Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6: (52 commits)
  init: Open /dev/console from rootfs
  mqueue: fix typo "failues" -> "failures"
  mqueue: only set error codes if they are really necessary
  mqueue: simplify do_open() error handling
  mqueue: apply mathematics distributivity on mq_bytes calculation
  mqueue: remove unneeded info->messages initialization
  mqueue: fix mq_open() file descriptor leak on user-space processes
  fix race in d_splice_alias()
  set S_DEAD on unlink() and non-directory rename() victims
  vfs: add NOFOLLOW flag to umount(2)
  get rid of ->mnt_parent in tomoyo/realpath
  hppfs can use existing proc_mnt, no need for do_kern_mount() in there
  Mirror MS_KERNMOUNT in ->mnt_flags
  get rid of useless vfsmount_lock use in put_mnt_ns()
  Take vfsmount_lock to fs/internal.h
  get rid of insanity with namespace roots in tomoyo
  take check for new events in namespace (guts of mounts_poll()) to namespace.c
  Don't mess with generic_permission() under ->d_lock in hpfs
  sanitize const/signedness for udf
  nilfs: sanitize const/signedness in dealing with ->d_name.name
  ...

Fix up fairly trivial (famous last words...) conflicts in
drivers/infiniband/core/uverbs_main.c and security/tomoyo/realpath.c

Documentation/filesystems/sharedsubtree.txt

@@ -837,6 +837,9 @@ replicas continue to be exactly same.
 	 individual lists does not affect propagation or the way propagation
 	 tree is modified by operations.
 
+	All vfsmounts in a peer group have the same ->mnt_master.  If it is
+	non-NULL, they form a contiguous (ordered) segment of slave list.
+
 	A example propagation tree looks as shown in the figure below.
 	[ NOTE: Though it looks like a forest, if we consider all the shared
 	mounts as a conceptual entity called 'pnode', it becomes a tree]
@@ -874,8 +877,19 @@ replicas continue to be exactly same.
 
 	NOTE: The propagation tree is orthogonal to the mount tree.
 
+8B Locking:
 
-8B Algorithm:
+	->mnt_share, ->mnt_slave, ->mnt_slave_list, ->mnt_master are protected
+	by namespace_sem (exclusive for modifications, shared for reading).
+
+	Normally we have ->mnt_flags modifications serialized by vfsmount_lock.
+	There are two exceptions: do_add_mount() and clone_mnt().
+	The former modifies a vfsmount that has not been visible in any shared
+	data structures yet.
+	The latter holds namespace_sem and the only references to vfsmount
+	are in lists that can't be traversed without namespace_sem.
+
+8C Algorithm:
 
 	The crux of the implementation resides in rbind/move operation.
 

arch/s390/hypfs/inode.c

@@ -288,46 +288,30 @@ static int hypfs_fill_super(struct super_block *sb, void *data, int silent)
 	sb->s_blocksize_bits = PAGE_CACHE_SHIFT;
 	sb->s_magic = HYPFS_MAGIC;
 	sb->s_op = &hypfs_s_ops;
-	if (hypfs_parse_options(data, sb)) {
+	if (hypfs_parse_options(data, sb))
-		rc = -EINVAL;
+		return -EINVAL;
-		goto err_alloc;
-	}
 	root_inode = hypfs_make_inode(sb, S_IFDIR | 0755);
-	if (!root_inode) {
+	if (!root_inode)
-		rc = -ENOMEM;
+		return -ENOMEM;
-		goto err_alloc;
-	}
 	root_inode->i_op = &simple_dir_inode_operations;
 	root_inode->i_fop = &simple_dir_operations;
-	root_dentry = d_alloc_root(root_inode);
+	sb->s_root = root_dentry = d_alloc_root(root_inode);
 	if (!root_dentry) {
 		iput(root_inode);
-		rc = -ENOMEM;
+		return -ENOMEM;
-		goto err_alloc;
 	}
 	if (MACHINE_IS_VM)
 		rc = hypfs_vm_create_files(sb, root_dentry);
 	else
 		rc = hypfs_diag_create_files(sb, root_dentry);
 	if (rc)
-		goto err_tree;
+		return rc;
 	sbi->update_file = hypfs_create_update_file(sb, root_dentry);
-	if (IS_ERR(sbi->update_file)) {
+	if (IS_ERR(sbi->update_file))
-		rc = PTR_ERR(sbi->update_file);
+		return PTR_ERR(sbi->update_file);
-		goto err_tree;
-	}
 	hypfs_update_update(sb);
-	sb->s_root = root_dentry;
 	pr_info("Hypervisor filesystem mounted\n");
 	return 0;
-
-err_tree:
-	hypfs_delete_tree(root_dentry);
-	d_genocide(root_dentry);
-	dput(root_dentry);
-err_alloc:
-	kfree(sbi);
-	return rc;
 }
 
 static int hypfs_get_super(struct file_system_type *fst, int flags,
@@ -340,12 +324,12 @@ static void hypfs_kill_super(struct super_block *sb)
 {
 	struct hypfs_sb_info *sb_info = sb->s_fs_info;
 
-	if (sb->s_root) {
+	if (sb->s_root)
 		hypfs_delete_tree(sb->s_root);
+	if (sb_info->update_file)
 		hypfs_remove(sb_info->update_file);
-		kfree(sb->s_fs_info);
+	kfree(sb->s_fs_info);
-		sb->s_fs_info = NULL;
+	sb->s_fs_info = NULL;
-	}
 	kill_litter_super(sb);
 }
 

arch/um/drivers/mconsole_kern.c

@@ -140,7 +140,7 @@ void mconsole_proc(struct mc_request *req)
 		goto out;
 	}
 
-	err = may_open(&nd.path, MAY_READ, FMODE_READ);
+	err = may_open(&nd.path, MAY_READ, O_RDONLY);
 	if (result) {
 		mconsole_reply(req, "Failed to open file", 1, 0);
 		path_put(&nd.path);

drivers/infiniband/core/uverbs.h

@@ -146,7 +146,7 @@ extern struct idr ib_uverbs_srq_idr;
 void idr_remove_uobj(struct idr *idp, struct ib_uobject *uobj);
 
 struct file *ib_uverbs_alloc_event_file(struct ib_uverbs_file *uverbs_file,
-					int is_async, int *fd);
+					int is_async);
 struct ib_uverbs_event_file *ib_uverbs_lookup_comp_file(int fd);
 
 void ib_uverbs_release_ucq(struct ib_uverbs_file *file,

drivers/infiniband/core/uverbs_cmd.c

@@ -301,10 +301,15 @@ ssize_t ib_uverbs_get_context(struct ib_uverbs_file *file,
 
 	resp.num_comp_vectors = file->device->num_comp_vectors;
 
-	filp = ib_uverbs_alloc_event_file(file, 1, &resp.async_fd);
+	ret = get_unused_fd();
+	if (ret < 0)
+		goto err_free;
+	resp.async_fd = ret;
+
+	filp = ib_uverbs_alloc_event_file(file, 1);
 	if (IS_ERR(filp)) {
 		ret = PTR_ERR(filp);
-		goto err_free;
+		goto err_fd;
 	}
 
 	if (copy_to_user((void __user *) (unsigned long) cmd.response,
@@ -332,9 +337,11 @@ ssize_t ib_uverbs_get_context(struct ib_uverbs_file *file,
 	return in_len;
 
 err_file:
-	put_unused_fd(resp.async_fd);
 	fput(filp);
 
+err_fd:
+	put_unused_fd(resp.async_fd);
+
 err_free:
 	ibdev->dealloc_ucontext(ucontext);
 
@@ -715,6 +722,7 @@ ssize_t ib_uverbs_create_comp_channel(struct ib_uverbs_file *file,
 	struct ib_uverbs_create_comp_channel	   cmd;
 	struct ib_uverbs_create_comp_channel_resp  resp;
 	struct file				  *filp;
+	int ret;
 
 	if (out_len < sizeof resp)
 		return -ENOSPC;
@@ -722,9 +730,16 @@ ssize_t ib_uverbs_create_comp_channel(struct ib_uverbs_file *file,
 	if (copy_from_user(&cmd, buf, sizeof cmd))
 		return -EFAULT;
 
-	filp = ib_uverbs_alloc_event_file(file, 0, &resp.fd);
+	ret = get_unused_fd();
-	if (IS_ERR(filp))
+	if (ret < 0)
+		return ret;
+	resp.fd = ret;
+
+	filp = ib_uverbs_alloc_event_file(file, 0);
+	if (IS_ERR(filp)) {
+		put_unused_fd(resp.fd);
 		return PTR_ERR(filp);
+	}
 
 	if (copy_to_user((void __user *) (unsigned long) cmd.response,
 			 &resp, sizeof resp)) {

drivers/infiniband/core/uverbs_main.c

@@ -484,11 +484,10 @@ void ib_uverbs_event_handler(struct ib_event_handler *handler,
 }
 
 struct file *ib_uverbs_alloc_event_file(struct ib_uverbs_file *uverbs_file,
-					int is_async, int *fd)
+					int is_async)
 {
 	struct ib_uverbs_event_file *ev_file;
 	struct file *filp;
-	int ret;
 
 	ev_file = kmalloc(sizeof *ev_file, GFP_KERNEL);
 	if (!ev_file)
@@ -503,27 +502,12 @@ struct file *ib_uverbs_alloc_event_file(struct ib_uverbs_file *uverbs_file,
 	ev_file->is_async    = is_async;
 	ev_file->is_closed   = 0;
 
-	*fd = get_unused_fd();
+	filp = anon_inode_getfile("[infinibandevent]", &uverbs_event_fops,
-	if (*fd < 0) {
-		ret = *fd;
-		goto err;
-	}
-
-	filp = anon_inode_getfile("[uverbs-event]", &uverbs_event_fops,
 				  ev_file, O_RDONLY);
-	if (!filp) {
+	if (IS_ERR(filp))
-		ret = -ENFILE;
+		kfree(ev_file);
-		goto err_fd;
-	}
 
 	return filp;
-
-err_fd:
-	put_unused_fd(*fd);
-
-err:
-	kfree(ev_file);
-	return ERR_PTR(ret);
 }
 
 /*

drivers/usb/gadget/f_mass_storage.c

@@ -1050,7 +1050,7 @@ static void invalidate_sub(struct fsg_lun *curlun)
 	unsigned long	rc;
 
 	rc = invalidate_mapping_pages(inode->i_mapping, 0, -1);
-	VLDBG(curlun, "invalidate_inode_pages -> %ld\n", rc);
+	VLDBG(curlun, "invalidate_mapping_pages -> %ld\n", rc);
 }
 
 static int do_verify(struct fsg_common *common)

drivers/usb/gadget/file_storage.c

@@ -1448,7 +1448,7 @@ static void invalidate_sub(struct fsg_lun *curlun)
 	unsigned long	rc;
 
 	rc = invalidate_mapping_pages(inode->i_mapping, 0, -1);
-	VLDBG(curlun, "invalidate_inode_pages -> %ld\n", rc);
+	VLDBG(curlun, "invalidate_mapping_pages -> %ld\n", rc);
 }
 
 static int do_verify(struct fsg_dev *fsg)

fs/autofs4/autofs_i.h

@@ -60,11 +60,6 @@ do {							\
 		current->pid, __func__, ##args);	\
 } while (0)
 
-struct rehash_entry {
-	struct task_struct *task;
-	struct list_head list;
-};
-
 /* Unified info structure.  This is pointed to by both the dentry and
    inode structures.  Each file in the filesystem has an instance of this
    structure.  It holds a reference to the dentry, so dentries are never
@@ -81,7 +76,6 @@ struct autofs_info {
 
 	struct list_head active;
 	int active_count;
-	struct list_head rehash_list;
 
 	struct list_head expiring;
 
@@ -104,7 +98,6 @@ struct autofs_info {
 #define AUTOFS_INF_EXPIRING	(1<<0) /* dentry is in the process of expiring */
 #define AUTOFS_INF_MOUNTPOINT	(1<<1) /* mountpoint status for direct expire */
 #define AUTOFS_INF_PENDING	(1<<2) /* dentry pending mount */
-#define AUTOFS_INF_REHASH	(1<<3) /* dentry in transit to ->lookup() */
 
 struct autofs_wait_queue {
 	wait_queue_head_t queue;

fs/autofs4/dev-ioctl.c

@@ -544,10 +544,9 @@ static int autofs_dev_ioctl_ismountpoint(struct file *fp,
 			goto out;
 		devid = new_encode_dev(path.mnt->mnt_sb->s_dev);
 		err = 0;
-		if (path.dentry->d_inode &&
+		if (path.mnt->mnt_root == path.dentry) {
-		    path.mnt->mnt_root == path.dentry) {
 			err = 1;
-			magic = path.dentry->d_inode->i_sb->s_magic;
+			magic = path.mnt->mnt_sb->s_magic;
 		}
 	} else {
 		dev_t dev = sbi->sb->s_dev;
@@ -560,10 +559,8 @@ static int autofs_dev_ioctl_ismountpoint(struct file *fp,
 
 		err = have_submounts(path.dentry);
 
-		if (path.mnt->mnt_mountpoint != path.mnt->mnt_root) {
+		if (follow_down(&path))
-			if (follow_down(&path))
+			magic = path.mnt->mnt_sb->s_magic;
-				magic = path.mnt->mnt_sb->s_magic;
-		}
 	}
 
 	param->ismountpoint.out.devid = devid;

fs/autofs4/expire.c

@@ -279,7 +279,6 @@ struct dentry *autofs4_expire_direct(struct super_block *sb,
 			root->d_mounted--;
 		}
 		ino->flags |= AUTOFS_INF_EXPIRING;
-		autofs4_add_expiring(root);
 		init_completion(&ino->expire_complete);
 		spin_unlock(&sbi->fs_lock);
 		return root;
@@ -407,7 +406,6 @@ found:
 		expired, (int)expired->d_name.len, expired->d_name.name);
 	ino = autofs4_dentry_ino(expired);
 	ino->flags |= AUTOFS_INF_EXPIRING;
-	autofs4_add_expiring(expired);
 	init_completion(&ino->expire_complete);
 	spin_unlock(&sbi->fs_lock);
 	spin_lock(&dcache_lock);
@@ -435,7 +433,7 @@ int autofs4_expire_wait(struct dentry *dentry)
 
 		DPRINTK("expire done status=%d", status);
 
-		if (d_unhashed(dentry) && IS_DEADDIR(dentry->d_inode))
+		if (d_unhashed(dentry))
 			return -EAGAIN;
 
 		return status;
@@ -475,7 +473,6 @@ int autofs4_expire_run(struct super_block *sb,
 	spin_lock(&sbi->fs_lock);
 	ino = autofs4_dentry_ino(dentry);
 	ino->flags &= ~AUTOFS_INF_EXPIRING;
-	autofs4_del_expiring(dentry);
 	complete_all(&ino->expire_complete);
 	spin_unlock(&sbi->fs_lock);
 
@@ -506,7 +503,6 @@ int autofs4_do_expire_multi(struct super_block *sb, struct vfsmount *mnt,
 			ino->flags &= ~AUTOFS_INF_MOUNTPOINT;
 		}
 		ino->flags &= ~AUTOFS_INF_EXPIRING;
-		autofs4_del_expiring(dentry);
 		complete_all(&ino->expire_complete);
 		spin_unlock(&sbi->fs_lock);
 		dput(dentry);

fs/autofs4/inode.c

@@ -49,7 +49,6 @@ struct autofs_info *autofs4_init_ino(struct autofs_info *ino,
 		ino->dentry = NULL;
 		ino->size = 0;
 		INIT_LIST_HEAD(&ino->active);
-		INIT_LIST_HEAD(&ino->rehash_list);
 		ino->active_count = 0;
 		INIT_LIST_HEAD(&ino->expiring);
 		atomic_set(&ino->count, 0);
@@ -97,63 +96,6 @@ void autofs4_free_ino(struct autofs_info *ino)
 	kfree(ino);
 }
 
-/*
- * Deal with the infamous "Busy inodes after umount ..." message.
- *
- * Clean up the dentry tree. This happens with autofs if the user
- * space program goes away due to a SIGKILL, SIGSEGV etc.
- */
-static void autofs4_force_release(struct autofs_sb_info *sbi)
-{
-	struct dentry *this_parent = sbi->sb->s_root;
-	struct list_head *next;
-
-	if (!sbi->sb->s_root)
-		return;
-
-	spin_lock(&dcache_lock);
-repeat:
-	next = this_parent->d_subdirs.next;
-resume:
-	while (next != &this_parent->d_subdirs) {
-		struct dentry *dentry = list_entry(next, struct dentry, d_u.d_child);
-
-		/* Negative dentry - don`t care */
-		if (!simple_positive(dentry)) {
-			next = next->next;
-			continue;
-		}
-
-		if (!list_empty(&dentry->d_subdirs)) {
-			this_parent = dentry;
-			goto repeat;
-		}
-
-		next = next->next;
-		spin_unlock(&dcache_lock);
-
-		DPRINTK("dentry %p %.*s",
-			dentry, (int)dentry->d_name.len, dentry->d_name.name);
-
-		dput(dentry);
-		spin_lock(&dcache_lock);
-	}
-
-	if (this_parent != sbi->sb->s_root) {
-		struct dentry *dentry = this_parent;
-
-		next = this_parent->d_u.d_child.next;
-		this_parent = this_parent->d_parent;
-		spin_unlock(&dcache_lock);
-		DPRINTK("parent dentry %p %.*s",
-			dentry, (int)dentry->d_name.len, dentry->d_name.name);
-		dput(dentry);
-		spin_lock(&dcache_lock);
-		goto resume;
-	}
-	spin_unlock(&dcache_lock);
-}
-
 void autofs4_kill_sb(struct super_block *sb)
 {
 	struct autofs_sb_info *sbi = autofs4_sbi(sb);
@@ -170,15 +112,12 @@ void autofs4_kill_sb(struct super_block *sb)
 	/* Free wait queues, close pipe */
 	autofs4_catatonic_mode(sbi);
 
-	/* Clean up and release dangling references */
-	autofs4_force_release(sbi);
-
 	sb->s_fs_info = NULL;
 	kfree(sbi);
 
 out_kill_sb:
 	DPRINTK("shutting down");
-	kill_anon_super(sb);
+	kill_litter_super(sb);
 }
 
 static int autofs4_show_options(struct seq_file *m, struct vfsmount *mnt)

fs/autofs4/root.c

@@ -104,99 +104,6 @@ static void autofs4_del_active(struct dentry *dentry)
 	return;
 }
 
-static void autofs4_add_rehash_entry(struct autofs_info *ino,
-				     struct rehash_entry *entry)
-{
-	entry->task = current;
-	INIT_LIST_HEAD(&entry->list);
-	list_add(&entry->list, &ino->rehash_list);
-	return;
-}
-
-static void autofs4_remove_rehash_entry(struct autofs_info *ino)
-{
-	struct list_head *head = &ino->rehash_list;
-	struct rehash_entry *entry;
-	list_for_each_entry(entry, head, list) {
-		if (entry->task == current) {
-			list_del(&entry->list);
-			kfree(entry);
-			break;
-		}
-	}
-	return;
-}
-
-static void autofs4_remove_rehash_entrys(struct autofs_info *ino)
-{
-	struct autofs_sb_info *sbi = ino->sbi;
-	struct rehash_entry *entry, *next;
-	struct list_head *head;
-
-	spin_lock(&sbi->fs_lock);
-	spin_lock(&sbi->lookup_lock);
-	if (!(ino->flags & AUTOFS_INF_REHASH)) {
-		spin_unlock(&sbi->lookup_lock);
-		spin_unlock(&sbi->fs_lock);
-		return;
-	}
-	ino->flags &= ~AUTOFS_INF_REHASH;
-	head = &ino->rehash_list;
-	list_for_each_entry_safe(entry, next, head, list) {
-		list_del(&entry->list);
-		kfree(entry);
-	}
-	spin_unlock(&sbi->lookup_lock);
-	spin_unlock(&sbi->fs_lock);
-	dput(ino->dentry);
-
-	return;
-}
-
-static void autofs4_revalidate_drop(struct dentry *dentry,
-				    struct rehash_entry *entry)
-{
-	struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb);
-	struct autofs_info *ino = autofs4_dentry_ino(dentry);
-	/*
-	 * Add to the active list so we can pick this up in
-	 * ->lookup(). Also add an entry to a rehash list so
-	 * we know when there are no dentrys in flight so we
-	 * know when we can rehash the dentry.
-	 */
-	spin_lock(&sbi->lookup_lock);
-	if (list_empty(&ino->active))
-		list_add(&ino->active, &sbi->active_list);
-	autofs4_add_rehash_entry(ino, entry);
-	spin_unlock(&sbi->lookup_lock);
-	if (!(ino->flags & AUTOFS_INF_REHASH)) {
-		ino->flags |= AUTOFS_INF_REHASH;
-		dget(dentry);
-		spin_lock(&dentry->d_lock);
-		__d_drop(dentry);
-		spin_unlock(&dentry->d_lock);
-	}
-	return;
-}
-
-static void autofs4_revalidate_rehash(struct dentry *dentry)
-{
-	struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb);
-	struct autofs_info *ino = autofs4_dentry_ino(dentry);
-	if (ino->flags & AUTOFS_INF_REHASH) {
-		spin_lock(&sbi->lookup_lock);
-		autofs4_remove_rehash_entry(ino);
-		if (list_empty(&ino->rehash_list)) {
-			spin_unlock(&sbi->lookup_lock);
-			ino->flags &= ~AUTOFS_INF_REHASH;
-			d_rehash(dentry);
-			dput(ino->dentry);
-		} else
-			spin_unlock(&sbi->lookup_lock);
-	}
-	return;
-}
-
 static unsigned int autofs4_need_mount(unsigned int flags)
 {
 	unsigned int res = 0;
@@ -236,7 +143,7 @@ out:
 	return dcache_dir_open(inode, file);
 }
 
-static int try_to_fill_dentry(struct dentry *dentry)
+static int try_to_fill_dentry(struct dentry *dentry, int flags)
 {
 	struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb);
 	struct autofs_info *ino = autofs4_dentry_ino(dentry);
@@ -249,17 +156,55 @@ static int try_to_fill_dentry(struct dentry *dentry)
 	 * Wait for a pending mount, triggering one if there
 	 * isn't one already
 	 */
-	DPRINTK("waiting for mount name=%.*s",
+	if (dentry->d_inode == NULL) {
-		 dentry->d_name.len, dentry->d_name.name);
+		DPRINTK("waiting for mount name=%.*s",
+			 dentry->d_name.len, dentry->d_name.name);
 
-	status = autofs4_wait(sbi, dentry, NFY_MOUNT);
+		status = autofs4_wait(sbi, dentry, NFY_MOUNT);
 
-	DPRINTK("mount done status=%d", status);
+		DPRINTK("mount done status=%d", status);
 
-	/* Update expiry counter */
+		/* Turn this into a real negative dentry? */
-	ino->last_used = jiffies;
+		if (status == -ENOENT) {
+			spin_lock(&sbi->fs_lock);
+			ino->flags &= ~AUTOFS_INF_PENDING;
+			spin_unlock(&sbi->fs_lock);
+			return status;
+		} else if (status) {
+			/* Return a negative dentry, but leave it "pending" */
+			return status;
+		}
+	/* Trigger mount for path component or follow link */
+	} else if (ino->flags & AUTOFS_INF_PENDING ||
+			autofs4_need_mount(flags) ||
+			current->link_count) {
+		DPRINTK("waiting for mount name=%.*s",
+			dentry->d_name.len, dentry->d_name.name);
 
-	return status;
+		spin_lock(&sbi->fs_lock);
+		ino->flags |= AUTOFS_INF_PENDING;
+		spin_unlock(&sbi->fs_lock);
+		status = autofs4_wait(sbi, dentry, NFY_MOUNT);
+
+		DPRINTK("mount done status=%d", status);
+
+		if (status) {
+			spin_lock(&sbi->fs_lock);
+			ino->flags &= ~AUTOFS_INF_PENDING;
+			spin_unlock(&sbi->fs_lock);
+			return status;
+		}
+	}
+
+	/* Initialize expiry counter after successful mount */
+	if (ino)
+		ino->last_used = jiffies;
+
+	spin_lock(&sbi->fs_lock);
+	ino->flags &= ~AUTOFS_INF_PENDING;
+	spin_unlock(&sbi->fs_lock);
+
+	return 0;
 }
 
 /* For autofs direct mounts the follow link triggers the mount */
@@ -313,16 +258,10 @@ static void *autofs4_follow_link(struct dentry *dentry, struct nameidata *nd)
 	 */
 	if (ino->flags & AUTOFS_INF_PENDING ||
 	    (!d_mountpoint(dentry) && list_empty(&dentry->d_subdirs))) {
-		ino->flags |= AUTOFS_INF_PENDING;
 		spin_unlock(&dcache_lock);
 		spin_unlock(&sbi->fs_lock);
 
-		status = try_to_fill_dentry(dentry);
+		status = try_to_fill_dentry(dentry, 0);
-
-		spin_lock(&sbi->fs_lock);
-		ino->flags &= ~AUTOFS_INF_PENDING;
-		spin_unlock(&sbi->fs_lock);
-
 		if (status)
 			goto out_error;
 
@@ -361,47 +300,18 @@ static int autofs4_revalidate(struct dentry *dentry, struct nameidata *nd)
 {
 	struct inode *dir = dentry->d_parent->d_inode;
 	struct autofs_sb_info *sbi = autofs4_sbi(dir->i_sb);
-	struct autofs_info *ino = autofs4_dentry_ino(dentry);
+	int oz_mode = autofs4_oz_mode(sbi);
-	struct rehash_entry *entry;
 	int flags = nd ? nd->flags : 0;
-	unsigned int mutex_aquired;
+	int status = 1;
 
-	DPRINTK("name = %.*s oz_mode = %d",
-		dentry->d_name.len, dentry->d_name.name, oz_mode);
-
-	/* Daemon never causes a mount to trigger */
-	if (autofs4_oz_mode(sbi))
-		return 1;
-
-	entry = kmalloc(sizeof(struct rehash_entry), GFP_KERNEL);
-	if (!entry)
-		return -ENOMEM;
-
-	mutex_aquired = mutex_trylock(&dir->i_mutex);
-
-	spin_lock(&sbi->fs_lock);
-	spin_lock(&dcache_lock);
 	/* Pending dentry */
+	spin_lock(&sbi->fs_lock);
 	if (autofs4_ispending(dentry)) {
-		int status;
+		/* The daemon never causes a mount to trigger */
-
-		/*
-		 * We can only unhash and send this to ->lookup() if
-		 * the directory mutex is held over d_revalidate() and
-		 * ->lookup(). This prevents the VFS from incorrectly
-		 * seeing the dentry as non-existent.
-		 */
-		ino->flags |= AUTOFS_INF_PENDING;
-		if (!mutex_aquired) {
-			autofs4_revalidate_drop(dentry, entry);
-			spin_unlock(&dcache_lock);
-			spin_unlock(&sbi->fs_lock);
-			return 0;
-		}
-		spin_unlock(&dcache_lock);
 		spin_unlock(&sbi->fs_lock);
-		mutex_unlock(&dir->i_mutex);
+
-		kfree(entry);
+		if (oz_mode)
+			return 1;
 
 		/*
 		 * If the directory has gone away due to an expire
@@ -415,82 +325,45 @@ static int autofs4_revalidate(struct dentry *dentry, struct nameidata *nd)
 		 * A zero status is success otherwise we have a
 		 * negative error code.
 		 */
-		status = try_to_fill_dentry(dentry);
+		status = try_to_fill_dentry(dentry, flags);
-
-		spin_lock(&sbi->fs_lock);
-		ino->flags &= ~AUTOFS_INF_PENDING;
-		spin_unlock(&sbi->fs_lock);
-
 		if (status == 0)
 			return 1;
 
 		return status;
 	}
+	spin_unlock(&sbi->fs_lock);
+
+	/* Negative dentry.. invalidate if "old" */
+	if (dentry->d_inode == NULL)
+		return 0;
 
 	/* Check for a non-mountpoint directory with no contents */
+	spin_lock(&dcache_lock);
 	if (S_ISDIR(dentry->d_inode->i_mode) &&
 	    !d_mountpoint(dentry) && list_empty(&dentry->d_subdirs)) {
 		DPRINTK("dentry=%p %.*s, emptydir",
 			 dentry, dentry->d_name.len, dentry->d_name.name);
+		spin_unlock(&dcache_lock);
 
-		if (autofs4_need_mount(flags) || current->link_count) {
+		/* The daemon never causes a mount to trigger */
-			int status;
+		if (oz_mode)
+			return 1;
 
-			/*
+		/*
-			 * We can only unhash and send this to ->lookup() if
+		 * A zero status is success otherwise we have a
-			 * the directory mutex is held over d_revalidate() and
+		 * negative error code.
-			 * ->lookup(). This prevents the VFS from incorrectly
+		 */
-			 * seeing the dentry as non-existent.
+		status = try_to_fill_dentry(dentry, flags);
-			 */
+		if (status == 0)
-			ino->flags |= AUTOFS_INF_PENDING;
+			return 1;
-			if (!mutex_aquired) {
-				autofs4_revalidate_drop(dentry, entry);
-				spin_unlock(&dcache_lock);
-				spin_unlock(&sbi->fs_lock);
-				return 0;
-			}
-			spin_unlock(&dcache_lock);
-			spin_unlock(&sbi->fs_lock);
-			mutex_unlock(&dir->i_mutex);
-			kfree(entry);
 
-			/*
+		return status;
-			 * A zero status is success otherwise we have a
-			 * negative error code.
-			 */
-			status = try_to_fill_dentry(dentry);
-
-			spin_lock(&sbi->fs_lock);
-			ino->flags &= ~AUTOFS_INF_PENDING;
-			spin_unlock(&sbi->fs_lock);
-
-			if (status == 0)
-				return 1;
-
-			return status;
-		}
 	}
 	spin_unlock(&dcache_lock);
-	spin_unlock(&sbi->fs_lock);
-
-	if (mutex_aquired)
-		mutex_unlock(&dir->i_mutex);
-
-	kfree(entry);
 
 	return 1;
 }
 
-static void autofs4_free_rehash_entrys(struct autofs_info *inf)
-{
-	struct list_head *head = &inf->rehash_list;
-	struct rehash_entry *entry, *next;
-	list_for_each_entry_safe(entry, next, head, list) {
-		list_del(&entry->list);
-		kfree(entry);
-	}
-}
-
 void autofs4_dentry_release(struct dentry *de)
 {
 	struct autofs_info *inf;
@@ -509,8 +382,6 @@ void autofs4_dentry_release(struct dentry *de)
 				list_del(&inf->active);
 			if (!list_empty(&inf->expiring))
 				list_del(&inf->expiring);
-			if (!list_empty(&inf->rehash_list))
-				autofs4_free_rehash_entrys(inf);
 			spin_unlock(&sbi->lookup_lock);
 		}
 
@@ -543,7 +414,6 @@ static struct dentry *autofs4_lookup_active(struct dentry *dentry)
 	const unsigned char *str = name->name;
 	struct list_head *p, *head;
 
-restart:
 	spin_lock(&dcache_lock);
 	spin_lock(&sbi->lookup_lock);
 	head = &sbi->active_list;
@@ -561,19 +431,6 @@ restart:
 		if (atomic_read(&active->d_count) == 0)
 			goto next;
 
-		if (active->d_inode && IS_DEADDIR(active->d_inode)) {
-			if (!list_empty(&ino->rehash_list)) {
-				dget(active);
-				spin_unlock(&active->d_lock);
-				spin_unlock(&sbi->lookup_lock);
-				spin_unlock(&dcache_lock);
-				autofs4_remove_rehash_entrys(ino);
-				dput(active);
-				goto restart;
-			}
-			goto next;
-		}
-
 		qstr = &active->d_name;
 
 		if (active->d_name.hash != hash)
@@ -586,11 +443,13 @@ restart:
 		if (memcmp(qstr->name, str, len))
 			goto next;
 
-		dget(active);
+		if (d_unhashed(active)) {
-		spin_unlock(&active->d_lock);
+			dget(active);
-		spin_unlock(&sbi->lookup_lock);
+			spin_unlock(&active->d_lock);
-		spin_unlock(&dcache_lock);
+			spin_unlock(&sbi->lookup_lock);
-		return active;
+			spin_unlock(&dcache_lock);
+			return active;
+		}
 next:
 		spin_unlock(&active->d_lock);
 	}
@@ -639,11 +498,13 @@ static struct dentry *autofs4_lookup_expiring(struct dentry *dentry)
 		if (memcmp(qstr->name, str, len))
 			goto next;
 
-		dget(expiring);
+		if (d_unhashed(expiring)) {
-		spin_unlock(&expiring->d_lock);
+			dget(expiring);
-		spin_unlock(&sbi->lookup_lock);
+			spin_unlock(&expiring->d_lock);
-		spin_unlock(&dcache_lock);
+			spin_unlock(&sbi->lookup_lock);
-		return expiring;
+			spin_unlock(&dcache_lock);
+			return expiring;
+		}
 next:
 		spin_unlock(&expiring->d_lock);
 	}
@@ -653,48 +514,6 @@ next:
 	return NULL;
 }
 
-static struct autofs_info *init_new_dentry(struct autofs_sb_info *sbi,
-					   struct dentry *dentry, int oz_mode)
-{
-	struct autofs_info *ino;
-
-	/*
-	 * Mark the dentry incomplete but don't hash it. We do this
-	 * to serialize our inode creation operations (symlink and
-	 * mkdir) which prevents deadlock during the callback to
-	 * the daemon. Subsequent user space lookups for the same
-	 * dentry are placed on the wait queue while the daemon
-	 * itself is allowed passage unresticted so the create
-	 * operation itself can then hash the dentry. Finally,
-	 * we check for the hashed dentry and return the newly
-	 * hashed dentry.
-	 */
-	dentry->d_op = &autofs4_root_dentry_operations;
-
-	/*
-	 * And we need to ensure that the same dentry is used for
-	 * all following lookup calls until it is hashed so that
-	 * the dentry flags are persistent throughout the request.
-	 */
-	ino = autofs4_init_ino(NULL, sbi, 0555);
-	if (!ino)
-		return ERR_PTR(-ENOMEM);
-
-	dentry->d_fsdata = ino;
-	ino->dentry = dentry;
-
-	/*
-	 * Only set the mount pending flag for new dentrys not created
-	 * by the daemon.
-	 */
-	if (!oz_mode)
-		ino->flags |= AUTOFS_INF_PENDING;
-
-	d_instantiate(dentry, NULL);
-
-	return ino;
-}
-
 /* Lookups in the root directory */
 static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd)
 {
@@ -702,7 +521,6 @@ static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, s
 	struct autofs_info *ino;
 	struct dentry *expiring, *active;
 	int oz_mode;
-	int status = 0;
 
 	DPRINTK("name = %.*s",
 		dentry->d_name.len, dentry->d_name.name);
@@ -717,26 +535,44 @@ static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, s
 	DPRINTK("pid = %u, pgrp = %u, catatonic = %d, oz_mode = %d",
 		 current->pid, task_pgrp_nr(current), sbi->catatonic, oz_mode);
 
-	spin_lock(&sbi->fs_lock);
 	active = autofs4_lookup_active(dentry);
 	if (active) {
 		dentry = active;
 		ino = autofs4_dentry_ino(dentry);
-		/* If this came from revalidate, rehash it */
-		autofs4_revalidate_rehash(dentry);
-		spin_unlock(&sbi->fs_lock);
 	} else {
-		spin_unlock(&sbi->fs_lock);
+		/*
-		ino = init_new_dentry(sbi, dentry, oz_mode);
+		 * Mark the dentry incomplete but don't hash it. We do this
-		if (IS_ERR(ino))
+		 * to serialize our inode creation operations (symlink and
-			return (struct dentry *) ino;
+		 * mkdir) which prevents deadlock during the callback to
+		 * the daemon. Subsequent user space lookups for the same
+		 * dentry are placed on the wait queue while the daemon
+		 * itself is allowed passage unresticted so the create
+		 * operation itself can then hash the dentry. Finally,
+		 * we check for the hashed dentry and return the newly
+		 * hashed dentry.
+		 */
+		dentry->d_op = &autofs4_root_dentry_operations;
+
+		/*
+		 * And we need to ensure that the same dentry is used for
+		 * all following lookup calls until it is hashed so that
+		 * the dentry flags are persistent throughout the request.
+		 */
+		ino = autofs4_init_ino(NULL, sbi, 0555);
+		if (!ino)
+			return ERR_PTR(-ENOMEM);
+
+		dentry->d_fsdata = ino;
+		ino->dentry = dentry;
+
+		autofs4_add_active(dentry);
+
+		d_instantiate(dentry, NULL);
 	}
 
-	autofs4_add_active(dentry);
-
 	if (!oz_mode) {
-		expiring = autofs4_lookup_expiring(dentry);
 		mutex_unlock(&dir->i_mutex);
+		expiring = autofs4_lookup_expiring(dentry);
 		if (expiring) {
 			/*
 			 * If we are racing with expire the request might not
@@ -744,22 +580,23 @@ static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, s
 			 * so it must have been successful, so just wait for it.
 			 */
 			autofs4_expire_wait(expiring);
+			autofs4_del_expiring(expiring);
 			dput(expiring);
 		}
-		status = try_to_fill_dentry(dentry);
+
-		mutex_lock(&dir->i_mutex);
 		spin_lock(&sbi->fs_lock);
-		ino->flags &= ~AUTOFS_INF_PENDING;
+		ino->flags |= AUTOFS_INF_PENDING;
 		spin_unlock(&sbi->fs_lock);
+		if (dentry->d_op && dentry->d_op->d_revalidate)
+			(dentry->d_op->d_revalidate)(dentry, nd);
+		mutex_lock(&dir->i_mutex);
 	}
 
-	autofs4_del_active(dentry);
-
 	/*
-	 * If we had a mount fail, check if we had to handle
+	 * If we are still pending, check if we had to handle
 	 * a signal. If so we can force a restart..
 	 */
-	if (status) {
+	if (ino->flags & AUTOFS_INF_PENDING) {
 		/* See if we were interrupted */
 		if (signal_pending(current)) {
 			sigset_t *sigset = &current->pending.signal;
@@ -771,46 +608,43 @@ static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, s
 			    return ERR_PTR(-ERESTARTNOINTR);
 			}
 		}
-	}
+		if (!oz_mode) {
-
+			spin_lock(&sbi->fs_lock);
-	/*
+			ino->flags &= ~AUTOFS_INF_PENDING;
-	 * User space can (and has done in the past) remove and re-create
+			spin_unlock(&sbi->fs_lock);
-	 * this directory during the callback. This can leave us with an
-	 * unhashed dentry, but a successful mount!  So we need to
-	 * perform another cached lookup in case the dentry now exists.
-	 */
-	if (!oz_mode && !have_submounts(dentry)) {
-		struct dentry *new;
-		new = d_lookup(dentry->d_parent, &dentry->d_name);
-		if (new) {
-			if (active)
-				dput(active);
-			return new;
-		} else {
-			if (!status)
-				status = -ENOENT;
 		}
 	}
 
 	/*
-	 * If we had a mount failure, return status to user space.
+	 * If this dentry is unhashed, then we shouldn't honour this
-	 * If the mount succeeded and we used a dentry from the active queue
+	 * lookup.  Returning ENOENT here doesn't do the right thing
-	 * return it.
+	 * for all system calls, but it should be OK for the operations
+	 * we permit from an autofs.
 	 */
-	if (status) {
+	if (!oz_mode && d_unhashed(dentry)) {
-		dentry = ERR_PTR(status);
+		/*
+		 * A user space application can (and has done in the past)
+		 * remove and re-create this directory during the callback.
+		 * This can leave us with an unhashed dentry, but a
+		 * successful mount!  So we need to perform another
+		 * cached lookup in case the dentry now exists.
+		 */
+		struct dentry *parent = dentry->d_parent;
+		struct dentry *new = d_lookup(parent, &dentry->d_name);
+		if (new != NULL)
+			dentry = new;
+		else
+			dentry = ERR_PTR(-ENOENT);
+
 		if (active)
 			dput(active);
+
 		return dentry;
-	} else {
-		/*
-		 * Valid successful mount, return active dentry or NULL
-		 * for a new dentry.
-		 */
-		if (active)
-			return active;
 	}
 
+	if (active)
+		return active;
+
 	return NULL;
 }
 
@@ -834,6 +668,8 @@ static int autofs4_dir_symlink(struct inode *dir,
 	if (!ino)
 		return -ENOMEM;
 
+	autofs4_del_active(dentry);
+
 	ino->size = strlen(symname);
 	cp = kmalloc(ino->size + 1, GFP_KERNEL);
 	if (!cp) {
@@ -910,6 +746,7 @@ static int autofs4_dir_unlink(struct inode *dir, struct dentry *dentry)
 	dir->i_mtime = CURRENT_TIME;
 
 	spin_lock(&dcache_lock);
+	autofs4_add_expiring(dentry);
 	spin_lock(&dentry->d_lock);
 	__d_drop(dentry);
 	spin_unlock(&dentry->d_lock);
@@ -935,6 +772,7 @@ static int autofs4_dir_rmdir(struct inode *dir, struct dentry *dentry)
 		spin_unlock(&dcache_lock);
 		return -ENOTEMPTY;
 	}
+	autofs4_add_expiring(dentry);
 	spin_lock(&dentry->d_lock);
 	__d_drop(dentry);
 	spin_unlock(&dentry->d_lock);
@@ -972,6 +810,8 @@ static int autofs4_dir_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	if (!ino)
 		return -ENOMEM;
 
+	autofs4_del_active(dentry);
+
 	inode = autofs4_get_inode(dir->i_sb, ino);
 	if (!inode) {
 		if (!dentry->d_fsdata)

fs/cifs/file.c

@@ -2289,9 +2289,9 @@ cifs_oplock_break(struct slow_work *work)
 	if (inode && S_ISREG(inode->i_mode)) {
 #ifdef CONFIG_CIFS_EXPERIMENTAL
 		if (cinode->clientCanCacheAll == 0)
-			break_lease(inode, FMODE_READ);
+			break_lease(inode, O_RDONLY);
 		else if (cinode->clientCanCacheRead == 0)
-			break_lease(inode, FMODE_WRITE);
+			break_lease(inode, O_WRONLY);
 #endif
 		rc = filemap_fdatawrite(inode->i_mapping);
 		if (cinode->clientCanCacheRead == 0) {

fs/dcache.c

@@ -257,6 +257,7 @@ kill_it:
 	if (dentry)
 		goto repeat;
 }
+EXPORT_SYMBOL(dput);
 
 /**
  * d_invalidate - invalidate a dentry
@@ -314,6 +315,7 @@ int d_invalidate(struct dentry * dentry)
 	spin_unlock(&dcache_lock);
 	return 0;
 }
+EXPORT_SYMBOL(d_invalidate);
 
 /* This should be called _only_ with dcache_lock held */
 
@@ -328,6 +330,7 @@ struct dentry * dget_locked(struct dentry *dentry)
 {
 	return __dget_locked(dentry);
 }
+EXPORT_SYMBOL(dget_locked);
 
 /**
  * d_find_alias - grab a hashed alias of inode
@@ -384,6 +387,7 @@ struct dentry * d_find_alias(struct inode *inode)
 	}
 	return de;
 }
+EXPORT_SYMBOL(d_find_alias);
 
 /*
  *	Try to kill dentries associated with this inode.
@@ -408,6 +412,7 @@ restart:
 	}
 	spin_unlock(&dcache_lock);
 }
+EXPORT_SYMBOL(d_prune_aliases);
 
 /*
  * Throw away a dentry - free the inode, dput the parent.  This requires that
@@ -610,6 +615,7 @@ void shrink_dcache_sb(struct super_block * sb)
 {
 	__shrink_dcache_sb(sb, NULL, 0);
 }
+EXPORT_SYMBOL(shrink_dcache_sb);
 
 /*
  * destroy a single subtree of dentries for unmount
@@ -792,6 +798,7 @@ positive:
 	spin_unlock(&dcache_lock);
 	return 1;
 }
+EXPORT_SYMBOL(have_submounts);
 
 /*
  * Search the dentry child list for the specified parent,
@@ -876,6 +883,7 @@ void shrink_dcache_parent(struct dentry * parent)
 	while ((found = select_parent(parent)) != 0)
 		__shrink_dcache_sb(sb, &found, 0);
 }
+EXPORT_SYMBOL(shrink_dcache_parent);
 
 /*
  * Scan `nr' dentries and return the number which remain.
@@ -968,6 +976,7 @@ struct dentry *d_alloc(struct dentry * parent, const struct qstr *name)
 
 	return dentry;
 }
+EXPORT_SYMBOL(d_alloc);
 
 struct dentry *d_alloc_name(struct dentry *parent, const char *name)
 {
@@ -1012,6 +1021,7 @@ void d_instantiate(struct dentry *entry, struct inode * inode)
 	spin_unlock(&dcache_lock);
 	security_d_instantiate(entry, inode);
 }
+EXPORT_SYMBOL(d_instantiate);
 
 /**
  * d_instantiate_unique - instantiate a non-aliased dentry
@@ -1108,6 +1118,7 @@ struct dentry * d_alloc_root(struct inode * root_inode)
 	}
 	return res;
 }
+EXPORT_SYMBOL(d_alloc_root);
 
 static inline struct hlist_head *d_hash(struct dentry *parent,
 					unsigned long hash)
@@ -1211,7 +1222,6 @@ struct dentry *d_splice_alias(struct inode *inode, struct dentry *dentry)
 			BUG_ON(!(new->d_flags & DCACHE_DISCONNECTED));
 			spin_unlock(&dcache_lock);
 			security_d_instantiate(new, inode);
-			d_rehash(dentry);
 			d_move(new, dentry);
 			iput(inode);
 		} else {
@@ -1225,6 +1235,7 @@ struct dentry *d_splice_alias(struct inode *inode, struct dentry *dentry)
 		d_add(dentry, inode);
 	return new;
 }
+EXPORT_SYMBOL(d_splice_alias);
 
 /**
  * d_add_ci - lookup or allocate new dentry with case-exact name
@@ -1314,6 +1325,7 @@ err_out:
 	iput(inode);
 	return ERR_PTR(error);
 }
+EXPORT_SYMBOL(d_add_ci);
 
 /**
  * d_lookup - search for a dentry
@@ -1357,6 +1369,7 @@ struct dentry * d_lookup(struct dentry * parent, struct qstr * name)
 	} while (read_seqretry(&rename_lock, seq));
 	return dentry;
 }
+EXPORT_SYMBOL(d_lookup);
 
 struct dentry * __d_lookup(struct dentry * parent, struct qstr * name)
 {
@@ -1483,6 +1496,7 @@ int d_validate(struct dentry *dentry, struct dentry *dparent)
 out:
 	return 0;
 }
+EXPORT_SYMBOL(d_validate);
 
 /*
  * When a file is deleted, we have two options:
@@ -1528,6 +1542,7 @@ void d_delete(struct dentry * dentry)
 
 	fsnotify_nameremove(dentry, isdir);
 }
+EXPORT_SYMBOL(d_delete);
 
 static void __d_rehash(struct dentry * entry, struct hlist_head *list)
 {
@@ -1556,6 +1571,7 @@ void d_rehash(struct dentry * entry)
 	spin_unlock(&entry->d_lock);
 	spin_unlock(&dcache_lock);
 }
+EXPORT_SYMBOL(d_rehash);
 
 /*
  * When switching names, the actual string doesn't strictly have to
@@ -1702,6 +1718,7 @@ void d_move(struct dentry * dentry, struct dentry * target)
 	d_move_locked(dentry, target);
 	spin_unlock(&dcache_lock);
 }
+EXPORT_SYMBOL(d_move);
 
 /**
  * d_ancestor - search for an ancestor
@@ -1868,6 +1885,7 @@ shouldnt_be_hashed:
 	spin_unlock(&dcache_lock);
 	BUG();
 }
+EXPORT_SYMBOL_GPL(d_materialise_unique);
 
 static int prepend(char **buffer, int *buflen, const char *str, int namelen)
 {
@@ -2005,6 +2023,7 @@ char *d_path(const struct path *path, char *buf, int buflen)
 	path_put(&root);
 	return res;
 }
+EXPORT_SYMBOL(d_path);
 
 /*
  * Helper function for dentry_operations.d_dname() members
@@ -2171,6 +2190,30 @@ int is_subdir(struct dentry *new_dentry, struct dentry *old_dentry)
 	return result;
 }
 
+int path_is_under(struct path *path1, struct path *path2)
+{
+	struct vfsmount *mnt = path1->mnt;
+	struct dentry *dentry = path1->dentry;
+	int res;
+	spin_lock(&vfsmount_lock);
+	if (mnt != path2->mnt) {
+		for (;;) {
+			if (mnt->mnt_parent == mnt) {
+				spin_unlock(&vfsmount_lock);
+				return 0;
+			}
+			if (mnt->mnt_parent == path2->mnt)
+				break;
+			mnt = mnt->mnt_parent;
+		}
+		dentry = mnt->mnt_mountpoint;
+	}
+	res = is_subdir(dentry, path2->dentry);
+	spin_unlock(&vfsmount_lock);
+	return res;
+}
+EXPORT_SYMBOL(path_is_under);
+
 void d_genocide(struct dentry *root)
 {
 	struct dentry *this_parent = root;
@@ -2228,6 +2271,7 @@ ino_t find_inode_number(struct dentry *dir, struct qstr *name)
 	}
 	return ino;
 }
+EXPORT_SYMBOL(find_inode_number);
 
 static __initdata unsigned long dhash_entries;
 static int __init set_dhash_entries(char *str)
@@ -2297,6 +2341,7 @@ static void __init dcache_init(void)
 
 /* SLAB cache for __getname() consumers */
 struct kmem_cache *names_cachep __read_mostly;
+EXPORT_SYMBOL(names_cachep);
 
 EXPORT_SYMBOL(d_genocide);
 
@@ -2326,26 +2371,3 @@ void __init vfs_caches_init(unsigned long mempages)
 	bdev_cache_init();
 	chrdev_init();
 }
-
-EXPORT_SYMBOL(d_alloc);
-EXPORT_SYMBOL(d_alloc_root);
-EXPORT_SYMBOL(d_delete);
-EXPORT_SYMBOL(d_find_alias);
-EXPORT_SYMBOL(d_instantiate);
-EXPORT_SYMBOL(d_invalidate);
-EXPORT_SYMBOL(d_lookup);
-EXPORT_SYMBOL(d_move);
-EXPORT_SYMBOL_GPL(d_materialise_unique);
-EXPORT_SYMBOL(d_path);
-EXPORT_SYMBOL(d_prune_aliases);
-EXPORT_SYMBOL(d_rehash);
-EXPORT_SYMBOL(d_splice_alias);
-EXPORT_SYMBOL(d_add_ci);
-EXPORT_SYMBOL(d_validate);
-EXPORT_SYMBOL(dget_locked);
-EXPORT_SYMBOL(dput);
-EXPORT_SYMBOL(find_inode_number);
-EXPORT_SYMBOL(have_submounts);
-EXPORT_SYMBOL(names_cachep);
-EXPORT_SYMBOL(shrink_dcache_parent);
-EXPORT_SYMBOL(shrink_dcache_sb);

fs/debugfs/inode.c

@@ -496,7 +496,7 @@ struct dentry *debugfs_rename(struct dentry *old_dir, struct dentry *old_dentry,
 	}
 	d_move(old_dentry, dentry);
 	fsnotify_move(old_dir->d_inode, new_dir->d_inode, old_name,
-		old_dentry->d_name.name, S_ISDIR(old_dentry->d_inode->i_mode),
+		S_ISDIR(old_dentry->d_inode->i_mode),
 		NULL, old_dentry);
 	fsnotify_oldname_free(old_name);
 	unlock_rename(new_dir, old_dir);

fs/ext4/file.c

@@ -116,11 +116,9 @@ static int ext4_file_open(struct inode * inode, struct file * filp)
 		 * devices or filesystem images.
 		 */
 		memset(buf, 0, sizeof(buf));
-		path.mnt = mnt->mnt_parent;
+		path.mnt = mnt;
-		path.dentry = mnt->mnt_mountpoint;
+		path.dentry = mnt->mnt_root;
-		path_get(&path);
 		cp = d_path(&path, buf, sizeof(buf));
-		path_put(&path);
 		if (!IS_ERR(cp)) {
 			memcpy(sbi->s_es->s_last_mounted, cp,
 			       sizeof(sbi->s_es->s_last_mounted));

fs/gfs2/ops_inode.c

@@ -975,103 +975,12 @@ out:
 	return error;
 }
 
-/**
- * gfs2_readlinki - return the contents of a symlink
- * @ip: the symlink's inode
- * @buf: a pointer to the buffer to be filled
- * @len: a pointer to the length of @buf
- *
- * If @buf is too small, a piece of memory is kmalloc()ed and needs
- * to be freed by the caller.
- *
- * Returns: errno
- */
-
-static int gfs2_readlinki(struct gfs2_inode *ip, char **buf, unsigned int *len)
-{
-	struct gfs2_holder i_gh;
-	struct buffer_head *dibh;
-	unsigned int x;
-	int error;
-
-	gfs2_holder_init(ip->i_gl, LM_ST_SHARED, 0, &i_gh);
-	error = gfs2_glock_nq(&i_gh);
-	if (error) {
-		gfs2_holder_uninit(&i_gh);
-		return error;
-	}
-
-	if (!ip->i_disksize) {
-		gfs2_consist_inode(ip);
-		error = -EIO;
-		goto out;
-	}
-
-	error = gfs2_meta_inode_buffer(ip, &dibh);
-	if (error)
-		goto out;
-
-	x = ip->i_disksize + 1;
-	if (x > *len) {
-		*buf = kmalloc(x, GFP_NOFS);
-		if (!*buf) {
-			error = -ENOMEM;
-			goto out_brelse;
-		}
-	}
-
-	memcpy(*buf, dibh->b_data + sizeof(struct gfs2_dinode), x);
-	*len = x;
-
-out_brelse:
-	brelse(dibh);
-out:
-	gfs2_glock_dq_uninit(&i_gh);
-	return error;
-}
-
-/**
- * gfs2_readlink - Read the value of a symlink
- * @dentry: the symlink
- * @buf: the buffer to read the symlink data into
- * @size: the size of the buffer
- *
- * Returns: errno
- */
-
-static int gfs2_readlink(struct dentry *dentry, char __user *user_buf,
-			 int user_size)
-{
-	struct gfs2_inode *ip = GFS2_I(dentry->d_inode);
-	char array[GFS2_FAST_NAME_SIZE], *buf = array;
-	unsigned int len = GFS2_FAST_NAME_SIZE;
-	int error;
-
-	error = gfs2_readlinki(ip, &buf, &len);
-	if (error)
-		return error;
-
-	if (user_size > len - 1)
-		user_size = len - 1;
-
-	if (copy_to_user(user_buf, buf, user_size))
-		error = -EFAULT;
-	else
-		error = user_size;
-
-	if (buf != array)
-		kfree(buf);
-
-	return error;
-}
-
 /**
  * gfs2_follow_link - Follow a symbolic link
  * @dentry: The dentry of the link
  * @nd: Data that we pass to vfs_follow_link()
  *
- * This can handle symlinks of any size. It is optimised for symlinks
+ * This can handle symlinks of any size.
- * under GFS2_FAST_NAME_SIZE.
  *
  * Returns: 0 on success or error code
  */
@@ -1079,19 +988,50 @@ static int gfs2_readlink(struct dentry *dentry, char __user *user_buf,
 static void *gfs2_follow_link(struct dentry *dentry, struct nameidata *nd)
 {
 	struct gfs2_inode *ip = GFS2_I(dentry->d_inode);
-	char array[GFS2_FAST_NAME_SIZE], *buf = array;
+	struct gfs2_holder i_gh;
-	unsigned int len = GFS2_FAST_NAME_SIZE;
+	struct buffer_head *dibh;
+	unsigned int x;
+	char *buf;
 	int error;
 
-	error = gfs2_readlinki(ip, &buf, &len);
+	gfs2_holder_init(ip->i_gl, LM_ST_SHARED, 0, &i_gh);
-	if (!error) {
+	error = gfs2_glock_nq(&i_gh);
-		error = vfs_follow_link(nd, buf);
+	if (error) {
-		if (buf != array)
+		gfs2_holder_uninit(&i_gh);
-			kfree(buf);
+		nd_set_link(nd, ERR_PTR(error));
-	} else
+		return NULL;
-		path_put(&nd->path);
+	}
 
-	return ERR_PTR(error);
+	if (!ip->i_disksize) {
+		gfs2_consist_inode(ip);
+		buf = ERR_PTR(-EIO);
+		goto out;
+	}
+
+	error = gfs2_meta_inode_buffer(ip, &dibh);
+	if (error) {
+		buf = ERR_PTR(error);
+		goto out;
+	}
+
+	x = ip->i_disksize + 1;
+	buf = kmalloc(x, GFP_NOFS);
+	if (!buf)
+		buf = ERR_PTR(-ENOMEM);
+	else
+		memcpy(buf, dibh->b_data + sizeof(struct gfs2_dinode), x);
+	brelse(dibh);
+out:
+	gfs2_glock_dq_uninit(&i_gh);
+	nd_set_link(nd, buf);
+	return NULL;
+}
+
+static void gfs2_put_link(struct dentry *dentry, struct nameidata *nd, void *p)
+{
+	char *s = nd_get_link(nd);
+	if (!IS_ERR(s))
+		kfree(s);
 }
 
 /**
@@ -1426,8 +1366,9 @@ const struct inode_operations gfs2_dir_iops = {
 };
 
 const struct inode_operations gfs2_symlink_iops = {
-	.readlink = gfs2_readlink,
+	.readlink = generic_readlink,
 	.follow_link = gfs2_follow_link,
+	.put_link = gfs2_put_link,
 	.permission = gfs2_permission,
 	.setattr = gfs2_setattr,
 	.getattr = gfs2_getattr,

fs/hpfs/anode.c

@@ -353,7 +353,7 @@ int hpfs_ea_read(struct super_block *s, secno a, int ano, unsigned pos,
 }
 
 int hpfs_ea_write(struct super_block *s, secno a, int ano, unsigned pos,
-	     unsigned len, char *buf)
+	     unsigned len, const char *buf)
 {
 	struct buffer_head *bh;
 	char *data;

fs/hpfs/dentry.c

@@ -20,8 +20,8 @@ static int hpfs_hash_dentry(struct dentry *dentry, struct qstr *qstr)
 
 	if (l == 1) if (qstr->name[0]=='.') goto x;
 	if (l == 2) if (qstr->name[0]=='.' || qstr->name[1]=='.') goto x;
-	hpfs_adjust_length((char *)qstr->name, &l);
+	hpfs_adjust_length(qstr->name, &l);
-	/*if (hpfs_chk_name((char *)qstr->name,&l))*/
+	/*if (hpfs_chk_name(qstr->name,&l))*/
 		/*return -ENAMETOOLONG;*/
 		/*return -ENOENT;*/
 	x:
@@ -38,14 +38,16 @@ static int hpfs_compare_dentry(struct dentry *dentry, struct qstr *a, struct qst
 {
 	unsigned al=a->len;
 	unsigned bl=b->len;
-	hpfs_adjust_length((char *)a->name, &al);
+	hpfs_adjust_length(a->name, &al);
-	/*hpfs_adjust_length((char *)b->name, &bl);*/
+	/*hpfs_adjust_length(b->name, &bl);*/
 	/* 'a' is the qstr of an already existing dentry, so the name
 	 * must be valid. 'b' must be validated first.
 	 */
 
-	if (hpfs_chk_name((char *)b->name, &bl)) return 1;
+	if (hpfs_chk_name(b->name, &bl))
-	if (hpfs_compare_names(dentry->d_sb, (char *)a->name, al, (char *)b->name, bl, 0)) return 1;
+		return 1;
+	if (hpfs_compare_names(dentry->d_sb, a->name, al, b->name, bl, 0))
+		return 1;
 	return 0;
 }
 

fs/hpfs/dir.c

@@ -59,7 +59,7 @@ static int hpfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
 	struct hpfs_dirent *de;
 	int lc;
 	long old_pos;
-	char *tempname;
+	unsigned char *tempname;
 	int c1, c2 = 0;
 	int ret = 0;
 
@@ -158,11 +158,11 @@ static int hpfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
 		tempname = hpfs_translate_name(inode->i_sb, de->name, de->namelen, lc, de->not_8x3);
 		if (filldir(dirent, tempname, de->namelen, old_pos, de->fnode, DT_UNKNOWN) < 0) {
 			filp->f_pos = old_pos;
-			if (tempname != (char *)de->name) kfree(tempname);
+			if (tempname != de->name) kfree(tempname);
 			hpfs_brelse4(&qbh);
 			goto out;
 		}
-		if (tempname != (char *)de->name) kfree(tempname);
+		if (tempname != de->name) kfree(tempname);
 		hpfs_brelse4(&qbh);
 	}
 out:
@@ -187,7 +187,7 @@ out:
 
 struct dentry *hpfs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd)
 {
-	const char *name = dentry->d_name.name;
+	const unsigned char *name = dentry->d_name.name;
 	unsigned len = dentry->d_name.len;
 	struct quad_buffer_head qbh;
 	struct hpfs_dirent *de;
@@ -197,7 +197,7 @@ struct dentry *hpfs_lookup(struct inode *dir, struct dentry *dentry, struct name
 	struct hpfs_inode_info *hpfs_result;
 
 	lock_kernel();
-	if ((err = hpfs_chk_name((char *)name, &len))) {
+	if ((err = hpfs_chk_name(name, &len))) {
 		if (err == -ENAMETOOLONG) {
 			unlock_kernel();
 			return ERR_PTR(-ENAMETOOLONG);
@@ -209,7 +209,7 @@ struct dentry *hpfs_lookup(struct inode *dir, struct dentry *dentry, struct name
 	 * '.' and '..' will never be passed here.
 	 */
 
-	de = map_dirent(dir, hpfs_i(dir)->i_dno, (char *) name, len, NULL, &qbh);
+	de = map_dirent(dir, hpfs_i(dir)->i_dno, name, len, NULL, &qbh);
 
 	/*
 	 * This is not really a bailout, just means file not found.
@@ -250,7 +250,7 @@ struct dentry *hpfs_lookup(struct inode *dir, struct dentry *dentry, struct name
 	hpfs_result = hpfs_i(result);
 	if (!de->directory) hpfs_result->i_parent_dir = dir->i_ino;
 
-	hpfs_decide_conv(result, (char *)name, len);
+	hpfs_decide_conv(result, name, len);
 
 	if (de->has_acl || de->has_xtd_perm) if (!(dir->i_sb->s_flags & MS_RDONLY)) {
 		hpfs_error(result->i_sb, "ACLs or XPERM found. This is probably HPFS386. This driver doesn't support it now. Send me some info on these structures");

fs/hpfs/dnode.c

@@ -158,7 +158,8 @@ static void set_last_pointer(struct super_block *s, struct dnode *d, dnode_secno
 
 /* Add an entry to dnode and don't care if it grows over 2048 bytes */
 
-struct hpfs_dirent *hpfs_add_de(struct super_block *s, struct dnode *d, unsigned char *name,
+struct hpfs_dirent *hpfs_add_de(struct super_block *s, struct dnode *d,
+				const unsigned char *name,
 				unsigned namelen, secno down_ptr)
 {
 	struct hpfs_dirent *de;
@@ -223,7 +224,7 @@ static void fix_up_ptrs(struct super_block *s, struct dnode *d)
 /* Add an entry to dnode and do dnode splitting if required */
 
 static int hpfs_add_to_dnode(struct inode *i, dnode_secno dno,
-			     unsigned char *name, unsigned namelen,
+			     const unsigned char *name, unsigned namelen,
 			     struct hpfs_dirent *new_de, dnode_secno down_ptr)
 {
 	struct quad_buffer_head qbh, qbh1, qbh2;
@@ -231,7 +232,7 @@ static int hpfs_add_to_dnode(struct inode *i, dnode_secno dno,
 	dnode_secno adno, rdno;
 	struct hpfs_dirent *de;
 	struct hpfs_dirent nde;
-	char *nname;
+	unsigned char *nname;
 	int h;
 	int pos;
 	struct buffer_head *bh;
@@ -305,7 +306,9 @@ static int hpfs_add_to_dnode(struct inode *i, dnode_secno dno,
 		pos++;
 	}
 	copy_de(new_de = &nde, de);
-	memcpy(name = nname, de->name, namelen = de->namelen);
+	memcpy(nname, de->name, de->namelen);
+	name = nname;
+	namelen = de->namelen;
 	for_all_poss(i, hpfs_pos_subst, ((loff_t)dno << 4) | pos, 4);
 	down_ptr = adno;
 	set_last_pointer(i->i_sb, ad, de->down ? de_down_pointer(de) : 0);
@@ -368,7 +371,8 @@ static int hpfs_add_to_dnode(struct inode *i, dnode_secno dno,
  * I hope, now it's finally bug-free.
  */
 
-int hpfs_add_dirent(struct inode *i, unsigned char *name, unsigned namelen,
+int hpfs_add_dirent(struct inode *i,
+		    const unsigned char *name, unsigned namelen,
 		    struct hpfs_dirent *new_de, int cdepth)
 {
 	struct hpfs_inode_info *hpfs_inode = hpfs_i(i);
@@ -897,7 +901,8 @@ struct hpfs_dirent *map_pos_dirent(struct inode *inode, loff_t *posp,
 
 /* Find a dirent in tree */
 
-struct hpfs_dirent *map_dirent(struct inode *inode, dnode_secno dno, char *name, unsigned len,
+struct hpfs_dirent *map_dirent(struct inode *inode, dnode_secno dno,
+			       const unsigned char *name, unsigned len,
 			       dnode_secno *dd, struct quad_buffer_head *qbh)
 {
 	struct dnode *dnode;
@@ -988,8 +993,8 @@ void hpfs_remove_dtree(struct super_block *s, dnode_secno dno)
 struct hpfs_dirent *map_fnode_dirent(struct super_block *s, fnode_secno fno,
 				     struct fnode *f, struct quad_buffer_head *qbh)
 {
-	char *name1;
+	unsigned char *name1;
-	char *name2;
+	unsigned char *name2;
 	int name1len, name2len;
 	struct dnode *d;
 	dnode_secno dno, downd;

fs/hpfs/ea.c

@@ -62,8 +62,8 @@ static char *get_indirect_ea(struct super_block *s, int ano, secno a, int size)
 	return ret;
 }
 
-static void set_indirect_ea(struct super_block *s, int ano, secno a, char *data,
+static void set_indirect_ea(struct super_block *s, int ano, secno a,
-			    int size)
+			    const char *data, int size)
 {
 	hpfs_ea_write(s, a, ano, 0, size, data);
 }
@@ -186,7 +186,8 @@ char *hpfs_get_ea(struct super_block *s, struct fnode *fnode, char *key, int *si
  * This driver can't change sizes of eas ('cause I just don't need it).
  */
 
-void hpfs_set_ea(struct inode *inode, struct fnode *fnode, char *key, char *data, int size)
+void hpfs_set_ea(struct inode *inode, struct fnode *fnode, const char *key,
+		 const char *data, int size)
 {
 	fnode_secno fno = inode->i_ino;
 	struct super_block *s = inode->i_sb;

fs/hpfs/hpfs_fn.h

@@ -215,7 +215,7 @@ secno hpfs_bplus_lookup(struct super_block *, struct inode *, struct bplus_heade
 secno hpfs_add_sector_to_btree(struct super_block *, secno, int, unsigned);
 void hpfs_remove_btree(struct super_block *, struct bplus_header *);
 int hpfs_ea_read(struct super_block *, secno, int, unsigned, unsigned, char *);
-int hpfs_ea_write(struct super_block *, secno, int, unsigned, unsigned, char *);
+int hpfs_ea_write(struct super_block *, secno, int, unsigned, unsigned, const char *);
 void hpfs_ea_remove(struct super_block *, secno, int, unsigned);
 void hpfs_truncate_btree(struct super_block *, secno, int, unsigned);
 void hpfs_remove_fnode(struct super_block *, fnode_secno fno);
@@ -244,13 +244,17 @@ extern const struct file_operations hpfs_dir_ops;
 
 void hpfs_add_pos(struct inode *, loff_t *);
 void hpfs_del_pos(struct inode *, loff_t *);
-struct hpfs_dirent *hpfs_add_de(struct super_block *, struct dnode *, unsigned char *, unsigned, secno);
+struct hpfs_dirent *hpfs_add_de(struct super_block *, struct dnode *,
-int hpfs_add_dirent(struct inode *, unsigned char *, unsigned, struct hpfs_dirent *, int);
+				const unsigned char *, unsigned, secno);
+int hpfs_add_dirent(struct inode *, const unsigned char *, unsigned,
+		    struct hpfs_dirent *, int);
 int hpfs_remove_dirent(struct inode *, dnode_secno, struct hpfs_dirent *, struct quad_buffer_head *, int);
 void hpfs_count_dnodes(struct super_block *, dnode_secno, int *, int *, int *);
 dnode_secno hpfs_de_as_down_as_possible(struct super_block *, dnode_secno dno);
 struct hpfs_dirent *map_pos_dirent(struct inode *, loff_t *, struct quad_buffer_head *);
-struct hpfs_dirent *map_dirent(struct inode *, dnode_secno, char *, unsigned, dnode_secno *, struct quad_buffer_head *);
+struct hpfs_dirent *map_dirent(struct inode *, dnode_secno,
+			       const unsigned char *, unsigned, dnode_secno *,
+			       struct quad_buffer_head *);
 void hpfs_remove_dtree(struct super_block *, dnode_secno);
 struct hpfs_dirent *map_fnode_dirent(struct super_block *, fnode_secno, struct fnode *, struct quad_buffer_head *);
 
@@ -259,7 +263,8 @@ struct hpfs_dirent *map_fnode_dirent(struct super_block *, fnode_secno, struct f
 void hpfs_ea_ext_remove(struct super_block *, secno, int, unsigned);
 int hpfs_read_ea(struct super_block *, struct fnode *, char *, char *, int);
 char *hpfs_get_ea(struct super_block *, struct fnode *, char *, int *);
-void hpfs_set_ea(struct inode *, struct fnode *, char *, char *, int);
+void hpfs_set_ea(struct inode *, struct fnode *, const char *,
+		 const char *, int);
 
 /* file.c */
 
@@ -282,7 +287,7 @@ void hpfs_delete_inode(struct inode *);
 
 unsigned *hpfs_map_dnode_bitmap(struct super_block *, struct quad_buffer_head *);
 unsigned *hpfs_map_bitmap(struct super_block *, unsigned, struct quad_buffer_head *, char *);
-char *hpfs_load_code_page(struct super_block *, secno);
+unsigned char *hpfs_load_code_page(struct super_block *, secno);
 secno *hpfs_load_bitmap_directory(struct super_block *, secno bmp);
 struct fnode *hpfs_map_fnode(struct super_block *s, ino_t, struct buffer_head **);
 struct anode *hpfs_map_anode(struct super_block *s, anode_secno, struct buffer_head **);
@@ -292,12 +297,13 @@ dnode_secno hpfs_fnode_dno(struct super_block *s, ino_t ino);
 /* name.c */
 
 unsigned char hpfs_upcase(unsigned char *, unsigned char);
-int hpfs_chk_name(unsigned char *, unsigned *);
+int hpfs_chk_name(const unsigned char *, unsigned *);
-char *hpfs_translate_name(struct super_block *, unsigned char *, unsigned, int, int);
+unsigned char *hpfs_translate_name(struct super_block *, unsigned char *, unsigned, int, int);
-int hpfs_compare_names(struct super_block *, unsigned char *, unsigned, unsigned char *, unsigned, int);
+int hpfs_compare_names(struct super_block *, const unsigned char *, unsigned,
-int hpfs_is_name_long(unsigned char *, unsigned);
+		       const unsigned char *, unsigned, int);
-void hpfs_adjust_length(unsigned char *, unsigned *);
+int hpfs_is_name_long(const unsigned char *, unsigned);
-void hpfs_decide_conv(struct inode *, unsigned char *, unsigned);
+void hpfs_adjust_length(const unsigned char *, unsigned *);
+void hpfs_decide_conv(struct inode *, const unsigned char *, unsigned);
 
 /* namei.c */
 

fs/hpfs/inode.c

@@ -46,7 +46,7 @@ void hpfs_read_inode(struct inode *i)
 	struct fnode *fnode;
 	struct super_block *sb = i->i_sb;
 	struct hpfs_inode_info *hpfs_inode = hpfs_i(i);
-	unsigned char *ea;
+	void *ea;
 	int ea_size;
 
 	if (!(fnode = hpfs_map_fnode(sb, i->i_ino, &bh))) {
@@ -112,7 +112,7 @@ void hpfs_read_inode(struct inode *i)
 		}
 	}
 	if (fnode->dirflag) {
-		unsigned n_dnodes, n_subdirs;
+		int n_dnodes, n_subdirs;
 		i->i_mode |= S_IFDIR;
 		i->i_op = &hpfs_dir_iops;
 		i->i_fop = &hpfs_dir_ops;

fs/hpfs/map.c

@@ -35,7 +35,7 @@ unsigned int *hpfs_map_bitmap(struct super_block *s, unsigned bmp_block,
  * lowercasing table
  */
 
-char *hpfs_load_code_page(struct super_block *s, secno cps)
+unsigned char *hpfs_load_code_page(struct super_block *s, secno cps)
 {
 	struct buffer_head *bh;
 	secno cpds;
@@ -71,7 +71,7 @@ char *hpfs_load_code_page(struct super_block *s, secno cps)
 		brelse(bh);
 		return NULL;
 	}
-	ptr = (char *)cpd + cpd->offs[cpi] + 6;
+	ptr = (unsigned char *)cpd + cpd->offs[cpi] + 6;
 	if (!(cp_table = kmalloc(256, GFP_KERNEL))) {
 		printk("HPFS: out of memory for code page table\n");
 		brelse(bh);
@@ -217,7 +217,7 @@ struct dnode *hpfs_map_dnode(struct super_block *s, unsigned secno,
 	if ((dnode = hpfs_map_4sectors(s, secno, qbh, DNODE_RD_AHEAD)))
 		if (hpfs_sb(s)->sb_chk) {
 			unsigned p, pp = 0;
-			unsigned char *d = (char *)dnode;
+			unsigned char *d = (unsigned char *)dnode;
 			int b = 0;
 			if (dnode->magic != DNODE_MAGIC) {
 				hpfs_error(s, "bad magic on dnode %08x", secno);

fs/hpfs/name.c

@@ -8,16 +8,16 @@
 
 #include "hpfs_fn.h"
 
-static char *text_postfix[]={
+static const char *text_postfix[]={
 ".ASM", ".BAS", ".BAT", ".C", ".CC", ".CFG", ".CMD", ".CON", ".CPP", ".DEF",
 ".DOC", ".DPR", ".ERX", ".H", ".HPP", ".HTM", ".HTML", ".JAVA", ".LOG", ".PAS",
 ".RC", ".TEX", ".TXT", ".Y", ""};
 
-static char *text_prefix[]={
+static const char *text_prefix[]={
 "AUTOEXEC.", "CHANGES", "COPYING", "CONFIG.", "CREDITS", "FAQ", "FILE_ID.DIZ",
 "MAKEFILE", "READ.ME", "README", "TERMCAP", ""};
 
-void hpfs_decide_conv(struct inode *inode, unsigned char *name, unsigned len)
+void hpfs_decide_conv(struct inode *inode, const unsigned char *name, unsigned len)
 {
 	struct hpfs_inode_info *hpfs_inode = hpfs_i(inode);
 	int i;
@@ -71,7 +71,7 @@ static inline unsigned char locase(unsigned char *dir, unsigned char a)
 	return dir[a];
 }
 
-int hpfs_chk_name(unsigned char *name, unsigned *len)
+int hpfs_chk_name(const unsigned char *name, unsigned *len)
 {
 	int i;
 	if (*len > 254) return -ENAMETOOLONG;
@@ -83,10 +83,10 @@ int hpfs_chk_name(unsigned char *name, unsigned *len)
 	return 0;
 }
 
-char *hpfs_translate_name(struct super_block *s, unsigned char *from,
+unsigned char *hpfs_translate_name(struct super_block *s, unsigned char *from,
 			  unsigned len, int lc, int lng)
 {
-	char *to;
+	unsigned char *to;
 	int i;
 	if (hpfs_sb(s)->sb_chk >= 2) if (hpfs_is_name_long(from, len) != lng) {
 		printk("HPFS: Long name flag mismatch - name ");
@@ -103,8 +103,9 @@ char *hpfs_translate_name(struct super_block *s, unsigned char *from,
 	return to;
 }
 
-int hpfs_compare_names(struct super_block *s, unsigned char *n1, unsigned l1,
+int hpfs_compare_names(struct super_block *s,
-		       unsigned char *n2, unsigned l2, int last)
+		       const unsigned char *n1, unsigned l1,
+		       const unsigned char *n2, unsigned l2, int last)
 {
 	unsigned l = l1 < l2 ? l1 : l2;
 	unsigned i;
@@ -120,7 +121,7 @@ int hpfs_compare_names(struct super_block *s, unsigned char *n1, unsigned l1,
 	return 0;
 }
 
-int hpfs_is_name_long(unsigned char *name, unsigned len)
+int hpfs_is_name_long(const unsigned char *name, unsigned len)
 {
 	int i,j;
 	for (i = 0; i < len && name[i] != '.'; i++)
@@ -134,7 +135,7 @@ int hpfs_is_name_long(unsigned char *name, unsigned len)
 
 /* OS/2 clears dots and spaces at the end of file name, so we have to */
 
-void hpfs_adjust_length(unsigned char *name, unsigned *len)
+void hpfs_adjust_length(const unsigned char *name, unsigned *len)
 {
 	if (!*len) return;
 	if (*len == 1 && name[0] == '.') return;

fs/hpfs/namei.c

@@ -11,7 +11,7 @@
 
 static int hpfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 {
-	const char *name = dentry->d_name.name;
+	const unsigned char *name = dentry->d_name.name;
 	unsigned len = dentry->d_name.len;
 	struct quad_buffer_head qbh0;
 	struct buffer_head *bh;
@@ -24,7 +24,7 @@ static int hpfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	int r;
 	struct hpfs_dirent dee;
 	int err;
-	if ((err = hpfs_chk_name((char *)name, &len))) return err==-ENOENT ? -EINVAL : err;
+	if ((err = hpfs_chk_name(name, &len))) return err==-ENOENT ? -EINVAL : err;
 	lock_kernel();
 	err = -ENOSPC;
 	fnode = hpfs_alloc_fnode(dir->i_sb, hpfs_i(dir)->i_dno, &fno, &bh);
@@ -62,7 +62,7 @@ static int hpfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 		result->i_mode &= ~0222;
 
 	mutex_lock(&hpfs_i(dir)->i_mutex);
-	r = hpfs_add_dirent(dir, (char *)name, len, &dee, 0);
+	r = hpfs_add_dirent(dir, name, len, &dee, 0);
 	if (r == 1)
 		goto bail3;
 	if (r == -1) {
@@ -121,7 +121,7 @@ bail:
 
 static int hpfs_create(struct inode *dir, struct dentry *dentry, int mode, struct nameidata *nd)
 {
-	const char *name = dentry->d_name.name;
+	const unsigned char *name = dentry->d_name.name;
 	unsigned len = dentry->d_name.len;
 	struct inode *result = NULL;
 	struct buffer_head *bh;
@@ -130,7 +130,7 @@ static int hpfs_create(struct inode *dir, struct dentry *dentry, int mode, struc
 	int r;
 	struct hpfs_dirent dee;
 	int err;
-	if ((err = hpfs_chk_name((char *)name, &len)))
+	if ((err = hpfs_chk_name(name, &len)))
 		return err==-ENOENT ? -EINVAL : err;
 	lock_kernel();
 	err = -ENOSPC;
@@ -155,7 +155,7 @@ static int hpfs_create(struct inode *dir, struct dentry *dentry, int mode, struc
 	result->i_op = &hpfs_file_iops;
 	result->i_fop = &hpfs_file_ops;
 	result->i_nlink = 1;
-	hpfs_decide_conv(result, (char *)name, len);
+	hpfs_decide_conv(result, name, len);
 	hpfs_i(result)->i_parent_dir = dir->i_ino;
 	result->i_ctime.tv_sec = result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, dee.creation_date);
 	result->i_ctime.tv_nsec = 0;
@@ -170,7 +170,7 @@ static int hpfs_create(struct inode *dir, struct dentry *dentry, int mode, struc
 	hpfs_i(result)->mmu_private = 0;
 
 	mutex_lock(&hpfs_i(dir)->i_mutex);
-	r = hpfs_add_dirent(dir, (char *)name, len, &dee, 0);
+	r = hpfs_add_dirent(dir, name, len, &dee, 0);
 	if (r == 1)
 		goto bail2;
 	if (r == -1) {
@@ -211,7 +211,7 @@ bail:
 
 static int hpfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t rdev)
 {
-	const char *name = dentry->d_name.name;
+	const unsigned char *name = dentry->d_name.name;
 	unsigned len = dentry->d_name.len;
 	struct buffer_head *bh;
 	struct fnode *fnode;
@@ -220,7 +220,7 @@ static int hpfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t
 	struct hpfs_dirent dee;
 	struct inode *result = NULL;
 	int err;
-	if ((err = hpfs_chk_name((char *)name, &len))) return err==-ENOENT ? -EINVAL : err;
+	if ((err = hpfs_chk_name(name, &len))) return err==-ENOENT ? -EINVAL : err;
 	if (hpfs_sb(dir->i_sb)->sb_eas < 2) return -EPERM;
 	if (!new_valid_dev(rdev))
 		return -EINVAL;
@@ -256,7 +256,7 @@ static int hpfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t
 	init_special_inode(result, mode, rdev);
 
 	mutex_lock(&hpfs_i(dir)->i_mutex);
-	r = hpfs_add_dirent(dir, (char *)name, len, &dee, 0);
+	r = hpfs_add_dirent(dir, name, len, &dee, 0);
 	if (r == 1)
 		goto bail2;
 	if (r == -1) {
@@ -289,7 +289,7 @@ bail:
 
 static int hpfs_symlink(struct inode *dir, struct dentry *dentry, const char *symlink)
 {
-	const char *name = dentry->d_name.name;
+	const unsigned char *name = dentry->d_name.name;
 	unsigned len = dentry->d_name.len;
 	struct buffer_head *bh;
 	struct fnode *fnode;
@@ -298,7 +298,7 @@ static int hpfs_symlink(struct inode *dir, struct dentry *dentry, const char *sy
 	struct hpfs_dirent dee;
 	struct inode *result;
 	int err;
-	if ((err = hpfs_chk_name((char *)name, &len))) return err==-ENOENT ? -EINVAL : err;
+	if ((err = hpfs_chk_name(name, &len))) return err==-ENOENT ? -EINVAL : err;
 	lock_kernel();
 	if (hpfs_sb(dir->i_sb)->sb_eas < 2) {
 		unlock_kernel();
@@ -335,7 +335,7 @@ static int hpfs_symlink(struct inode *dir, struct dentry *dentry, const char *sy
 	result->i_data.a_ops = &hpfs_symlink_aops;
 
 	mutex_lock(&hpfs_i(dir)->i_mutex);
-	r = hpfs_add_dirent(dir, (char *)name, len, &dee, 0);
+	r = hpfs_add_dirent(dir, name, len, &dee, 0);
 	if (r == 1)
 		goto bail2;
 	if (r == -1) {
@@ -345,7 +345,7 @@ static int hpfs_symlink(struct inode *dir, struct dentry *dentry, const char *sy
 	fnode->len = len;
 	memcpy(fnode->name, name, len > 15 ? 15 : len);
 	fnode->up = dir->i_ino;
-	hpfs_set_ea(result, fnode, "SYMLINK", (char *)symlink, strlen(symlink));
+	hpfs_set_ea(result, fnode, "SYMLINK", symlink, strlen(symlink));
 	mark_buffer_dirty(bh);
 	brelse(bh);
 
@@ -369,7 +369,7 @@ bail:
 
 static int hpfs_unlink(struct inode *dir, struct dentry *dentry)
 {
-	const char *name = dentry->d_name.name;
+	const unsigned char *name = dentry->d_name.name;
 	unsigned len = dentry->d_name.len;
 	struct quad_buffer_head qbh;
 	struct hpfs_dirent *de;
@@ -381,12 +381,12 @@ static int hpfs_unlink(struct inode *dir, struct dentry *dentry)
 	int err;
 
 	lock_kernel();
-	hpfs_adjust_length((char *)name, &len);
+	hpfs_adjust_length(name, &len);
 again:
 	mutex_lock(&hpfs_i(inode)->i_parent_mutex);
 	mutex_lock(&hpfs_i(dir)->i_mutex);
 	err = -ENOENT;
-	de = map_dirent(dir, hpfs_i(dir)->i_dno, (char *)name, len, &dno, &qbh);
+	de = map_dirent(dir, hpfs_i(dir)->i_dno, name, len, &dno, &qbh);
 	if (!de)
 		goto out;
 
@@ -413,22 +413,25 @@ again:
 
 		mutex_unlock(&hpfs_i(dir)->i_mutex);
 		mutex_unlock(&hpfs_i(inode)->i_parent_mutex);
-		d_drop(dentry);
+		dentry_unhash(dentry);
-		spin_lock(&dentry->d_lock);
+		if (!d_unhashed(dentry)) {
-		if (atomic_read(&dentry->d_count) > 1 ||
+			dput(dentry);
-		    generic_permission(inode, MAY_WRITE, NULL) ||
+			unlock_kernel();
+			return -ENOSPC;
+		}
+		if (generic_permission(inode, MAY_WRITE, NULL) ||
 		    !S_ISREG(inode->i_mode) ||
 		    get_write_access(inode)) {
-			spin_unlock(&dentry->d_lock);
 			d_rehash(dentry);
+			dput(dentry);
 		} else {
 			struct iattr newattrs;
-			spin_unlock(&dentry->d_lock);
 			/*printk("HPFS: truncating file before delete.\n");*/
 			newattrs.ia_size = 0;
 			newattrs.ia_valid = ATTR_SIZE | ATTR_CTIME;
 			err = notify_change(dentry, &newattrs);
 			put_write_access(inode);
+			dput(dentry);
 			if (!err)
 				goto again;
 		}
@@ -451,7 +454,7 @@ out:
 
 static int hpfs_rmdir(struct inode *dir, struct dentry *dentry)
 {
-	const char *name = dentry->d_name.name;
+	const unsigned char *name = dentry->d_name.name;
 	unsigned len = dentry->d_name.len;
 	struct quad_buffer_head qbh;
 	struct hpfs_dirent *de;
@@ -462,12 +465,12 @@ static int hpfs_rmdir(struct inode *dir, struct dentry *dentry)
 	int err;
 	int r;
 
-	hpfs_adjust_length((char *)name, &len);
+	hpfs_adjust_length(name, &len);
 	lock_kernel();
 	mutex_lock(&hpfs_i(inode)->i_parent_mutex);
 	mutex_lock(&hpfs_i(dir)->i_mutex);
 	err = -ENOENT;
-	de = map_dirent(dir, hpfs_i(dir)->i_dno, (char *)name, len, &dno, &qbh);
+	de = map_dirent(dir, hpfs_i(dir)->i_dno, name, len, &dno, &qbh);
 	if (!de)
 		goto out;
 
@@ -546,10 +549,10 @@ const struct address_space_operations hpfs_symlink_aops = {
 static int hpfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		struct inode *new_dir, struct dentry *new_dentry)
 {
-	char *old_name = (char *)old_dentry->d_name.name;
+	const unsigned char *old_name = old_dentry->d_name.name;
-	int old_len = old_dentry->d_name.len;
+	unsigned old_len = old_dentry->d_name.len;
-	char *new_name = (char *)new_dentry->d_name.name;
+	const unsigned char *new_name = new_dentry->d_name.name;
-	int new_len = new_dentry->d_name.len;
+	unsigned new_len = new_dentry->d_name.len;
 	struct inode *i = old_dentry->d_inode;
 	struct inode *new_inode = new_dentry->d_inode;
 	struct quad_buffer_head qbh, qbh1;
@@ -560,9 +563,9 @@ static int hpfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	struct buffer_head *bh;
 	struct fnode *fnode;
 	int err;
-	if ((err = hpfs_chk_name((char *)new_name, &new_len))) return err;
+	if ((err = hpfs_chk_name(new_name, &new_len))) return err;
 	err = 0;
-	hpfs_adjust_length((char *)old_name, &old_len);
+	hpfs_adjust_length(old_name, &old_len);
 
 	lock_kernel();
 	/* order doesn't matter, due to VFS exclusion */
@@ -579,7 +582,7 @@ static int hpfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		goto end1;
 	}
 
-	if (!(dep = map_dirent(old_dir, hpfs_i(old_dir)->i_dno, (char *)old_name, old_len, &dno, &qbh))) {
+	if (!(dep = map_dirent(old_dir, hpfs_i(old_dir)->i_dno, old_name, old_len, &dno, &qbh))) {
 		hpfs_error(i->i_sb, "lookup succeeded but map dirent failed");
 		err = -ENOENT;
 		goto end1;
@@ -590,7 +593,7 @@ static int hpfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	if (new_inode) {
 		int r;
 		if ((r = hpfs_remove_dirent(old_dir, dno, dep, &qbh, 1)) != 2) {
-			if ((nde = map_dirent(new_dir, hpfs_i(new_dir)->i_dno, (char *)new_name, new_len, NULL, &qbh1))) {
+			if ((nde = map_dirent(new_dir, hpfs_i(new_dir)->i_dno, new_name, new_len, NULL, &qbh1))) {
 				clear_nlink(new_inode);
 				copy_de(nde, &de);
 				memcpy(nde->name, new_name, new_len);
@@ -618,7 +621,7 @@ static int hpfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	}
 	
 	if (new_dir == old_dir)
-		if (!(dep = map_dirent(old_dir, hpfs_i(old_dir)->i_dno, (char *)old_name, old_len, &dno, &qbh))) {
+		if (!(dep = map_dirent(old_dir, hpfs_i(old_dir)->i_dno, old_name, old_len, &dno, &qbh))) {
 			hpfs_unlock_creation(i->i_sb);
 			hpfs_error(i->i_sb, "lookup succeeded but map dirent failed at #2");
 			err = -ENOENT;
@@ -648,7 +651,7 @@ static int hpfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		brelse(bh);
 	}
 	hpfs_i(i)->i_conv = hpfs_sb(i->i_sb)->sb_conv;
-	hpfs_decide_conv(i, (char *)new_name, new_len);
+	hpfs_decide_conv(i, new_name, new_len);
 end1:
 	if (old_dir != new_dir)
 		mutex_unlock(&hpfs_i(new_dir)->i_mutex);

fs/hppfs/hppfs.c

@@ -718,7 +718,7 @@ static int hppfs_fill_super(struct super_block *sb, void *d, int silent)
 	struct vfsmount *proc_mnt;
 	int err = -ENOENT;
 
-	proc_mnt = do_kern_mount("proc", 0, "proc", NULL);
+	proc_mnt = mntget(current->nsproxy->pid_ns->proc_mnt);
 	if (IS_ERR(proc_mnt))
 		goto out;
 

fs/internal.h

@@ -70,6 +70,8 @@ extern struct vfsmount *copy_tree(struct vfsmount *, struct dentry *, int);
 
 extern void __init mnt_init(void);
 
+extern spinlock_t vfsmount_lock;
+
 /*
  * fs_struct.c
  */

fs/libfs.c

@@ -338,28 +338,14 @@ int simple_readpage(struct file *file, struct page *page)
 	return 0;
 }
 
-int simple_prepare_write(struct file *file, struct page *page,
-			unsigned from, unsigned to)
-{
-	if (!PageUptodate(page)) {
-		if (to - from != PAGE_CACHE_SIZE)
-			zero_user_segments(page,
-				0, from,
-				to, PAGE_CACHE_SIZE);
-	}
-	return 0;
-}
-
 int simple_write_begin(struct file *file, struct address_space *mapping,
 			loff_t pos, unsigned len, unsigned flags,
 			struct page **pagep, void **fsdata)
 {
 	struct page *page;
 	pgoff_t index;
-	unsigned from;
 
 	index = pos >> PAGE_CACHE_SHIFT;
-	from = pos & (PAGE_CACHE_SIZE - 1);
 
 	page = grab_cache_page_write_begin(mapping, index, flags);
 	if (!page)
@@ -367,14 +353,48 @@ int simple_write_begin(struct file *file, struct address_space *mapping,
 
 	*pagep = page;
 
-	return simple_prepare_write(file, page, from, from+len);
+	if (!PageUptodate(page) && (len != PAGE_CACHE_SIZE)) {
+		unsigned from = pos & (PAGE_CACHE_SIZE - 1);
+
+		zero_user_segments(page, 0, from, from + len, PAGE_CACHE_SIZE);
+	}
+	return 0;
 }
 
-static int simple_commit_write(struct file *file, struct page *page,
+/**
-			       unsigned from, unsigned to)
+ * simple_write_end - .write_end helper for non-block-device FSes
+ * @available: See .write_end of address_space_operations
+ * @file: 		"
+ * @mapping: 		"
+ * @pos: 		"
+ * @len: 		"
+ * @copied: 		"
+ * @page: 		"
+ * @fsdata: 		"
+ *
+ * simple_write_end does the minimum needed for updating a page after writing is
+ * done. It has the same API signature as the .write_end of
+ * address_space_operations vector. So it can just be set onto .write_end for
+ * FSes that don't need any other processing. i_mutex is assumed to be held.
+ * Block based filesystems should use generic_write_end().
+ * NOTE: Even though i_size might get updated by this function, mark_inode_dirty
+ * is not called, so a filesystem that actually does store data in .write_inode
+ * should extend on what's done here with a call to mark_inode_dirty() in the
+ * case that i_size has changed.
+ */
+int simple_write_end(struct file *file, struct address_space *mapping,
+			loff_t pos, unsigned len, unsigned copied,
+			struct page *page, void *fsdata)
 {
 	struct inode *inode = page->mapping->host;
-	loff_t pos = ((loff_t)page->index << PAGE_CACHE_SHIFT) + to;
+	loff_t last_pos = pos + copied;
+
+	/* zero the stale part of the page if we did a short copy */
+	if (copied < len) {
+		unsigned from = pos & (PAGE_CACHE_SIZE - 1);
+
+		zero_user(page, from + copied, len - copied);
+	}
 
 	if (!PageUptodate(page))
 		SetPageUptodate(page);
@@ -382,28 +402,10 @@ static int simple_commit_write(struct file *file, struct page *page,
 	 * No need to use i_size_read() here, the i_size
 	 * cannot change under us because we hold the i_mutex.
 	 */
-	if (pos > inode->i_size)
+	if (last_pos > inode->i_size)
-		i_size_write(inode, pos);
+		i_size_write(inode, last_pos);
+
 	set_page_dirty(page);
-	return 0;
-}
-
-int simple_write_end(struct file *file, struct address_space *mapping,
-			loff_t pos, unsigned len, unsigned copied,
-			struct page *page, void *fsdata)
-{
-	unsigned from = pos & (PAGE_CACHE_SIZE - 1);
-
-	/* zero the stale part of the page if we did a short copy */
-	if (copied < len) {
-		void *kaddr = kmap_atomic(page, KM_USER0);
-		memset(kaddr + from + copied, 0, len - copied);
-		flush_dcache_page(page);
-		kunmap_atomic(kaddr, KM_USER0);
-	}
-
-	simple_commit_write(file, page, from, from+copied);
-
 	unlock_page(page);
 	page_cache_release(page);
 
@@ -853,7 +855,6 @@ EXPORT_SYMBOL(simple_getattr);
 EXPORT_SYMBOL(simple_link);
 EXPORT_SYMBOL(simple_lookup);
 EXPORT_SYMBOL(simple_pin_fs);
-EXPORT_UNUSED_SYMBOL(simple_prepare_write);
 EXPORT_SYMBOL(simple_readpage);
 EXPORT_SYMBOL(simple_release_fs);
 EXPORT_SYMBOL(simple_rename);

fs/locks.c

@@ -1182,8 +1182,9 @@ int __break_lease(struct inode *inode, unsigned int mode)
 	struct file_lock *fl;
 	unsigned long break_time;
 	int i_have_this_lease = 0;
+	int want_write = (mode & O_ACCMODE) != O_RDONLY;
 
-	new_fl = lease_alloc(NULL, mode & FMODE_WRITE ? F_WRLCK : F_RDLCK);
+	new_fl = lease_alloc(NULL, want_write ? F_WRLCK : F_RDLCK);
 
 	lock_kernel();
 
@@ -1197,7 +1198,7 @@ int __break_lease(struct inode *inode, unsigned int mode)
 		if (fl->fl_owner == current->files)
 			i_have_this_lease = 1;
 
-	if (mode & FMODE_WRITE) {
+	if (want_write) {
 		/* If we want write access, we have to revoke any lease. */
 		future = F_UNLCK | F_INPROGRESS;
 	} else if (flock->fl_type & F_INPROGRESS) {

fs/namei.c

@@ -689,33 +689,20 @@ static __always_inline void follow_dotdot(struct nameidata *nd)
 	set_root(nd);
 
 	while(1) {
-		struct vfsmount *parent;
 		struct dentry *old = nd->path.dentry;
 
 		if (nd->path.dentry == nd->root.dentry &&
 		    nd->path.mnt == nd->root.mnt) {
 			break;
 		}
-		spin_lock(&dcache_lock);
 		if (nd->path.dentry != nd->path.mnt->mnt_root) {
-			nd->path.dentry = dget(nd->path.dentry->d_parent);
+			/* rare case of legitimate dget_parent()... */
-			spin_unlock(&dcache_lock);
+			nd->path.dentry = dget_parent(nd->path.dentry);
 			dput(old);
 			break;
 		}
-		spin_unlock(&dcache_lock);
+		if (!follow_up(&nd->path))
-		spin_lock(&vfsmount_lock);
-		parent = nd->path.mnt->mnt_parent;
-		if (parent == nd->path.mnt) {
-			spin_unlock(&vfsmount_lock);
 			break;
-		}
-		mntget(parent);
-		nd->path.dentry = dget(nd->path.mnt->mnt_mountpoint);
-		spin_unlock(&vfsmount_lock);
-		dput(old);
-		mntput(nd->path.mnt);
-		nd->path.mnt = parent;
 	}
 	follow_mount(&nd->path);
 }
@@ -1347,7 +1334,7 @@ static int may_delete(struct inode *dir,struct dentry *victim,int isdir)
 		return -ENOENT;
 
 	BUG_ON(victim->d_parent->d_inode != dir);
-	audit_inode_child(victim->d_name.name, victim, dir);
+	audit_inode_child(victim, dir);
 
 	error = inode_permission(dir, MAY_WRITE | MAY_EXEC);
 	if (error)
@@ -1503,7 +1490,7 @@ int may_open(struct path *path, int acc_mode, int flag)
 	 * An append-only file must be opened in append mode for writing.
 	 */
 	if (IS_APPEND(inode)) {
-		if  ((flag & FMODE_WRITE) && !(flag & O_APPEND))
+		if  ((flag & O_ACCMODE) != O_RDONLY && !(flag & O_APPEND))
 			return -EPERM;
 		if (flag & O_TRUNC)
 			return -EPERM;
@@ -1547,7 +1534,7 @@ static int handle_truncate(struct path *path)
  * what get passed to sys_open().
  */
 static int __open_namei_create(struct nameidata *nd, struct path *path,
-				int flag, int mode)
+				int open_flag, int mode)
 {
 	int error;
 	struct dentry *dir = nd->path.dentry;
@@ -1565,7 +1552,7 @@ out_unlock:
 	if (error)
 		return error;
 	/* Don't check for write permission, don't truncate */
-	return may_open(&nd->path, 0, flag & ~O_TRUNC);
+	return may_open(&nd->path, 0, open_flag & ~O_TRUNC);
 }
 
 /*
@@ -1736,7 +1723,7 @@ do_last:
 		error = mnt_want_write(nd.path.mnt);
 		if (error)
 			goto exit_mutex_unlock;
-		error = __open_namei_create(&nd, &path, flag, mode);
+		error = __open_namei_create(&nd, &path, open_flag, mode);
 		if (error) {
 			mnt_drop_write(nd.path.mnt);
 			goto exit;
@@ -1798,7 +1785,7 @@ ok:
 		if (error)
 			goto exit;
 	}
-	error = may_open(&nd.path, acc_mode, flag);
+	error = may_open(&nd.path, acc_mode, open_flag);
 	if (error) {
 		if (will_truncate)
 			mnt_drop_write(nd.path.mnt);
@@ -2275,8 +2262,11 @@ int vfs_unlink(struct inode *dir, struct dentry *dentry)
 		error = -EBUSY;
 	else {
 		error = security_inode_unlink(dir, dentry);
-		if (!error)
+		if (!error) {
 			error = dir->i_op->unlink(dir, dentry);
+			if (!error)
+				dentry->d_inode->i_flags |= S_DEAD;
+		}
 	}
 	mutex_unlock(&dentry->d_inode->i_mutex);
 
@@ -2629,6 +2619,8 @@ static int vfs_rename_other(struct inode *old_dir, struct dentry *old_dentry,
 	else
 		error = old_dir->i_op->rename(old_dir, old_dentry, new_dir, new_dentry);
 	if (!error) {
+		if (target)
+			target->i_flags |= S_DEAD;
 		if (!(old_dir->i_sb->s_type->fs_flags & FS_RENAME_DOES_D_MOVE))
 			d_move(old_dentry, new_dentry);
 	}
@@ -2671,11 +2663,9 @@ int vfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		error = vfs_rename_dir(old_dir,old_dentry,new_dir,new_dentry);
 	else
 		error = vfs_rename_other(old_dir,old_dentry,new_dir,new_dentry);
-	if (!error) {
+	if (!error)
-		const char *new_name = old_dentry->d_name.name;
+		fsnotify_move(old_dir, new_dir, old_name, is_dir,
-		fsnotify_move(old_dir, new_dir, old_name, new_name, is_dir,
 			      new_dentry->d_inode, old_dentry);
-	}
 	fsnotify_oldname_free(old_name);
 
 	return error;

fs/namespace.c

@@ -573,7 +573,7 @@ static struct vfsmount *clone_mnt(struct vfsmount *old, struct dentry *root,
 			mnt->mnt_master = old;
 			CLEAR_MNT_SHARED(mnt);
 		} else if (!(flag & CL_PRIVATE)) {
-			if ((flag & CL_PROPAGATION) || IS_MNT_SHARED(old))
+			if ((flag & CL_MAKE_SHARED) || IS_MNT_SHARED(old))
 				list_add(&mnt->mnt_share, &old->mnt_share);
 			if (IS_MNT_SLAVE(old))
 				list_add(&mnt->mnt_slave, &old->mnt_slave);
@@ -737,6 +737,21 @@ static void m_stop(struct seq_file *m, void *v)
 	up_read(&namespace_sem);
 }
 
+int mnt_had_events(struct proc_mounts *p)
+{
+	struct mnt_namespace *ns = p->ns;
+	int res = 0;
+
+	spin_lock(&vfsmount_lock);
+	if (p->event != ns->event) {
+		p->event = ns->event;
+		res = 1;
+	}
+	spin_unlock(&vfsmount_lock);
+
+	return res;
+}
+
 struct proc_fs_info {
 	int flag;
 	const char *str;
@@ -1121,8 +1136,15 @@ SYSCALL_DEFINE2(umount, char __user *, name, int, flags)
 {
 	struct path path;
 	int retval;
+	int lookup_flags = 0;
 
-	retval = user_path(name, &path);
+	if (flags & ~(MNT_FORCE | MNT_DETACH | MNT_EXPIRE | UMOUNT_NOFOLLOW))
+		return -EINVAL;
+
+	if (!(flags & UMOUNT_NOFOLLOW))
+		lookup_flags |= LOOKUP_FOLLOW;
+
+	retval = user_path_at(AT_FDCWD, name, lookup_flags, &path);
 	if (retval)
 		goto out;
 	retval = -EINVAL;
@@ -1246,6 +1268,21 @@ void drop_collected_mounts(struct vfsmount *mnt)
 	release_mounts(&umount_list);
 }
 
+int iterate_mounts(int (*f)(struct vfsmount *, void *), void *arg,
+		   struct vfsmount *root)
+{
+	struct vfsmount *mnt;
+	int res = f(root, arg);
+	if (res)
+		return res;
+	list_for_each_entry(mnt, &root->mnt_list, mnt_list) {
+		res = f(mnt, arg);
+		if (res)
+			return res;
+	}
+	return 0;
+}
+
 static void cleanup_group_ids(struct vfsmount *mnt, struct vfsmount *end)
 {
 	struct vfsmount *p;
@@ -1538,7 +1575,7 @@ static int do_remount(struct path *path, int flags, int mnt_flags,
 		err = do_remount_sb(sb, flags, data, 0);
 	if (!err) {
 		spin_lock(&vfsmount_lock);
-		mnt_flags |= path->mnt->mnt_flags & MNT_PNODE_MASK;
+		mnt_flags |= path->mnt->mnt_flags & MNT_PROPAGATION_MASK;
 		path->mnt->mnt_flags = mnt_flags;
 		spin_unlock(&vfsmount_lock);
 	}
@@ -1671,7 +1708,7 @@ int do_add_mount(struct vfsmount *newmnt, struct path *path,
 {
 	int err;
 
-	mnt_flags &= ~(MNT_SHARED | MNT_WRITE_HOLD);
+	mnt_flags &= ~(MNT_SHARED | MNT_WRITE_HOLD | MNT_INTERNAL);
 
 	down_write(&namespace_sem);
 	/* Something was mounted here while we slept */
@@ -2314,17 +2351,13 @@ void __init mnt_init(void)
 
 void put_mnt_ns(struct mnt_namespace *ns)
 {
-	struct vfsmount *root;
 	LIST_HEAD(umount_list);
 
-	if (!atomic_dec_and_lock(&ns->count, &vfsmount_lock))
+	if (!atomic_dec_and_test(&ns->count))
 		return;
-	root = ns->root;
-	ns->root = NULL;
-	spin_unlock(&vfsmount_lock);
 	down_write(&namespace_sem);
 	spin_lock(&vfsmount_lock);
-	umount_tree(root, 0, &umount_list);
+	umount_tree(ns->root, 0, &umount_list);
 	spin_unlock(&vfsmount_lock);
 	up_write(&namespace_sem);
 	release_mounts(&umount_list);

fs/nfs/inode.c

@@ -574,14 +574,14 @@ void nfs_close_context(struct nfs_open_context *ctx, int is_sync)
 	nfs_revalidate_inode(server, inode);
 }
 
-static struct nfs_open_context *alloc_nfs_open_context(struct vfsmount *mnt, struct dentry *dentry, struct rpc_cred *cred)
+static struct nfs_open_context *alloc_nfs_open_context(struct path *path, struct rpc_cred *cred)
 {
 	struct nfs_open_context *ctx;
 
 	ctx = kmalloc(sizeof(*ctx), GFP_KERNEL);
 	if (ctx != NULL) {
-		ctx->path.dentry = dget(dentry);
+		ctx->path = *path;
-		ctx->path.mnt = mntget(mnt);
+		path_get(&ctx->path);
 		ctx->cred = get_rpccred(cred);
 		ctx->state = NULL;
 		ctx->lockowner = current->files;
@@ -686,7 +686,7 @@ int nfs_open(struct inode *inode, struct file *filp)
 	cred = rpc_lookup_cred();
 	if (IS_ERR(cred))
 		return PTR_ERR(cred);
-	ctx = alloc_nfs_open_context(filp->f_path.mnt, filp->f_path.dentry, cred);
+	ctx = alloc_nfs_open_context(&filp->f_path, cred);
 	put_rpccred(cred);
 	if (ctx == NULL)
 		return -ENOMEM;

fs/nfs/nfs4proc.c

@@ -724,8 +724,8 @@ static struct nfs4_opendata *nfs4_opendata_alloc(struct path *path,
 	p->o_arg.seqid = nfs_alloc_seqid(&sp->so_seqid);
 	if (p->o_arg.seqid == NULL)
 		goto err_free;
-	p->path.mnt = mntget(path->mnt);
+	path_get(path);
-	p->path.dentry = dget(path->dentry);
+	p->path = *path;
 	p->dir = parent;
 	p->owner = sp;
 	atomic_inc(&sp->so_count);
@@ -1944,8 +1944,8 @@ int nfs4_do_close(struct path *path, struct nfs4_state *state, int wait)
 	calldata->res.seqid = calldata->arg.seqid;
 	calldata->res.server = server;
 	calldata->res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE;
-	calldata->path.mnt = mntget(path->mnt);
+	path_get(path);
-	calldata->path.dentry = dget(path->dentry);
+	calldata->path = *path;
 
 	msg.rpc_argp = &calldata->arg,
 	msg.rpc_resp = &calldata->res,

fs/nfsctl.c

@@ -36,10 +36,9 @@ static struct file *do_open(char *name, int flags)
 		return ERR_PTR(error);
 
 	if (flags == O_RDWR)
-		error = may_open(&nd.path, MAY_READ|MAY_WRITE,
+		error = may_open(&nd.path, MAY_READ|MAY_WRITE, flags);
-					   FMODE_READ|FMODE_WRITE);
 	else
-		error = may_open(&nd.path, MAY_WRITE, FMODE_WRITE);
+		error = may_open(&nd.path, MAY_WRITE, flags);
 
 	if (!error)
 		return dentry_open(nd.path.dentry, nd.path.mnt, flags,

fs/nfsd/nfs4xdr.c

@@ -2121,9 +2121,15 @@ out_acl:
 		 * and this is the root of a cross-mounted filesystem.
 		 */
 		if (ignore_crossmnt == 0 &&
-		    exp->ex_path.mnt->mnt_root->d_inode == dentry->d_inode) {
+		    dentry == exp->ex_path.mnt->mnt_root) {
-			err = vfs_getattr(exp->ex_path.mnt->mnt_parent,
+			struct path path = exp->ex_path;
-				exp->ex_path.mnt->mnt_mountpoint, &stat);
+			path_get(&path);
+			while (follow_up(&path)) {
+				if (path.dentry != path.mnt->mnt_root)
+					break;
+			}
+			err = vfs_getattr(path.mnt, path.dentry, &stat);
+			path_put(&path);
 			if (err)
 				goto out_nfserr;
 		}

fs/nfsd/vfs.c

@@ -361,7 +361,7 @@ nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap,
 		 * If we are changing the size of the file, then
 		 * we need to break all leases.
 		 */
-		host_err = break_lease(inode, FMODE_WRITE | O_NONBLOCK);
+		host_err = break_lease(inode, O_WRONLY | O_NONBLOCK);
 		if (host_err == -EWOULDBLOCK)
 			host_err = -ETIMEDOUT;
 		if (host_err) /* ENOMEM or EWOULDBLOCK */
@@ -734,7 +734,7 @@ nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, int type,
 	 * Check to see if there are any leases on this file.
 	 * This may block while leases are broken.
 	 */
-	host_err = break_lease(inode, O_NONBLOCK | ((access & NFSD_MAY_WRITE) ? FMODE_WRITE : 0));
+	host_err = break_lease(inode, O_NONBLOCK | ((access & NFSD_MAY_WRITE) ? O_WRONLY : 0));
 	if (host_err == -EWOULDBLOCK)
 		host_err = -ETIMEDOUT;
 	if (host_err) /* NOMEM or WOULDBLOCK */

fs/nilfs2/dir.c

@@ -224,7 +224,7 @@ fail:
  * len <= NILFS_NAME_LEN and de != NULL are guaranteed by caller.
  */
 static int
-nilfs_match(int len, const char * const name, struct nilfs_dir_entry *de)
+nilfs_match(int len, const unsigned char *name, struct nilfs_dir_entry *de)
 {
 	if (len != de->name_len)
 		return 0;
@@ -349,11 +349,11 @@ done:
  * Entry is guaranteed to be valid.
  */
 struct nilfs_dir_entry *
-nilfs_find_entry(struct inode *dir, struct dentry *dentry,
+nilfs_find_entry(struct inode *dir, const struct qstr *qstr,
 		 struct page **res_page)
 {
-	const char *name = dentry->d_name.name;
+	const unsigned char *name = qstr->name;
-	int namelen = dentry->d_name.len;
+	int namelen = qstr->len;
 	unsigned reclen = NILFS_DIR_REC_LEN(namelen);
 	unsigned long start, n;
 	unsigned long npages = dir_pages(dir);
@@ -424,13 +424,13 @@ struct nilfs_dir_entry *nilfs_dotdot(struct inode *dir, struct page **p)
 	return de;
 }
 
-ino_t nilfs_inode_by_name(struct inode *dir, struct dentry *dentry)
+ino_t nilfs_inode_by_name(struct inode *dir, const struct qstr *qstr)
 {
 	ino_t res = 0;
 	struct nilfs_dir_entry *de;
 	struct page *page;
 
-	de = nilfs_find_entry(dir, dentry, &page);
+	de = nilfs_find_entry(dir, qstr, &page);
 	if (de) {
 		res = le64_to_cpu(de->inode);
 		kunmap(page);
@@ -465,7 +465,7 @@ void nilfs_set_link(struct inode *dir, struct nilfs_dir_entry *de,
 int nilfs_add_link(struct dentry *dentry, struct inode *inode)
 {
 	struct inode *dir = dentry->d_parent->d_inode;
-	const char *name = dentry->d_name.name;
+	const unsigned char *name = dentry->d_name.name;
 	int namelen = dentry->d_name.len;
 	unsigned chunk_size = nilfs_chunk_size(dir);
 	unsigned reclen = NILFS_DIR_REC_LEN(namelen);

fs/nilfs2/namei.c

@@ -67,7 +67,7 @@ nilfs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd)
 	if (dentry->d_name.len > NILFS_NAME_LEN)
 		return ERR_PTR(-ENAMETOOLONG);
 
-	ino = nilfs_inode_by_name(dir, dentry);
+	ino = nilfs_inode_by_name(dir, &dentry->d_name);
 	inode = NULL;
 	if (ino) {
 		inode = nilfs_iget(dir->i_sb, ino);
@@ -81,10 +81,7 @@ struct dentry *nilfs_get_parent(struct dentry *child)
 {
 	unsigned long ino;
 	struct inode *inode;
-	struct dentry dotdot;
+	struct qstr dotdot = {.name = "..", .len = 2};
-
-	dotdot.d_name.name = "..";
-	dotdot.d_name.len = 2;
 
 	ino = nilfs_inode_by_name(child->d_inode, &dotdot);
 	if (!ino)
@@ -296,7 +293,7 @@ static int nilfs_do_unlink(struct inode *dir, struct dentry *dentry)
 	int err;
 
 	err = -ENOENT;
-	de = nilfs_find_entry(dir, dentry, &page);
+	de = nilfs_find_entry(dir, &dentry->d_name, &page);
 	if (!de)
 		goto out;
 
@@ -389,7 +386,7 @@ static int nilfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		return err;
 
 	err = -ENOENT;
-	old_de = nilfs_find_entry(old_dir, old_dentry, &old_page);
+	old_de = nilfs_find_entry(old_dir, &old_dentry->d_name, &old_page);
 	if (!old_de)
 		goto out;
 
@@ -409,7 +406,7 @@ static int nilfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 			goto out_dir;
 
 		err = -ENOENT;
-		new_de = nilfs_find_entry(new_dir, new_dentry, &new_page);
+		new_de = nilfs_find_entry(new_dir, &new_dentry->d_name, &new_page);
 		if (!new_de)
 			goto out_dir;
 		inc_nlink(old_inode);

fs/nilfs2/nilfs.h

@@ -217,10 +217,10 @@ static inline int nilfs_init_acl(struct inode *inode, struct inode *dir)
 
 /* dir.c */
 extern int nilfs_add_link(struct dentry *, struct inode *);
-extern ino_t nilfs_inode_by_name(struct inode *, struct dentry *);
+extern ino_t nilfs_inode_by_name(struct inode *, const struct qstr *);
 extern int nilfs_make_empty(struct inode *, struct inode *);
 extern struct nilfs_dir_entry *
-nilfs_find_entry(struct inode *, struct dentry *, struct page **);
+nilfs_find_entry(struct inode *, const struct qstr *, struct page **);
 extern int nilfs_delete_entry(struct nilfs_dir_entry *, struct page *);
 extern int nilfs_empty_dir(struct inode *);
 extern struct nilfs_dir_entry *nilfs_dotdot(struct inode *, struct page **);

fs/notify/inotify/inotify_user.c

@@ -29,14 +29,12 @@
 #include <linux/init.h> /* module_init */
 #include <linux/inotify.h>
 #include <linux/kernel.h> /* roundup() */
-#include <linux/magic.h> /* superblock magic number */
-#include <linux/mount.h> /* mntget */
 #include <linux/namei.h> /* LOOKUP_FOLLOW */
-#include <linux/path.h> /* struct path */
 #include <linux/sched.h> /* struct user */
 #include <linux/slab.h> /* struct kmem_cache */
 #include <linux/syscalls.h>
 #include <linux/types.h>
+#include <linux/anon_inodes.h>
 #include <linux/uaccess.h>
 #include <linux/poll.h>
 #include <linux/wait.h>
@@ -45,8 +43,6 @@
 
 #include <asm/ioctls.h>
 
-static struct vfsmount *inotify_mnt __read_mostly;
-
 /* these are configurable via /proc/sys/fs/inotify/ */
 static int inotify_max_user_instances __read_mostly;
 static int inotify_max_queued_events __read_mostly;
@@ -645,9 +641,7 @@ SYSCALL_DEFINE1(inotify_init1, int, flags)
 {
 	struct fsnotify_group *group;
 	struct user_struct *user;
-	struct file *filp;
+	int ret;
-	struct path path;
-	int fd, ret;
 
 	/* Check the IN_* constants for consistency.  */
 	BUILD_BUG_ON(IN_CLOEXEC != O_CLOEXEC);
@@ -656,10 +650,6 @@ SYSCALL_DEFINE1(inotify_init1, int, flags)
 	if (flags & ~(IN_CLOEXEC | IN_NONBLOCK))
 		return -EINVAL;
 
-	fd = get_unused_fd_flags(flags & O_CLOEXEC);
-	if (fd < 0)
-		return fd;
-
 	user = get_current_user();
 	if (unlikely(atomic_read(&user->inotify_devs) >=
 			inotify_max_user_instances)) {
@@ -676,27 +666,14 @@ SYSCALL_DEFINE1(inotify_init1, int, flags)
 
 	atomic_inc(&user->inotify_devs);
 
-	path.mnt = inotify_mnt;
+	ret = anon_inode_getfd("inotify", &inotify_fops, group,
-	path.dentry = inotify_mnt->mnt_root;
+				  O_RDONLY | flags);
-	path_get(&path);
+	if (ret >= 0)
-	filp = alloc_file(&path, FMODE_READ, &inotify_fops);
+		return ret;
-	if (!filp)
-		goto Enfile;
 
-	filp->f_flags = O_RDONLY | (flags & O_NONBLOCK);
-	filp->private_data = group;
-
-	fd_install(fd, filp);
-
-	return fd;
-
-Enfile:
-	ret = -ENFILE;
-	path_put(&path);
 	atomic_dec(&user->inotify_devs);
 out_free_uid:
 	free_uid(user);
-	put_unused_fd(fd);
 	return ret;
 }
 
@@ -783,20 +760,6 @@ out:
 	return ret;
 }
 
-static int
-inotify_get_sb(struct file_system_type *fs_type, int flags,
-	       const char *dev_name, void *data, struct vfsmount *mnt)
-{
-	return get_sb_pseudo(fs_type, "inotify", NULL,
-			INOTIFYFS_SUPER_MAGIC, mnt);
-}
-
-static struct file_system_type inotify_fs_type = {
-    .name	= "inotifyfs",
-    .get_sb	= inotify_get_sb,
-    .kill_sb	= kill_anon_super,
-};
-
 /*
  * inotify_user_setup - Our initialization function.  Note that we cannnot return
  * error because we have compiled-in VFS hooks.  So an (unlikely) failure here
@@ -804,16 +767,6 @@ static struct file_system_type inotify_fs_type = {
  */
 static int __init inotify_user_setup(void)
 {
-	int ret;
-
-	ret = register_filesystem(&inotify_fs_type);
-	if (unlikely(ret))
-		panic("inotify: register_filesystem returned %d!\n", ret);
-
-	inotify_mnt = kern_mount(&inotify_fs_type);
-	if (IS_ERR(inotify_mnt))
-		panic("inotify: kern_mount ret %ld!\n", PTR_ERR(inotify_mnt));
-
 	inotify_inode_mark_cachep = KMEM_CACHE(inotify_inode_mark_entry, SLAB_PANIC);
 	event_priv_cachep = KMEM_CACHE(inotify_event_private_data, SLAB_PANIC);
 

fs/open.c

@@ -271,7 +271,7 @@ static long do_sys_truncate(const char __user *pathname, loff_t length)
 	 * Make sure that there are no leases.  get_write_access() protects
 	 * against the truncate racing with a lease-granting setlease().
 	 */
-	error = break_lease(inode, FMODE_WRITE);
+	error = break_lease(inode, O_WRONLY);
 	if (error)
 		goto put_write_and_out;
 

fs/pnode.c

@@ -86,7 +86,7 @@ static int do_make_slave(struct vfsmount *mnt)
 
 	/*
 	 * slave 'mnt' to a peer mount that has the
-	 * same root dentry. If none is available than
+	 * same root dentry. If none is available then
 	 * slave it to anything that is available.
 	 */
 	while ((peer_mnt = next_peer(peer_mnt)) != mnt &&
@@ -147,6 +147,11 @@ void change_mnt_propagation(struct vfsmount *mnt, int type)
  * get the next mount in the propagation tree.
  * @m: the mount seen last
  * @origin: the original mount from where the tree walk initiated
+ *
+ * Note that peer groups form contiguous segments of slave lists.
+ * We rely on that in get_source() to be able to find out if
+ * vfsmount found while iterating with propagation_next() is
+ * a peer of one we'd found earlier.
  */
 static struct vfsmount *propagation_next(struct vfsmount *m,
 					 struct vfsmount *origin)
@@ -186,10 +191,6 @@ static struct vfsmount *get_source(struct vfsmount *dest,
 {
 	struct vfsmount *p_last_src = NULL;
 	struct vfsmount *p_last_dest = NULL;
-	*type = CL_PROPAGATION;
-
-	if (IS_MNT_SHARED(dest))
-		*type |= CL_MAKE_SHARED;
 
 	while (last_dest != dest->mnt_master) {
 		p_last_dest = last_dest;
@@ -202,13 +203,18 @@ static struct vfsmount *get_source(struct vfsmount *dest,
 		do {
 			p_last_dest = next_peer(p_last_dest);
 		} while (IS_MNT_NEW(p_last_dest));
+		/* is that a peer of the earlier? */
+		if (dest == p_last_dest) {
+			*type = CL_MAKE_SHARED;
+			return p_last_src;
+		}
 	}
-
+	/* slave of the earlier, then */
-	if (dest != p_last_dest) {
+	*type = CL_SLAVE;
-		*type |= CL_SLAVE;
+	/* beginning of peer group among the slaves? */
-		return last_src;
+	if (IS_MNT_SHARED(dest))
-	} else
+		*type |= CL_MAKE_SHARED;
-		return p_last_src;
+	return last_src;
 }
 
 /*

fs/pnode.h

@@ -21,12 +21,11 @@
 #define CL_SLAVE     		0x02
 #define CL_COPY_ALL 		0x04
 #define CL_MAKE_SHARED 		0x08
-#define CL_PROPAGATION 		0x10
+#define CL_PRIVATE 		0x10
-#define CL_PRIVATE 		0x20
 
 static inline void set_mnt_shared(struct vfsmount *mnt)
 {
-	mnt->mnt_flags &= ~MNT_PNODE_MASK;
+	mnt->mnt_flags &= ~MNT_SHARED_MASK;
 	mnt->mnt_flags |= MNT_SHARED;
 }
 

fs/proc/base.c

@@ -647,17 +647,11 @@ static int mounts_release(struct inode *inode, struct file *file)
 static unsigned mounts_poll(struct file *file, poll_table *wait)
 {
 	struct proc_mounts *p = file->private_data;
-	struct mnt_namespace *ns = p->ns;
 	unsigned res = POLLIN | POLLRDNORM;
 
-	poll_wait(file, &ns->poll, wait);
+	poll_wait(file, &p->ns->poll, wait);
-
+	if (mnt_had_events(p))
-	spin_lock(&vfsmount_lock);
-	if (p->event != ns->event) {
-		p->event = ns->event;
 		res |= POLLERR | POLLPRI;
-	}
-	spin_unlock(&vfsmount_lock);
 
 	return res;
 }

fs/proc/generic.c

@@ -662,6 +662,7 @@ struct proc_dir_entry *proc_symlink(const char *name,
 	}
 	return ent;
 }
+EXPORT_SYMBOL(proc_symlink);
 
 struct proc_dir_entry *proc_mkdir_mode(const char *name, mode_t mode,
 		struct proc_dir_entry *parent)
@@ -700,6 +701,7 @@ struct proc_dir_entry *proc_mkdir(const char *name,
 {
 	return proc_mkdir_mode(name, S_IRUGO | S_IXUGO, parent);
 }
+EXPORT_SYMBOL(proc_mkdir);
 
 struct proc_dir_entry *create_proc_entry(const char *name, mode_t mode,
 					 struct proc_dir_entry *parent)
@@ -728,6 +730,7 @@ struct proc_dir_entry *create_proc_entry(const char *name, mode_t mode,
 	}
 	return ent;
 }
+EXPORT_SYMBOL(create_proc_entry);
 
 struct proc_dir_entry *proc_create_data(const char *name, mode_t mode,
 					struct proc_dir_entry *parent,
@@ -762,6 +765,7 @@ out_free:
 out:
 	return NULL;
 }
+EXPORT_SYMBOL(proc_create_data);
 
 static void free_proc_entry(struct proc_dir_entry *de)
 {
@@ -853,3 +857,4 @@ continue_removing:
 			de->parent->name, de->name, de->subdir->name);
 	pde_put(de);
 }
+EXPORT_SYMBOL(remove_proc_entry);

fs/proc/root.c

@@ -220,9 +220,3 @@ void pid_ns_release_proc(struct pid_namespace *ns)
 {
 	mntput(ns->proc_mnt);
 }
-
-EXPORT_SYMBOL(proc_symlink);
-EXPORT_SYMBOL(proc_mkdir);
-EXPORT_SYMBOL(create_proc_entry);
-EXPORT_SYMBOL(proc_create_data);
-EXPORT_SYMBOL(remove_proc_entry);

fs/super.c

@@ -568,7 +568,7 @@ out:
 int do_remount_sb(struct super_block *sb, int flags, void *data, int force)
 {
 	int retval;
-	int remount_rw;
+	int remount_rw, remount_ro;
 
 	if (sb->s_frozen != SB_UNFROZEN)
 		return -EBUSY;
@@ -583,9 +583,12 @@ int do_remount_sb(struct super_block *sb, int flags, void *data, int force)
 	shrink_dcache_sb(sb);
 	sync_filesystem(sb);
 
+	remount_ro = (flags & MS_RDONLY) && !(sb->s_flags & MS_RDONLY);
+	remount_rw = !(flags & MS_RDONLY) && (sb->s_flags & MS_RDONLY);
+
 	/* If we are remounting RDONLY and current sb is read/write,
 	   make sure there are no rw files opened */
-	if ((flags & MS_RDONLY) && !(sb->s_flags & MS_RDONLY)) {
+	if (remount_ro) {
 		if (force)
 			mark_files_ro(sb);
 		else if (!fs_may_remount_ro(sb))
@@ -594,7 +597,6 @@ int do_remount_sb(struct super_block *sb, int flags, void *data, int force)
 		if (retval < 0 && retval != -ENOSYS)
 			return -EBUSY;
 	}
-	remount_rw = !(flags & MS_RDONLY) && (sb->s_flags & MS_RDONLY);
 
 	if (sb->s_op->remount_fs) {
 		retval = sb->s_op->remount_fs(sb, &flags, data);
@@ -604,6 +606,16 @@ int do_remount_sb(struct super_block *sb, int flags, void *data, int force)
 	sb->s_flags = (sb->s_flags & ~MS_RMT_MASK) | (flags & MS_RMT_MASK);
 	if (remount_rw)
 		vfs_dq_quota_on_remount(sb);
+	/*
+	 * Some filesystems modify their metadata via some other path than the
+	 * bdev buffer cache (eg. use a private mapping, or directories in
+	 * pagecache, etc). Also file data modifications go via their own
+	 * mappings. So If we try to mount readonly then copy the filesystem
+	 * from bdev, we could get stale data, so invalidate it to give a best
+	 * effort at coherency.
+	 */
+	if (remount_ro && sb->s_bdev)
+		invalidate_bdev(sb->s_bdev);
 	return 0;
 }
 
@@ -925,6 +937,9 @@ vfs_kern_mount(struct file_system_type *type, int flags, const char *name, void
 	if (!mnt)
 		goto out;
 
+	if (flags & MS_KERNMOUNT)
+		mnt->mnt_flags = MNT_INTERNAL;
+
 	if (data && !(type->fs_flags & FS_BINARY_MOUNTDATA)) {
 		secdata = alloc_secdata();
 		if (!secdata)

fs/udf/balloc.c

@@ -547,7 +547,7 @@ static void udf_table_free_blocks(struct super_block *sb,
 		}
 
 		if (epos.offset + (2 * adsize) > sb->s_blocksize) {
-			char *sptr, *dptr;
+			unsigned char *sptr, *dptr;
 			int loffset;
 
 			brelse(oepos.bh);

fs/udf/dir.c

@@ -45,8 +45,8 @@ static int do_udf_readdir(struct inode *dir, struct file *filp,
 	int block, iblock;
 	loff_t nf_pos = (filp->f_pos - 1) << 2;
 	int flen;
-	char *fname = NULL;
+	unsigned char *fname = NULL;
-	char *nameptr;
+	unsigned char *nameptr;
 	uint16_t liu;
 	uint8_t lfi;
 	loff_t size = udf_ext0_offset(dir) + dir->i_size;

fs/udf/inode.c

@@ -1672,7 +1672,7 @@ int8_t udf_add_aext(struct inode *inode, struct extent_position *epos,
 		return -1;
 
 	if (epos->offset + (2 * adsize) > inode->i_sb->s_blocksize) {
-		char *sptr, *dptr;
+		unsigned char *sptr, *dptr;
 		struct buffer_head *nbh;
 		int err, loffset;
 		struct kernel_lb_addr obloc = epos->block;

fs/udf/namei.c

@@ -34,8 +34,8 @@
 #include <linux/crc-itu-t.h>
 #include <linux/exportfs.h>
 
-static inline int udf_match(int len1, const char *name1, int len2,
+static inline int udf_match(int len1, const unsigned char *name1, int len2,
-			    const char *name2)
+			    const unsigned char *name2)
 {
 	if (len1 != len2)
 		return 0;
@@ -142,15 +142,15 @@ int udf_write_fi(struct inode *inode, struct fileIdentDesc *cfi,
 }
 
 static struct fileIdentDesc *udf_find_entry(struct inode *dir,
-					    struct qstr *child,
+					    const struct qstr *child,
 					    struct udf_fileident_bh *fibh,
 					    struct fileIdentDesc *cfi)
 {
 	struct fileIdentDesc *fi = NULL;
 	loff_t f_pos;
 	int block, flen;
-	char *fname = NULL;
+	unsigned char *fname = NULL;
-	char *nameptr;
+	unsigned char *nameptr;
 	uint8_t lfi;
 	uint16_t liu;
 	loff_t size;
@@ -308,7 +308,7 @@ static struct fileIdentDesc *udf_add_entry(struct inode *dir,
 {
 	struct super_block *sb = dir->i_sb;
 	struct fileIdentDesc *fi = NULL;
-	char *name = NULL;
+	unsigned char *name = NULL;
 	int namelen;
 	loff_t f_pos;
 	loff_t size = udf_ext0_offset(dir) + dir->i_size;
@@ -885,16 +885,16 @@ static int udf_symlink(struct inode *dir, struct dentry *dentry,
 {
 	struct inode *inode;
 	struct pathComponent *pc;
-	char *compstart;
+	const char *compstart;
 	struct udf_fileident_bh fibh;
 	struct extent_position epos = {};
 	int eoffset, elen = 0;
 	struct fileIdentDesc *fi;
 	struct fileIdentDesc cfi;
-	char *ea;
+	uint8_t *ea;
 	int err;
 	int block;
-	char *name = NULL;
+	unsigned char *name = NULL;
 	int namelen;
 	struct buffer_head *bh;
 	struct udf_inode_info *iinfo;
@@ -970,7 +970,7 @@ static int udf_symlink(struct inode *dir, struct dentry *dentry,
 
 		pc = (struct pathComponent *)(ea + elen);
 
-		compstart = (char *)symname;
+		compstart = symname;
 
 		do {
 			symname++;

fs/udf/symlink.c

@@ -32,12 +32,12 @@
 #include <linux/buffer_head.h>
 #include "udf_i.h"
 
-static void udf_pc_to_char(struct super_block *sb, char *from, int fromlen,
+static void udf_pc_to_char(struct super_block *sb, unsigned char *from,
-			   char *to)
+			   int fromlen, unsigned char *to)
 {
 	struct pathComponent *pc;
 	int elen = 0;
-	char *p = to;
+	unsigned char *p = to;
 
 	while (elen < fromlen) {
 		pc = (struct pathComponent *)(from + elen);
@@ -75,9 +75,9 @@ static int udf_symlink_filler(struct file *file, struct page *page)
 {
 	struct inode *inode = page->mapping->host;
 	struct buffer_head *bh = NULL;
-	char *symlink;
+	unsigned char *symlink;
 	int err = -EIO;
-	char *p = kmap(page);
+	unsigned char *p = kmap(page);
 	struct udf_inode_info *iinfo;
 
 	lock_kernel();

fs/ufs/dir.c

@@ -31,7 +31,7 @@
  * len <= UFS_MAXNAMLEN and de != NULL are guaranteed by caller.
  */
 static inline int ufs_match(struct super_block *sb, int len,
-		const char * const name, struct ufs_dir_entry * de)
+		const unsigned char *name, struct ufs_dir_entry *de)
 {
 	if (len != ufs_get_de_namlen(sb, de))
 		return 0;
@@ -70,7 +70,7 @@ static inline unsigned long ufs_dir_pages(struct inode *inode)
 	return (inode->i_size+PAGE_CACHE_SIZE-1)>>PAGE_CACHE_SHIFT;
 }
 
-ino_t ufs_inode_by_name(struct inode *dir, struct qstr *qstr)
+ino_t ufs_inode_by_name(struct inode *dir, const struct qstr *qstr)
 {
 	ino_t res = 0;
 	struct ufs_dir_entry *de;
@@ -249,11 +249,11 @@ struct ufs_dir_entry *ufs_dotdot(struct inode *dir, struct page **p)
  * (as a parameter - res_dir). Page is returned mapped and unlocked.
  * Entry is guaranteed to be valid.
  */
-struct ufs_dir_entry *ufs_find_entry(struct inode *dir, struct qstr *qstr,
+struct ufs_dir_entry *ufs_find_entry(struct inode *dir, const struct qstr *qstr,
 				     struct page **res_page)
 {
 	struct super_block *sb = dir->i_sb;
-	const char *name = qstr->name;
+	const unsigned char *name = qstr->name;
 	int namelen = qstr->len;
 	unsigned reclen = UFS_DIR_REC_LEN(namelen);
 	unsigned long start, n;
@@ -313,7 +313,7 @@ found:
 int ufs_add_link(struct dentry *dentry, struct inode *inode)
 {
 	struct inode *dir = dentry->d_parent->d_inode;
-	const char *name = dentry->d_name.name;
+	const unsigned char *name = dentry->d_name.name;
 	int namelen = dentry->d_name.len;
 	struct super_block *sb = dir->i_sb;
 	unsigned reclen = UFS_DIR_REC_LEN(namelen);

fs/ufs/ufs.h

@@ -86,9 +86,9 @@ extern void ufs_put_cylinder (struct super_block *, unsigned);
 /* dir.c */
 extern const struct inode_operations ufs_dir_inode_operations;
 extern int ufs_add_link (struct dentry *, struct inode *);
-extern ino_t ufs_inode_by_name(struct inode *, struct qstr *);
+extern ino_t ufs_inode_by_name(struct inode *, const struct qstr *);
 extern int ufs_make_empty(struct inode *, struct inode *);
-extern struct ufs_dir_entry *ufs_find_entry(struct inode *, struct qstr *, struct page **);
+extern struct ufs_dir_entry *ufs_find_entry(struct inode *, const struct qstr *, struct page **);
 extern int ufs_delete_entry(struct inode *, struct ufs_dir_entry *, struct page *);
 extern int ufs_empty_dir (struct inode *);
 extern struct ufs_dir_entry *ufs_dotdot(struct inode *, struct page **);

include/linux/audit.h

@@ -424,7 +424,7 @@ extern void audit_syscall_exit(int failed, long return_code);
 extern void __audit_getname(const char *name);
 extern void audit_putname(const char *name);
 extern void __audit_inode(const char *name, const struct dentry *dentry);
-extern void __audit_inode_child(const char *dname, const struct dentry *dentry,
+extern void __audit_inode_child(const struct dentry *dentry,
 				const struct inode *parent);
 extern void __audit_ptrace(struct task_struct *t);
 
@@ -442,11 +442,10 @@ static inline void audit_inode(const char *name, const struct dentry *dentry) {
 	if (unlikely(!audit_dummy_context()))
 		__audit_inode(name, dentry);
 }
-static inline void audit_inode_child(const char *dname, 
+static inline void audit_inode_child(const struct dentry *dentry,
-				     const struct dentry *dentry,
 				     const struct inode *parent) {
 	if (unlikely(!audit_dummy_context()))
-		__audit_inode_child(dname, dentry, parent);
+		__audit_inode_child(dentry, parent);
 }
 void audit_core_dumps(long signr);
 
@@ -544,9 +543,9 @@ extern int audit_signals;
 #define audit_getname(n) do { ; } while (0)
 #define audit_putname(n) do { ; } while (0)
 #define __audit_inode(n,d) do { ; } while (0)
-#define __audit_inode_child(d,i,p) do { ; } while (0)
+#define __audit_inode_child(i,p) do { ; } while (0)
 #define audit_inode(n,d) do { ; } while (0)
-#define audit_inode_child(d,i,p) do { ; } while (0)
+#define audit_inode_child(i,p) do { ; } while (0)
 #define audit_core_dumps(i) do { ; } while (0)
 #define auditsc_get_stamp(c,t,s) (0)
 #define audit_get_loginuid(t) (-1)

include/linux/fs.h

@@ -1305,6 +1305,8 @@ extern int send_sigurg(struct fown_struct *fown);
 #define MNT_FORCE	0x00000001	/* Attempt to forcibily umount */
 #define MNT_DETACH	0x00000002	/* Just detach from the tree */
 #define MNT_EXPIRE	0x00000004	/* Mark for expiry */
+#define UMOUNT_NOFOLLOW	0x00000008	/* Don't follow symlink on umount */
+#define UMOUNT_UNUSED	0x80000000	/* Flag guaranteed to be unused */
 
 extern struct list_head super_blocks;
 extern spinlock_t sb_lock;
@@ -1314,9 +1316,9 @@ extern spinlock_t sb_lock;
 struct super_block {
 	struct list_head	s_list;		/* Keep this first */
 	dev_t			s_dev;		/* search index; _not_ kdev_t */
-	unsigned long		s_blocksize;
-	unsigned char		s_blocksize_bits;
 	unsigned char		s_dirt;
+	unsigned char		s_blocksize_bits;
+	unsigned long		s_blocksize;
 	loff_t			s_maxbytes;	/* Max file size */
 	struct file_system_type	*s_type;
 	const struct super_operations	*s_op;
@@ -1357,16 +1359,16 @@ struct super_block {
 	void 			*s_fs_info;	/* Filesystem private info */
 	fmode_t			s_mode;
 
+	/* Granularity of c/m/atime in ns.
+	   Cannot be worse than a second */
+	u32		   s_time_gran;
+
 	/*
 	 * The next field is for VFS *only*. No filesystems have any business
 	 * even looking at it. You had been warned.
 	 */
 	struct mutex s_vfs_rename_mutex;	/* Kludge */
 
-	/* Granularity of c/m/atime in ns.
-	   Cannot be worse than a second */
-	u32		   s_time_gran;
-
 	/*
 	 * Filesystem subtype.  If non-empty the filesystem type field
 	 * in /proc/mounts will be "type.subtype"
@@ -1794,7 +1796,8 @@ extern int may_umount(struct vfsmount *);
 extern long do_mount(char *, char *, char *, unsigned long, void *);
 extern struct vfsmount *collect_mounts(struct path *);
 extern void drop_collected_mounts(struct vfsmount *);
-
+extern int iterate_mounts(int (*)(struct vfsmount *, void *), void *,
+			  struct vfsmount *);
 extern int vfs_statfs(struct dentry *, struct kstatfs *);
 
 extern int current_umask(void);
@@ -2058,12 +2061,6 @@ extern int invalidate_inodes(struct super_block *);
 unsigned long invalidate_mapping_pages(struct address_space *mapping,
 					pgoff_t start, pgoff_t end);
 
-static inline unsigned long __deprecated
-invalidate_inode_pages(struct address_space *mapping)
-{
-	return invalidate_mapping_pages(mapping, 0, ~0UL);
-}
-
 static inline void invalidate_remote_inode(struct inode *inode)
 {
 	if (S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) ||
@@ -2132,6 +2129,7 @@ extern struct file * open_exec(const char *);
  
 /* fs/dcache.c -- generic fs support functions */
 extern int is_subdir(struct dentry *, struct dentry *);
+extern int path_is_under(struct path *, struct path *);
 extern ino_t find_inode_number(struct dentry *, struct qstr *);
 
 #include <linux/err.h>
@@ -2340,8 +2338,6 @@ extern int simple_rename(struct inode *, struct dentry *, struct inode *, struct
 extern int simple_sync_file(struct file *, struct dentry *, int);
 extern int simple_empty(struct dentry *);
 extern int simple_readpage(struct file *file, struct page *page);
-extern int simple_prepare_write(struct file *file, struct page *page,
-			unsigned offset, unsigned to);
 extern int simple_write_begin(struct file *file, struct address_space *mapping,
 			loff_t pos, unsigned len, unsigned flags,
 			struct page **pagep, void **fsdata);

include/linux/fsnotify.h

@@ -65,7 +65,7 @@ static inline void fsnotify_link_count(struct inode *inode)
  * fsnotify_move - file old_name at old_dir was moved to new_name at new_dir
  */
 static inline void fsnotify_move(struct inode *old_dir, struct inode *new_dir,
-				 const char *old_name, const char *new_name,
+				 const char *old_name,
 				 int isdir, struct inode *target, struct dentry *moved)
 {
 	struct inode *source = moved->d_inode;
@@ -73,6 +73,7 @@ static inline void fsnotify_move(struct inode *old_dir, struct inode *new_dir,
 	u32 fs_cookie = fsnotify_get_cookie();
 	__u32 old_dir_mask = (FS_EVENT_ON_CHILD | FS_MOVED_FROM);
 	__u32 new_dir_mask = (FS_EVENT_ON_CHILD | FS_MOVED_TO);
+	const char *new_name = moved->d_name.name;
 
 	if (old_dir == new_dir)
 		old_dir_mask |= FS_DN_RENAME;
@@ -103,7 +104,7 @@ static inline void fsnotify_move(struct inode *old_dir, struct inode *new_dir,
 		inotify_inode_queue_event(source, IN_MOVE_SELF, 0, NULL, NULL);
 		fsnotify(source, FS_MOVE_SELF, moved->d_inode, FSNOTIFY_EVENT_INODE, NULL, 0);
 	}
-	audit_inode_child(new_name, moved, new_dir);
+	audit_inode_child(moved, new_dir);
 }
 
 /*
@@ -146,7 +147,7 @@ static inline void fsnotify_create(struct inode *inode, struct dentry *dentry)
 {
 	inotify_inode_queue_event(inode, IN_CREATE, 0, dentry->d_name.name,
 				  dentry->d_inode);
-	audit_inode_child(dentry->d_name.name, dentry, inode);
+	audit_inode_child(dentry, inode);
 
 	fsnotify(inode, FS_CREATE, dentry->d_inode, FSNOTIFY_EVENT_INODE, dentry->d_name.name, 0);
 }
@@ -161,7 +162,7 @@ static inline void fsnotify_link(struct inode *dir, struct inode *inode, struct
 	inotify_inode_queue_event(dir, IN_CREATE, 0, new_dentry->d_name.name,
 				  inode);
 	fsnotify_link_count(inode);
-	audit_inode_child(new_dentry->d_name.name, new_dentry, dir);
+	audit_inode_child(new_dentry, dir);
 
 	fsnotify(dir, FS_CREATE, inode, FSNOTIFY_EVENT_INODE, new_dentry->d_name.name, 0);
 }
@@ -175,7 +176,7 @@ static inline void fsnotify_mkdir(struct inode *inode, struct dentry *dentry)
 	struct inode *d_inode = dentry->d_inode;
 
 	inotify_inode_queue_event(inode, mask, 0, dentry->d_name.name, d_inode);
-	audit_inode_child(dentry->d_name.name, dentry, inode);
+	audit_inode_child(dentry, inode);
 
 	fsnotify(inode, mask, d_inode, FSNOTIFY_EVENT_INODE, dentry->d_name.name, 0);
 }

include/linux/magic.h

@@ -52,7 +52,6 @@
 #define CGROUP_SUPER_MAGIC	0x27e0eb
 
 #define FUTEXFS_SUPER_MAGIC	0xBAD1DEA
-#define INOTIFYFS_SUPER_MAGIC	0x2BAD1DEA
 
 #define STACK_END_MAGIC		0x57AC6E9D
 

include/linux/mnt_namespace.h

@@ -35,6 +35,7 @@ static inline void get_mnt_ns(struct mnt_namespace *ns)
 extern const struct seq_operations mounts_op;
 extern const struct seq_operations mountinfo_op;
 extern const struct seq_operations mountstats_op;
+extern int mnt_had_events(struct proc_mounts *);
 
 #endif
 #endif

include/linux/mount.h

@@ -34,7 +34,18 @@ struct mnt_namespace;
 
 #define MNT_SHARED	0x1000	/* if the vfsmount is a shared mount */
 #define MNT_UNBINDABLE	0x2000	/* if the vfsmount is a unbindable mount */
-#define MNT_PNODE_MASK	0x3000	/* propagation flag mask */
+/*
+ * MNT_SHARED_MASK is the set of flags that should be cleared when a
+ * mount becomes shared.  Currently, this is only the flag that says a
+ * mount cannot be bind mounted, since this is how we create a mount
+ * that shares events with another mount.  If you add a new MNT_*
+ * flag, consider how it interacts with shared mounts.
+ */
+#define MNT_SHARED_MASK	(MNT_UNBINDABLE)
+#define MNT_PROPAGATION_MASK	(MNT_SHARED | MNT_UNBINDABLE)
+
+
+#define MNT_INTERNAL	0x4000
 
 struct vfsmount {
 	struct list_head mnt_hash;
@@ -123,7 +134,6 @@ extern int do_add_mount(struct vfsmount *newmnt, struct path *path,
 
 extern void mark_mounts_for_expiry(struct list_head *mounts);
 
-extern spinlock_t vfsmount_lock;
 extern dev_t name_to_dev_t(char *name);
 
 #endif /* _LINUX_MOUNT_H */

init/do_mounts_initrd.c

@@ -30,11 +30,7 @@ static int __init do_linuxrc(void * shell)
 	extern char * envp_init[];
 
 	sys_close(old_fd);sys_close(root_fd);
-	sys_close(0);sys_close(1);sys_close(2);
 	sys_setsid();
-	(void) sys_open("/dev/console",O_RDWR,0);
-	(void) sys_dup(0);
-	(void) sys_dup(0);
 	return kernel_execve(shell, argv, envp_init);
 }
 

init/main.c

@@ -822,11 +822,6 @@ static noinline int init_post(void)
 	system_state = SYSTEM_RUNNING;
 	numa_default_policy();
 
-	if (sys_open((const char __user *) "/dev/console", O_RDWR, 0) < 0)
-		printk(KERN_WARNING "Warning: unable to open an initial console.\n");
-
-	(void) sys_dup(0);
-	(void) sys_dup(0);
 
 	current->signal->flags |= SIGNAL_UNKILLABLE;
 
@@ -889,6 +884,12 @@ static int __init kernel_init(void * unused)
 
 	do_basic_setup();
 
+	/* Open the /dev/console on the rootfs, this should never fail */
+	if (sys_open((const char __user *) "/dev/console", O_RDWR, 0) < 0)
+		printk(KERN_WARNING "Warning: unable to open an initial console.\n");
+
+	(void) sys_dup(0);
+	(void) sys_dup(0);
 	/*
 	 * check if there is an early userspace init.  If yes, let it do all
 	 * the work

ipc/mqueue.c

@@ -134,7 +134,6 @@ static struct inode *mqueue_get_inode(struct super_block *sb,
 			init_waitqueue_head(&info->wait_q);
 			INIT_LIST_HEAD(&info->e_wait_q[0].list);
 			INIT_LIST_HEAD(&info->e_wait_q[1].list);
-			info->messages = NULL;
 			info->notify_owner = NULL;
 			info->qsize = 0;
 			info->user = NULL;	/* set when all is ok */
@@ -146,6 +145,10 @@ static struct inode *mqueue_get_inode(struct super_block *sb,
 				info->attr.mq_msgsize = attr->mq_msgsize;
 			}
 			mq_msg_tblsz = info->attr.mq_maxmsg * sizeof(struct msg_msg *);
+			info->messages = kmalloc(mq_msg_tblsz, GFP_KERNEL);
+			if (!info->messages)
+				goto out_inode;
+
 			mq_bytes = (mq_msg_tblsz +
 				(info->attr.mq_maxmsg * info->attr.mq_msgsize));
 
@@ -154,18 +157,12 @@ static struct inode *mqueue_get_inode(struct super_block *sb,
 		 	    u->mq_bytes + mq_bytes >
 			    p->signal->rlim[RLIMIT_MSGQUEUE].rlim_cur) {
 				spin_unlock(&mq_lock);
+				kfree(info->messages);
 				goto out_inode;
 			}
 			u->mq_bytes += mq_bytes;
 			spin_unlock(&mq_lock);
 
-			info->messages = kmalloc(mq_msg_tblsz, GFP_KERNEL);
-			if (!info->messages) {
-				spin_lock(&mq_lock);
-				u->mq_bytes -= mq_bytes;
-				spin_unlock(&mq_lock);
-				goto out_inode;
-			}
 			/* all is ok */
 			info->user = get_uid(u);
 		} else if (S_ISDIR(mode)) {
@@ -187,7 +184,7 @@ static int mqueue_fill_super(struct super_block *sb, void *data, int silent)
 {
 	struct inode *inode;
 	struct ipc_namespace *ns = data;
-	int error = 0;
+	int error;
 
 	sb->s_blocksize = PAGE_CACHE_SIZE;
 	sb->s_blocksize_bits = PAGE_CACHE_SHIFT;
@@ -205,7 +202,9 @@ static int mqueue_fill_super(struct super_block *sb, void *data, int silent)
 	if (!sb->s_root) {
 		iput(inode);
 		error = -ENOMEM;
+		goto out;
 	}
+	error = 0;
 
 out:
 	return error;
@@ -264,8 +263,9 @@ static void mqueue_delete_inode(struct inode *inode)
 
 	clear_inode(inode);
 
-	mq_bytes = (info->attr.mq_maxmsg * sizeof(struct msg_msg *) +
+	/* Total amount of bytes accounted for the mqueue */
-		   (info->attr.mq_maxmsg * info->attr.mq_msgsize));
+	mq_bytes = info->attr.mq_maxmsg * (sizeof(struct msg_msg *)
+	    + info->attr.mq_msgsize);
 	user = info->user;
 	if (user) {
 		spin_lock(&mq_lock);
@@ -604,8 +604,8 @@ static int mq_attr_ok(struct ipc_namespace *ipc_ns, struct mq_attr *attr)
 	/* check for overflow */
 	if (attr->mq_msgsize > ULONG_MAX/attr->mq_maxmsg)
 		return 0;
-	if ((unsigned long)(attr->mq_maxmsg * attr->mq_msgsize) +
+	if ((unsigned long)(attr->mq_maxmsg * (attr->mq_msgsize
-	    (attr->mq_maxmsg * sizeof (struct msg_msg *)) <
+	    + sizeof (struct msg_msg *))) <
 	    (unsigned long)(attr->mq_maxmsg * attr->mq_msgsize))
 		return 0;
 	return 1;
@@ -623,9 +623,10 @@ static struct file *do_create(struct ipc_namespace *ipc_ns, struct dentry *dir,
 	int ret;
 
 	if (attr) {
-		ret = -EINVAL;
+		if (!mq_attr_ok(ipc_ns, attr)) {
-		if (!mq_attr_ok(ipc_ns, attr))
+			ret = -EINVAL;
 			goto out;
+		}
 		/* store for use during create */
 		dentry->d_fsdata = attr;
 	}
@@ -659,24 +660,28 @@ out:
 static struct file *do_open(struct ipc_namespace *ipc_ns,
 				struct dentry *dentry, int oflag)
 {
+	int ret;
 	const struct cred *cred = current_cred();
 
 	static const int oflag2acc[O_ACCMODE] = { MAY_READ, MAY_WRITE,
 						  MAY_READ | MAY_WRITE };
 
 	if ((oflag & O_ACCMODE) == (O_RDWR | O_WRONLY)) {
-		dput(dentry);
+		ret = -EINVAL;
-		mntput(ipc_ns->mq_mnt);
+		goto err;
-		return ERR_PTR(-EINVAL);
 	}
 
 	if (inode_permission(dentry->d_inode, oflag2acc[oflag & O_ACCMODE])) {
-		dput(dentry);
+		ret = -EACCES;
-		mntput(ipc_ns->mq_mnt);
+		goto err;
-		return ERR_PTR(-EACCES);
 	}
 
 	return dentry_open(dentry, ipc_ns->mq_mnt, oflag, cred);
+
+err:
+	dput(dentry);
+	mntput(ipc_ns->mq_mnt);
+	return ERR_PTR(ret);
 }
 
 SYSCALL_DEFINE4(mq_open, const char __user *, u_name, int, oflag, mode_t, mode,
@@ -705,16 +710,17 @@ SYSCALL_DEFINE4(mq_open, const char __user *, u_name, int, oflag, mode_t, mode,
 	dentry = lookup_one_len(name, ipc_ns->mq_mnt->mnt_root, strlen(name));
 	if (IS_ERR(dentry)) {
 		error = PTR_ERR(dentry);
-		goto out_err;
+		goto out_putfd;
 	}
 	mntget(ipc_ns->mq_mnt);
 
 	if (oflag & O_CREAT) {
 		if (dentry->d_inode) {	/* entry already exists */
 			audit_inode(name, dentry);
-			error = -EEXIST;
+			if (oflag & O_EXCL) {
-			if (oflag & O_EXCL)
+				error = -EEXIST;
 				goto out;
+			}
 			filp = do_open(ipc_ns, dentry, oflag);
 		} else {
 			filp = do_create(ipc_ns, ipc_ns->mq_mnt->mnt_root,
@@ -722,9 +728,10 @@ SYSCALL_DEFINE4(mq_open, const char __user *, u_name, int, oflag, mode_t, mode,
 						u_attr ? &attr : NULL);
 		}
 	} else {
-		error = -ENOENT;
+		if (!dentry->d_inode) {
-		if (!dentry->d_inode)
+			error = -ENOENT;
 			goto out;
+		}
 		audit_inode(name, dentry);
 		filp = do_open(ipc_ns, dentry, oflag);
 	}
@@ -742,7 +749,6 @@ out:
 	mntput(ipc_ns->mq_mnt);
 out_putfd:
 	put_unused_fd(fd);
-out_err:
 	fd = error;
 out_upsem:
 	mutex_unlock(&ipc_ns->mq_mnt->mnt_root->d_inode->i_mutex);
@@ -872,19 +878,24 @@ SYSCALL_DEFINE5(mq_timedsend, mqd_t, mqdes, const char __user *, u_msg_ptr,
 	audit_mq_sendrecv(mqdes, msg_len, msg_prio, p);
 	timeout = prepare_timeout(p);
 
-	ret = -EBADF;
 	filp = fget(mqdes);
-	if (unlikely(!filp))
+	if (unlikely(!filp)) {
+		ret = -EBADF;
 		goto out;
+	}
 
 	inode = filp->f_path.dentry->d_inode;
-	if (unlikely(filp->f_op != &mqueue_file_operations))
+	if (unlikely(filp->f_op != &mqueue_file_operations)) {
+		ret = -EBADF;
 		goto out_fput;
+	}
 	info = MQUEUE_I(inode);
 	audit_inode(NULL, filp->f_path.dentry);
 
-	if (unlikely(!(filp->f_mode & FMODE_WRITE)))
+	if (unlikely(!(filp->f_mode & FMODE_WRITE))) {
+		ret = -EBADF;
 		goto out_fput;
+	}
 
 	if (unlikely(msg_len > info->attr.mq_msgsize)) {
 		ret = -EMSGSIZE;
@@ -961,19 +972,24 @@ SYSCALL_DEFINE5(mq_timedreceive, mqd_t, mqdes, char __user *, u_msg_ptr,
 	audit_mq_sendrecv(mqdes, msg_len, 0, p);
 	timeout = prepare_timeout(p);
 
-	ret = -EBADF;
 	filp = fget(mqdes);
-	if (unlikely(!filp))
+	if (unlikely(!filp)) {
+		ret = -EBADF;
 		goto out;
+	}
 
 	inode = filp->f_path.dentry->d_inode;
-	if (unlikely(filp->f_op != &mqueue_file_operations))
+	if (unlikely(filp->f_op != &mqueue_file_operations)) {
+		ret = -EBADF;
 		goto out_fput;
+	}
 	info = MQUEUE_I(inode);
 	audit_inode(NULL, filp->f_path.dentry);
 
-	if (unlikely(!(filp->f_mode & FMODE_READ)))
+	if (unlikely(!(filp->f_mode & FMODE_READ))) {
+		ret = -EBADF;
 		goto out_fput;
+	}
 
 	/* checks if buffer is big enough */
 	if (unlikely(msg_len < info->attr.mq_msgsize)) {
@@ -1063,13 +1079,14 @@ SYSCALL_DEFINE2(mq_notify, mqd_t, mqdes,
 
 			/* create the notify skb */
 			nc = alloc_skb(NOTIFY_COOKIE_LEN, GFP_KERNEL);
-			ret = -ENOMEM;
+			if (!nc) {
-			if (!nc)
+				ret = -ENOMEM;
 				goto out;
-			ret = -EFAULT;
+			}
 			if (copy_from_user(nc->data,
 					notification.sigev_value.sival_ptr,
 					NOTIFY_COOKIE_LEN)) {
+				ret = -EFAULT;
 				goto out;
 			}
 
@@ -1078,9 +1095,10 @@ SYSCALL_DEFINE2(mq_notify, mqd_t, mqdes,
 			/* and attach it to the socket */
 retry:
 			filp = fget(notification.sigev_signo);
-			ret = -EBADF;
+			if (!filp) {
-			if (!filp)
+				ret = -EBADF;
 				goto out;
+			}
 			sock = netlink_getsockbyfilp(filp);
 			fput(filp);
 			if (IS_ERR(sock)) {
@@ -1092,7 +1110,7 @@ retry:
 			timeo = MAX_SCHEDULE_TIMEOUT;
 			ret = netlink_attachskb(sock, nc, &timeo, NULL);
 			if (ret == 1)
-		       		goto retry;
+				goto retry;
 			if (ret) {
 				sock = NULL;
 				nc = NULL;
@@ -1101,14 +1119,17 @@ retry:
 		}
 	}
 
-	ret = -EBADF;
 	filp = fget(mqdes);
-	if (!filp)
+	if (!filp) {
+		ret = -EBADF;
 		goto out;
+	}
 
 	inode = filp->f_path.dentry->d_inode;
-	if (unlikely(filp->f_op != &mqueue_file_operations))
+	if (unlikely(filp->f_op != &mqueue_file_operations)) {
+		ret = -EBADF;
 		goto out_fput;
+	}
 	info = MQUEUE_I(inode);
 
 	ret = 0;
@@ -1171,14 +1192,17 @@ SYSCALL_DEFINE3(mq_getsetattr, mqd_t, mqdes,
 			return -EINVAL;
 	}
 
-	ret = -EBADF;
 	filp = fget(mqdes);
-	if (!filp)
+	if (!filp) {
+		ret = -EBADF;
 		goto out;
+	}
 
 	inode = filp->f_path.dentry->d_inode;
-	if (unlikely(filp->f_op != &mqueue_file_operations))
+	if (unlikely(filp->f_op != &mqueue_file_operations)) {
+		ret = -EBADF;
 		goto out_fput;
+	}
 	info = MQUEUE_I(inode);
 
 	spin_lock(&info->lock);
@@ -1272,7 +1296,7 @@ static int __init init_mqueue_fs(void)
 	if (mqueue_inode_cachep == NULL)
 		return -ENOMEM;
 
-	/* ignore failues - they are not fatal */
+	/* ignore failures - they are not fatal */
 	mq_sysctl_table = mq_register_sysctl_table();
 
 	error = register_filesystem(&mqueue_fs_type);

kernel/audit_tree.c

@@ -548,6 +548,11 @@ int audit_remove_tree_rule(struct audit_krule *rule)
 	return 0;
 }
 
+static int compare_root(struct vfsmount *mnt, void *arg)
+{
+	return mnt->mnt_root->d_inode == arg;
+}
+
 void audit_trim_trees(void)
 {
 	struct list_head cursor;
@@ -559,7 +564,6 @@ void audit_trim_trees(void)
 		struct path path;
 		struct vfsmount *root_mnt;
 		struct node *node;
-		struct list_head list;
 		int err;
 
 		tree = container_of(cursor.next, struct audit_tree, list);
@@ -577,24 +581,16 @@ void audit_trim_trees(void)
 		if (!root_mnt)
 			goto skip_it;
 
-		list_add_tail(&list, &root_mnt->mnt_list);
 		spin_lock(&hash_lock);
 		list_for_each_entry(node, &tree->chunks, list) {
-			struct audit_chunk *chunk = find_chunk(node);
+			struct inode *inode = find_chunk(node)->watch.inode;
-			struct inode *inode = chunk->watch.inode;
-			struct vfsmount *mnt;
 			node->index |= 1U<<31;
-			list_for_each_entry(mnt, &list, mnt_list) {
+			if (iterate_mounts(compare_root, inode, root_mnt))
-				if (mnt->mnt_root->d_inode == inode) {
+				node->index &= ~(1U<<31);
-					node->index &= ~(1U<<31);
-					break;
-				}
-			}
 		}
 		spin_unlock(&hash_lock);
 		trim_marked(tree);
 		put_tree(tree);
-		list_del_init(&list);
 		drop_collected_mounts(root_mnt);
 skip_it:
 		mutex_lock(&audit_filter_mutex);
@@ -603,22 +599,6 @@ skip_it:
 	mutex_unlock(&audit_filter_mutex);
 }
 
-static int is_under(struct vfsmount *mnt, struct dentry *dentry,
-		    struct path *path)
-{
-	if (mnt != path->mnt) {
-		for (;;) {
-			if (mnt->mnt_parent == mnt)
-				return 0;
-			if (mnt->mnt_parent == path->mnt)
-					break;
-			mnt = mnt->mnt_parent;
-		}
-		dentry = mnt->mnt_mountpoint;
-	}
-	return is_subdir(dentry, path->dentry);
-}
-
 int audit_make_tree(struct audit_krule *rule, char *pathname, u32 op)
 {
 
@@ -638,13 +618,17 @@ void audit_put_tree(struct audit_tree *tree)
 	put_tree(tree);
 }
 
+static int tag_mount(struct vfsmount *mnt, void *arg)
+{
+	return tag_chunk(mnt->mnt_root->d_inode, arg);
+}
+
 /* called with audit_filter_mutex */
 int audit_add_tree_rule(struct audit_krule *rule)
 {
 	struct audit_tree *seed = rule->tree, *tree;
 	struct path path;
-	struct vfsmount *mnt, *p;
+	struct vfsmount *mnt;
-	struct list_head list;
 	int err;
 
 	list_for_each_entry(tree, &tree_list, list) {
@@ -670,16 +654,9 @@ int audit_add_tree_rule(struct audit_krule *rule)
 		err = -ENOMEM;
 		goto Err;
 	}
-	list_add_tail(&list, &mnt->mnt_list);
 
 	get_tree(tree);
-	list_for_each_entry(p, &list, mnt_list) {
+	err = iterate_mounts(tag_mount, tree, mnt);
-		err = tag_chunk(p->mnt_root->d_inode, tree);
-		if (err)
-			break;
-	}
-
-	list_del(&list);
 	drop_collected_mounts(mnt);
 
 	if (!err) {
@@ -714,31 +691,23 @@ int audit_tag_tree(char *old, char *new)
 {
 	struct list_head cursor, barrier;
 	int failed = 0;
-	struct path path;
+	struct path path1, path2;
 	struct vfsmount *tagged;
-	struct list_head list;
-	struct vfsmount *mnt;
-	struct dentry *dentry;
 	int err;
 
-	err = kern_path(new, 0, &path);
+	err = kern_path(new, 0, &path2);
 	if (err)
 		return err;
-	tagged = collect_mounts(&path);
+	tagged = collect_mounts(&path2);
-	path_put(&path);
+	path_put(&path2);
 	if (!tagged)
 		return -ENOMEM;
 
-	err = kern_path(old, 0, &path);
+	err = kern_path(old, 0, &path1);
 	if (err) {
 		drop_collected_mounts(tagged);
 		return err;
 	}
-	mnt = mntget(path.mnt);
-	dentry = dget(path.dentry);
-	path_put(&path);
-
-	list_add_tail(&list, &tagged->mnt_list);
 
 	mutex_lock(&audit_filter_mutex);
 	list_add(&barrier, &tree_list);
@@ -746,7 +715,7 @@ int audit_tag_tree(char *old, char *new)
 
 	while (cursor.next != &tree_list) {
 		struct audit_tree *tree;
-		struct vfsmount *p;
+		int good_one = 0;
 
 		tree = container_of(cursor.next, struct audit_tree, list);
 		get_tree(tree);
@@ -754,30 +723,19 @@ int audit_tag_tree(char *old, char *new)
 		list_add(&cursor, &tree->list);
 		mutex_unlock(&audit_filter_mutex);
 
-		err = kern_path(tree->pathname, 0, &path);
+		err = kern_path(tree->pathname, 0, &path2);
-		if (err) {
+		if (!err) {
+			good_one = path_is_under(&path1, &path2);
+			path_put(&path2);
+		}
+
+		if (!good_one) {
 			put_tree(tree);
 			mutex_lock(&audit_filter_mutex);
 			continue;
 		}
 
-		spin_lock(&vfsmount_lock);
+		failed = iterate_mounts(tag_mount, tree, tagged);
-		if (!is_under(mnt, dentry, &path)) {
-			spin_unlock(&vfsmount_lock);
-			path_put(&path);
-			put_tree(tree);
-			mutex_lock(&audit_filter_mutex);
-			continue;
-		}
-		spin_unlock(&vfsmount_lock);
-		path_put(&path);
-
-		list_for_each_entry(p, &list, mnt_list) {
-			failed = tag_chunk(p->mnt_root->d_inode, tree);
-			if (failed)
-				break;
-		}
-
 		if (failed) {
 			put_tree(tree);
 			mutex_lock(&audit_filter_mutex);
@@ -818,10 +776,8 @@ int audit_tag_tree(char *old, char *new)
 	}
 	list_del(&barrier);
 	list_del(&cursor);
-	list_del(&list);
 	mutex_unlock(&audit_filter_mutex);
-	dput(dentry);
+	path_put(&path1);
-	mntput(mnt);
 	drop_collected_mounts(tagged);
 	return failed;
 }

kernel/auditsc.c

@@ -1988,7 +1988,6 @@ void __audit_inode(const char *name, const struct dentry *dentry)
 
 /**
  * audit_inode_child - collect inode info for created/removed objects
- * @dname: inode's dentry name
  * @dentry: dentry being audited
  * @parent: inode of dentry parent
  *
@@ -2000,13 +1999,14 @@ void __audit_inode(const char *name, const struct dentry *dentry)
  * must be hooked prior, in order to capture the target inode during
  * unsuccessful attempts.
  */
-void __audit_inode_child(const char *dname, const struct dentry *dentry,
+void __audit_inode_child(const struct dentry *dentry,
 			 const struct inode *parent)
 {
 	int idx;
 	struct audit_context *context = current->audit_context;
 	const char *found_parent = NULL, *found_child = NULL;
 	const struct inode *inode = dentry->d_inode;
+	const char *dname = dentry->d_name.name;
 	int dirlen = 0;
 
 	if (!context->in_syscall)
@@ -2014,9 +2014,6 @@ void __audit_inode_child(const char *dname, const struct dentry *dentry,
 
 	if (inode)
 		handle_one(inode);
-	/* determine matching parent */
-	if (!dname)
-		goto add_names;
 
 	/* parent is more likely, look for it first */
 	for (idx = 0; idx < context->name_count; idx++) {

kernel/sysctl_binary.c

@@ -1331,7 +1331,7 @@ static ssize_t binary_sysctl(const int *name, int nlen,
 	ssize_t result;
 	char *pathname;
 	int flags;
-	int acc_mode, fmode;
+	int acc_mode;
 
 	pathname = sysctl_getname(name, nlen, &table);
 	result = PTR_ERR(pathname);
@@ -1342,15 +1342,12 @@ static ssize_t binary_sysctl(const int *name, int nlen,
 	if (oldval && oldlen && newval && newlen) {
 		flags = O_RDWR;
 		acc_mode = MAY_READ | MAY_WRITE;
-		fmode = FMODE_READ | FMODE_WRITE;
 	} else if (newval && newlen) {
 		flags = O_WRONLY;
 		acc_mode = MAY_WRITE;
-		fmode = FMODE_WRITE;
 	} else if (oldval && oldlen) {
 		flags = O_RDONLY;
 		acc_mode = MAY_READ;
-		fmode = FMODE_READ;
 	} else {
 		result = 0;
 		goto out_putname;
@@ -1361,7 +1358,7 @@ static ssize_t binary_sysctl(const int *name, int nlen,
 	if (result)
 		goto out_putname;
 
-	result = may_open(&nd.path, acc_mode, fmode);
+	result = may_open(&nd.path, acc_mode, flags);
 	if (result)
 		goto out_putpath;
 

mm/filemap.c

@@ -1117,7 +1117,7 @@ readpage:
 			if (!PageUptodate(page)) {
 				if (page->mapping == NULL) {
 					/*
-					 * invalidate_inode_pages got it
+					 * invalidate_mapping_pages got it
 					 */
 					unlock_page(page);
 					page_cache_release(page);

net/sunrpc/rpc_pipe.c

@@ -999,19 +999,14 @@ rpc_fill_super(struct super_block *sb, void *data, int silent)
 	inode = rpc_get_inode(sb, S_IFDIR | 0755);
 	if (!inode)
 		return -ENOMEM;
-	root = d_alloc_root(inode);
+	sb->s_root = root = d_alloc_root(inode);
 	if (!root) {
 		iput(inode);
 		return -ENOMEM;
 	}
 	if (rpc_populate(root, files, RPCAUTH_lockd, RPCAUTH_RootEOF, NULL))
-		goto out;
+		return -ENOMEM;
-	sb->s_root = root;
 	return 0;
-out:
-	d_genocide(root);
-	dput(root);
-	return -ENOMEM;
 }
 
 static int

security/smack/smack_lsm.c

@@ -387,7 +387,7 @@ static int smack_sb_umount(struct vfsmount *mnt, int flags)
 	struct smk_audit_info ad;
 
 	smk_ad_init(&ad, __func__, LSM_AUDIT_DATA_FS);
-	smk_ad_setfield_u_fs_path_dentry(&ad, mnt->mnt_mountpoint);
+	smk_ad_setfield_u_fs_path_dentry(&ad, mnt->mnt_root);
 	smk_ad_setfield_u_fs_path_mnt(&ad, mnt);
 
 	sbp = mnt->mnt_sb->s_security;

security/tomoyo/realpath.c

@@ -88,29 +88,14 @@ int tomoyo_realpath_from_path2(struct path *path, char *newname,
 		sp = dentry->d_op->d_dname(dentry, newname + offset,
 					   newname_len - offset);
 	} else {
-		/* Taken from d_namespace_path(). */
+		struct path ns_root = {.mnt = NULL, .dentry = NULL};
-		struct path root;
-		struct path ns_root = { };
-		struct path tmp;
 
-		read_lock(&current->fs->lock);
-		root = current->fs->root;
-		path_get(&root);
-		read_unlock(&current->fs->lock);
-		spin_lock(&vfsmount_lock);
-		if (root.mnt && root.mnt->mnt_ns)
-			ns_root.mnt = mntget(root.mnt->mnt_ns->root);
-		if (ns_root.mnt)
-			ns_root.dentry = dget(ns_root.mnt->mnt_root);
-		spin_unlock(&vfsmount_lock);
 		spin_lock(&dcache_lock);
-		tmp = ns_root;
+		/* go to whatever namespace root we are under */
-		sp = __d_path(path, &tmp, newname, newname_len);
+		sp = __d_path(path, &ns_root, newname, newname_len);
 		spin_unlock(&dcache_lock);
-		path_put(&root);
-		path_put(&ns_root);
 		/* Prepend "/proc" prefix if using internal proc vfs mount. */
-		if (!IS_ERR(sp) && (path->mnt->mnt_parent == path->mnt) &&
+		if (!IS_ERR(sp) && (path->mnt->mnt_flags & MNT_INTERNAL) &&
 		    (path->mnt->mnt_sb->s_magic == PROC_SUPER_MAGIC)) {
 			sp -= 5;
 			if (sp >= newname)