Rpi Pcie
Raspberry Pi 5 PCIe link health (negotiated vs. capable speed/width per device, HAT+ ID EEPROM, recent pcie dmesg lines) via sysfs, as a typed, versioned Swamp model — catches link training down to a lower speed/width, the common silent cause of a mysteriously slow NVMe SSD.
Documentation, descriptions and setup brought up to date with running every workflow under a single swamp serve (2026-11-05).
Global Arguments
| Argument | Type | Description |
|---|---|---|
| pciDevicesBasePath? | string | sysfs path enumerating PCI devices. |
| hatBasePath? | string | Device-tree path for HAT+ ID EEPROM data. Its absence is a normal, |
| dmesgLineLimit? | number | Maximum number of pcie-matching dmesg lines to keep, most recent last. |
Resources
Read PCIe device link state from pcie-local and assert linkDegraded == false — the silent-performance-loss signal this model exists to catch (link training down to a lower speed/width presents only as "my SSD is mysteriously slow"). On this machine (no NVMe base attached yet) the assert exercises the always-present internal RP1 link, which is expected to stay healthy; re-verify against the actual NVMe device once the base is physically attached. Scheduled every 5 minutes, the same cadence as the
Ship the workflow in the package so a single ops repo can pull it and run it under swamp serve's own scheduler.
Added 1 workflows
Judge link degradation against the upstream port's max speed/width as well as the device's own, so an x4 NVMe behind the Pi 5's x1 port is no longer reported as degraded.
- Has README or module doc2/2earned
- README has a code example1/1earned
- README is substantive1/1earned
- Most symbols documented1/1earned
- No slow types (deprecated)1/1earned
- Dependencies pass trust audit2/2earned
- Docs keep secrets in a vault1/1earned
- Has description1/1earned
- Platform support declared (or universal)2/2earned
- License1/2partial
no SPDX license declared
- Verified public repository2/2earned