memcg: fast hierarchy-aware child test

Currently, we use cgroups' provided list of children to verify if it is
safe to proceed with any value change that is dependent on the cgroup
being empty.

This is less than ideal, because it enforces a dependency over cgroup
core that we would be better off without.  The solution proposed here is
to iterate over the child cgroups and if any is found that is already
online, we bounce and return: we don't really care how many children we
have, only if we have any.

This is also made to be hierarchy aware.  IOW, cgroups with hierarchy
disabled, while they still exist, will be considered for the purpose of
this interface as having no children.

[akpm@linux-foundation.org: tweak comments]
Signed-off-by: Glauber Costa <glommer@parallels.com>
Acked-by: Michal Hocko <mhocko@suse.cz>
Cc: Tejun Heo <tj@kernel.org>
Cc: Hiroyuki Kamezawa <kamezawa.hiroyuki@gmail.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index a931896..2bc3fbe 100644 (file)
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -4763,6 +4763,33 @@ static void mem_cgroup_reparent_charges(struct mem_cgroup *memcg)
 }
 
 /*
+ * This mainly exists for tests during the setting of set of use_hierarchy.
+ * Since this is the very setting we are changing, the current hierarchy value
+ * is meaningless
+ */
+static inline bool __memcg_has_children(struct mem_cgroup *memcg)
+{
+       struct cgroup *pos;
+
+       /* bounce at first found */
+       cgroup_for_each_child(pos, memcg->css.cgroup)
+               return true;
+       return false;
+}
+
+/*
+ * Must be called with cgroup_lock held, unless the cgroup is guaranteed to be
+ * already dead (in mem_cgroup_force_empty(), for instance).  This is different
+ * from mem_cgroup_count_children(), in the sense that we don't really care how
+ * many children we have; we only need to know if we have any.  It also counts
+ * any memcg without hierarchy as infertile.
+ */
+static inline bool memcg_has_children(struct mem_cgroup *memcg)
+{
+       return memcg->use_hierarchy && __memcg_has_children(memcg);
+}
+
+/*
  * Reclaims as many pages from the given memcg as possible and moves
  * the rest to the parent.
  *
@@ -4847,7 +4874,7 @@ static int mem_cgroup_hierarchy_write(struct cgroup *cont, struct cftype *cft,
         */
        if ((!parent_memcg || !parent_memcg->use_hierarchy) &&
                                (val == 1 || val == 0)) {
-               if (list_empty(&cont->children))
+               if (!__memcg_has_children(memcg))
                        memcg->use_hierarchy = val;
                else
                        retval = -EBUSY;
@@ -4964,8 +4991,7 @@ static int memcg_update_kmem_limit(struct cgroup *cont, u64 val)
        cgroup_lock();
        mutex_lock(&set_limit_mutex);
        if (!memcg->kmem_account_flags && val != RESOURCE_MAX) {
-               if (cgroup_task_count(cont) || (memcg->use_hierarchy &&
-                                               !list_empty(&cont->children))) {
+               if (cgroup_task_count(cont) || memcg_has_children(memcg)) {
                        ret = -EBUSY;
                        goto out;
                }
@@ -5373,8 +5399,7 @@ static int mem_cgroup_swappiness_write(struct cgroup *cgrp, struct cftype *cft,
        cgroup_lock();
 
        /* If under hierarchy, only empty-root can set this value */
-       if ((parent->use_hierarchy) ||
-           (memcg->use_hierarchy && !list_empty(&cgrp->children))) {
+       if ((parent->use_hierarchy) || memcg_has_children(memcg)) {
                cgroup_unlock();
                return -EINVAL;
        }
@@ -5709,8 +5734,7 @@ static int mem_cgroup_oom_control_write(struct cgroup *cgrp,
 
        cgroup_lock();
        /* oom-kill-disable is a flag for subhierarchy. */
-       if ((parent->use_hierarchy) ||
-           (memcg->use_hierarchy && !list_empty(&cgrp->children))) {
+       if ((parent->use_hierarchy) || memcg_has_children(memcg)) {
                cgroup_unlock();
                return -EINVAL;
        }