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:
  reiserfs: fix j_last_flush_trans_id type
  fs: Mark get_filesystem_list() as __init function.
  kill vfs_stat_fd / vfs_lstat_fd
  Separate out common fstatat code into vfs_fstatat
  ecryptfs: use memdup_user()
  ncpfs: use memdup_user()
  xfs: use memdup_user()
  sysfs: use memdup_user()
  btrfs: use memdup_user()
  xattr: use memdup_user()
  autofs4: use memchr() in invalid_string()
  Documentation/filesystems: remove out of date reference to BKL being held
  Fix i_mutex vs. readdir handling in nfsd
  fs/compat_ioctl: fix build when !BLOCK
  Fix autofs_expire()
  No need for crossing to mountpoint in audit_tag_tree()
  Safer nfsd_cross_mnt()
  Touch all affected namespaces on propagation of mount
  Fix AUTOFS_DEV_IOCTL_REQUESTER_CMD

Documentation/filesystems/vfs.txt

@@ -277,8 +277,7 @@ or bottom half).
   unfreeze_fs: called when VFS is unlocking a filesystem and making it writable
   	again.
 
-  statfs: called when the VFS needs to get filesystem statistics. This
-	is called with the kernel lock held
+  statfs: called when the VFS needs to get filesystem statistics.
 
   remount_fs: called when the filesystem is remounted. This is called
 	with the kernel lock held

arch/arm/kernel/sys_oabi-compat.c

@@ -177,21 +177,12 @@ asmlinkage long sys_oabi_fstatat64(int dfd,
 				   int flag)
 {
 	struct kstat stat;
-	int error = -EINVAL;
+	int error;
 
-	if ((flag & ~AT_SYMLINK_NOFOLLOW) != 0)
-		goto out;
-
-	if (flag & AT_SYMLINK_NOFOLLOW)
-		error = vfs_lstat_fd(dfd, filename, &stat);
-	else
-		error = vfs_stat_fd(dfd, filename, &stat);
-
-	if (!error)
-	error = cp_oldabi_stat64(&stat, statbuf);
-
-out:
-	return error;
+	error = vfs_fstatat(dfd, filename, &stat, flag);
+	if (error)
+		return error;
+	return cp_oldabi_stat64(&stat, statbuf);
 }
 
 struct oabi_flock64 {

arch/s390/kernel/compat_linux.c

@@ -702,20 +702,12 @@ asmlinkage long sys32_fstatat64(unsigned int dfd, char __user *filename,
 				struct stat64_emu31 __user* statbuf, int flag)
 {
 	struct kstat stat;
-	int error = -EINVAL;
+	int error;
 
-	if ((flag & ~AT_SYMLINK_NOFOLLOW) != 0)
-		goto out;
-
-	if (flag & AT_SYMLINK_NOFOLLOW)
-		error = vfs_lstat_fd(dfd, filename, &stat);
-	else
-		error = vfs_stat_fd(dfd, filename, &stat);
-
-	if (!error)
-		error = cp_stat64(statbuf, &stat);
-out:
-	return error;
+	error = vfs_fstatat(dfd, filename, &stat, flag);
+	if (error)
+		return error;
+	return cp_stat64(statbuf, &stat);
 }
 
 /*

arch/sparc/kernel/sys_sparc32.c

@@ -206,21 +206,12 @@ asmlinkage long compat_sys_fstatat64(unsigned int dfd, char __user *filename,
 		struct compat_stat64 __user * statbuf, int flag)
 {
 	struct kstat stat;
-	int error = -EINVAL;
+	int error;
 
-	if ((flag & ~AT_SYMLINK_NOFOLLOW) != 0)
-		goto out;
-
-	if (flag & AT_SYMLINK_NOFOLLOW)
-		error = vfs_lstat_fd(dfd, filename, &stat);
-	else
-		error = vfs_stat_fd(dfd, filename, &stat);
-
-	if (!error)
-		error = cp_compat_stat64(&stat, statbuf);
-
-out:
-	return error;
+	error = vfs_fstatat(dfd, filename, &stat, flag);
+	if (error)
+		return error;
+	return cp_compat_stat64(&stat, statbuf);
 }
 
 asmlinkage long compat_sys_sysfs(int option, u32 arg1, u32 arg2)

arch/x86/ia32/sys_ia32.c

@@ -129,21 +129,12 @@ asmlinkage long sys32_fstatat(unsigned int dfd, char __user *filename,
 			      struct stat64 __user *statbuf, int flag)
 {
 	struct kstat stat;
-	int error = -EINVAL;
+	int error;
 
