Skip to main content

NVIDIA GPU Driver CVE-2026-47599

| EUVDEUVD-2026-90139 HIGH
Improper Preservation of Permissions (CWE-281)
2026-09-30 nvidia GHSA-v4fm-m6gv-jvp3
7.8
CVSS 3.1 · Vendor: nvidia
View

Attack ChainAIDerived

Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain

  1. Access
    Gain unprivileged local account on Linux host
  2. Delivery
    Invoke DMA mapping via NVIDIA kernel module
  3. Exploit
    Improper preservation of access permissions
  4. Execution
    Write/access protected kernel memory
  5. Impact
    Escalate privileges or execute code in kernel

Vulnerability Assessment

Weakness
CWE
OWASP Top 10:2025
A01:2025 - Broken Access Control
ASVS 5.0
V8 - Authorization Permission preservation
Exploitation Conditions

Requires an unprivileged local user account (PR:L, AV:L) on a Linux system running the NVIDIA open-source GPU kernel module; exploitation works against default configurations of that module but is limited to the open-source kernel module variant (not the proprietary module) and demands a triggerable DMA mapping code path. No user interaction is needed (UI:N). Limiting factors: attacker must already have local execution on the host and the affected open-source kernel module must be installed and loaded, making this most relevant to multi-user, VDI, cloud-GPU, and HPC deployments rather than remote attack surface.

Impact

All exploitation signals point to a local privilege-escalation risk rather than a remote-critical one. The CVSS 3.1 vector (AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H, 7.8 HIGH) requires local access and low privileges - an unprivileged local user account already on the Linux host. There is no EPSS score, no CISA KEV listing, and no public POC provided, so this is 'no public exploit identified at time of analysis.' No SSVC decision data is supplied. The 'RCE' tag overstates reach: with AV:L this is local code execution within the kernel context, not remote code execution - treat the tag as loose. The described impact (code execution, privilege escalation, DoS, information disclosure, data tampering) is broad but gated behind local access; NVIDIA self-reported and patched via bulletin 5861, consistent with a defense-in-depth kernel fix. Net: a genuine but not emergency issue - meaningful on shared/multi-tenant Linux GPU hosts (VDI, cloud GPU, HPC), lower urgency on single-user workstations.

Exploit ScenarioDerived

An unprivileged local user on a Linux system running the NVIDIA open-source kernel module triggers a DMA mapping operation that improperly preserves memory access permissions, gaining write or access rights the mapping should not grant. By leveraging the retained permissions against kernel-owned or cross-context memory, the attacker escalates to kernel-level code execution or leaks sensitive data. No public exploit code is currently identified, and exploitation requires an existing local account on the target.

This Week Schedule remediation this week. Why: HIGH severity (CVSS 7.0-8.9).
No patch listed yet 0.2% exploitation probability (EPSS)

Get an e-mail when this CVE is exploited, patched or re-scored. Free account.

Alert me

AnalysisAIv1 – Sep 30, 2026

Local privilege escalation in the NVIDIA GPU Display Driver for Linux (open-source kernel module variant) allows any unprivileged local user with an account on the host to abuse a DMA mapping code path where memory access permissions are not properly preserved, potentially yielding code execution, privilege escalation, denial of service, information disclosure, or data tampering. The vulnerability is rated CVSS 7.8 (AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H), meaning exploitation requires only local access, no user interaction, and no elevated privileges to trigger, but it is confined to systems running the open-source kernel module - the proprietary module is not affected. There is no public exploit identified at time of analysis and no CISA KEV listing; risk is concentrated in multi-user, VDI, cloud-GPU, and HPC hosts where untrusted local users share GPU-equipped machines.

RemediationAI

No vendor-released patch or exact fixed driver version is identified in the provided intelligence; the authoritative remediation source is NVIDIA's product security advisory at https://github.com/NVIDIA/product-security/tree/main/2026/5861, which should be consulted for the specific patched driver release and version mapping for GeForce, RTX/Quadro/NVS, and Tesla products - apply that driver update as soon as it is confirmed available. Because the flaw is confined to the open-source kernel module, the most direct compensating control is to avoid or disable that module: install/use the proprietary NVIDIA kernel module instead, or blacklist/unload the nvidia-open kernel modules on hosts where they are not required, accepting the trade-off that the proprietary module is not upstreamed and may lack features or kernel-version support you rely on (and that GPU acceleration is lost if the module is fully unloaded). Where the open-source module must stay, reduce exposure by tightening local access: remove untrusted interactive users from affected GPU hosts, restrict who can open GPU device nodes (/dev/nvidia*) via permissions or device cgroup rules, and isolate multi-tenant GPU workloads in containers or VMs with GPU passthrough so an unprivileged local process cannot reach the vulnerable DMA mapping path - noting that these controls reduce, but do not eliminate, the underlying kernel-level flaw. Confirm IOMMU enforcement is enabled (intel_iommu=on/amd_iommu=on with iommu=pt considerations) as an additional containment layer, understanding that IOMMU settings are a hardening measure, not a substitute for the vendor fix.

