[marcin-kordas-hoc]

Context

The PR lowers the threshold of the instruction-count verdict from 0.5% to 0.25%. It is one of four alternatives to #1852, stacked on it, and changes only that value.

• Why: smaller regressions get named. In a control run with a fixed 200,000-iteration loop in setCellContents (HF-150 (throwaway control, do not merge): known extra work for the blocking instruction check #1863, run 41010016546), four scenarios besides the 5 regressions rose by +0.261% to +0.380% in executed instructions and were reported ok at 0.5%.
• What changed:
  • .github/workflows/performance.yml: PERF_INSTRUCTIONS_THRESHOLD: '0.25'.
  • Paired tests branch: the README sentence that states the value.
• Trade-off: less margin over effects that are not code. HF-150: one performance report per pull request, with the instruction count as the verdict #1852 found heap state still moving the executed count: +0.17% in a scenario that runs none of the changed code (locally, Node 22). At 0.25% such a shift is close to a false regression.
• Unchanged: the report is advisory; the comparison script keeps 0.5% as its default outside the workflow.

How did you test your changes?

• Paired tests branch: feat/hf-150-alt-threshold-025 in hyperformula-tests (README only).
• npm run test:jest -- unit/performance with the paired tests: 5 suites, 101 tests passed. The tests branch changes no lintable file.
• CI on this PR, with engine code identical to the base (run 41012215331): all three performance jobs passed. The comment reads "no regression above +0.25% in 15 scenarios"; the largest executed change is -0.011%. Wall-clock: 14 ok, 1 inconclusive.

Compatibility with Excel and Google Sheets

Not applicable.

Public API changes

None.

Out of scope and follow-ups

• Attributing the remaining heap effects (allocation locks) to garbage collection, which a lower threshold would need.

Types of changes

• Breaking change (existing functionality no longer works as before) – agreed with the maintainers
• New feature or improvement (a non-breaking change that adds functionality)
• Bug fix (a non-breaking change that fixes an issue)
• Additional language file, or a change to an existing language file (translations)
• Change to the documentation

CI tooling only.

Related issues

1. HF-150 (internal task)

Checklist

• My pull request meets the Definition of Done. External contributors: see also Contributing to HyperFormula.
• I have signed the Contributor License Agreement.
• I read my own diff end to end before requesting a review.
• I added a CHANGELOG.md entry, written by the changelog rules – or none is needed (docs, tests or tooling only).
• I updated the documentation and JSDoc to match the change.
• Breaking change only: I added a section to the migration guide.
• I updated the dev-docs/ pages this change affects – or none needed it.

🤖 Generated with Claude Code

[cloudflare-workers-and-pages]

Deploying with    Cloudflare Workers

The latest updates on your project. Learn more about integrating Git with Workers.

Status	Name	Latest Commit	Preview URL	Updated (UTC)
✅ Deployment successful! View logs	hyperformula-docs	1b51b4f	Commit Preview URL Branch Preview URL	Nov 03 2026, 07:08 AM

[github-actions]

Performance

✅ Executed instructions: no regression above +0.25% in 15 scenarios.
Wall-clock time (information only): ok: 15.

Scenario	Executed instructions	Verdict	GC / compilation (information)	Wall-clock	Verdict (information)
Sheet A	+0.000%	ok	-0.000% / 0.000%	+0.23%	ok
Sheet B	-0.007%	ok	-0.000% / 0.000%	+0.01%	ok
Sheet T	-0.006%	ok	+0.000% / 0.000%	-0.33%	ok
Column ranges	-0.001%	ok	+0.000% / 0.000%	+0.70%	ok
Sorted lookup	-0.004%	ok	+0.000% / 0.000%	+0.47%	ok
Sheet A: change value, add/remove row/column	-0.002%	ok	-0.000% / 0.000%	+4.01%	ok
Sheet B: change value, add/remove row/column	+0.000%	ok	-0.000% / 0.000%	-6.37%	ok
Column ranges - add column	+0.000%	ok	+0.000% / 0.000%	-3.96%	ok
Column ranges - without batch	-0.000%	ok	0.000% / 0.000%	-1.65%	ok
Column ranges - batch	+0.000%	ok	+0.000% / 0.000%	-1.90%	ok
Recalculation fan-out - one edit, 10k dependents	-0.000%	ok	0.000% / 0.000%	-2.40%	ok
Recalculation fan-out - 5 edits without batch	-0.000%	ok	-0.000% / 0.000%	-2.75%	ok
Recalculation fan-out - 5 edits in batch	+0.000%	ok	0.000% / 0.000%	-1.63%	ok
Fan-out sheet - build, 30k rows	-0.000%	ok	-0.000% / 0.000%	-	-
Recalculation fan-out - one edit, 30k dependents	-0.000%	ok	0.000% / 0.000%	-	-
Fan-out sheet - build, 100k rows	-	-	-	+0.17%	ok
Recalculation fan-out - one edit, 100k dependents	-	-	-	-2.91%	ok

- in a whole column: the scenario is measured by one suite only. The instruction count runs the 100k-row scenarios at 30k rows, under names that say so.

Instruction counts

Instructions counted by callgrind in the measured region of each scenario. The verdict compares the executed instructions: the count without garbage collection and JIT compilation, whose share moves with the state of the heap and the compiler, not only with the code under test. A change larger than ±0.25% of the base is reported. GC, Compilation and Total give the change of those parts and of the whole count, in percent of the base count.

