From 15a9155fe3e8215c02b80df51ec2cac7c0d726ad Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 16 Feb 2011 15:08:54 -0500
Subject: fix race in audit_get_nd()

don't rely on pathname resolution ending up twice at the same point...

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 kernel/audit_watch.c | 85 ++++++++++++++++++++--------------------------------
 1 file changed, 32 insertions(+), 53 deletions(-)

(limited to 'kernel')

diff --git a/kernel/audit_watch.c b/kernel/audit_watch.c
index d2e3c7866460..20b9fe6907d0 100644
--- a/kernel/audit_watch.c
+++ b/kernel/audit_watch.c
@@ -144,9 +144,9 @@ int audit_watch_compare(struct audit_watch *watch, unsigned long ino, dev_t dev)
 }
 
 /* Initialize a parent watch entry. */
-static struct audit_parent *audit_init_parent(struct nameidata *ndp)
+static struct audit_parent *audit_init_parent(struct path *path)
 {
-	struct inode *inode = ndp->path.dentry->d_inode;
+	struct inode *inode = path->dentry->d_inode;
 	struct audit_parent *parent;
 	int ret;
 
@@ -353,53 +353,40 @@ static void audit_remove_parent_watches(struct audit_parent *parent)
 }
 
 /* Get path information necessary for adding watches. */
-static int audit_get_nd(char *path, struct nameidata **ndp, struct nameidata **ndw)
+static int audit_get_nd(struct audit_watch *watch, struct path *parent)
 {
-	struct nameidata *ndparent, *ndwatch;
+	struct nameidata nd;
+	struct dentry *d;
 	int err;
 
-	ndparent = kmalloc(sizeof(*ndparent), GFP_KERNEL);
-	if (unlikely(!ndparent))
-		return -ENOMEM;
+	err = path_lookup(watch->path, LOOKUP_PARENT, &nd);
+	if (err)
+		return err;
 
-	ndwatch = kmalloc(sizeof(*ndwatch), GFP_KERNEL);
-	if (unlikely(!ndwatch)) {
-		kfree(ndparent);
-		return -ENOMEM;
+	if (nd.last_type != LAST_NORM) {
+		path_put(&nd.path);
+		return -EINVAL;
 	}
 
-	err = path_lookup(path, LOOKUP_PARENT, ndparent);
-	if (err) {
-		kfree(ndparent);
-		kfree(ndwatch);
-		return err;
+	mutex_lock_nested(&nd.path.dentry->d_inode->i_mutex, I_MUTEX_PARENT);
+	d = lookup_one_len(nd.last.name, nd.path.dentry, nd.last.len);
+	if (IS_ERR(d)) {
+		mutex_unlock(&nd.path.dentry->d_inode->i_mutex);
+		path_put(&nd.path);
+		return PTR_ERR(d);
 	}
-
-	err = path_lookup(path, 0, ndwatch);
-	if (err) {
-		kfree(ndwatch);
-		ndwatch = NULL;
+	if (d->d_inode) {
+		/* update watch filter fields */
+		watch->dev = d->d_inode->i_sb->s_dev;
+		watch->ino = d->d_inode->i_ino;
 	}
+	mutex_unlock(&nd.path.dentry->d_inode->i_mutex);
 
-	*ndp = ndparent;
-	*ndw = ndwatch;
-
+	*parent = nd.path;
+	dput(d);
 	return 0;
 }
 
-/* Release resources used for watch path information. */
-static void audit_put_nd(struct nameidata *ndp, struct nameidata *ndw)
-{
-	if (ndp) {
-		path_put(&ndp->path);
-		kfree(ndp);
-	}
-	if (ndw) {
-		path_put(&ndw->path);
-		kfree(ndw);
-	}
-}
-
 /* Associate the given rule with an existing parent.
  * Caller must hold audit_filter_mutex. */
 static void audit_add_to_parent(struct audit_krule *krule,
@@ -440,31 +427,24 @@ int audit_add_watch(struct audit_krule *krule, struct list_head **list)
 {
 	struct audit_watch *watch = krule->watch;
 	struct audit_parent *parent;
-	struct nameidata *ndp = NULL, *ndw = NULL;
+	struct path parent_path;
 	int h, ret = 0;
 
 	mutex_unlock(&audit_filter_mutex);
 
 	/* Avoid calling path_lookup under audit_filter_mutex. */
-	ret = audit_get_nd(watch->path, &ndp, &ndw);
-	if (ret) {
-		/* caller expects mutex locked */
-		mutex_lock(&audit_filter_mutex);
-		goto error;
-	}
+	ret = audit_get_nd(watch, &parent_path);
 
+	/* caller expects mutex locked */
 	mutex_lock(&audit_filter_mutex);
 
-	/* update watch filter fields */
-	if (ndw) {
-		watch->dev = ndw->path.dentry->d_inode->i_sb->s_dev;
-		watch->ino = ndw->path.dentry->d_inode->i_ino;
-	}
+	if (ret)
+		return ret;
 
 	/* either find an old parent or attach a new one */
-	parent = audit_find_parent(ndp->path.dentry->d_inode);
+	parent = audit_find_parent(parent_path.dentry->d_inode);
 	if (!parent) {
-		parent = audit_init_parent(ndp);
+		parent = audit_init_parent(&parent_path);
 		if (IS_ERR(parent)) {
 			ret = PTR_ERR(parent);
 			goto error;
@@ -479,9 +459,8 @@ int audit_add_watch(struct audit_krule *krule, struct list_head **list)
 	h = audit_hash_ino((u32)watch->ino);
 	*list = &audit_inode_hash[h];
 error:
-	audit_put_nd(ndp, ndw);		/* NULL args OK */
+	path_put(&parent_path);
 	return ret;
-
 }
 
 void audit_remove_watch_rule(struct audit_krule *krule)
-- 
cgit v1.2.3


From c9c6cac0c2bdbda42e7b804838648d0bc60ddb13 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 16 Feb 2011 15:15:47 -0500
Subject: kill path_lookup()

all remaining callers pass LOOKUP_PARENT to it, so
flags argument can die; renamed to kern_path_parent()