Affected VersionsENISA

GeForce All driver versions prior to 580.178.04
Tesla All driver versions prior to 615.71.09
RTX, Quadro, NVS All driver versions prior to 615.71.09
GeForce All driver versions prior to 615.71.09
GeForce All driver versions prior to 610.57.04
RTX, Quadro, NVS All driver versions prior to 595.91.07
RTX, Quadro, NVS All driver versions prior to 610.57.04
Tesla All driver versions prior to 610.57.04
RTX, Quadro, NVS All driver versions prior to 580.178.04
GeForce All driver versions prior to 595.91.07
Tesla All driver versions prior to 580.178.04
Tesla All driver versions prior to 595.91.07

Affected Packages

euvd geforce_all_driver_versions_prior_to no fix
euvd rtx_quadro_nvs_all_driver_versions_prior_to no fix
euvd tesla_all_driver_versions_prior_to no fix

Source: OSV (Open Source Vulnerabilities)

Affects your stack

Affects 3 packages in euvd

Checking this CVE against your own stacks and servers is part of the Basic plan and up.

Technical ContextAI · show details

The flaw maps to CWE-281 (Improper Preservation of Permissions) and lives in NVIDIA's open-source Linux GPU kernel module, which implements DMA mapping for GPU memory in the upstream-style kernel driver stack. During DMA mapping - where the kernel builds scatter-gather/DMA mappings so the device can access host memory - the driver fails to correctly preserve the access permissions associated with those mappings, so a mapping intended to be restricted can end up granting broader read/write reach than the calling context should have. Because the affected functionality is inside the kernel module and reachable through normal driver ioctl/mmap-style interfaces, an unprivileged local user can drive the faulty path (CVSS PR:L, AV:L, UI:N) against default configurations of the open-source module only. Successful abuse breaks the memory-isolation boundary between an unprivileged process and kernel/GPU-accessible memory, which is why the impact vector spans high confidentiality, integrity, and availability loss (C:H/I:H/A:H). The CPE data tags NVIDIA GeForce, RTX/Quadro/NVS, and Tesla product families as affected, i.e., the consumer, workstation, and datacenter GPU lines that ship the Linux display driver with the open-source kernel module option.

Know when Geforce gets its next vulnerability

  • Watch products and get an email when something new is published
  • Check this CVE against your own stack and servers
  • Exploit and PoC references, threat intelligence from 22 sources

No credit card · 7-day full trial

Severity by source

Vendor (nvidia) PRIMARY 7.8 HIGH
AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
vuln.today AI 7.8 HIGH

Local unprivileged user (AV:L/PR:L) triggers a kernel DMA permission flaw with no interaction, yielding full kernel-context confidentiality, integrity, and availability impact.

3.1 AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
4.0 AV:L/AC:L/AT:N/PR:L/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N
SUSE HIGH
Red Hat 7.8 HIGH

Primary rating from Vendor (nvidia).

CVSS VectorVendor: nvidia

Attack Vector
Local
Attack Complexity
Low
Privileges Required
Low
User Interaction
None
Scope
Unchanged
Confidentiality
High
Integrity
High
Availability
High

DescriptionCVE.org

NVIDIA GPU Display Driver for Linux contains a vulnerability in the open-source kernel module where an unprivileged local user could cause improper preservation of memory access permissions during DMA mapping. A successful exploit of this vulnerability might lead to code execution, escalation of privileges, denial of service, information disclosure, and data tampering.

Executive ViewAI

Why It Matters

The NVIDIA GPU Display Driver for Linux in its open-source kernel module variant contains a HIGH-severity local privilege escalation flaw (CVE-2026-47599, CVSS 7.8) that lets any ordinary user with an account on a GPU-equipped host gain elevated control of that machine. The risk is concentrated in shared-compute environments - VDI pools, multi-tenant cloud GPU instances, and HPC clusters - where many users, including contractors or customers, already hold local accounts, and where a single compromised host can expose other tenants' data or workloads. No vendor patch is currently available, so leadership should treat this as an exposure-management problem requiring compensating controls rather than a routine patching task.

Business Impact: High
Compensating Controls

No vendor-released patch identified at time of analysis. Interim mitigations: migrate affected Linux hosts to NVIDIA's proprietary kernel module variant, which the advisory states is not affected; restrict local account and shell access on multi-user, VDI, cloud-GPU, and HPC hosts; enforce least privilege for local users and isolate tenant workloads; monitor GPU driver and DMA-related activity for anomalous behavior; and disable or block loading of the open-source kernel module on the highest-risk shared systems.

