From 07b20889e3052c7e77d6a6a54e7e83446eb1ba84 Mon Sep 17 00:00:00 2001
From: Ram Pai <linuxram@us.ibm.com>
Date: Mon, 7 Nov 2005 17:19:07 -0500
Subject: [PATCH] beginning of the shared-subtree proper

A private mount does not forward or receive propagation.  This patch
provides user the ability to convert any mount to private.

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/pnode.c | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)
 create mode 100644 fs/pnode.c

(limited to 'fs/pnode.c')

diff --git a/fs/pnode.c b/fs/pnode.c
new file mode 100644
index 000000000000..aaa0dffda12a
--- /dev/null
+++ b/fs/pnode.c
@@ -0,0 +1,17 @@
+/*
+ *  linux/fs/pnode.c
+ *
+ * (C) Copyright IBM Corporation 2005.
+ *	Released under GPL v2.
+ *	Author : Ram Pai (linuxram@us.ibm.com)
+ *
+ */
+#include <linux/namespace.h>
+#include <linux/mount.h>
+#include <linux/fs.h>
+#include "pnode.h"
+
+void change_mnt_propagation(struct vfsmount *mnt, int type)
+{
+	mnt->mnt_flags &= ~MNT_PNODE_MASK;
+}
-- 
cgit v1.2.3


From 03e06e68ff76294e53ffa898cb844d2a997b043e Mon Sep 17 00:00:00 2001
From: Ram Pai <linuxram@us.ibm.com>
Date: Mon, 7 Nov 2005 17:19:33 -0500
Subject: [PATCH] introduce shared mounts

This creates shared mounts.  A shared mount when bind-mounted to some
mountpoint, propagates mount/umount events to each other.  All the
shared mounts that propagate events to each other belong to the same
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        |  3 ++-
 fs/pnode.c            | 13 ++++++++++++-
 fs/pnode.h            |  4 ++++
 include/linux/fs.h    |  1 +
 include/linux/mount.h |  2 ++
 5 files changed, 21 insertions(+), 2 deletions(-)

(limited to 'fs/pnode.c')

diff --git a/fs/namespace.c b/fs/namespace.c
index 3782923d6d4d..f6861a5487df 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -68,6 +68,7 @@ struct vfsmount *alloc_vfsmnt(const char *name)
 		INIT_LIST_HEAD(&mnt->mnt_mounts);
 		INIT_LIST_HEAD(&mnt->mnt_list);
 		INIT_LIST_HEAD(&mnt->mnt_expire);
+		INIT_LIST_HEAD(&mnt->mnt_share);
 		if (name) {
 			int size = strlen(name) + 1;
 			char *newname = kmalloc(size, GFP_KERNEL);
@@ -1113,7 +1114,7 @@ long do_mount(char *dev_name, char *dir_name, char *type_page,
 				    data_page);
 	else if (flags & MS_BIND)
 		retval = do_loopback(&nd, dev_name, flags & MS_REC);
-	else if (flags & MS_PRIVATE)
+	else if (flags & (MS_SHARED | MS_PRIVATE))
 		retval = do_change_type(&nd, flags);
 	else if (flags & MS_MOVE)
 		retval = do_move_mount(&nd, dev_name);
diff --git a/fs/pnode.c b/fs/pnode.c
index aaa0dffda12a..1e22165ea41f 100644
--- a/fs/pnode.c
+++ b/fs/pnode.c
@@ -11,7 +11,18 @@
 #include <linux/fs.h>
 #include "pnode.h"
 
+/* return the next shared peer mount of @p */
+static inline struct vfsmount *next_peer(struct vfsmount *p)
+{
+	return list_entry(p->mnt_share.next, struct vfsmount, mnt_share);
+}
+
 void change_mnt_propagation(struct vfsmount *mnt, int type)
 {
-	mnt->mnt_flags &= ~MNT_PNODE_MASK;
+	if (type == MS_SHARED) {
+		mnt->mnt_flags |= MNT_SHARED;
+	} else {
+		list_del_init(&mnt->mnt_share);
+		mnt->mnt_flags &= ~MNT_PNODE_MASK;
+	}
 }
diff --git a/fs/pnode.h b/fs/pnode.h
index 33549a3a74eb..ab1bdaee4e08 100644
--- a/fs/pnode.h
+++ b/fs/pnode.h
@@ -10,5 +10,9 @@
 
