Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs

Pull vfs pile #2 from Al Viro:
 "Next pile (and there'll be one or two more).

  The large piece in this one is getting rid of /proc/*/ns/* weirdness;
  among other things, it allows to (finally) make nameidata completely
  opaque outside of fs/namei.c, making for easier further cleanups in
  there"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
  coda_venus_readdir(): use file_inode()
  fs/namei.c: fold link_path_walk() call into path_init()
  path_init(): don't bother with LOOKUP_PARENT in argument
  fs/namei.c: new helper (path_cleanup())
  path_init(): store the "base" pointer to file in nameidata itself
  make default ->i_fop have ->open() fail with ENXIO
  make nameidata completely opaque outside of fs/namei.c
  kill proc_ns completely
  take the targets of /proc/*/ns/* symlinks to separate fs
  bury struct proc_ns in fs/proc
  copy address of proc_ns_ops into ns_common
  new helpers: ns_alloc_inum/ns_free_inum
  make proc_ns_operations work with struct ns_common * instead of void *
  switch the rest of proc_ns_operations to working with &...->ns
  netns: switch ->get()/->put()/->install()/->inum() to working with &net->ns
  make mntns ->get()/->put()/->install()/->inum() work with &mnt_ns->ns
  common object embedded into various struct ....ns

arch/ia64/kernel/perfmon.c

@@ -2145,22 +2145,12 @@ doit:
 	return 0;
 }
 
-static int
-pfm_no_open(struct inode *irrelevant, struct file *dontcare)
-{
-	DPRINT(("pfm_no_open called\n"));
-	return -ENXIO;
-}
-
-
-
 static const struct file_operations pfm_file_ops = {
 	.llseek		= no_llseek,
 	.read		= pfm_read,
 	.write		= pfm_write,
 	.poll		= pfm_poll,
 	.unlocked_ioctl = pfm_ioctl,
-	.open		= pfm_no_open,	/* special open code to disallow open via /proc */
 	.fasync		= pfm_fasync,
 	.release	= pfm_close,
 	.flush		= pfm_flush

fs/Makefile

@@ -11,7 +11,7 @@ obj-y :=	open.o read_write.o file_table.o super.o \
 		attr.o bad_inode.o file.o filesystems.o namespace.o \
 		seq_file.o xattr.o libfs.o fs-writeback.o \
 		pnode.o splice.o sync.o utimes.o \
-		stack.o fs_struct.o statfs.o fs_pin.o
+		stack.o fs_struct.o statfs.o fs_pin.o nsfs.o
 
 ifeq ($(CONFIG_BLOCK),y)
 obj-y +=	buffer.o block_dev.o direct-io.o mpage.o

fs/coda/dir.c

@@ -426,7 +426,6 @@ static int coda_venus_readdir(struct file *coda_file, struct dir_context *ctx)
 	struct coda_file_info *cfi;
 	struct coda_inode_info *cii;
 	struct file *host_file;
-	struct dentry *de;
 	struct venus_dirent *vdir;
 	unsigned long vdir_size = offsetof(struct venus_dirent, d_name);
 	unsigned int type;
@@ -438,8 +437,7 @@ static int coda_venus_readdir(struct file *coda_file, struct dir_context *ctx)
 	BUG_ON(!cfi || cfi->cfi_magic != CODA_MAGIC);
 	host_file = cfi->cfi_container;
 
-	de = coda_file->f_path.dentry;
-	cii = ITOC(de->d_inode);
+	cii = ITOC(file_inode(coda_file));
 
 	vdir = kmalloc(sizeof(*vdir), GFP_KERNEL);
 	if (!vdir) return -ENOMEM;

fs/inode.c

@@ -114,6 +114,11 @@ int proc_nr_inodes(struct ctl_table *table, int write,
 }
 #endif
 
+static int no_open(struct inode *inode, struct file *file)
+{
+	return -ENXIO;
+}
+
 /**
  * inode_init_always - perform inode structure intialisation
  * @sb: superblock inode belongs to
@@ -125,7 +130,7 @@ int proc_nr_inodes(struct ctl_table *table, int write,
 int inode_init_always(struct super_block *sb, struct inode *inode)
 {
 	static const struct inode_operations empty_iops;
-	static const struct file_operations empty_fops;
+	static const struct file_operations no_open_fops = {.open = no_open};
 	struct address_space *const mapping = &inode->i_data;
 
 	inode->i_sb = sb;
@@ -133,7 +138,7 @@ int inode_init_always(struct super_block *sb, struct inode *inode)
 	inode->i_flags = 0;
 	atomic_set(&inode->i_count, 1);
 	inode->i_op = &empty_iops;
-	inode->i_fop = &empty_fops;
+	inode->i_fop = &no_open_fops;
 	inode->__i_nlink = 1;
 	inode->i_opflags = 0;
 	i_uid_write(inode, 0);
@@ -1798,7 +1803,7 @@ void init_special_inode(struct inode *inode, umode_t mode, dev_t rdev)
 	} else if (S_ISFIFO(mode))
 		inode->i_fop = &pipefifo_fops;
 	else if (S_ISSOCK(mode))
-		inode->i_fop = &bad_sock_fops;
+		;	/* leave it no_open_fops */
 	else
 		printk(KERN_DEBUG "init_special_inode: bogus i_mode (%o) for"
 				  " inode %s:%lu\n", mode, inode->i_sb->s_id,

