feat(memory): apply kernel memory tunables before memory-mode runs

[not-matthias]

Benchmark repos running codspeed --mode memory currently need a hand-written CI step to stabilise kernel memory behaviour before measuring: disable THP, zero vm.compaction_proactiveness / vm.swappiness / kernel.numa_balancing, swapoff -a, and drop the page cache before each suite.

The runner now does this itself at the start of every memory-mode execution.

Behaviour

• Applied only when a CI run environment is detected, so developer machines are not silently mutated. Skipped entirely when privileges cannot be elevated without a password prompt.
• Best effort: a knob that cannot be applied is a warning, never fatal.
• MemoryTunables::apply() returns a guard holding the previous value of every knob it changed, and restores it on drop — including on the error paths, where teardown() never runs. Only knobs that were not already at the target are captured, and THP is restored to the exact mode it held (madvise stays madvise).
• swapoff -a is guarded: skipped on zram devices (it permanently resets their disksize) and whenever the swapped pages would not fit in MemAvailable. Restoring re-enables the recorded entries one by one, since swapon -a would miss a swap file absent from /etc/fstab.
• The page cache drop has no previous state to restore; the node is a write-only trigger.
• Applied in run() rather than setup(), so the page cache is dropped before every suite and --skip-setup does not bypass it. The sysctl/THP writes are no-ops on later suites since ensure_sysctl compares before writing.

Commits

1. refactor(executor): move linux_sysctl into executor helpers — pure move, the sysctl primitive is no longer walltime-specific.
2. feat(memory): apply kernel memory tunables before memory-mode runs

Verification

• cargo clippy --all-targets clean, cargo test --lib (pre-existing failures on this host are the sudo-requiring tests, which write kernel.kptr_restrict and are identical on main).
• Local non-CI run logs Not running in CI, skipping kernel memory tunables; host THP unchanged.
• Privileged container: THP enabled/defrag become [never], vm.compaction_proactiveness = 0, vm.swappiness = 0, and the swap guard correctly logged Leaving swap enabled: swapped pages do not fit in available memory.

[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

---

_{Comparing cod-3300-kernel-memory-tunables (e89df29) with main (8868af5)}

[greptile-apps]

Greptile Summary

The PR adds best-effort Linux memory stabilization around memory-mode executions and restores reversible host settings after each run. It also centralizes Linux profiling sysctl handling and gives wall-time executors ownership of their restoration guards.

• Adds THP, swap, and page-cache tuning for memory benchmarks.
• Records successfully disabled swap areas and restores their paths and explicit priorities.
• Moves profiling sysctl handling into shared executor helpers and restores changed values when the wall-time executor is dropped.

Confidence Score: 4/5

The PR is not yet safe to merge because local privileged memory runs still mutate host-global memory settings despite the stated CI-only boundary.

MemoryExecutor invokes the tunables unconditionally, and MemoryTunables gates only on passwordless elevation, so a local root or passwordless-sudo run still disables THP and eligible swap and irreversibly drops the local page cache.

Files Needing Attention: src/executor/memory/executor.rs, src/executor/memory/tunables.rs

Important Files Changed

Filename	Overview
src/executor/memory/tunables.rs	Implements best-effort THP, swap, and page-cache tuning with drop-based restoration.
src/executor/memory/executor.rs	Applies the new tunables guard for the lifetime of each memory-mode execution.
src/executor/helpers/linux_sysctl.rs	Moves Linux sysctl handling into a shared helper and adds restoration guards.
src/executor/wall_time/executor.rs	Retains changed profiling sysctls on the executor so their original values can be restored.
src/executor/tests.rs	Updates wall-time test setup so each executor owns its sysctl restoration guard.

Flowchart

%%{init: {'theme': 'neutral'}}%%
flowchart TD
  A[MemoryExecutor run] --> B[Apply memory tunables]
  B --> C[Disable THP]
  B --> D[Disable eligible swap]
  B --> E[Drop page cache]
  C --> F[Run memtrack benchmark]
  D --> F
  E --> F
  F --> G[Drop MemoryTunables guard]
  G --> H[Restore swap entries]
  G --> I[Restore THP mode]

Loading

_{Reviews (5): Last reviewed commit: "fix(walltime): restore profiling sysctls" | Re-trigger Greptile}

[art049]

I know it was the case before but it hurts readability a bit

[not-matthias]

@greptileai review again