Skip to content

docs(bench): record the object door's cost on both sides of #4716, on both stack shapes (#9859)#10040

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
os-elon merged 1 commit into from
Aug 19, 2026
+31 −0
Merged

docs(bench): record the object door's cost on both sides of #4716, on both stack shapes (#9859) #10040

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
31 changes: 31 additions & 0 deletions scripts/bench/runtime-publish-gate.bench.mts
View file
Open in desktop
Original file line numberDiff line numberDiff line change
Expand Up@@ -95,6 +95,37 @@
* that app is edited. Deliberate — it is the only real, spec-parsed corpus in
* the repo, and a frozen copy would silently stop describing shipped metadata.
* Quote the ref you measured at.
*
* ── THE OBJECT DOOR, BOTH SIDES OF #4716 — MEASURED ON BOTH SHAPES (#9859) ──
*
* #9825's cost table for the object-door widening was published from the `stub`
* shape alone, so its `+0.31 / +3.25 ms` delta describes a stack in which the
* rules that judge validation rules and autonumber formats find nothing to
* judge. Both legs re-measured here at `17854cba0`, one machine, one session,
* median of 30 with warmup 5. The BEFORE leg is built from the PRE-WIDENING
* registry — `git restore --source=1408ae337^ -- packages/lint/src/authoring-
* rules.ts`, rebuild `@objectstack/lint`, then read `rules dispatched (2)` off
* the run's own header before believing any number (`ablation-dist-preflight`
* both ways: marker `P2 (#4463): judges an object/field declaration` PRESENT in
* `dist` on the BEFORE leg, ABSENT again after restoring):
*
* N=420, object write BEFORE (2 rules) AFTER (7 rules) delta
* stub 3.6-4.4 ms 8.5-8.6 ms ~ +4.6 ms
* real (showcase decls) 25.5-29.2 ms 282.8-304.4 ms ~ +265 ms
*
* N=21, object write BEFORE (2 rules) AFTER (7 rules) delta
* stub 0.19-0.21 ms 0.41-0.45 ms ~ +0.23 ms
* real (showcase decls) 1.20-1.32 ms 16.7-18.9 ms ~ +16.5 ms
*
* ⛔ The real-shape delta is NOT the stub delta scaled by the totals ratio, and
* this is the measurement that shows it: at N=420 the totals ratio is ~34x, so
* `+3.25 x 34` would predict ~+110 ms where the measured delta is ~+265 ms. A
* ratio measured on TOTALS does not transfer to a DELTA between two rule sets,
* because the sets walk different keys. `--mode per-rule` says exactly why —
* on the real shape at N=420 `validateRuleCompilability` alone is ~233 ms (82%
* of the whole gate) while on the stub shape it is ~0.04 ms (0.5%), the stub
* carrying no validation rules for it to judge. The two rules that were already
* at the door cost the same on both legs (~23-25 ms, real, N=420), as they must.
*/

import {
Expand Down
Loading