Fix Safari dormant overhead in Live Debugger instrumentation

[watson]

What and why?

The Live Debugger runtime benchmark surfaced a large dormant-probe overhead on Safari: a tiny instrumented function cost ~8.9 ns/call (~1,100% over baseline) versus ~0 ns on Chrome and ~1.6 ns on Firefox. The root cause is that the transform hoisted each function's captured parameters into a per-call arrow closure (const $dd_eN = () => ({...})). A nested closure that captures parameters forces JavaScriptCore — Safari and all iOS browsers, since they're all WebKit — to heap-allocate the function's scope on every invocation, even when probes are dormant and the closure is never created or called. V8 and SpiderMonkey sink that allocation; JavaScriptCore does not, so it dominated the dormant overhead. Because every instrumented iOS app pays this cost on every call, it's worth eliminating.

How?

Inline the captured-arguments object literal ({a, b}) directly at each probe-guarded call site instead of hoisting it into a per-call closure. The object now lives inside the if (probe) / probe ? guards, so dormant calls allocate nothing; gating the closure behind the probe check was confirmed insufficient because the captured scope is still allocated. The active path is unchanged — the same arguments object is built the same number of times when a probe fires. Verified with the runtime benchmark: Safari per-call dormant overhead drops ~73% (8.9 → 2.4 ns) and the Hot workload from ~10 to ~7 ns, with no change on Chrome or Firefox. Transform unit-test snapshots and the plugin README/EXAMPLES output catalog are updated to match the new generated shape.

[watson]

Warning

This pull request is not mergeable via GitHub because a downstack PR is open. Once all requirements are satisfied, merge this PR as a stack on Graphite.
Learn more

• Add decorator support to Live Debugger instrumentation #448
• Fix Live Debugger parenthesized arrow body instrumentation #447
• Fix Live Debugger returned arrow instrumentation #442
• Fix live debugger curried function instrumentation #441
• Fix Safari dormant overhead in Live Debugger instrumentation #438 👈 (View in Graphite)
• Add Live Debugger runtime benchmark coverage #429: 1 other dependent PR (#440)
• master

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

[datadog-official]

🎉 All green!

🧪 All tests passed
❄️ No new flaky tests detected

_{This comment will be updated automatically if new data arrives.
🔗 Commit SHA: 937be82 | Docs | Datadog PR Page | Give us feedback!}

[github-actions]

Live Debugger Runtime Benchmark

SDK-loaded dormant-probe runtime overhead, measured against an uninstrumented bundle in the same browser session.

Browser	Workload	Quality	Per-call overhead upper
chrome	Hot	clean	<= 2.59 ns
chrome	Tiny	clean	<= 0.01 ns
firefox	Hot	clean	<= 10.02 ns
firefox	Tiny	clean	<= 0.01 ns
safari	Hot	clean	<= 5.71 ns
safari	Tiny	clean	<= 1.60 ns

Full diagnostics

browser workload quality per-call overhead upper overhead upper 95% CI A/A diag block CI acf(1) baseline instrumented samples
------- -------- ------- ----------------------- -------------- -------------- -------------- -------------- ------ --------- ------------ ------------------------------
chrome Hot clean <= 2.59 ns <= 5.21% 2.54..2.58 ns -0.02..0.02 ns 2.54..2.59 ns 0.04 50.858 ms 53.467 ms 102 (trim 20%, outliers 4.9%)
chrome Tiny clean <= 0.01 ns <= 0.12% -0.02..0.01 ns -0.00..0.03 ns -0.02..0.01 ns 0.03 76.842 ms 76.990 ms 102 (trim 20%, outliers 13.7%)
firefox Hot clean <= 10.02 ns <= 39.92% 9.99..10.02 ns -0.00..0.01 ns 9.99..10.02 ns -0.30 38.540 ms 53.940 ms 102 (trim 20%, outliers 2.9%)
firefox Tiny clean <= 0.01 ns <= 0.08% 0.00..0.01 ns -0.00..0.00 ns 0.00..0.01 ns 0.07 97.860 ms 97.910 ms 102 (trim 20%, outliers 11.8%)
safari Hot clean <= 5.71 ns <= 22.85% 5.70..5.71 ns 0.00..0.01 ns 5.70..5.71 ns 0.05 51.200 ms 62.900 ms 102 (trim 20%, outliers 4.9%)
safari Tiny clean <= 1.60 ns <= 42.75% 1.59..1.60 ns -0.00..0.00 ns 1.59..1.60 ns -0.08 37.520 ms 53.570 ms 102 (trim 20%, outliers 7.8%)

Raw samples are in the live-debugger-runtime-bench-results artifact.