CVE-2023-54235

What it actually does

Race condition in PCI/DOE workqueue handling where destroy_work_on_stack() is called after signaling completion, allowing the work struct to go out of scope prematurely and triggering debug object warnings.

Why it matters to the business

May cause kernel warnings or potential instability in systems using affected Linux kernels with PCI DOE devices, impacting reliability for organizations relying on stable kernel operations. Update promptly to avoid issues, though not associated with active exploitation.

How an attack unfolds

An attacker with local access could attempt to manipulate PCI device interactions to trigger the DOE state machine race, but the issue primarily manifests as a debug warning rather than a security breach. No known threat actors exploiting this.

How to fix it

Update to fixed kernel versions: 6.1.53 or later, or apply commit d96799ee3b78962c80e4b6653734f488f999ca09 in the stable git repository.

How to confirm the fix worked

Verify kernel version is updated to one of the fixed releases and test PCI DOE functionality without observing debug object splats.

Typical attack path

1. Gain local access to system
2. Interact with PCI DOE capable device to trigger state machine
3. Exploit race in work destruction to cause warning or crash

How to detect it

Check system logs for debug object warnings mentioning doe_statemachine_work or work_struct in PCI contexts; examine kernel crash dumps for related stack traces.

What your security team should do

• Prioritize kernel updates on affected systems
• Monitor for similar race conditions in workqueue usage
• Review logs for debug object warnings related to doe_statemachine_work