CVE-2021-47454 is a vulnerability identified in the Linux kernel specifically affecting the PowerPC architecture's CPU hotplug functionality. The issue arises in the handling of the idle task's preempt count during CPU offline and online cycles when the kernel is compiled with PREEMPT_COUNT=y. The vulnerability is rooted in the powerpc/smp code where the idle task's preempt count is decremented incorrectly when a CPU is offlined and then brought back online. This leads to a kernel BUG triggered by scheduling operations occurring while in an atomic context, which is not allowed. The root cause is a mismatch between the expected preempt count state in the start_secondary() function and the actual state caused by legacy code that re-enabled preemption on the idle task during CPU offline. Recent commits in the Linux kernel had changed the idle task initialization and CPU hotplug behavior, invalidating the previous assumptions and making the decrement of the preempt count inappropriate. The fix involves removing the vestigial preempt_enable() call from the CPU offline path to prevent the preempt count from being decremented erroneously. This vulnerability was tested on PowerPC platforms including pseries and powernv in QEMU and on pseries running PowerVM. It does not affect other architectures. No known exploits are reported in the wild, and no CVSS score has been assigned yet.

For European organizations running Linux on PowerPC hardware, particularly in specialized environments such as telecommunications, embedded systems, or legacy enterprise servers, this vulnerability could cause system instability or kernel panics during CPU hotplug operations. This may lead to unexpected downtime or service interruptions, impacting availability. Since the issue triggers a kernel BUG due to scheduling in atomic context, it could potentially disrupt critical processes or services relying on CPU hotplug capabilities. However, the vulnerability does not appear to allow privilege escalation, arbitrary code execution, or data leakage, so confidentiality and integrity impacts are limited. The impact is mainly on system reliability and availability. Organizations using PowerPC-based Linux systems in data centers or industrial control systems in Europe should be aware of this risk, especially if they perform CPU hotplugging dynamically for maintenance or load balancing. Given the niche hardware affected, the overall impact on the broader European IT landscape is limited but significant for affected sectors.

European organizations should ensure that their Linux kernel versions are updated to include the patch that removes the erroneous preempt count decrement during CPU offline on PowerPC architectures. Since this vulnerability is architecture-specific, organizations should audit their infrastructure to identify any PowerPC-based Linux systems. For those systems, kernel upgrades to versions including the fix (post commit a7c2bb8279d2 and related changes) are critical. Additionally, organizations should review CPU hotplug usage policies and avoid unnecessary CPU offlining/onlining until patched. Testing kernel updates in staging environments that replicate production PowerPC workloads is recommended to confirm stability. Monitoring kernel logs for BUG messages related to scheduling while atomic can help detect attempts to trigger this issue. Finally, coordinating with hardware vendors and Linux distribution maintainers to obtain timely patches and advisories is essential for maintaining system availability.