inotify, memcg: account inotify instances to kmemcg

Currently the fs sysctl inotify/max_user_instances is used to limit the
number of inotify instances on the system. For systems running multiple
workloads, the per-user namespace sysctl max_inotify_instances can be
used to further partition inotify instances. However there is no easy
way to set a sensible system level max limit on inotify instances and
further partition it between the workloads. It is much easier to charge
the underlying resource (i.e. memory) behind the inotify instances to
the memcg of the workload and let their memory limits limit the number
of inotify instances they can create.

With inotify instances charged to memcg, the admin can simply set
max_user_instances to INT_MAX and let the memcg limits of the jobs limit
their inotify instances.

Link: https://lore.kernel.org/r/20201220044608.1258123-1-shakeelb@google.com
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Shakeel Butt <shakeelb@google.com>
Signed-off-by: Jan Kara <jack@suse.cz>

fs/notify/fanotify/fanotify_user.c

@@ -976,7 +976,7 @@ SYSCALL_DEFINE2(fanotify_init, unsigned int, flags, unsigned int, event_f_flags)
 		f_flags |= O_NONBLOCK;
 
 	/* fsnotify_alloc_group takes a ref.  Dropped in fanotify_release */
-	group = fsnotify_alloc_group(&fanotify_fsnotify_ops);
+	group = fsnotify_alloc_user_group(&fanotify_fsnotify_ops);
 	if (IS_ERR(group)) {
 		free_uid(user);
 		return PTR_ERR(group);

fs/notify/group.c

@@ -111,14 +111,12 @@ void fsnotify_put_group(struct fsnotify_group *group)
 }
 EXPORT_SYMBOL_GPL(fsnotify_put_group);
 
-/*
- * Create a new fsnotify_group and hold a reference for the group returned.
- */
-struct fsnotify_group *fsnotify_alloc_group(const struct fsnotify_ops *ops)
+static struct fsnotify_group *__fsnotify_alloc_group(
+				const struct fsnotify_ops *ops, gfp_t gfp)
 {
 	struct fsnotify_group *group;
 
-	group = kzalloc(sizeof(struct fsnotify_group), GFP_KERNEL);
+	group = kzalloc(sizeof(struct fsnotify_group), gfp);
 	if (!group)
 		return ERR_PTR(-ENOMEM);
 
@@ -139,8 +137,25 @@ struct fsnotify_group *fsnotify_alloc_group(const struct fsnotify_ops *ops)
 
 	return group;
 }
+
+/*
+ * Create a new fsnotify_group and hold a reference for the group returned.
+ */
+struct fsnotify_group *fsnotify_alloc_group(const struct fsnotify_ops *ops)
+{
+	return __fsnotify_alloc_group(ops, GFP_KERNEL);
+}
 EXPORT_SYMBOL_GPL(fsnotify_alloc_group);
 
+/*
+ * Create a new fsnotify_group and hold a reference for the group returned.
+ */
+struct fsnotify_group *fsnotify_alloc_user_group(const struct fsnotify_ops *ops)
+{
+	return __fsnotify_alloc_group(ops, GFP_KERNEL_ACCOUNT);
+}
+EXPORT_SYMBOL_GPL(fsnotify_alloc_user_group);
+
 int fsnotify_fasync(int fd, struct file *file, int on)
 {
 	struct fsnotify_group *group = file->private_data;

fs/notify/inotify/inotify_user.c

@@ -632,11 +632,11 @@ static struct fsnotify_group *inotify_new_group(unsigned int max_events)
 	struct fsnotify_group *group;
 	struct inotify_event_info *oevent;
 
-	group = fsnotify_alloc_group(&inotify_fsnotify_ops);
+	group = fsnotify_alloc_user_group(&inotify_fsnotify_ops);
 	if (IS_ERR(group))
 		return group;
 
-	oevent = kmalloc(sizeof(struct inotify_event_info), GFP_KERNEL);
+	oevent = kmalloc(sizeof(struct inotify_event_info), GFP_KERNEL_ACCOUNT);
 	if (unlikely(!oevent)) {
 		fsnotify_destroy_group(group);
 		return ERR_PTR(-ENOMEM);

include/linux/fsnotify_backend.h

@@ -470,6 +470,7 @@ static inline void fsnotify_update_flags(struct dentry *dentry)
 
 /* create a new group */
 extern struct fsnotify_group *fsnotify_alloc_group(const struct fsnotify_ops *ops);
+extern struct fsnotify_group *fsnotify_alloc_user_group(const struct fsnotify_ops *ops);
 /* get reference to a group */
 extern void fsnotify_get_group(struct fsnotify_group *group);
 /* drop reference on a group from fsnotify_alloc_group */