verbatim,
and a walked or overlapping head still offering a sign-in link reads whole.
The refreshSlotBound over-ceiling test is kirodotdev#13236's own.
billygerhard added a commit to billygerhard/KiroCrew that referenced this pull request Oct 2, 2026
Rebased onto kirodotdev#13236, which already bounds refreshSlot and switchSlot with a
clamp plus an older-walking window. This keeps that walk and adds only what a
kept head needs on top of it: a marked view or cache (a serving change) reads
whole instead of walking, since the walk keeps rows above its anchor verbatim,
and a walked or overlapping head still offering a sign-in link reads whole.
The refreshSlotBound over-ceiling test is kirodotdev#13236's own.
billygerhard added a commit to billygerhard/KiroCrew that referenced this pull request Oct 2, 2026
Rebased onto kirodotdev#13236, which already bounds refreshSlot and switchSlot with a
clamp plus an older-walking window. This keeps that walk and adds only what a
kept head needs on top of it: a marked view or cache (a serving change) reads
whole instead of walking, since the walk keeps rows above its anchor verbatim,
and a walked or overlapping head still offering a sign-in link reads whole.
The refreshSlotBound over-ceiling test is kirodotdev#13236's own.
billygerhard added a commit to billygerhard/KiroCrew that referenced this pull request Oct 2, 2026
Rebased onto kirodotdev#13236, which already bounds refreshSlot and switchSlot with a
clamp plus an older-walking window. This keeps that walk and adds only what a
kept head needs on top of it: a marked view or cache (a serving change) reads
whole instead of walking, since the walk keeps rows above its anchor verbatim,
and a walked or overlapping head still offering a sign-in link reads whole.
The refreshSlotBound over-ceiling test is kirodotdev#13236's own.
billygerhard added a commit to billygerhard/KiroCrew that referenced this pull request Oct 2, 2026
Rebased onto kirodotdev#13236, which already bounds refreshSlot and switchSlot with a
clamp plus an older-walking window. This keeps that walk and adds only what a
kept head needs on top of it: a marked view or cache (a serving change) reads
whole instead of walking, since the walk keeps rows above its anchor verbatim,
and a walked or overlapping head still offering a sign-in link reads whole.
The refreshSlotBound over-ceiling test is kirodotdev#13236's own.
billygerhard added a commit to billygerhard/KiroCrew that referenced this pull request Oct 2, 2026
Rebased onto kirodotdev#13236 and kirodotdev#16005, which bound refreshSlot, switchSlot and
warmSlotCache with a clamp plus an older-walking window. This keeps that walk
on all three and adds only what a kept head needs on top of it: a marked view
or cache (a serving change) reads whole instead of walking, since the walk
keeps rows above its anchor verbatim, a walked or overlapping head still
offering a sign-in link reads whole, and a background warm that anchors
nothing in a cache with a raw retained count reads whole once so a remote
clear is still read as a shrink. The refreshSlotBound over-ceiling test is
kirodotdev#13236's own.
billygerhard added a commit to billygerhard/KiroCrew that referenced this pull request Oct 2, 2026
Rebased onto kirodotdev#13236 and kirodotdev#16005, which bound refreshSlot, switchSlot and
warmSlotCache with a clamp plus an older-walking window. This keeps that walk
on all three and adds only what a kept head needs on top of it: a marked view
or cache (a serving change) reads whole instead of walking, since the walk
keeps rows above its anchor verbatim, a walked or overlapping head still
offering a sign-in link reads whole, and a background warm that anchors
nothing in a cache with a raw retained count reads whole once so a remote
clear is still read as a shrink. The refreshSlotBound over-ceiling test is
kirodotdev#13236's own.
billygerhard added a commit to billygerhard/KiroCrew that referenced this pull request Oct 2, 2026
Rebased onto kirodotdev#13236 and kirodotdev#16005, which bound refreshSlot, switchSlot and
warmSlotCache with a clamp plus an older-walking window. This keeps that walk
on all three and adds only what a kept head needs on top of it: a marked view
or cache (a serving change) reads whole instead of walking, since the walk
keeps rows above its anchor verbatim, a walked or overlapping head still
offering a sign-in link reads whole, and a background warm that anchors
nothing in a cache with a raw retained count reads whole once so a remote
clear is still read as a shrink. The refreshSlotBound over-ceiling test is
kirodotdev#13236's own.