fs/internal.h

@@ -147,3 +147,8 @@ extern const struct file_operations pipefifo_fops;
  */
 extern void sb_pin_kill(struct super_block *sb);
 extern void mnt_pin_kill(struct mount *m);
+
+/*
+ * fs/nsfs.c
+ */
+extern struct dentry_operations ns_dentry_operations;

fs/mount.h

@@ -1,10 +1,11 @@
 #include <linux/mount.h>
 #include <linux/seq_file.h>
 #include <linux/poll.h>
+#include <linux/ns_common.h>
 
 struct mnt_namespace {
 	atomic_t		count;
-	unsigned int		proc_inum;
+	struct ns_common	ns;
 	struct mount *	root;
 	struct list_head	list;
 	struct user_namespace	*user_ns;

fs/namei.c

@@ -487,6 +487,19 @@ void path_put(const struct path *path)
 }
 EXPORT_SYMBOL(path_put);
 
+struct nameidata {
+	struct path	path;
+	struct qstr	last;
+	struct path	root;
+	struct inode	*inode; /* path.dentry.d_inode */
+	unsigned int	flags;
+	unsigned	seq, m_seq;
+	int		last_type;
+	unsigned	depth;
+	struct file	*base;
+	char *saved_names[MAX_NESTED_LINKS + 1];
+};
+
 /*
  * Path walking has 2 modes, rcu-walk and ref-walk (see
  * Documentation/filesystems/path-lookup.txt).  In situations when we can't
@@ -695,6 +708,18 @@ void nd_jump_link(struct nameidata *nd, struct path *path)
 	nd->flags |= LOOKUP_JUMPED;
 }
 
+void nd_set_link(struct nameidata *nd, char *path)
+{
+	nd->saved_names[nd->depth] = path;
+}
+EXPORT_SYMBOL(nd_set_link);
+
+char *nd_get_link(struct nameidata *nd)
+{
+	return nd->saved_names[nd->depth];
+}
+EXPORT_SYMBOL(nd_get_link);
+
 static inline void put_link(struct nameidata *nd, struct path *link, void *cookie)
 {
 	struct inode *inode = link->dentry->d_inode;
@@ -1821,13 +1846,14 @@ static int link_path_walk(const char *name, struct nameidata *nd)
 }
 
 static int path_init(int dfd, const char *name, unsigned int flags,
-		     struct nameidata *nd, struct file **fp)
+		     struct nameidata *nd)
 {
 	int retval = 0;
 
 	nd->last_type = LAST_ROOT; /* if there are only slashes... */
-	nd->flags = flags | LOOKUP_JUMPED;
+	nd->flags = flags | LOOKUP_JUMPED | LOOKUP_PARENT;
 	nd->depth = 0;
+	nd->base = NULL;
 	if (flags & LOOKUP_ROOT) {
 		struct dentry *root = nd->root.dentry;
 		struct inode *inode = root->d_inode;
@@ -1847,7 +1873,7 @@ static int path_init(int dfd, const char *name, unsigned int flags,
 		} else {
 			path_get(&nd->path);
 		}
-		return 0;
+		goto done;
 	}
 
 	nd->root.mnt = NULL;
@@ -1897,7 +1923,7 @@ static int path_init(int dfd, const char *name, unsigned int flags,
 		nd->path = f.file->f_path;
 		if (flags & LOOKUP_RCU) {
 			if (f.flags & FDPUT_FPUT)
-				*fp = f.file;
+				nd->base = f.file;
 			nd->seq = __read_seqcount_begin(&nd->path.dentry->d_seq);
 			rcu_read_lock();
 		} else {
@@ -1908,13 +1934,26 @@ static int path_init(int dfd, const char *name, unsigned int flags,
 
 	nd->inode = nd->path.dentry->d_inode;
 	if (!(flags & LOOKUP_RCU))
-		return 0;
+		goto done;
 	if (likely(!read_seqcount_retry(&nd->path.dentry->d_seq, nd->seq)))
-		return 0;
+		goto done;
 	if (!(nd->flags & LOOKUP_ROOT))
 		nd->root.mnt = NULL;
 	rcu_read_unlock();
 	return -ECHILD;
+done:
+	current->total_link_count = 0;
+	return link_path_walk(name, nd);
+}
+
+static void path_cleanup(struct nameidata *nd)
+{
+	if (nd->root.mnt && !(nd->flags & LOOKUP_ROOT)) {
+		path_put(&nd->root);
+		nd->root.mnt = NULL;
+	}
+	if (unlikely(nd->base))
+		fput(nd->base);
 }
 
 static inline int lookup_last(struct nameidata *nd, struct path *path)
@@ -1930,7 +1969,6 @@ static inline int lookup_last(struct nameidata *nd, struct path *path)
 static int path_lookupat(int dfd, const char *name,
 				unsigned int flags, struct nameidata *nd)
 {
-	struct file *base = NULL;
 	struct path path;
 	int err;
 
