Merge branch 'for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup.gitNotice: this object is not reachable from any branch.

Notice: this object is not reachable from any branch.

26 files changed, 245 insertions, 187 deletions

diff --git a/Documentation/admin-guide/cgroup-v1/cgroups.rst b/Documentation/admin-guide/cgroup-v1/cgroups.rst
index 9343148ee99366..a3e2edb3d2745a 100644
--- a/Documentation/admin-guide/cgroup-v1/cgroups.rst
+++ b/Documentation/admin-guide/cgroup-v1/cgroups.rst
@@ -570,7 +570,7 @@ visible to cgroup_for_each_child/descendant_*() iterators. The
 subsystem may choose to fail creation by returning -errno. This
 callback can be used to implement reliable state sharing and
 propagation along the hierarchy. See the comment on
-cgroup_for_each_descendant_pre() for details.
+cgroup_for_each_live_descendant_pre() for details.
 
 ``void css_offline(struct cgroup *cgrp);``
 (cgroup_mutex held by caller)
diff --git a/Documentation/admin-guide/cgroup-v2.rst b/Documentation/admin-guide/cgroup-v2.rst
index 2dcbb040e15991..6f6fe744452ec6 100644
--- a/Documentation/admin-guide/cgroup-v2.rst
+++ b/Documentation/admin-guide/cgroup-v2.rst
@@ -1058,12 +1058,15 @@ cpufreq governor about the minimum desired frequency which should always be
 provided by a CPU, as well as the maximum desired frequency, which should not
 be exceeded by a CPU.
 
-WARNING: cgroup2 doesn't yet support control of realtime processes and
-the cpu controller can only be enabled when all RT processes are in
-the root cgroup.  Be aware that system management software may already
-have placed RT processes into nonroot cgroups during the system boot
-process, and these processes may need to be moved to the root cgroup
-before the cpu controller can be enabled.
+WARNING: cgroup2 doesn't yet support control of realtime processes. For
+a kernel built with the CONFIG_RT_GROUP_SCHED option enabled for group
+scheduling of realtime processes, the cpu controller can only be enabled
+when all RT processes are in the root cgroup.  This limitation does
+not apply if CONFIG_RT_GROUP_SCHED is disabled.  Be aware that system
+management software may already have placed RT processes into nonroot
+cgroups during the system boot process, and these processes may need
+to be moved to the root cgroup before the cpu controller can be enabled
+with a CONFIG_RT_GROUP_SCHED enabled kernel.
 
 
 CPU Interface Files
@@ -2185,11 +2188,25 @@ PID Interface Files
 	Hard limit of number of processes.
 
   pids.current
-	A read-only single value file which exists on all cgroups.
+	A read-only single value file which exists on non-root cgroups.
 
 	The number of processes currently in the cgroup and its
 	descendants.
 
+  pids.peak
+	A read-only single value file which exists on non-root cgroups.
+
+	The maximum value that the number of processes in the cgroup and its
+	descendants has ever reached.
+
+  pids.events
+	A read-only flat-keyed file which exists on non-root cgroups. The
+	following entries are defined. Unless specified otherwise, a value
+	change in this file generates a file modified event.
+
+	  max
+		Number of times fork failed because limit was hit.
+
 Organisational operations are not blocked by cgroup policies, so it is
 possible to have pids.current > pids.max.  This can be done by either
 setting the limit to be smaller than pids.current, or attaching enough
diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h
index 34aaf0e87def8f..2150ca60394ba1 100644
--- a/include/linux/cgroup.h
+++ b/include/linux/cgroup.h
@@ -690,7 +690,7 @@ static inline void cgroup_path_from_kernfs_id(u64 id, char *buf, size_t buflen)
 void cgroup_rstat_updated(struct cgroup *cgrp, int cpu);
 void cgroup_rstat_flush(struct cgroup *cgrp);
 void cgroup_rstat_flush_hold(struct cgroup *cgrp);
-void cgroup_rstat_flush_release(void);
+void cgroup_rstat_flush_release(struct cgroup *cgrp);
 
 /*
  * Basic resource stats.
diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h
index 0ce6ff0d9c9aad..de4cf0ee96f79e 100644
--- a/include/linux/cpuset.h
+++ b/include/linux/cpuset.h
@@ -70,7 +70,6 @@ extern int cpuset_init(void);
 extern void cpuset_init_smp(void);
 extern void cpuset_force_rebuild(void);
 extern void cpuset_update_active_cpus(void);
-extern void cpuset_wait_for_hotplug(void);
 extern void inc_dl_tasks_cs(struct task_struct *task);
 extern void dec_dl_tasks_cs(struct task_struct *task);
 extern void cpuset_lock(void);
@@ -185,8 +184,6 @@ static inline void cpuset_update_active_cpus(void)
 	partition_sched_domains(1, NULL, NULL);
 }
 
-static inline void cpuset_wait_for_hotplug(void) { }
-
 static inline void inc_dl_tasks_cs(struct task_struct *task) { }
 static inline void dec_dl_tasks_cs(struct task_struct *task) { }
 static inline void cpuset_lock(void) { }
diff --git a/include/trace/events/cgroup.h b/include/trace/events/cgroup.h
index dd7d7c9efecdf3..13f3758001355c 100644
--- a/include/trace/events/cgroup.h
+++ b/include/trace/events/cgroup.h
@@ -204,6 +204,54 @@ DEFINE_EVENT(cgroup_event, cgroup_notify_frozen,
 	TP_ARGS(cgrp, path, val)
 );
 
