In the Linux kernel a logic error in is_bpf_migration_disabled() may incorrectly report that a task is not migration‑disabled when CONFIG_PREEMPT_RCU is not enabled. Because the BPF prolog only calls migrate_disable() under this configuration, the function may allow a truly migration‑disabled task to be scheduled on a remote CPU. That dispatch can trigger an scx_error inside task_can_run_on_remote_rq(), which might result in a kernel panic. The flaw is classified as CWE‑372.

Any Linux kernel build that includes BPF support and does not contain the patch commit 8e4f0b1ebcf2 is affected. This includes all architectures that ship the unpatched kernel, regardless of the CONFIG_PREEMPT_RCU setting, as long as BPF programs can be loaded.

The likely attack vector involves the ability to load or run BPF programs, which normally requires elevated privileges such as CAP_SYS_ADMIN or CAP_BPF; the CVE description does not explicitly enumerate the required privileges. No public exploit is known. The EPSS score of <1% indicates a very low probability of exploitation, and the vulnerability is not listed in CISA's KEV catalog. The CVSS score of 5.5 classifies the flaw as medium severity, reflecting a moderate risk of causing a kernel panic through placement of a migration‑disabled task on a remote CPU.