CVE-2026-47501

A flaw was found in the NVIDIA GPU Display Driver. A local user can exploit an out-of-bounds write vulnerability in the kernel mode layer by supplying mismatched memory buffers during event buffer setup. Successful exploitation could lead to privilege escalation, arbitrary code execution, denial of service (DoS), information disclosure, or data tampering.

This vulnerability is rated as Important because an authenticated local user can achieve arbitrary kernel-level code execution and full administrative compromise on systems running the driver, despite requiring local access rather than being remotely exploitable. Red Hat Enterprise Linux deployments are only exposed if the proprietary NVIDIA GPU driver packages are explicitly installed and active, as default installations rely on open-source graphics drivers. Exploitation requires the attacker to hold an active local account capable of interacting with the kernel-mode driver interfaces, but once triggered, the flaw allows total compromise of host confidentiality, integrity, and system availability.

The CVSS score(s) listed for this vulnerability do not reflect the associated product's status, and are included for informational purposes to better understand the severity of this vulnerability.

To prevent exposure to this vulnerability on systems where GPU acceleration or computing workloads are not actively required, prevent the NVIDIA kernel driver module from loading:

1. Create a configuration file in /etc/modprobe.d/ to blacklist and prevent the module from loading:

echo "blacklist nvidia" > /etc/modprobe.d/disable-nvidia.conf
echo "install nvidia /bin/false" >> /etc/modprobe.d/disable-nvidia.conf

2. If the module is currently loaded, reboot the system to unload it:

reboot

Warning: Disabling the NVIDIA kernel module prevents graphical display servers, CUDA acceleration, and GPU-dependent workloads from utilizing the hardware. A system reboot will interrupt running applications and active sessions.