Skip to main content
No exploitation reportedFix availableAwaiting NVD analysis

CVE-2026-99463

mm/rmap: don't discard lazyfree THPs at PMD level

No exploitation reported by the sources we read.

  • Published2026-11-19
  • Fixed in7.3.2 · 6.18.61 · 6.12.117

Am I affected?

With a domain, we read the version its public headers and pages expose. A distribution build can carry the fix under an older number.

Description

In the Linux kernel, the following vulnerability has been resolved: mm/rmap: don't discard lazyfree THPs at PMD level __discard_anon_folio_pmd_locked() clears a lazyfree THP's PMD before it determines whether the folio can be discarded, and restores it if the folio is dirty or has extra references. It is done under only the PMD lock and the anon_vma read lock inside try_to_unmap_one() and no mmap lock is held. The temporarily none PMD causes UAF. A parallel munmap() or mremap() is supposed to serialize with it on the PMD lock, but zap_pmd_range() (called by munmap()) and move_page_tables() (called by mremap()) do not take the lock when they see the temporarily none PMD and skip the PMD. There is no more page table walk after the none PMD is restored: - munmap(): free_pgtables() waits in unlink_anon_vmas() for the anon_vma lock, then free_pmd_range() drops the restored PMD as bad without removing the rmap, so the folio keeps a stale mapcount. - whole-VMA MREMAP_DONTUNMAP: dontunmap_complete() unlinks the source VMA from its anon_vma while the restored PMD still maps the folio. - plain mremap(): the THP is not moved but unmapped, so the THP's data is lost. In the first two cases, the folio stays mapped, but no VMA that maps it is linked to its anon_vma. When the anon_vma is freed, it is still reachable when one rmap walks the folio. Racing lazyfree reclaim against MREMAP_DONTUNMAP as an unprivileged user reproduces a KASAN use-after-free in folio_lock_anon_vma_read() as reported in the Closes link. One solution is to use an invalidated PMD instead of a cleared one until the discard is committed, but it needs arch fixes first. 1. GUP-fast still follows an invalidated PMD on riscv, LoongArch and s390. 2. sparc64 miscounts thp_pte_count when an invalidated PMD is set again or cleared. To keep the fix simple for backport, stop discarding lazyfree THPs at PMD level instead. try_to_unmap_one() splits the PMD and discards the folio at PTE level. unmap_huge_pmd_locked() and __discard_anon_folio_pmd_locked() are left in place, unused, to minimize this fix for backport and the invalidated PMD approach is planned as a follow-up series to use them again.

Timeline

dates in UTC
  1. 2026-11-1914:42CVE publishedCNA

Versions per branch

OSV
Linux kernel
BranchAffectedFixed in
7.3≥ 6.19.0, < 7.3.27.3.2
6.x≥ 6.13.0, < 6.18.616.18.61
6.x≥ 6.11.0, < 6.12.1176.12.117

as publishedRanges as NVD and OSV publish them; a distribution’s build can be fixed under an older number.

Products declared by the CNA2 products

As Linux, the CNA that assigned this CVE, declared them in the CVE record; versions as written.

  • Linuxother versions: not affected

    https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git

    Affected
    • ≥ 735ecdfaf4e802209caf34b7ac45adf448c54ccc, < 8cf9517996067cfbfc5abdb4cfb3d7e15a7b34a3fixed in 8cf9517996067cfbfc5abdb4cfb3d7e15a7b34a3git
    • ≥ 735ecdfaf4e802209caf34b7ac45adf448c54ccc, < 3594027c292b09d9301b4be80fa8e0c815ffaf2afixed in 3594027c292b09d9301b4be80fa8e0c815ffaf2agit
    • ≥ 735ecdfaf4e802209caf34b7ac45adf448c54ccc, < b8c683bd548268453d50e933aab2e7d50a7ac2befixed in b8c683bd548268453d50e933aab2e7d50a7ac2begit
    • ≥ 735ecdfaf4e802209caf34b7ac45adf448c54ccc, < 984c0700854e35779a2bcce6b02783e709b6efa2fixed in 984c0700854e35779a2bcce6b02783e709b6efa2git
  • Linuxother versions: affected

    https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git

    Affected
    • = 6.11no fix stated

    Not affected < 6.11 · ≥ 6.12.117, ≤ 6.12.* · ≥ 6.18.61, ≤ 6.18.* · ≥ 7.3.2, ≤ 7.3.* · ≥ 7.4-rc1

Source: the CNA record (CVE Program), © The MITRE Corporation,CVE terms of useCVE recordupdated 2026-11-19

Distribution builds and vendor advisories

Packages from OSV.dev and products from the vendors' CSAF advisories.

version not verifiedVersions not verified: read as OSV.dev and the vendors publish them, not checked against each vendor's release list.