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 arch/powerpc/platforms/cell/spufs/syscalls.c | 2 +-
 fs/namei.c                                   | 7 +++----
 fs/ocfs2/refcounttree.c                      | 2 +-
 include/linux/namei.h                        | 2 +-
 kernel/audit_watch.c                         | 2 +-
 net/unix/af_unix.c                           | 2 +-
 6 files changed, 8 insertions(+), 9 deletions(-)

(limited to 'kernel')

diff --git a/arch/powerpc/platforms/cell/spufs/syscalls.c b/arch/powerpc/platforms/cell/spufs/syscalls.c
index 187a7d32f86a..a3d2ce54ea2e 100644
--- a/arch/powerpc/platforms/cell/spufs/syscalls.c
+++ b/arch/powerpc/platforms/cell/spufs/syscalls.c
@@ -70,7 +70,7 @@ static long do_spu_create(const char __user *pathname, unsigned int flags,
 	if (!IS_ERR(tmp)) {
 		struct nameidata nd;
 
-		ret = path_lookup(tmp, LOOKUP_PARENT, &nd);
+		ret = kern_path_parent(tmp, &nd);
 		if (!ret) {
 			nd.flags |= LOOKUP_OPEN | LOOKUP_CREATE;
 			ret = spufs_create(&nd, flags, mode, neighbor);
diff --git a/fs/namei.c b/fs/namei.c
index a4689eb2df28..1d6bc8151553 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -1747,10 +1747,9 @@ static int do_path_lookup(int dfd, const char *name,
 	return retval;
 }
 
-int path_lookup(const char *name, unsigned int flags,
-			struct nameidata *nd)
+int kern_path_parent(const char *name, struct nameidata *nd)
 {
-	return do_path_lookup(AT_FDCWD, name, flags, nd);
+	return do_path_lookup(AT_FDCWD, name, LOOKUP_PARENT, nd);
 }
 
 int kern_path(const char *name, unsigned int flags, struct path *path)
@@ -3586,7 +3585,7 @@ EXPORT_SYMBOL(page_readlink);
 EXPORT_SYMBOL(__page_symlink);
 EXPORT_SYMBOL(page_symlink);
 EXPORT_SYMBOL(page_symlink_inode_operations);
-EXPORT_SYMBOL(path_lookup);
+EXPORT_SYMBOL(kern_path_parent);
 EXPORT_SYMBOL(kern_path);
 EXPORT_SYMBOL(vfs_path_lookup);
 EXPORT_SYMBOL(inode_permission);
diff --git a/fs/ocfs2/refcounttree.c b/fs/ocfs2/refcounttree.c
index 19ebc5aad391..29623da133cc 100644
--- a/fs/ocfs2/refcounttree.c
+++ b/fs/ocfs2/refcounttree.c
@@ -4379,7 +4379,7 @@ static int ocfs2_user_path_parent(const char __user *path,
 	if (IS_ERR(s))
 		return PTR_ERR(s);
 
-	error = path_lookup(s, LOOKUP_PARENT, nd);
+	error = kern_path_parent(s, nd);
 	if (error)
 		putname(s);
 	else
diff --git a/include/linux/namei.h b/include/linux/namei.h
index f276d4fa01fc..58ce3433d4ec 100644
--- a/include/linux/namei.h
+++ b/include/linux/namei.h
@@ -72,7 +72,7 @@ extern int user_path_at(int, const char __user *, unsigned, struct path *);
 
 extern int kern_path(const char *, unsigned, struct path *);
 
-extern int path_lookup(const char *, unsigned, struct nameidata *);
+extern int kern_path_parent(const char *, struct nameidata *);
 extern int vfs_path_lookup(struct dentry *, struct vfsmount *,
 			   const char *, unsigned int, struct nameidata *);
 
diff --git a/kernel/audit_watch.c b/kernel/audit_watch.c
index 20b9fe6907d0..e683869365d9 100644
--- a/kernel/audit_watch.c
+++ b/kernel/audit_watch.c
@@ -359,7 +359,7 @@ static int audit_get_nd(struct audit_watch *watch, struct path *parent)
 	struct dentry *d;
 	int err;
 
-	err = path_lookup(watch->path, LOOKUP_PARENT, &nd);
+	err = kern_path_parent(watch->path, &nd);
 	if (err)
 		return err;
 
diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index dd419d286204..d8c04a602cf1 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -850,7 +850,7 @@ static int unix_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len)
 		 * Get the parent directory, calculate the hash for last
 		 * component.
 		 */
-		err = path_lookup(sunaddr->sun_path, LOOKUP_PARENT, &nd);
+		err = kern_path_parent(sunaddr->sun_path, &nd);
 		if (err)
 			goto out_mknod_parent;
 
-- 
cgit v1.2.3


From 73d049a40fc6269189c4e2ba6792cb5dd054883c Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Fri, 11 Mar 2011 12:08:24 -0500
Subject: open-style analog of vfs_path_lookup()

new function: file_open_root(dentry, mnt, name, flags) opens the file
vfs_path_lookup would arrive to.

