Look up another CVE · open CVE-2026-89910 in the interactive tracker
CVE-2026-89910
LoongArch: KVM: Fix uninitialized stack variable issue with dmsintc
Published 2026-09-16 (33 days ago) by the Linux kernel CNA. Subsystem arch/loongarch. Debian urgency: unimportant.
What to do, per Debian release
Debian's own status for CVE-2026-89910 in every release this site tracks, and what each one means in practice. Every status below is the one the Debian Security Team publishes for this CVE. source
| Release | Status | Version | What this means |
|---|---|---|---|
| Debian 12bookworm | not affected | 6.1.190-1 | Nothing to do. bookworm never shipped the vulnerable code. |
| Debian 13trixie | not affected | 6.12.113-1 | Nothing to do. trixie never shipped the vulnerable code. |
| Debian 14forky | fixed | 7.2.6-1 | Update the kernel and reboot. Fixed in 7.2.6-1. A running kernel keeps the old code until the machine restarts. |
| Debian unstablesid | fixed | 7.2.6-1 | Update the kernel and reboot. Fixed in 7.2.6-1. A running kernel keeps the old code until the machine restarts. |
The vulnerability
What the bug is
Variable vector[] is declared on stack in function dmsintc_inject_irq() and sometimes it is used without initialized. Here fix this issue.
Where it lives
The fix touches these files. If the subsystem is one you do not use, the practical exposure is lower, though the package is still the vulnerable one.
arch/loongarch/kvm/intc/dmsintc.c
- What this is
- KERNEL VIRTUAL MACHINE FOR LOONGARCH (KVM/LoongArch)MAINTAINERS section matched on arch/loongarch/kvm/ · kvm@vger.kernel.org · Maintained
Show whether Debian builds this, per release
Not established. This site could not find the build rule for the affected file in the kernel trees Debian ships, so it has no answer about whether Debian compiles this code. Read that as unknown, never as absent.
- bookworm, trixie, forky, sid?
- ?
- not established - no Makefile entry for the affected file
Where a release above reads "not established", no build rule for the affected file was found in that release's kernel tree. That is a gap in this lookup, not a finding that the code is absent.
From the configuration Debian publishes for every kernel flavour it builds. It covers only the architectures Debian ships a linux-config package for, and nothing about a kernel you built yourself. Debian's own status above stays authoritative about the source package.MAINTAINERS
Signals
- Published
- 2026-09-16 (33 days ago)source
- CVSS
- 7.3 high
CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:C/C:H/I:L/A:Nsource recompute it - Attack vector
- local (AV:L)source
- EPSS
- 0.15%, higher than 3.8% of all CVEssource
- CISA KEV
- not listedthe KEV catalogue
- Debian urgency
- unimportantsource
- Subsystem
- arch/loongarch
Upstream fixes
The versions the CNA records as introducing and fixing this, and the commit for each. source
| Introduced | Fixed in | Commit |
|---|---|---|
| 7.1 | 7.2.5 | f56060b0e10f |
| 7.1 | 7.3 | 81aa3a58b542 |
Why the CNA scored it this way
Quoted from the kernel CNA's own scoring file, word for word. A highlighted metric carries the most severe value CVSS v3.1 defines for it, which is what pushes the score up. source
- AV:LAttack vector, local
- The uninitialized vector[] is used in dmsintc_inject_irq(), called from kvm_deliver_intr() on KVM_RUN (kvm_pre_enter_guest) and KVM_GET_ONE_REG ESTAT after irqfd/KVM_SIGNAL_MSI or KVM_INTERRUPT queues INT_AVEC. That is a local /dev/kvm and guest MSI path, not a network or physical protocol handler.
- AC:LAttack complexity, low
- A kvm-group VMM can create a MSGINT vCPU, queue INT_AVEC via KVM_INTERRUPT or irqfd MSI, then KVM_RUN or GET_ONE_REG ESTAT; any zero vector_map word leaves the matching stack slot uninitialized and typically nonzero. A guest retriggers the same inject on ordinary virtio/MSI I/O with no race.
- PR:LPrivileges required, low
- kvm_dev_ioctl_create_vm() has no capable() check, so kvm-group (or equivalent) access is enough to create dmsintc, enable CPUCFG MSGINT, and inject AVEC. A LoongArch KVM tenant can also cause irqfd MSI injection and read guest ISR CSRs, which is Low privilege relative to the hypervisor, not init-namespace root.
- UI:NUser interaction, none
- The attacker issues KVM ioctls or guest device I/O themselves; no separate victim action such as mounting a filesystem or opening a file is required.
- S:CScope, changed
- dmsintc_inject_irq() writes host kernel stack contents into hardware guest ISR CSRs (LOONGARCH_CSR_ISR0-3) that the guest can read with csrrd and that the VMM can read after vcpu_put. That discloses hypervisor memory across the KVM guest-to-host boundary.
- C:HConfidentiality impact, high
- Up to four uninitialized unsigned long stack slots are OR'd into guest-visible ISR registers on every AVEC injection. Repeated MSI or KVM_RUN samples resample kernel stack including pointers, matching CNA stack-leak scoring for KASLR defeat.
- I:LIntegrity impact, low
- The same uninitialized words are OR'd into guest ISR0-3, setting extra AVEC pending bits the VMM did not inject. That is limited corruption of guest interrupt-controller state, not an arbitrary host write or control-flow hijack.
- A:NAvailability impact, none
- The uninitialized values are consumed only as ISR bitmasks via gcsr_write; they are not used as pointers and do not oops, panic, hang, or otherwise deny host service.