From f47ec3f28354795f000c14bf18ed967ec81a3ec3 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Mon, 21 Nov 2011 21:15:42 -0500
Subject: trim fs/internal.h

some stuff in there can actually become static; some belongs to pnode.h
as it's a private interface between namespace.c and pnode.c...

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/internal.h | 24 ------------------------
 1 file changed, 24 deletions(-)

(limited to 'fs/internal.h')

diff --git a/fs/internal.h b/fs/internal.h
index fe327c20af83..839d3f9e9173 100644
--- a/fs/internal.h
+++ b/fs/internal.h
@@ -20,14 +20,8 @@ struct path;
  * block_dev.c
  */
 #ifdef CONFIG_BLOCK
-extern struct super_block *blockdev_superblock;
 extern void __init bdev_cache_init(void);
 
-static inline int sb_is_blkdev_sb(struct super_block *sb)
-{
-	return sb == blockdev_superblock;
-}
-
 extern int __sync_blockdev(struct block_device *bdev, int wait);
 
 #else
@@ -35,11 +29,6 @@ static inline void bdev_cache_init(void)
 {
 }
 
-static inline int sb_is_blkdev_sb(struct super_block *sb)
-{
-	return 0;
-}
-
 static inline int __sync_blockdev(struct block_device *bdev, int wait)
 {
 	return 0;
@@ -51,25 +40,14 @@ static inline int __sync_blockdev(struct block_device *bdev, int wait)
  */
 extern void __init chrdev_init(void);
 
-/*
- * exec.c
- */
-extern int check_unsafe_exec(struct linux_binprm *);
-
 /*
  * namespace.c
  */
 extern int copy_mount_options(const void __user *, unsigned long *);
 extern int copy_mount_string(const void __user *, char **);
 
-extern unsigned int mnt_get_count(struct vfsmount *mnt);
 extern struct vfsmount *__lookup_mnt(struct vfsmount *, struct dentry *, int);
 extern struct vfsmount *lookup_mnt(struct path *);
-extern void mnt_set_mountpoint(struct vfsmount *, struct dentry *,
-				struct vfsmount *);
-extern void release_mounts(struct list_head *);
-extern void umount_tree(struct vfsmount *, int, struct list_head *);
-extern struct vfsmount *copy_tree(struct vfsmount *, struct dentry *, int);
 extern int finish_automount(struct vfsmount *, struct path *);
 
 extern void mnt_make_longterm(struct vfsmount *);
@@ -98,8 +76,6 @@ extern struct file *get_empty_filp(void);
  */
 extern int do_remount_sb(struct super_block *, int, void *, int);
 extern bool grab_super_passive(struct super_block *sb);
-extern void __put_super(struct super_block *sb);
-extern void put_super(struct super_block *sb);
 extern struct dentry *mount_fs(struct file_system_type *,
 			       int, const char *, void *);
 
-- 
cgit 


From cf31e70d6cf93f19fe9bf1144966ef40991ac723 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Mon, 2 Jan 2012 22:28:36 -0500
Subject: vfs: new helper - vfs_ustat()

... and bury user_get_super()/statfs_by_dentry() - they are
purely internal now.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 arch/parisc/hpux/sys_hpux.c |  9 +--------
 fs/compat.c                 |  9 +--------
 fs/internal.h               |  1 +
 fs/statfs.c                 | 21 +++++++++++++--------
 include/linux/fs.h          |  3 +--
 5 files changed, 17 insertions(+), 26 deletions(-)

(limited to 'fs/internal.h')

diff --git a/arch/parisc/hpux/sys_hpux.c b/arch/parisc/hpux/sys_hpux.c
index 6ab9580b0b00..d9dc6cd3b7d2 100644
--- a/arch/parisc/hpux/sys_hpux.c
+++ b/arch/parisc/hpux/sys_hpux.c
@@ -136,16 +136,9 @@ struct hpux_ustat {
  */
 static int hpux_ustat(dev_t dev, struct hpux_ustat __user *ubuf)
 {
-	struct super_block *s;
 	struct hpux_ustat tmp;  /* Changed to hpux_ustat */
 	struct kstatfs sbuf;
-	int err = -EINVAL;
-
-	s = user_get_super(dev);
-	if (s == NULL)
-		goto out;
-	err = statfs_by_dentry(s->s_root, &sbuf);
-	drop_super(s);
+	int err = vfs_ustat(dev, &sbuf);
 	if (err)
 		goto out;
 