Note that name can be empty; in that case the usual requirement that
dentry should be a directory is lifted.

open-coded equivalents switched to it, may_open() got down exactly
one caller and became static.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 arch/um/drivers/mconsole_kern.c | 21 ++---------
 fs/internal.h                   |  2 ++
 fs/namei.c                      | 80 ++++++++++++++++++++++++++---------------
 fs/nfsctl.c                     | 21 +++--------
 fs/open.c                       | 14 ++++++++
 include/linux/fs.h              |  4 +--
 kernel/sysctl_binary.c          | 19 +---------
 7 files changed, 77 insertions(+), 84 deletions(-)

(limited to 'kernel')

diff --git a/arch/um/drivers/mconsole_kern.c b/arch/um/drivers/mconsole_kern.c
index 975613b23dcf..c70e047eed72 100644
--- a/arch/um/drivers/mconsole_kern.c
+++ b/arch/um/drivers/mconsole_kern.c
@@ -124,35 +124,18 @@ void mconsole_log(struct mc_request *req)
 #if 0
 void mconsole_proc(struct mc_request *req)
 {
-	struct nameidata nd;
 	struct vfsmount *mnt = current->nsproxy->pid_ns->proc_mnt;
 	struct file *file;
-	int n, err;
+	int n;
 	char *ptr = req->request.data, *buf;
 	mm_segment_t old_fs = get_fs();
 
 	ptr += strlen("proc");
 	ptr = skip_spaces(ptr);
 
-	err = vfs_path_lookup(mnt->mnt_root, mnt, ptr, LOOKUP_FOLLOW, &nd);
-	if (err) {
-		mconsole_reply(req, "Failed to look up file", 1, 0);
-		goto out;
-	}
-
-	err = may_open(&nd.path, MAY_READ, O_RDONLY);
-	if (result) {
-		mconsole_reply(req, "Failed to open file", 1, 0);
-		path_put(&nd.path);
-		goto out;
-	}
-
-	file = dentry_open(nd.path.dentry, nd.path.mnt, O_RDONLY,
-			   current_cred());
-	err = PTR_ERR(file);
+	file = file_open_root(mnt->mnt_root, mnt, ptr, O_RDONLY);
 	if (IS_ERR(file)) {
 		mconsole_reply(req, "Failed to open file", 1, 0);
-		path_put(&nd.path);
 		goto out;
 	}
 
diff --git a/fs/internal.h b/fs/internal.h
index 6fdbdf2c6047..52abc5287f50 100644
--- a/fs/internal.h
+++ b/fs/internal.h
@@ -114,6 +114,8 @@ struct open_flags {
 };
 extern struct file *do_filp_open(int dfd, const char *pathname,
 		const struct open_flags *op, int lookup_flags);
+extern struct file *do_file_open_root(struct dentry *, struct vfsmount *,
+		const char *, const struct open_flags *, int lookup_flags);
 
 /*
  * inode.c
diff --git a/fs/namei.c b/fs/namei.c
index 8ee7785d5642..abc8d2df121c 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -1487,11 +1487,13 @@ static int path_init(int dfd, const char *name, unsigned int flags,
 	nd->depth = 0;
 	if (flags & LOOKUP_ROOT) {
 		struct inode *inode = nd->root.dentry->d_inode;
-		if (!inode->i_op->lookup)
-			return -ENOTDIR;
-		retval = inode_permission(inode, MAY_EXEC);
-		if (retval)
-			return retval;
+		if (*name) {
+			if (!inode->i_op->lookup)
+				return -ENOTDIR;
+			retval = inode_permission(inode, MAY_EXEC);
+			if (retval)
+				return retval;
+		}
 		nd->path = nd->root;
 		nd->inode = inode;
 		if (flags & LOOKUP_RCU) {
@@ -1937,7 +1939,7 @@ int vfs_create(struct inode *dir, struct dentry *dentry, int mode,
 	return error;
 }
 
-int may_open(struct path *path, int acc_mode, int flag)
+static int may_open(struct path *path, int acc_mode, int flag)
 {
 	struct dentry *dentry = path->dentry;
 	struct inode *inode = dentry->d_inode;
@@ -2250,11 +2252,10 @@ exit:
 }
 
 static struct file *path_openat(int dfd, const char *pathname,
-		const struct open_flags *op, int flags)
+		struct nameidata *nd, const struct open_flags *op, int flags)
 {
 	struct file *base = NULL;
 	struct file *filp;
-	struct nameidata nd;
 	struct path path;
 	int count = 0;
 	int error;
@@ -2264,27 +2265,27 @@ static struct file *path_openat(int dfd, const char *pathname,
 		return ERR_PTR(-ENFILE);
 
 	filp->f_flags = op->open_flag;
-	nd.intent.open.file = filp;
-	nd.intent.open.flags = open_to_namei_flags(op->open_flag);
-	nd.intent.open.create_mode = op->mode;
+	nd->intent.open.file = filp;
+	nd->intent.open.flags = open_to_namei_flags(op->open_flag);
+	nd->intent.open.create_mode = op->mode;
 
-	error = path_init(dfd, pathname, flags | LOOKUP_PARENT, &nd, &base);
+	error = path_init(dfd, pathname, flags | LOOKUP_PARENT, nd, &base);
 	if (unlikely(error))
 		goto out_filp;
 
 	current->total_link_count = 0;
-	error = link_path_walk(pathname, &nd);
+	error = link_path_walk(pathname, nd);
 	if (unlikely(error))
 		goto out_filp;
 