+DECLARE_EVENT_CLASS(cgroup_rstat,
+
+	TP_PROTO(struct cgroup *cgrp, int cpu_in_loop, bool contended),
+
+	TP_ARGS(cgrp, cpu_in_loop, contended),
+
+	TP_STRUCT__entry(
+		__field(	int,		root			)
+		__field(	int,		level			)
+		__field(	u64,		id			)
+		__field(	int,		cpu_in_loop		)
+		__field(	bool,		contended		)
+	),
+
+	TP_fast_assign(
+		__entry->root = cgrp->root->hierarchy_id;
+		__entry->id = cgroup_id(cgrp);
+		__entry->level = cgrp->level;
+		__entry->cpu_in_loop = cpu_in_loop;
+		__entry->contended = contended;
+	),
+
+	TP_printk("root=%d id=%llu level=%d cpu_in_loop=%d lock contended:%d",
+		  __entry->root, __entry->id, __entry->level,
+		  __entry->cpu_in_loop, __entry->contended)
+);
+
+DEFINE_EVENT(cgroup_rstat, cgroup_rstat_lock_contended,
+
+	TP_PROTO(struct cgroup *cgrp, int cpu, bool contended),
+
+	TP_ARGS(cgrp, cpu, contended)
+);
+
+DEFINE_EVENT(cgroup_rstat, cgroup_rstat_locked,
+
+	TP_PROTO(struct cgroup *cgrp, int cpu, bool contended),
+
+	TP_ARGS(cgrp, cpu, contended)
+);
+
+DEFINE_EVENT(cgroup_rstat, cgroup_rstat_unlock,
+
+	TP_PROTO(struct cgroup *cgrp, int cpu, bool contended),
+
+	TP_ARGS(cgrp, cpu, contended)
+);
+
 #endif /* _TRACE_CGROUP_H */
 
 /* This part must be outside protection */
diff --git a/kernel/cgroup/cgroup-v1.c b/kernel/cgroup/cgroup-v1.c
index 520a11cb12f44c..b9dbf6bf2779d6 100644
--- a/kernel/cgroup/cgroup-v1.c
+++ b/kernel/cgroup/cgroup-v1.c
@@ -1335,6 +1335,7 @@ static int __init cgroup_no_v1(char *str)
 				continue;
 
 			cgroup_no_v1_mask |= 1 << i;
+			break;
 		}
 	}
 	return 1;
diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c
index a66c088c851cfb..e32b6972c47840 100644
--- a/kernel/cgroup/cgroup.c
+++ b/kernel/cgroup/cgroup.c
@@ -5368,7 +5368,8 @@ static void css_free_rwork_fn(struct work_struct *work)
 	} else {
 		/* cgroup free path */
 		atomic_dec(&cgrp->root->nr_cgrps);
-		cgroup1_pidlist_destroy_all(cgrp);
+		if (!cgroup_on_dfl(cgrp))
+			cgroup1_pidlist_destroy_all(cgrp);
 		cancel_work_sync(&cgrp->release_agent_work);
 		bpf_cgrp_storage_free(cgrp);
 
diff --git a/kernel/cgroup/cpuset.c b/kernel/cgroup/cpuset.c
index 4237c8748715d4..a10e4bd0c0c16c 100644
--- a/kernel/cgroup/cpuset.c
+++ b/kernel/cgroup/cpuset.c
@@ -202,6 +202,14 @@ struct cpuset {
 };
 
 /*
+ * Legacy hierarchy call to cgroup_transfer_tasks() is handled asynchrously
+ */
+struct cpuset_remove_tasks_struct {
+	struct work_struct work;
+	struct cpuset *cs;
+};
+
+/*
  * Exclusive CPUs distributed out to sub-partitions of top_cpuset
  */
 static cpumask_var_t	subpartitions_cpus;
@@ -360,9 +368,10 @@ static inline void notify_partition_change(struct cpuset *cs, int old_prs)
 }
 
 static struct cpuset top_cpuset = {
-	.flags = ((1 << CS_ONLINE) | (1 << CS_CPU_EXCLUSIVE) |
-		  (1 << CS_MEM_EXCLUSIVE)),
+	.flags = BIT(CS_ONLINE) | BIT(CS_CPU_EXCLUSIVE) |
+		 BIT(CS_MEM_EXCLUSIVE) | BIT(CS_SCHED_LOAD_BALANCE),
 	.partition_root_state = PRS_ROOT,
+	.relax_domain_level = -1,
 	.remote_sibling = LIST_HEAD_INIT(top_cpuset.remote_sibling),
 };
 
@@ -449,12 +458,6 @@ static DEFINE_SPINLOCK(callback_lock);
 
 static struct workqueue_struct *cpuset_migrate_mm_wq;
 
-/*
- * CPU / memory hotplug is handled asynchronously.
- */
-static void cpuset_hotplug_workfn(struct work_struct *work);
-static DECLARE_WORK(cpuset_hotplug_work, cpuset_hotplug_workfn);
-
 static DECLARE_WAIT_QUEUE_HEAD(cpuset_attach_wq);
 
 static inline void check_insane_mems_config(nodemask_t *nodes)
@@ -540,22 +543,10 @@ static void guarantee_online_cpus(struct task_struct *tsk,
 	rcu_read_lock();
 	cs = task_cs(tsk);
 
-	while (!cpumask_intersects(cs->effective_cpus, pmask)) {
+	while (!cpumask_intersects(cs->effective_cpus, pmask))
 		cs = parent_cs(cs);
-		if (unlikely(!cs)) {
-			/*
-			 * The top cpuset doesn't have any online cpu as a
-			 * consequence of a race between cpuset_hotplug_work
-			 * and cpu hotplug notifier.  But we know the top
-			 * cpuset's effective_cpus is on its way to be
-			 * identical to cpu_online_mask.
-			 */
-			goto out_unlock;
-		}
-	}
-	cpumask_and(pmask, pmask, cs->effective_cpus);
 