 #include <linux/list.h>
 #include <linux/mount.h>
+
+#define IS_MNT_SHARED(mnt) (mnt->mnt_flags & MNT_SHARED)
+#define CLEAR_MNT_SHARED(mnt) (mnt->mnt_flags &= ~MNT_SHARED)
+
 void change_mnt_propagation(struct vfsmount *, int);
 #endif /* _LINUX_PNODE_H */
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 6c431086abb3..551fba303cf8 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -105,6 +105,7 @@ extern int dir_notify_enable;
 #define MS_REC		16384
 #define MS_VERBOSE	32768
 #define MS_PRIVATE	(1<<18)	/* change to private */
+#define MS_SHARED	(1<<20)	/* change to shared */
 #define MS_POSIXACL	(1<<16)	/* VFS does not apply the umask */
 #define MS_ACTIVE	(1<<30)
 #define MS_NOUSER	(1<<31)
diff --git a/include/linux/mount.h b/include/linux/mount.h
index 8eadd3b65899..2582559718fc 100644
--- a/include/linux/mount.h
+++ b/include/linux/mount.h
@@ -20,6 +20,7 @@
 #define MNT_NOSUID	0x01
 #define MNT_NODEV	0x02
 #define MNT_NOEXEC	0x04
+#define MNT_SHARED	0x10	/* if the vfsmount is a shared mount */
 #define MNT_PNODE_MASK	0x30	/* propogation flag mask */
 
 struct vfsmount {
@@ -36,6 +37,7 @@ struct vfsmount {
 	char *mnt_devname;		/* Name of device e.g. /dev/dsk/hda1 */
 	struct list_head mnt_list;
 	struct list_head mnt_expire;	/* link in fs-specific expiry list */
+	struct list_head mnt_share;	/* circular list of shared mounts */
 	struct namespace *mnt_namespace; /* containing namespace */
 	int mnt_pinned;
 };
-- 
cgit v1.2.3


From b90fa9ae8f51f098ee480bbaabd6867992e9fc58 Mon Sep 17 00:00:00 2001
From: Ram Pai <linuxram@us.ibm.com>
Date: Mon, 7 Nov 2005 17:19:50 -0500
Subject: [PATCH] shared mount handling: bind and rbind

Implement handling of MS_BIND 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     | 126 ++++++++++++++++++++++++++++++++++++++++++++---------
 fs/pnode.c         |  81 +++++++++++++++++++++++++++++++++-
 fs/pnode.h         |  14 ++++++
 include/linux/fs.h |   5 +++
 4 files changed, 204 insertions(+), 22 deletions(-)

(limited to 'fs/pnode.c')

diff --git a/fs/namespace.c b/fs/namespace.c
index f6861a5487df..9f5a084b239f 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -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 @@ static void detach_mnt(struct vfsmount *mnt, struct nameidata *old_nd)
 	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 @@ static struct vfsmount *clone_mnt(struct vfsmount *old, struct dentry *root,
 		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 @@ int may_umount(struct vfsmount *mnt)
 
 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 release_mounts(struct list_head *head)
 	}
 }
 
-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 int lives_below_in_same_fs(struct dentry *d, struct dentry *dentry)
 	}
 }
 
-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 @@ Enomem:
 	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 @@ static int graft_tree(struct vfsmount *mnt, struct nameidata *nd)
 		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 --git a/fs/pnode.c b/fs/pnode.c
index 1e22165ea41f..2d572b88e6f6 100644
--- a/fs/pnode.c
+++ b/fs/pnode.c
@@ -20,9 +20,88 @@ static inline struct vfsmount *next_peer(struct vfsmount *p)
 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 --git a/fs/pnode.h b/fs/pnode.h
index ab1bdaee4e08..c62c72fad212 100644
--- a/fs/pnode.h
+++ b/fs/pnode.h
@@ -12,7 +12,21 @@
 #include <linux/mount.h>
 
 #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 --git a/include/linux/fs.h b/include/linux/fs.h
index 551fba303cf8..5e188b773e9c 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1251,7 +1251,12 @@ 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 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 *);
 
-- 
cgit v1.2.3


From a05964f3917c7c55368c229d7985f8e7c9977e97 Mon Sep 17 00:00:00 2001
From: Ram Pai <linuxram@us.ibm.com>
Date: Mon, 7 Nov 2005 17:20:17 -0500
Subject: [PATCH] shared mounts handling: umount

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         | 56 +++++++++++++++++++++-----------
 fs/pnode.c             | 87 +++++++++++++++++++++++++++++++++++++++++++++++++-
 fs/pnode.h             |  2 ++
 include/linux/dcache.h |  1 +
 include/linux/fs.h     |  2 +-
 5 files changed, 128 insertions(+), 20 deletions(-)

