Skip to content

Commit 4c6a2ee

Browse files
ryansolidClaude via Cursor
andcommitted
docs(plans): reader-seat rebuild §4.2 — three-seed 50k class count, post-fix re-run, plain-build abort indices, untraced residue
The 50k row counted seed 3289's classes (133) against the three-seed reads/passes/instances; the aggregate is 220 classes / 8,821 instances before the two 50k table fixes and 209 / 8,706 re-run with the table as committed (the 11 classes the fixes cover go, none appear). The infinite loop abort on each seed is verified against a dist built without SIGNALS_TEST_DEFINE, at the same case index on both builds (7: 1158, 3289: 3963, 91501: 5807). Classes no §6 finding or §7 item names literally (623 instances / 43 classes) are listed and left untraced. Co-authored-by: Claude via Cursor <[email protected]>
1 parent b71aafe commit 4c6a2ee

1 file changed

Lines changed: 41 additions & 20 deletions

File tree

‎documentation/plans/reader-seat-rebuild.md‎

Lines changed: 41 additions & 20 deletions
Original file line numberDiff line numberDiff line change
@@ -271,32 +271,53 @@ go.
271271

272272
Counts are the shadow's: reads compared / passes seated / mismatch
273273
classes / instances. Every mismatch is bucketed in §6 (table fixes,
274-
applied) or §7 (two readings, awaiting a ruling); no class was found where
275-
the new slot is right and today wrong by a rule, so no issue was filed.
276-
277-
| run | reads | passes | classes | instances |
278-
| --------------------------------------------------- | ---------: | ---------: | ------: | --------: |
279-
| signals suite (375 files, green) | 965,402 | 238,364 | 40 | 137 |
280-
| visibility oracle (posture / oracle / store) | 11,379 | 8,322 | 0 | 0 |
281-
| solid suite | 2,389 | 1,546 | 0 | 0 |
282-
| web suite (the 27 `.spec.ts` files that load here¹) | 254,902 | 155,740 | 25 | 74 |
283-
| generator 2,000 × {3289, 91501, 7}, every pair | 6,969,627 | 5,153,328 | 143 | 4,127 |
284-
| generator 50,000 × {3289, 91501, 7}, every pair² | 67,328,944 | 43,755,988 | 133 | 8,821 |
274+
applied), §7 (two readings, awaiting a ruling) or the untraced residue
275+
below; no class was found where the new slot is right and today wrong by
276+
a rule, so no issue was filed.
277+
278+
| run | reads | passes | classes | instances |
279+
| --------------------------------------------------------------------------- | ---------: | ---------: | ------: | --------: |
280+
| signals suite (376 files, green; table as committed) | 965,386 | 238,343 | 40 | 151 |
281+
| visibility oracle (posture / oracle / store) | 11,379 | 8,322 | 0 | 0 |
282+
| solid suite | 2,389 | 1,546 | 0 | 0 |
283+
| web suite (the 27 `.spec.ts` files that load here¹) | 254,902 | 155,740 | 25 | 74 |
284+
| generator 2,000 × {3289, 91501, 7}, every pair | 6,969,627 | 5,153,328 | 143 | 4,127 |
285+
| generator 50,000 × {3289, 91501, 7}, every pair² — before the two 50k fixes | 67,328,944 | 43,755,988 | 220 | 8,821 |
286+
| generator 50,000 × {3289, 91501, 7}, every pair² — table as committed | 67,328,944 | 43,755,988 | 209 | 8,706 |
285287

286288
¹ The worktree has no `@solidjs/compiler` binary, so the 153 `.spec.tsx`
287289
files fail to load on `next` too; CI runs them.
288290
² All three seeds abort on `next` before 50,000 cases with `Potential
289-
Infinite Loop Detected. Kept alive by scheduled work` (seed 7 inside the
290-
first 2,000; 3289 at ~427 s, 91501 at ~638 s), identically on the plain
291-
build without the shadow — a generator finding on `next`, not a shadow
292-
finding, reported in the PR and not minimized here. Seeds 3289 and 91501
293-
also report one `equivalence/E7` signature each on both builds.
291+
Infinite Loop Detected. Kept alive by scheduled work`, and at the same
292+
case on a dist built without `SIGNALS_TEST_DEFINE` (no shadow): seed 7 at
293+
case 1158 (the E8 `remount` run; `last staged node: computed`), 3289 at
294+
case 3963 (the base run; `value`), 91501 at case 5807 (the E8 `remount`
295+
run; `signal`) — a generator finding on `next`, not a shadow finding,
296+
reported in the PR and not minimized here. (The campaign does not print
297+
the index; a throwaway replay of its survey loop did.) Seeds 3289 and
298+
91501 also report one `equivalence/E7` signature each on both builds.
294299

295300
The 50k run added two table rows to §6 (L4/L2/V1' staged error, V3 under
296-
published inputs) and no new two-readings item; by bucket (2k → 50k):
297-
F-ahead-posture 1,461 → 2,922; F-vlane 1,054 → 2,431; F-vworld 1,101 →
298-
2,308; F-transitive 301 → 684; B2 display-ahead residue 104 → 257; F-mount
299-
97 → 180.
301+
published inputs) and no new two-readings item. Re-run with the table as
302+
committed, the three seeds lose exactly the 11 classes / 115 instances
303+
those two fixes cover and gain none (reads and passes are identical). By
304+
bucket (2k → 50k, the previous triage's reading): F-ahead-posture 1,461 →
305+
2,922; F-vlane 1,054 → 2,431; F-vworld 1,101 → 2,308; F-transitive 301 →
306+
684; B2 display-ahead residue 104 → 257; F-mount 97 → 180.
307+
308+
Read literally against the class signatures §7 names, 623 instances / 43
309+
classes of the 50k (unchanged by the two fixes) match no §6 finding or §7
310+
item. They are listed with counts under the PR body's residue row and
311+
left **untraced** here: `B1 ahead (old true, new false; creator running
312+
true)` ×406 / 4 classes; `D5` E vs NR for an errored node in flight
313+
(`u f` · `f ip` · `u f ip` · one `f d vlane`; D and V postures; 4 under
314+
`oldVlane`) ×101 / 17; `L8` E vs NR, the same shape at a leaf ×34 / 6; `V1` C vs P (`s h`,
315+
`oldNoLane`) ×57 / 2; `V4` C vs NR (`f ip`; 8 under `oldVlane`, 4 not)
316+
×12 / 5; `V1''` P vs E (`s h u ip` — the D-41 staged-error shape on a row
317+
the §6 fix did not touch) ×2 / 2; `V1'` P vs C ×2 / 2 and `Vl1` P vs C
318+
×1 / 1 (`s h`, `oldNoLane`, world 2 / 0); world after the read old 2 →
319+
new 1 with equal slots (`Vl1` / `Vp1` / `Vl5`) ×7 / 3; `B2 born held (old
320+
true, new false; world 1, no mount, rows L8)` ×1 / 1.
300321

301322
## 5. Swap plan
302323

0 commit comments

Comments
 (0)