-out_unlock:
+	cpumask_and(pmask, pmask, cs->effective_cpus);
 	rcu_read_unlock();
 }
 
@@ -1217,7 +1208,7 @@ static void rebuild_sched_domains_locked(void)
 	/*
 	 * If we have raced with CPU hotplug, return early to avoid
 	 * passing doms with offlined cpu to partition_sched_domains().
-	 * Anyways, cpuset_hotplug_workfn() will rebuild sched domains.
+	 * Anyways, cpuset_handle_hotplug() will rebuild sched domains.
 	 *
 	 * With no CPUs in any subpartitions, top_cpuset's effective CPUs
 	 * should be the same as the active CPUs, so checking only top_cpuset
@@ -1260,12 +1251,17 @@ static void rebuild_sched_domains_locked(void)
 }
 #endif /* CONFIG_SMP */
 
-void rebuild_sched_domains(void)
+static void rebuild_sched_domains_cpuslocked(void)
 {
-	cpus_read_lock();
 	mutex_lock(&cpuset_mutex);
 	rebuild_sched_domains_locked();
 	mutex_unlock(&cpuset_mutex);
+}
+
+void rebuild_sched_domains(void)
+{
+	cpus_read_lock();
+	rebuild_sched_domains_cpuslocked();
 	cpus_read_unlock();
 }
 
@@ -2079,14 +2075,11 @@ write_error:
 
 	/*
 	 * For partcmd_update without newmask, it is being called from
-	 * cpuset_hotplug_workfn() where cpus_read_lock() wasn't taken.
-	 * Update the load balance flag and scheduling domain if
-	 * cpus_read_trylock() is successful.
+	 * cpuset_handle_hotplug(). Update the load balance flag and
+	 * scheduling domain accordingly.
 	 */
-	if ((cmd == partcmd_update) && !newmask && cpus_read_trylock()) {
+	if ((cmd == partcmd_update) && !newmask)
 		update_partition_sd_lb(cs, old_prs);
-		cpus_read_unlock();
-	}
 
 	notify_partition_change(cs, old_prs);
 	return 0;
@@ -3599,8 +3592,8 @@ static ssize_t cpuset_write_resmask(struct kernfs_open_file *of,
 	 * proceeding, so that we don't end up keep removing tasks added
 	 * after execution capability is restored.
 	 *
-	 * cpuset_hotplug_work calls back into cgroup core via
-	 * cgroup_transfer_tasks() and waiting for it from a cgroupfs
+	 * cpuset_handle_hotplug may call back into cgroup core asynchronously
+	 * via cgroup_transfer_tasks() and waiting for it from a cgroupfs
 	 * operation like this one can lead to a deadlock through kernfs
 	 * active_ref protection.  Let's break the protection.  Losing the
 	 * protection is okay as we check whether @cs is online after
@@ -3609,7 +3602,6 @@ static ssize_t cpuset_write_resmask(struct kernfs_open_file *of,
 	 */
 	css_get(&cs->css);
 	kernfs_break_active_protection(of->kn);
-	flush_work(&cpuset_hotplug_work);
 
 	cpus_read_lock();
 	mutex_lock(&cpuset_mutex);
@@ -3782,9 +3774,6 @@ static ssize_t sched_partition_write(struct kernfs_open_file *of, char *buf,
 
 	buf = strstrip(buf);
 
-	/*
-	 * Convert "root" to ENABLED, and convert "member" to DISABLED.
-	 */
 	if (!strcmp(buf, "root"))
 		val = PRS_ROOT;
 	else if (!strcmp(buf, "member"))
@@ -4060,11 +4049,6 @@ static int cpuset_css_online(struct cgroup_subsys_state *css)
 		cs->effective_mems = parent->effective_mems;
 		cs->use_parent_ecpus = true;
 		parent->child_ecpus_count++;
-		/*
-		 * Clear CS_SCHED_LOAD_BALANCE if parent is isolated
-		 */
-		if (!is_sched_load_balance(parent))
-			clear_bit(CS_SCHED_LOAD_BALANCE, &cs->flags);
 	}
 
 	/*
@@ -4318,8 +4302,6 @@ int __init cpuset_init(void)
 	nodes_setall(top_cpuset.effective_mems);
 
 	fmeter_init(&top_cpuset.fmeter);
-	set_bit(CS_SCHED_LOAD_BALANCE, &top_cpuset.flags);
-	top_cpuset.relax_domain_level = -1;
 	INIT_LIST_HEAD(&remote_children);
 
 	BUG_ON(!alloc_cpumask_var(&cpus_attach, GFP_KERNEL));
@@ -4354,6 +4336,16 @@ static void remove_tasks_in_empty_cpuset(struct cpuset *cs)
 	}
 }
 
+static void cpuset_migrate_tasks_workfn(struct work_struct *work)
+{
+	struct cpuset_remove_tasks_struct *s;
+
+	s = container_of(work, struct cpuset_remove_tasks_struct, work);
+	remove_tasks_in_empty_cpuset(s->cs);
+	css_put(&s->cs->css);
+	kfree(s);
+}
+
 static void
 hotplug_update_tasks_legacy(struct cpuset *cs,
 			    struct cpumask *new_cpus, nodemask_t *new_mems,
@@ -4383,12 +4375,21 @@ hotplug_update_tasks_legacy(struct cpuset *cs,
 	/*
 	 * Move tasks to the nearest ancestor with execution resources,
 	 * This is full cgroup operation which will also call back into
-	 * cpuset. Should be done outside any lock.
+	 * cpuset. Execute it asynchronously using workqueue.
 	 */
-	if (is_empty) {
-		mutex_unlock(&cpuset_mutex);
-		remove_tasks_in_empty_cpuset(cs);
-		mutex_lock(&cpuset_mutex);
+	if (is_empty && cs->css.cgroup->nr_populated_csets &&
+	    css_tryget_online(&cs->css)) {
+		struct cpuset_remove_tasks_struct *s;
+
+		s = kzalloc(sizeof(*s), GFP_KERNEL);
+		if (WARN_ON_ONCE(!s)) {
+			css_put(&cs->css);
+			return;
+		}
+
+		s->cs = cs;
+		INIT_WORK(&s->work, cpuset_migrate_tasks_workfn);
+		schedule_work(&s->work);
 	}
 }
 
