Cloudanix VulnDB

CVE-2026-89756

None

CVSS

—

EPSS

0.2%

Published

Sep 11, 2026

CISA due date

—

Cloudanix analysis

Schedule a fix SSVC: Track

A kernel bug can make certain memory-cleanup tasks freeze parts of the system for minutes on virtual-machine hosts, slowing everything down until the kernel is updated.

Exploitation
Local Privilege
Attacker needs
Local User
Complexity
Moderate
Impact
Denial Of Service
Component
OS kernel
Class
Denial of service
Remediation
Upgrade
Effort
Medium
Patch risk
Low

What it actually does

Long page migration batches in migrate_pages_batch() fail to report Tasks-RCU quiescent states on PREEMPTION kernels because cond_resched() is a no-op, allowing kcompactd and similar tasks to become RCU holdouts and stall grace periods for minutes during expensive KVM MMU notifier callbacks.

Why it matters to the business

Memory compaction and migration workloads on KVM hosts can trigger prolonged system stalls or RCU-related warnings, degrading performance and availability in large-scale virtualization environments.

How an attack unfolds

An attacker or workload triggers repeated memory compaction or migration (for example via memory pressure) causing migrate_pages_batch() to run extended unmap loops on KVM hosts; without quiescent state reporting the migrating task stalls Tasks-RCU grace periods, amplifying latency for dependent operations.

How to fix it

Update the Linux kernel to a version containing the fix that replaces cond_resched() with cond_resched_tasks_rcu_qs() inside migrate_pages_batch().

How to confirm the fix worked

Confirm the presence of the cond_resched_tasks_rcu_qs() call in migrate_pages_batch() after applying the update and verify that long migration batches no longer produce RCU-tasks stall warnings under the same KVM workload.

Typical attack path

  1. Create memory pressure to invoke compaction or migration
  2. kcompactd or similar kernel thread enters migrate_pages_batch()
  3. Large folio batches trigger repeated expensive KVM MMU notifier invalidations
  4. cond_resched() does nothing on PREEMPTION kernels so no Tasks-RCU quiescent state is reported
  5. Migrating task becomes RCU holdout and stalls grace periods for minutes

How to detect it

Search kernel logs for RCU-tasks stall messages referencing kcompactd, migrate_pages_batch, or tdp_mmu_zap_leafs; correlate with KVM MMU notifier activity during memory compaction.

What your security team should do

  • Prioritize kernel updates on KVM hosts that perform frequent memory compaction
  • Monitor for RCU-tasks stall messages in dmesg on systems running memory-intensive workloads
  • Test the updated kernel under representative compaction loads before broad deployment

The record

In the Linux kernel, the following vulnerability has been resolved: mm/migrate: report RCU-tasks quiescent states in migrate_pages_batch() migrate_pages_batch() unmaps each folio before moving it, and every unmap runs the mmu_notifier invalidate callbacks. On KVM hosts try_to_migrate() ends up in kvm_mmu_notifier_invalidate_range_start() -> tdp_mmu_zap_leafs(), which is expensive, so unmapping a large batch keeps the CPU busy for a long time. The loop already calls cond_resched(), but on PREEMPTION kernels that is a no-op, and involuntary preemption is not a Tasks-RCU quiescent state. A long batch therefore never reports a quiescent state, and the migrating task (e.g. kcompactd) becomes a Tasks-RCU holdout, stalling the Tasks-RCU grace period for minutes, which is common at Meta fleet: INFO: rcu_tasks detected stalls on tasks: 0000000055349ecc: .. nvcsw: 1157401/1157401 holdout: 1 idle_cpu: -1/56 task:kcompactd0 state:R running task Call Trace: tdp_mmu_zap_leafs tdp_mmu_next_root gfn_to_pfn_cache_invalidate_start kvm_mmu_notifier_invalidate_range_start __mmu_notifier_invalidate_range_start try_to_migrate_one try_to_migrate migrate_pages_batch migrate_pages compact_zone compact_node kcompactd kthread Use cond_resched_tasks_rcu_qs() so a quiescent state is reported even when cond_resched() does nothing. This has also been discussed at [1]

Affected packages

Ecosystem Package Introduced Fixed