Executive brief
A vulnerability in the Linux kernel's memory management system can cause a system to become unresponsive or crash under heavy load. This occurs when the system is managing large amounts of memory and swap space, leading to a 'soft lockup' where the processor is stuck in a loop and cannot perform other tasks. This could result in a service outage or system instability in high-performance computing environments.
Technical details
A soft lockup vulnerability exists in the Linux kernel's mm/swap component within the swap_reclaim_full_clusters function. On large-scale systems (e.g., 320 CPUs, 1TB RAM), heavy memory stress can cause the swap reclaim process to iterate through a large number of full clusters without yielding the CPU. This lack of scheduling points triggers the kernel's watchdog timer, leading to a kernel panic. The fix introduces periodic cond_resched() calls within the reclaim loop to ensure the CPU can handle other tasks. This is primarily reachable under local stress conditions and impacts system availability.
Affected products
- Linux Linux Kernel 6.12, 6.18, 7.1
Timeline
- 2026-05-06: disclosed: Initial patch submission by Zijiang Huang
- 2026-07-18: patched: Commits merged into stable branches
- 2026-07-25: advisory: CVE-2026-64415 published