test(fspy): benchmark tracking overhead

[wan9chi]

Motivation

fspy does not have a stable cross-platform benchmark that isolates the wall-clock cost it adds to tracked child processes. A small synthetic workload makes regressions visible without introducing real-world repository variance.

Changes

• Add one workload: four threads each open and immediately close 2,048 pre-created files.
• Measure the same dynamically linked target untracked and tracked on Linux, macOS, and Windows Namespace runners.
• Add a statically linked Linux variant to exercise seccomp user notification as well as LD_PRELOAD.
• Normalize each result as tracked/untracked overhead, then compare with the latest main result.
• Persist results as workflow artifacts and update a sticky PR comment for same-repository PRs; fork PRs get job summaries and artifacts.

[wan9chi]

• test(fspy): benchmark tracking overhead #566 👈 (View in Graphite)
• main

This stack of pull requests is managed by Graphite. Learn more about stacking.

[socket-security]

Review the following changes in direct dependencies. Learn more about Socket for GitHub.

Diff	Package	Supply Chain Security	Vulnerability	Quality	Maintenance	License
	cargo/​serde_json@​1.0.149 ⏵ 1.0.151
	cargo/​criterion2@​3.0.4

View full report

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 8541d088c2

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[github-actions]

fspy benchmark

linux

fspy/dynamic/launch     time:   [49.993 % 50.696 % 51.392 %]
Found 2 outliers among 10 measurements (20.00%)
  1 (10.00%) low mild
  1 (10.00%) high mild
mean   [49.993 % 51.392 %] std. dev.      [0.4529 % 1.6045 %]
median [50.058 % 51.601 %] med. abs. dev. [0.1568 % 2.1289 %]
fspy/dynamic/access     time:   [52.268 % 52.335 % 52.406 %]
mean   [52.268 % 52.406 %] std. dev.      [0.0693 % 0.1492 %]
median [52.220 % 52.424 %] med. abs. dev. [0.0148 % 0.2204 %]
fspy/dynamic/access-tail
                        time:   [69.527 % 70.654 % 71.827 %]
mean   [69.527 % 71.827 %] std. dev.      [1.1625 % 2.4853 %]
median [68.866 % 71.922 %] med. abs. dev. [0.1687 % 3.3857 %]
fspy/dynamic/access-concurrent
                        time:   [55.101 % 55.200 % 55.319 %]
Found 1 outliers among 10 measurements (10.00%)
  1 (10.00%) high mild
mean   [55.101 % 55.319 %] std. dev.      [0.0791 % 0.2627 %]
median [55.032 % 55.270 %] med. abs. dev. [0.0150 % 0.2775 %]
fspy/static/access      time:   [1049.5 % 1050.2 % 1050.9 %]
mean   [1049.5 % 1050.9 %] std. dev.      [0.6442 % 1.5497 %]
median [1049.3 % 1051.2 %] med. abs. dev. [0.1330 % 2.0698 %]
fspy/static/access-concurrent
                        time:   [818.69 % 819.40 % 820.09 %]
mean   [818.69 % 820.09 %] std. dev.      [0.5323 % 1.6119 %]
median [818.59 % 820.03 %] med. abs. dev. [0.1123 % 2.0363 %]

macos

fspy/dynamic/launch     time:   [171.80 % 171.96 % 172.10 %]
Found 1 outliers among 10 measurements (10.00%)
  1 (10.00%) low mild
mean   [171.80 % 172.10 %] std. dev.      [0.1178 % 0.3482 %]
median [171.85 % 172.17 %] med. abs. dev. [0.0072 % 0.4067 %]
fspy/dynamic/access     time:   [66.952 % 67.880 % 69.115 %]
Found 1 outliers among 10 measurements (10.00%)
  1 (10.00%) high severe
mean   [66.952 % 69.115 %] std. dev.      [0.5907 % 2.7437 %]
median [66.679 % 68.348 %] med. abs. dev. [0.2749 % 2.1284 %]
fspy/dynamic/access-tail
                        time:   [66.198 % 66.591 % 66.959 %]
mean   [66.198 % 66.959 %] std. dev.      [0.3433 % 0.8124 %]
median [65.914 % 67.053 %] med. abs. dev. [0.0177 % 1.1374 %]

windows

fspy/dynamic/launch     time:   [35.174 % 35.554 % 35.908 %]
Found 1 outliers among 10 measurements (10.00%)
  1 (10.00%) low mild
mean   [35.174 % 35.908 %] std. dev.      [0.2881 % 0.8402 %]
median [35.230 % 35.981 %] med. abs. dev. [0.0890 % 1.0454 %]
fspy/dynamic/access     time:   [8.0348 % 8.2790 % 8.4726 %]
Found 1 outliers among 10 measurements (10.00%)
  1 (10.00%) low severe
mean   [8.0348 % 8.4726 %] std. dev.      [0.1289 % 0.5452 %]
median [8.1653 % 8.5350 %] med. abs. dev. [0.0130 % 0.4802 %]
fspy/dynamic/access-tail
                        time:   [9.5181 % 9.9708 % 10.390 %]
Found 1 outliers among 10 measurements (10.00%)
  1 (10.00%) low mild
mean   [9.5181 % 10.390 %] std. dev.      [0.3168 % 1.0130 %]
median [9.5634 % 10.359 %] med. abs. dev. [0.0370 % 1.2649 %]
fspy/dynamic/access-concurrent
                        time:   [8.9963 % 9.1425 % 9.2973 %]
mean   [8.9963 % 9.2973 %] std. dev.      [0.1482 % 0.3077 %]
median [8.9463 % 9.4118 %] med. abs. dev. [0.0264 % 0.4118 %]