diff --git a/fs/compat.c b/fs/compat.c
index c98787536bb8..9db5a6076610 100644
--- a/fs/compat.c
+++ b/fs/compat.c
@@ -342,16 +342,9 @@ asmlinkage long compat_sys_fstatfs64(unsigned int fd, compat_size_t sz, struct c
  */
 asmlinkage long compat_sys_ustat(unsigned dev, struct compat_ustat __user *u)
 {
-	struct super_block *sb;
 	struct compat_ustat tmp;
 	struct kstatfs sbuf;
-	int err;
-
-	sb = user_get_super(new_decode_dev(dev));
-	if (!sb)
-		return -EINVAL;
-	err = statfs_by_dentry(sb->s_root, &sbuf);
-	drop_super(sb);
+	int err = vfs_ustat(new_decode_dev(dev), &sbuf);
 	if (err)
 		return err;
 
diff --git a/fs/internal.h b/fs/internal.h
index 839d3f9e9173..7b1cb1528ac2 100644
--- a/fs/internal.h
+++ b/fs/internal.h
@@ -78,6 +78,7 @@ extern int do_remount_sb(struct super_block *, int, void *, int);
 extern bool grab_super_passive(struct super_block *sb);
 extern struct dentry *mount_fs(struct file_system_type *,
 			       int, const char *, void *);
+extern struct super_block *user_get_super(dev_t);
 
 /*
  * open.c
diff --git a/fs/statfs.c b/fs/statfs.c
index 9cf04a118965..2aa6a22e0be2 100644
--- a/fs/statfs.c
+++ b/fs/statfs.c
@@ -7,6 +7,7 @@
 #include <linux/statfs.h>
 #include <linux/security.h>
 #include <linux/uaccess.h>
+#include "internal.h"
 
 static int flags_by_mnt(int mnt_flags)
 {
@@ -45,7 +46,7 @@ static int calculate_f_flags(struct vfsmount *mnt)
 		flags_by_sb(mnt->mnt_sb->s_flags);
 }
 
-int statfs_by_dentry(struct dentry *dentry, struct kstatfs *buf)
+static int statfs_by_dentry(struct dentry *dentry, struct kstatfs *buf)
 {
 	int retval;
 
@@ -205,19 +206,23 @@ SYSCALL_DEFINE3(fstatfs64, unsigned int, fd, size_t, sz, struct statfs64 __user
 	return error;
 }
 
-SYSCALL_DEFINE2(ustat, unsigned, dev, struct ustat __user *, ubuf)
+int vfs_ustat(dev_t dev, struct kstatfs *sbuf)
 {
-	struct super_block *s;
-	struct ustat tmp;
-	struct kstatfs sbuf;
+	struct super_block *s = user_get_super(dev);
 	int err;
-
-	s = user_get_super(new_decode_dev(dev));
 	if (!s)
 		return -EINVAL;
 
-	err = statfs_by_dentry(s->s_root, &sbuf);
+	err = statfs_by_dentry(s->s_root, sbuf);
 	drop_super(s);
+	return err;
+}
+
+SYSCALL_DEFINE2(ustat, unsigned, dev, struct ustat __user *, ubuf)
+{
+	struct ustat tmp;
+	struct kstatfs sbuf;
+	int err = vfs_ustat(new_decode_dev(dev), &sbuf);
 	if (err)
 		return err;
 