@@ -4421,30 +4422,6 @@ void cpuset_force_rebuild(void)
 	force_rebuild = true;
 }
 
-/*
- * Attempt to acquire a cpus_read_lock while a hotplug operation may be in
- * progress.
- * Return: true if successful, false otherwise
- *
- * To avoid circular lock dependency between cpuset_mutex and cpus_read_lock,
- * cpus_read_trylock() is used here to acquire the lock.
- */
-static bool cpuset_hotplug_cpus_read_trylock(void)
-{
-	int retries = 0;
-
-	while (!cpus_read_trylock()) {
-		/*
-		 * CPU hotplug still in progress. Retry 5 times
-		 * with a 10ms wait before bailing out.
-		 */
-		if (++retries > 5)
-			return false;
-		msleep(10);
-	}
-	return true;
-}
-
 /**
  * cpuset_hotplug_update_tasks - update tasks in a cpuset for hotunplug
  * @cs: cpuset in interest
@@ -4493,13 +4470,11 @@ retry:
 		compute_partition_effective_cpumask(cs, &new_cpus);
 
 	if (remote && cpumask_empty(&new_cpus) &&
-	    partition_is_populated(cs, NULL) &&
-	    cpuset_hotplug_cpus_read_trylock()) {
+	    partition_is_populated(cs, NULL)) {
 		remote_partition_disable(cs, tmp);
 		compute_effective_cpumask(&new_cpus, cs, parent);
 		remote = false;
 		cpuset_force_rebuild();
-		cpus_read_unlock();
 	}
 
 	/*
@@ -4519,18 +4494,8 @@ retry:
 	else if (is_partition_valid(parent) && is_partition_invalid(cs))
 		partcmd = partcmd_update;
 
-	/*
-	 * cpus_read_lock needs to be held before calling
-	 * update_parent_effective_cpumask(). To avoid circular lock
-	 * dependency between cpuset_mutex and cpus_read_lock,
-	 * cpus_read_trylock() is used here to acquire the lock.
-	 */
 	if (partcmd >= 0) {
-		if (!cpuset_hotplug_cpus_read_trylock())
-			goto update_tasks;
-
 		update_parent_effective_cpumask(cs, partcmd, NULL, tmp);
-		cpus_read_unlock();
 		if ((partcmd == partcmd_invalidate) || is_partition_valid(cs)) {
 			compute_partition_effective_cpumask(cs, &new_cpus);
 			cpuset_force_rebuild();
@@ -4558,8 +4523,7 @@ unlock:
 }
 
 /**
- * cpuset_hotplug_workfn - handle CPU/memory hotunplug for a cpuset
- * @work: unused
+ * cpuset_handle_hotplug - handle CPU/memory hot{,un}plug for a cpuset
  *
  * This function is called after either CPU or memory configuration has
  * changed and updates cpuset accordingly.  The top_cpuset is always
@@ -4573,8 +4537,10 @@ unlock:
  *
  * Note that CPU offlining during suspend is ignored.  We don't modify
  * cpusets across suspend/resume cycles at all.
+ *
+ * CPU / memory hotplug is handled synchronously.
  */
-static void cpuset_hotplug_workfn(struct work_struct *work)
+static void cpuset_handle_hotplug(void)
 {
 	static cpumask_t new_cpus;
 	static nodemask_t new_mems;
@@ -4585,6 +4551,7 @@ static void cpuset_hotplug_workfn(struct work_struct *work)
 	if (on_dfl && !alloc_cpumasks(NULL, &tmp))
 		ptmp = &tmp;
 
+	lockdep_assert_cpus_held();
 	mutex_lock(&cpuset_mutex);
 
 	/* fetch the available cpus/mems and find out which changed how */
@@ -4666,7 +4633,7 @@ static void cpuset_hotplug_workfn(struct work_struct *work)
 	/* rebuild sched domains if cpus_allowed has changed */
 	if (cpus_updated || force_rebuild) {
 		force_rebuild = false;
-		rebuild_sched_domains();
+		rebuild_sched_domains_cpuslocked();
 	}
 
 	free_cpumasks(NULL, ptmp);
@@ -4679,12 +4646,7 @@ void cpuset_update_active_cpus(void)
 	 * inside cgroup synchronization.  Bounce actual hotplug processing
 	 * to a work item to avoid reverse locking order.
 	 */
-	schedule_work(&cpuset_hotplug_work);
-}
-
-void cpuset_wait_for_hotplug(void)
-{
-	flush_work(&cpuset_hotplug_work);
+	cpuset_handle_hotplug();
 }
 
