cgroup: Include dying leaders with live threads in PROCS iterations

commit c03cd7738a83b13739f00546166969342c8ff014 upstream.

CSS_TASK_ITER_PROCS currently iterates live group leaders; however,
this means that a process with dying leader and live threads will be
skipped.  IOW, cgroup.procs might be empty while cgroup.threads isn't,
which is confusing to say the least.

Fix it by making cset track dying tasks and include dying leaders with
live threads in PROCS iteration.

Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-and-tested-by: Topi Miettinen <toiwoton@gmail.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

3 files changed, 39 insertions, 7 deletions

diff --git a/include/linux/cgroup-defs.h b/include/linux/cgroup-defs.h
index eb396f71285f..8d4b92185a09 100644
--- a/include/linux/cgroup-defs.h
+++ b/include/linux/cgroup-defs.h
@@ -201,6 +201,7 @@ struct css_set {
 	 */
 	struct list_head tasks;
 	struct list_head mg_tasks;
+	struct list_head dying_tasks;
 
 	/* all css_task_iters currently walking this cset */
 	struct list_head task_iters;
diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h
index a5f6253b014d..0e21619f1c03 100644
--- a/include/linux/cgroup.h
+++ b/include/linux/cgroup.h
@@ -59,6 +59,7 @@ struct css_task_iter {
 	struct list_head		*task_pos;
 	struct list_head		*tasks_head;
 	struct list_head		*mg_tasks_head;
+	struct list_head		*dying_tasks_head;
 
 	struct css_set			*cur_cset;
 	struct css_set			*cur_dcset;
diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c
index c72f0727db2c..8eb37e732253 100644
--- a/kernel/cgroup/cgroup.c
+++ b/kernel/cgroup/cgroup.c
@@ -643,6 +643,7 @@ struct css_set init_css_set = {
 	.dom_cset		= &init_css_set,
 	.tasks			= LIST_HEAD_INIT(init_css_set.tasks),
 	.mg_tasks		= LIST_HEAD_INIT(init_css_set.mg_tasks),
+	.dying_tasks		= LIST_HEAD_INIT(init_css_set.dying_tasks),
 	.task_iters		= LIST_HEAD_INIT(init_css_set.task_iters),
 	.threaded_csets		= LIST_HEAD_INIT(init_css_set.threaded_csets),
 	.cgrp_links		= LIST_HEAD_INIT(init_css_set.cgrp_links),
@@ -1107,6 +1108,7 @@ static struct css_set *find_css_set(struct css_set *old_cset,
 	cset->dom_cset = cset;
 	INIT_LIST_HEAD(&cset->tasks);
 	INIT_LIST_HEAD(&cset->mg_tasks);
+	INIT_LIST_HEAD(&cset->dying_tasks);
 	INIT_LIST_HEAD(&cset->task_iters);
 	INIT_LIST_HEAD(&cset->threaded_csets);
 	INIT_HLIST_NODE(&cset->hlist);
@@ -4046,15 +4048,18 @@ static void css_task_iter_advance_css_set(struct css_task_iter *it)
 			it->task_pos = NULL;
 			return;
 		}
-	} while (!css_set_populated(cset));
+	} while (!css_set_populated(cset) && !list_empty(&cset->dying_tasks));
 
 	if (!list_empty(&cset->tasks))
 		it->task_pos = cset->tasks.next;
-	else
+	else if (!list_empty(&cset->mg_tasks))
 		it->task_pos = cset->mg_tasks.next;
+	else
+		it->task_pos = cset->dying_tasks.next;
 
 	it->tasks_head = &cset->tasks;
 	it->mg_tasks_head = &cset->mg_tasks;
+	it->dying_tasks_head = &cset->dying_tasks;
 
 	/*
 	 * We don't keep css_sets locked across iteration steps and thus
@@ -4093,6 +4098,8 @@ static void css_task_iter_skip(struct css_task_iter *it,
 
 static void css_task_iter_advance(struct css_task_iter *it)
 {
+	struct task_struct *task;
+
 	lockdep_assert_held(&css_set_lock);
 repeat:
 	if (it->task_pos) {
@@ -4109,17 +4116,32 @@ repeat:
 		if (it->task_pos == it->tasks_head)
 			it->task_pos = it->mg_tasks_head->next;
 		if (it->task_pos == it->mg_tasks_head)
+			it->task_pos = it->dying_tasks_head->next;
+		if (it->task_pos == it->dying_tasks_head)
 			css_task_iter_advance_css_set(it);
 	} else {
 		/* called from start, proceed to the first cset */
 		css_task_iter_advance_css_set(it);
 	}
 
-	/* if PROCS, skip over tasks which aren't group leaders */
-	if ((it->flags & CSS_TASK_ITER_PROCS) && it->task_pos &&
-	    !thread_group_leader(list_entry(it->task_pos, struct task_struct,
-					    cg_list)))
-		goto repeat;
+	if (!it->task_pos)
+		return;
+
+	task = list_entry(it->task_pos, struct task_struct, cg_list);
+
+	if (it->flags & CSS_TASK_ITER_PROCS) {
+		/* if PROCS, skip over tasks which aren't group leaders */
+		if (!thread_group_leader(task))
+			goto repeat;
+
+		/* and dying leaders w/o live member threads */
+		if (!atomic_read(&task->signal->live))
+			goto repeat;
+	} else {
+		/* skip all dying ones */
+		if (task->flags & PF_EXITING)
+			goto repeat;
+	}
 }
 
 /**
@@ -5552,6 +5574,7 @@ void cgroup_exit(struct task_struct *tsk)
 	if (!list_empty(&tsk->cg_list)) {
 		spin_lock_irq(&css_set_lock);
 		css_set_move_task(tsk, cset, NULL, false);
+		list_add_tail(&tsk->cg_list, &cset->dying_tasks);
 		cset->nr_tasks--;
 		spin_unlock_irq(&css_set_lock);
 	} else {
@@ -5572,6 +5595,13 @@ void cgroup_release(struct task_struct *task)
 	do_each_subsys_mask(ss, ssid, have_release_callback) {
 		ss->release(task);
 	} while_each_subsys_mask();
+
+	if (use_task_css_set_links) {
+		spin_lock_irq(&css_set_lock);
+		css_set_skip_task_iters(task_css_set(task), task);
+		list_del_init(&task->cg_list);
+		spin_unlock_irq(&css_set_lock);
+	}
 }
 
 void cgroup_free(struct task_struct *task)