The issue is related to the Linux kernel, where a recent fix for making the takeover of the broadcast timer more reliable retrieves a per CPU pointer in preemptible context. This went unnoticed as compilers hoist the access into the non-preemptible region where the pointer is actually used, triggering a bug when using `smp processor id()` in preemptible code. The caller is `hotplug cpu broadcast tick pull+0x1c/0xc0`. The fix involves moving the per CPU pointer access into the atomic section.