@@ -1948,14 +1986,7 @@ static int path_lookupat(int dfd, const char *name,
 	 * be handled by restarting a traditional ref-walk (which will always
 	 * be able to complete).
 	 */
-	err = path_init(dfd, name, flags | LOOKUP_PARENT, nd, &base);
-
-	if (unlikely(err))
-		goto out;
-
-	current->total_link_count = 0;
-	err = link_path_walk(name, nd);
-
+	err = path_init(dfd, name, flags, nd);
 	if (!err && !(flags & LOOKUP_PARENT)) {
 		err = lookup_last(nd, &path);
 		while (err > 0) {
@@ -1983,14 +2014,7 @@ static int path_lookupat(int dfd, const char *name,
 		}
 	}
 
-out:
-	if (base)
-		fput(base);
-
-	if (nd->root.mnt && !(nd->flags & LOOKUP_ROOT)) {
-		path_put(&nd->root);
-		nd->root.mnt = NULL;
-	}
+	path_cleanup(nd);
 	return err;
 }
 
@@ -2297,19 +2321,13 @@ out:
 static int
 path_mountpoint(int dfd, const char *name, struct path *path, unsigned int flags)
 {
-	struct file *base = NULL;
 	struct nameidata nd;
 	int err;
 
-	err = path_init(dfd, name, flags | LOOKUP_PARENT, &nd, &base);
+	err = path_init(dfd, name, flags, &nd);
 	if (unlikely(err))
 		goto out;
 
-	current->total_link_count = 0;
-	err = link_path_walk(name, &nd);
-	if (err)
-		goto out;
-
 	err = mountpoint_last(&nd, path);
 	while (err > 0) {
 		void *cookie;
@@ -2325,12 +2343,7 @@ path_mountpoint(int dfd, const char *name, struct path *path, unsigned int flags
 		put_link(&nd, &link, cookie);
 	}
 out:
-	if (base)
-		fput(base);
-
-	if (nd.root.mnt && !(nd.flags & LOOKUP_ROOT))
-		path_put(&nd.root);
-
+	path_cleanup(&nd);
 	return err;
 }
 
@@ -3181,7 +3194,6 @@ out:
 static struct file *path_openat(int dfd, struct filename *pathname,
 		struct nameidata *nd, const struct open_flags *op, int flags)
 {
-	struct file *base = NULL;
 	struct file *file;
 	struct path path;
 	int opened = 0;
@@ -3198,12 +3210,7 @@ static struct file *path_openat(int dfd, struct filename *pathname,
 		goto out;
 	}
 
-	error = path_init(dfd, pathname->name, flags | LOOKUP_PARENT, nd, &base);
-	if (unlikely(error))
-		goto out;
-
-	current->total_link_count = 0;
-	error = link_path_walk(pathname->name, nd);
+	error = path_init(dfd, pathname->name, flags, nd);
 	if (unlikely(error))
 		goto out;
 
@@ -3229,10 +3236,7 @@ static struct file *path_openat(int dfd, struct filename *pathname,
 		put_link(nd, &link, cookie);
 	}
 out:
-	if (nd->root.mnt && !(nd->flags & LOOKUP_ROOT))
-		path_put(&nd->root);
-	if (base)
-		fput(base);
+	path_cleanup(nd);
 	if (!(opened & FILE_OPENED)) {
 		BUG_ON(!error);
 		put_filp(file);

fs/namespace.c

@@ -1569,17 +1569,13 @@ SYSCALL_DEFINE1(oldumount, char __user *, name)
 static bool is_mnt_ns_file(struct dentry *dentry)
 {
 	/* Is this a proxy for a mount namespace? */
-	struct inode *inode = dentry->d_inode;
-	struct proc_ns *ei;
+	return dentry->d_op == &ns_dentry_operations &&
+	       dentry->d_fsdata == &mntns_operations;
+}
 
-	if (!proc_ns_inode(inode))
-		return false;
-
-	ei = get_proc_ns(inode);
-	if (ei->ns_ops != &mntns_operations)
-		return false;
-
-	return true;
+struct mnt_namespace *to_mnt_ns(struct ns_common *ns)
+{
+	return container_of(ns, struct mnt_namespace, ns);
 }
 
 static bool mnt_ns_loop(struct dentry *dentry)
@@ -1591,7 +1587,7 @@ static bool mnt_ns_loop(struct dentry *dentry)
 	if (!is_mnt_ns_file(dentry))
 		return false;
 
-	mnt_ns = get_proc_ns(dentry->d_inode)->ns;
+	mnt_ns = to_mnt_ns(get_proc_ns(dentry->d_inode));
 	return current->nsproxy->mnt_ns->seq >= mnt_ns->seq;
 }
 