(limited to 'fs/pnode.c')

diff --git a/fs/namespace.c b/fs/namespace.c
index 1487982dbc24..4b1af01c2fb4 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -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);
diff --git a/fs/pnode.c b/fs/pnode.c
index 2d572b88e6f6..7bc942d047cd 100644
--- a/fs/pnode.c
+++ b/fs/pnode.c
@@ -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;
+}
diff --git a/fs/pnode.h b/fs/pnode.h
index c62c72fad212..9b88ba06794a 100644
--- a/fs/pnode.h
+++ b/fs/pnode.h
@@ -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 */
diff --git a/include/linux/dcache.h b/include/linux/dcache.h
index ab04b4f9b0db..46a2ba617595 100644
--- a/include/linux/dcache.h
+++ b/include/linux/dcache.h
@@ -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;
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 5e188b773e9c..8bdb504be73b 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -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);
-- 
cgit v1.2.3


From a58b0eb8e64b78d9315a5491955e78b1391d42e5 Mon Sep 17 00:00:00 2001
From: Ram Pai <linuxram@us.ibm.com>
Date: Mon, 7 Nov 2005 17:20:48 -0500
Subject: [PATCH] introduce slave mounts

A slave mount always has a master mount from which it receives
mount/umount events.  Unlike shared mount the event propagation does not
flow from the slave mount to the master.

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        |  4 +++-
 fs/pnode.c            | 54 ++++++++++++++++++++++++++++++++++++++++++++++++---
 fs/pnode.h            |  2 ++
 include/linux/fs.h    |  1 +
 include/linux/mount.h |  3 +++
 5 files changed, 60 insertions(+), 4 deletions(-)

(limited to 'fs/pnode.c')

diff --git a/fs/namespace.c b/fs/namespace.c
index 4b1af01c2fb4..46f99bc585bd 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -67,6 +67,8 @@ struct vfsmount *alloc_vfsmnt(const char *name)
 		INIT_LIST_HEAD(&mnt->mnt_list);
 		INIT_LIST_HEAD(&mnt->mnt_expire);
 		INIT_LIST_HEAD(&mnt->mnt_share);
+		INIT_LIST_HEAD(&mnt->mnt_slave_list);
+		INIT_LIST_HEAD(&mnt->mnt_slave);
 		if (name) {
 			int size = strlen(name) + 1;
 			char *newname = kmalloc(size, GFP_KERNEL);
@@ -1243,7 +1245,7 @@ long do_mount(char *dev_name, char *dir_name, char *type_page,
 				    data_page);
 	else if (flags & MS_BIND)
 		retval = do_loopback(&nd, dev_name, flags & MS_REC);
-	else if (flags & (MS_SHARED | MS_PRIVATE))
+	else if (flags & (MS_SHARED | MS_PRIVATE | MS_SLAVE))
 		retval = do_change_type(&nd, flags);
 	else if (flags & MS_MOVE)
 		retval = do_move_mount(&nd, dev_name);
diff --git a/fs/pnode.c b/fs/pnode.c
index 7bc942d047cd..f73eba24f1d1 100644
--- a/fs/pnode.c
+++ b/fs/pnode.c
@@ -17,13 +17,61 @@ static inline struct vfsmount *next_peer(struct vfsmount *p)
 	return list_entry(p->mnt_share.next, struct vfsmount, mnt_share);
 }
 
