mm: Fix a hmm_range_fault() livelock / starvation problem
Linux
Affected versions
| Product | Affected | Fixed in |
|---|---|---|
| Linux Linux |
| No fix yet |
| Linux Linux |
|
|
Linux
| Product | Affected | Fixed in |
|---|---|---|
| Linux Linux |
| No fix yet |
| Linux Linux |
|
|
6.18.196.19.97.0In the Linux kernel, the following vulnerability has been resolved:
mm: Fix a hmm_range_fault() livelock / starvation problem
If hmm_range_fault() fails a folio_trylock() in do_swap_page, trying to acquire the lock of a device-private folio for migration, to ram, the function will spin until it succeeds grabbing the lock.
However, if the process holding the lock is depending on a work item to be completed, which is scheduled on the same CPU as the spinning hmm_range_fault(), that work item might be starved and we end up in a livelock / starvation situation which is never resolved.
This can happen, for example if the process holding the device-private folio lock is stuck in migrate_device_unmap()->lru_add_drain_all() sinc lru_add_drain_all() requires a short work-item to be run on all online cpus to complete.
A prerequisite for this to happen is: a) Both zone device and system memory folios are considered in migrate_device_unmap(), so that there is a reason to call lru_add_drain_all() for a system memory folio while a folio lock is held on a zone device folio. b) The zone device folio has an initial mapcount > 1 which causes at least one migration PTE entry insertion to be deferred to try_to_migrate(), which can happen after the call to lru_add_drain_all(). c) No or voluntary only preemption.
This all seems pretty unlikely to happen, but indeed is hit by the "xe_exec_system_allocator" igt test.
Resolve this by waiting for the folio to be unlocked if the folio_trylock() fails in do_swap_page().
From the CVE record · Linux
A locking issue in the Linux kernel’s memory migration path can cause hmm_range_fault() to spin indefinitely, starving work needed to release a device-private folio lock. It requires specific folio migration and mapping conditions and no or only voluntary preemption. The input does not identify an attack route, authentication requirement, or user-interaction requirement.
Engineers running Linux kernels with the affected memory migration code should check their vendor’s advisory; affected versions are not specified.