CVE-2022-49462 is a vulnerability identified in the Linux kernel, specifically within the Direct Rendering Manager (DRM) subsystem for Qualcomm Adreno GPUs (a6xx series). The issue arises from a reference count leak in the function a6xx_gpu_init. The root cause is improper management of device tree node references: the function of_parse_phandle() returns a node pointer with an incremented reference count, which must be decremented using of_node_put() when the node is no longer needed. In the affected code path, a6xx_gmu_init() passes this node to of_find_device_by_node() and of_dma_configure(). While of_find_device_by_node() takes its own reference, of_dma_configure() does not require the node after use, but the missing call to of_node_put() leads to a reference count leak. Over time, this leak can cause resource exhaustion within the kernel's device tree node management, potentially leading to degraded system performance or instability. The vulnerability affects Linux kernel versions identified by the given commit hashes (all identical in the data), and no known exploits have been reported in the wild. The issue is primarily a resource management bug rather than a direct code execution or privilege escalation vulnerability. No CVSS score has been assigned yet, and no patches or exploit indicators are currently linked.

For European organizations relying on Linux-based systems with Qualcomm Adreno a6xx GPUs—commonly found in embedded devices, mobile platforms, or specialized hardware—this vulnerability could lead to kernel resource leaks that degrade system stability or cause crashes. While it does not directly enable remote code execution or privilege escalation, prolonged exploitation or heavy usage scenarios could result in denial of service conditions due to resource exhaustion. This may affect critical infrastructure or industrial control systems using embedded Linux with these GPUs, causing operational disruptions. The impact on confidentiality and integrity is minimal, but availability could be impaired. Given the widespread use of Linux in servers, desktops, and embedded devices across Europe, organizations with hardware using affected GPU drivers should be aware. However, typical enterprise servers and desktops without this specific GPU hardware are unlikely to be affected. The lack of known exploits and the nature of the bug reduce immediate risk, but unpatched systems could face stability issues over time.

Organizations should identify Linux systems running kernels with the affected a6xx DRM driver, particularly those using Qualcomm Adreno a6xx GPUs. Applying the official Linux kernel patches that add the missing of_node_put() calls to fix the reference count leak is the primary mitigation. If immediate patching is not feasible, monitoring system logs for kernel warnings or errors related to device tree node management and GPU initialization may help detect issues early. Limiting workload intensity on affected GPUs can reduce the risk of resource exhaustion. For embedded or specialized devices, coordinate with hardware vendors or Linux distribution maintainers to obtain updated kernel versions. Additionally, implementing kernel hardening and resource monitoring tools can help detect abnormal resource usage patterns. Regularly updating Linux kernels and drivers to the latest stable releases is recommended to incorporate this and other fixes.