+static int do_make_slave(struct vfsmount *mnt)
+{
+	struct vfsmount *peer_mnt = mnt, *master = mnt->mnt_master;
+	struct vfsmount *slave_mnt;
+
+	/*
+	 * slave 'mnt' to a peer mount that has the
+	 * same root dentry. If none is available than
+	 * slave it to anything that is available.
+	 */
+	while ((peer_mnt = next_peer(peer_mnt)) != mnt &&
+	       peer_mnt->mnt_root != mnt->mnt_root) ;
+
+	if (peer_mnt == mnt) {
+		peer_mnt = next_peer(mnt);
+		if (peer_mnt == mnt)
+			peer_mnt = NULL;
+	}
+	list_del_init(&mnt->mnt_share);
+
+	if (peer_mnt)
+		master = peer_mnt;
+
+	if (master) {
+		list_for_each_entry(slave_mnt, &mnt->mnt_slave_list, mnt_slave)
+			slave_mnt->mnt_master = master;
+		list_del(&mnt->mnt_slave);
+		list_add(&mnt->mnt_slave, &master->mnt_slave_list);
+		list_splice(&mnt->mnt_slave_list, master->mnt_slave_list.prev);
+		INIT_LIST_HEAD(&mnt->mnt_slave_list);
+	} else {
+		struct list_head *p = &mnt->mnt_slave_list;
+		while (!list_empty(p)) {
+                        slave_mnt = list_entry(p->next,
+					struct vfsmount, mnt_slave);
+			list_del_init(&slave_mnt->mnt_slave);
+			slave_mnt->mnt_master = NULL;
+		}
+	}
+	mnt->mnt_master = master;
+	CLEAR_MNT_SHARED(mnt);
+	INIT_LIST_HEAD(&mnt->mnt_slave_list);
+	return 0;
+}
+
 void change_mnt_propagation(struct vfsmount *mnt, int type)
 {
 	if (type == MS_SHARED) {
 		set_mnt_shared(mnt);
-	} else {
-		list_del_init(&mnt->mnt_share);
-		mnt->mnt_flags &= ~MNT_PNODE_MASK;
+		return;
+	}
+	do_make_slave(mnt);
+	if (type != MS_SLAVE) {
+		list_del_init(&mnt->mnt_slave);
+		mnt->mnt_master = NULL;
 	}
 }
 
diff --git a/fs/pnode.h b/fs/pnode.h
index 9b88ba06794a..b59f0e9fe6b1 100644
--- a/fs/pnode.h
+++ b/fs/pnode.h
@@ -12,10 +12,12 @@
 #include <linux/mount.h>
 
 #define IS_MNT_SHARED(mnt) (mnt->mnt_flags & MNT_SHARED)
+#define IS_MNT_SLAVE(mnt) (mnt->mnt_master)
 #define IS_MNT_NEW(mnt)  (!mnt->mnt_namespace)
 #define CLEAR_MNT_SHARED(mnt) (mnt->mnt_flags &= ~MNT_SHARED)
 
 #define CL_EXPIRE    		0x01
+#define CL_SLAVE     		0x02
 #define CL_COPY_ALL 		0x04
 #define CL_MAKE_SHARED 		0x08
 #define CL_PROPAGATION 		0x10
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 8bdb504be73b..eef66f54c017 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -105,6 +105,7 @@ extern int dir_notify_enable;
 #define MS_REC		16384
 #define MS_VERBOSE	32768
 #define MS_PRIVATE	(1<<18)	/* change to private */
+#define MS_SLAVE	(1<<19)	/* change to slave */
 #define MS_SHARED	(1<<20)	/* change to shared */
 #define MS_POSIXACL	(1<<16)	/* VFS does not apply the umask */
 #define MS_ACTIVE	(1<<30)
diff --git a/include/linux/mount.h b/include/linux/mount.h
index 2582559718fc..7e133ae2a94f 100644
--- a/include/linux/mount.h
+++ b/include/linux/mount.h
@@ -38,6 +38,9 @@ struct vfsmount {
 	struct list_head mnt_list;
 	struct list_head mnt_expire;	/* link in fs-specific expiry list */
 	struct list_head mnt_share;	/* circular list of shared mounts */
+	struct list_head mnt_slave_list;/* list of slave mounts */
+	struct list_head mnt_slave;	/* slave list entry */
+	struct vfsmount *mnt_master;	/* slave is on master->mnt_slave_list */
 	struct namespace *mnt_namespace; /* containing namespace */
 	int mnt_pinned;
 };
-- 
cgit v1.2.3


From 5afe00221389998a25d611dc7941c06580c29eb6 Mon Sep 17 00:00:00 2001
From: Ram Pai <linuxram@us.ibm.com>
Date: Mon, 7 Nov 2005 17:21:01 -0500
Subject: [PATCH] handling of slave mounts

This makes bind, rbind, move, clone namespace and umount operations
aware of the semantics of slave mount (see Documentation/sharedsubtree.txt
in the last patch of the 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 | 77 +++++++++++++++++++++++++++++++++++--------------------
 fs/pnode.c     | 81 +++++++++++++++++++++++++++++++++++++++++++++++++++-------
 2 files changed, 121 insertions(+), 37 deletions(-)

(limited to 'fs/pnode.c')

diff --git a/fs/namespace.c b/fs/namespace.c
index 46f99bc585bd..089670363704 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -227,8 +227,17 @@ static struct vfsmount *clone_mnt(struct vfsmount *old, struct dentry *root,
 		mnt->mnt_mountpoint = mnt->mnt_root;
 		mnt->mnt_parent = mnt;
 
