PT-2026-152465
CVE-2026-99585
·
Published
2026-11-20
·
Updated
2026-11-20
Description by
Score
N/A
Not availableNo severity ratings or metrics are available. When they are, we'll update the corresponding info on the page.
In the Linux kernel, the following vulnerability has been resolved:
KVM: arm64: Use kvm s2 fault vma info in gmem abort()
gmem abort() maps at s2fd->fault ipa, which HPFAR EL2 holds at 4K
granularity whatever the page size. The generic page-table code aligns
it, but pkvm pgtable stage2 map() looks up existing mappings over
[addr, addr + size), so on a pKVM host with pages larger than 4K a
guest memfd-backed guest that faults past the first 4K of a page can
find its neighbour's mapping, get -EAGAIN and take the same fault
forever.
The memory fault exit reports s2fd->fault ipa too, which for a nested
guest's fault is the nested IPA rather than the canonical one
kvm gmem get pfn() failed on.
Take the addresses from kvm s2 fault vma info instead, as
user mem abort() does, and report the canonical gfn in the memory fault
exit. kvm s2 fault get vma info() itself isn't called: a guest memfd
memslot's userspace addr doesn't need to be backed by a VMA.
Found an issue in the description? Have something to add? Feel free to write us 👾