[PATCH 01/13] vfs: add i_op->dentry_open()

[Miklos Szeredi]

From: Miklos Szeredi <mszeredi@xxxxxxx>

Add a new inode operation i_op->dentry_open().  This is for stacked filesystems
that want to return a struct file from a different filesystem.

Signed-off-by: Miklos Szeredi <mszeredi@xxxxxxx>
---
 Documentation/filesystems/Locking |    2 ++
 Documentation/filesystems/vfs.txt |    7 +++++++
 fs/namei.c                        |    9 ++++++---
 fs/open.c                         |   23 +++++++++++++++++++++--
 include/linux/fs.h                |    2 ++
 5 files changed, 38 insertions(+), 5 deletions(-)

diff --git a/Documentation/filesystems/Locking b/Documentation/filesystems/Locking
index e540a24..019fc72 100644
--- a/Documentation/filesystems/Locking
+++ b/Documentation/filesystems/Locking
@@ -64,6 +64,7 @@ prototypes:
 	int (*atomic_open)(struct inode *, struct dentry *,
 				struct file *, unsigned open_flag,
 				umode_t create_mode, int *opened);
+	int (*dentry_open)(struct dentry *, struct file *, const struct cred *);
 
 locking rules:
 	all may block
@@ -92,6 +93,7 @@ removexattr:	yes
 fiemap:		no
 update_time:	no
 atomic_open:	yes
+dentry_open:	no
 
 	Additionally, ->rmdir(), ->unlink() and ->rename() have ->i_mutex on
 victim.
diff --git a/Documentation/filesystems/vfs.txt b/Documentation/filesystems/vfs.txt
index 2ee133e..ea1379c 100644
--- a/Documentation/filesystems/vfs.txt
+++ b/Documentation/filesystems/vfs.txt
@@ -363,6 +363,7 @@ struct inode_operations {
 	int (*atomic_open)(struct inode *, struct dentry *,
 				struct file *, unsigned open_flag,
 				umode_t create_mode, int *opened);
+	int (*dentry_open)(struct dentry *, struct file *, const struct cred *);
 };
 
 Again, all methods are called without any locks being held, unless
@@ -692,6 +693,12 @@ struct address_space_operations {
   	but instead uses bmap to find out where the blocks in the file
   	are and uses those addresses directly.
 
+  dentry_open: this is an alternative to f_op->open(), the difference is that
+	this method may open a file not necessarily originating from the same
+	filesystem as the one i_op->open() was called on.  It may be
+	useful for stacking filesystems which want to allow native I/O directly
+	on underlying files.
+
 
   invalidatepage: If a page has PagePrivate set, then invalidatepage
         will be called when part or all of the page is to be removed
diff --git a/fs/namei.c b/fs/namei.c
index dd1ed1b..829404a 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -2822,9 +2822,12 @@ finish_open_created:
 	error = may_open(&nd->path, acc_mode, open_flag);
 	if (error)
 		goto out;
-	file->f_path.mnt = nd->path.mnt;
-	error = finish_open(file, nd->path.dentry, NULL, opened);
-	if (error) {
+
+	BUG_ON(*opened & FILE_OPENED); /* once it's opened, it's opened */
+	error = vfs_open(&nd->path, file, current_cred());
+	if (!error) {
+		*opened |= FILE_OPENED;
+	} else {
 		if (error == -EOPENSTALE)
 			goto stale_open;
 		goto out;
diff --git a/fs/open.c b/fs/open.c
index e1f2cdb..d0f0ec7 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -787,8 +787,7 @@ struct file *dentry_open(const struct path *path, int flags,
 		return ERR_PTR(error);
 
 	f->f_flags = flags;
-	f->f_path = *path;
-	error = do_dentry_open(f, NULL, cred);
+	error = vfs_open(path, f, cred);
 	if (!error) {
 		error = open_check_o_direct(f);
 		if (error) {
@@ -803,6 +802,26 @@ struct file *dentry_open(const struct path *path, int flags,
 }
 EXPORT_SYMBOL(dentry_open);
 
+/**
+ * vfs_open - open the file at the given path
+ * @path: path to open
+ * @filp: newly allocated file with f_flag initialized
+ * @cred: credentials to use
+ */
+int vfs_open(const struct path *path, struct file *filp,
+	     const struct cred *cred)
+{
+	struct inode *inode = path->dentry->d_inode;
+
+	if (inode->i_op->dentry_open)
+		return inode->i_op->dentry_open(path->dentry, filp, cred);
+	else {
+		filp->f_path = *path;
+		return do_dentry_open(filp, NULL, cred);
+	}
+}
+EXPORT_SYMBOL(vfs_open);
+
 static void __put_unused_fd(struct files_struct *files, unsigned int fd)
 {
 	struct fdtable *fdt = files_fdtable(files);
diff --git a/include/linux/fs.h b/include/linux/fs.h
index aa11047..e372113 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1835,6 +1835,7 @@ struct inode_operations {
 	int (*atomic_open)(struct inode *, struct dentry *,
 			   struct file *, unsigned open_flag,
 			   umode_t create_mode, int *opened);
+	int (*dentry_open)(struct dentry *, struct file *, const struct cred *);
 } ____cacheline_aligned;
 
 struct seq_file;
@@ -2199,6 +2200,7 @@ extern long do_sys_open(int dfd, const char __user *filename, int flags,
 extern struct file *filp_open(const char *, int, umode_t);
 extern struct file *file_open_root(struct dentry *, struct vfsmount *,
 				   const char *, int);
+extern int vfs_open(const struct path *, struct file *, const struct cred *);
 extern struct file * dentry_open(const struct path *, int, const struct cred *);
 extern int filp_close(struct file *, fl_owner_t id);
 extern char * getname(const char __user *);
-- 
1.7.7

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/