Merge branch 'for-5.2-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup

Pull cgroup fixes from Tejun Heo:
 "This has an unusually high density of tricky fixes:

   - task_get_css() could deadlock when it races against a dying cgroup.

   - cgroup.procs didn't list thread group leaders with live threads.

     This could mislead readers to think that a cgroup is empty when
     it's not. Fixed by making PROCS iterator include dead tasks. I made
     a couple mistakes making this change and this pull request contains
     a couple follow-up patches.

   - When cpusets run out of online cpus, it updates cpusmasks of member
     tasks in bizarre ways. Joel improved the behavior significantly"

* 'for-5.2-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup:
  cpuset: restore sanity to cpuset_cpus_allowed_fallback()
  cgroup: Fix css_task_iter_advance_css_set() cset skip condition
  cgroup: css_task_iter_skip()'d iterators must be advanced before accessed
  cgroup: Include dying leaders with live threads in PROCS iterations
  cgroup: Implement css_task_iter_skip()
  cgroup: Call cgroup_release() before __exit_signal()
  docs cgroups: add another example size for hugetlb
  cgroup: Use css_tryget() instead of css_tryget_online() in task_get_css()

2 files changed, 13 insertions, 2 deletions

diff --git a/include/linux/cgroup-defs.h b/include/linux/cgroup-defs.h
index 11e215d7937e..b4e766e93f6e 100644
--- a/include/linux/cgroup-defs.h
+++ b/include/linux/cgroup-defs.h
@@ -221,6 +221,7 @@ 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 c0077adeea83..0297f930a56e 100644
--- a/include/linux/cgroup.h
+++ b/include/linux/cgroup.h
@@ -43,6 +43,9 @@
 /* walk all threaded css_sets in the domain */
 #define CSS_TASK_ITER_THREADED		(1U << 1)
 
+/* internal flags */
+#define CSS_TASK_ITER_SKIPPED		(1U << 16)
+
 /* a css_task_iter should be treated as an opaque object */
 struct css_task_iter {
 	struct cgroup_subsys		*ss;
@@ -57,6 +60,7 @@ 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;
@@ -487,7 +491,7 @@ static inline struct cgroup_subsys_state *task_css(struct task_struct *task,
  *
  * Find the css for the (@task, @subsys_id) combination, increment a
  * reference on and return it.  This function is guaranteed to return a
- * valid css.
+ * valid css.  The returned css may already have been offlined.
  */
 static inline struct cgroup_subsys_state *
 task_get_css(struct task_struct *task, int subsys_id)
@@ -497,7 +501,13 @@ task_get_css(struct task_struct *task, int subsys_id)
 	rcu_read_lock();
 	while (true) {
 		css = task_css(task, subsys_id);
-		if (likely(css_tryget_online(css)))
+		/*
+		 * Can't use css_tryget_online() here.  A task which has
+		 * PF_EXITING set may stay associated with an offline css.
+		 * If such task calls this function, css_tryget_online()
+		 * will keep failing.
+		 */
+		if (likely(css_tryget(css)))
 			break;
 		cpu_relax();
 	}