CVE-2026-99653: mm: fix unkillable fault loop on MAP_DROPPABLE memory
In the Linux kernel, the following vulnerability has been resolved:
mm: fix unkillable fault loop on MAPDROPPABLE memory
handlemmfault() suppresses VMFAULTOOM for VMDROPPABLE mappings.
For user-mode faults, pending signals are processed before retrying the access. Kernel-mode faults on user buffers instead return directly to the faulting instruction.
With persistent allocation failures (esp. memcg limits), a syscall, such as write(), can therefore loop while not dropping locks.
Only suppress VMFAULTOOM when FAULTFLAGUSER is set, allowing kernel-mode faults and GUP to fail and signal allocation errors.
Fix generated by an LLM and sent by Chris Mason in private.
[ david: simplify description ]
Affected Software
Remediation
Recommended actions to resolve this vulnerability, in priority order.
- Upgrade
Upgrade
linux/Kernelto a version that resolves this vulnerability.Fixed in 6.12.118 - Upgrade
Upgrade
linux/Kernelto a version that resolves this vulnerability.Fixed in 6.18.64 - Upgrade
Upgrade
linux/Kernelto a version that resolves this vulnerability.Fixed in 7.3.5
Frequently Asked Questions
What conditions are required to trigger the loop?
A kernel-mode access to a user buffer backed by a MAP_DROPPABLE mapping must encounter persistent memory-allocation failures. Memory cgroup limits are specifically identified as a condition that can make those failures persistent.
What impact can this have on an affected system?
A syscall such as write() can repeatedly fault on the buffer without dropping locks, creating an unkillable fault loop. User-mode faults are different because pending signals are processed before the access is retried.
What behavior changes with the fix?
The fix limits suppression of VM_FAULT_OOM for droppable mappings to faults marked FAULT_FLAG_USER. Kernel-mode faults on user buffers and GUP can then fail and report allocation errors instead of retrying indefinitely.