ci(memtrack): benchmark memtrack's own tracking overhead

[not-matthias]

Add walltime benchmarks that measure codspeed-memtrack's own tracking overhead.

The memory benchmarks only report the tracked program's numbers, so the tracker's
fixed costs (BPF load, uprobe attach, ring-buffer drain, event encoding) were
invisible: regressions in the tracker itself looked like noise in the workload.
Each benchmark times a full codspeed-memtrack track invocation of a
representative command (ls, dd, tar) through the walltime instrument.

Every workload runs twice — RSS-only and RSS+rmap — so rmap's overhead is directly
comparable in CI instead of being benchmarked by hand on random machines. The
tracked commands redirect their output away and run with generous warmup/max
times, since one tracked run already pays the fixed BPF setup cost.

Stacked follow-up will use these benches to drive and verify tracking-overhead
reductions.

Refs COD-3179

[greptile-apps]

Greptile Summary

The PR adds a required Linux CI job that measures codspeed-memtrack startup and tracking overhead through the wall-time runner.

• Installs memtrack, grants its BPF capabilities, and builds the release CodSpeed CLI.
• Adds RSS-only and RSS+rmap benchmarks for ls, tar, and dd.
• Makes the new benchmark job a dependency of the aggregate CI check.

Confidence Score: 5/5

The PR appears safe to merge, with no concrete blocking or independently actionable non-blocking issue identified.

The workflow installs and configures the same memtrack executable used by the configured targets, and the benchmark commands have bounded execution settings and compatible output behavior.

Important Files Changed

Filename	Overview
.github/workflows/ci.yml	Adds the memtrack benchmark setup and execution job and makes it part of the required aggregate check; no actionable defect was established.
crates/memtrack/codspeed.yml	Defines six wall-time targets comparing RSS-only and RSS+rmap tracker overhead with bounded warmup and measurement times.

Flowchart

%%{init: {'theme': 'neutral'}}%%
flowchart LR
  CI[GitHub Actions job] --> Install[Install memtrack and grant capabilities]
  Install --> Build[Build release CodSpeed CLI]
  Build --> Runner[Run wall-time benchmark configuration]
  Runner --> RSS[RSS-only ls, tar, and dd]
  Runner --> RMap[RSS plus rmap ls, tar, and dd]
  RSS --> Check[Required aggregate CI check]
  RMap --> Check

Loading

_{Reviews (1): Last reviewed commit: "fix(memtrack): launch rmap benchmarks th..." | Re-trigger Greptile}

[codspeed-hq]

Merging this PR will not alter performance

⚠️ 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.

✅ 17 untouched benchmarks
🆕 6 new benchmarks

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
🆕	WallTime	memtrack track dd	N/A	1.5 s	N/A
🆕	WallTime	memtrack track ls	N/A	1.5 s	N/A
🆕	WallTime	memtrack track tar	N/A	5 s	N/A
🆕	WallTime	memtrack track dd (rmap)	N/A	1.6 s	N/A
🆕	WallTime	memtrack track ls (rmap)	N/A	1.6 s	N/A
🆕	WallTime	memtrack track tar (rmap)	N/A	5 s	N/A

---

_{Comparing memtrack-walltime-benches (a2d6c6f) with main (8868af5)}