diff -urN FS-shared_mount/fs/namespace.c FS-shared_rbind/fs/namespace.c --- FS-shared_mount/fs/namespace.c 2005-11-03 07:20:53.000000000 -0500 +++ FS-shared_rbind/fs/namespace.c 2005-11-03 07:21:44.000000000 -0500 @@ -28,8 +28,6 @@ extern int __init init_rootfs(void); -#define CL_EXPIRE 0x01 - #ifdef CONFIG_SYSFS extern int __init sysfs_init(void); #else @@ -145,13 +143,43 @@ old_nd->dentry->d_mounted--; } +void mnt_set_mountpoint(struct vfsmount *mnt, struct dentry *dentry, + struct vfsmount *child_mnt) +{ + child_mnt->mnt_parent = mntget(mnt); + child_mnt->mnt_mountpoint = dget(dentry); + dentry->d_mounted++; +} + static void attach_mnt(struct vfsmount *mnt, struct nameidata *nd) { - mnt->mnt_parent = mntget(nd->mnt); - mnt->mnt_mountpoint = dget(nd->dentry); - list_add(&mnt->mnt_hash, mount_hashtable + hash(nd->mnt, nd->dentry)); + mnt_set_mountpoint(nd->mnt, nd->dentry, mnt); + list_add_tail(&mnt->mnt_hash, mount_hashtable + + hash(nd->mnt, nd->dentry)); list_add_tail(&mnt->mnt_child, &nd->mnt->mnt_mounts); - nd->dentry->d_mounted++; +} + +/* + * the caller must hold vfsmount_lock + */ +static void commit_tree(struct vfsmount *mnt) +{ + struct vfsmount *parent = mnt->mnt_parent; + struct vfsmount *m; + LIST_HEAD(head); + struct namespace *n = parent->mnt_namespace; + + BUG_ON(parent == mnt); + + list_add_tail(&head, &mnt->mnt_list); + list_for_each_entry(m, &head, mnt_list) + m->mnt_namespace = n; + list_splice(&head, n->list.prev); + + list_add_tail(&mnt->mnt_hash, mount_hashtable + + hash(parent, mnt->mnt_mountpoint)); + list_add_tail(&mnt->mnt_child, &parent->mnt_mounts); + touch_namespace(n); } static struct vfsmount *next_mnt(struct vfsmount *p, struct vfsmount *root) @@ -183,7 +211,11 @@ mnt->mnt_root = dget(root); mnt->mnt_mountpoint = mnt->mnt_root; mnt->mnt_parent = mnt; - mnt->mnt_namespace = current->namespace; + + if ((flag & CL_PROPAGATION) || IS_MNT_SHARED(old)) + list_add(&mnt->mnt_share, &old->mnt_share); + if (flag & CL_MAKE_SHARED) + set_mnt_shared(mnt); /* stick the duplicate mount on the same expiry list * as the original if that was on one */ @@ -379,7 +411,7 @@ EXPORT_SYMBOL(may_umount); -static void release_mounts(struct list_head *head) +void release_mounts(struct list_head *head) { struct vfsmount *mnt; while(!list_empty(head)) { @@ -401,7 +433,7 @@ } } -static void umount_tree(struct vfsmount *mnt, struct list_head *kill) +void umount_tree(struct vfsmount *mnt, struct list_head *kill) { struct vfsmount *p; @@ -581,7 +613,7 @@ } } -static struct vfsmount *copy_tree(struct vfsmount *mnt, struct dentry *dentry, +struct vfsmount *copy_tree(struct vfsmount *mnt, struct dentry *dentry, int flag) { struct vfsmount *res, *p, *q, *r, *s; @@ -626,6 +658,67 @@ return NULL; } +/* + * @source_mnt : mount tree to be attached + * @nd : place the mount tree @source_mnt is attached + * + * 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. + * --------------------------------------------- + * | BIND MOUNT OPERATION | + * |******************************************** + * | source-->| shared | private | + * | dest | | | + * | | | | | + * | v | | | + * |******************************************** + * | shared | shared (++) | shared (+) | + * | | | | + * |non-shared| shared (+) | private | + * ********************************************* + * A bind operation clones the source mount and mounts the clone on the + * destination mount. + * + * (++) the cloned mount is propagated to all the mounts in the propagation + * tree of the destination mount and the cloned mount is added to + * the peer group of the source mount. + * (+) 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. + * + * if the source mount is a tree, the operations explained above is + * applied to each mount in the tree. + * Must be called without spinlocks held, since this function can sleep + * in allocations. + */ +static int attach_recursive_mnt(struct vfsmount *source_mnt, + struct nameidata *nd) +{ + LIST_HEAD(tree_list); + struct vfsmount *dest_mnt = nd->mnt; + struct dentry *dest_dentry = nd->dentry; + struct vfsmount *child, *p; + + if (propagate_mnt(dest_mnt, dest_dentry, source_mnt, &tree_list)) + return -EINVAL; + + if (IS_MNT_SHARED(dest_mnt)) { + for (p = source_mnt; p; p = next_mnt(p, source_mnt)) + set_mnt_shared(p); + } + + spin_lock(&vfsmount_lock); + 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); + commit_tree(child); + } + spin_unlock(&vfsmount_lock); + return 