Packages

  • linuxDebian:14all versions · fixed in 7.3.3-1
  • linuxDebian:13all versions · fixed in 6.12.117-1
  • linux-6.12Debian:12all versions · fixed in 6.12.117-1~deb12u1
  • linux-ltsAlpaquita:streamfrom 6.18.58-r0 · fixed in 6.18.62-r0
  • linux-ltsAlpaquita:25from 6.12.115-r0 · no fixed version listed

Vendor advisories

  • CVE-2026-99463 Red Hat Product Security16 known affected

    kernel: kernel: Privilege escalation via use-after-free in transparent huge page reclaim

    kernel.src as a component of Red Hat Enterprise Linux 10 · kernel as a component of Red Hat Enterprise Linux 6 · kernel-abi-whitelists as a component of Red Hat Enterprise Linux 6 · +13 more

Full record on NVD

Fix

  1. Update Linux kernel to 7.3.2 · 6.18.61 · 6.12.117
  2. No fix available, says the vendor Red Hat Product Security CVE-2026-99463 · 16 productsAffected

Detection

Nuclei template
none so far
ProjectDiscovery
Exploit-DB
no entry
Exploit-DB
Proof of concept
Metasploit modules
none listed
Metasploit Framework

Sources

Identifiers

Technical details

CVSS v3 vector
CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
Published
2026-11-19
Last modified
2026-11-22
NVD status
Received
GitHub advisory
GHSA-rvfh-j369-c8qj

Frequently asked questions

What is CVE-2026-99463?

In the Linux kernel, the following vulnerability has been resolved: mm/rmap: don't discard lazyfree THPs at PMD level __discard_anon_folio_pmd_locked() clears a lazyfree THP's PMD before it determines whether the folio can be discarded, and restores it if the folio is dirty or has extra references. It is done under only the PMD lock and the anon_vma read lock inside try_to_unmap_one() and no mmap lock is held. The temporarily none PMD causes UAF. A parallel munmap() or mremap() is supposed to serialize with it on the PMD lock, but zap_pmd_range() (called by munmap()) and move_page_tables() (called by mremap()) do not take the lock when they see the temporarily none PMD and skip the PMD. There is no more page table walk after the none PMD is restored: - munmap(): free_pgtables() waits in unlink_anon_vmas() for the anon_vma lock, then free_pmd_range() drops the restored PMD as bad without removing the rmap, so the folio keeps a stale mapcount. - whole-VMA MREMAP_DONTUNMAP: dontunmap_complete() unlinks the source VMA from its anon_vma while the restored PMD still maps the folio. - plain mremap(): the THP is not moved but unmapped, so the THP's data is lost. In the first two cases, the folio stays mapped, but no VMA that maps it is linked to its anon_vma. When the anon_vma is freed, it is still reachable when one rmap walks the folio. Racing lazyfree reclaim against MREMAP_DONTUNMAP as an unprivileged user reproduces a KASAN use-after-free in folio_lock_anon_vma_read() as reported in the Closes link. One solution is to use an invalidated PMD instead of a cleared one until the discard is committed, but it needs arch fixes first. 1. GUP-fast still follows an invalidated PMD on riscv, LoongArch and s390. 2. sparc64 miscounts thp_pte_count when an invalidated PMD is set again or cleared. To keep the fix simple for backport, stop discarding lazyfree THPs at PMD level instead. try_to_unmap_one() splits the PMD and discards the folio at PTE level. unmap_huge_pmd_locked() and __discard_anon_folio_pmd_locked() are left in place, unused, to minimize this fix for backport and the invalidated PMD approach is planned as a follow-up series to use them again.

Is CVE-2026-99463 actively exploited?

Active exploitation of CVE-2026-99463 has not been confirmed. Its EPSS score was 0.2% on 2026-11-21, the estimated probability of exploitation in the next 30 days.

What is the CVSS score for CVE-2026-99463?

CVE-2026-99463 has a CVSS v3 base score of 7.8 (High severity), with vector string CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H.

Running Linux kernel? Hear about its next vulnerability

Surveillance reads the versions in your stack — a domain, a lockfile or a list — checks each CVE against the ranges its sources publish, and alerts you when one you run is exploited or enters CISA KEV. A version it cannot read stays « version not verified ».

No credit card required · No monthly cap on website checks · 250 API requests/month

Sources for this record