diff --git a/include/linux/fs.h b/include/linux/fs.h
index ed17e54fd204..cec429d76ab0 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1939,7 +1939,7 @@ extern int iterate_mounts(int (*)(struct vfsmount *, void *), void *,
 extern int vfs_statfs(struct path *, struct kstatfs *);
 extern int user_statfs(const char __user *, struct kstatfs *);
 extern int fd_statfs(int, struct kstatfs *);
-extern int statfs_by_dentry(struct dentry *, struct kstatfs *);
+extern int vfs_ustat(dev_t, struct kstatfs *);
 extern int freeze_super(struct super_block *super);
 extern int thaw_super(struct super_block *super);
 extern bool our_mnt(struct vfsmount *mnt);
@@ -2531,7 +2531,6 @@ extern void put_filesystem(struct file_system_type *fs);
 extern struct file_system_type *get_fs_type(const char *name);
 extern struct super_block *get_super(struct block_device *);
 extern struct super_block *get_active_super(struct block_device *bdev);
-extern struct super_block *user_get_super(dev_t);
 extern void drop_super(struct super_block *sb);
 extern void iterate_supers(void (*)(struct super_block *, void *), void *);
 extern void iterate_supers_type(struct file_system_type *,
-- 
cgit 


From a218d0fdc5f9004164ff151d274487f6799907d0 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Mon, 21 Nov 2011 14:59:34 -0500
Subject: switch open and mkdir syscalls to umode_t

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/compat.c              |  4 ++--
 fs/internal.h            |  2 +-
 fs/namei.c               |  6 +++---
 fs/open.c                | 12 ++++++------
 include/linux/compat.h   |  4 ++--
 include/linux/fs.h       |  4 ++--
 include/linux/syscalls.h | 10 +++++-----
 7 files changed, 21 insertions(+), 21 deletions(-)

(limited to 'fs/internal.h')

diff --git a/fs/compat.c b/fs/compat.c
index 9db5a6076610..fa9d721ecfee 100644
--- a/fs/compat.c
+++ b/fs/compat.c
@@ -1281,7 +1281,7 @@ compat_sys_vmsplice(int fd, const struct compat_iovec __user *iov32,
  * O_LARGEFILE flag.
  */
 asmlinkage long
-compat_sys_open(const char __user *filename, int flags, int mode)
+compat_sys_open(const char __user *filename, int flags, umode_t mode)
 {
 	return do_sys_open(AT_FDCWD, filename, flags, mode);
 }
@@ -1291,7 +1291,7 @@ compat_sys_open(const char __user *filename, int flags, int mode)
  * O_LARGEFILE flag.
  */
 asmlinkage long
-compat_sys_openat(unsigned int dfd, const char __user *filename, int flags, int mode)
+compat_sys_openat(unsigned int dfd, const char __user *filename, int flags, umode_t mode)
 {
 	return do_sys_open(dfd, filename, flags, mode);
 }
diff --git a/fs/internal.h b/fs/internal.h
index 7b1cb1528ac2..23599f88d1a5 100644
--- a/fs/internal.h
+++ b/fs/internal.h
@@ -88,7 +88,7 @@ extern struct file *nameidata_to_filp(struct nameidata *);
 extern void release_open_intent(struct nameidata *);
 struct open_flags {
 	int open_flag;
-	int mode;
+	umode_t mode;
 	int acc_mode;
 	int intent;
 };
diff --git a/fs/namei.c b/fs/namei.c
index e275dc36d7c5..afd5876cd072 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -2177,7 +2177,7 @@ static struct file *do_last(struct nameidata *nd, struct path *path,
 
 	/* Negative dentry, just create the file */
 	if (!dentry->d_inode) {
-		int mode = op->mode;
+		umode_t mode = op->mode;
 		if (!IS_POSIXACL(dir->d_inode))
 			mode &= ~current_umask();
 		/*
@@ -2562,7 +2562,7 @@ int vfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
 	return error;
 }
 
-SYSCALL_DEFINE3(mkdirat, int, dfd, const char __user *, pathname, int, mode)
+SYSCALL_DEFINE3(mkdirat, int, dfd, const char __user *, pathname, umode_t, mode)
 {
 	struct dentry *dentry;
 	struct path path;
@@ -2590,7 +2590,7 @@ out_dput:
 	return error;
 }
 
-SYSCALL_DEFINE2(mkdir, const char __user *, pathname, int, mode)
+SYSCALL_DEFINE2(mkdir, const char __user *, pathname, umode_t, mode)
 {
 	return sys_mkdirat(AT_FDCWD, pathname, mode);
 }
diff --git a/fs/open.c b/fs/open.c
index 834e3e1adeb9..2659f596f4c5 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -877,7 +877,7 @@ void fd_install(unsigned int fd, struct file *file)
 
 EXPORT_SYMBOL(fd_install);
 
-static inline int build_open_flags(int flags, int mode, struct open_flags *op)
+static inline int build_open_flags(int flags, umode_t mode, struct open_flags *op)
 {
 	int lookup_flags = 0;
 	int acc_mode;
@@ -948,7 +948,7 @@ static inline int build_open_flags(int flags, int mode, struct open_flags *op)
  * have to.  But in generally you should not do this, so please move
  * along, nothing to see here..
  */
-struct file *filp_open(const char *filename, int flags, int mode)
+struct file *filp_open(const char *filename, int flags, umode_t mode)
 {
 	struct open_flags op;
 	int lookup = build_open_flags(flags, mode, &op);
@@ -970,7 +970,7 @@ struct file *file_open_root(struct dentry *dentry, struct vfsmount *mnt,
 }
 EXPORT_SYMBOL(file_open_root);
 
-long do_sys_open(int dfd, const char __user *filename, int flags, int mode)
+long do_sys_open(int dfd, const char __user *filename, int flags, umode_t mode)
 {
 	struct open_flags op;
 	int lookup = build_open_flags(flags, mode, &op);
@@ -994,7 +994,7 @@ long do_sys_open(int dfd, const char __user *filename, int flags, int mode)
 	return fd;
 }
 
-SYSCALL_DEFINE3(open, const char __user *, filename, int, flags, int, mode)
+SYSCALL_DEFINE3(open, const char __user *, filename, int, flags, umode_t, mode)
 {
 	long ret;
 
@@ -1008,7 +1008,7 @@ SYSCALL_DEFINE3(open, const char __user *, filename, int, flags, int, mode)
 }
 
 SYSCALL_DEFINE4(openat, int, dfd, const char __user *, filename, int, flags,
-		int, mode)
+		umode_t, mode)
 {
 	long ret;
 
@@ -1027,7 +1027,7 @@ SYSCALL_DEFINE4(openat, int, dfd, const char __user *, filename, int, flags,
  * For backward compatibility?  Maybe this should be moved
  * into arch/i386 instead?
  */
-SYSCALL_DEFINE2(creat, const char __user *, pathname, int, mode)
+SYSCALL_DEFINE2(creat, const char __user *, pathname, umode_t, mode)
 {
 	return sys_open(pathname, O_CREAT | O_WRONLY | O_TRUNC, mode);
 }
diff --git a/include/linux/compat.h b/include/linux/compat.h
index 66ed067fb729..41c9f6515f46 100644
--- a/include/linux/compat.h
+++ b/include/linux/compat.h
@@ -422,9 +422,9 @@ asmlinkage long compat_sys_getdents64(unsigned int fd,
 asmlinkage long compat_sys_vmsplice(int fd, const struct compat_iovec __user *,
 				    unsigned int nr_segs, unsigned int flags);
 asmlinkage long compat_sys_open(const char __user *filename, int flags,
-				int mode);
+				umode_t mode);
 asmlinkage long compat_sys_openat(unsigned int dfd, const char __user *filename,
-				  int flags, int mode);
+				  int flags, umode_t mode);
 asmlinkage long compat_sys_open_by_handle_at(int mountdirfd,
 					     struct file_handle __user *handle,
 					     int flags);
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 9d02fab420c6..f0e57b7e4297 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -2054,8 +2054,8 @@ extern int do_truncate(struct dentry *, loff_t start, unsigned int time_attrs,
 extern int do_fallocate(struct file *file, int mode, loff_t offset,
 			loff_t len);
 extern long do_sys_open(int dfd, const char __user *filename, int flags,
-			int mode);
-extern struct file *filp_open(const char *, int, int);
+			umode_t mode);
+extern struct file *filp_open(const char *, int, umode_t);
 extern struct file *file_open_root(struct dentry *, struct vfsmount *,
 				   const char *, int);
 extern struct file * dentry_open(struct dentry *, struct vfsmount *, int,
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index b25621476316..515669fa3c1d 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -517,9 +517,9 @@ asmlinkage long sys_sendfile64(int out_fd, int in_fd,
 			       loff_t __user *offset, size_t count);
 asmlinkage long sys_readlink(const char __user *path,
 				char __user *buf, int bufsiz);
-asmlinkage long sys_creat(const char __user *pathname, int mode);
+asmlinkage long sys_creat(const char __user *pathname, umode_t mode);
 asmlinkage long sys_open(const char __user *filename,
-				int flags, int mode);
+				int flags, umode_t mode);
 asmlinkage long sys_close(unsigned int fd);
 asmlinkage long sys_access(const char __user *filename, int mode);
 asmlinkage long sys_vhangup(void);
@@ -582,7 +582,7 @@ asmlinkage long sys_preadv(unsigned long fd, const struct iovec __user *vec,
 asmlinkage long sys_pwritev(unsigned long fd, const struct iovec __user *vec,
 			    unsigned long vlen, unsigned long pos_l, unsigned long pos_h);
 asmlinkage long sys_getcwd(char __user *buf, unsigned long size);
-asmlinkage long sys_mkdir(const char __user *pathname, int mode);
+asmlinkage long sys_mkdir(const char __user *pathname, umode_t mode);
 asmlinkage long sys_chdir(const char __user *filename);
 asmlinkage long sys_fchdir(unsigned int fd);
 asmlinkage long sys_rmdir(const char __user *pathname);
@@ -757,7 +757,7 @@ asmlinkage long sys_spu_create(const char __user *name,
 