-	filp = do_last(&nd, &path, op, pathname);
+	filp = do_last(nd, &path, op, pathname);
 	while (unlikely(!filp)) { /* trailing symlink */
 		struct path link = path;
 		struct inode *linki = link.dentry->d_inode;
 		void *cookie;
-		if (!(nd.flags & LOOKUP_FOLLOW) || count++ == 32) {
-			path_put_conditional(&path, &nd);
-			path_put(&nd.path);
+		if (!(nd->flags & LOOKUP_FOLLOW) || count++ == 32) {
+			path_put_conditional(&path, nd);
+			path_put(&nd->path);
 			filp = ERR_PTR(-ELOOP);
 			break;
 		}
@@ -2299,23 +2300,23 @@ static struct file *path_openat(int dfd, const char *pathname,
 		 * have to putname() it when we are done. Procfs-like symlinks
 		 * just set LAST_BIND.
 		 */
-		nd.flags |= LOOKUP_PARENT;
-		nd.flags &= ~(LOOKUP_OPEN|LOOKUP_CREATE|LOOKUP_EXCL);
-		error = __do_follow_link(&link, &nd, &cookie);
+		nd->flags |= LOOKUP_PARENT;
+		nd->flags &= ~(LOOKUP_OPEN|LOOKUP_CREATE|LOOKUP_EXCL);
+		error = __do_follow_link(&link, nd, &cookie);
 		if (unlikely(error))
 			filp = ERR_PTR(error);
 		else
-			filp = do_last(&nd, &path, op, pathname);
+			filp = do_last(nd, &path, op, pathname);
 		if (!IS_ERR(cookie) && linki->i_op->put_link)
-			linki->i_op->put_link(link.dentry, &nd, cookie);
+			linki->i_op->put_link(link.dentry, nd, cookie);
 		path_put(&link);
 	}
 out:
-	if (nd.root.mnt && !(nd.flags & LOOKUP_ROOT))
-		path_put(&nd.root);
+	if (nd->root.mnt && !(nd->flags & LOOKUP_ROOT))
+		path_put(&nd->root);
 	if (base)
 		fput(base);
-	release_open_intent(&nd);
+	release_open_intent(nd);
 	return filp;
 
 out_filp:
@@ -2326,16 +2327,39 @@ out_filp:
 struct file *do_filp_open(int dfd, const char *pathname,
 		const struct open_flags *op, int flags)
 {
+	struct nameidata nd;
 	struct file *filp;
 
-	filp = path_openat(dfd, pathname, op, flags | LOOKUP_RCU);
+	filp = path_openat(dfd, pathname, &nd, op, flags | LOOKUP_RCU);
 	if (unlikely(filp == ERR_PTR(-ECHILD)))
-		filp = path_openat(dfd, pathname, op, flags);
+		filp = path_openat(dfd, pathname, &nd, op, flags);
 	if (unlikely(filp == ERR_PTR(-ESTALE)))
-		filp = path_openat(dfd, pathname, op, flags | LOOKUP_REVAL);
+		filp = path_openat(dfd, pathname, &nd, op, flags | LOOKUP_REVAL);
 	return filp;
 }
 
+struct file *do_file_open_root(struct dentry *dentry, struct vfsmount *mnt,
+		const char *name, const struct open_flags *op, int flags)
+{
+	struct nameidata nd;
+	struct file *file;
+
+	nd.root.mnt = mnt;
+	nd.root.dentry = dentry;
+
+	flags |= LOOKUP_ROOT;
+
+	if (dentry->d_inode->i_op->follow_link)
+		return ERR_PTR(-ELOOP);
+
+	file = path_openat(-1, name, &nd, op, flags | LOOKUP_RCU);
+	if (unlikely(file == ERR_PTR(-ECHILD)))
+		file = path_openat(-1, name, &nd, op, flags);
+	if (unlikely(file == ERR_PTR(-ESTALE)))
+		file = path_openat(-1, name, &nd, op, flags | LOOKUP_REVAL);
+	return file;
+}
+
 /**
  * lookup_create - lookup a dentry, creating it if it doesn't exist
  * @nd: nameidata info
diff --git a/fs/nfsctl.c b/fs/nfsctl.c
index bf9cbd242ddd..124e8fcb0dd6 100644
--- a/fs/nfsctl.c
+++ b/fs/nfsctl.c
@@ -22,30 +22,17 @@
 
 static struct file *do_open(char *name, int flags)
 {
-	struct nameidata nd;
 	struct vfsmount *mnt;
-	int error;
+	struct file *file;
 
 	mnt = do_kern_mount("nfsd", 0, "nfsd", NULL);
 	if (IS_ERR(mnt))
 		return (struct file *)mnt;
 
-	error = vfs_path_lookup(mnt->mnt_root, mnt, name, 0, &nd);
-	mntput(mnt);	/* drop do_kern_mount reference */
-	if (error)
-		return ERR_PTR(error);
-
-	if (flags == O_RDWR)
-		error = may_open(&nd.path, MAY_READ|MAY_WRITE, flags);
-	else
-		error = may_open(&nd.path, MAY_WRITE, flags);
+	file = file_open_root(mnt->mnt_root, mnt, name, flags);
 
-	if (!error)
-		return dentry_open(nd.path.dentry, nd.path.mnt, flags,
-				   current_cred());
-
-	path_put(&nd.path);
-	return ERR_PTR(error);
+	mntput(mnt);	/* drop do_kern_mount reference */
+	return file;
 }
 
 static struct {
diff --git a/fs/open.c b/fs/open.c
index d05e18c60bae..48afc5c139d2 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -959,6 +959,20 @@ struct file *filp_open(const char *filename, int flags, int mode)
 }
 EXPORT_SYMBOL(filp_open);
 
