CVE-2026-47514

A flaw was found in nvidia-driver. A local user can trigger an issue in the driver's kernel mode layer that exposes sensitive kernel stack contents, including return addresses and memory pointers. A successful exploit could allow an attacker to achieve arbitrary code execution, escalate privileges, cause a denial of service (DoS), or tamper with system data.

This vulnerability is rated as Important severity for Red Hat Enterprise Linux because an authenticated local user can abuse the proprietary driver to expose sensitive kernel stack data and bypass system security boundaries, potentially leading to full privilege escalation. NVIDIA driver kernel modules are not active by default and are deployed primarily on systems configured for GPU-accelerated computing or dedicated graphical workloads. Systems without NVIDIA hardware or where the driver module is not loaded remain unaffected by this flaw.

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 mitigate this vulnerability on systems where GPU hardware acceleration or compute capabilities are not required, prevent the driver module from loading:

1. Create a modprobe configuration to blacklist and disable the module: echo "blacklist nvidia" > /etc/modprobe.d/nvidia-blacklist.conf echo "install nvidia /bin/false" >> /etc/modprobe.d/nvidia-blacklist.conf

2. Unload the driver if it is currently active: modprobe -r nvidia

Caveats: Disabling this module prevents proprietary NVIDIA GPU hardware acceleration, graphical desktop sessions driven by NVIDIA hardware, and CUDA compute applications from functioning. A system reboot or display manager restart may be required to completely unload active kernel modules.