cgroup/cpuset: Fix a race between cpuset_attach() and cpu hotplug

As previously discussed(https://lkml.org/lkml/2022/1/20/51),
cpuset_attach() is affected with similar cpu hotplug race,
as follow scenario:

     cpuset_attach()				cpu hotplug
    ---------------------------            ----------------------
    down_write(cpuset_rwsem)
    guarantee_online_cpus() // (load cpus_attach)
					sched_cpu_deactivate
					  set_cpu_active()
					  // will change cpu_active_mask
    set_cpus_allowed_ptr(cpus_attach)
      __set_cpus_allowed_ptr_locked()
       // (if the intersection of cpus_attach and
         cpu_active_mask is empty, will return -EINVAL)
    up_write(cpuset_rwsem)

To avoid races such as described above, protect cpuset_attach() call
with cpu_hotplug_lock.

Fixes: be367d0 ("cgroups: let ss->can_attach and ss->attach do whole threadgroups at a time")
Cc: [email protected] # v2.6.32+
Reported-by: Zhao Gongyi <[email protected]>
Signed-off-by: Zhang Qiao <[email protected]>
Acked-by: Waiman Long <[email protected]>
Reviewed-by: Michal Koutný <[email protected]>
Signed-off-by: Tejun Heo <[email protected]>

kernel/cgroup/cpuset.c

@@ -2289,6 +2289,7 @@ static void cpuset_attach(struct cgroup_taskset *tset)
 	cgroup_taskset_first(tset, &css);
 	cs = css_cs(css);
 
+	cpus_read_lock();
 	percpu_down_write(&cpuset_rwsem);
 
 	guarantee_online_mems(cs, &cpuset_attach_nodemask_to);
@@ -2342,6 +2343,7 @@ static void cpuset_attach(struct cgroup_taskset *tset)
 		wake_up(&cpuset_attach_wq);
 
 	percpu_up_write(&cpuset_rwsem);
+	cpus_read_unlock();
 }
 
 /* The various types of files and directories in a cpuset file system */