---
id: "CVE-2026-98505"
kind: "vulnerability"
aliases: ["CVE-2026-98505"]
title: "Linux kernel: DAMON hugetlb pages can lose hardware dirty-bit updates"
severity: "unknown"
cvss: null
epss: null
exploited: false
status: "open"
category: "runtime"
products: []
ecosystems: []
tags: ["damon", "hugetlb", "linux-kernel", "runtime"]
not_affected_if: ["damon-disabled", "no-hugetlb-monitoring"]
published_at: "2026-10-18T16:17:01+00:00"
first_seen_at: "2026-10-23T21:57:15+00:00"
updated_at: "2026-10-25T05:25:29+00:00"
sources: [{"source": "nvd", "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-98505", "first_seen_at": "2026-10-23T21:57:15+00:00"}]
---

# Linux kernel: DAMON hugetlb pages can lose hardware dirty-bit updates

In the Linux kernel's DAMON memory monitor, clearing the accessed bit on hugetlb pages could overwrite hardware-set dirty bits during a race. The sources describe this as a data-integrity bug with no confirmed security impact.

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

## How to check

- `uname -r` to see the running kernel version
- unknown whether DAMON hugetlb monitoring is enabled on the system; the sources do not say how to check

## Fix

- Move to a kernel build that includes the upstream fix for mm/damon/vaddr hugetlb handling (the sources do not list fixed versions)

## Timeline

- 2026-10-23T21:57:15+00:00 nvd https://nvd.nist.gov/vuln/detail/CVE-2026-98505

## From the sources

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

mm/damon/vaddr: avoid hw-driven pte updates during damon_hugetlb_mkold()

damon_hugetlb_mkold() reads the page table entry into a local variable,
unsets the accessed bit in the variable, and updates the page table entry
with the updated variable value.  If hardware updates the same page table
entry in parallel, the hw updates could be lost.  For example,
hardware-updated dirty bits might be lost.

Avoid the parallel updates by clearing the page table entry when reading
it together, using huge_ptep_get_and_clear().  If a parallel write to the
memory is made after the clearing, the hw will see the page table entry is
cleared, trigger page fault and wait until it is handled.  The page fault
handling will wait for damon_hugetlb_mkold() due to the page table lock.

Because hugetlbfs is an in-memory file system and hugetlb pages cannot be
reclaimed, no critical issue is expected to my best knowledge.  But
definitely this is a nasty bug that should be fixed sooner rather than
later.

The issue was discovered [1] by Sashiko.

---

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