aboutsummaryrefslogtreecommitdiffstats
path: root/fs/notify/group.c
diff options
context:
space:
mode:
authorEric Paris <eparis@redhat.com>2009-05-21 17:01:20 -0400
committerEric Paris <eparis@redhat.com>2009-06-11 14:57:52 -0400
commit90586523eb4b349806887c62ee70685a49415124 (patch)
tree2ba6da637f897bbb2309b141b81801e4151d87b0 /fs/notify/group.c
parentc9059598ea8981d02356eead3188bf7fa4d717b8 (diff)
fsnotify: unified filesystem notification backend
fsnotify is a backend for filesystem notification. fsnotify does not provide any userspace interface but does provide the basis needed for other notification schemes such as dnotify. fsnotify can be extended to be the backend for inotify or the upcoming fanotify. fsnotify provides a mechanism for "groups" to register for some set of filesystem events and to then deliver those events to those groups for processing. fsnotify has a number of benefits, the first being actually shrinking the size of an inode. Before fsnotify to support both dnotify and inotify an inode had unsigned long i_dnotify_mask; /* Directory notify events */ struct dnotify_struct *i_dnotify; /* for directory notifications */ struct list_head inotify_watches; /* watches on this inode */ struct mutex inotify_mutex; /* protects the watches list But with fsnotify this same functionallity (and more) is done with just __u32 i_fsnotify_mask; /* all events for this inode */ struct hlist_head i_fsnotify_mark_entries; /* marks on this inode */ That's right, inotify, dnotify, and fanotify all in 64 bits. We used that much space just in inotify_watches alone, before this patch set. fsnotify object lifetime and locking is MUCH better than what we have today. inotify locking is incredibly complex. See 8f7b0ba1c8539 as an example of what's been busted since inception. inotify needs to know internal semantics of superblock destruction and unmounting to function. The inode pinning and vfs contortions are horrible. no fsnotify implementers do allocation under locks. This means things like f04b30de3 which (due to an overabundance of caution) changes GFP_KERNEL to GFP_NOFS can be reverted. There are no longer any allocation rules when using or implementing your own fsnotify listener. fsnotify paves the way for fanotify. In brief fanotify is a notification mechanism that delivers the lisener both an 'event' and an open file descriptor to the object in question. This means that fanotify is pathname agnostic. Some on lkml may not care for the original companies or users that pushed for TALPA, but fanotify was designed with flexibility and input for other users in mind. The readahead group expressed interest in fanotify as it could be used to profile disk access on boot without breaking the audit system. The desktop search groups have also expressed interest in fanotify as it solves a number of the race conditions and problems present with managing inotify when more than a limited number of specific files are of interest. fanotify can provide for a userspace access control system which makes it a clean interface for AV vendors to hook without trying to do binary patching on the syscall table, LSM, and everywhere else they do their things today. With this patch series fanotify can be implemented in less than 1200 lines of easy to review code. Almost all of which is the socket based user interface. This patch series builds fsnotify to the point that it can implement dnotify and inotify_user. Patches exist and will be sent soon after acceptance to finish the in kernel inotify conversion (audit) and implement fanotify. Signed-off-by: Eric Paris <eparis@redhat.com> Acked-by: Al Viro <viro@zeniv.linux.org.uk> Cc: Christoph Hellwig <hch@lst.de>
Diffstat (limited to 'fs/notify/group.c')
-rw-r--r--fs/notify/group.c198
1 files changed, 198 insertions, 0 deletions
diff --git a/fs/notify/group.c b/fs/notify/group.c
new file mode 100644
index 00000000000..c6812953b96
--- /dev/null
+++ b/fs/notify/group.c
@@ -0,0 +1,198 @@
+/*
+ * Copyright (C) 2008 Red Hat, Inc., Eric Paris <eparis@redhat.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2, or (at your option)
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; see the file COPYING. If not, write to
+ * the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#include <linux/list.h>
+#include <linux/mutex.h>
+#include <linux/slab.h>
+#include <linux/srcu.h>
+#include <linux/rculist.h>
+#include <linux/wait.h>
+
+#include <linux/fsnotify_backend.h>
+#include "fsnotify.h"
+
+#include <asm/atomic.h>
+
+/* protects writes to fsnotify_groups and fsnotify_mask */
+static DEFINE_MUTEX(fsnotify_grp_mutex);
+/* protects reads while running the fsnotify_groups list */
+struct srcu_struct fsnotify_grp_srcu;
+/* all groups registered to receive filesystem notifications */
+LIST_HEAD(fsnotify_groups);
+/* bitwise OR of all events (FS_*) interesting to some group on this system */
+__u32 fsnotify_mask;
+
+/*
+ * When a new group registers or changes it's set of interesting events
+ * this function updates the fsnotify_mask to contain all interesting events
+ */
+void fsnotify_recalc_global_mask(void)
+{
+ struct fsnotify_group *group;
+ __u32 mask = 0;
+ int idx;
+
+ idx = srcu_read_lock(&fsnotify_grp_srcu);
+ list_for_each_entry_rcu(group, &fsnotify_groups, group_list)
+ mask |= group->mask;
+ srcu_read_unlock(&fsnotify_grp_srcu, idx);
+ fsnotify_mask = mask;
+}
+
+/*
+ * Take a reference to a group so things found under the fsnotify_grp_mutex
+ * can't get freed under us
+ */
+static void fsnotify_get_group(struct fsnotify_group *group)
+{
+ atomic_inc(&group->refcnt);
+}
+
+/*
+ * Final freeing of a group
+ */
+static void fsnotify_destroy_group(struct fsnotify_group *group)
+{
+ if (group->ops->free_group_priv)
+ group->ops->free_group_priv(group);
+
+ kfree(group);
+}
+
+/*
+ * Remove this group from the global list of groups that will get events
+ * this can be done even if there are still references and things still using
+ * this group. This just stops the group from getting new events.
