CVE-2026-47591

A flaw was found in the NVIDIA GPU display driver. An incorrect authorization issue in the kernel mode layer allows a local unprivileged user to bypass read-only memory protections and obtain write access to restricted memory. A local attacker could exploit this flaw to escalate privileges, execute arbitrary code, or cause a denial of service (DoS).

Red Hat has evaluated this vulnerability as having an Important severity rating because an authenticated local user can bypass memory access restrictions in the kernel layer, potentially enabling privilege escalation to root. The flaw is contained to systems with physical or virtualized NVIDIA GPU hardware utilizing the proprietary display driver. Systems without NVIDIA hardware, environments using standard open-source drivers, or configurations where untrusted users cannot execute local code are not affected.

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.

Prevent the NVIDIA kernel driver module from being loaded if GPU acceleration is not required:

1. Add a configuration file under /etc/modprobe.d/ to prevent loading the module:

echo 'install nvidia /bin/true' > /etc/modprobe.d/disable-nvidia.conf

2. If the module is currently loaded in the kernel, attempt to unload it:

rmmod nvidia

Warning: Unloading or disabling this module immediately interrupts GPU-accelerated computing workloads and graphical sessions relying on the NVIDIA driver. If the driver is actively in use by running processes, a system reboot is required for the configuration to take effect.