fix(signals): a merged transaction's verdict lane is the same hold's at the D-28 seat (#4427) by ryansolid · Pull Request #4455 · solidjs/solid

ryansolid · 2026-10-30T10:31:56Z

Summary

Before: a render effect seated as a hold's verdict reader (it reads latest() of a node B holds) left that seat when it read a verdict-lane derivation of a transaction merged into B. That derivation was stamped before the merge and still sits in A._verdict.

D-28's reset compared lane objects, so B's own merged lane looked like another hold's. The leaf was then served the staging as B's held work, and its run waited for B's landing, while the derivation it read was already on screen. The screen showed m = 20 beside a leaf still at 1/10, a tear (F-1).

After: the reset compares holds, so a merged transaction's verdict lane counts as the same hold's (D-28, D-4). The leaf stays seated, reads the lane's value, and shows it together with the derivation (1/20).

How

In laneRead, a seated leaf now leaves the seat only in three cases:

the node is in flight and has never shown;
the lane is not a verdict lane;
the lane's parent resolves to a different hold than the seat's.

Otherwise the leaf reads the lane as its own: it skips the stale-reader block, and enterLane links the two verdict lanes of the one hold. This is the V6 row of #4438's table.

This only runs with the lane engine installed. Nothing changes in recompute, and no effects are added. The structural fix for lanes keeping their pre-merge identities (#3905) is out of scope.

Also: readsHeld's loop (from #4449) is rewritten as for (let d = r._deps; d; d = d._nextDep), as retryReaches does, per the #4449 review. That saves 7 B minified. The #4438 seat-shadow known #4427 entry now gets 0 hits and is left in place for later cleanup.

Size vs next @ cddaa76 (measured locally, linux; next's ledger was re-recorded by #4462):

scenario	brotli	minified	cap
compiled floor (hello world)	0 B (9,963)	0 B	unchanged
`signals: + isPending/latest`	+17 B	+45 B	11.10 → 11.12 KB
`app: hydrating + every store primitive family`	+93 B	+45 B	32.45 → 32.48 KB
`page: live server components` (frozen floor)	+22 B	+45 B	38.83 → 38.85 KB
`page: compiled live server components`	+20 B	+40 B	42.71 → 42.73 KB
`page: live + router`	+29 B	+45 B	unchanged (recorded minified re-recorded)

The brotli on the store scenario is mostly layout: the lanes.ts change renames a mangled property in store.js, which moves brotli on bundles whose minified size barely changes.

Size-Exception: the live server-components page carries the D-28 same-hold check that fixes #4427 (+45 B minified, +22 B brotli); correctness.

How did you test this change?

New packages/signals/tests/verdict-seat-merged-lane-4427.test.ts. On next the leaf stays at 1/10 beside m = 20; with the fix it passes.
Suites:
- signals: 5,999 passed.
- web: 1,559 passed.
- hydrate: 715 passed.
- server: server-functions-adapter-request fails the same way on next.
On the test(signals): reader seat — the slot table and shadow differential as a CI spec (S0+M1, byte-identical) #4438 shadow:
- On next, the new test hits only known #4427; with the fix it hits nothing.
- Generator 7 / 3289 / 91501 × 5,000 cases: the hit count goes from 1 to 0, with 0 mismatch classes. Screens, finals and findings are identical.

Fixes #4427

Generated by Claude Code

…ane objects (#4427) `laneRead`'s D-28 reset (#4275, #4284) makes a leaf seated as a hold's verdict reader leave the seat when it reads a lane that is not that hold's. It decided that by lane identity (`passLane !== l`). A transaction merged into the hold keeps its own verdict lane: a derivation stamped before the merge stays in it. So the leaf read that lane as another hold's. It was served the derivation's staging as the hold's own work (L7'), and its run waited for the hold's landing while the derivation it read was already on screen. The reset now compares the holds. It leaves only when the node is in flight and has never shown, when the lane is not a verdict lane, or when the lane's parent resolves to a different hold than the seat's (D-4: merged holds are one). The seated leaf then reads that lane as the hold's own: the lane's value, the two lanes linked by `enterLane` as for any pass that reads both. Fixes #4427 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_015hPxMkC3jg3Uivh65gvaf2

…-x1ore2-4427

…golf readsHeld's loop `readsHeld` (#4449) now walks its deps as `retryReaches` does (`for (let d = r._deps; d; d = d._nextDep)`), which saves 7 B minified on isPending/latest. Measured locally (linux) over next @ 7135a77. On page: live server components (a frozen floor): 38,764 B brotli / 117,272 B minified, +45 B minified, so the cap goes 38.76 -> 38.78 KB and floor-caps.json's recorded minified 117,227 -> 117,272. On page: compiled live server components: 42,690 / 125,414, so the cap goes 42.67 -> 42.70 KB and the recorded minified 125,374 -> 125,414. Caps are at measured + 10 B rounded up to 0.01 KB. Every other scenario passes the gate. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_015hPxMkC3jg3Uivh65gvaf2

changeset-bot · 2026-10-30T10:32:03Z

🦋 Changeset detected

Latest commit: 828aa89

The changes in this PR will be included in the next version bump.