-		if ((flag & CL_PROPAGATION) || IS_MNT_SHARED(old))
-			list_add(&mnt->mnt_share, &old->mnt_share);
+		if (flag & CL_SLAVE) {
+			list_add(&mnt->mnt_slave, &old->mnt_slave_list);
+			mnt->mnt_master = old;
+			CLEAR_MNT_SHARED(mnt);
+		} else {
+			if ((flag & CL_PROPAGATION) || IS_MNT_SHARED(old))
+				list_add(&mnt->mnt_share, &old->mnt_share);
+			if (IS_MNT_SLAVE(old))
+				list_add(&mnt->mnt_slave, &old->mnt_slave);
+			mnt->mnt_master = old->mnt_master;
+		}
 		if (flag & CL_MAKE_SHARED)
 			set_mnt_shared(mnt);
 
@@ -689,18 +698,18 @@ Enomem:
  *
  *  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   |
- * 	*********************************************
+ * 	-------------------------------------------------------------
+ * 	|         BIND MOUNT OPERATION                               |
+ * 	|*************************************************************
+ * 	| source-->| shared        |       private  |       slave    |
+ * 	| dest     |               |                |                |
+ * 	|   |      |               |                |                |
+ * 	|   v      |               |                |                |
+ * 	|*************************************************************
+ * 	|  shared  | shared (++)   |     shared (+) |     shared(+++)|
+ * 	|          |               |                |                |
+ * 	|non-shared| shared (+)    |      private   |      slave (*) |
+ * 	**************************************************************
  * A bind operation clones the source mount and mounts the clone on the
  * destination mount.
  *
@@ -710,21 +719,33 @@ 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 propagated to all the mounts in the propagation tree
+ *       of the destination mount and the cloned mount is made slave
+ *       of the same master as that of the source mount. The cloned mount
+ *       is marked as 'shared and slave'.
+ * (*)   the cloned mount is made a slave of the same master as that of the
+ * 	 source mount.
+ *
+ * 	--------------------------------------------------------------
+ * 	|         		MOVE MOUNT OPERATION                 |
+ * 	|*************************************************************
+ * 	| source-->| shared        |       private  |       slave    |
+ * 	| dest     |               |                |                |
+ * 	|   |      |               |                |                |
+ * 	|   v      |               |                |                |
+ * 	|*************************************************************
+ * 	|  shared  | shared (+)    |     shared (+) |    shared(+++) |
+ * 	|          |               |                |                |
+ * 	|non-shared| shared (+*)   |      private   |    slave (*)   |
+ * 	**************************************************************
+ *
+ * (+)  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.
+ * (+++)  the mount is moved to the destination and is then propagated to
+ * 	all the mounts belonging to the destination mount's propagation tree.
+ * 	the mount is marked as 'shared and slave'.
+ * (*)	the mount continues to be a slave at the new location.
  *
  * if the source mount is a tree, the operations explained above is
  * applied to each mount in the tree.
diff --git a/fs/pnode.c b/fs/pnode.c
index f73eba24f1d1..3e266c5a3071 100644
--- a/fs/pnode.c
+++ b/fs/pnode.c
@@ -17,6 +17,16 @@ static inline struct vfsmount *next_peer(struct vfsmount *p)
 	return list_entry(p->mnt_share.next, struct vfsmount, mnt_share);
 }
 