+struct file *file_open_root(struct dentry *dentry, struct vfsmount *mnt,
+			    const char *filename, int flags)
+{
+	struct open_flags op;
+	int lookup = build_open_flags(flags, 0, &op);
+	if (flags & O_CREAT)
+		return ERR_PTR(-EINVAL);
+	if (!filename && (flags & O_DIRECTORY))
+		if (!dentry->d_inode->i_op->lookup)
+			return ERR_PTR(-ENOTDIR);
+	return do_file_open_root(dentry, mnt, filename, &op, lookup);
+}
+EXPORT_SYMBOL(file_open_root);
+
 long do_sys_open(int dfd, const char __user *filename, int flags, int mode)
 {
 	struct open_flags op;
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 9c75714f92c1..bf5c3c896072 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1990,6 +1990,8 @@ extern int do_fallocate(struct file *file, int mode, loff_t offset,
 extern long do_sys_open(int dfd, const char __user *filename, int flags,
 			int mode);
 extern struct file *filp_open(const char *, int, int);
+extern struct file *file_open_root(struct dentry *, struct vfsmount *,
+				   const char *, int);
 extern struct file * dentry_open(struct dentry *, struct vfsmount *, int,
 				 const struct cred *);
 extern int filp_close(struct file *, fl_owner_t id);
@@ -2205,8 +2207,6 @@ extern struct file *create_read_pipe(struct file *f, int flags);
 extern struct file *create_write_pipe(int flags);
 extern void free_write_pipe(struct file *);
 
-extern int may_open(struct path *, int, int);
-
 extern int kernel_read(struct file *, loff_t, char *, unsigned long);
 extern struct file * open_exec(const char *);
  
diff --git a/kernel/sysctl_binary.c b/kernel/sysctl_binary.c
index b875bedf7c9a..3b8e028b9601 100644
--- a/kernel/sysctl_binary.c
+++ b/kernel/sysctl_binary.c
@@ -1321,13 +1321,11 @@ static ssize_t binary_sysctl(const int *name, int nlen,
 	void __user *oldval, size_t oldlen, void __user *newval, size_t newlen)
 {
 	const struct bin_table *table = NULL;
-	struct nameidata nd;
 	struct vfsmount *mnt;
 	struct file *file;
 	ssize_t result;
 	char *pathname;
 	int flags;
-	int acc_mode;
 
 	pathname = sysctl_getname(name, nlen, &table);
 	result = PTR_ERR(pathname);
@@ -1337,28 +1335,17 @@ static ssize_t binary_sysctl(const int *name, int nlen,
 	/* How should the sysctl be accessed? */
 	if (oldval && oldlen && newval && newlen) {
 		flags = O_RDWR;
-		acc_mode = MAY_READ | MAY_WRITE;
 	} else if (newval && newlen) {
 		flags = O_WRONLY;
-		acc_mode = MAY_WRITE;
 	} else if (oldval && oldlen) {
 		flags = O_RDONLY;
-		acc_mode = MAY_READ;
 	} else {
 		result = 0;
 		goto out_putname;
 	}
 
 	mnt = current->nsproxy->pid_ns->proc_mnt;
-	result = vfs_path_lookup(mnt->mnt_root, mnt, pathname, 0, &nd);
-	if (result)
-		goto out_putname;
-
-	result = may_open(&nd.path, acc_mode, flags);
-	if (result)
-		goto out_putpath;
-
-	file = dentry_open(nd.path.dentry, nd.path.mnt, flags, current_cred());
+	file = file_open_root(mnt->mnt_root, mnt, pathname, flags);
 	result = PTR_ERR(file);
 	if (IS_ERR(file))
 		goto out_putname;
@@ -1370,10 +1357,6 @@ out_putname:
 	putname(pathname);
 out:
 	return result;
-
-out_putpath:
-	path_put(&nd.path);
-	goto out_putname;
 }
 
 
-- 
cgit v1.2.3


From 990d6c2d7aee921e3bce22b2d6a750fd552262be Mon Sep 17 00:00:00 2001
From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Date: Sat, 29 Jan 2011 18:43:26 +0530
Subject: vfs: Add name to file handle conversion support

The syscall also return mount id which can be used
to lookup file system specific information such as uuid
in /proc/<pid>/mountinfo

Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/Kconfig               |   2 +-
 fs/Makefile              |   2 +
 fs/fhandle.c             | 107 +++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/exportfs.h |   3 ++
 include/linux/fs.h       |   7 ++++
 include/linux/syscalls.h |   5 ++-
 init/Kconfig             |  12 ++++++
 kernel/sys_ni.c          |   3 ++
 8 files changed, 139 insertions(+), 2 deletions(-)
 create mode 100644 fs/fhandle.c

(limited to 'kernel')

diff --git a/fs/Kconfig b/fs/Kconfig
index 3db9caa57edc..7cb53aafac1e 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -47,7 +47,7 @@ config FS_POSIX_ACL
 	def_bool n
 
 config EXPORTFS
-	tristate
+	bool
 
 config FILE_LOCKING
 	bool "Enable POSIX file locking API" if EXPERT
diff --git a/fs/Makefile b/fs/Makefile
index a7f7cef0c0c8..ba01202844c5 100644
--- a/fs/Makefile
+++ b/fs/Makefile
@@ -48,6 +48,8 @@ obj-$(CONFIG_FS_POSIX_ACL)	+= posix_acl.o xattr_acl.o
 obj-$(CONFIG_NFS_COMMON)	+= nfs_common/
 obj-$(CONFIG_GENERIC_ACL)	+= generic_acl.o
 
