Merge branch 'for-5.1' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup

Pull cgroup updates from Tejun Heo:

 - Oleg's pids controller accounting update which gets rid of rcu delay
   in pids accounting updates

 - rstat (cgroup hierarchical stat collection mechanism) optimization

 - Doc updates

* 'for-5.1' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup:
  cpuset: remove unused task_has_mempolicy()
  cgroup, rstat: Don't flush subtree root unless necessary
  cgroup: add documentation for pids.events file
  Documentation: cgroup-v2: eliminate markup warnings
  MAINTAINERS: Update cgroup entry
  cgroup/pids: turn cgroup_subsys->free() into cgroup_subsys->release() to fix the accounting

Documentation/admin-guide/cgroup-v2.rst

@@ -1519,7 +1519,7 @@ protected workload.
 
 The limits are only applied at the peer level in the hierarchy.  This means that
 in the diagram below, only groups A, B, and C will influence each other, and
-groups D and F will influence each other.  Group G will influence nobody.
+groups D and F will influence each other.  Group G will influence nobody::
 
 			[root]
 		/	   |		\

Documentation/cgroup-v1/pids.txt

@@ -33,6 +33,9 @@ limit in the hierarchy is followed).
 pids.current tracks all child cgroup hierarchies, so parent/pids.current is a
 superset of parent/child/pids.current.
 
+The pids.events file contains event counters:
+  - max: Number of times fork failed because limit was hit.
+
 Example
 -------
 

MAINTAINERS

@@ -3970,9 +3970,10 @@ M:	Johannes Weiner <hannes@cmpxchg.org>
 L:	cgroups@vger.kernel.org
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup.git
 S:	Maintained
-F:	Documentation/cgroup*
+F:	Documentation/admin-guide/cgroup-v2.rst
+F:	Documentation/cgroup-v1/
 F:	include/linux/cgroup*
-F:	kernel/cgroup*
+F:	kernel/cgroup/
 
 CONTROL GROUP - CPUSET
 M:	Li Zefan <lizefan@huawei.com>

include/linux/cgroup-defs.h

@@ -606,7 +606,7 @@ struct cgroup_subsys {
 	void (*cancel_fork)(struct task_struct *task);
 	void (*fork)(struct task_struct *task);
 	void (*exit)(struct task_struct *task);
-	void (*free)(struct task_struct *task);
+	void (*release)(struct task_struct *task);
 	void (*bind)(struct cgroup_subsys_state *root_css);
 
 	bool early_init:1;

include/linux/cgroup.h

@@ -121,6 +121,7 @@ extern int cgroup_can_fork(struct task_struct *p);
 extern void cgroup_cancel_fork(struct task_struct *p);
 extern void cgroup_post_fork(struct task_struct *p);
 void cgroup_exit(struct task_struct *p);
+void cgroup_release(struct task_struct *p);
 void cgroup_free(struct task_struct *p);
 
 int cgroup_init_early(void);
@@ -697,6 +698,7 @@ static inline int cgroup_can_fork(struct task_struct *p) { return 0; }
 static inline void cgroup_cancel_fork(struct task_struct *p) {}
 static inline void cgroup_post_fork(struct task_struct *p) {}
 static inline void cgroup_exit(struct task_struct *p) {}
+static inline void cgroup_release(struct task_struct *p) {}
 static inline void cgroup_free(struct task_struct *p) {}
 
 static inline int cgroup_init_early(void) { return 0; }

kernel/cgroup/cgroup.c

@@ -197,7 +197,7 @@ static u64 css_serial_nr_next = 1;
  */
 static u16 have_fork_callback __read_mostly;
 static u16 have_exit_callback __read_mostly;
-static u16 have_free_callback __read_mostly;
+static u16 have_release_callback __read_mostly;
 static u16 have_canfork_callback __read_mostly;
 
 /* cgroup namespace for init task */
@@ -5326,7 +5326,7 @@ static void __init cgroup_init_subsys(struct cgroup_subsys *ss, bool early)
 
 	have_fork_callback |= (bool)ss->fork << ss->id;
 	have_exit_callback |= (bool)ss->exit << ss->id;
-	have_free_callback |= (bool)ss->free << ss->id;
+	have_release_callback |= (bool)ss->release << ss->id;
 	have_canfork_callback |= (bool)ss->can_fork << ss->id;
 
