A division by zero error in the Linux kernel AMDGPU driver occurs when H264 decoding processes frames whose width or height is less than 16 pixels. The driver calculates buffer sizes from these dimensions without validating them, causing an invalid division and a kernel crash. This flaw satisfies the definition of improper input validation and leads directly to a denial of service by crashing the operating system. No exploit code has been published, but the issue can be triggered by crafting a malicious H264 stream that contains framed dimensions below the threshold.

The vulnerability affects all Linux kernel releases that include the AMDGPU driver with the buggy calculation before the patch is applied. Any such system—regardless of distribution—running a kernel that incorporates the affected code path is at risk. Because the affected code is identical across all kernel versions prior to the commit that adds frame‑size validation, the scope is broad across all unpatched AMDGPU-enabled kernels. Specific version numbers are not listed in the CVE record.

The flaw results in an immediate kernel crash, and its CVSS score of 7.8 indicates high severity. The EPSS score is less than 1%, indicating a very low but non‑zero exploitation probability, and the vulnerability is not listed in the CISA KEV catalog, so no widespread exploitation has been documented. The likely attack vector is the delivery of crafted H264 video data to the AMDGPU decoder, which could be achieved remotely through media services or locally by any application that processes such content. While no formal exploit has been published, the kernel crash can be reliably triggered by malformed frames, making the vulnerability a significant risk to availability.