-	if ((flag & ~AT_SYMLINK_NOFOLLOW) != 0)
-		goto out;
-
-	if (flag & AT_SYMLINK_NOFOLLOW)
-		error = vfs_lstat_fd(dfd, filename, &stat);
-	else
-		error = vfs_stat_fd(dfd, filename, &stat);
-
-	if (!error)
-		error = cp_stat64(statbuf, &stat);
-
-out:
-	return error;
+	error = vfs_fstatat(dfd, filename, &stat, flag);
+	if (error)
+		return error;
+	return cp_stat64(statbuf, &stat);
 }
 
 /*

fs/autofs/dirhash.c

@@ -39,10 +39,12 @@ struct autofs_dir_ent *autofs_expire(struct super_block *sb,
 {
 	struct autofs_dirhash *dh = &sbi->dirhash;
 	struct autofs_dir_ent *ent;
-	struct dentry *dentry;
 	unsigned long timeout = sbi->exp_timeout;
 
 	while (1) {
+		struct path path;
+		int umount_ok;
+
 		if ( list_empty(&dh->expiry_head) || sbi->catatonic )
 			return NULL;	/* No entries */
 		/* We keep the list sorted by last_usage and want old stuff */
@@ -57,17 +59,17 @@ struct autofs_dir_ent *autofs_expire(struct super_block *sb,
 			return ent; /* Symlinks are always expirable */
 
 		/* Get the dentry for the autofs subdirectory */
-		dentry = ent->dentry;
+		path.dentry = ent->dentry;
 
-		if ( !dentry ) {
+		if (!path.dentry) {
 			/* Should only happen in catatonic mode */
 			printk("autofs: dentry == NULL but inode range is directory, entry %s\n", ent->name);
 			autofs_delete_usage(ent);
 			continue;
 		}
 
-		if ( !dentry->d_inode ) {
-			dput(dentry);
+		if (!path.dentry->d_inode) {
+			dput(path.dentry);
 			printk("autofs: negative dentry on expiry queue: %s\n",
 			       ent->name);
 			autofs_delete_usage(ent);
@@ -76,29 +78,29 @@ struct autofs_dir_ent *autofs_expire(struct super_block *sb,
 
 		/* Make sure entry is mounted and unused; note that dentry will
 		   point to the mounted-on-top root. */
-		if (!S_ISDIR(dentry->d_inode->i_mode)||!d_mountpoint(dentry)) {
+		if (!S_ISDIR(path.dentry->d_inode->i_mode) ||
+		    !d_mountpoint(path.dentry)) {
 			DPRINTK(("autofs: not expirable (not a mounted directory): %s\n", ent->name));
 			continue;
 		}
-		mntget(mnt);
-		dget(dentry);
-		if (!follow_down(&mnt, &dentry)) {
-			dput(dentry);
-			mntput(mnt);
+		path.mnt = mnt;
+		path_get(&path);
+		if (!follow_down(&path.mnt, &path.dentry)) {
+			path_put(&path);
 			DPRINTK(("autofs: not expirable (not a mounted directory): %s\n", ent->name));
 			continue;
 		}
-		while (d_mountpoint(dentry) && follow_down(&mnt, &dentry))
+		while (d_mountpoint(path.dentry) &&
+		       follow_down(&path.mnt, &path.dentry))
 			;
-		dput(dentry);
+		umount_ok = may_umount(path.mnt);
+		path_put(&path);
 
-		if ( may_umount(mnt) ) {
-			mntput(mnt);
+		if (umount_ok) {
 			DPRINTK(("autofs: signaling expire on %s\n", ent->name));
 			return ent; /* Expirable! */
 		}
 		DPRINTK(("autofs: didn't expire due to may_umount: %s\n", ent->name));
-		mntput(mnt);
 	}
 	return NULL;		/* No expirable entries */
 }

fs/autofs4/dev-ioctl.c

@@ -54,11 +54,10 @@ static int check_name(const char *name)
  * Check a string doesn't overrun the chunk of
  * memory we copied from user land.
  */
-static int invalid_str(char *str, void *end)
+static int invalid_str(char *str, size_t size)
 {
-	while ((void *) str <= end)
-		if (!*str++)
-			return 0;
+	if (memchr(str, 0, size))
+		return 0;
 	return -EINVAL;
 }
 
