cpuset: Treat cpusets in attaching as populated

Currently, the check for whether a partition is populated does not
account for tasks in the cpuset of attaching. This is a corner case
that can leave a task stuck in a partition with no effective CPUs.

The race condition occurs as follows:

cpu0				cpu1
				//cpuset A  with cpu N
migrate task p to A
cpuset_can_attach
// with effective cpus
// check ok

// cpuset_mutex is not held	// clear cpuset.cpus.exclusive
				// making effective cpus empty
				update_exclusive_cpumask
				// tasks_nocpu_error check ok
				// empty effective cpus, partition valid
cpuset_attach
...
// task p stays in A, with non-effective cpus.

To fix this issue, this patch introduces cs_is_populated, which considers
tasks in the attaching cpuset. This new helper is used in validate_change
and partition_is_populated.

Fixes: e2d5990 ("cgroup/cpuset: Allow no-task partition to have empty cpuset.cpus.effective")
Signed-off-by: Chen Ridong <[email protected]>
Reviewed-by: Waiman Long <[email protected]>
Signed-off-by: Tejun Heo <[email protected]>

Author: Chen Ridong

kernel/cgroup/cpuset.c

@@ -356,6 +356,15 @@ static inline bool is_in_v2_mode(void)
 	      (cpuset_cgrp_subsys.root->flags & CGRP_ROOT_CPUSET_V2_MODE);
 }
 
+static inline bool cpuset_is_populated(struct cpuset *cs)
+{
+	lockdep_assert_held(&cpuset_mutex);
+
+	/* Cpusets in the process of attaching should be considered as populated */
+	return cgroup_is_populated(cs->css.cgroup) ||
+		cs->attach_in_progress;
+}
+
 /**
  * partition_is_populated - check if partition has tasks
  * @cs: partition root to be checked
@@ -373,19 +382,29 @@ static inline bool is_in_v2_mode(void)
 static inline bool partition_is_populated(struct cpuset *cs,
 					  struct cpuset *excluded_child)
 {
-	struct cgroup_subsys_state *css;
-	struct cpuset *child;
+	struct cpuset *cp;
+	struct cgroup_subsys_state *pos_css;
 
-	if (cs->css.cgroup->nr_populated_csets)
+	/*
+	 * We cannot call cs_is_populated(cs) directly, as
+	 * nr_populated_domain_children may include populated
+	 * csets from descendants that are partitions.
+	 */
+	if (cs->css.cgroup->nr_populated_csets ||
+	    cs->attach_in_progress)
 		return true;
 
 	rcu_read_lock();
-	cpuset_for_each_child(child, css, cs) {
-		if (child == excluded_child)
+	cpuset_for_each_descendant_pre(cp, pos_css, cs) {
+		if (cp == cs || cp == excluded_child)
 			continue;
-		if (is_partition_valid(child))
+
+		if (is_partition_valid(cp)) {
+			pos_css = css_rightmost_descendant(pos_css);
 			continue;
-		if (cgroup_is_populated(child->css.cgroup)) {
+		}
+
+		if (cpuset_is_populated(cp)) {
 			rcu_read_unlock();
 			return true;
 		}
@@ -670,7 +689,7 @@ static int validate_change(struct cpuset *cur, struct cpuset *trial)
 	 * be changed to have empty cpus_allowed or mems_allowed.
 	 */
 	ret = -ENOSPC;
-	if ((cgroup_is_populated(cur->css.cgroup) || cur->attach_in_progress)) {
+	if (cpuset_is_populated(cur)) {
 		if (!cpumask_empty(cur->cpus_allowed) &&
 		    cpumask_empty(trial->cpus_allowed))
 			goto out;