---
id: "CVE-2026-98894"
kind: "vulnerability"
aliases: ["CVE-2026-98894"]
title: "Linux amdgpu RAS: corrupt CPER entry can hang the record-writing worker"
severity: "unknown"
cvss: null
epss: 0.00198
exploited: false
status: "open"
category: "infrastructure"
products: []
ecosystems: []
tags: ["amd-gpu", "infrastructure", "kernel", "linux"]
not_affected_if: []
published_at: "2026-10-23T12:18:23+00:00"
first_seen_at: "2026-10-23T21:57:33+00:00"
updated_at: "2026-10-25T12:38:50+00:00"
sources: [{"source": "nvd", "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-98894", "first_seen_at": "2026-10-23T21:57:33+00:00"}, {"source": "epss", "url": "https://api.first.org/data/v1/epss?cve=CVE-2026-98894", "first_seen_at": "2026-10-25T12:38:50+00:00"}]
---

# Linux amdgpu RAS: corrupt CPER entry can hang the record-writing worker

A corrupt CPER record can produce an entry size that stops the ring read pointer from advancing, so the recovery loop spins while holding the CPER ring mutex. The fix resets the ring to empty when no progress is made, which drops the corrupt data and lets the writer continue.

_Summary, check and fix written by a model from the sources; labelled as such._

## Timeline

- 2026-10-23T21:57:33+00:00 nvd https://nvd.nist.gov/vuln/detail/CVE-2026-98894
- 2026-10-25T12:38:50+00:00 epss https://api.first.org/data/v1/epss?cve=CVE-2026-98894

## From the sources

In the Linux kernel, the following vulnerability has been resolved:

drm/amd/ras: reset CPER ring on corrupt entry size

When CPER ring overflow handling advances the read pointer, it trusts the
parsed entry size from the current ring contents. Corrupt CPER data can
produce an entry size that does not advance rptr after dword conversion
and pointer masking.

In that case the recovery loop keeps testing the same location while
holding the CPER ring mutex. This can hang the worker that is writing the
next CPER record.

Detect a no-progress rptr update and reset the CPER ring to an empty
state instead. This drops the corrupt contents and lets the writer leave
the recovery path without spinning.

---

_Free to read at a human pace. An API with keys and an MCP server are in preparation; early access: hello@infrastack.ai_