fix: emit perf maps when not parsing the full inlining logs from V8

[GuillaumeLagrange]

No description provided.

[greptile-apps]

Greptile Summary

This PR restores the --perf-basic-prof V8 flag for walltime mode when CODSPEED_V8_LOG is not set, and removes the now-released alpha runner version pin from the codspeed-walltime CI job.

• introspection.ts: Adds an else branch so that when CODSPEED_V8_LOG is unset in walltime mode, --perf-basic-prof is pushed — this writes a /tmp/perf-<pid>.map file that linux perf / samply uses for JS symbol resolution alongside the always-emitted jitdump (--perf-prof).
• codspeed.yml: Drops the runner-version: v4.17.7-alpha.2 pin and its associated TODO comment from the codspeed-walltime job, confirming the runner is now generally available.

Confidence Score: 5/5

Safe to merge — restores a flag that was accidentally removed and cleans up a released-runner pin.

Both changes are narrow and correct: the --perf-basic-prof fallback is only active in walltime mode when CODSPEED_V8_LOG is absent, and it complements (not conflicts with) the always-present --perf-prof jitdump flag. The CI change simply removes a version pin whose TODO comment explicitly marked it for removal once the runner shipped.

No files require special attention.

Important Files Changed

Filename	Overview
packages/core/src/introspection.ts	Adds --perf-basic-prof fallback in walltime mode when CODSPEED_V8_LOG is unset, restoring the perf.map output path for linux perf/samply symbol resolution.
.github/workflows/codspeed.yml	Removes the runner-version: v4.17.7-alpha.2 pin (and its TODO comment) from the codspeed-walltime job, indicating the runner has been officially released.

Flowchart

%%{init: {'theme': 'neutral'}}%%
flowchart TD
A[getV8Flags - walltime mode] --> B["push --perf-prof\n(always emit jitdump)"]
B --> C{CODSPEED_V8_LOG\nenv var set?}
C -- Yes --> D["push --log-code\n--no-log-source-code\n--no-logfile-per-isolate\n--logfile=<dir>/codspeed-v8-%p.log"]
C -- No --> E["push --perf-basic-prof\n(write /tmp/perf-PID.map)"]
D --> F[Return flags]
E --> F

Loading

%%{init: {'theme': 'base', 'themeVariables': {"darkMode": true, "background": "#0d1117", "primaryColor": "#21262d", "primaryTextColor": "#e6edf3", "primaryBorderColor": "#8b949e", "lineColor": "#8b949e", "textColor": "#e6edf3", "edgeLabelBackground": "#161b22", "actorBkg": "#21262d", "actorBorder": "#8b949e", "actorTextColor": "#e6edf3", "actorLineColor": "#8b949e", "signalColor": "#8b949e", "signalTextColor": "#e6edf3", "noteBkgColor": "#373320", "noteBorderColor": "#d4a72c", "noteTextColor": "#f0e6c0", "labelBoxBkgColor": "#21262d", "labelBoxBorderColor": "#8b949e", "labelTextColor": "#e6edf3", "loopTextColor": "#e6edf3", "activationBkgColor": "#30363d", "activationBorderColor": "#8b949e"}}}%%
flowchart TD
A[getV8Flags - walltime mode] --> B["push --perf-prof\n(always emit jitdump)"]
B --> C{CODSPEED_V8_LOG\nenv var set?}
C -- Yes --> D["push --log-code\n--no-log-source-code\n--no-logfile-per-isolate\n--logfile=<dir>/codspeed-v8-%p.log"]
C -- No --> E["push --perf-basic-prof\n(write /tmp/perf-PID.map)"]
D --> F[Return flags]
E --> F

Loading

_{Reviews (2): Last reviewed commit: "fix: have node output perf maps when not..." | Re-trigger Greptile}

[codspeed-hq]

Merging this PR will improve performance by 58.32%

⚠️Unknown Walltime execution environment detected

Using the Walltime instrument on standard Hosted Runners will lead to inconsistent data.

For the most accurate results, we recommend using CodSpeed Macro Runners: bare-metal machines fine-tuned for performance measurement consistency.

⚠️Different runtime environments detected

Some benchmarks with significant performance changes were compared across different runtime environments,
which may affect the accuracy of the results.

Open the report in CodSpeed to investigate

⚡ 4 improved benchmarks
❌ 7 regressed benchmarks
✅ 227 untouched benchmarks

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	Memory	short body	536 B	3,656 B	-85.34%
❌	Simulation	test_recursive_fibo_10	140.2 µs	275.4 µs	-49.1%
❌	WallTime	short body	2.4 µs	2.8 µs	-15.52%
❌	WallTime	short body	2.3 µs	2.6 µs	-12.33%
❌	WallTime	short body	2.3 µs	2.6 µs	-12.09%
❌	WallTime	short body	2.3 µs	2.6 µs	-11.27%
❌	WallTime	test_recursive_cached_fibo_10	1 µs	1.1 µs	-10.53%
⚡	Memory	fibo 15	5,120 B	5 B	×1,000
⚡	Simulation	recursive fibo 10	316.7 µs	138.1 µs	×2.3
⚡	Memory	wait 1ms	10 B	7 B	+42.86%
⚡	WallTime	test_iterative_fibo_10	276 ns	228 ns	+21.05%

Tip

Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing cod-2987-e2e-tests-show-no-flamegraph-for-vitest (c88994c) with main (36834e8)}