[PATCH] shared mounts handling: umount
authorRam Pai <linuxram@us.ibm.com>
Mon, 7 Nov 2005 22:20:17 +0000 (17:20 -0500)
committerLinus Torvalds <torvalds@g5.osdl.org>
Tue, 8 Nov 2005 02:18:11 +0000 (18:18 -0800)
An unmount of a mount creates a umount event on the parent.  If the
parent is a shared mount, it gets propagated to all mounts in the peer
group.

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
fs/pnode.c
fs/pnode.h
include/linux/dcache.h
include/linux/fs.h

index 1487982dbc24dcdc6569d546b77ae701fc5aefb6..4b1af01c2fb44ced138b8ac4a909d27924aad5b6 100644 (file)
@@ -86,31 +86,44 @@ void free_vfsmnt(struct vfsmount *mnt)
 }
 
 /*
- * Now, lookup_mnt increments the ref count before returning
- * the vfsmount struct.
+ * find the first or last mount at @dentry on vfsmount @mnt depending on
+ * @dir. If @dir is set return the first mount else return the last mount.
  */
-struct vfsmount *lookup_mnt(struct vfsmount *mnt, struct dentry *dentry)
+struct vfsmount *__lookup_mnt(struct vfsmount *mnt, struct dentry *dentry,
+                             int dir)
 {
        struct list_head *head = mount_hashtable + hash(mnt, dentry);
        struct list_head *tmp = head;
        struct vfsmount *p, *found = NULL;
 
-       spin_lock(&vfsmount_lock);
        for (;;) {
-               tmp = tmp->next;
+               tmp = dir ? tmp->next : tmp->prev;
                p = NULL;
                if (tmp == head)
                        break;
                p = list_entry(tmp, struct vfsmount, mnt_hash);
                if (p->mnt_parent == mnt && p->mnt_mountpoint == dentry) {
-                       found = mntget(p);
+                       found = p;
                        break;
                }
        }
-       spin_unlock(&vfsmount_lock);
        return found;
 }
 
+/*
+ * lookup_mnt increments the ref count before returning
+ * the vfsmount struct.
+ */
+struct vfsmount *lookup_mnt(struct vfsmount *mnt, struct dentry *dentry)
+{
+       struct vfsmount *child_mnt;
+       spin_lock(&vfsmount_lock);
+       if ((child_mnt = __lookup_mnt(mnt, dentry, 1)))
+               mntget(child_mnt);
+       spin_unlock(&vfsmount_lock);
+       return child_mnt;
+}
+
 static inline int check_mnt(struct vfsmount *mnt)
 {
        return mnt->mnt_namespace == current->namespace;
@@ -404,9 +417,12 @@ EXPORT_SYMBOL(may_umount_tree);
  */
 int may_umount(struct vfsmount *mnt)
 {
-       if (atomic_read(&mnt->mnt_count) > 2)
-               return -EBUSY;
-       return 0;
+       int ret = 0;
+       spin_lock(&vfsmount_lock);
+       if (propagate_mount_busy(mnt, 2))
+               ret = -EBUSY;
+       spin_unlock(&vfsmount_lock);
+       return ret;
 }
 
 EXPORT_SYMBOL(may_umount);
@@ -433,7 +449,7 @@ void release_mounts(struct list_head *head)
        }
 }
 
-void umount_tree(struct vfsmount *mnt, struct list_head *kill)
+void umount_tree(struct vfsmount *mnt, int propagate, struct list_head *kill)
 {
        struct vfsmount *p;
 
@@ -442,6 +458,9 @@ void umount_tree(struct vfsmount *mnt, struct list_head *kill)
                list_add(&p->mnt_hash, kill);
        }
 
+       if (propagate)
+               propagate_umount(kill);
+
        list_for_each_entry(p, kill, mnt_hash) {
                list_del_init(&p->mnt_expire);
                list_del_init(&p->mnt_list);
@@ -450,6 +469,7 @@ void umount_tree(struct vfsmount *mnt, struct list_head *kill)
                list_del_init(&p->mnt_child);
                if (p->mnt_parent != p)
                        mnt->mnt_mountpoint->d_mounted--;
+               change_mnt_propagation(p, MS_PRIVATE);
        }
 }
 
