drivers/ras

Reliability, Availability, and Serviceability (RAS) infrastructure

Common kernel-side plumbing for Reliability, Availability, and Serviceability features — the framework that collects, tracks, and reports hardware error events (such as memory ECC errors and CPU machine checks) so that other subsystems and userspace tools like rasdaemon can act on them. It is not tied to any single device or vendor.

not-a-driver conf=1.00 deploy=none replacement=none subsystem=ras category=not-a-driver

100%

recommendation

This is not actually a hardware driver directory; it is shared infrastructure that other drivers and subsystems build on top of to surface hardware error information. It should remain in the tree as a foundational component, but it falls outside the scope of a per-driver keep/remove decision.

repository signals

19 files

6,508 source lines

45 commits, 5y

+6,062 / −209 lines added / removed, 5y

20 authors, 5y

monthly commits · 2021-04-21 → 2026-04-21 · 45 total · active in 20/61 months

sources

No sources cited.

codex reasoning notes (technical)

Not a driver directory: generic RAS subsystem support/helpers under drivers/ras, not kernel-bound hardware driver code for a distinct device family.