A flaw in the Linux kernel’s early boot routine causes the debug PREEMPT_RT build to call a memory‑allocation helper while in hard interrupt context. The helper can acquire a lock that the pool allocator also needs, leading to a deadlock that stalls the kernel. The impact is a loss of availability as the system can freeze during boot or become unresponsive, potentially affecting any process that requires debug or real‑time capabilities.

This issue affects Linux kernel builds that include the debug PREEMPT_RT configuration, particularly on ARM64 architectures. The specific affected versions are those before the patch referenced by commits 0d046ae106255cba5eb83b23f78ee93f3620247d, 44b8b03a9fb5c575548fc72c674653d6baba142a, and 7bc71bdb1c1526c7f02a6adab324394ff1327b0a. Versions containing this fix are effectively all kernels updated after those commits.

Based on the description, it is inferred that the flaw does not provide a direct path for code execution or privilege escalation; the primary impact is the potential for the kernel to become locked during early boot when interrupts are enabled but the scheduler is not yet running. The CVSS score is not listed and no EPSS data is available, indicating limited publicly measured risk data. The vulnerability has not been listed in the CISA KEV catalog. The likely attack vector would require an attacker with physical or firmware control to influence the boot sequence or trigger the interrupt that accesses a region holding the conflicting lock; no remote exploitation path is disclosed. Consequently, the exploitation probability is considered low in the absence of evidence of public exploits, but any successful occurrence would cause a severe unavailability issue during system startup.