takedown_cpu() has a comment about "all preempt/rcu users must observe
!cpu_active()" which is kind of meaningless in this function. This
comment was originally introduced by commit
6acce3ef8452 ("sched: Remove
get_online_cpus() usage") when _cpu_down() was setting cpu_active_mask
and synchronize_rcu()/synchronize_sched() were added after that.
Later commit
40190a78f85f ("sched/hotplug: Convert cpu_[in]active
notifiers to state machine") added a new CPUHP_AP_ACTIVE hotplug
state to set/clear cpu_active_mask. The following commit
b2454caa8977
("sched/hotplug: Move sync_rcu to be with set_cpu_active(false)")
move the synchronize_*() calls to sched_cpu_deactivate() associated
with the new hotplug state, but left the comment behind.
Remove this comment as it is no longer relevant in takedown_cpu().
Signed-off-by: Waiman Long <longman@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/all/20250729191232.664931-1-longman@redhat.com
*/
irq_lock_sparse();
- /*
- * So now all preempt/rcu users must observe !cpu_active().
- */
err = stop_machine_cpuslocked(take_cpu_down, NULL, cpumask_of(cpu));
if (err) {
/* CPU refused to die */