 asmlinkage long sys_mknodat(int dfd, const char __user * filename, umode_t mode,
 			    unsigned dev);
-asmlinkage long sys_mkdirat(int dfd, const char __user * pathname, int mode);
+asmlinkage long sys_mkdirat(int dfd, const char __user * pathname, umode_t mode);
 asmlinkage long sys_unlinkat(int dfd, const char __user * pathname, int flag);
 asmlinkage long sys_symlinkat(const char __user * oldname,
 			      int newdfd, const char __user * newname);
@@ -773,7 +773,7 @@ asmlinkage long sys_fchmodat(int dfd, const char __user * filename,
 asmlinkage long sys_fchownat(int dfd, const char __user *filename, uid_t user,
 			     gid_t group, int flag);
 asmlinkage long sys_openat(int dfd, const char __user *filename, int flags,
-			   int mode);
+			   umode_t mode);
 asmlinkage long sys_newfstatat(int dfd, const char __user *filename,
 			       struct stat __user *statbuf, int flag);
 asmlinkage long sys_fstatat64(int dfd, const char __user *filename,
-- 
cgit 


From c71053659e3bb27d44b79da0bb4abf5838c2060a Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Thu, 24 Nov 2011 18:22:03 -0500
Subject: vfs: spread struct mount - __lookup_mnt() result

switch __lookup_mnt() to returning struct mount *; callers adjusted.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/internal.h  |  2 +-
 fs/mount.h     |  2 ++
 fs/namei.c     | 13 +++++++------
 fs/namespace.c | 23 ++++++++++++++---------
 fs/pnode.c     | 13 +++++++------
 5 files changed, 31 insertions(+), 22 deletions(-)

(limited to 'fs/internal.h')

diff --git a/fs/internal.h b/fs/internal.h
index 7b1cb1528ac2..6aab61a4f36f 100644
--- a/fs/internal.h
+++ b/fs/internal.h
@@ -15,6 +15,7 @@ struct super_block;
 struct file_system_type;
 struct linux_binprm;
 struct path;
+struct mount;
 
 /*
  * block_dev.c
@@ -46,7 +47,6 @@ extern void __init chrdev_init(void);
 extern int copy_mount_options(const void __user *, unsigned long *);
 extern int copy_mount_string(const void __user *, char **);
 
-extern struct vfsmount *__lookup_mnt(struct vfsmount *, struct dentry *, int);
 extern struct vfsmount *lookup_mnt(struct path *);
 extern int finish_automount(struct vfsmount *, struct path *);
 
diff --git a/fs/mount.h b/fs/mount.h
index 47da8163e1f4..44e5b6f54b7e 100644
--- a/fs/mount.h
+++ b/fs/mount.h
@@ -13,3 +13,5 @@ static inline int mnt_has_parent(struct vfsmount *mnt)
 {
 	return mnt != mnt->mnt_parent;
 }
+
+extern struct mount *__lookup_mnt(struct vfsmount *, struct dentry *, int);
diff --git a/fs/namei.c b/fs/namei.c
index 5008f01787f5..d1c6a559f8f0 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -36,6 +36,7 @@
 #include <asm/uaccess.h>
 
 #include "internal.h"
+#include "mount.h"
 
 /* [Feb-1997 T. Schoebel-Theuer]
  * Fundamental changes in the pathname lookup mechanisms (namei)
@@ -884,7 +885,7 @@ static bool __follow_mount_rcu(struct nameidata *nd, struct path *path,
 			       struct inode **inode)
 {
 	for (;;) {
-		struct vfsmount *mounted;
+		struct mount *mounted;
 		/*
 		 * Don't forget we might have a non-mountpoint managed dentry
 		 * that wants to block transit.
@@ -898,8 +899,8 @@ static bool __follow_mount_rcu(struct nameidata *nd, struct path *path,
 		mounted = __lookup_mnt(path->mnt, path->dentry, 1);
 		if (!mounted)
 			break;
-		path->mnt = mounted;
-		path->dentry = mounted->mnt_root;
+		path->mnt = &mounted->mnt;
+		path->dentry = mounted->mnt.mnt_root;
 		nd->flags |= LOOKUP_JUMPED;
 		nd->seq = read_seqcount_begin(&path->dentry->d_seq);
 		/*
@@ -915,12 +916,12 @@ static bool __follow_mount_rcu(struct nameidata *nd, struct path *path,
 static void follow_mount_rcu(struct nameidata *nd)
 {
 	while (d_mountpoint(nd->path.dentry)) {
-		struct vfsmount *mounted;
+		struct mount *mounted;
 		mounted = __lookup_mnt(nd->path.mnt, nd->path.dentry, 1);
 		if (!mounted)
 			break;
-		nd->path.mnt = mounted;
-		nd->path.dentry = mounted->mnt_root;
+		nd->path.mnt = &mounted->mnt;
+		nd->path.dentry = mounted->mnt.mnt_root;
 		nd->seq = read_seqcount_begin(&nd->path.dentry->d_seq);
 	}
 }
diff --git a/fs/namespace.c b/fs/namespace.c
index dda47fee6fdf..398eb2421494 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -464,20 +464,20 @@ static void free_vfsmnt(struct vfsmount *mnt)
  * @dir. If @dir is set return the first mount else return the last mount.
  * vfsmount_lock must be held for read or write.
  */
-struct vfsmount *__lookup_mnt(struct vfsmount *mnt, struct dentry *dentry,
+struct mount *__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;
+	struct mount *p, *found = NULL;
 
 	for (;;) {
 		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) {
+		p = list_entry(tmp, struct mount, mnt.mnt_hash);
+		if (p->mnt.mnt_parent == mnt && p->mnt.mnt_mountpoint == dentry) {
 			found = p;
 			break;
 		}
@@ -491,13 +491,18 @@ struct vfsmount *__lookup_mnt(struct vfsmount *mnt, struct dentry *dentry,
  */
 struct vfsmount *lookup_mnt(struct path *path)
 {
-	struct vfsmount *child_mnt;
+	struct mount *child_mnt;
 
 	br_read_lock(vfsmount_lock);
-	if ((child_mnt = __lookup_mnt(path->mnt, path->dentry, 1)))
-		mntget(child_mnt);
-	br_read_unlock(vfsmount_lock);
-	return child_mnt;
+	child_mnt = __lookup_mnt(path->mnt, path->dentry, 1);
+	if (child_mnt) {
+		mnt_add_count(child_mnt, 1);
+		br_read_unlock(vfsmount_lock);
+		return &child_mnt->mnt;
+	} else {
+		br_read_unlock(vfsmount_lock);
+		return NULL;
+	}
 }
 
 static inline int check_mnt(struct vfsmount *mnt)
diff --git a/fs/pnode.c b/fs/pnode.c
index 4d5a06ea57a2..e996d039c0f2 100644
--- a/fs/pnode.c
+++ b/fs/pnode.c
@@ -289,7 +289,8 @@ static inline int do_refcount_check(struct vfsmount *mnt, int count)
  */
 int propagate_mount_busy(struct vfsmount *mnt, int refcnt)
 {
-	struct vfsmount *m, *child;
+	struct vfsmount *m;
+	struct mount *child;
 	struct vfsmount *parent = mnt->mnt_parent;
 	int ret = 0;
 
@@ -307,8 +308,8 @@ int propagate_mount_busy(struct vfsmount *mnt, int refcnt)
 	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)))
+		if (child && list_empty(&child->mnt.mnt_mounts) &&
+		    (ret = do_refcount_check(&child->mnt, 1)))
 			break;
 	}
 	return ret;
@@ -328,14 +329,14 @@ static void __propagate_umount(struct vfsmount *mnt)
 	for (m = propagation_next(parent, parent); m;
 			m = propagation_next(m, parent)) {
 
-		struct vfsmount *child = __lookup_mnt(m,
+		struct mount *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_move_tail(&child->mnt_hash, &mnt->mnt_hash);
+		if (child && list_empty(&child->mnt.mnt_mounts))
+			list_move_tail(&child->mnt.mnt_hash, &mnt->mnt_hash);
 	}
 }
 
-- 
cgit 


From 4ed5e82fe77f4147cf386327c9a63a2dd7eff518 Mon Sep 17 00:00:00 2001
From: Miklos Szeredi <mszeredi@suse.cz>
Date: Mon, 21 Nov 2011 12:11:31 +0100
Subject: vfs: protect remounting superblock read-only

Currently remouting superblock read-only is racy in a major way.

With the per mount read-only infrastructure it is now possible to
prevent most races, which this patch attempts.

Before starting the remount read-only, iterate through all mounts
belonging to the superblock and if none of them have any pending
writes, set sb->s_readonly_remount.  This indicates that remount is in
progress and no further write requests are allowed.  If the remount
succeeds set MS_RDONLY and reset s_readonly_remount.

If the remounting is unsuccessful just reset s_readonly_remount.
This can result in transient EROFS errors, despite the fact the
remount failed.  Unfortunately hodling off writes is difficult as
remount itself may touch the filesystem (e.g. through load_nls())
which would deadlock.

A later patch deals with delayed writes due to nlink going to zero.

Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Tested-by: Toshiyuki Okajima <toshi.okajima@jp.fujitsu.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/internal.h      |  1 +
 fs/namespace.c     | 40 +++++++++++++++++++++++++++++++++++++++-
 fs/super.c         | 22 ++++++++++++++++++----
 include/linux/fs.h |  3 +++
 4 files changed, 61 insertions(+), 5 deletions(-)

(limited to 'fs/internal.h')

diff --git a/fs/internal.h b/fs/internal.h
index 2523a4029452..9962c59ba280 100644
--- a/fs/internal.h
+++ b/fs/internal.h
@@ -52,6 +52,7 @@ extern int finish_automount(struct vfsmount *, struct path *);
 
 extern void mnt_make_longterm(struct vfsmount *);
 extern void mnt_make_shortterm(struct vfsmount *);
+extern int sb_prepare_remount_readonly(struct super_block *);
 
 extern void __init mnt_init(void);
 
diff --git a/fs/namespace.c b/fs/namespace.c
index 145217b088d1..98ebc78b21ab 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -273,6 +273,15 @@ static unsigned int mnt_get_writers(struct mount *mnt)
 #endif
 }
 
