1 files changed, 166 insertions, 35 deletions

diff --git a/fs/notify/fsnotify.c b/fs/notify/fsnotify.c
index fcc2f064af83..4d2a82c1ceb1 100644
--- a/fs/notify/fsnotify.c
+++ b/fs/notify/fsnotify.c
@@ -21,6 +21,7 @@
 #include <linux/gfp.h>
 #include <linux/init.h>
 #include <linux/module.h>
+#include <linux/mount.h>
 #include <linux/srcu.h>
 
 #include <linux/fsnotify_backend.h>
@@ -35,6 +36,11 @@ void __fsnotify_inode_delete(struct inode *inode)
 }
 EXPORT_SYMBOL_GPL(__fsnotify_inode_delete);
 
+void __fsnotify_vfsmount_delete(struct vfsmount *mnt)
+{
+        fsnotify_clear_marks_by_mount(mnt);
+}
+
 /*
  * Given an inode, first check if we care what happens to our children.  Inotify
  * and dnotify both tell their parents about events.  If we care about any event
@@ -78,13 +84,16 @@ void __fsnotify_update_child_dentry_flags(struct inode *inode)
 }
 
 /* Notify this dentry's parent about a child's events. */
-void __fsnotify_parent(struct dentry *dentry, __u32 mask)
+void __fsnotify_parent(struct file *file, struct dentry *dentry, __u32 mask)
 {
         struct dentry *parent;
         struct inode *p_inode;
         bool send = false;
         bool should_update_children = false;
 
+        if (!dentry)
+                dentry = file->f_path.dentry;
+
         if (!(dentry->d_flags & DCACHE_FSNOTIFY_PARENT_WATCHED))
                 return;
 
@@ -115,8 +124,12 @@ void __fsnotify_parent(struct dentry *dentry, __u32 mask)
                  * specifies these are events which came from a child. */
                 mask |= FS_EVENT_ON_CHILD;
 
-                fsnotify(p_inode, mask, dentry->d_inode, FSNOTIFY_EVENT_INODE,
-                         dentry->d_name.name, 0);
+                if (file)
+                        fsnotify(p_inode, mask, file, FSNOTIFY_EVENT_FILE,
+                                 dentry->d_name.name, 0);
+                else
+                        fsnotify(p_inode, mask, dentry->d_inode, FSNOTIFY_EVENT_INODE,
+                                 dentry->d_name.name, 0);
                 dput(parent);
         }
 
@@ -127,63 +140,181 @@ void __fsnotify_parent(struct dentry *dentry, __u32 mask)
 }
 EXPORT_SYMBOL_GPL(__fsnotify_parent);
 
+static int send_to_group(struct inode *to_tell, struct vfsmount *mnt,
+                         struct fsnotify_mark *inode_mark,
+                         struct fsnotify_mark *vfsmount_mark,
+                         __u32 mask, void *data,
+                         int data_is, u32 cookie,
+                         const unsigned char *file_name,
+                         struct fsnotify_event **event)
+{
+        struct fsnotify_group *group = inode_mark->group;
+        __u32 inode_test_mask = (mask & ~FS_EVENT_ON_CHILD);
+        __u32 vfsmount_test_mask = (mask & ~FS_EVENT_ON_CHILD);
+
+        pr_debug("%s: group=%p to_tell=%p mnt=%p mark=%p mask=%x data=%p"
+                 " data_is=%d cookie=%d event=%p\n", __func__, group, to_tell,
+                 mnt, inode_mark, mask, data, data_is, cookie, *event);
+
+        /* clear ignored on inode modification */
+        if (mask & FS_MODIFY) {
+                if (inode_mark &&
+                    !(inode_mark->flags & FSNOTIFY_MARK_FLAG_IGNORED_SURV_MODIFY))
+                        inode_mark->ignored_mask = 0;
+                if (vfsmount_mark &&
+                    !(vfsmount_mark->flags & FSNOTIFY_MARK_FLAG_IGNORED_SURV_MODIFY))
+                        vfsmount_mark->ignored_mask = 0;
+        }
+
+        /* does the inode mark tell us to do something? */
+        if (inode_mark) {
+                inode_test_mask &= inode_mark->mask;
+                inode_test_mask &= ~inode_mark->ignored_mask;
+        }
+
+        /* does the vfsmount_mark tell us to do something? */
+        if (vfsmount_mark) {
+                vfsmount_test_mask &= vfsmount_mark->mask;
+                vfsmount_test_mask &= ~vfsmount_mark->ignored_mask;
+                if (inode_mark)
+                        vfsmount_test_mask &= ~inode_mark->ignored_mask;
+        }
+
+        if (!inode_test_mask && !vfsmount_test_mask)
+                return 0;
+
+        if (group->ops->should_send_event(group, to_tell, inode_mark,
+                                          vfsmount_mark, mask, data,
+                                          data_is) == false)
+                return 0;
+
+        if (!*event) {
+                *event = fsnotify_create_event(to_tell, mask, data,
+                                                data_is, file_name,
+                                                cookie, GFP_KERNEL);
+                if (!*event)
+                        return -ENOMEM;
+        }
+        return group->ops->handle_event(group, inode_mark, vfsmount_mark, *event);
+}
+
 /*
  * This is the main call to fsnotify.  The VFS calls into hook specific functions
  * in linux/fsnotify.h.  Those functions then in turn call here.  Here will call
  * out to all of the registered fsnotify_group.  Those groups can then use the
  * notification event in whatever means they feel necessary.
  */
-void fsnotify(struct inode *to_tell, __u32 mask, void *data, int data_is, const char *file_name, u32 cookie)
+int fsnotify(struct inode *to_tell, __u32 mask, void *data, int data_is,
+             const unsigned char *file_name, u32 cookie)
 {
-        struct fsnotify_group *group;
+        struct hlist_node *inode_node, *vfsmount_node;
+        struct fsnotify_mark *inode_mark = NULL, *vfsmount_mark = NULL;
+        struct fsnotify_group *inode_group, *vfsmount_group;
         struct fsnotify_event *event = NULL;
-        int idx;
+        struct vfsmount *mnt;
+        int idx, ret = 0;
+        bool used_inode = false, used_vfsmount = false;
         /* global tests shouldn't care about events on child only the specific event */
         __u32 test_mask = (mask & ~FS_EVENT_ON_CHILD);
 