@@ -138,8 +137,7 @@ static int validate_dev_ioctl(int cmd, struct autofs_dev_ioctl *param)
 	}
 
 	if (param->size > sizeof(*param)) {
-		err = invalid_str(param->path,
-				 (void *) ((size_t) param + param->size));
+		err = invalid_str(param->path, param->size - sizeof(*param));
 		if (err) {
 			AUTOFS_WARN(
 			  "path string terminator missing for cmd(0x%08x)",
@@ -488,7 +486,7 @@ static int autofs_dev_ioctl_requester(struct file *fp,
 	}
 
 	path = param->path;
-	devid = sbi->sb->s_dev;
+	devid = new_encode_dev(sbi->sb->s_dev);
 
 	param->requester.uid = param->requester.gid = -1;
 

fs/btrfs/ioctl.c

@@ -461,15 +461,9 @@ static int btrfs_ioctl_resize(struct btrfs_root *root, void __user *arg)
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
 
-	vol_args = kmalloc(sizeof(*vol_args), GFP_NOFS);
-
-	if (!vol_args)
-		return -ENOMEM;
-
-	if (copy_from_user(vol_args, arg, sizeof(*vol_args))) {
-		ret = -EFAULT;
-		goto out;
-	}
+	vol_args = memdup_user(arg, sizeof(*vol_args));
+	if (IS_ERR(vol_args))
+		return PTR_ERR(vol_args);
 
 	vol_args->name[BTRFS_PATH_NAME_MAX] = '\0';
 	namelen = strlen(vol_args->name);
@@ -545,7 +539,6 @@ static int btrfs_ioctl_resize(struct btrfs_root *root, void __user *arg)
 
 out_unlock:
 	mutex_unlock(&root->fs_info->volume_mutex);
-out:
 	kfree(vol_args);
 	return ret;
 }
@@ -565,15 +558,9 @@ static noinline int btrfs_ioctl_snap_create(struct file *file,
 	if (root->fs_info->sb->s_flags & MS_RDONLY)
 		return -EROFS;
 
-	vol_args = kmalloc(sizeof(*vol_args), GFP_NOFS);
-
-	if (!vol_args)
-		return -ENOMEM;
-
-	if (copy_from_user(vol_args, arg, sizeof(*vol_args))) {
-		ret = -EFAULT;
-		goto out;
-	}
+	vol_args = memdup_user(arg, sizeof(*vol_args));
+	if (IS_ERR(vol_args))
+		return PTR_ERR(vol_args);
 
 	vol_args->name[BTRFS_PATH_NAME_MAX] = '\0';
 	namelen = strlen(vol_args->name);
@@ -675,19 +662,13 @@ static long btrfs_ioctl_add_dev(struct btrfs_root *root, void __user *arg)
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
 
-	vol_args = kmalloc(sizeof(*vol_args), GFP_NOFS);
+	vol_args = memdup_user(arg, sizeof(*vol_args));
+	if (IS_ERR(vol_args))
+		return PTR_ERR(vol_args);
 
-	if (!vol_args)
-		return -ENOMEM;
-
-	if (copy_from_user(vol_args, arg, sizeof(*vol_args))) {
-		ret = -EFAULT;
-		goto out;
-	}
 	vol_args->name[BTRFS_PATH_NAME_MAX] = '\0';
 	ret = btrfs_init_new_device(root, vol_args->name);
 
-out:
 	kfree(vol_args);
 	return ret;
 }
@@ -703,19 +684,13 @@ static long btrfs_ioctl_rm_dev(struct btrfs_root *root, void __user *arg)
 	if (root->fs_info->sb->s_flags & MS_RDONLY)
 		return -EROFS;
 
-	vol_args = kmalloc(sizeof(*vol_args), GFP_NOFS);
+	vol_args = memdup_user(arg, sizeof(*vol_args));
+	if (IS_ERR(vol_args))
+		return PTR_ERR(vol_args);
 
-	if (!vol_args)
-		return -ENOMEM;
-
-	if (copy_from_user(vol_args, arg, sizeof(*vol_args))) {
-		ret = -EFAULT;
-		goto out;
-	}
 	vol_args->name[BTRFS_PATH_NAME_MAX] = '\0';
 	ret = btrfs_rm_device(root, vol_args->name);
 
-out:
 	kfree(vol_args);
 	return ret;
 }

fs/btrfs/super.c

@@ -635,14 +635,9 @@ static long btrfs_control_ioctl(struct file *file, unsigned int cmd,
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
 
-	vol = kmalloc(sizeof(*vol), GFP_KERNEL);
-	if (!vol)
-		return -ENOMEM;
-
-	if (copy_from_user(vol, (void __user *)arg, sizeof(*vol))) {
-		ret = -EFAULT;
-		goto out;
-	}
+	vol = memdup_user((void __user *)arg, sizeof(*vol));
+	if (IS_ERR(vol))
+		return PTR_ERR(vol);
 
 	switch (cmd) {
 	case BTRFS_IOC_SCAN_DEV:
@@ -650,7 +645,7 @@ static long btrfs_control_ioctl(struct file *file, unsigned int cmd,
 					    &btrfs_fs_type, &fs_devices);
 		break;
 	}
-out:
+
 	kfree(vol);
 	return ret;
 }

fs/compat.c

@@ -181,22 +181,24 @@ asmlinkage long compat_sys_newstat(char __user * filename,
 		struct compat_stat __user *statbuf)
 {
 	struct kstat stat;
-	int error = vfs_stat_fd(AT_FDCWD, filename, &stat);
+	int error;
 
-	if (!error)
-		error = cp_compat_stat(&stat, statbuf);
-	return error;
+	error = vfs_stat(filename, &stat);
+	if (error)
+		return error;
+	return cp_compat_stat(&stat, statbuf);
 }
 
 asmlinkage long compat_sys_newlstat(char __user * filename,
 		struct compat_stat __user *statbuf)
 {
 	struct kstat stat;
-	int error = vfs_lstat_fd(AT_FDCWD, filename, &stat);
+	int error;
 
-	if (!error)
-		error = cp_compat_stat(&stat, statbuf);
-	return error;
+	error = vfs_lstat(filename, &stat);
+	if (error)
+		return error;
+	return cp_compat_stat(&stat, statbuf);
 }
 
 #ifndef __ARCH_WANT_STAT64
@@ -204,21 +206,12 @@ asmlinkage long compat_sys_newfstatat(unsigned int dfd, char __user *filename,
 		struct compat_stat __user *statbuf, int flag)
 {
 	struct kstat stat;
-	int error = -EINVAL;
+	int error;
 
-	if ((flag & ~AT_SYMLINK_NOFOLLOW) != 0)
-		goto out;
-
-	if (flag & AT_SYMLINK_NOFOLLOW)
-		error = vfs_lstat_fd(dfd, filename, &stat);
-	else
-		error = vfs_stat_fd(dfd, filename, &stat);
-
-	if (!error)
-		error = cp_compat_stat(&stat, statbuf);
-
-out:
-	return error;
+	error = vfs_fstatat(dfd, filename, &stat, flag);
+	if (error)
+		return error;
+	return cp_compat_stat(&stat, statbuf);
 }
 #endif
 

fs/compat_ioctl.c

@@ -58,7 +58,6 @@
 #include <linux/i2c.h>
 #include <linux/i2c-dev.h>
 #include <linux/atalk.h>
-#include <linux/loop.h>
 
 #include <net/bluetooth/bluetooth.h>
 #include <net/bluetooth/hci.h>
@@ -68,6 +67,7 @@
 #include <linux/gigaset_dev.h>
 
 #ifdef CONFIG_BLOCK
+#include <linux/loop.h>
 #include <scsi/scsi.h>
 #include <scsi/scsi_ioctl.h>
 #include <scsi/sg.h>
@@ -2660,6 +2660,8 @@ HANDLE_IOCTL(SONET_GETFRAMING, do_atm_ioctl)
 HANDLE_IOCTL(SONET_GETFRSENSE, do_atm_ioctl)
 /* block stuff */
 #ifdef CONFIG_BLOCK
+/* loop */
+IGNORE_IOCTL(LOOP_CLR_FD)
 /* Raw devices */
 HANDLE_IOCTL(RAW_SETBIND, raw_ioctl)
 HANDLE_IOCTL(RAW_GETBIND, raw_ioctl)
@@ -2728,9 +2730,6 @@ HANDLE_IOCTL(LPSETTIMEOUT, lp_timeout_trans)
 IGNORE_IOCTL(VFAT_IOCTL_READDIR_BOTH32)
 IGNORE_IOCTL(VFAT_IOCTL_READDIR_SHORT32)
 
-/* loop */
-IGNORE_IOCTL(LOOP_CLR_FD)
-
 #ifdef CONFIG_SPARC
 /* Sparc framebuffers, handled in sbusfb_compat_ioctl() */
 IGNORE_IOCTL(FBIOGTYPE)

fs/dcache.c

@@ -2149,7 +2149,6 @@ int is_subdir(struct dentry *new_dentry, struct dentry *old_dentry)
 	int result;
 	unsigned long seq;
 
