cgroup: Show # of subsystem CSSes in cgroup.stat

Cgroup subsystem state (CSS) is an abstraction in the cgroup layer to
help manage different structures in various cgroup subsystems by being
an embedded element inside a larger structure like cpuset or mem_cgroup.

The /proc/cgroups file shows the number of cgroups for each of the
subsystems.  With cgroup v1, the number of CSSes is the same as the
number of cgroups.  That is not the case anymore with cgroup v2. The
/proc/cgroups file cannot show the actual number of CSSes for the
subsystems that are bound to cgroup v2.

So if a v2 cgroup subsystem is leaking cgroups (usually memory cgroup),
we can't tell by looking at /proc/cgroups which cgroup subsystems may
be responsible.

As cgroup v2 had deprecated the use of /proc/cgroups, the hierarchical
cgroup.stat file is now being extended to show the number of live and
dying CSSes associated with all the non-inhibited cgroup subsystems that
have been bound to cgroup v2. The number includes CSSes in the current
cgroup as well as in all the descendants underneath it.  This will help
us pinpoint which subsystems are responsible for the increasing number
of dying (nr_dying_descendants) cgroups.

The CSSes dying counts are stored in the cgroup structure itself
instead of inside the CSS as suggested by Johannes. This will allow
us to accurately track dying counts of cgroup subsystems that have
recently been disabled in a cgroup. It is now possible that a zero
subsystem number is coupled with a non-zero dying subsystem number.

The cgroup-v2.rst file is updated to discuss this new behavior.

With this patch applied, a sample output from root cgroup.stat file
was shown below.

	nr_descendants 56
	nr_subsys_cpuset 1
	nr_subsys_cpu 43
	nr_subsys_io 43
	nr_subsys_memory 56
	nr_subsys_perf_event 57
	nr_subsys_hugetlb 1
	nr_subsys_pids 56
	nr_subsys_rdma 1
	nr_subsys_misc 1
	nr_dying_descendants 30
	nr_dying_subsys_cpuset 0
	nr_dying_subsys_cpu 0
	nr_dying_subsys_io 0
	nr_dying_subsys_memory 30
	nr_dying_subsys_perf_event 0
	nr_dying_subsys_hugetlb 0
	nr_dying_subsys_pids 0
	nr_dying_subsys_rdma 0
	nr_dying_subsys_misc 0

Another sample output from system.slice/cgroup.stat was:

	nr_descendants 34
	nr_subsys_cpuset 0
	nr_subsys_cpu 32
	nr_subsys_io 32
	nr_subsys_memory 34
	nr_subsys_perf_event 35
	nr_subsys_hugetlb 0
	nr_subsys_pids 34
	nr_subsys_rdma 0
	nr_subsys_misc 0
	nr_dying_descendants 30
	nr_dying_subsys_cpuset 0
	nr_dying_subsys_cpu 0
	nr_dying_subsys_io 0
	nr_dying_subsys_memory 30
	nr_dying_subsys_perf_event 0
	nr_dying_subsys_hugetlb 0
	nr_dying_subsys_pids 0
	nr_dying_subsys_rdma 0
	nr_dying_subsys_misc 0

Note that 'debug' controller wasn't used to provide this information because
the controller is not recommended in productions kernels, also many of them
won't enable CONFIG_CGROUP_DEBUG by default.

Similar information could be retrieved with debuggers like drgn but that's
also not always available (e.g. lockdown) and the additional cost of runtime
tracking here is deemed marginal.

tj: Added Michal's paragraphs on why this is not added the debug controller
    to the commit message.

Signed-off-by: Waiman Long <longman@redhat.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Roman Gushchin <roman.gushchin@linux.dev>
Reviewed-by: Kamalesh Babulal <kamalesh.babulal@oracle.com>
Cc: Michal Koutný <mkoutny@suse.com>
Link: http://lkml.kernel.org/r/20240715150034.2583772-1-longman@redhat.com
Signed-off-by: Tejun Heo <tj@kernel.org>

Documentation/admin-guide/cgroup-v2.rst

@@ -981,6 +981,14 @@ All cgroup core files are prefixed with "cgroup."
 		A dying cgroup can consume system resources not exceeding
 		limits, which were active at the moment of cgroup deletion.
 
+	  nr_subsys_<cgroup_subsys>
+		Total number of live cgroup subsystems (e.g memory
+		cgroup) at and beneath the current cgroup.
+
+	  nr_dying_subsys_<cgroup_subsys>
+		Total number of dying cgroup subsystems (e.g. memory
+		cgroup) at and beneath the current cgroup.
+
   cgroup.freeze
 	A read-write single value file which exists on non-root cgroups.
 	Allowed values are "0" and "1". The default is "0".
@@ -2939,8 +2947,8 @@ Deprecated v1 Core Features
 
 - "cgroup.clone_children" is removed.
 
-- /proc/cgroups is meaningless for v2.  Use "cgroup.controllers" file
-  at the root instead.
+- /proc/cgroups is meaningless for v2.  Use "cgroup.controllers" or
+  "cgroup.stat" files at the root instead.
 
