In the Linux kernel, the following vulnerability has been resolved: mm/hugetlb: fix missing hugetlb_lock for resv uncharge There is a recent report on UFFDIO_COPY over hugetlb: https://lore.kernel.org/all/[email protected]/ 350: lockdep_assert_held(&hugetlb_lock); Should be an issue in hugetlb but triggered in an userfault context, where it goes into the unlikely path where two threads modifying the resv map together. Mike has a fix in that path for resv uncharge but it looks like the locking criteria was overlooked: hugetlb_cgroup_uncharge_folio_rsvd() will update the cgroup pointer, so it requires to be called with the lock held.
OS | Version | Architecture | Package | Version | Filename |
---|---|---|---|---|---|
Debian | 12 | all | linux | < 6.1.94-1 | linux_6.1.94-1_all.deb |
Debian | 11 | all | linux | <= 5.10.223-1 | linux_5.10.223-1_all.deb |
Debian | 999 | all | linux | < 6.8.9-1 | linux_6.8.9-1_all.deb |
Debian | 13 | all | linux | < 6.8.9-1 | linux_6.8.9-1_all.deb |