Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S933392Ab3CMOQ5 (ORCPT ); Wed, 13 Mar 2013 10:16:57 -0400 Received: from mail-bk0-f53.google.com ([209.85.214.53]:46411 "EHLO mail-bk0-f53.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932391Ab3CMOQy (ORCPT ); Wed, 13 Mar 2013 10:16:54 -0400 From: Miklos Szeredi To: viro@ZenIV.linux.org.uk, torvalds@linux-foundation.org Cc: linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, hch@infradead.org, akpm@linux-foundation.org, apw@canonical.com, nbd@openwrt.org, neilb@suse.de, jordipujolp@gmail.com, ezk@fsl.cs.sunysb.edu, dhowells@redhat.com, sedat.dilek@googlemail.com, hooanon05@yahoo.co.jp, mszeredi@suse.cz Subject: [PATCH 1/9] vfs: add i_op->dentry_open() Date: Wed, 13 Mar 2013 15:16:25 +0100 Message-Id: <1363184193-1796-2-git-send-email-miklos@szeredi.hu> X-Mailer: git-send-email 1.7.10.4 In-Reply-To: <1363184193-1796-1-git-send-email-miklos@szeredi.hu> References: <1363184193-1796-1-git-send-email-miklos@szeredi.hu> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 5335 Lines: 148 From: Miklos Szeredi 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 --- 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 0706d32..4331290 100644 --- a/Documentation/filesystems/Locking +++ b/Documentation/filesystems/Locking @@ -66,6 +66,7 @@ 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 @@ -93,6 +94,7 @@ 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 bc4b06b..f64a4d1 100644 --- a/Documentation/filesystems/vfs.txt +++ b/Documentation/filesystems/vfs.txt @@ -362,6 +362,7 @@ 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 @@ -681,6 +682,12 @@ 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 57ae9c8..c1c0c15 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -2867,9 +2867,12 @@ 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 6835446..b9d9f9e 100644 --- a/fs/open.c +++ b/fs/open.c @@ -828,8 +828,7 @@ struct file *dentry_open(const struct path *path, int flags, f = get_empty_filp(); if (!IS_ERR(f)) { f->f_flags = flags; - f->f_path = *path; - error = do_dentry_open(f, NULL, cred); + error = vfs_open(path, f, cred); if (!error) { /* from now on we need fput() to dispose of f */ error = open_check_o_direct(f); @@ -846,6 +845,26 @@ 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 inline int build_open_flags(int flags, umode_t mode, struct open_flags *op) { int lookup_flags = 0; diff --git a/include/linux/fs.h b/include/linux/fs.h index 2c28271..891c93d 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1573,6 +1573,7 @@ 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; ssize_t rw_copy_check_uvector(int type, const struct iovec __user * uvector, @@ -2006,6 +2007,7 @@ extern struct file *file_open_name(struct filename *, int, umode_t); 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); -- 1.7.10.4 -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/