[PATCH] shared mounts handling: move
authorRam Pai <linuxram@us.ibm.com>
Mon, 7 Nov 2005 22:20:03 +0000 (17:20 -0500)
committerLinus Torvalds <torvalds@g5.osdl.org>
Tue, 8 Nov 2005 02:18:11 +0000 (18:18 -0800)
Implement handling of mount --move in presense of shared mounts (see
Documentation/sharedsubtree.txt in the end of patch series for detailed
description).

Signed-off-by: Ram Pai <linuxram@us.ibm.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
fs/namespace.c

index 9f5a084b239fccfb025af06f34beb42c987b73be..1487982dbc24dcdc6569d546b77ae701fc5aefb6 100644 (file)
@@ -660,7 +660,10 @@ Enomem:
 
 /*
  *  @source_mnt : mount tree to be attached
- *  @nd        : place the mount tree @source_mnt is attached
+ *  @nd         : place the mount tree @source_mnt is attached
+ *  @parent_nd  : if non-null, detach the source_mnt from its parent and
+ *                store the parent mount and mountpoint dentry.
+ *                (done when source_mnt is moved)
  *
  *  NOTE: in the table below explains the semantics when a source mount
  *  of a given type is attached to a destination mount of a given type.
@@ -685,6 +688,21 @@ Enomem:
  * (+)   the cloned mount is created under the destination mount and is marked
  *       as shared. The cloned mount is added to the peer group of the source
  *       mount.
+ *     ---------------------------------------------
+ *     |               MOVE MOUNT OPERATION        |
+ *     |********************************************
+ *     | source-->| shared        |       private  |
+ *     | dest     |               |                |
+ *     |   |      |               |                |
+ *     |   v      |               |                |
+ *     |********************************************
+ *     |  shared  | shared (+)    |     shared (+) |
+ *     |          |               |                |
+ *     |non-shared| shared (+*)   |      private   |
+ *     *********************************************
+ * (+)  the mount is moved to the destination. And is then propagated to all
+ *     the mounts in the propagation tree of the destination mount.
+ * (+*)  the mount is moved to the destination.
  *
  * if the source mount is a tree, the operations explained above is
  * applied to each mount in the tree.
@@ -692,7 +710,7 @@ Enomem:
  * in allocations.
  */
 static int attach_recursive_mnt(struct vfsmount *source_mnt,
-                               struct nameidata *nd)
+                       struct nameidata *nd, struct nameidata *parent_nd)
 {
        LIST_HEAD(tree_list);
        struct vfsmount *dest_mnt = nd->mnt;
@@ -708,8 +726,14 @@ static int attach_recursive_mnt(struct vfsmount *source_mnt,
        }
 
        spin_lock(&vfsmount_lock);
-       mnt_set_mountpoint(dest_mnt, dest_dentry, source_mnt);
-       commit_tree(source_mnt);
+       if (parent_nd) {
+               detach_mnt(source_mnt, parent_nd);
+               attach_mnt(source_mnt, nd);
+               touch_namespace(current->namespace);
+       } else {
+               mnt_set_mountpoint(dest_mnt, dest_dentry, source_mnt);
+               commit_tree(source_mnt);
+       }
 
        list_for_each_entry_safe(child, p, &tree_list, mnt_hash) {
                list_del_init(&child->mnt_hash);
@@ -740,7 +764,7 @@ static int graft_tree(struct vfsmount *mnt, struct nameidata *nd)
 
        err = -ENOENT;
        if (IS_ROOT(nd->dentry) || !d_unhashed(nd->dentry))
-               err = attach_recursive_mnt(mnt, nd);
+               err = attach_recursive_mnt(mnt, nd, NULL);
 out_unlock:
        up(&nd->dentry->d_inode->i_sem);
        if (!err)
@@ -869,35 +893,36 @@ static int do_move_mount(struct nameidata *nd, char *old_name)
        if (IS_DEADDIR(nd->dentry->d_inode))
                goto out1;
 
-       spin_lock(&vfsmount_lock);
        if (!IS_ROOT(nd->dentry) && d_unhashed(nd->dentry))
-               goto out2;
+               goto out1;
 
        err = -EINVAL;
        if (old_nd.dentry != old_nd.mnt->mnt_root)
-               goto out2;
+               goto out1;
 
        if (old_nd.mnt == old_nd.mnt->mnt_parent)
-               goto out2;
+               goto out1;
 
        if (S_ISDIR(nd->dentry->d_inode->i_mode) !=
              S_ISDIR(old_nd.dentry->d_inode->i_mode))
-               goto out2;
-
+               goto out1;
+       /*
+        * Don't move a mount residing in a shared parent.
+        */
+       if (old_nd.mnt->mnt_parent && IS_MNT_SHARED(old_nd.mnt->mnt_parent))
+               goto out1;
        err = -ELOOP;
        for (p = nd->mnt; p->mnt_parent != p; p = p->mnt_parent)
                if (p == old_nd.mnt)
-                       goto out2;
-       err = 0;
+                       goto out1;
 
-       detach_mnt(old_nd.mnt, &parent_nd);
-       attach_mnt(old_nd.mnt, nd);
-       touch_namespace(current->namespace);
+       if ((err = attach_recursive_mnt(old_nd.mnt, nd, &parent_nd)))
+               goto out1;
 
+       spin_lock(&vfsmount_lock);
        /* if the mount is moved, it should no longer be expire
         * automatically */
        list_del_init(&old_nd.mnt->mnt_expire);
-out2:
        spin_unlock(&vfsmount_lock);
 out1:
        up(&nd->dentry->d_inode->i_sem);
@@ -1467,6 +1492,10 @@ asmlinkage long sys_pivot_root(const char __user * new_root,
        down_write(&namespace_sem);
        down(&old_nd.dentry->d_inode->i_sem);
        error = -EINVAL;
+       if (IS_MNT_SHARED(old_nd.mnt) ||
+               IS_MNT_SHARED(new_nd.mnt->mnt_parent) ||
+               IS_MNT_SHARED(user_nd.mnt->mnt_parent))
+               goto out2;
        if (!check_mnt(user_nd.mnt))
                goto out2;
        error = -ENOENT;