CVE-2023-52671

What it actually does

Race condition in AMD display driver where disabling an OPTC fails to fully disconnect its assigned OPPs, allowing subsequent reassignment to another OPTC and resulting in display pipeline hang or underflow.

Why it matters to the business

Systems using affected AMD GPUs may experience intermittent display freezes, black screens, or full system hangs during mode changes or multi-monitor transitions, leading to user disruption and potential loss of work on Linux desktops and servers.

How an attack unfolds

An attacker with local access configures display outputs to trigger repeated enable/disable cycles on OPTC blocks; the missing unassignment leaves OPPs in an inconsistent state, causing the display hardware to hang or underflow when the kernel attempts to reclaim resources for a new OPTC configuration.

How to fix it

Update the Linux kernel to a version containing commit ae62f1dde66a6f0eee98defc4c7a346bd5acd239 or later; rebuild and reboot with the patched kernel. Where immediate patching is not possible, avoid complex multi-monitor or dynamic display reconfiguration on affected AMD hardware.

How to confirm the fix worked

Confirm the running kernel contains commit ae62f1dde66a6f0eee98defc4c7a346bd5acd239 via git log or package version; test display transitions under load to verify absence of hangs; monitor dmesg for underflow or OPP assignment warnings that were present before the fix.

Typical attack path

1. Obtain local user access on a Linux system with vulnerable AMD display hardware
2. Trigger graphics mode transitions or multi-monitor reconfiguration that exercises ODM4:1 paths
3. Repeatedly enable and disable OPTC instances to race OPP reclamation
4. Observe or force kernel hang/underflow in the display driver

How to detect it

Review dmesg and kernel logs around display mode changes for messages indicating OPTC disablement followed by OPP underflow or assignment conflicts; correlate with user-reported freezes during monitor hotplug or resolution changes.

What your security team should do

• Prioritize kernel updates on all AMD GPU Linux systems within standard patch windows
• Add kernel version checks to asset inventory for affected stable trees
• Test display reconfiguration workloads in staging before broad deployment
• Review any downstream distributions for backported fixes matching the referenced commit