@@ -2020,7 +2016,10 @@ static int do_loopback(struct path *path, const char *old_name,
 	if (IS_MNT_UNBINDABLE(old))
 		goto out2;
 
-	if (!check_mnt(parent) || !check_mnt(old))
+	if (!check_mnt(parent))
+		goto out2;
+
+	if (!check_mnt(old) && old_path.dentry->d_op != &ns_dentry_operations)
 		goto out2;
 
 	if (!recurse && has_locked_children(old, old_path.dentry))
@@ -2640,7 +2639,7 @@ dput_out:
 
 static void free_mnt_ns(struct mnt_namespace *ns)
 {
-	proc_free_inum(ns->proc_inum);
+	ns_free_inum(&ns->ns);
 	put_user_ns(ns->user_ns);
 	kfree(ns);
 }
@@ -2662,11 +2661,12 @@ static struct mnt_namespace *alloc_mnt_ns(struct user_namespace *user_ns)
 	new_ns = kmalloc(sizeof(struct mnt_namespace), GFP_KERNEL);
 	if (!new_ns)
 		return ERR_PTR(-ENOMEM);
-	ret = proc_alloc_inum(&new_ns->proc_inum);
+	ret = ns_alloc_inum(&new_ns->ns);
 	if (ret) {
 		kfree(new_ns);
 		return ERR_PTR(ret);
 	}
+	new_ns->ns.ops = &mntns_operations;
 	new_ns->seq = atomic64_add_return(1, &mnt_ns_seq);
 	atomic_set(&new_ns->count, 1);
 	new_ns->root = NULL;
@@ -3144,31 +3144,31 @@ found:
 	return visible;
 }
 
-static void *mntns_get(struct task_struct *task)
+static struct ns_common *mntns_get(struct task_struct *task)
 {
-	struct mnt_namespace *ns = NULL;
+	struct ns_common *ns = NULL;
 	struct nsproxy *nsproxy;
 
 	task_lock(task);
 	nsproxy = task->nsproxy;
 	if (nsproxy) {
-		ns = nsproxy->mnt_ns;
-		get_mnt_ns(ns);
+		ns = &nsproxy->mnt_ns->ns;
+		get_mnt_ns(to_mnt_ns(ns));
 	}
 	task_unlock(task);
 
 	return ns;
 }
 
-static void mntns_put(void *ns)
+static void mntns_put(struct ns_common *ns)
 {
-	put_mnt_ns(ns);
+	put_mnt_ns(to_mnt_ns(ns));
 }
 
-static int mntns_install(struct nsproxy *nsproxy, void *ns)
+static int mntns_install(struct nsproxy *nsproxy, struct ns_common *ns)
 {
 	struct fs_struct *fs = current->fs;
-	struct mnt_namespace *mnt_ns = ns;
+	struct mnt_namespace *mnt_ns = to_mnt_ns(ns);
 	struct path root;
 
 	if (!ns_capable(mnt_ns->user_ns, CAP_SYS_ADMIN) ||
@@ -3198,17 +3198,10 @@ static int mntns_install(struct nsproxy *nsproxy, void *ns)
 	return 0;
 }
 
-static unsigned int mntns_inum(void *ns)
-{
-	struct mnt_namespace *mnt_ns = ns;
-	return mnt_ns->proc_inum;
-}
-
 const struct proc_ns_operations mntns_operations = {
 	.name		= "mnt",
 	.type		= CLONE_NEWNS,
 	.get		= mntns_get,
 	.put		= mntns_put,
 	.install	= mntns_install,
-	.inum		= mntns_inum,
 };