-        if (list_empty(&fsnotify_groups))
-                return;
+        if (data_is == FSNOTIFY_EVENT_FILE)
+                mnt = ((struct file *)data)->f_path.mnt;
+        else
+                mnt = NULL;
 
-        if (!(test_mask & fsnotify_mask))
-                return;
-
-        if (!(test_mask & to_tell->i_fsnotify_mask))
-                return;
         /*
-         * SRCU!!  the groups list is very very much read only and the path is
-         * very hot.  The VAST majority of events are not going to need to do
-         * anything other than walk the list so it's crazy to pre-allocate.
+         * if this is a modify event we may need to clear the ignored masks
+         * otherwise return if neither the inode nor the vfsmount care about
+         * this type of event.
          */
-        idx = srcu_read_lock(&fsnotify_grp_srcu);
-        list_for_each_entry_rcu(group, &fsnotify_groups, group_list) {
-                if (test_mask & group->mask) {
-                        if (!group->ops->should_send_event(group, to_tell, mask))
-                                continue;
-                        if (!event) {
-                                event = fsnotify_create_event(to_tell, mask, data,
-                                                              data_is, file_name, cookie,
-                                                              GFP_KERNEL);
-                                /* shit, we OOM'd and now we can't tell, maybe
-                                 * someday someone else will want to do something
-                                 * here */
-                                if (!event)
-                                        break;
-                        }
-                        group->ops->handle_event(group, event);
+        if (!(mask & FS_MODIFY) &&
+            !(test_mask & to_tell->i_fsnotify_mask) &&
+            !(mnt && test_mask & mnt->mnt_fsnotify_mask))
+                return 0;
+
+        idx = srcu_read_lock(&fsnotify_mark_srcu);
+
+        if ((mask & FS_MODIFY) ||
+            (test_mask & to_tell->i_fsnotify_mask))
+                inode_node = srcu_dereference(to_tell->i_fsnotify_marks.first,
+                                              &fsnotify_mark_srcu);
+        else
+                inode_node = NULL;
+
+        if (mnt) {
+                if ((mask & FS_MODIFY) ||
+                    (test_mask & mnt->mnt_fsnotify_mask))
+                        vfsmount_node = srcu_dereference(mnt->mnt_fsnotify_marks.first,
+                                                         &fsnotify_mark_srcu);
+                else
+                        vfsmount_node = NULL;
+        } else {
+                mnt = NULL;
+                vfsmount_node = NULL;
+        }
+
+        while (inode_node || vfsmount_node) {
+                if (inode_node) {
+                        inode_mark = hlist_entry(srcu_dereference(inode_node, &fsnotify_mark_srcu),
+                                                 struct fsnotify_mark, i.i_list);
+                        inode_group = inode_mark->group;
+                } else
+                        inode_group = (void *)-1;
+
+                if (vfsmount_node) {
+                        vfsmount_mark = hlist_entry(srcu_dereference(vfsmount_node, &fsnotify_mark_srcu),
+                                                        struct fsnotify_mark, m.m_list);
+                        vfsmount_group = vfsmount_mark->group;
+                } else
+                        vfsmount_group = (void *)-1;
+
+                if (inode_group < vfsmount_group) {
+                        /* handle inode */
+                        send_to_group(to_tell, NULL, inode_mark, NULL, mask, data,
+                                      data_is, cookie, file_name, &event);
+                        used_inode = true;
+                } else if (vfsmount_group < inode_group) {
+                        send_to_group(to_tell, mnt, NULL, vfsmount_mark, mask, data,
+                                      data_is, cookie, file_name, &event);
+                        used_vfsmount = true;
+                } else {
+                        send_to_group(to_tell, mnt, inode_mark, vfsmount_mark,
+                                      mask, data, data_is, cookie, file_name,
+                                      &event);
+                        used_vfsmount = true;
+                        used_inode = true;
                 }
+
+                if (used_inode)
+                        inode_node = srcu_dereference(inode_node->next,
+                                                      &fsnotify_mark_srcu);
+                if (used_vfsmount)
+                        vfsmount_node = srcu_dereference(vfsmount_node->next,
+                                                         &fsnotify_mark_srcu);
         }
-        srcu_read_unlock(&fsnotify_grp_srcu, idx);
+
+        srcu_read_unlock(&fsnotify_mark_srcu, idx);
         /*
          * fsnotify_create_event() took a reference so the event can't be cleaned
          * up while we are still trying to add it to lists, drop that one.
          */
         if (event)
                 fsnotify_put_event(event);
+
+        return ret;
 }
 EXPORT_SYMBOL_GPL(fsnotify);
 
 static __init int fsnotify_init(void)
 {
-        return init_srcu_struct(&fsnotify_grp_srcu);
+        int ret;
+
+        BUG_ON(hweight32(ALL_FSNOTIFY_EVENTS) != 23);
+
+        ret = init_srcu_struct(&fsnotify_mark_srcu);
+        if (ret)
+                panic("initializing fsnotify_mark_srcu");
+
+        return 0;
 }
-subsys_initcall(fsnotify_init);
+core_initcall(fsnotify_init);