This PR includes changesets to release 15 packages

Name	Type
@solidjs/signals	Patch
test-integration	Patch
@solidjs/web	Patch
@solidjs/babel-plugin	Patch
@solidjs/compiler	Patch
@solidjs/diagnostics	Patch
@solidjs/element	Patch
@solidjs/h	Patch
@solidjs/html	Patch
solid-js	Patch
@solidjs/universal	Patch
board-example	Patch
canvas-example	Patch
server-dbmon-example	Patch
todos-server-example	Patch

Not sure what this means? Click here to learn what changesets are.

Click here if you're a maintainer who wants to add another changeset to this PR

github-actions · 2026-10-30T10:35:12Z

Size (brotli, eager entry chunk)

scenario	head	vs base	minified vs base	minified vs recorded	cap		lazy chunks (not counted)
signals: core floor (createSignal/Memo/Effect/Root/flush)	7.42 KB	0 B	0 B	+13 B	7.43 KB	✅
signals: + createStore	14.95 KB	−2 B (−0.0%)	0 B	0 B	14.97 KB	✅
signals: + isPending/latest	11.11 KB	+17 B (+0.2%)	+45 B	0 B	11.12 KB	✅
signals: + until/refresh (truth readers, no optimism)	8.04 KB	0 B	0 B	0 B	8.06 KB	✅
app: render + one signal (the simple-app floor)	9.78 KB	0 B	0 B	+13 B	9.78 KB	✅
app: hydrating (no stores) with Show/For/Loading/Errored/lazy	18.75 KB	0 B	0 B	+13 B	18.74 KB	⚠️ over by 5 B, 7 B minified headroom	lazy-page.js 0.04 KB
app: hydrating + every store primitive family	32.47 KB	+93 B (+0.3%)	+45 B	0 B	32.48 KB	✅	lazy-page.js 0.04 KB
app: CSR with Show/For/Loading/Errored/lazy	13.45 KB	0 B	0 B	0 B	13.47 KB	✅	lazy-page.js 0.04 KB
app: CSR, observe tier (same app on the `observe` artifacts)	15.18 KB	0 B	0 B	0 B	15.20 KB	✅	lazy-page.js 0.04 KB
app: CSR, observe tier + attribution engine enabled	29.76 KB	0 B	0 B	0 B	29.78 KB	✅	lazy-page.js 0.04 KB
app: compiled floor (one template, one text hole, one delegated click)	9.96 KB	0 B	0 B	0 B	9.98 KB	✅
app: compiled CSR (JSX todo app: spread/merge/omit, events, class/style, keyed For, Show, Loading + lazy, store)	26.30 KB	−13 B (−0.0%)	0 B	0 B	26.33 KB	✅	stats.js 0.18 KB
app: compiled hydrating (the same JSX todo app through hydrate(), compiled hydratable)	32.51 KB	+41 B (+0.1%)	0 B	0 B	32.52 KB	✅	stats.js 0.20 KB
frames: eager client consumer (frames client + transport, lazy codec)	11.75 KB	0 B	0 B	0 B	11.77 KB	✅
page: base server components (hydrating + dynamic + frames + sf reference)	33.76 KB	0 B	0 B	0 B	33.77 KB	✅	assets.js 0.78 KB, decode.js 7.09 KB, lazy-page.js 0.04 KB, regions.js 1.06 KB, trace.js 9.03 KB, wire.js 3.47 KB
page: live server components (base + live/GET + action + isPending/latest)	38.84 KB	+22 B (+0.1%)	+45 B	0 B	38.85 KB	✅	assets.js 0.78 KB, decode.js 7.09 KB, lazy-page.js 0.04 KB, regions.js 1.06 KB, trace.js 9.03 KB, wire.js 3.47 KB
page: compiled base server components (the base page as JSX: templates with class/style/attributes/events, For/Show; no spread)	35.70 KB	0 B	0 B	0 B	35.72 KB	✅	assets.js 0.78 KB, decode.js 7.09 KB, regions.js 1.06 KB, sc-comments.js 0.20 KB, trace.js 9.00 KB, wire.js 3.47 KB
page: compiled live server components (the compiled base page + live/GET + action + isPending/latest)	42.72 KB	+20 B (+0.0%)	+40 B	0 B	42.73 KB	✅	eager (counted): web.js 22.77 KB; assets.js 0.78 KB, decode.js 7.09 KB, regions.js 1.07 KB, sc-comments.js 0.19 KB, trace.js 9.07 KB, wire.js 3.47 KB
page: base + router (base page + @solidjs/router: createRouter, two routes, preload, useNavigate)	41.18 KB	0 B	0 B	0 B	41.23 KB	✅	assets.js 0.78 KB, decode.js 7.09 KB, lazy-page.js 0.04 KB, regions.js 1.06 KB, server.js 1.03 KB, serverForms.js 3.57 KB, trace.js 9.01 KB, wire.js 3.48 KB
page: live + router (live page + @solidjs/router: createRouter, two routes, preload, useNavigate)	48.39 KB	+29 B (+0.1%)	+45 B	0 B	48.44 KB	✅	eager (counted): client.js 27.91 KB; assets.js 0.78 KB, decode.js 7.09 KB, lazy-page.js 0.04 KB, regions.js 1.07 KB, server.js 1.03 KB, serverForms.js 3.30 KB, trace.js 9.04 KB, wire.js 3.49 KB
server: floor (getRequestEvent + isServer)	1.27 KB	0 B	0 B	0 B	1.29 KB	✅
server: renderToString (the server-render floor)	22.79 KB	0 B	0 B	0 B	22.80 KB	✅