+obj-$(CONFIG_FHANDLE)		+= fhandle.o
+
 obj-y				+= quota/
 
 obj-$(CONFIG_PROC_FS)		+= proc/
diff --git a/fs/fhandle.c b/fs/fhandle.c
new file mode 100644
index 000000000000..9f79e743a840
--- /dev/null
+++ b/fs/fhandle.c
@@ -0,0 +1,107 @@
+#include <linux/syscalls.h>
+#include <linux/slab.h>
+#include <linux/fs.h>
+#include <linux/file.h>
+#include <linux/mount.h>
+#include <linux/namei.h>
+#include <linux/exportfs.h>
+#include <asm/uaccess.h>
+#include "internal.h"
+
+static long do_sys_name_to_handle(struct path *path,
+				  struct file_handle __user *ufh,
+				  int __user *mnt_id)
+{
+	long retval;
+	struct file_handle f_handle;
+	int handle_dwords, handle_bytes;
+	struct file_handle *handle = NULL;
+
+	/*
+	 * We need t make sure wether the file system
+	 * support decoding of the file handle
+	 */
+	if (!path->mnt->mnt_sb->s_export_op ||
+	    !path->mnt->mnt_sb->s_export_op->fh_to_dentry)
+		return -EOPNOTSUPP;
+
+	if (copy_from_user(&f_handle, ufh, sizeof(struct file_handle)))
+		return -EFAULT;
+
+	if (f_handle.handle_bytes > MAX_HANDLE_SZ)
+		return -EINVAL;
+
+	handle = kmalloc(sizeof(struct file_handle) + f_handle.handle_bytes,
+			 GFP_KERNEL);
+	if (!handle)
+		return -ENOMEM;
+
+	/* convert handle size to  multiple of sizeof(u32) */
+	handle_dwords = f_handle.handle_bytes >> 2;
+
+	/* we ask for a non connected handle */
+	retval = exportfs_encode_fh(path->dentry,
+				    (struct fid *)handle->f_handle,
+				    &handle_dwords,  0);
+	handle->handle_type = retval;
+	/* convert handle size to bytes */
+	handle_bytes = handle_dwords * sizeof(u32);
+	handle->handle_bytes = handle_bytes;
+	if ((handle->handle_bytes > f_handle.handle_bytes) ||
+	    (retval == 255) || (retval == -ENOSPC)) {
+		/* As per old exportfs_encode_fh documentation
+		 * we could return ENOSPC to indicate overflow
+		 * But file system returned 255 always. So handle
+		 * both the values
+		 */
+		/*
+		 * set the handle size to zero so we copy only
+		 * non variable part of the file_handle
+		 */
+		handle_bytes = 0;
+		retval = -EOVERFLOW;
+	} else
+		retval = 0;
+	/* copy the mount id */
+	if (copy_to_user(mnt_id, &path->mnt->mnt_id, sizeof(*mnt_id)) ||
+	    copy_to_user(ufh, handle,
+			 sizeof(struct file_handle) + handle_bytes))
+		retval = -EFAULT;
+	kfree(handle);
+	return retval;
+}
+
+/**
+ * sys_name_to_handle_at: convert name to handle
+ * @dfd: directory relative to which name is interpreted if not absolute
+ * @name: name that should be converted to handle.
+ * @handle: resulting file handle
+ * @mnt_id: mount id of the file system containing the file
+ * @flag: flag value to indicate whether to follow symlink or not
+ *
+ * @handle->handle_size indicate the space available to store the
+ * variable part of the file handle in bytes. If there is not
+ * enough space, the field is updated to return the minimum
+ * value required.
+ */
+SYSCALL_DEFINE5(name_to_handle_at, int, dfd, const char __user *, name,
+		struct file_handle __user *, handle, int __user *, mnt_id,
+		int, flag)
+{
+	struct path path;
+	int lookup_flags;
+	int err;
+
+	if ((flag & ~(AT_SYMLINK_FOLLOW | AT_EMPTY_PATH)) != 0)
+		return -EINVAL;
+
+	lookup_flags = (flag & AT_SYMLINK_FOLLOW) ? LOOKUP_FOLLOW : 0;
+	if (flag & AT_EMPTY_PATH)
+		lookup_flags |= LOOKUP_EMPTY;
+	err = user_path_at(dfd, name, lookup_flags, &path);
+	if (!err) {
+		err = do_sys_name_to_handle(&path, handle, mnt_id);
+		path_put(&path);
+	}
+	return err;
+}
diff --git a/include/linux/exportfs.h b/include/linux/exportfs.h
index 65afdfd31b7b..33a42f24b275 100644
--- a/include/linux/exportfs.h
+++ b/include/linux/exportfs.h
@@ -8,6 +8,9 @@ struct inode;
 struct super_block;
 struct vfsmount;
 
+/* limit the handle size to NFSv4 handle size now */
+#define MAX_HANDLE_SZ 128
+
 /*
  * The fileid_type identifies how the file within the filesystem is encoded.
  * In theory this is freely set and parsed by the filesystem, but we try to
diff --git a/include/linux/fs.h b/include/linux/fs.h
index b7178b05cf3a..3f64630c0e10 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -978,6 +978,13 @@ struct file {
 #endif
 };
 
+struct file_handle {
+	__u32 handle_bytes;
+	int handle_type;
+	/* file identifier */
+	unsigned char f_handle[0];
+};
+
 #define get_file(x)	atomic_long_inc(&(x)->f_count)
 #define fput_atomic(x)	atomic_long_add_unless(&(x)->f_count, -1, 1)
 #define file_count(x)	atomic_long_read(&(x)->f_count)
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index 98664db1be47..970112613fb4 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -62,6 +62,7 @@ struct robust_list_head;
 struct getcpu_cache;
 struct old_linux_dirent;
 struct perf_event_attr;
+struct file_handle;
 