Scenario	Base	Head	Change	GC	Compilation	Total	Verdict
Sheet A	4,705,373,458	4,705,385,548	+0.000%	-0.000%	0.000%	+0.000%	ok
Sheet B	1,647,048,997	1,646,928,279	-0.007%	-0.000%	0.000%	-0.004%	ok
Sheet T	1,510,478,293	1,510,384,123	-0.006%	+0.000%	0.000%	-0.004%	ok
Column ranges	5,679,680,641	5,679,620,591	-0.001%	+0.000%	0.000%	-0.001%	ok
Sorted lookup	4,968,289,110	4,968,115,032	-0.004%	+0.000%	0.000%	-0.003%	ok
Sheet A: change value, add/remove row/column	1,067,411,071	1,067,392,994	-0.002%	-0.000%	0.000%	-0.001%	ok
Sheet B: change value, add/remove row/column	1,332,665,644	1,332,665,949	+0.000%	-0.000%	0.000%	+0.000%	ok
Column ranges - add column	1,660,972,455	1,660,976,299	+0.000%	+0.000%	0.000%	+0.000%	ok
Column ranges - without batch	4,552,900,274	4,552,894,521	-0.000%	0.000%	0.000%	-0.000%	ok
Column ranges - batch	1,247,170,503	1,247,172,568	+0.000%	+0.000%	0.000%	+0.000%	ok
Recalculation fan-out - one edit, 10k dependents	219,233,976	219,233,576	-0.000%	0.000%	0.000%	-0.000%	ok
Recalculation fan-out - 5 edits without batch	690,424,164	690,423,683	-0.000%	-0.000%	0.000%	-0.000%	ok
Recalculation fan-out - 5 edits in batch	222,429,289	222,429,647	+0.000%	0.000%	0.000%	+0.000%	ok
Fan-out sheet - build, 30k rows	8,305,905,853	8,305,905,037	-0.000%	-0.000%	0.000%	-0.000%	ok
Recalculation fan-out - one edit, 30k dependents	481,466,254	481,466,152	-0.000%	0.000%	0.000%	-0.000%	ok

Wall-clock times

Scenario	base median (ms)	head median (ms)	change	threshold	noise (CV base/head)	verdict
Sheet A	452.8	453.8	+0.23%	±5.0% †	1.9% / 1.6%	ok
Sheet B	188.5	188.6	+0.01%	±7.4% †	2.5% / 2.7%	ok
Sheet T	156.9	156.3	-0.33%	±8.0% †	4.2% / 3.7%	ok
Column ranges	441.0	444.1	+0.70%	±5.3% †	1.5% / 2.8%	ok
Sorted lookup	11486.8	11540.3	+0.47%	±50.0% ‡	1.1% / 1.8%	ok
Fan-out sheet - build, 100k rows	5001.0	5009.5	+0.17%	±50.0% ‡	0.4% / 1.9%	ok
Sheet A: change value, add/remove row/column	133.4	138.8	+4.01%	±12.0%	9.6% / 14.7%	ok
Sheet B: change value, add/remove row/column	194.8	182.4	-6.37%	±13.3% †	5.7% / 3.4%	ok
Column ranges - add column	158.5	152.2	-3.96%	±6.4% †	2.2% / 2.2%	ok
Column ranges - without batch	412.0	405.3	-1.65%	±6.3% †	2.3% / 5.3%	ok
Column ranges - batch	133.7	131.2	-1.90%	±5.1% †	2.4% / 3.5%	ok
Recalculation fan-out - one edit, 10k dependents	52.1	50.8	-2.40%	±6.4% †	2.7% / 4.7%	ok
Recalculation fan-out - 5 edits without batch	115.1	112.0	-2.75%	±9.0% †	3.8% / 4.5%	ok
Recalculation fan-out - 5 edits in batch	52.7	51.8	-1.63%	±6.8% †	4.5% / 2.8%	ok
Recalculation fan-out - one edit, 100k dependents	330.4	320.8	-2.91%	±50.0% ‡	10.4% / 7.8%	ok

Threshold of an unmarked scenario = max(5%, 3 x the standard error of the difference of the two medians). † = the band of changes observed on identical code in earlier runs, instead of that rule. ‡ = a fixed threshold for a scenario that guards against a change of complexity class. This check is advisory: it reports regressions but does not fail the build.

head 1b51b4fec · base 8869f4703 · tests af17e1009 · run

Executed instructions decide the verdict: they do not depend on the load of the runner. Garbage collection and JIT compilation are counted too but shown apart, in percent of the base count, because they move with the state of the heap and the compiler, not only with the code under test. Wall-clock times vary with the load of the runner and are shown for information. This report does not block merging.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 97.28%. Comparing base (8869f47) to head (1b51b4f).

Additional details and impacted files

@@                   Coverage Diff                    @@
##           feat/hf-150-perf-report    #1864   +/-   ##
========================================================
  Coverage                    97.28%   97.28%           
========================================================
  Files                          207      207           
  Lines                        16218    16218           
  Branches                      3556     3556           
========================================================
  Hits                         15777    15777           
  Misses                         433      433           
  Partials                         8        8           

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.