 /*
@@ -4695,7 +4657,7 @@ void cpuset_wait_for_hotplug(void)
 static int cpuset_track_online_nodes(struct notifier_block *self,
 				unsigned long action, void *arg)
 {
-	schedule_work(&cpuset_hotplug_work);
+	cpuset_handle_hotplug();
 	return NOTIFY_OK;
 }
 
diff --git a/kernel/cgroup/legacy_freezer.c b/kernel/cgroup/legacy_freezer.c
index 66d1708042a72b..074653f964c1d0 100644
--- a/kernel/cgroup/legacy_freezer.c
+++ b/kernel/cgroup/legacy_freezer.c
@@ -106,8 +106,7 @@ freezer_css_alloc(struct cgroup_subsys_state *parent_css)
  * @css: css being created
  *
  * We're committing to creation of @css.  Mark it online and inherit
- * parent's freezing state while holding both parent's and our
- * freezer->lock.
+ * parent's freezing state while holding cpus read lock and freezer_mutex.
  */
 static int freezer_css_online(struct cgroup_subsys_state *css)
 {
@@ -133,7 +132,7 @@ static int freezer_css_online(struct cgroup_subsys_state *css)
  * freezer_css_offline - initiate destruction of a freezer css
  * @css: css being destroyed
  *
- * @css is going away.  Mark it dead and decrement system_freezing_count if
+ * @css is going away.  Mark it dead and decrement freezer_active if
  * it was holding one.
  */
 static void freezer_css_offline(struct cgroup_subsys_state *css)
diff --git a/kernel/cgroup/pids.c b/kernel/cgroup/pids.c
index 7695e60bcb4092..0e5ec7d59b4d21 100644
--- a/kernel/cgroup/pids.c
+++ b/kernel/cgroup/pids.c
@@ -75,9 +75,7 @@ pids_css_alloc(struct cgroup_subsys_state *parent)
 	if (!pids)
 		return ERR_PTR(-ENOMEM);
 
-	atomic64_set(&pids->counter, 0);
 	atomic64_set(&pids->limit, PIDS_MAX);
-	atomic64_set(&pids->events_limit, 0);
 	return &pids->css;
 }
 
diff --git a/kernel/cgroup/rstat.c b/kernel/cgroup/rstat.c
index 07e2284bb49971..52e3b0ed1ceea3 100644
--- a/kernel/cgroup/rstat.c
+++ b/kernel/cgroup/rstat.c
@@ -7,6 +7,8 @@
 #include <linux/btf.h>
 #include <linux/btf_ids.h>
 
+#include <trace/events/cgroup.h>
+
 static DEFINE_SPINLOCK(cgroup_rstat_lock);
 static DEFINE_PER_CPU(raw_spinlock_t, cgroup_rstat_cpu_lock);
 
@@ -222,6 +224,35 @@ __weak noinline void bpf_rstat_flush(struct cgroup *cgrp,
 
 __bpf_hook_end();
 
+/*
+ * Helper functions for locking cgroup_rstat_lock.
+ *
+ * This makes it easier to diagnose locking issues and contention in
+ * production environments.  The parameter @cpu_in_loop indicate lock
+ * was released and re-taken when collection data from the CPUs. The
+ * value -1 is used when obtaining the main lock else this is the CPU
+ * number processed last.
+ */
+static inline void __cgroup_rstat_lock(struct cgroup *cgrp, int cpu_in_loop)
+	__acquires(&cgroup_rstat_lock)
+{
+	bool contended;
+
+	contended = !spin_trylock_irq(&cgroup_rstat_lock);
+	if (contended) {
+		trace_cgroup_rstat_lock_contended(cgrp, cpu_in_loop, contended);
+		spin_lock_irq(&cgroup_rstat_lock);
+	}
+	trace_cgroup_rstat_locked(cgrp, cpu_in_loop, contended);
+}
+
+static inline void __cgroup_rstat_unlock(struct cgroup *cgrp, int cpu_in_loop)
+	__releases(&cgroup_rstat_lock)
+{
+	trace_cgroup_rstat_unlock(cgrp, cpu_in_loop, false);
+	spin_unlock_irq(&cgroup_rstat_lock);
+}
+
 /* see cgroup_rstat_flush() */
 static void cgroup_rstat_flush_locked(struct cgroup *cgrp)
 	__releases(&cgroup_rstat_lock) __acquires(&cgroup_rstat_lock)
@@ -248,10 +279,10 @@ static void cgroup_rstat_flush_locked(struct cgroup *cgrp)
 
 		/* play nice and yield if necessary */
 		if (need_resched() || spin_needbreak(&cgroup_rstat_lock)) {
-			spin_unlock_irq(&cgroup_rstat_lock);
+			__cgroup_rstat_unlock(cgrp, cpu);
 			if (!cond_resched())
 				cpu_relax();
-			spin_lock_irq(&cgroup_rstat_lock);
+			__cgroup_rstat_lock(cgrp, cpu);
 		}
 	}
 }
@@ -273,9 +304,9 @@ __bpf_kfunc void cgroup_rstat_flush(struct cgroup *cgrp)
 {
 	might_sleep();
 
-	spin_lock_irq(&cgroup_rstat_lock);
+	__cgroup_rstat_lock(cgrp, -1);
 	cgroup_rstat_flush_locked(cgrp);
-	spin_unlock_irq(&cgroup_rstat_lock);
+	__cgroup_rstat_unlock(cgrp, -1);
 }
 
 /**
@@ -291,17 +322,18 @@ void cgroup_rstat_flush_hold(struct cgroup *cgrp)
 	__acquires(&cgroup_rstat_lock)
 {
 	might_sleep();
-	spin_lock_irq(&cgroup_rstat_lock);
+	__cgroup_rstat_lock(cgrp, -1);
 	cgroup_rstat_flush_locked(cgrp);
 }
 