 #include <linux/types.h>
 #include <linux/aio_abi.h>
@@ -832,5 +833,7 @@ asmlinkage long sys_mmap_pgoff(unsigned long addr, unsigned long len,
 			unsigned long prot, unsigned long flags,
 			unsigned long fd, unsigned long pgoff);
 asmlinkage long sys_old_mmap(struct mmap_arg_struct __user *arg);
-
+asmlinkage long sys_name_to_handle_at(int dfd, const char __user *name,
+				      struct file_handle __user *handle,
+				      int __user *mnt_id, int flag);
 #endif
diff --git a/init/Kconfig b/init/Kconfig
index be788c0957d4..e72fa17fe559 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -287,6 +287,18 @@ config BSD_PROCESS_ACCT_V3
 	  for processing it. A preliminary version of these tools is available
 	  at <http://www.gnu.org/software/acct/>.
 
+config FHANDLE
+	bool "open by fhandle syscalls"
+	select EXPORTFS
+	help
+	  If you say Y here, a user level program will be able to map
+	  file names to handle and then later use the handle for
+	  different file system operations. This is useful in implementing
+	  userspace file servers, which now track files using handles instead
+	  of names. The handle would remain the same even if file names
+	  get renamed. Enables open_by_handle_at(2) and name_to_handle_at(2)
+	  syscalls.
+
 config TASKSTATS
 	bool "Export task/process statistics through netlink (EXPERIMENTAL)"
 	depends on NET
diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c
index c782fe9924c7..4e013439ac28 100644
--- a/kernel/sys_ni.c
+++ b/kernel/sys_ni.c
@@ -186,3 +186,6 @@ cond_syscall(sys_perf_event_open);
 /* fanotify! */
 cond_syscall(sys_fanotify_init);
 cond_syscall(sys_fanotify_mark);
+
+/* open by handle */
+cond_syscall(sys_name_to_handle_at);
-- 
cgit v1.2.3


From becfd1f37544798cbdfd788f32c827160fab98c1 Mon Sep 17 00:00:00 2001
From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Date: Sat, 29 Jan 2011 18:43:26 +0530
Subject: vfs: Add open by file handle support

[AV: duplicate of open() guts removed; file_open_root() used instead]

Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/compat.c              |  13 ++++
 fs/exportfs/expfs.c      |   2 +
 fs/fhandle.c             | 158 +++++++++++++++++++++++++++++++++++++++++++++++
 fs/internal.h            |   3 +
 include/linux/syscalls.h |   3 +
 kernel/sys_ni.c          |   2 +
 6 files changed, 181 insertions(+)

(limited to 'kernel')

diff --git a/fs/compat.c b/fs/compat.c
index a071775f3bb3..c6d31a3bab88 100644
--- a/fs/compat.c
+++ b/fs/compat.c
@@ -2284,3 +2284,16 @@ asmlinkage long compat_sys_timerfd_gettime(int ufd,
 }
 
 #endif /* CONFIG_TIMERFD */
+
+#ifdef CONFIG_FHANDLE
+/*
+ * Exactly like fs/open.c:sys_open_by_handle_at(), except that it
+ * doesn't set the O_LARGEFILE flag.
+ */
+asmlinkage long
+compat_sys_open_by_handle_at(int mountdirfd,
+			     struct file_handle __user *handle, int flags)
+{
+	return do_handle_open(mountdirfd, handle, flags);
+}
+#endif
diff --git a/fs/exportfs/expfs.c b/fs/exportfs/expfs.c
index cfe55731b6dc..b05acb796135 100644
--- a/fs/exportfs/expfs.c
+++ b/fs/exportfs/expfs.c
@@ -374,6 +374,8 @@ struct dentry *exportfs_decode_fh(struct vfsmount *mnt, struct fid *fid,
 	/*
 	 * Try to get any dentry for the given file handle from the filesystem.
 	 */
+	if (!nop || !nop->fh_to_dentry)
+		return ERR_PTR(-ESTALE);
 	result = nop->fh_to_dentry(mnt->mnt_sb, fid, fh_len, fileid_type);
 	if (!result)
 		result = ERR_PTR(-ESTALE);
diff --git a/fs/fhandle.c b/fs/fhandle.c
index 9f79e743a840..bf93ad2bee07 100644
--- a/fs/fhandle.c
+++ b/fs/fhandle.c
@@ -5,6 +5,8 @@
 #include <linux/mount.h>
 #include <linux/namei.h>
 #include <linux/exportfs.h>
+#include <linux/fs_struct.h>
+#include <linux/fsnotify.h>
 #include <asm/uaccess.h>
 #include "internal.h"
 
