- Veröffentlicht
- 20.11.2026 06:50
- Zuletzt geändert
- 20.11.2026
- EPSS (30 Tage)
- unbekannt
- CVSS-Vektor
- CVSS:3.1/AV:L/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H
Beschreibung
Originaltext der Quelle, nicht übersetzt.
In the Linux kernel, the following vulnerability has been resolved:
mm/vma: call ->may_split() before a partial merge
A driver can use the ->may_split() hook to forbid a VMA boundary at
certain addresses.
A device-dax device is a character device that gives raw access to a
range of memory through mmap(). device-dax refuses any address that is not
a multiple of the device's alignment through vm_ops->may_split.
vma_merge_existing_range() can move boundaries, but does not call
vm_ops->may_split.
If a range at the edge of a VMA gets the attributes of the adjacent VMA,
the function moves only that range over to the neighbour: one VMA shrinks
and the other grows. This is a partial merge, and vma_modify() tries it
before any split.
Take a device-dax VMA that lies next to another VMA of the same file.
The user calls mprotect() on the one page of the first VMA that touches
the second VMA, and asks for the protection that the second VMA already
has. The result is a partial merge. It puts the boundary inside a 1 GiB
mapping, which is a single leaf entry at the PUD level.
As part of the merge, the kernel looks for a huge PMD to split at the
new boundary. That code does not expect a PUD leaf:
vma_merge_existing_range()
commit_merge()
vma_adjust_trans_huge()
split_huge_pmd_address()
mm_find_pmd()
treats the PUD leaf as a pointer to a PMD table,
so the "table" is the device memory that the leaf maps
__split_huge_pmd()
reads a PMD that the user wrote to the device memory
folio_put() on the folio that this PMD names
Bad. While the code must at some point be taught to split PUD entries as
well, the issue here is that we move VMA boundaries such that they are
incompatible with the driver.
Fix this by calling ->may_split() on the address that a partial merge
would move the boundary to, and by abandoning the merge if the driver
refuses. vma_modify() then tries the split, which fails with -EINVAL.
Factor out existing split handling into check_split() and reuse it.
Fix generated by an LLM and sent by Chris Mason in private.
[ david: improve description; cleanup the code ]
Status beim Hersteller
Red Hat stuft seine Produktlinien ein: 1 betroffen, 6 nicht betroffen.
Red Hat Enterprise Linux 10
BetroffenFix ausstehend
kernel
Nicht betroffen (6)
Red Hat Enterprise Linux 6Nicht betroffenverwundbarer Code nicht enthaltenkernel, kernel-abi-whitelists, kernel-bootwrapper, kernel-debug, kernel-debug-devel, kernel-devel, kernel-doc, kernel-firmware, kernel-headers, kernel-kdump, kernel-kdump-devel, perf, python-perf
Red Hat Enterprise Linux 7Nicht betroffenverwundbarer Code nicht enthaltenbpftool, kernel, kernel-abi-whitelists, kernel-bootwrapper, kernel-debug, kernel-debug-devel, kernel-devel, kernel-doc, kernel-headers, kernel-kdump, kernel-kdump-devel, kernel-rt, kernel-rt-debug, kernel-rt-debug-devel, kernel-rt-debug-kvm, kernel-rt-devel, kernel-rt-doc, kernel-rt-kvm, kernel-rt-trace, kernel-rt-trace-devel, kernel-rt-trace-kvm, kernel-tools, kernel-tools-libs, kernel-tools-libs-devel, perf und 1 weitere
Red Hat Enterprise Linux 8Nicht betroffenverwundbarer Code nicht enthaltenbpftool, kernel, kernel-abi-stablelists, kernel-abi-whitelists, kernel-core, kernel-cross-headers, kernel-debug, kernel-debug-core, kernel-debug-devel, kernel-debug-modules, kernel-debug-modules-extra, kernel-debug-modules-internal, kernel-devel, kernel-doc, kernel-headers, kernel-ipaclones-internal, kernel-modules, kernel-modules-extra, kernel-modules-internal, kernel-rt, kernel-rt-core, kernel-rt-debug, kernel-rt-debug-core, kernel-rt-debug-devel, kernel-rt-debug-kvm und 21 weitere
Red Hat Enterprise Linux 9Nicht betroffenverwundbarer Code nicht enthaltenbpftool, kernel, kernel-64k, kernel-64k-core, kernel-64k-debug, kernel-64k-debug-core, kernel-64k-debug-devel, kernel-64k-debug-devel-matched, kernel-64k-debug-modules, kernel-64k-debug-modules-core, kernel-64k-debug-modules-extra, kernel-64k-debug-modules-internal, kernel-64k-debug-modules-partner, kernel-64k-devel, kernel-64k-devel-matched, kernel-64k-modules, kernel-64k-modules-core, kernel-64k-modules-extra, kernel-64k-modules-internal, kernel-64k-modules-partner, kernel-abi-stablelists, kernel-core, kernel-cross-headers, kernel-debug, kernel-debug-core und 82 weitere
Red Hat OpenShift Container Platform 4Nicht betroffenverwundbarer Code nicht enthaltenopenshift/ose-rhel-coreos-8, openshift/ose-rhel-coreos-9
Red Hat OpenShift Container Platform 5Nicht betroffenverwundbarer Code nicht enthaltenopenshift/ose-rhel-coreos-10, openshift/ose-rhel-coreos-9
Betroffene Pakete im beobachteten Stack
Debian:14linuxkein Fix bekanntDEBIAN-CVE-2026-99648
Red Hat:enterprise_linux:10kernelkein Fix bekanntCVE-2026-99648
Weitere betroffene Pakete (3)
CVELinux/Linuxkein Fix bekanntCVE-2026-99648
CVELinux/Linuxkein Fix bekanntCVE-2026-99648
LinuxKernel6.15.0 → 6.18.64, 6.19.0 → 7.3.5CVE-2026-99648
Empfohlene Prüfung
- Feststellen, ob und in welcher Version die betroffenen Pakete in Ihren Systemen, Images und Abhängigkeiten vorkommen.
- Wo eine Fix-Version genannt ist: aktualisieren. Wo nicht: Hersteller-Advisory und Referenzen auf Workarounds prüfen.
- Erreichbarkeit beachten: Der Angriffsweg (Lokal) ist aus dem CVSS-Vektor abgeleitet.
Diese Seite sagt nichts darüber, ob Ihre Systeme betroffen sind. Das klärt erst eine Prüfung Ihrer Versionen.
Referenzen
- git.kernel.org/stable/c/f437fcb76d448a396940498aba278ac0ad8d715a
- git.kernel.org/stable/c/456932cdc6eda70b5ad4c628556098846699ab03
- git.kernel.org/stable/c/0193e83368022a308d19ee0293f55a4f5d548e40
- git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git
- github.com/CVEProject/cvelistV5/tree/main/cves/2026/99xxx/CVE-2026-99648.json
- nvd.nist.gov/vuln/detail/CVE-2026-99648
- security-tracker.debian.org/tracker/CVE-2026-99648
- security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-99648.json
- access.redhat.com/security/cve/CVE-2026-99648
- bugzilla.redhat.com/show_bug.cgi
- www.cve.org/CVERecord