billygerhard added a commit to billygerhard/KiroCrew that referenced this pull request Oct 2, 2026
Rebased onto kirodotdev#13236 and kirodotdev#16005, which bound refreshSlot, switchSlot and
warmSlotCache with a clamp plus an older-walking window. This keeps that walk
on all three and adds only what a kept head needs on top of it: a marked view
or cache (a serving change) reads whole instead of walking, since the walk
keeps rows above its anchor verbatim, a walked or overlapping head still
offering a sign-in link reads whole, and a background warm that anchors
nothing in a cache with a raw retained count reads whole once so a remote
clear is still read as a shrink. The refreshSlotBound over-ceiling test is
billygerhard added a commit to billygerhard/KiroCrew that referenced this pull request Oct 2, 2026
Rebased onto kirodotdev#13236 and kirodotdev#16005, which bound refreshSlot, switchSlot and
warmSlotCache with a clamp plus an older-walking window. This keeps that walk
on all three and adds only what a kept head needs on top of it: a marked view
or cache (a serving change) reads whole instead of walking, since the walk
keeps rows above its anchor verbatim, a walked or overlapping head still
offering a sign-in link reads whole, and a background warm that anchors
nothing in a cache with a raw retained count reads whole once so a remote
clear is still read as a shrink. The refreshSlotBound over-ceiling test is
billygerhard added a commit to billygerhard/KiroCrew that referenced this pull request Oct 2, 2026
Rebased onto kirodotdev#13236 and kirodotdev#16005, which bound refreshSlot, switchSlot and
warmSlotCache with a clamp plus an older-walking window. This keeps that walk
on all three and adds only what a kept head needs on top of it: a marked view
or cache (a serving change) reads whole instead of walking, since the walk
keeps rows above its anchor verbatim, a walked or overlapping head still
offering a sign-in link reads whole, and a background warm that anchors
nothing in a cache with a raw retained count reads whole once so a remote
clear is still read as a shrink. The refreshSlotBound over-ceiling test is
billygerhard added a commit to billygerhard/KiroCrew that referenced this pull request Oct 2, 2026
Rebased onto kirodotdev#13236 and kirodotdev#16005, which bound refreshSlot, switchSlot and
warmSlotCache with a clamp plus an older-walking window. This keeps that walk
on all three and adds only what a kept head needs on top of it: a marked view
or cache (a serving change) reads whole instead of walking, since the walk
keeps rows above its anchor verbatim, a walked or overlapping head still
offering a sign-in link reads whole, and a background warm that anchors
nothing in a cache with a raw retained count reads whole once so a remote
clear is still read as a shrink. The refreshSlotBound over-ceiling test is
billygerhard added a commit to billygerhard/KiroCrew that referenced this pull request Oct 2, 2026
Rebased onto kirodotdev#13236 and kirodotdev#16005, which bound refreshSlot, switchSlot and
warmSlotCache with a clamp plus an older-walking window. This keeps that walk
on all three and adds only what a kept head needs on top of it: a marked view
or cache (a serving change) reads whole instead of walking, since the walk
keeps rows above its anchor verbatim, a walked or overlapping head still
offering a sign-in link reads whole, and a background warm that anchors
nothing in a cache with a raw retained count reads whole once so a remote
clear is still read as a shrink. The refreshSlotBound over-ceiling test is
billygerhard added a commit to billygerhard/KiroCrew that referenced this pull request Oct 3, 2026
Rebased onto kirodotdev#13236 and kirodotdev#16005, which bound refreshSlot, switchSlot and
warmSlotCache with a clamp plus an older-walking window. This keeps that walk
on all three and adds only what a kept head needs on top of it: a marked view
or cache (a serving change) reads whole instead of walking, since the walk
keeps rows above its anchor verbatim, a walked or overlapping head still
offering a sign-in link reads whole, and a background warm that anchors
nothing in a cache with a raw retained count reads whole once so a remote
clear is still read as a shrink. The refreshSlotBound over-ceiling test is
billygerhard added a commit to billygerhard/KiroCrew that referenced this pull request Oct 3, 2026
Rebased onto kirodotdev#13236 and kirodotdev#16005, which bound refreshSlot, switchSlot and
warmSlotCache with a clamp plus an older-walking window. This keeps that walk
on all three and adds only what a kept head needs on top of it: a marked view
or cache (a serving change) reads whole instead of walking, since the walk
keeps rows above its anchor verbatim, a walked or overlapping head still
offering a sign-in link reads whole, and a background warm that anchors
nothing in a cache with a raw retained count reads whole once so a remote
clear is still read as a shrink. The refreshSlotBound over-ceiling test is
billygerhard added a commit to billygerhard/KiroCrew that referenced this pull request Oct 3, 2026
Rebased onto kirodotdev#13236 and kirodotdev#16005, which bound refreshSlot, switchSlot and
warmSlotCache with a clamp plus an older-walking window. This keeps that walk
on all three and adds only what a kept head needs on top of it: a marked view
or cache (a serving change) reads whole instead of walking, since the walk
keeps rows above its anchor verbatim, a walked or overlapping head still
offering a sign-in link reads whole, and a background warm that anchors
nothing in a cache with a raw retained count reads whole once so a remote
clear is still read as a shrink. The refreshSlotBound over-ceiling test is
billygerhard added a commit to billygerhard/KiroCrew that referenced this pull request Oct 3, 2026
Rebased onto kirodotdev#13236 and kirodotdev#16005, which bound refreshSlot, switchSlot and
warmSlotCache with a clamp plus an older-walking window. This keeps that walk
on all three and adds only what a kept head needs on top of it: a marked view
or cache (a serving change) reads whole instead of walking, since the walk
keeps rows above its anchor verbatim, a walked or overlapping head still
offering a sign-in link reads whole, and a background warm that anchors
nothing in a cache with a raw retained count reads whole once so a remote
clear is still read as a shrink. The refreshSlotBound over-ceiling test is
billygerhard added a commit to billygerhard/KiroCrew that referenced this pull request Oct 5, 2026
Rebased onto kirodotdev#13236 and kirodotdev#16005, which bound refreshSlot, switchSlot and
warmSlotCache with a clamp plus an older-walking window. This keeps that walk
on all three and adds only what a kept head needs on top of it: a marked view
or cache (a serving change) reads whole instead of walking, since the walk
keeps rows above its anchor verbatim, a walked or overlapping head still
offering a sign-in link reads whole, and a background warm that anchors
nothing in a cache with a raw retained count reads whole once so a remote
clear is still read as a shrink. The refreshSlotBound over-ceiling test is
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants