CVE-2024-43851

What it actually does

The per-CPU variable initialization in xlnx_event_manager_probe passes an unused argument to __request_percpu_irq, causing __kmem_cache_alloc_node to be invoked while in_atomic() under CONFIG_DEBUG_ATOMIC_SLEEP and triggering a sleeping-function-from-invalid-context BUG.

Why it matters to the business

Systems using the Xilinx Versal driver may experience boot-time kernel warnings or crashes when the debug atomic-sleep option is enabled, increasing operational risk and forcing unplanned kernel updates across fleets that depend on the affected Linux versions.

How an attack unfolds

An attacker with local access or the ability to trigger driver probe on an unpatched kernel with CONFIG_DEBUG_ATOMIC_SLEEP enabled can cause the kernel to hit the BUG() path during percpu IRQ allocation; the listed espionage and ransomware campaigns have targeted similar Linux kernel resource-initialization flaws but no public weaponization of this specific issue is recorded.

How to fix it

Update the Linux kernel to commit e8f897f4afef0031fe618a8e94127a0934896aba or later; rebuild and deploy the fixed kernel image to all affected Xilinx Versal and similar platforms; disable CONFIG_DEBUG_ATOMIC_SLEEP only as a temporary workaround.

How to confirm the fix worked

Boot the updated kernel with CONFIG_DEBUG_ATOMIC_SLEEP enabled and confirm absence of the BUG trace containing xlnx_event_manager_probe and __request_percpu_irq; verify the commit hash appears in the running kernel source or package changelog.

Typical attack path

1. Obtain local execution or trigger device probe on target kernel
2. Ensure CONFIG_DEBUG_ATOMIC_SLEEP is active
3. Load or probe the xilinx soc driver to reach xlnx_event_manager_probe
4. Trigger the kmalloc call inside __request_percpu_irq while preempt_count > 0
5. Cause kernel BUG and potential denial of service or information leak

How to detect it

Search kernel logs for the exact BUG trace mentioning xlnx_event_manager_probe, __request_percpu_irq, and __might_sleep; inspect boot logs on systems that enable CONFIG_DEBUG_ATOMIC_SLEEP for the call trace shown in the advisory.

What your security team should do

• Prioritize kernel upgrade to the listed fixed commit on all Xilinx-based systems within the next maintenance window
• Add the fixed commit ID to internal kernel patch review and CI pipelines
• Monitor EPSS and KEV feeds for any future elevation in exploitability
• Test driver probe paths under debug configurations in staging environments