 	/* At system boot, before all subsystems have been
@@ -5762,16 +5762,19 @@ void cgroup_exit(struct task_struct *tsk)
 	} while_each_subsys_mask();
 }
 
-void cgroup_free(struct task_struct *task)
+void cgroup_release(struct task_struct *task)
 {
-	struct css_set *cset = task_css_set(task);
 	struct cgroup_subsys *ss;
 	int ssid;
 
-	do_each_subsys_mask(ss, ssid, have_free_callback) {
+	do_each_subsys_mask(ss, ssid, have_release_callback) {
-		ss->free(task);
+		ss->release(task);
 	} while_each_subsys_mask();
+}
 
+void cgroup_free(struct task_struct *task)
+{
+	struct css_set *cset = task_css_set(task);
 	put_css_set(cset);
 }
 

kernel/cgroup/cpuset.c

@@ -203,19 +203,6 @@ static inline struct cpuset *parent_cs(struct cpuset *cs)
 	return css_cs(cs->css.parent);
 }
 
-#ifdef CONFIG_NUMA
-static inline bool task_has_mempolicy(struct task_struct *task)
-{
-	return task->mempolicy;
-}
-#else
-static inline bool task_has_mempolicy(struct task_struct *task)
-{
-	return false;
-}
-#endif
-
-
 /* bits in struct cpuset flags field */
 typedef enum {
 	CS_ONLINE,

kernel/cgroup/pids.c

@@ -247,7 +247,7 @@ static void pids_cancel_fork(struct task_struct *task)
 	pids_uncharge(pids, 1);
 }
 
-static void pids_free(struct task_struct *task)
+static void pids_release(struct task_struct *task)
 {
 	struct pids_cgroup *pids = css_pids(task_css(task, pids_cgrp_id));
 
@@ -342,7 +342,7 @@ struct cgroup_subsys pids_cgrp_subsys = {
 	.cancel_attach 	= pids_cancel_attach,
 	.can_fork	= pids_can_fork,
 	.cancel_fork	= pids_cancel_fork,
-	.free		= pids_free,
+	.release	= pids_release,
 	.legacy_cftypes	= pids_files,
 	.dfl_cftypes	= pids_files,
 	.threaded	= true,

kernel/cgroup/rstat.c

@@ -87,7 +87,6 @@ static struct cgroup *cgroup_rstat_cpu_pop_updated(struct cgroup *pos,
 						   struct cgroup *root, int cpu)
 {
 	struct cgroup_rstat_cpu *rstatc;
-	struct cgroup *parent;
 
 	if (pos == root)
 		return NULL;
@@ -115,8 +114,8 @@ static struct cgroup *cgroup_rstat_cpu_pop_updated(struct cgroup *pos,
 	 * However, due to the way we traverse, @pos will be the first
 	 * child in most cases. The only exception is @root.
 	 */
-	parent = cgroup_parent(pos);
+	if (rstatc->updated_next) {
-	if (parent && rstatc->updated_next) {
+		struct cgroup *parent = cgroup_parent(pos);
 		struct cgroup_rstat_cpu *prstatc = cgroup_rstat_cpu(parent, cpu);
 		struct cgroup_rstat_cpu *nrstatc;
 		struct cgroup **nextp;
@@ -140,11 +139,14 @@ static struct cgroup *cgroup_rstat_cpu_pop_updated(struct cgroup *pos,
 		 * updated stat.
 		 */
 		smp_mb();
-	}
 
 		return pos;
 	}
 
+	/* only happens for @root */
+	return NULL;
+}
+
 /* see cgroup_rstat_flush() */
 static void cgroup_rstat_flush_locked(struct cgroup *cgrp, bool may_sleep)
 	__releases(&cgroup_rstat_lock) __acquires(&cgroup_rstat_lock)

kernel/exit.c

@@ -219,6 +219,7 @@ repeat:
 	}
 
 	write_unlock_irq(&tasklist_lock);
+	cgroup_release(p);
 	release_thread(p);
 	call_rcu(&p->rcu, delayed_put_task_struct);