@@ -105,3 +107,159 @@ SYSCALL_DEFINE5(name_to_handle_at, int, dfd, const char __user *, name,
 	}
 	return err;
 }
+
+static struct vfsmount *get_vfsmount_from_fd(int fd)
+{
+	struct path path;
+
+	if (fd == AT_FDCWD) {
+		struct fs_struct *fs = current->fs;
+		spin_lock(&fs->lock);
+		path = fs->pwd;
+		mntget(path.mnt);
+		spin_unlock(&fs->lock);
+	} else {
+		int fput_needed;
+		struct file *file = fget_light(fd, &fput_needed);
+		if (!file)
+			return ERR_PTR(-EBADF);
+		path = file->f_path;
+		mntget(path.mnt);
+		fput_light(file, fput_needed);
+	}
+	return path.mnt;
+}
+
+static int vfs_dentry_acceptable(void *context, struct dentry *dentry)
+{
+	return 1;
+}
+
+static int do_handle_to_path(int mountdirfd, struct file_handle *handle,
+			     struct path *path)
+{
+	int retval = 0;
+	int handle_dwords;
+
+	path->mnt = get_vfsmount_from_fd(mountdirfd);
+	if (IS_ERR(path->mnt)) {
+		retval = PTR_ERR(path->mnt);
+		goto out_err;
+	}
+	/* change the handle size to multiple of sizeof(u32) */
+	handle_dwords = handle->handle_bytes >> 2;
+	path->dentry = exportfs_decode_fh(path->mnt,
+					  (struct fid *)handle->f_handle,
+					  handle_dwords, handle->handle_type,
+					  vfs_dentry_acceptable, NULL);
+	if (IS_ERR(path->dentry)) {
+		retval = PTR_ERR(path->dentry);
+		goto out_mnt;
+	}
+	return 0;
+out_mnt:
+	mntput(path->mnt);
+out_err:
+	return retval;
+}
+
+static int handle_to_path(int mountdirfd, struct file_handle __user *ufh,
+		   struct path *path)
+{
+	int retval = 0;
+	struct file_handle f_handle;
+	struct file_handle *handle = NULL;
+
+	/*
+	 * With handle we don't look at the execute bit on the
+	 * the directory. Ideally we would like CAP_DAC_SEARCH.
+	 * But we don't have that
+	 */
+	if (!capable(CAP_DAC_READ_SEARCH)) {
+		retval = -EPERM;
+		goto out_err;
+	}
+	if (copy_from_user(&f_handle, ufh, sizeof(struct file_handle))) {
+		retval = -EFAULT;
+		goto out_err;
+	}
+	if ((f_handle.handle_bytes > MAX_HANDLE_SZ) ||
+	    (f_handle.handle_bytes == 0)) {
+		retval = -EINVAL;
+		goto out_err;
+	}
+	handle = kmalloc(sizeof(struct file_handle) + f_handle.handle_bytes,
+			 GFP_KERNEL);
+	if (!handle) {
+		retval = -ENOMEM;
+		goto out_err;
+	}
+	/* copy the full handle */
+	if (copy_from_user(handle, ufh,
+			   sizeof(struct file_handle) +
+			   f_handle.handle_bytes)) {
+		retval = -EFAULT;
+		goto out_handle;
+	}
+
+	retval = do_handle_to_path(mountdirfd, handle, path);
+
+out_handle:
+	kfree(handle);
+out_err:
+	return retval;
+}
+
+long do_handle_open(int mountdirfd,
+		    struct file_handle __user *ufh, int open_flag)
+{
+	long retval = 0;
+	struct path path;
+	struct file *file;
+	int fd;
+
+	retval = handle_to_path(mountdirfd, ufh, &path);
+	if (retval)
+		return retval;
+
+	fd = get_unused_fd_flags(open_flag);
+	if (fd < 0) {
+		path_put(&path);
+		return fd;
+	}
+	file = file_open_root(path.dentry, path.mnt, "", open_flag);
+	if (IS_ERR(file)) {
+		put_unused_fd(fd);
+		retval =  PTR_ERR(file);
+	} else {
+		retval = fd;
+		fsnotify_open(file);
+		fd_install(fd, file);
+	}
+	path_put(&path);
+	return retval;
+}
+
+/**
+ * sys_open_by_handle_at: Open the file handle
+ * @mountdirfd: directory file descriptor
+ * @handle: file handle to be opened
+ * @flag: open flags.
+ *
+ * @mountdirfd indicate the directory file descriptor
+ * of the mount point. file handle is decoded relative
+ * to the vfsmount pointed by the @mountdirfd. @flags
+ * value is same as the open(2) flags.
+ */
+SYSCALL_DEFINE3(open_by_handle_at, int, mountdirfd,
+		struct file_handle __user *, handle,
+		int, flags)
+{
+	long ret;
+
+	if (force_o_largefile())
+		flags |= O_LARGEFILE;
+
+	ret = do_handle_open(mountdirfd, handle, flags);
+	return ret;
+}
diff --git a/fs/internal.h b/fs/internal.h
index 52abc5287f50..f3d15de44b15 100644
--- a/fs/internal.h
+++ b/fs/internal.h
@@ -117,6 +117,9 @@ extern struct file *do_filp_open(int dfd, const char *pathname,
 extern struct file *do_file_open_root(struct dentry *, struct vfsmount *,
 		const char *, const struct open_flags *, int lookup_flags);
 
+extern long do_handle_open(int mountdirfd,
+			   struct file_handle __user *ufh, int open_flag);
+
 /*
  * inode.c
  */
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index 970112613fb4..2d9b79c0f224 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -836,4 +836,7 @@ asmlinkage long sys_old_mmap(struct mmap_arg_struct __user *arg);
 asmlinkage long sys_name_to_handle_at(int dfd, const char __user *name,
 				      struct file_handle __user *handle,
 				      int __user *mnt_id, int flag);
+asmlinkage long sys_open_by_handle_at(int mountdirfd,
+				      struct file_handle __user *handle,
+				      int flags);
 #endif
diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c
index 4e013439ac28..25cc41cd8f33 100644
--- a/kernel/sys_ni.c
+++ b/kernel/sys_ni.c
@@ -189,3 +189,5 @@ cond_syscall(sys_fanotify_mark);
 
 /* open by handle */
 cond_syscall(sys_name_to_handle_at);
+cond_syscall(sys_open_by_handle_at);
+cond_syscall(compat_sys_open_by_handle_at);
-- 
cgit v1.2.3