fs/nsfs.c

@@ -0,0 +1,161 @@
+#include <linux/mount.h>
+#include <linux/file.h>
+#include <linux/fs.h>
+#include <linux/proc_ns.h>
+#include <linux/magic.h>
+#include <linux/ktime.h>
+
+static struct vfsmount *nsfs_mnt;
+
+static const struct file_operations ns_file_operations = {
+	.llseek		= no_llseek,
+};
+
+static char *ns_dname(struct dentry *dentry, char *buffer, int buflen)
+{
+	struct inode *inode = dentry->d_inode;
+	const struct proc_ns_operations *ns_ops = dentry->d_fsdata;
+
+	return dynamic_dname(dentry, buffer, buflen, "%s:[%lu]",
+		ns_ops->name, inode->i_ino);
+}
+
+static void ns_prune_dentry(struct dentry *dentry)
+{
+	struct inode *inode = dentry->d_inode;
+	if (inode) {
+		struct ns_common *ns = inode->i_private;
+		atomic_long_set(&ns->stashed, 0);
+	}
+}
+
+const struct dentry_operations ns_dentry_operations =
+{
+	.d_prune	= ns_prune_dentry,
+	.d_delete	= always_delete_dentry,
+	.d_dname	= ns_dname,
+};
+
+static void nsfs_evict(struct inode *inode)
+{
+	struct ns_common *ns = inode->i_private;
+	clear_inode(inode);
+	ns->ops->put(ns);
+}
+
+void *ns_get_path(struct path *path, struct task_struct *task,
+			const struct proc_ns_operations *ns_ops)
+{
+	struct vfsmount *mnt = mntget(nsfs_mnt);
+	struct qstr qname = { .name = "", };
+	struct dentry *dentry;
+	struct inode *inode;
+	struct ns_common *ns;
+	unsigned long d;
+
+again:
+	ns = ns_ops->get(task);
+	if (!ns) {
+		mntput(mnt);
+		return ERR_PTR(-ENOENT);
+	}
+	rcu_read_lock();
+	d = atomic_long_read(&ns->stashed);
+	if (!d)
+		goto slow;
+	dentry = (struct dentry *)d;
+	if (!lockref_get_not_dead(&dentry->d_lockref))
+		goto slow;
+	rcu_read_unlock();
+	ns_ops->put(ns);
+got_it:
+	path->mnt = mnt;
+	path->dentry = dentry;
+	return NULL;
+slow:
+	rcu_read_unlock();
+	inode = new_inode_pseudo(mnt->mnt_sb);
+	if (!inode) {
+		ns_ops->put(ns);
+		mntput(mnt);
+		return ERR_PTR(-ENOMEM);
+	}
+	inode->i_ino = ns->inum;
+	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
+	inode->i_flags |= S_IMMUTABLE;
+	inode->i_mode = S_IFREG | S_IRUGO;
+	inode->i_fop = &ns_file_operations;
+	inode->i_private = ns;
+
+	dentry = d_alloc_pseudo(mnt->mnt_sb, &qname);
+	if (!dentry) {
+		iput(inode);
+		mntput(mnt);
+		return ERR_PTR(-ENOMEM);
+	}
+	d_instantiate(dentry, inode);
+	dentry->d_fsdata = (void *)ns_ops;
+	d = atomic_long_cmpxchg(&ns->stashed, 0, (unsigned long)dentry);
+	if (d) {
+		d_delete(dentry);	/* make sure ->d_prune() does nothing */
+		dput(dentry);
+		cpu_relax();
+		goto again;
+	}
+	goto got_it;
+}
+
+int ns_get_name(char *buf, size_t size, struct task_struct *task,
+			const struct proc_ns_operations *ns_ops)
+{
+	struct ns_common *ns;
+	int res = -ENOENT;
+	ns = ns_ops->get(task);
+	if (ns) {
+		res = snprintf(buf, size, "%s:[%u]", ns_ops->name, ns->inum);
+		ns_ops->put(ns);
+	}
+	return res;
+}
+
+struct file *proc_ns_fget(int fd)
+{
+	struct file *file;
+
+	file = fget(fd);
+	if (!file)
+		return ERR_PTR(-EBADF);
+
+	if (file->f_op != &ns_file_operations)
+		goto out_invalid;
+
+	return file;
+
+out_invalid:
+	fput(file);
+	return ERR_PTR(-EINVAL);
+}
+
+static const struct super_operations nsfs_ops = {
+	.statfs = simple_statfs,
+	.evict_inode = nsfs_evict,
+};
+static struct dentry *nsfs_mount(struct file_system_type *fs_type,
+			int flags, const char *dev_name, void *data)
+{
+	return mount_pseudo(fs_type, "nsfs:", &nsfs_ops,
+			&ns_dentry_operations, NSFS_MAGIC);
+}
+static struct file_system_type nsfs = {
+	.name = "nsfs",
+	.mount = nsfs_mount,
+	.kill_sb = kill_anon_super,
+};
+
+void __init nsfs_init(void)
+{
+	nsfs_mnt = kern_mount(&nsfs);
+	if (IS_ERR(nsfs_mnt))
+		panic("can't set nsfs up\n");
+	nsfs_mnt->mnt_sb->s_flags &= ~MS_NOUSER;
+}

fs/proc/inode.c

@@ -32,8 +32,6 @@ static void proc_evict_inode(struct inode *inode)
 {
 	struct proc_dir_entry *de;
 	struct ctl_table_header *head;
-	const struct proc_ns_operations *ns_ops;
-	void *ns;
 
 	truncate_inode_pages_final(&inode->i_data);
 	clear_inode(inode);
@@ -50,11 +48,6 @@ static void proc_evict_inode(struct inode *inode)
 		RCU_INIT_POINTER(PROC_I(inode)->sysctl, NULL);
 		sysctl_head_put(head);
 	}
-	/* Release any associated namespace */
-	ns_ops = PROC_I(inode)->ns.ns_ops;
-	ns = PROC_I(inode)->ns.ns;
-	if (ns_ops && ns)
-		ns_ops->put(ns);
 }
 
 static struct kmem_cache * proc_inode_cachep;
@@ -73,8 +66,7 @@ static struct inode *proc_alloc_inode(struct super_block *sb)
 	ei->pde = NULL;
 	ei->sysctl = NULL;
 	ei->sysctl_entry = NULL;
-	ei->ns.ns = NULL;
-	ei->ns.ns_ops = NULL;
+	ei->ns_ops = NULL;
 	inode = &ei->vfs_inode;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
 	return inode;