 /**
  * cgroup_rstat_flush_release - release cgroup_rstat_flush_hold()
+ * @cgrp: cgroup used by tracepoint
  */
-void cgroup_rstat_flush_release(void)
+void cgroup_rstat_flush_release(struct cgroup *cgrp)
 	__releases(&cgroup_rstat_lock)
 {
-	spin_unlock_irq(&cgroup_rstat_lock);
+	__cgroup_rstat_unlock(cgrp, -1);
 }
 
 int cgroup_rstat_init(struct cgroup *cgrp)
@@ -533,7 +565,7 @@ void cgroup_base_stat_cputime_show(struct seq_file *seq)
 #ifdef CONFIG_SCHED_CORE
 		forceidle_time = cgrp->bstat.forceidle_sum;
 #endif
-		cgroup_rstat_flush_release();
+		cgroup_rstat_flush_release(cgrp);
 	} else {
 		root_cgroup_cputime(&bstat);
 		usage = bstat.cputime.sum_exec_runtime;
diff --git a/kernel/cpu.c b/kernel/cpu.c
index 63447eb85dab6b..563877d6c28b65 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -1208,52 +1208,6 @@ void __init cpuhp_threads_init(void)
 	kthread_unpark(this_cpu_read(cpuhp_state.thread));
 }
 
-/*
- *
- * Serialize hotplug trainwrecks outside of the cpu_hotplug_lock
- * protected region.
- *
- * The operation is still serialized against concurrent CPU hotplug via
- * cpu_add_remove_lock, i.e. CPU map protection.  But it is _not_
- * serialized against other hotplug related activity like adding or
- * removing of state callbacks and state instances, which invoke either the
- * startup or the teardown callback of the affected state.
- *
- * This is required for subsystems which are unfixable vs. CPU hotplug and
- * evade lock inversion problems by scheduling work which has to be
- * completed _before_ cpu_up()/_cpu_down() returns.
- *
- * Don't even think about adding anything to this for any new code or even
- * drivers. It's only purpose is to keep existing lock order trainwrecks
- * working.
- *
- * For cpu_down() there might be valid reasons to finish cleanups which are
- * not required to be done under cpu_hotplug_lock, but that's a different
- * story and would be not invoked via this.
- */
-static void cpu_up_down_serialize_trainwrecks(bool tasks_frozen)
-{
-	/*
-	 * cpusets delegate hotplug operations to a worker to "solve" the
-	 * lock order problems. Wait for the worker, but only if tasks are
-	 * _not_ frozen (suspend, hibernate) as that would wait forever.
-	 *
-	 * The wait is required because otherwise the hotplug operation
-	 * returns with inconsistent state, which could even be observed in
-	 * user space when a new CPU is brought up. The CPU plug uevent
-	 * would be delivered and user space reacting on it would fail to
-	 * move tasks to the newly plugged CPU up to the point where the
-	 * work has finished because up to that point the newly plugged CPU
-	 * is not assignable in cpusets/cgroups. On unplug that's not
-	 * necessarily a visible issue, but it is still inconsistent state,
-	 * which is the real problem which needs to be "fixed". This can't
-	 * prevent the transient state between scheduling the work and
-	 * returning from waiting for it.
-	 */
-	if (!tasks_frozen)
-		cpuset_wait_for_hotplug();
-}
-
 #ifdef CONFIG_HOTPLUG_CPU
 #ifndef arch_clear_mm_cpumask_cpu
 #define arch_clear_mm_cpumask_cpu(cpu, mm) cpumask_clear_cpu(cpu, mm_cpumask(mm))
@@ -1494,7 +1448,6 @@ out:
 	 */
 	lockup_detector_cleanup();
 	arch_smt_update();
-	cpu_up_down_serialize_trainwrecks(tasks_frozen);
 	return ret;
 }
 
@@ -1728,7 +1681,6 @@ static int _cpu_up(unsigned int cpu, int tasks_frozen, enum cpuhp_state target)
 out:
 	cpus_write_unlock();
 	arch_smt_update();
-	cpu_up_down_serialize_trainwrecks(tasks_frozen);
 	return ret;
 }
 
diff --git a/kernel/power/process.c b/kernel/power/process.c
index cae81a87cc91e3..66ac067d9ae64e 100644
--- a/kernel/power/process.c
+++ b/kernel/power/process.c
@@ -194,8 +194,6 @@ void thaw_processes(void)
 	__usermodehelper_set_disable_depth(UMH_FREEZING);
 	thaw_workqueues();
 
-	cpuset_wait_for_hotplug();
-
 	read_lock(&tasklist_lock);
 	for_each_process_thread(g, p) {
 		/* No other threads should have PF_SUSPEND_TASK set */
diff --git a/tools/testing/selftests/cgroup/Makefile b/tools/testing/selftests/cgroup/Makefile
index 00b44192890961..16461dc0ffdf9d 100644
--- a/tools/testing/selftests/cgroup/Makefile
+++ b/tools/testing/selftests/cgroup/Makefile
@@ -4,7 +4,7 @@ CFLAGS += -Wall -pthread
 all: ${HELPER_PROGS}
 
 TEST_FILES     := with_stress.sh
-TEST_PROGS     := test_stress.sh test_cpuset_prs.sh
+TEST_PROGS     := test_stress.sh test_cpuset_prs.sh test_cpuset_v1_hp.sh
 TEST_GEN_FILES := wait_inotify
 TEST_GEN_PROGS = test_memcontrol
 TEST_GEN_PROGS += test_kmem
diff --git a/tools/testing/selftests/cgroup/cgroup_util.c b/tools/testing/selftests/cgroup/cgroup_util.c
index 0340d4ca8f51cb..432db923bced0c 100644
--- a/tools/testing/selftests/cgroup/cgroup_util.c
+++ b/tools/testing/selftests/cgroup/cgroup_util.c
@@ -195,10 +195,10 @@ int cg_write_numeric(const char *cgroup, const char *control, long value)
 	return cg_write(cgroup, control, buf);
 }
 
