CVE-2026-98223
NoneCVSS
—
EPSS
0.2%
Published
Oct 6, 2026
CISA due date
—
Cloudanix analysis
A Linux kernel memory bug can produce unexpected warnings or crashes when the system handles certain file operations under load.
- Exploitation
- Local Privilege
- Attacker needs
- Low Privilege
- Complexity
- Moderate
- Impact
- Denial Of Service
- Component
- OS kernel
- Class
- Race Condition
- Remediation
- Upgrade
- Effort
- Medium
- Patch risk
- Low
What it actually does
Race condition in Linux kernel filemap handling allows mapped dropbehind folios to trigger folio_unmap_invalidate() calling unmap_mapping_folio() under i_mmap_rwsem in non-preemptible task context, producing sleeping-in-atomic warnings.
Why it matters to the business
Linux systems may experience kernel warnings or instability during memory and file I/O workloads, reducing reliability and requiring unplanned reboots or patches.
How an attack unfolds
An attacker triggers fault-around mapping of ready dropbehind folios followed by writeback completion that invokes the unmap path without page-cache lookup clearing dropbehind state, but no public exploits or threat actors are known.
How to fix it
Apply upstream kernel patches that retain mapped dropbehind folios instead of forcibly unmapping them; rebuild and deploy updated kernel images across affected distributions.
How to confirm the fix worked
Boot patched kernel and exercise file mapping plus writeback workloads while monitoring dmesg for absence of sleeping-in-atomic warnings referencing folio_unmap_invalidate or filemap.
Typical attack path
- Map file pages to establish competing cached user
- Mark folios for dropbehind
- Initiate writeback while holding non-preemptible context
- Trigger folio_end_dropbehind to call unmap_mapping_folio under atomic constraints
How to detect it
Search kernel logs and dmesg for sleeping-in-atomic or WARNING messages containing folio_unmap_invalidate, filemap, or dropbehind; correlate with processes performing mmap and writeback.
What your security team should do
- Prioritize kernel updates in next maintenance window given 17-day publication window
- Add kernel log monitoring for folio-related atomic sleep warnings
- Test memory-pressure and file-I/O workloads in staging before production rollout
The record
In the Linux kernel, the following vulnerability has been resolved: mm: filemap: retain mapped dropbehind folios Fault-around can map ready dropbehind folios without going through the normal page-cache lookup that clears dropbehind. A mapping represents a competing cached user, so retain the folio instead of forcibly unmapping it when writeback completes. For a mapped folio, folio_unmap_invalidate() can call unmap_mapping_folio(), which takes i_mmap_rwsem and may sleep. Retaining mapped folios avoids this path when folio_end_dropbehind() runs in non-preemptible task context. Tal was able to trigger a sleeping-in-atomic warning due to this [1]. Unmapped dropbehind folios continue through the existing invalidation path.
Affected packages
| Ecosystem | Package | Introduced | Fixed |
|---|