⚠️ Over the brotli cap within the minified allowance (passes)

app: hydrating (no stores) with Show/For/Loading/Errored/lazy: over brotli cap by 5 B; minified 53,598 B vs 53,585 B recorded with the cap (+13 B) — 7 B of the 20 B minified allowance left; +0 B minified over this PR's base

_{Bundled with Rolldown (what Vite ships), brotli q11, decimal KB. A scenario fails only when it is over its brotli cap and its minified size is more than 20 B over the minified recorded with the cap; over the cap within that allowance is brotli layout noise and passes with a warning. Caps and their recorded minified in scripts/size/scenarios.js; the floor and page caps in floor-caps.json are frozen (lower only, or Size-Exception: in the PR body). npm run ratchet lowers caps per RC; it never raises one (scripts/size/README.md).}

coveralls · 2026-10-30T10:41:29Z

Coverage Report for CI Build 41065411805

Coverage remained the same at 78.1%

Details

Coverage remained the same as the base build.
Patch coverage: No coverable lines changed in this PR.
No coverage regressions found.

Uncovered Changes

No uncovered changes found.

Coverage Regressions

No coverage regressions found.

Coverage Stats


Relevant Lines:	1362
Covered Lines:	1125
Line Coverage:	82.6%
Relevant Branches:	1090
Covered Branches:	790
Branch Coverage:	72.48%
Branches in Coverage %:	Yes
Coverage Strength:	29.82 hits per line

💛 - Coveralls

codspeed · 2026-10-30T10:57:16Z

Merging this PR will degrade performance by 13.54%

2 warnings

⚠️ 3 benchmarks are dominated by system calls

System calls cannot be simulated consistently and are left out of the measure. The reported value understates the real cost. Switch to the Walltime instrument to measure them.

Learn more about system calls

See affected benchmarks

⚠️ 1 benchmark with a significant change ran on different hardware

Its base and head results were measured on machines with a different CPU, core count or memory, so the change might not come from the code. Run base and head on the same runner type.

Learn more about Macro Runners

See affected benchmarks

❌ 1 regressed benchmark
✅ 199 untouched benchmarks
⏩ 5 skipped benchmarks¹

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Benchmark	`BASE`	`HEAD`	Efficiency
❌	`projection derive: write one NESTED field (reference)`	2.1 ms	2.5 ms	-13.54%

Tip

Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.

_{Comparing claude/project-thread-x1ore2-4427 (828aa89) with next (cddaa76)}

5 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩

The #4400 case writes s, waits 3 ms, and writes again while a 10 ms ask is in flight. On CI's coverage run the real gap outlived the ask, which landed before the second write and asked `m1:23` (seen on #4455 and #4460). The gaps are part of each scenario, so the file now advances fake timers instead of sleeping. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_015hPxMkC3jg3Uivh65gvaf2

…er it lands (#4460) * fix(signals): don't retain the transaction a latest() read opened after it lands `pendingVerdict` (#4453) remembered the flush's transaction a `latest` read of an unheld flight opened in a module-level `latestJoined`, compared only against `flushTransaction` and never cleared: after the seam it kept the last landed transaction (its `_nodes`, `_queues`, `_reruns`) alive until the next such read overwrote it. The mark now lives on the transaction itself (`Transaction.latest`), so it ends with it. Same answer for every probe: a transaction stays the flush's from its opening to the seam, a merge keeps the flush's own, and one merged away is no longer anyone's flush transaction, as before. The property is not `_`-prefixed (as `folded`), so the mangler's assignment is untouched; only bundles importing `latest`/`isPending` change, and they shrink (isPending/latest -13 B brotli, -10 B minified). Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_015hPxMkC3jg3Uivh65gvaf2 * test(signals): the #4393/#4400 two-reader specs run on fake timers The #4400 case writes s, waits 3 ms, and writes again while a 10 ms ask is in flight. On CI's coverage run the real gap outlived the ask, which landed before the second write and asked `m1:23` (seen on #4455 and #4460). The gaps are part of each scenario, so the file now advances fake timers instead of sleeping. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_015hPxMkC3jg3Uivh65gvaf2 --------- Co-authored-by: Claude <noreply@anthropic.com>

