Merge tag 'cgroup-for-6.2' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup

Pull cgroup updates from Tejun Heo:
 "Nothing too interesting:

   - Add CONFIG_DEBUG_GROUP_REF which makes cgroup refcnt operations
     kprobable

   - A couple cpuset optimizations

   - Other misc changes including doc and test updates"

* tag 'cgroup-for-6.2' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup:
  cgroup: remove rcu_read_lock()/rcu_read_unlock() in critical section of spin_lock_irq()
  cgroup/cpuset: Improve cpuset_css_alloc() description
  kselftest/cgroup: Add cleanup() to test_cpuset_prs.sh
  cgroup/cpuset: Optimize cpuset_attach() on v2
  cgroup/cpuset: Skip spread flags update on v2
  kselftest/cgroup: Fix gathering number of CPUs
  cgroup: cgroup refcnt functions should be exported when CONFIG_DEBUG_CGROUP_REF
  cgroup: Implement DEBUG_CGROUP_REF

2 files changed, 45 insertions, 11 deletions

diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c
index 2319946715e0..15cc26513596 100644
--- a/kernel/cgroup/cgroup.c
+++ b/kernel/cgroup/cgroup.c
@@ -248,6 +248,12 @@ static int cgroup_addrm_files(struct cgroup_subsys_state *css,
 			      struct cgroup *cgrp, struct cftype cfts[],
 			      bool is_add);
 
+#ifdef CONFIG_DEBUG_CGROUP_REF
+#define CGROUP_REF_FN_ATTRS	noinline
+#define CGROUP_REF_EXPORT(fn)	EXPORT_SYMBOL_GPL(fn);
+#include <linux/cgroup_refcnt.h>
+#endif
+
 /**
  * cgroup_ssid_enabled - cgroup subsys enabled test by subsys ID
  * @ssid: subsys ID of interest
@@ -2860,14 +2866,12 @@ int cgroup_migrate(struct task_struct *leader, bool threadgroup,
 	 * take an rcu_read_lock.
 	 */
 	spin_lock_irq(&css_set_lock);
-	rcu_read_lock();
 	task = leader;
 	do {
 		cgroup_migrate_add_task(task, mgctx);
 		if (!threadgroup)
 			break;
 	} while_each_thread(leader, task);
-	rcu_read_unlock();
 	spin_unlock_irq(&css_set_lock);
 
 	return cgroup_migrate_execute(mgctx);
diff --git a/kernel/cgroup/cpuset.c b/kernel/cgroup/cpuset.c
index b474289c15b8..589827ccda8b 100644
--- a/kernel/cgroup/cpuset.c
+++ b/kernel/cgroup/cpuset.c
@@ -550,11 +550,15 @@ static void guarantee_online_mems(struct cpuset *cs, nodemask_t *pmask)
 /*
  * update task's spread flag if cpuset's page/slab spread flag is set
  *
- * Call with callback_lock or cpuset_rwsem held.
+ * Call with callback_lock or cpuset_rwsem held. The check can be skipped
+ * if on default hierarchy.
  */
-static void cpuset_update_task_spread_flag(struct cpuset *cs,
+static void cpuset_update_task_spread_flags(struct cpuset *cs,
 					struct task_struct *tsk)
 {
+	if (cgroup_subsys_on_dfl(cpuset_cgrp_subsys))
+		return;
+
 	if (is_spread_page(cs))
 		task_set_spread_page(tsk);
 	else
@@ -2153,7 +2157,7 @@ static void update_tasks_flags(struct cpuset *cs)
 
 	css_task_iter_start(&cs->css, 0, &it);
 	while ((task = css_task_iter_next(&it)))
-		cpuset_update_task_spread_flag(cs, task);
+		cpuset_update_task_spread_flags(cs, task);
 	css_task_iter_end(&it);
 }
 
@@ -2509,12 +2513,28 @@ static void cpuset_attach(struct cgroup_taskset *tset)
 	struct cgroup_subsys_state *css;
 	struct cpuset *cs;
 	struct cpuset *oldcs = cpuset_attach_old_cs;
+	bool cpus_updated, mems_updated;
 
 	cgroup_taskset_first(tset, &css);
 	cs = css_cs(css);
 
 	lockdep_assert_cpus_held();	/* see cgroup_attach_lock() */
 	percpu_down_write(&cpuset_rwsem);
+	cpus_updated = !cpumask_equal(cs->effective_cpus,
+				      oldcs->effective_cpus);
+	mems_updated = !nodes_equal(cs->effective_mems, oldcs->effective_mems);
+
+	/*
+	 * In the default hierarchy, enabling cpuset in the child cgroups
+	 * will trigger a number of cpuset_attach() calls with no change
+	 * in effective cpus and mems. In that case, we can optimize out
+	 * by skipping the task iteration and update.
+	 */
+	if (cgroup_subsys_on_dfl(cpuset_cgrp_subsys) &&
+	    !cpus_updated && !mems_updated) {
+		cpuset_attach_nodemask_to = cs->effective_mems;
+		goto out;
+	}
 
 	guarantee_online_mems(cs, &cpuset_attach_nodemask_to);
 
@@ -2530,14 +2550,19 @@ static void cpuset_attach(struct cgroup_taskset *tset)
 		WARN_ON_ONCE(set_cpus_allowed_ptr(task, cpus_attach));
 
 		cpuset_change_task_nodemask(task, &cpuset_attach_nodemask_to);
-		cpuset_update_task_spread_flag(cs, task);
+		cpuset_update_task_spread_flags(cs, task);
 	}
 
 	/*
 	 * Change mm for all threadgroup leaders. This is expensive and may
-	 * sleep and should be moved outside migration path proper.
+	 * sleep and should be moved outside migration path proper. Skip it
+	 * if there is no change in effective_mems and CS_MEMORY_MIGRATE is
+	 * not set.
 	 */
 	cpuset_attach_nodemask_to = cs->effective_mems;
+	if (!is_memory_migrate(cs) && !mems_updated)
+		goto out;
+
 	cgroup_taskset_for_each_leader(leader, css, tset) {
 		struct mm_struct *mm = get_task_mm(leader);
 
@@ -2560,6 +2585,7 @@ static void cpuset_attach(struct cgroup_taskset *tset)
 		}
 	}
 
+out:
 	cs->old_mems_allowed = cpuset_attach_nodemask_to;
 
 	cs->attach_in_progress--;
@@ -3046,11 +3072,15 @@ static struct cftype dfl_files[] = {
 };
 
 
-/*
- *	cpuset_css_alloc - allocate a cpuset css
- *	cgrp:	control group that the new cpuset will be part of
+/**
+ * cpuset_css_alloc - Allocate a cpuset css
+ * @parent_css: Parent css of the control group that the new cpuset will be
+ *              part of
+ * Return: cpuset css on success, -ENOMEM on failure.
+ *
+ * Allocate and initialize a new cpuset css, for non-NULL @parent_css, return
+ * top cpuset css otherwise.
  */
-
 static struct cgroup_subsys_state *
 cpuset_css_alloc(struct cgroup_subsys_state *parent_css)
 {