* Bypass the panic_cpu check and call __crash_kexec directly.
*/
if (!_crash_kexec_post_notifiers) {
- printk_safe_flush_on_panic();
__crash_kexec(NULL);
/*
*/
atomic_notifier_call_chain(&panic_notifier_list, 0, buf);
- /* Call flush even twice. It tries harder with a single online CPU */
- printk_safe_flush_on_panic();
kmsg_dump(KMSG_DUMP_PANIC);
/*