…-x1ore2-4427 # Conflicts: # scripts/size/floor-caps.json # scripts/size/scenarios.js

…-x1ore2-4427

…-x1ore2-4427 # Conflicts: # scripts/size/scenarios.js

ryansolid · 2026-10-30T14:07:18Z

I don't think CodSpeed's −13.5% on projection-root-write "projection derive: write one NESTED field (reference)" comes from this PR. This bench file's single-run cases swing ±13–15% with the store module's byte size alone, because a V8 lazy sweep after the pre-run gc() either lands in the one measured run or misses it. #4456 is on the same base (next cddaa76) and shows the same bench at −13.2% with an unrelated diff. This PR changes only the verdict-lane pick at the D-28 seat, and the bench is a synchronous projection write with no hold or lane. The other 199 benches are untouched. I'm treating this as noise.

Generated by Claude Code

… held-frame clause covers it laneRead's D-28 seat reset goes back to comparing verdict lanes by identity. With the held-frame clause, no shape depends on the hold comparison, and the seat table's #4427 known entry no longer matches anything. -48 B minified on the verdict pages. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_015hPxMkC3jg3Uivh65gvaf2

claude added 4 commits October 30, 2026 08:42

Merge remote-tracking branch 'origin/next' into claude/project-thread…

f2ade0a

…-x1ore2-4427

Merge remote-tracking branch 'origin/next' into claude/project-thread…

8495f9c

…-x1ore2-4427

ryansolid mentioned this pull request Oct 30, 2026

fix(signals): don't retain the transaction a latest() read opened after it lands #4460

Merged

claude added 3 commits October 30, 2026 13:18

Merge remote-tracking branch 'origin/next' into claude/project-thread…

5ce3134

…-x1ore2-4427 # Conflicts: # scripts/size/floor-caps.json # scripts/size/scenarios.js

Merge remote-tracking branch 'origin/next' into claude/project-thread…

f55f743

…-x1ore2-4427

Merge remote-tracking branch 'origin/next' into claude/project-thread…

828aa89

…-x1ore2-4427 # Conflicts: # scripts/size/scenarios.js

ryansolid merged commit 131ceb8 into next Oct 30, 2026
8 of 9 checks passed

Navigation Menu

Repository navigation

fix(signals): a merged transaction's verdict lane is the same hold's at the D-28 seat (#4427) - #4455

fix(signals): a merged transaction's verdict lane is the same hold's at the D-28 seat (#4427)#4455
ryansolid merged 7 commits into
nextfrom
claude/project-thread-x1ore2-4427

ryansolid commented Oct 30, 2026 •

edited

Loading

Uh oh!

changeset-bot Bot commented Oct 30, 2026 •

edited

Loading

Uh oh!

github-actions Bot commented Oct 30, 2026 •

edited

Loading

Uh oh!

coveralls commented Oct 30, 2026 •

edited

Loading

Uh oh!

codspeed Bot commented Oct 30, 2026 •

edited

Loading

Uh oh!

ryansolid commented Oct 30, 2026

Uh oh!

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Conversation

ryansolid commented Oct 30, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

How

How did you test this change?

Uh oh!

changeset-bot Bot commented Oct 30, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🦋 Changeset detected

Uh oh!

github-actions Bot commented Oct 30, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Size (brotli, eager entry chunk)

⚠️ Over the brotli cap within the minified allowance (passes)

Uh oh!

coveralls commented Oct 30, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Coverage Report for CI Build 41065411805

Coverage remained the same at 78.1%

Details

Uncovered Changes

Coverage Regressions

Coverage Stats

💛 - Coveralls

Uh oh!

codspeed Bot commented Oct 30, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Merging this PR will degrade performance by 13.54%

Performance Changes

Footnotes

Uh oh!

ryansolid commented Oct 30, 2026

Uh oh!

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

ryansolid commented Oct 30, 2026 •

edited

Loading

changeset-bot Bot commented Oct 30, 2026 •

edited

Loading

github-actions Bot commented Oct 30, 2026 •

edited

Loading

coveralls commented Oct 30, 2026 •

edited

Loading

codspeed Bot commented Oct 30, 2026 •

edited

Loading