@@ -526,9 +546,9 @@ static int do_umount(struct vfsmount *mnt, int flags)
        event++;
 
        retval = -EBUSY;
-       if (atomic_read(&mnt->mnt_count) == 2 || flags & MNT_DETACH) {
+       if (flags & MNT_DETACH || !propagate_mount_busy(mnt, 2)) {
                if (!list_empty(&mnt->mnt_list))
-                       umount_tree(mnt, &umount_list);
+                       umount_tree(mnt, 1, &umount_list);
                retval = 0;
        }
        spin_unlock(&vfsmount_lock);
@@ -651,7 +671,7 @@ Enomem:
        if (res) {
                LIST_HEAD(umount_list);
                spin_lock(&vfsmount_lock);
-               umount_tree(res, &umount_list);
+               umount_tree(res, 0, &umount_list);
                spin_unlock(&vfsmount_lock);
                release_mounts(&umount_list);
        }
@@ -827,7 +847,7 @@ static int do_loopback(struct nameidata *nd, char *old_name, int recurse)
        if (err) {
                LIST_HEAD(umount_list);
                spin_lock(&vfsmount_lock);
-               umount_tree(mnt, &umount_list);
+               umount_tree(mnt, 0, &umount_list);
                spin_unlock(&vfsmount_lock);
                release_mounts(&umount_list);
        }
@@ -1023,12 +1043,12 @@ static void expire_mount(struct vfsmount *mnt, struct list_head *mounts,
         * Check that it is still dead: the count should now be 2 - as
         * contributed by the vfsmount parent and the mntget above
         */
-       if (atomic_read(&mnt->mnt_count) == 2) {
+       if (!propagate_mount_busy(mnt, 2)) {
                /* delete from the namespace */
                touch_namespace(mnt->mnt_namespace);
                list_del_init(&mnt->mnt_list);
                mnt->mnt_namespace = NULL;
-               umount_tree(mnt, umounts);
+               umount_tree(mnt, 1, umounts);
                spin_unlock(&vfsmount_lock);
        } else {
                /*
@@ -1647,7 +1667,7 @@ void __put_namespace(struct namespace *namespace)
        spin_unlock(&vfsmount_lock);
        down_write(&namespace_sem);
        spin_lock(&vfsmount_lock);
-       umount_tree(root, &umount_list);
+       umount_tree(root, 0, &umount_list);
        spin_unlock(&vfsmount_lock);
        up_write(&namespace_sem);
        release_mounts(&umount_list);
index 2d572b88e6f6e87cc4fa579bfed573a798bfd8aa..7bc942d047cd299df0e76282664ba1d052c81c1b 100644 (file)
@@ -99,9 +99,94 @@ out:
        while (!list_empty(&tmp_list)) {
                child = list_entry(tmp_list.next, struct vfsmount, mnt_hash);
                list_del_init(&child->mnt_hash);
-               umount_tree(child, &umount_list);
+               umount_tree(child, 0, &umount_list);
        }
        spin_unlock(&vfsmount_lock);
        release_mounts(&umount_list);
        return ret;
 }
+
+/*
+ * return true if the refcount is greater than count
+ */
+static inline int do_refcount_check(struct vfsmount *mnt, int count)
+{
+       int mycount = atomic_read(&mnt->mnt_count);
+       return (mycount > count);
+}
+
+/*
+ * check if the mount 'mnt' can be unmounted successfully.
+ * @mnt: the mount to be checked for unmount
+ * NOTE: unmounting 'mnt' would naturally propagate to all
+ * other mounts its parent propagates to.
+ * Check if any of these mounts that **do not have submounts**
+ * have more references than 'refcnt'. If so return busy.
+ */
+int propagate_mount_busy(struct vfsmount *mnt, int refcnt)
+{
+       struct vfsmount *m, *child;
+       struct vfsmount *parent = mnt->mnt_parent;
+       int ret = 0;
+
