CVE-2024-56697

What it actually does

Null pointer dereference and potential integer multiplication overflow in amdgpu_discovery_get_nps_info during mem_ranges allocation, allowing kernel memory corruption on allocation failure paths.

Why it matters to the business

Affected older Ubuntu Linux kernels on systems with AMD GPUs can experience kernel instability or crashes, disrupting workloads on legacy deployments with high downstream package prevalence.

How an attack unfolds

Local attackers invoke AMD GPU discovery code paths that reach the vulnerable allocation; failure to check for null or overflow lets the function proceed and corrupt kernel state or panic, consistent with patterns seen in Linux kernel out-of-bounds and uninitialized-resource campaigns by espionage and financially motivated actors.

How to fix it

Upgrade to the listed fixed package versions: linux 3.13.0-98.145, linux-aws 4.4.0-1002.2, linux-lts-xenial 4.4.0-42.62~14.04.1, linux 4.4.0-42.62, linux-aws 4.4.0-1001.10, linux-gke 4.4.0-1003.3, linux-hwe 4.8.0-36.36~16.04.1, linux-raspi2 4.4.0-1027.33, linux-snapdragon 4.4.0-1030.33, or linux-lts-vivid 3.19.0-71.79~14.04.1 depending on the installed package.

How to confirm the fix worked

Confirm the running kernel matches one of the fixed versions above and inspect amdgpu_discovery.c for kvcalloc usage followed by null check before output parameter assignment.

Typical attack path

1. Gain local execution on a system running an affected Ubuntu kernel package
2. Trigger amdgpu device initialization or NPS info query
3. Force allocation failure or multiplication overflow in mem_ranges
4. Dereference null or use corrupted buffer leading to kernel panic or memory corruption

How to detect it

Review kernel logs for oops/panic entries mentioning amdgpu_discovery_get_nps_info or null pointer dereferences during GPU initialization; correlate with process activity invoking DRM or amdgpu modules.

What your security team should do

• Apply the exact fixed kernel packages immediately on all affected Ubuntu 14.04/16.04 systems
• Prioritize systems with AMD GPUs given the targeted driver function
• Remove or isolate unsupported legacy kernels after patching
• Monitor EPSS and threat actor activity for any rise in Linux kernel memory corruption exploits