Threat Intelligence Overview

EPSS Score
0.2%
Top 96% exploitable
Prevalence
High
NVIDIA GPU drivers across GeForce/RTX/Quadro/Tesla widely deployed on Linux systems
Data Quality
5 sources
Confidence: Medium
EUVD NVD CVE.org EPSS Vulnrichment Vendor GitHub

How this decision was reached

The engine reads each signal below and applies a fixed list of rules from top to bottom. The first rule that matches sets the recommended action – like a statement, every line is shown.

SignalValue
Severity HIGH
CISA KEV No
Public POC No
EPSS 0.2%
Patch available No
Active threat No
Deciding rule
HIGH severity (CVSS 7.0-8.9)
What every action level means
Emergency Immediate action required. Criteria: CRITICAL + in CISA KEV (actively exploited) without patch, OR CRITICAL + EPSS > 30% without patch, OR CRITICAL + active threat signals without patch.
Act Now Prioritize within 24-48h. Criteria: in KEV with patch available, OR CRITICAL + public POC, OR HIGH + EPSS > 10%, OR CRITICAL + active threat, OR any CRITICAL severity.
This Week Schedule remediation this week. Criteria: HIGH severity (CVSS 7.0-8.9). Patch within normal change window.
This Month Plan within 30 days. Criteria: MEDIUM severity (CVSS 4.0-6.9), lower exploitation probability. Include in regular patch cycle.
Monitor Track and reassess. Criteria: LOW severity (CVSS 0.1-3.9), minimal exploitation risk. No immediate action needed.

SSVC AssessmentENISA

SSVC Decision
Scheduled
Exploitation
None
System Exposure· derived
Small
Automatable
No
Human Impact
High
Technical Impact
Total

Priority Score

39
Low Medium High Critical
KEV: 0
EPSS: 0
CVSS: +39
PoC: 0
Weaponized: 0
Threat intel: 0
Gov advisory: 0

Attack Technique

Information disclosure occurs when an application unintentionally exposes sensitive data that aids attackers in reconnaissance or directly compromises security.

Remote Code Execution represents the critical moment when an attacker successfully runs arbitrary code on a target system without physical access.

Denial of Service attacks render applications or systems unavailable by overwhelming resources or triggering failure conditions.

CVE-2026-24187 HIGH
8.8 May 26

Local privilege escalation and code execution in the NVIDIA Display Driver for Linux (GeForce, RTX/Quadro/NVS, Tesla, an

CVE-2026-47592 HIGH
7.8 Sep 30

An out-of-bounds read in the kernel-mode layer of NVIDIA GPU Display Driver on Windows and Linux allows a local, authent

CVE-2026-47573 HIGH
7.8 Sep 30

An out-of-bounds write in NVIDIA NVAPI for Windows allows a locally authenticated attacker holding standard user privile

CVE-2026-47572 HIGH
7.8 Sep 30

Type confusion in the kernel-mode layer of the NVIDIA GPU Display Driver for Linux allows a local low-privileged user (C

CVE-2026-47570 HIGH
7.8 Sep 30

Local privilege escalation in NVIDIA GPU Display Driver for Windows (CUDA driver component) allows an authenticated atta

CVE-2026-47593 HIGH
7.8 Sep 30

Local privilege escalation in NVIDIA Display Driver for Windows allows an authenticated low-privileged user to trigger a

CVE-2026-47594 HIGH
7.8 Sep 30

Local privilege escalation and potential code execution in NVIDIA GPU Display Driver for Windows and Linux arise from a

CVE-2026-47590 HIGH
7.8 Sep 30

Local privilege escalation in NVIDIA Display Driver for Windows and Linux allows an unprivileged local attacker to trigg

CVE-2026-47589 HIGH
7.8 Sep 30

Local privilege escalation and memory corruption in NVIDIA GPU Display Driver on Windows and Linux, affecting GeForce, R

CVE-2026-47588 HIGH
7.8 Sep 30

Local privilege escalation in the NVIDIA GPU Display Driver for Linux allows an authenticated, unprivileged user to trig

CVE-2026-47591 HIGH
7.8 Sep 30

Local privilege escalation in the kernel-mode layer of the NVIDIA GPU Display Driver for Linux allows an unprivileged lo

CVE-2026-47571 HIGH
7.8 Sep 30

Privilege escalation in NVIDIA's GPU Display Driver for Windows lets a local, low-privileged authenticated user bypass a

Vendor StatusVendor

SUSE

Severity: Important
Product Status
SUSE Linux Enterprise Desktop 15 SP7 Affected
SUSE Linux Enterprise Desktop 15 SP7 Affected
SUSE Linux Enterprise Micro 5.3 Affected
SUSE Linux Enterprise Micro 5.4 Affected
SUSE Linux Enterprise Micro 5.5 Affected

This site uses cookies essential for authentication and security. No tracking or analytics cookies are used. Privacy Policy