-int cg_find_unified_root(char *root, size_t len)
+int cg_find_unified_root(char *root, size_t len, bool *nsdelegate)
 {
 	char buf[10 * PAGE_SIZE];
-	char *fs, *mount, *type;
+	char *fs, *mount, *type, *options;
 	const char delim[] = "\n\t ";
 
 	if (read_text("/proc/self/mounts", buf, sizeof(buf)) <= 0)
@@ -211,12 +211,14 @@ int cg_find_unified_root(char *root, size_t len)
 	for (fs = strtok(buf, delim); fs; fs = strtok(NULL, delim)) {
 		mount = strtok(NULL, delim);
 		type = strtok(NULL, delim);
-		strtok(NULL, delim);
+		options = strtok(NULL, delim);
 		strtok(NULL, delim);
 		strtok(NULL, delim);
 
 		if (strcmp(type, "cgroup2") == 0) {
 			strncpy(root, mount, len);
+			if (nsdelegate)
+				*nsdelegate = !!strstr(options, "nsdelegate");
 			return 0;
 		}
 	}
diff --git a/tools/testing/selftests/cgroup/cgroup_util.h b/tools/testing/selftests/cgroup/cgroup_util.h
index 1df7f202214afc..89e8519fb2719c 100644
--- a/tools/testing/selftests/cgroup/cgroup_util.h
+++ b/tools/testing/selftests/cgroup/cgroup_util.h
@@ -21,7 +21,7 @@ static inline int values_close(long a, long b, int err)
 	return abs(a - b) <= (a + b) / 100 * err;
 }
 
-extern int cg_find_unified_root(char *root, size_t len);
+extern int cg_find_unified_root(char *root, size_t len, bool *nsdelegate);
 extern char *cg_name(const char *root, const char *name);
 extern char *cg_name_indexed(const char *root, const char *name, int index);
 extern char *cg_control(const char *cgroup, const char *control);
diff --git a/tools/testing/selftests/cgroup/test_core.c b/tools/testing/selftests/cgroup/test_core.c
index 80aa6b2373b966..a5672a91d273ce 100644
--- a/tools/testing/selftests/cgroup/test_core.c
+++ b/tools/testing/selftests/cgroup/test_core.c
@@ -18,6 +18,8 @@
 #include "../kselftest.h"
 #include "cgroup_util.h"
 
+static bool nsdelegate;
+
 static int touch_anon(char *buf, size_t size)
 {
 	int fd;
@@ -775,6 +777,9 @@ static int test_cgcore_lesser_ns_open(const char *root)
 	pid_t pid;
 	int status;
 
+	if (!nsdelegate)
+		return KSFT_SKIP;
+
 	cg_test_a = cg_name(root, "cg_test_a");
 	cg_test_b = cg_name(root, "cg_test_b");
 
@@ -862,7 +867,7 @@ int main(int argc, char *argv[])
 	char root[PATH_MAX];
 	int i, ret = EXIT_SUCCESS;
 
-	if (cg_find_unified_root(root, sizeof(root)))
+	if (cg_find_unified_root(root, sizeof(root), &nsdelegate))
 		ksft_exit_skip("cgroup v2 isn't mounted\n");
 
 	if (cg_read_strstr(root, "cgroup.subtree_control", "memory"))
diff --git a/tools/testing/selftests/cgroup/test_cpu.c b/tools/testing/selftests/cgroup/test_cpu.c
index 24020a2c68dcdd..186bf96f6a2846 100644
--- a/tools/testing/selftests/cgroup/test_cpu.c
+++ b/tools/testing/selftests/cgroup/test_cpu.c
@@ -700,7 +700,7 @@ int main(int argc, char *argv[])
 	char root[PATH_MAX];
 	int i, ret = EXIT_SUCCESS;
 
-	if (cg_find_unified_root(root, sizeof(root)))
+	if (cg_find_unified_root(root, sizeof(root), NULL))
 		ksft_exit_skip("cgroup v2 isn't mounted\n");
 
 	if (cg_read_strstr(root, "cgroup.subtree_control", "cpu"))
diff --git a/tools/testing/selftests/cgroup/test_cpuset.c b/tools/testing/selftests/cgroup/test_cpuset.c
index b061ed1e05b4d0..4034d14ba69ac0 100644
--- a/tools/testing/selftests/cgroup/test_cpuset.c
+++ b/tools/testing/selftests/cgroup/test_cpuset.c
@@ -249,7 +249,7 @@ int main(int argc, char *argv[])
 	char root[PATH_MAX];
 	int i, ret = EXIT_SUCCESS;
 
-	if (cg_find_unified_root(root, sizeof(root)))
+	if (cg_find_unified_root(root, sizeof(root), NULL))
 		ksft_exit_skip("cgroup v2 isn't mounted\n");
 