+ */
+static void __fsnotify_evict_group(struct fsnotify_group *group)
+{
+ BUG_ON(!mutex_is_locked(&fsnotify_grp_mutex));
+
+ if (group->on_group_list)
+ list_del_rcu(&group->group_list);
+ group->on_group_list = 0;
+}
+
+/*
+ * Called when a group is no longer interested in getting events. This can be
+ * used if a group is misbehaving or if for some reason a group should no longer
+ * get any filesystem events.
+ */
+void fsnotify_evict_group(struct fsnotify_group *group)
+{
+ mutex_lock(&fsnotify_grp_mutex);
+ __fsnotify_evict_group(group);
+ mutex_unlock(&fsnotify_grp_mutex);
+}
+
+/*
+ * Drop a reference to a group. Free it if it's through.
+ */
+void fsnotify_put_group(struct fsnotify_group *group)
+{
+ if (!atomic_dec_and_mutex_lock(&group->refcnt, &fsnotify_grp_mutex))
+ return;
+
+ /*
+ * OK, now we know that there's no other users *and* we hold mutex,
+ * so no new references will appear
+ */
+ __fsnotify_evict_group(group);
+
+ /*
+ * now it's off the list, so the only thing we might care about is
+ * srcu access....
+ */
+ mutex_unlock(&fsnotify_grp_mutex);
+ synchronize_srcu(&fsnotify_grp_srcu);
+
+ /* and now it is really dead. _Nothing_ could be seeing it */
+ fsnotify_recalc_global_mask();
+ fsnotify_destroy_group(group);
+}
+
+/*
+ * Simply run the fsnotify_groups list and find a group which matches
+ * the given parameters. If a group is found we take a reference to that
+ * group.
+ */
+static struct fsnotify_group *fsnotify_find_group(unsigned int group_num, __u32 mask,
+ const struct fsnotify_ops *ops)
+{
+ struct fsnotify_group *group_iter;
+ struct fsnotify_group *group = NULL;
+
+ BUG_ON(!mutex_is_locked(&fsnotify_grp_mutex));
+
+ list_for_each_entry_rcu(group_iter, &fsnotify_groups, group_list) {
+ if (group_iter->group_num == group_num) {
+ if ((group_iter->mask == mask) &&
+ (group_iter->ops == ops)) {
+ fsnotify_get_group(group_iter);
+ group = group_iter;
+ } else
+ group = ERR_PTR(-EEXIST);
+ }
+ }
+ return group;
+}
+
+/*
+ * Either finds an existing group which matches the group_num, mask, and ops or
+ * creates a new group and adds it to the global group list. In either case we
+ * take a reference for the group returned.
+ */
+struct fsnotify_group *fsnotify_obtain_group(unsigned int group_num, __u32 mask,
+ const struct fsnotify_ops *ops)
+{
+ struct fsnotify_group *group, *tgroup;
+
+ /* very low use, simpler locking if we just always alloc */
+ group = kmalloc(sizeof(struct fsnotify_group), GFP_KERNEL);
+ if (!group)
+ return ERR_PTR(-ENOMEM);
+
+ atomic_set(&group->refcnt, 1);
+
+ group->on_group_list = 0;
+ group->group_num = group_num;
+ group->mask = mask;
+
+ group->ops = ops;
+
+ mutex_lock(&fsnotify_grp_mutex);
+ tgroup = fsnotify_find_group(group_num, mask, ops);
+ if (tgroup) {
+ /* group already exists */
+ mutex_unlock(&fsnotify_grp_mutex);
+ /* destroy the new one we made */
+ fsnotify_put_group(group);
+ return tgroup;
+ }
+
+ /* group not found, add a new one */
+ list_add_rcu(&group->group_list, &fsnotify_groups);
+ group->on_group_list = 1;
+
+ mutex_unlock(&fsnotify_grp_mutex);
+
+ if (mask)
+ fsnotify_recalc_global_mask();
+
+ return group;
+}