-	/* FIXME: This is old behavior, needed? Please check callers. */
 	if (new_dentry == old_dentry)
 		return 1;
 

fs/ecryptfs/miscdev.c

@@ -418,18 +418,13 @@ ecryptfs_miscdev_write(struct file *file, const char __user *buf,
 
 	if (count == 0)
 		goto out;
-	data = kmalloc(count, GFP_KERNEL);
-	if (!data) {
-		printk(KERN_ERR "%s: Out of memory whilst attempting to "
-		       "kmalloc([%zd], GFP_KERNEL)\n", __func__, count);
+
+	data = memdup_user(buf, count);
+	if (IS_ERR(data)) {
+		printk(KERN_ERR "%s: memdup_user returned error [%ld]\n",
+		       __func__, PTR_ERR(data));
 		goto out;
 	}
-	rc = copy_from_user(data, buf, count);
-	if (rc) {
-		printk(KERN_ERR "%s: copy_from_user returned error [%d]\n",
-		       __func__, rc);
-		goto out_free;
-	}
 	sz = count;
 	i = 0;
 	switch (data[i++]) {

fs/filesystems.c

@@ -199,7 +199,7 @@ SYSCALL_DEFINE3(sysfs, int, option, unsigned long, arg1, unsigned long, arg2)
 	return retval;
 }
 
-int get_filesystem_list(char * buf)
+int __init get_filesystem_list(char *buf)
 {
 	int len = 0;
 	struct file_system_type * tmp;

fs/namei.c

@@ -1248,6 +1248,8 @@ struct dentry *lookup_one_len(const char *name, struct dentry *base, int len)
 	int err;
 	struct qstr this;
 
+	WARN_ON_ONCE(!mutex_is_locked(&base->d_inode->i_mutex));
+
 	err = __lookup_one_len(name, &this, base, len);
 	if (err)
 		return ERR_PTR(err);

fs/namespace.c

@@ -1377,7 +1377,7 @@ static int attach_recursive_mnt(struct vfsmount *source_mnt,
 	if (parent_path) {
 		detach_mnt(source_mnt, parent_path);
 		attach_mnt(source_mnt, path);
-		touch_mnt_namespace(current->nsproxy->mnt_ns);
+		touch_mnt_namespace(parent_path->mnt->mnt_ns);
 	} else {
 		mnt_set_mountpoint(dest_mnt, dest_dentry, source_mnt);
 		commit_tree(source_mnt);

fs/ncpfs/ioctl.c

@@ -660,13 +660,10 @@ outrel:
 			if (user.object_name_len > NCP_OBJECT_NAME_MAX_LEN)
 				return -ENOMEM;
 			if (user.object_name_len) {
-				newname = kmalloc(user.object_name_len, GFP_USER);
-				if (!newname)
-					return -ENOMEM;
-				if (copy_from_user(newname, user.object_name, user.object_name_len)) {
-					kfree(newname);
-					return -EFAULT;
-				}
+				newname = memdup_user(user.object_name,
+						      user.object_name_len);
+				if (IS_ERR(newname))
+					return PTR_ERR(newname);
 			} else {
 				newname = NULL;
 			}
@@ -760,13 +757,9 @@ outrel:
 			if (user.len > NCP_PRIVATE_DATA_MAX_LEN)
 				return -ENOMEM;
 			if (user.len) {
-				new = kmalloc(user.len, GFP_USER);
-				if (!new)
-					return -ENOMEM;
-				if (copy_from_user(new, user.data, user.len)) {
-					kfree(new);
-					return -EFAULT;
-				}
+				new = memdup_user(user.data, user.len);
+				if (IS_ERR(new))
+					return PTR_ERR(new);
 			} else {
 				new = NULL;
 			}

fs/nfsd/nfs4recover.c

@@ -229,21 +229,23 @@ nfsd4_list_rec_dir(struct dentry *dir, recdir_func *f)
 		goto out;
 	status = vfs_readdir(filp, nfsd4_build_namelist, &names);
 	fput(filp);
+	mutex_lock(&dir->d_inode->i_mutex);
 	while (!list_empty(&names)) {
 		entry = list_entry(names.next, struct name_list, list);
 
 		dentry = lookup_one_len(entry->name, dir, HEXDIR_LEN-1);
 		if (IS_ERR(dentry)) {
 			status = PTR_ERR(dentry);
-			goto out;
+			break;
 		}
 		status = f(dir, dentry);
 		dput(dentry);
 		if (status)
-			goto out;
+			break;
 		list_del(&entry->list);
 		kfree(entry);
 	}
+	mutex_unlock(&dir->d_inode->i_mutex);
 out:
 	while (!list_empty(&names)) {
 		entry = list_entry(names.next, struct name_list, list);
@@ -254,36 +256,6 @@ out:
 	return status;
 }
 