fs/proc/internal.h

@@ -65,7 +65,7 @@ struct proc_inode {
 	struct proc_dir_entry *pde;
 	struct ctl_table_header *sysctl;
 	struct ctl_table *sysctl_entry;
-	struct proc_ns ns;
+	const struct proc_ns_operations *ns_ops;
 	struct inode vfs_inode;
 };
 

fs/proc/namespaces.c

@@ -1,10 +1,6 @@
 #include <linux/proc_fs.h>
 #include <linux/nsproxy.h>
-#include <linux/sched.h>
 #include <linux/ptrace.h>
-#include <linux/fs_struct.h>
-#include <linux/mount.h>
-#include <linux/path.h>
 #include <linux/namei.h>
 #include <linux/file.h>
 #include <linux/utsname.h>
@@ -34,138 +30,45 @@ static const struct proc_ns_operations *ns_entries[] = {
 	&mntns_operations,
 };
 
-static const struct file_operations ns_file_operations = {
-	.llseek		= no_llseek,
-};
-
-static const struct inode_operations ns_inode_operations = {
-	.setattr	= proc_setattr,
-};
-
-static char *ns_dname(struct dentry *dentry, char *buffer, int buflen)
-{
-	struct inode *inode = dentry->d_inode;
-	const struct proc_ns_operations *ns_ops = PROC_I(inode)->ns.ns_ops;
-
-	return dynamic_dname(dentry, buffer, buflen, "%s:[%lu]",
-		ns_ops->name, inode->i_ino);
-}
-
-const struct dentry_operations ns_dentry_operations =
-{
-	.d_delete	= always_delete_dentry,
-	.d_dname	= ns_dname,
-};
-
-static struct dentry *proc_ns_get_dentry(struct super_block *sb,
-	struct task_struct *task, const struct proc_ns_operations *ns_ops)
-{
-	struct dentry *dentry, *result;
-	struct inode *inode;
-	struct proc_inode *ei;
-	struct qstr qname = { .name = "", };
-	void *ns;
-
-	ns = ns_ops->get(task);
-	if (!ns)
-		return ERR_PTR(-ENOENT);
-
-	dentry = d_alloc_pseudo(sb, &qname);
-	if (!dentry) {
-		ns_ops->put(ns);
-		return ERR_PTR(-ENOMEM);
-	}
-
-	inode = iget_locked(sb, ns_ops->inum(ns));
-	if (!inode) {
-		dput(dentry);
-		ns_ops->put(ns);
-		return ERR_PTR(-ENOMEM);
-	}
-
-	ei = PROC_I(inode);
-	if (inode->i_state & I_NEW) {
-		inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
-		inode->i_op = &ns_inode_operations;
-		inode->i_mode = S_IFREG | S_IRUGO;
-		inode->i_fop = &ns_file_operations;
-		ei->ns.ns_ops = ns_ops;
-		ei->ns.ns = ns;
-		unlock_new_inode(inode);
-	} else {
-		ns_ops->put(ns);
-	}
-
-	d_set_d_op(dentry, &ns_dentry_operations);
-	result = d_instantiate_unique(dentry, inode);
-	if (result) {
-		dput(dentry);
-		dentry = result;
-	}
-
-	return dentry;
-}
-
 static void *proc_ns_follow_link(struct dentry *dentry, struct nameidata *nd)
 {
 	struct inode *inode = dentry->d_inode;
-	struct super_block *sb = inode->i_sb;
-	struct proc_inode *ei = PROC_I(inode);
+	const struct proc_ns_operations *ns_ops = PROC_I(inode)->ns_ops;
 	struct task_struct *task;
 	struct path ns_path;
 	void *error = ERR_PTR(-EACCES);
 
 	task = get_proc_task(inode);
 	if (!task)
-		goto out;
+		return error;
 
-	if (!ptrace_may_access(task, PTRACE_MODE_READ))
-		goto out_put_task;
-
-	ns_path.dentry = proc_ns_get_dentry(sb, task, ei->ns.ns_ops);
-	if (IS_ERR(ns_path.dentry)) {
-		error = ERR_CAST(ns_path.dentry);
-		goto out_put_task;
+	if (ptrace_may_access(task, PTRACE_MODE_READ)) {
+		error = ns_get_path(&ns_path, task, ns_ops);
+		if (!error)
+			nd_jump_link(nd, &ns_path);
 	}
-
-	ns_path.mnt = mntget(nd->path.mnt);
-	nd_jump_link(nd, &ns_path);
-	error = NULL;
-
-out_put_task:
 	put_task_struct(task);
-out:
 	return error;
 }
 
 static int proc_ns_readlink(struct dentry *dentry, char __user *buffer, int buflen)
 {
 	struct inode *inode = dentry->d_inode;
-	struct proc_inode *ei = PROC_I(inode);
-	const struct proc_ns_operations *ns_ops = ei->ns.ns_ops;
+	const struct proc_ns_operations *ns_ops = PROC_I(inode)->ns_ops;
 	struct task_struct *task;
-	void *ns;
 	char name[50];
 	int res = -EACCES;
 
 	task = get_proc_task(inode);
 	if (!task)
-		goto out;
+		return res;
 
