A NULL pointer dereference occurs in the Linux kernel’s amdgpu driver during device cleanup when GPU initialization fails and certain hardware blocks lack a valid version pointer. The driver accesses adev->ip_blocks[i].version without checking for NULL, causing a fault that can crash the system. The impact is a local denial of service, as an attacker can provoke the crash by forcing the driver to clean up a failed GPU initialization. No additional privilege escalation is documented in the advisory.

The flaw affects the amdgpu driver within the Linux kernel on all machines using AMD GPUs that run kernel versions prior to the patch commits referenced in the advisory. The specific commit identifiers that introduced the fix are available in the provided references, indicating that any kernel lacking those commits is potentially vulnerable.

No CVSS or EPSS score is published, and the vulnerability is not listed in CISA KEV, suggesting limited or no public exploitation. The likely attack vector is local or privileged users who can influence GPU initialization, leading to a system crash. Exploitation requires the ability to trigger a GPU teardown; the crash is the primary impact and carries high operational risk.