+       if (mnt == parent)
+               return do_refcount_check(mnt, refcnt);
+
+       /*
+        * quickly check if the current mount can be unmounted.
+        * If not, we don't have to go checking for all other
+        * mounts
+        */
+       if (!list_empty(&mnt->mnt_mounts) || do_refcount_check(mnt, refcnt))
+               return 1;
+
+       for (m = propagation_next(parent, parent); m;
+                       m = propagation_next(m, parent)) {
+               child = __lookup_mnt(m, mnt->mnt_mountpoint, 0);
+               if (child && list_empty(&child->mnt_mounts) &&
+                   (ret = do_refcount_check(child, 1)))
+                       break;
+       }
+       return ret;
+}
+
+/*
+ * NOTE: unmounting 'mnt' naturally propagates to all other mounts its
+ * parent propagates to.
+ */
+static void __propagate_umount(struct vfsmount *mnt)
+{
+       struct vfsmount *parent = mnt->mnt_parent;
+       struct vfsmount *m;
+
+       BUG_ON(parent == mnt);
+
+       for (m = propagation_next(parent, parent); m;
+                       m = propagation_next(m, parent)) {
+
+               struct vfsmount *child = __lookup_mnt(m,
+                                       mnt->mnt_mountpoint, 0);
+               /*
+                * umount the child only if the child has no
+                * other children
+                */
+               if (child && list_empty(&child->mnt_mounts)) {
+                       list_del(&child->mnt_hash);
+                       list_add_tail(&child->mnt_hash, &mnt->mnt_hash);
+               }
+       }
+}
+
+/*
+ * collect all mounts that receive propagation from the mount in @list,
+ * and return these additional mounts in the same list.
+ * @list: the list of mounts to be unmounted.
+ */
+int propagate_umount(struct list_head *list)
+{
+       struct vfsmount *mnt;
+
+       list_for_each_entry(mnt, list, mnt_hash)
+               __propagate_umount(mnt);
+       return 0;
+}
index c62c72fad212eac4a373d46943e3e1c3a1b27268..9b88ba06794a90260c86c9c982531bdfc086bc6b 100644 (file)
@@ -29,4 +29,6 @@ static inline void set_mnt_shared(struct vfsmount *mnt)
 void change_mnt_propagation(struct vfsmount *, int);
 int propagate_mnt(struct vfsmount *, struct dentry *, struct vfsmount *,
                struct list_head *);
+int propagate_umount(struct list_head *);
+int propagate_mount_busy(struct vfsmount *, int);
 #endif /* _LINUX_PNODE_H */
index ab04b4f9b0db0e822dc08bc5a2abccdab32cadd3..46a2ba6175954880fcdfa89bc9a9a063f4e76443 100644 (file)
@@ -329,6 +329,7 @@ static inline int d_mountpoint(struct dentry *dentry)
 }
 
 extern struct vfsmount *lookup_mnt(struct vfsmount *, struct dentry *);
+extern struct vfsmount *__lookup_mnt(struct vfsmount *, struct dentry *, int);
 extern struct dentry *lookup_create(struct nameidata *nd, int is_dir);
 
 extern int sysctl_vfs_cache_pressure;
index 5e188b773e9ce3398ad21bf8f7571f0138b68585..8bdb504be73b99fc4fda201b88a5dc6868266c4a 100644 (file)
@@ -1251,7 +1251,7 @@ extern int unregister_filesystem(struct file_system_type *);
 extern struct vfsmount *kern_mount(struct file_system_type *);
 extern int may_umount_tree(struct vfsmount *);
 extern int may_umount(struct vfsmount *);
-extern void umount_tree(struct vfsmount *, struct list_head *);
+extern void umount_tree(struct vfsmount *, int, struct list_head *);
 extern void release_mounts(struct list_head *);
 extern long do_mount(char *, char *, char *, unsigned long, void *);
 extern struct vfsmount *copy_tree(struct vfsmount *, struct dentry *, int);