-static int
-nfsd4_remove_clid_file(struct dentry *dir, struct dentry *dentry)
-{
-	int status;
-
-	if (!S_ISREG(dir->d_inode->i_mode)) {
-		printk("nfsd4: non-file found in client recovery directory\n");
-		return -EINVAL;
-	}
-	mutex_lock_nested(&dir->d_inode->i_mutex, I_MUTEX_PARENT);
-	status = vfs_unlink(dir->d_inode, dentry);
-	mutex_unlock(&dir->d_inode->i_mutex);
-	return status;
-}
-
-static int
-nfsd4_clear_clid_dir(struct dentry *dir, struct dentry *dentry)
-{
-	int status;
-
-	/* For now this directory should already be empty, but we empty it of
-	 * any regular files anyway, just in case the directory was created by
-	 * a kernel from the future.... */
-	nfsd4_list_rec_dir(dentry, nfsd4_remove_clid_file);
-	mutex_lock_nested(&dir->d_inode->i_mutex, I_MUTEX_PARENT);
-	status = vfs_rmdir(dir->d_inode, dentry);
-	mutex_unlock(&dir->d_inode->i_mutex);
-	return status;
-}
-
 static int
 nfsd4_unlink_clid_dir(char *name, int namlen)
 {
@@ -294,18 +266,18 @@ nfsd4_unlink_clid_dir(char *name, int namlen)
 
 	mutex_lock(&rec_dir.dentry->d_inode->i_mutex);
 	dentry = lookup_one_len(name, rec_dir.dentry, namlen);
-	mutex_unlock(&rec_dir.dentry->d_inode->i_mutex);
 	if (IS_ERR(dentry)) {
 		status = PTR_ERR(dentry);
-		return status;
+		goto out_unlock;
 	}
 	status = -ENOENT;
 	if (!dentry->d_inode)
 		goto out;
-
-	status = nfsd4_clear_clid_dir(rec_dir.dentry, dentry);
+	status = vfs_rmdir(rec_dir.dentry->d_inode, dentry);
 out:
 	dput(dentry);
+out_unlock:
+	mutex_unlock(&rec_dir.dentry->d_inode->i_mutex);
 	return status;
 }
 
@@ -348,7 +320,7 @@ purge_old(struct dentry *parent, struct dentry *child)
 	if (nfs4_has_reclaimed_state(child->d_name.name, false))
 		return 0;
 
-	status = nfsd4_clear_clid_dir(parent, child);
+	status = vfs_rmdir(parent->d_inode, child);
 	if (status)
 		printk("failed to remove client recovery directory %s\n",
 				child->d_name.name);

fs/nfsd/vfs.c

@@ -116,10 +116,15 @@ nfsd_cross_mnt(struct svc_rqst *rqstp, struct dentry **dpp,
 	}
 	if ((exp->ex_flags & NFSEXP_CROSSMOUNT) || EX_NOHIDE(exp2)) {
 		/* successfully crossed mount point */
-		exp_put(exp);
-		*expp = exp2;
+		/*
+		 * This is subtle: dentry is *not* under mnt at this point.
+		 * The only reason we are safe is that original mnt is pinned
+		 * down by exp, so we should dput before putting exp.
+		 */
 		dput(dentry);
 		*dpp = mounts;
+		exp_put(exp);
+		*expp = exp2;
 	} else {
 		exp_put(exp2);
 		dput(mounts);
@@ -1885,8 +1890,8 @@ static int nfsd_buffered_filldir(void *__buf, const char *name, int namlen,
 	return 0;
 }
 
-static int nfsd_buffered_readdir(struct file *file, filldir_t func,
-				 struct readdir_cd *cdp, loff_t *offsetp)
+static __be32 nfsd_buffered_readdir(struct file *file, filldir_t func,
+				    struct readdir_cd *cdp, loff_t *offsetp)
 {
 	struct readdir_data buf;
 	struct buffered_dirent *de;
@@ -1896,11 +1901,12 @@ static int nfsd_buffered_readdir(struct file *file, filldir_t func,
 
 	buf.dirent = (void *)__get_free_page(GFP_KERNEL);
 	if (!buf.dirent)
-		return -ENOMEM;
+		return nfserrno(-ENOMEM);
 
 	offset = *offsetp;
 
 	while (1) {
+		struct inode *dir_inode = file->f_path.dentry->d_inode;
 		unsigned int reclen;
 
 		cdp->err = nfserr_eof; /* will be cleared on successful read */
@@ -1919,26 +1925,38 @@ static int nfsd_buffered_readdir(struct file *file, filldir_t func,
 		if (!size)
 			break;
 
+		/*
+		 * Various filldir functions may end up calling back into
+		 * lookup_one_len() and the file system's ->lookup() method.
+		 * These expect i_mutex to be held, as it would within readdir.
+		 */
+		host_err = mutex_lock_killable(&dir_inode->i_mutex);
+		if (host_err)
+			break;
+
 		de = (struct buffered_dirent *)buf.dirent;
 		while (size > 0) {
 			offset = de->offset;
 
 			if (func(cdp, de->name, de->namlen, de->offset,
 				 de->ino, de->d_type))
-				goto done;
+				break;
 
 			if (cdp->err != nfs_ok)
-				goto done;
+				break;
 
 			reclen = ALIGN(sizeof(*de) + de->namlen,
 				       sizeof(u64));
 			size -= reclen;
 			de = (struct buffered_dirent *)((char *)de + reclen);
 		}
+		mutex_unlock(&dir_inode->i_mutex);
+		if (size > 0) /* We bailed out early */
+			break;
+
 		offset = vfs_llseek(file, 0, SEEK_CUR);
 	}
 
- done:
 	free_page((unsigned long)(buf.dirent));
 
 	if (host_err)

fs/stat.c

@@ -55,46 +55,6 @@ int vfs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat)
 
 EXPORT_SYMBOL(vfs_getattr);
 
-int vfs_stat_fd(int dfd, char __user *name, struct kstat *stat)
-{
-	struct path path;
-	int error;
-
-	error = user_path_at(dfd, name, LOOKUP_FOLLOW, &path);
-	if (!error) {
-		error = vfs_getattr(path.mnt, path.dentry, stat);
-		path_put(&path);
-	}
-	return error;
-}
-
-int vfs_stat(char __user *name, struct kstat *stat)
-{
-	return vfs_stat_fd(AT_FDCWD, name, stat);
-}
-
-EXPORT_SYMBOL(vfs_stat);
-
-int vfs_lstat_fd(int dfd, char __user *name, struct kstat *stat)
-{
-	struct path path;
-	int error;
-
-	error = user_path_at(dfd, name, 0, &path);
-	if (!error) {
-		error = vfs_getattr(path.mnt, path.dentry, stat);
-		path_put(&path);
-	}
-	return error;
-}
-
-int vfs_lstat(char __user *name, struct kstat *stat)
-{
-	return vfs_lstat_fd(AT_FDCWD, name, stat);
-}
-
-EXPORT_SYMBOL(vfs_lstat);
-
 int vfs_fstat(unsigned int fd, struct kstat *stat)
 {
 	struct file *f = fget(fd);
@@ -106,9 +66,44 @@ int vfs_fstat(unsigned int fd, struct kstat *stat)
 	}
 	return error;
 }
-
 EXPORT_SYMBOL(vfs_fstat);
 
+int vfs_fstatat(int dfd, char __user *filename, struct kstat *stat, int flag)
+{
+	struct path path;
+	int error = -EINVAL;
+	int lookup_flags = 0;
+
+	if ((flag & ~AT_SYMLINK_NOFOLLOW) != 0)
+		goto out;
+
+	if (!(flag & AT_SYMLINK_NOFOLLOW))
+		lookup_flags |= LOOKUP_FOLLOW;
+
+	error = user_path_at(dfd, filename, lookup_flags, &path);
+	if (error)
+		goto out;
+
+	error = vfs_getattr(path.mnt, path.dentry, stat);
+	path_put(&path);
+out:
+	return error;
+}
+EXPORT_SYMBOL(vfs_fstatat);
+
+int vfs_stat(char __user *name, struct kstat *stat)
+{
+	return vfs_fstatat(AT_FDCWD, name, stat, 0);
+}
+EXPORT_SYMBOL(vfs_stat);
+
+int vfs_lstat(char __user *name, struct kstat *stat)
+{
+	return vfs_fstatat(AT_FDCWD, name, stat, AT_SYMLINK_NOFOLLOW);
+}
+EXPORT_SYMBOL(vfs_lstat);
+
+
 #ifdef __ARCH_WANT_OLD_STAT
 
 /*
@@ -155,23 +150,25 @@ static int cp_old_stat(struct kstat *stat, struct __old_kernel_stat __user * sta
 SYSCALL_DEFINE2(stat, char __user *, filename, struct __old_kernel_stat __user *, statbuf)
 {
 	struct kstat stat;
-	int error = vfs_stat_fd(AT_FDCWD, filename, &stat);
+	int error;
 
-	if (!error)
-		error = cp_old_stat(&stat, statbuf);
+	error = vfs_stat(filename, &stat);
+	if (error)
+		return error;
 
-	return error;
+	return cp_old_stat(&stat, statbuf);
 }
 
 SYSCALL_DEFINE2(lstat, char __user *, filename, struct __old_kernel_stat __user *, statbuf)
 {
 	struct kstat stat;
-	int error = vfs_lstat_fd(AT_FDCWD, filename, &stat);
+	int error;
 
-	if (!error)
-		error = cp_old_stat(&stat, statbuf);
+	error = vfs_lstat(filename, &stat);
+	if (error)
+		return error;
 
-	return error;
+	return cp_old_stat(&stat, statbuf);
 }
 
 SYSCALL_DEFINE2(fstat, unsigned int, fd, struct __old_kernel_stat __user *, statbuf)
@@ -240,23 +237,23 @@ static int cp_new_stat(struct kstat *stat, struct stat __user *statbuf)
 SYSCALL_DEFINE2(newstat, char __user *, filename, struct stat __user *, statbuf)
 {
 	struct kstat stat;
-	int error = vfs_stat_fd(AT_FDCWD, filename, &stat);
+	int error = vfs_stat(filename, &stat);
 
-	if (!error)
-		error = cp_new_stat(&stat, statbuf);
-
-	return error;
+	if (error)
+		return error;
+	return cp_new_stat(&stat, statbuf);
 }
 
 SYSCALL_DEFINE2(newlstat, char __user *, filename, struct stat __user *, statbuf)
 {
 	struct kstat stat;
-	int error = vfs_lstat_fd(AT_FDCWD, filename, &stat);
+	int error;
 
-	if (!error)
-		error = cp_new_stat(&stat, statbuf);
+	error = vfs_lstat(filename, &stat);
+	if (error)
+		return error;
 
-	return error;
+	return cp_new_stat(&stat, statbuf);
 }
 
 #if !defined(__ARCH_WANT_STAT64) || defined(__ARCH_WANT_SYS_NEWFSTATAT)
@@ -264,21 +261,12 @@ SYSCALL_DEFINE4(newfstatat, int, dfd, char __user *, filename,
 		struct stat __user *, statbuf, int, flag)
 {
 	struct kstat stat;
-	int error = -EINVAL;
+	int error;
 
-	if ((flag & ~AT_SYMLINK_NOFOLLOW) != 0)
-		goto out;
-
-	if (flag & AT_SYMLINK_NOFOLLOW)
-		error = vfs_lstat_fd(dfd, filename, &stat);
-	else
-		error = vfs_stat_fd(dfd, filename, &stat);
-
-	if (!error)
-		error = cp_new_stat(&stat, statbuf);
-
-out:
-	return error;
+	error = vfs_fstatat(dfd, filename, &stat, flag);
+	if (error)
+		return error;
+	return cp_new_stat(&stat, statbuf);
 }
 #endif
 
@@ -404,21 +392,12 @@ SYSCALL_DEFINE4(fstatat64, int, dfd, char __user *, filename,
 		struct stat64 __user *, statbuf, int, flag)
 {
 	struct kstat stat;
-	int error = -EINVAL;
+	int error;
 
-	if ((flag & ~AT_SYMLINK_NOFOLLOW) != 0)
-		goto out;
-
-	if (flag & AT_SYMLINK_NOFOLLOW)
-		error = vfs_lstat_fd(dfd, filename, &stat);
-	else
-		error = vfs_stat_fd(dfd, filename, &stat);
-
-	if (!error)
-		error = cp_new_stat64(&stat, statbuf);
-
-out:
-	return error;
+	error = vfs_fstatat(dfd, filename, &stat, flag);
+	if (error)
+		return error;
+	return cp_new_stat64(&stat, statbuf);
 }
 #endif /* __ARCH_WANT_STAT64 */