+static inline struct vfsmount *first_slave(struct vfsmount *p)
+{
+	return list_entry(p->mnt_slave_list.next, struct vfsmount, mnt_slave);
+}
+
+static inline struct vfsmount *next_slave(struct vfsmount *p)
+{
+	return list_entry(p->mnt_slave.next, struct vfsmount, mnt_slave);
+}
+
 static int do_make_slave(struct vfsmount *mnt)
 {
 	struct vfsmount *peer_mnt = mnt, *master = mnt->mnt_master;
@@ -83,10 +93,64 @@ void change_mnt_propagation(struct vfsmount *mnt, int type)
 static struct vfsmount *propagation_next(struct vfsmount *m,
 					 struct vfsmount *origin)
 {
-	m = next_peer(m);
-	if (m == origin)
-		return NULL;
-	return m;
+	/* are there any slaves of this mount? */
+	if (!IS_MNT_NEW(m) && !list_empty(&m->mnt_slave_list))
+		return first_slave(m);
+
+	while (1) {
+		struct vfsmount *next;
+		struct vfsmount *master = m->mnt_master;
+
+		if ( master == origin->mnt_master ) {
+			next = next_peer(m);
+			return ((next == origin) ? NULL : next);
+		} else if (m->mnt_slave.next != &master->mnt_slave_list)
+			return next_slave(m);
+
+		/* back at master */
+		m = master;
+	}
+}
+
+/*
+ * return the source mount to be used for cloning
+ *
+ * @dest 	the current destination mount
+ * @last_dest  	the last seen destination mount
+ * @last_src  	the last seen source mount
+ * @type	return CL_SLAVE if the new mount has to be
+ * 		cloned as a slave.
+ */
+static struct vfsmount *get_source(struct vfsmount *dest,
+					struct vfsmount *last_dest,
+					struct vfsmount *last_src,
+					int *type)
+{
+	struct vfsmount *p_last_src = NULL;
+	struct vfsmount *p_last_dest = NULL;
+	*type = CL_PROPAGATION;;
+
+	if (IS_MNT_SHARED(dest))
+		*type |= CL_MAKE_SHARED;
+
+	while (last_dest != dest->mnt_master) {
+		p_last_dest = last_dest;
+		p_last_src = last_src;
+		last_dest = last_dest->mnt_master;
+		last_src = last_src->mnt_master;
+	}
+
+	if (p_last_dest) {
+		do {
+			p_last_dest = next_peer(p_last_dest);
+		} while (IS_MNT_NEW(p_last_dest));
+	}
+
+	if (dest != p_last_dest) {
+		*type |= CL_SLAVE;
+		return last_src;
+	} else
+		return p_last_src;
 }
 
 /*
@@ -114,16 +178,15 @@ int propagate_mnt(struct vfsmount *dest_mnt, struct dentry *dest_dentry,
 
 	for (m = propagation_next(dest_mnt, dest_mnt); m;
 			m = propagation_next(m, dest_mnt)) {
-		int type = CL_PROPAGATION;
+		int type;
+		struct vfsmount *source;
 
 		if (IS_MNT_NEW(m))
 			continue;
 
-		if (IS_MNT_SHARED(m))
-			type |= CL_MAKE_SHARED;
+		source =  get_source(m, prev_dest_mnt, prev_src_mnt, &type);
 
-		if (!(child = copy_tree(source_mnt, source_mnt->mnt_root,
-						type))) {
+		if (!(child = copy_tree(source, source->mnt_root, type))) {
 			ret = -ENOMEM;
 			list_splice(tree_list, tmp_list.prev);
 			goto out;
-- 
cgit v1.2.3


From 9676f0c6389b62bd6b24d77d4b3abdbcfa32d0f2 Mon Sep 17 00:00:00 2001
From: Ram Pai <linuxram@us.ibm.com>
Date: Mon, 7 Nov 2005 17:21:20 -0500
Subject: [PATCH] unbindable mounts

An unbindable mount does not forward or receive propagation.  Also
unbindable mount disallows bind mounts.  The semantics is as follows.

Bind semantics:
  It is invalid to bind mount an unbindable mount.

Move semantics:
  It is invalid to move an unbindable mount under shared mount.

Clone-namespace semantics:
  If a mount is unbindable in the parent namespace, the corresponding
  cloned mount in the child namespace becomes unbindable too.  Note:
  there is subtle difference, unbindable mounts cannot be bind mounted
  but can be cloned during clone-namespace.

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        | 88 ++++++++++++++++++++++++++++++++++++---------------
 fs/pnode.c            |  2 ++
 fs/pnode.h            |  1 +
 include/linux/fs.h    |  1 +
 include/linux/mount.h |  1 +
 5 files changed, 67 insertions(+), 26 deletions(-)

(limited to 'fs/pnode.c')

diff --git a/fs/namespace.c b/fs/namespace.c
index 089670363704..caa9187f67e5 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -213,6 +213,16 @@ static struct vfsmount *next_mnt(struct vfsmount *p, struct vfsmount *root)
 	return list_entry(next, struct vfsmount, mnt_child);
 }
 
+static struct vfsmount *skip_mnt_tree(struct vfsmount *p)
+{
+	struct list_head *prev = p->mnt_mounts.prev;
+	while (prev != &p->mnt_mounts) {
+		p = list_entry(prev, struct vfsmount, mnt_child);
+		prev = p->mnt_mounts.prev;
+	}
+	return p;
+}
+
 static struct vfsmount *clone_mnt(struct vfsmount *old, struct dentry *root,
 					int flag)
 {
@@ -650,6 +660,9 @@ struct vfsmount *copy_tree(struct vfsmount *mnt, struct dentry *dentry,
 	struct vfsmount *res, *p, *q, *r, *s;
 	struct nameidata nd;
 
+	if (!(flag & CL_COPY_ALL) && IS_MNT_UNBINDABLE(mnt))
+		return NULL;
+
 	res = q = clone_mnt(mnt, dentry, flag);
 	if (!q)
 		goto Enomem;
@@ -661,6 +674,10 @@ struct vfsmount *copy_tree(struct vfsmount *mnt, struct dentry *dentry,
 			continue;
 
 		for (s = r; s; s = next_mnt(s, r)) {
+			if (!(flag & CL_COPY_ALL) && IS_MNT_UNBINDABLE(s)) {
+				s = skip_mnt_tree(s);
+				continue;
+			}
 			while (p != s->mnt_parent) {
 				p = p->mnt_parent;
 				q = q->mnt_parent;
@@ -698,18 +715,18 @@ Enomem:
  *
  *  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  |       slave    |
- * 	| dest     |               |                |                |
- * 	|   |      |               |                |                |
- * 	|   v      |               |                |                |
- * 	|*************************************************************
- * 	|  shared  | shared (++)   |     shared (+) |     shared(+++)|
- * 	|          |               |                |                |
- * 	|non-shared| shared (+)    |      private   |      slave (*) |
- * 	**************************************************************
+ * ---------------------------------------------------------------------------
+ * |         BIND MOUNT OPERATION                                            |
+ * |**************************************************************************
+ * | source-->| shared        |       private  |       slave    | unbindable |
+ * | dest     |               |                |                |            |
+ * |   |      |               |                |                |            |
+ * |   v      |               |                |                |            |
+ * |**************************************************************************
+ * |  shared  | shared (++)   |     shared (+) |     shared(+++)|  invalid   |
+ * |          |               |                |                |            |
+ * |non-shared| shared (+)    |      private   |      slave (*) |  invalid   |
+ * ***************************************************************************
  * A bind operation clones the source mount and mounts the clone on the
  * destination mount.
  *
@@ -726,18 +743,18 @@ Enomem:
  * (*)   the cloned mount is made a slave of the same master as that of the
  * 	 source mount.
  *
- * 	--------------------------------------------------------------
- * 	|         		MOVE MOUNT OPERATION                 |
- * 	|*************************************************************
- * 	| source-->| shared        |       private  |       slave    |
- * 	| dest     |               |                |                |
- * 	|   |      |               |                |                |
- * 	|   v      |               |                |                |
- * 	|*************************************************************
- * 	|  shared  | shared (+)    |     shared (+) |    shared(+++) |
- * 	|          |               |                |                |
- * 	|non-shared| shared (+*)   |      private   |    slave (*)   |
- * 	**************************************************************
+ * ---------------------------------------------------------------------------
+ * |         		MOVE MOUNT OPERATION                                 |
+ * |**************************************************************************
+ * | source-->| shared        |       private  |       slave    | unbindable |
+ * | dest     |               |                |                |            |
+ * |   |      |               |                |                |            |
+ * |   v      |               |                |                |            |
+ * |**************************************************************************
+ * |  shared  | shared (+)    |     shared (+) |    shared(+++) |  invalid   |
+ * |          |               |                |                |            |
+ * |non-shared| shared (+*)   |      private   |    slave (*)   | unbindable |
+ * ***************************************************************************
  *
  * (+)  the mount is moved to the destination. And is then propagated to
  * 	all the mounts in the propagation tree of the destination mount.
@@ -854,6 +871,9 @@ static int do_loopback(struct nameidata *nd, char *old_name, int recurse)
 
 	down_write(&namespace_sem);
 	err = -EINVAL;
+	if (IS_MNT_UNBINDABLE(old_nd.mnt))
+ 		goto out;
+
 	if (!check_mnt(nd->mnt) || !check_mnt(old_nd.mnt))
 		goto out;
 
@@ -911,6 +931,16 @@ static int do_remount(struct nameidata *nd, int flags, int mnt_flags,
 	return err;
 }
 
