memcg: replace cgroup_lock with memcg specific memcg_lock

After the preparation work done in earlier patches, the cgroup_lock can
be trivially replaced with a memcg-specific lock.  This is an automatic
translation at every site where the values involved were queried.

The sites where values are written, however, used to be naturally called
under cgroup_lock.  This is the case for instance in the css_online
callback.  For those, we now need to explicitly add the memcg lock.

With this, all the calls to cgroup_lock outside cgroup core are gone.

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>

1 files changed, 21 insertions, 18 deletions

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 2bc3fbe93154..46cdaef78b01 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -488,6 +488,13 @@ enum res_type {
 #define MEM_CGROUP_RECLAIM_SHRINK_BIT   0x1
 #define MEM_CGROUP_RECLAIM_SHRINK       (1 << MEM_CGROUP_RECLAIM_SHRINK_BIT)
 
+/*
+ * The memcg_create_mutex will be held whenever a new cgroup is created.
+ * As a consequence, any change that needs to protect against new child cgroups
+ * appearing has to hold it as well.
+ */
+static DEFINE_MUTEX(memcg_create_mutex);
+
 static void mem_cgroup_get(struct mem_cgroup *memcg);
 static void mem_cgroup_put(struct mem_cgroup *memcg);
 
@@ -4778,8 +4785,8 @@ static inline bool __memcg_has_children(struct mem_cgroup *memcg)
 }
 
 /*
- * 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
+ * Must be called with memcg_create_mutex held, unless the cgroup is guaranteed
+ * to be already dead (as in mem_cgroup_force_empty, for instance).  This is
  * 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.
@@ -4859,7 +4866,7 @@ static int mem_cgroup_hierarchy_write(struct cgroup *cont, struct cftype *cft,
         if (parent)
                 parent_memcg = mem_cgroup_from_cont(parent);
 
-        cgroup_lock();
+        mutex_lock(&memcg_create_mutex);
 
         if (memcg->use_hierarchy == val)
                 goto out;
@@ -4882,7 +4889,7 @@ static int mem_cgroup_hierarchy_write(struct cgroup *cont, struct cftype *cft,
                 retval = -EINVAL;
 
 out:
-        cgroup_unlock();
+        mutex_unlock(&memcg_create_mutex);
 
         return retval;
 }
@@ -4981,14 +4988,8 @@ static int memcg_update_kmem_limit(struct cgroup *cont, u64 val)
          *
          * After it first became limited, changes in the value of the limit are
          * of course permitted.
-         *
-         * Taking the cgroup_lock is really offensive, but it is so far the only
-         * way to guarantee that no children will appear. There are plenty of
-         * other offenders, and they should all go away. Fine grained locking
-         * is probably the way to go here. When we are fully hierarchical, we
-         * can also get rid of the use_hierarchy check.
          */
-        cgroup_lock();
+        mutex_lock(&memcg_create_mutex);
         mutex_lock(&set_limit_mutex);
         if (!memcg->kmem_account_flags && val != RESOURCE_MAX) {
                 if (cgroup_task_count(cont) || memcg_has_children(memcg)) {
@@ -5015,7 +5016,7 @@ static int memcg_update_kmem_limit(struct cgroup *cont, u64 val)
                 ret = res_counter_set_limit(&memcg->kmem, val);
 out:
         mutex_unlock(&set_limit_mutex);
-        cgroup_unlock();
+        mutex_unlock(&memcg_create_mutex);
 
         /*
          * We are by now familiar with the fact that we can't inc the static
@@ -5396,17 +5397,17 @@ static int mem_cgroup_swappiness_write(struct cgroup *cgrp, struct cftype *cft,
 
         parent = mem_cgroup_from_cont(cgrp->parent);
 
-        cgroup_lock();
+        mutex_lock(&memcg_create_mutex);
 
         /* If under hierarchy, only empty-root can set this value */
         if ((parent->use_hierarchy) || memcg_has_children(memcg)) {
-                cgroup_unlock();
+                mutex_unlock(&memcg_create_mutex);
                 return -EINVAL;
         }
 
         memcg->swappiness = val;
 
-        cgroup_unlock();
+        mutex_unlock(&memcg_create_mutex);
 
         return 0;
 }
@@ -5732,16 +5733,16 @@ static int mem_cgroup_oom_control_write(struct cgroup *cgrp,
 
         parent = mem_cgroup_from_cont(cgrp->parent);
 
-        cgroup_lock();
+        mutex_lock(&memcg_create_mutex);
         /* oom-kill-disable is a flag for subhierarchy. */
         if ((parent->use_hierarchy) || memcg_has_children(memcg)) {
-                cgroup_unlock();
+                mutex_unlock(&memcg_create_mutex);
                 return -EINVAL;
         }
         memcg->oom_kill_disable = val;
         if (!val)
                 memcg_oom_recover(memcg);
-        cgroup_unlock();
+        mutex_unlock(&memcg_create_mutex);
         return 0;
 }
 
@@ -6170,6 +6171,7 @@ mem_cgroup_css_online(struct cgroup *cont)
         if (!cont->parent)
                 return 0;
 
+        mutex_lock(&memcg_create_mutex);
         memcg = mem_cgroup_from_cont(cont);
         parent = mem_cgroup_from_cont(cont->parent);
 
@@ -6203,6 +6205,7 @@ mem_cgroup_css_online(struct cgroup *cont)
         }
 
         error = memcg_init_kmem(memcg, &mem_cgroup_subsys);
+        mutex_unlock(&memcg_create_mutex);
         if (error) {
                 /*
                  * We call put now because our (and parent's) refcnts