+static int mnt_is_readonly(struct vfsmount *mnt)
+{
+	if (mnt->mnt_sb->s_readonly_remount)
+		return 1;
+	/* Order wrt setting s_flags/s_readonly_remount in do_remount() */
+	smp_rmb();
+	return __mnt_is_readonly(mnt);
+}
+
 /*
  * Most r/o checks on a fs are for operations that take
  * discrete amounts of time, like a write() or unlink().
@@ -312,7 +321,7 @@ int mnt_want_write(struct vfsmount *m)
 	 * MNT_WRITE_HOLD is cleared.
 	 */
 	smp_rmb();
-	if (__mnt_is_readonly(m)) {
+	if (mnt_is_readonly(m)) {
 		mnt_dec_writers(mnt);
 		ret = -EROFS;
 		goto out;
@@ -435,6 +444,35 @@ static void __mnt_unmake_readonly(struct mount *mnt)
 	br_write_unlock(vfsmount_lock);
 }
 
+int sb_prepare_remount_readonly(struct super_block *sb)
+{
+	struct mount *mnt;
+	int err = 0;
+
+	br_write_lock(vfsmount_lock);
+	list_for_each_entry(mnt, &sb->s_mounts, mnt_instance) {
+		if (!(mnt->mnt.mnt_flags & MNT_READONLY)) {
+			mnt->mnt.mnt_flags |= MNT_WRITE_HOLD;
+			smp_mb();
+			if (mnt_get_writers(mnt) > 0) {
+				err = -EBUSY;
+				break;
+			}
+		}
+	}
+	if (!err) {
+		sb->s_readonly_remount = 1;
+		smp_wmb();
+	}
+	list_for_each_entry(mnt, &sb->s_mounts, mnt_instance) {
+		if (mnt->mnt.mnt_flags & MNT_WRITE_HOLD)
+			mnt->mnt.mnt_flags &= ~MNT_WRITE_HOLD;
+	}
+	br_write_unlock(vfsmount_lock);
+
+	return err;
+}
+
 static void free_vfsmnt(struct mount *mnt)
 {
 	kfree(mnt->mnt_devname);
diff --git a/fs/super.c b/fs/super.c
index 993ca8f128d6..6acc02237e3e 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -723,23 +723,33 @@ int do_remount_sb(struct super_block *sb, int flags, void *data, int force)
 	/* If we are remounting RDONLY and current sb is read/write,
 	   make sure there are no rw files opened */
 	if (remount_ro) {
-		if (force)
+		if (force) {
 			mark_files_ro(sb);
-		else if (!fs_may_remount_ro(sb))
-			return -EBUSY;
+		} else {
+			retval = sb_prepare_remount_readonly(sb);
+			if (retval)
+				return retval;
+
+			retval = -EBUSY;
+			if (!fs_may_remount_ro(sb))
+				goto cancel_readonly;
+		}
 	}
 
 	if (sb->s_op->remount_fs) {
 		retval = sb->s_op->remount_fs(sb, &flags, data);
 		if (retval) {
 			if (!force)
-				return retval;
+				goto cancel_readonly;
 			/* If forced remount, go ahead despite any errors */
 			WARN(1, "forced remount of a %s fs returned %i\n",
 			     sb->s_type->name, retval);
 		}
 	}
 	sb->s_flags = (sb->s_flags & ~MS_RMT_MASK) | (flags & MS_RMT_MASK);
+	/* Needs to be ordered wrt mnt_is_readonly() */
+	smp_wmb();
+	sb->s_readonly_remount = 0;
 
 	/*
 	 * Some filesystems modify their metadata via some other path than the
@@ -752,6 +762,10 @@ int do_remount_sb(struct super_block *sb, int flags, void *data, int force)
 	if (remount_ro && sb->s_bdev)
 		invalidate_bdev(sb->s_bdev);
 	return 0;
+
+cancel_readonly:
+	sb->s_readonly_remount = 0;
+	return retval;
 }
 
 static void do_emergency_remount(struct work_struct *work)
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 03385acd71e8..7b8a681b1ef4 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1482,6 +1482,9 @@ struct super_block {
 	int cleancache_poolid;
 
 	struct shrinker s_shrink;	/* per-sb shrinker handle */
+
+	/* Being remounted read-only */
+	int s_readonly_remount;
 };
 
 /* superblock cache pruning functions */
-- 
cgit