+static inline int tree_contains_unbindable(struct vfsmount *mnt)
+{
+	struct vfsmount *p;
+	for (p = mnt; p; p = next_mnt(p, mnt)) {
+		if (IS_MNT_UNBINDABLE(p))
+			return 1;
+	}
+	return 0;
+}
+
 static int do_move_mount(struct nameidata *nd, char *old_name)
 {
 	struct nameidata old_nd, parent_nd;
@@ -954,6 +984,12 @@ static int do_move_mount(struct nameidata *nd, char *old_name)
 	 */
 	if (old_nd.mnt->mnt_parent && IS_MNT_SHARED(old_nd.mnt->mnt_parent))
 		goto out1;
+	/*
+	 * Don't move a mount tree containing unbindable mounts to a destination
+	 * mount which is shared.
+	 */
+	if (IS_MNT_SHARED(nd->mnt) && tree_contains_unbindable(old_nd.mnt))
+		goto out1;
 	err = -ELOOP;
 	for (p = nd->mnt; p->mnt_parent != p; p = p->mnt_parent)
 		if (p == old_nd.mnt)
@@ -1266,7 +1302,7 @@ long do_mount(char *dev_name, char *dir_name, char *type_page,
 				    data_page);
 	else if (flags & MS_BIND)
 		retval = do_loopback(&nd, dev_name, flags & MS_REC);
-	else if (flags & (MS_SHARED | MS_PRIVATE | MS_SLAVE))
+	else if (flags & (MS_SHARED | MS_PRIVATE | MS_SLAVE | MS_UNBINDABLE))
 		retval = do_change_type(&nd, flags);
 	else if (flags & MS_MOVE)
 		retval = do_move_mount(&nd, dev_name);
@@ -1311,7 +1347,7 @@ int copy_namespace(int flags, struct task_struct *tsk)
 	down_write(&namespace_sem);
 	/* First pass: copy the tree topology */
 	new_ns->root = copy_tree(namespace->root, namespace->root->mnt_root,
-					CL_EXPIRE);
+					CL_COPY_ALL | CL_EXPIRE);
 	if (!new_ns->root) {
 		up_write(&namespace_sem);
 		kfree(new_ns);
diff --git a/fs/pnode.c b/fs/pnode.c
index 3e266c5a3071..aeeec8ba8dd2 100644
--- a/fs/pnode.c
+++ b/fs/pnode.c
@@ -82,6 +82,8 @@ void change_mnt_propagation(struct vfsmount *mnt, int type)
 	if (type != MS_SLAVE) {
 		list_del_init(&mnt->mnt_slave);
 		mnt->mnt_master = NULL;
+		if (type == MS_UNBINDABLE)
+			mnt->mnt_flags |= MNT_UNBINDABLE;
 	}
 }
 
diff --git a/fs/pnode.h b/fs/pnode.h
index b59f0e9fe6b1..020e1bb60fdb 100644
--- a/fs/pnode.h
+++ b/fs/pnode.h
@@ -15,6 +15,7 @@
 #define IS_MNT_SLAVE(mnt) (mnt->mnt_master)
 #define IS_MNT_NEW(mnt)  (!mnt->mnt_namespace)
 #define CLEAR_MNT_SHARED(mnt) (mnt->mnt_flags &= ~MNT_SHARED)
+#define IS_MNT_UNBINDABLE(mnt) (mnt->mnt_flags & MNT_UNBINDABLE)
 
 #define CL_EXPIRE    		0x01
 #define CL_SLAVE     		0x02
diff --git a/include/linux/fs.h b/include/linux/fs.h
index eef66f54c017..1b5f502a4b8f 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -104,6 +104,7 @@ extern int dir_notify_enable;
 #define MS_MOVE		8192
 #define MS_REC		16384
 #define MS_VERBOSE	32768
+#define MS_UNBINDABLE	(1<<17)	/* change to unbindable */
 #define MS_PRIVATE	(1<<18)	/* change to private */
 #define MS_SLAVE	(1<<19)	/* change to slave */
 #define MS_SHARED	(1<<20)	/* change to shared */
diff --git a/include/linux/mount.h b/include/linux/mount.h
index 7e133ae2a94f..dd4e83eba933 100644
--- a/include/linux/mount.h
+++ b/include/linux/mount.h
@@ -21,6 +21,7 @@
 #define MNT_NODEV	0x02
 #define MNT_NOEXEC	0x04
 #define MNT_SHARED	0x10	/* if the vfsmount is a shared mount */
+#define MNT_UNBINDABLE	0x20	/* if the vfsmount is a unbindable mount */
 #define MNT_PNODE_MASK	0x30	/* propogation flag mask */
 
 struct vfsmount {
-- 
cgit v1.2.3