 
 Issues with v1 and Rationales for v2

include/linux/cgroup-defs.h

@@ -210,6 +210,14 @@ struct cgroup_subsys_state {
 	 * fields of the containing structure.
 	 */
 	struct cgroup_subsys_state *parent;
+
+	/*
+	 * Keep track of total numbers of visible descendant CSSes.
+	 * The total number of dying CSSes is tracked in
+	 * css->cgroup->nr_dying_subsys[ssid].
+	 * Protected by cgroup_mutex.
+	 */
+	int nr_descendants;
 };
 
 /*
@@ -470,6 +478,12 @@ struct cgroup {
 	/* Private pointers for each registered subsystem */
 	struct cgroup_subsys_state __rcu *subsys[CGROUP_SUBSYS_COUNT];
 
+	/*
+	 * Keep track of total number of dying CSSes at and below this cgroup.
+	 * Protected by cgroup_mutex.
+	 */
+	int nr_dying_subsys[CGROUP_SUBSYS_COUNT];
+
 	struct cgroup_root *root;
 
 	/*

kernel/cgroup/cgroup.c

@@ -3669,12 +3669,40 @@ static int cgroup_events_show(struct seq_file *seq, void *v)
 static int cgroup_stat_show(struct seq_file *seq, void *v)
 {
 	struct cgroup *cgroup = seq_css(seq)->cgroup;
+	struct cgroup_subsys_state *css;
+	int dying_cnt[CGROUP_SUBSYS_COUNT];
+	int ssid;
 
 	seq_printf(seq, "nr_descendants %d\n",
 		   cgroup->nr_descendants);
+
+	/*
+	 * Show the number of live and dying csses associated with each of
+	 * non-inhibited cgroup subsystems that is bound to cgroup v2.
+	 *
+	 * Without proper lock protection, racing is possible. So the
+	 * numbers may not be consistent when that happens.
+	 */
+	rcu_read_lock();
+	for (ssid = 0; ssid < CGROUP_SUBSYS_COUNT; ssid++) {
+		dying_cnt[ssid] = -1;
+		if ((BIT(ssid) & cgrp_dfl_inhibit_ss_mask) ||
+		    (cgroup_subsys[ssid]->root !=  &cgrp_dfl_root))
+			continue;
+		css = rcu_dereference_raw(cgroup->subsys[ssid]);
+		dying_cnt[ssid] = cgroup->nr_dying_subsys[ssid];
+		seq_printf(seq, "nr_subsys_%s %d\n", cgroup_subsys[ssid]->name,
+			   css ? (css->nr_descendants + 1) : 0);
+	}
+
 	seq_printf(seq, "nr_dying_descendants %d\n",
 		   cgroup->nr_dying_descendants);
-
+	for (ssid = 0; ssid < CGROUP_SUBSYS_COUNT; ssid++) {
+		if (dying_cnt[ssid] >= 0)
+			seq_printf(seq, "nr_dying_subsys_%s %d\n",
+				   cgroup_subsys[ssid]->name, dying_cnt[ssid]);
+	}
+	rcu_read_unlock();
 	return 0;
 }
 
@@ -5424,6 +5452,8 @@ static void css_release_work_fn(struct work_struct *work)
 	list_del_rcu(&css->sibling);
 
 	if (ss) {
+		struct cgroup *parent_cgrp;
+
 		/* css release path */
 		if (!list_empty(&css->rstat_css_node)) {
 			cgroup_rstat_flush(cgrp);
@@ -5433,6 +5463,14 @@ static void css_release_work_fn(struct work_struct *work)
 		cgroup_idr_replace(&ss->css_idr, NULL, css->id);
 		if (ss->css_released)
 			ss->css_released(css);
+
+		cgrp->nr_dying_subsys[ss->id]--;
+		WARN_ON_ONCE(css->nr_descendants || cgrp->nr_dying_subsys[ss->id]);
+		parent_cgrp = cgroup_parent(cgrp);
+		while (parent_cgrp) {
+			parent_cgrp->nr_dying_subsys[ss->id]--;
+			parent_cgrp = cgroup_parent(parent_cgrp);
+		}
 	} else {
 		struct cgroup *tcgrp;
 
@@ -5517,8 +5555,11 @@ static int online_css(struct cgroup_subsys_state *css)
 		rcu_assign_pointer(css->cgroup->subsys[ss->id], css);
 
 		atomic_inc(&css->online_cnt);
-		if (css->parent)
+		if (css->parent) {
 			atomic_inc(&css->parent->online_cnt);
+			while ((css = css->parent))
+				css->nr_descendants++;
+		}
 	}
 	return ret;
 }
@@ -5540,6 +5581,16 @@ static void offline_css(struct cgroup_subsys_state *css)
 	RCU_INIT_POINTER(css->cgroup->subsys[ss->id], NULL);
 
 	wake_up_all(&css->cgroup->offline_waitq);
+
+	css->cgroup->nr_dying_subsys[ss->id]++;
+	/*
+	 * Parent css and cgroup cannot be freed until after the freeing
+	 * of child css, see css_free_rwork_fn().
+	 */
+	while ((css = css->parent)) {
+		css->nr_descendants--;
+		css->cgroup->nr_dying_subsys[ss->id]++;
+	}
 }
 
 /**