 	if (cg_read_strstr(root, "cgroup.subtree_control", "cpuset"))
diff --git a/tools/testing/selftests/cgroup/test_cpuset_v1_hp.sh b/tools/testing/selftests/cgroup/test_cpuset_v1_hp.sh
new file mode 100755
index 00000000000000..3f45512fb512eb
--- /dev/null
+++ b/tools/testing/selftests/cgroup/test_cpuset_v1_hp.sh
@@ -0,0 +1,46 @@
+#!/bin/sh
+# SPDX-License-Identifier: GPL-2.0
+#
+# Test the special cpuset v1 hotplug case where a cpuset become empty of
+# CPUs will force migration of tasks out to an ancestor.
+#
+
+skip_test() {
+	echo "$1"
+	echo "Test SKIPPED"
+	exit 4 # ksft_skip
+}
+
+[[ $(id -u) -eq 0 ]] || skip_test "Test must be run as root!"
+
+# Find cpuset v1 mount point
+CPUSET=$(mount -t cgroup | grep cpuset | head -1 | awk -e '{print $3}')
+[[ -n "$CPUSET" ]] || skip_test "cpuset v1 mount point not found!"
+
+#
+# Create a test cpuset, put a CPU and a task there and offline that CPU
+#
+TDIR=test$$
+[[ -d $CPUSET/$TDIR ]] || mkdir $CPUSET/$TDIR
+echo 1 > $CPUSET/$TDIR/cpuset.cpus
+echo 0 > $CPUSET/$TDIR/cpuset.mems
+sleep 10&
+TASK=$!
+echo $TASK > $CPUSET/$TDIR/tasks
+NEWCS=$(cat /proc/$TASK/cpuset)
+[[ $NEWCS != "/$TDIR" ]] && {
+	echo "Unexpected cpuset $NEWCS, test FAILED!"
+	exit 1
+}
+
+echo 0 > /sys/devices/system/cpu/cpu1/online
+sleep 0.5
+echo 1 > /sys/devices/system/cpu/cpu1/online
+NEWCS=$(cat /proc/$TASK/cpuset)
+rmdir $CPUSET/$TDIR
+[[ $NEWCS != "/" ]] && {
+	echo "cpuset $NEWCS, test FAILED!"
+	exit 1
+}
+echo "Test PASSED"
+exit 0
diff --git a/tools/testing/selftests/cgroup/test_freezer.c b/tools/testing/selftests/cgroup/test_freezer.c
index 8845353aca53bb..8730645d363a73 100644
--- a/tools/testing/selftests/cgroup/test_freezer.c
+++ b/tools/testing/selftests/cgroup/test_freezer.c
@@ -827,7 +827,7 @@ int main(int argc, char *argv[])
 	char root[PATH_MAX];
 	int i, ret = EXIT_SUCCESS;
 
-	if (cg_find_unified_root(root, sizeof(root)))
+	if (cg_find_unified_root(root, sizeof(root), NULL))
 		ksft_exit_skip("cgroup v2 isn't mounted\n");
 	for (i = 0; i < ARRAY_SIZE(tests); i++) {
 		switch (tests[i].fn(root)) {
diff --git a/tools/testing/selftests/cgroup/test_hugetlb_memcg.c b/tools/testing/selftests/cgroup/test_hugetlb_memcg.c
index f0fefeb4cc24c4..856f9508ea562e 100644
--- a/tools/testing/selftests/cgroup/test_hugetlb_memcg.c
+++ b/tools/testing/selftests/cgroup/test_hugetlb_memcg.c
@@ -214,7 +214,7 @@ int main(int argc, char **argv)
 		return ret;
 	}
 
-	if (cg_find_unified_root(root, sizeof(root)))
+	if (cg_find_unified_root(root, sizeof(root), NULL))
 		ksft_exit_skip("cgroup v2 isn't mounted\n");
 
 	switch (test_hugetlb_memcg(root)) {
diff --git a/tools/testing/selftests/cgroup/test_kill.c b/tools/testing/selftests/cgroup/test_kill.c
index 6153690319c9c8..0e5bb6c7307a50 100644
--- a/tools/testing/selftests/cgroup/test_kill.c
+++ b/tools/testing/selftests/cgroup/test_kill.c
@@ -276,7 +276,7 @@ int main(int argc, char *argv[])
 	char root[PATH_MAX];
 	int i, ret = EXIT_SUCCESS;
 
-	if (cg_find_unified_root(root, sizeof(root)))
+	if (cg_find_unified_root(root, sizeof(root), NULL))
 		ksft_exit_skip("cgroup v2 isn't mounted\n");
 	for (i = 0; i < ARRAY_SIZE(tests); i++) {
 		switch (tests[i].fn(root)) {
diff --git a/tools/testing/selftests/cgroup/test_kmem.c b/tools/testing/selftests/cgroup/test_kmem.c
index c82f974b85c94d..137506db03127f 100644
--- a/tools/testing/selftests/cgroup/test_kmem.c
+++ b/tools/testing/selftests/cgroup/test_kmem.c
@@ -420,7 +420,7 @@ int main(int argc, char **argv)
 	char root[PATH_MAX];
 	int i, ret = EXIT_SUCCESS;
 
-	if (cg_find_unified_root(root, sizeof(root)))
+	if (cg_find_unified_root(root, sizeof(root), NULL))
 		ksft_exit_skip("cgroup v2 isn't mounted\n");
 
 	/*
diff --git a/tools/testing/selftests/cgroup/test_memcontrol.c b/tools/testing/selftests/cgroup/test_memcontrol.c
index c7c9572003a8c9..b462416b380612 100644
--- a/tools/testing/selftests/cgroup/test_memcontrol.c
+++ b/tools/testing/selftests/cgroup/test_memcontrol.c
@@ -1314,7 +1314,7 @@ int main(int argc, char **argv)
 	char root[PATH_MAX];
 	int i, proc_status, ret = EXIT_SUCCESS;
 
-	if (cg_find_unified_root(root, sizeof(root)))
+	if (cg_find_unified_root(root, sizeof(root), NULL))
 		ksft_exit_skip("cgroup v2 isn't mounted\n");
 
 	/*
diff --git a/tools/testing/selftests/cgroup/test_zswap.c b/tools/testing/selftests/cgroup/test_zswap.c
index f0e488ed90d895..ef7f3954531738 100644
--- a/tools/testing/selftests/cgroup/test_zswap.c
+++ b/tools/testing/selftests/cgroup/test_zswap.c
@@ -440,7 +440,7 @@ int main(int argc, char **argv)
 	char root[PATH_MAX];
 	int i, ret = EXIT_SUCCESS;
 
-	if (cg_find_unified_root(root, sizeof(root)))
+	if (cg_find_unified_root(root, sizeof(root), NULL))
 		ksft_exit_skip("cgroup v2 isn't mounted\n");
 
 	if (!zswap_configured())