-	if (!ptrace_may_access(task, PTRACE_MODE_READ))
-		goto out_put_task;
-
-	res = -ENOENT;
-	ns = ns_ops->get(task);
-	if (!ns)
-		goto out_put_task;
-
-	snprintf(name, sizeof(name), "%s:[%u]", ns_ops->name, ns_ops->inum(ns));
-	res = readlink_copy(buffer, buflen, name);
-	ns_ops->put(ns);
-out_put_task:
+	if (ptrace_may_access(task, PTRACE_MODE_READ)) {
+		res = ns_get_name(name, sizeof(name), task, ns_ops);
+		if (res >= 0)
+			res = readlink_copy(buffer, buflen, name);
+	}
 	put_task_struct(task);
-out:
 	return res;
 }
 
@@ -189,7 +92,7 @@ static int proc_ns_instantiate(struct inode *dir,
 	ei = PROC_I(inode);
 	inode->i_mode = S_IFLNK|S_IRWXUGO;
 	inode->i_op = &proc_ns_link_inode_operations;
-	ei->ns.ns_ops = ns_ops;
+	ei->ns_ops = ns_ops;
 
 	d_set_d_op(dentry, &pid_dentry_operations);
 	d_add(dentry, inode);
@@ -267,31 +170,3 @@ const struct inode_operations proc_ns_dir_inode_operations = {
 	.getattr	= pid_getattr,
 	.setattr	= proc_setattr,
 };
-
-struct file *proc_ns_fget(int fd)
-{
-	struct file *file;
-
-	file = fget(fd);
-	if (!file)
-		return ERR_PTR(-EBADF);
-
-	if (file->f_op != &ns_file_operations)
-		goto out_invalid;
-
-	return file;
-
-out_invalid:
-	fput(file);
-	return ERR_PTR(-EINVAL);
-}
-
-struct proc_ns *get_proc_ns(struct inode *inode)
-{
-	return &PROC_I(inode)->ns;
-}
-
-bool proc_ns_inode(struct inode *inode)
-{
-	return inode->i_fop == &ns_file_operations;
-}

include/linux/fs.h

@@ -2176,7 +2176,6 @@ static inline int sb_is_blkdev_sb(struct super_block *sb)
 extern int sync_filesystem(struct super_block *);
 extern const struct file_operations def_blk_fops;
 extern const struct file_operations def_chr_fops;
-extern const struct file_operations bad_sock_fops;
 #ifdef CONFIG_BLOCK
 extern int ioctl_by_bdev(struct block_device *, unsigned, unsigned long);
 extern int blkdev_ioctl(struct block_device *, fmode_t, unsigned, unsigned long);

include/linux/ipc_namespace.h

@@ -6,6 +6,7 @@
 #include <linux/rwsem.h>
 #include <linux/notifier.h>
 #include <linux/nsproxy.h>
+#include <linux/ns_common.h>
 
 struct user_namespace;
 
@@ -58,7 +59,7 @@ struct ipc_namespace {
 	/* user_ns which owns the ipc ns */
 	struct user_namespace *user_ns;
 
-	unsigned int	proc_inum;
+	struct ns_common ns;
 };
 
 extern struct ipc_namespace init_ipc_ns;

include/linux/namei.h

@@ -7,21 +7,10 @@
 #include <linux/path.h>
 
 struct vfsmount;
+struct nameidata;
 
 enum { MAX_NESTED_LINKS = 8 };
 
-struct nameidata {
-	struct path	path;
-	struct qstr	last;
-	struct path	root;
-	struct inode	*inode; /* path.dentry.d_inode */
-	unsigned int	flags;
-	unsigned	seq, m_seq;
-	int		last_type;
-	unsigned	depth;
-	char *saved_names[MAX_NESTED_LINKS + 1];
-};
-
 /*
  * Type of the last component on LOOKUP_PARENT
  */
@@ -82,16 +71,8 @@ extern struct dentry *lock_rename(struct dentry *, struct dentry *);
 extern void unlock_rename(struct dentry *, struct dentry *);
 
 extern void nd_jump_link(struct nameidata *nd, struct path *path);
-
-static inline void nd_set_link(struct nameidata *nd, char *path)
-{
-	nd->saved_names[nd->depth] = path;
-}
-
-static inline char *nd_get_link(struct nameidata *nd)
-{
-	return nd->saved_names[nd->depth];
-}
+extern void nd_set_link(struct nameidata *nd, char *path);
+extern char *nd_get_link(struct nameidata *nd);
 