0; +} + static int graft_tree(struct vfsmount *mnt, struct nameidata *nd) { int err; @@ -646,17 +739,8 @@ goto out_unlock; err = -ENOENT; - spin_lock(&vfsmount_lock); - if (IS_ROOT(nd->dentry) || !d_unhashed(nd->dentry)) { - struct list_head head; - - attach_mnt(mnt, nd); - list_add_tail(&head, &mnt->mnt_list); - list_splice(&head, current->namespace->list.prev); - err = 0; - touch_namespace(current->namespace); - } - spin_unlock(&vfsmount_lock); + if (IS_ROOT(nd->dentry) || !d_unhashed(nd->dentry)) + err = attach_recursive_mnt(mnt, nd); out_unlock: up(&nd->dentry->d_inode->i_sem); if (!err) diff -urN FS-shared_mount/fs/pnode.c FS-shared_rbind/fs/pnode.c --- FS-shared_mount/fs/pnode.c 2005-11-03 06:59:23.000000000 -0500 +++ FS-shared_rbind/fs/pnode.c 2005-11-03 06:59:24.000000000 -0500 @@ -20,9 +20,88 @@ void change_mnt_propagation(struct vfsmount *mnt, int type) { if (type == MS_SHARED) { - mnt->mnt_flags |= MNT_SHARED; + set_mnt_shared(mnt); } else { list_del_init(&mnt->mnt_share); mnt->mnt_flags &= ~MNT_PNODE_MASK; } } + +/* + * get the next mount in the propagation tree. + * @m: the mount seen last + * @origin: the original mount from where the tree walk initiated + */ +static struct vfsmount *propagation_next(struct vfsmount *m, + struct vfsmount *origin) +{ + m = next_peer(m); + if (m == origin) + return NULL; + return m; +} + +/* + * mount 'source_mnt' under the destination 'dest_mnt' at + * dentry 'dest_dentry'. And propagate that mount to + * all the peer and slave mounts of 'dest_mnt'. + * Link all the new mounts into a propagation tree headed at + * source_mnt. Also link all the new mounts using ->mnt_list + * headed at source_mnt's ->mnt_list + * + * @dest_mnt: destination mount. + * @dest_dentry: destination dentry. + * @source_mnt: source mount. + * @tree_list : list of heads of trees to be attached. + */ +int propagate_mnt(struct vfsmount *dest_mnt, struct dentry *dest_dentry, + struct vfsmount *source_mnt, struct list_head *tree_list) +{ + struct vfsmount *m, *child; + int ret = 0; + struct vfsmount *prev_dest_mnt = dest_mnt; + struct vfsmount *prev_src_mnt = source_mnt; + LIST_HEAD(tmp_list); + LIST_HEAD(umount_list); + + for (m = propagation_next(dest_mnt, dest_mnt); m; + m = propagation_next(m, dest_mnt)) { + int type = CL_PROPAGATION; + + if (IS_MNT_NEW(m)) + continue; + + if (IS_MNT_SHARED(m)) + type |= CL_MAKE_SHARED; + + if (!(child = copy_tree(source_mnt, source_mnt->mnt_root, + type))) { + ret = -ENOMEM; + list_splice(tree_list, tmp_list.prev); + goto out; + } + + if (is_subdir(dest_dentry, m->mnt_root)) { + mnt_set_mountpoint(m, dest_dentry, child); + list_add_tail(&child->mnt_hash, tree_list); + } else { + /* + * This can happen if the parent mount was bind mounted + * on some subdirectory of a shared/slave mount. + */ + list_add_tail(&child->mnt_hash, &tmp_list); + } + prev_dest_mnt = m; + prev_src_mnt = child; + } +out: + spin_lock(&vfsmount_lock); + 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); + } + spin_unlock(&vfsmount_lock); + release_mounts(&umount_list); + return ret; +} diff -urN FS-shared_mount/fs/pnode.h FS-shared_rbind/fs/pnode.h --- FS-shared_mount/fs/pnode.h 2005-11-03 06:59:23.000000000 -0500 +++ FS-shared_rbind/fs/pnode.h 2005-11-03 06:59:24.000000000 -0500 @@ -12,7 +12,21 @@ #include #define IS_MNT_SHARED(mnt) (mnt->mnt_flags & MNT_SHARED) +#define IS_MNT_NEW(mnt) (!mnt->mnt_namespace) #define CLEAR_MNT_SHARED(mnt) (mnt->mnt_flags &= ~MNT_SHARED) +#define CL_EXPIRE 0x01 +#define CL_COPY_ALL 0x04 +#define CL_MAKE_SHARED 0x08 +#define CL_PROPAGATION 0x10 + +static inline void set_mnt_shared(struct vfsmount *mnt) +{ + mnt->mnt_flags &= ~MNT_PNODE_MASK; + mnt->mnt_flags |= MNT_SHARED; +} + void change_mnt_propagation(struct vfsmount *, int); +int propagate_mnt(struct vfsmount *, struct dentry *, struct vfsmount *, + struct list_head *); #endif /* _LINUX_PNODE_H */ diff -urN FS-shared_mount/include/linux/fs.h FS-shared_rbind/include/linux/fs.h --- FS-shared_mount/include/linux/fs.h 2005-11-03 06:59:23.000000000 -0500 +++ FS-shared_rbind/include/linux/fs.h 2005-11-03 06:59:24.000000000 -0500 @@ -1241,7 +1241,12 @@ 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 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); +extern void mnt_set_mountpoint(struct vfsmount *, struct dentry *, + struct vfsmount *); extern int vfs_statfs(struct super_block *, struct kstatfs *);