assay-runner-linux 3.34.0

Internal/experimental substrate for Assay measured-run workflows. Linux-only platform adapter for the Assay-Runner candidate; hosts cgroup placement primitives (Phase 2D Slice 3). No standalone product guarantee; API surface remains narrow and intentionally undocumented for third-party use; semver tracks the Assay workspace.
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30

// Cgroup v2 placement uses raw libc syscalls (pidfd_open, pidfd_send_signal)
// for race-free PID handling during graceful kill. The same allow lived on
// `assay-cli/src/main.rs` before the Slice 3 move. The workspace default
// remains `unsafe_code = "deny"`; this crate is the platform adapter where
// such syscalls are intentional and unavoidable.
#![allow(unsafe_code)]
#![warn(unsafe_op_in_unsafe_fn)]
#![warn(clippy::undocumented_unsafe_blocks)]

//! Linux-only platform adapter for the Assay-Runner candidate.
//!
//! This crate is the Phase 2D Slice 3 result of the Assay-Runner extraction
//! roadmap (see `docs/reference/runner/extraction-roadmap.md`). It hosts
//! Linux-specific runner platform primitives that previously lived inside
//! `crates/assay-cli/` and therefore prevented the runner candidate from
//! being moved without dragging `assay-cli` along.
//!
//! Current scope is intentionally narrow: only cgroup v2 placement
//! (`CgroupManager`, `SessionCgroup`). No eBPF or monitor adapter lives
//! here yet — that boundary work is Slice 4 per the roadmap, and is not
//! opened by this crate's existence. Likewise, this crate does not
//! provide a platform-abstraction trait; macOS and Windows support are
//! out of scope until separate platform spikes open under
//! `platform-and-extraction-readiness.md`.
//!
//! The crate is `publish = false` until Slice 7 (repository extraction).

mod cgroup;

pub use cgroup::{CgroupManager, SessionCgroup};