 static inline void nd_terminate_link(void *name, size_t len, size_t maxlen)
 {

include/linux/ns_common.h

@@ -0,0 +1,12 @@
+#ifndef _LINUX_NS_COMMON_H
+#define _LINUX_NS_COMMON_H
+
+struct proc_ns_operations;
+
+struct ns_common {
+	atomic_long_t stashed;
+	const struct proc_ns_operations *ops;
+	unsigned int inum;
+};
+
+#endif

include/linux/pid_namespace.h

@@ -8,6 +8,7 @@
 #include <linux/threads.h>
 #include <linux/nsproxy.h>
 #include <linux/kref.h>
+#include <linux/ns_common.h>
 
 struct pidmap {
        atomic_t nr_free;
@@ -43,7 +44,7 @@ struct pid_namespace {
 	kgid_t pid_gid;
 	int hide_pid;
 	int reboot;	/* group exit code if this pidns was rebooted */
-	unsigned int proc_inum;
+	struct ns_common ns;
 };
 
 extern struct pid_namespace init_pid_ns;

include/linux/proc_ns.h

@@ -4,21 +4,18 @@
 #ifndef _LINUX_PROC_NS_H
 #define _LINUX_PROC_NS_H
 
+#include <linux/ns_common.h>
+
 struct pid_namespace;
 struct nsproxy;
+struct path;
 
 struct proc_ns_operations {
 	const char *name;
 	int type;
-	void *(*get)(struct task_struct *task);
-	void (*put)(void *ns);
-	int (*install)(struct nsproxy *nsproxy, void *ns);
-	unsigned int (*inum)(void *ns);
-};
-
-struct proc_ns {
-	void *ns;
-	const struct proc_ns_operations *ns_ops;
+	struct ns_common *(*get)(struct task_struct *task);
+	void (*put)(struct ns_common *ns);
+	int (*install)(struct nsproxy *nsproxy, struct ns_common *ns);
 };
 
 extern const struct proc_ns_operations netns_operations;
@@ -43,32 +40,38 @@ enum {
 
 extern int pid_ns_prepare_proc(struct pid_namespace *ns);
 extern void pid_ns_release_proc(struct pid_namespace *ns);
-extern struct file *proc_ns_fget(int fd);
-extern struct proc_ns *get_proc_ns(struct inode *);
 extern int proc_alloc_inum(unsigned int *pino);
 extern void proc_free_inum(unsigned int inum);
-extern bool proc_ns_inode(struct inode *inode);
 
 #else /* CONFIG_PROC_FS */
 
 static inline int pid_ns_prepare_proc(struct pid_namespace *ns) { return 0; }
 static inline void pid_ns_release_proc(struct pid_namespace *ns) {}
 
-static inline struct file *proc_ns_fget(int fd)
-{
-	return ERR_PTR(-EINVAL);
-}
-
-static inline struct proc_ns *get_proc_ns(struct inode *inode) { return NULL; }
-
 static inline int proc_alloc_inum(unsigned int *inum)
 {
 	*inum = 1;
 	return 0;
 }
 static inline void proc_free_inum(unsigned int inum) {}
-static inline bool proc_ns_inode(struct inode *inode) { return false; }
 
 #endif /* CONFIG_PROC_FS */
 
+static inline int ns_alloc_inum(struct ns_common *ns)
+{
+	atomic_long_set(&ns->stashed, 0);
+	return proc_alloc_inum(&ns->inum);
+}
+
+#define ns_free_inum(ns) proc_free_inum((ns)->inum)
+
+extern struct file *proc_ns_fget(int fd);
+#define get_proc_ns(inode) ((struct ns_common *)(inode)->i_private)
+extern void *ns_get_path(struct path *path, struct task_struct *task,
+			const struct proc_ns_operations *ns_ops);
+
+extern int ns_get_name(char *buf, size_t size, struct task_struct *task,
+			const struct proc_ns_operations *ns_ops);
+extern void nsfs_init(void);
+
 #endif /* _LINUX_PROC_NS_H */

include/linux/user_namespace.h

@@ -3,6 +3,7 @@
 
 #include <linux/kref.h>
 #include <linux/nsproxy.h>
+#include <linux/ns_common.h>
 #include <linux/sched.h>
 #include <linux/err.h>
 
@@ -26,7 +27,7 @@ struct user_namespace {
 	int			level;
 	kuid_t			owner;
 	kgid_t			group;
-	unsigned int		proc_inum;
+	struct ns_common	ns;
 
 	/* Register of per-UID persistent keyrings for this namespace */
 #ifdef CONFIG_PERSISTENT_KEYRINGS

include/linux/utsname.h

@@ -5,6 +5,7 @@
 #include <linux/sched.h>
 #include <linux/kref.h>
 #include <linux/nsproxy.h>
+#include <linux/ns_common.h>
 #include <linux/err.h>
 #include <uapi/linux/utsname.h>
 
@@ -23,7 +24,7 @@ struct uts_namespace {
 	struct kref kref;
 	struct new_utsname name;
 	struct user_namespace *user_ns;
-	unsigned int proc_inum;
+	struct ns_common ns;
 };
 extern struct uts_namespace init_uts_ns;