CVE-2025-58407
Official description Straight from the sourceThe vendor's or NVD's own wording, published unedited. Authoritative, but often terse — it says what broke, rarely what to do.
NVD · uneditedKernel or driver software installed on a Guest VM may post improper commands to the GPU Firmware to exploit a TOCTOU race condition and trigger a read and/or write of data outside the allotted memory escaping the virtual machine.
Technical summary Written by usOur analysis, written from the advisory, the CVSS vector and the affected-version data. It adds context the advisory leaves out, and never invents facts that are not in the source.
dbcve analysis · moderate confidenceA TOCTOU race condition in GPU Firmware allows malicious kernel or driver code running inside a Guest VM to send improper commands that bypass memory isolation boundaries, enabling read or write operations outside the VM's allocated memory space and achieving VM escape.
Verify against the referenced sources before acting — the references below are authoritative for this CVE, this summary is not.
Affected products & versions What the vendor confirmedThe version ranges the vendor confirmed as vulnerable. If your version sits inside a range here, treat yourself as exposed until you have upgraded.
NVD · CPE data= 25.2CVSS breakdown How the score is builtThe industry scoring standard. It rates how the flaw is reached, what it takes to exploit, and what an attacker gains — the score is derived from those, not the other way round.
From the vector- Attack vector
- Network
- Complexity
- High
- Privileges
- None
- User interaction
- None
- Scope
- Unchanged
- Confidentiality
- High
- Integrity
- High
- Availability
- None
CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:N
Am I affected? How to checkSteps we derive from the advisory and the affected-version data, so you can decide whether this CVE reaches your setup. They are a guide, not a scan — your own configuration is the authority.
dbcve checksWork through these to decide whether this CVE applies to you.
-
Identify Imagination Technologies GPU driver presenceCheck system for installed Imagination Technologies GPU driver files or modules. On Linux, look for files matching 'powervr', 'imagination', or 'pvrsrv' in /lib/modules or check dmesg for Imagination GPU kernel messages.Affected if No Imagination Technologies GPU driver is found - the system is not affected.
-
Determine Imaginationtech Ddk versionCheck the installed Ddk version. On Linux systems with Imagination GPU drivers, examine /sys/module/pvrsrvkm/version, or check /usr/lib/libIMGdxvfn.so version, or run 'modinfo pvrsrvkm' to retrieve version information.Affected if Version is exactly 25.2 - the system is affected by this CVE. Versions other than 25.2 are not in the affected range.
-
Verify GPU passthrough or virtualization configurationCheck if the host or VM environment uses GPU passthrough (VT-d/AMD-Vi), mediated device passthrough (vGPU), or direct hardware assignment of the Imagination GPU to a virtual machine. On the hypervisor, list assigned PCI devices or check VM configuration for GPU hardware assignment.Affected if GPU passthrough to a Guest VM is NOT configured - the attack surface does not exist, so the vulnerability cannot be exploited even with version 25.2.
-
Confirm firmware version in useQuery the GPU firmware version through the driver interface if accessible. On systems with Imagination GPU tools, use any available firmware querying utilities or check dmesg output for firmware version strings during driver load.Affected if Firmware version information is unavailable to query - detection relies on driver version confirmation.
The system is affected only if Imaginationtech Ddk version 25.2 is installed AND GPU passthrough to a Guest VM is configured, allowing the malicious VM code to directly interact with the GPU hardware.
Generated from the published advisory. Verify against your own configuration.
Remediation Closing itWhat it takes to close this. Where a vendor fix exists we point at it; where none exists we say so plainly, and can build one. Effort estimates are scoped from the advisory, not from your codebase.
From vendor dataApply GPU firmware updates from the hardware vendor; review and restrict GPU passthrough configurations if firmware updates are unavailable; implement hypervisor-level monitoring for anomalous GPU command patterns.
- Consultation8.0 h
- Implementation16.0 h
- Testing12.0 h
- Review / QA8.0 h
An estimate, not a bill — we confirm scope with you before any work starts. Need it this week? Rush from $12,352.
Scan for this in your stack
Free · runs locallyCheck whether your project pulls in CVE-2025-58407 — or any other known-vulnerable package — straight from your lock files. Free and open source; it runs locally and uploads nothing.
References Go to the primary sourcePrimary sources — vendor advisories, patches and trackers. Where our summary and a reference disagree, the reference wins.
Primary sourcesPractitioner notes
ContributedPeer-ranked notes from engineers who’ve handled CVE-2025-58407 in production — separate from our analysis above.
The advisory tells you what broke. It rarely tells you what actually worked. If you’ve dealt with this one, that detail is what the next engineer is searching for.
- The version that genuinely resolved it — not the one the vendor claimed
- A config change or rule that shut the vector down
- A gotcha in the upgrade path that cost you an afternoon
No notes yet
Be the first to add a field note for this CVE — a mitigation you’ve verified, a version caveat, or a link to a working fix. Sign in above to contribute.
A place for practitioners to share what actually worked: a mitigation you’ve tested, a configuration change, a version- or environment-specific caveat, or a link to a verified patch. The most useful notes rise to the top as peers upvote them, so the signal stays high.
- Verified mitigations, workarounds, and config changes
- Version or environment caveats, and links to real fixes
- No weaponised exploit code, or anything meant to cause harm
- No spam, self-promotion, credentials, or personal data