Merge branch 'work.misc' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs

Pull misc vfs updates from Al Viro:
"Assorted stuff all over the place (the largest group here is
Christoph's stat cleanups)"

* 'work.misc' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
fs: remove KSTAT_QUERY_FLAGS
fs: remove vfs_stat_set_lookup_flags
fs: move vfs_fstatat out of line
fs: implement vfs_stat and vfs_lstat in terms of vfs_fstatat
fs: remove vfs_statx_fd
fs: omfs: use kmemdup() rather than kmalloc+memcpy
[PATCH] reduce boilerplate in fsid handling
fs: Remove duplicated flag O_NDELAY occurring twice in VALID_OPEN_FLAGS
selftests: mount: add nosymfollow tests
Add a "nosymfollow" mount option.

fs/9p/vfs_super.c

···
 			buf->f_bavail = rs.bavail;
 			buf->f_files = rs.files;
 			buf->f_ffree = rs.ffree;
-			buf->f_fsid.val[0] = rs.fsid & 0xFFFFFFFFUL;
-			buf->f_fsid.val[1] = (rs.fsid >> 32) & 0xFFFFFFFFUL;
+			buf->f_fsid = u64_to_fsid(rs.fsid);
 			buf->f_namelen = rs.namelen;
 		}
 		if (res != -ENOSYS)

fs/adfs/super.c

···
 	buf->f_namelen = sbi->s_namelen;
 	buf->f_bsize   = sb->s_blocksize;
 	buf->f_ffree   = (long)(buf->f_bfree * buf->f_files) / (long)buf->f_blocks;
-	buf->f_fsid.val[0] = (u32)id;
-	buf->f_fsid.val[1] = (u32)(id >> 32);
+	buf->f_fsid    = u64_to_fsid(id);
 
 	return 0;
 }

fs/affs/super.c

···
 	buf->f_blocks  = AFFS_SB(sb)->s_partition_size - AFFS_SB(sb)->s_reserved;
 	buf->f_bfree   = free;
 	buf->f_bavail  = free;
-	buf->f_fsid.val[0] = (u32)id;
-	buf->f_fsid.val[1] = (u32)(id >> 32);
+	buf->f_fsid    = u64_to_fsid(id);
 	buf->f_namelen = AFFSNAMEMAX;
 	return 0;
 }

fs/befs/linuxvfs.c

···
 	buf->f_bavail = buf->f_bfree;
 	buf->f_files = 0;	/* UNKNOWN */
 	buf->f_ffree = 0;	/* UNKNOWN */
-	buf->f_fsid.val[0] = (u32)id;
-	buf->f_fsid.val[1] = (u32)(id >> 32);
+	buf->f_fsid = u64_to_fsid(id);
 	buf->f_namelen = BEFS_NAME_LEN;
 
 	befs_debug(sb, "<--- %s", __func__);

fs/bfs/inode.c

···
 	buf->f_bfree = buf->f_bavail = info->si_freeb;
 	buf->f_files = info->si_lasti + 1 - BFS_ROOT_INO;
 	buf->f_ffree = info->si_freei;
-	buf->f_fsid.val[0] = (u32)id;
-	buf->f_fsid.val[1] = (u32)(id >> 32);
+	buf->f_fsid = u64_to_fsid(id);
 	buf->f_namelen = BFS_NAMELEN;
 	return 0;
 }

fs/ceph/super.c

···
 	       le64_to_cpu(*((__le64 *)&monc->monmap->fsid + 1));
 	mutex_unlock(&monc->mutex);
 
-	buf->f_fsid.val[0] = fsid & 0xffffffff;
-	buf->f_fsid.val[1] = fsid >> 32;
+	buf->f_fsid = u64_to_fsid(fsid);
 
 	return 0;
 }

fs/cramfs/inode.c

···
 	buf->f_bavail = 0;
 	buf->f_files = CRAMFS_SB(sb)->files;
 	buf->f_ffree = 0;
-	buf->f_fsid.val[0] = (u32)id;
-	buf->f_fsid.val[1] = (u32)(id >> 32);
+	buf->f_fsid = u64_to_fsid(id);
 	buf->f_namelen = CRAMFS_MAXPATHLEN;
 	return 0;
 }

fs/efs/super.c

···
 			sbi->inode_blocks *
 			(EFS_BLOCKSIZE / sizeof(struct efs_dinode));
 	buf->f_ffree   = sbi->inode_free;	/* free inodes */
-	buf->f_fsid.val[0] = (u32)id;
-	buf->f_fsid.val[1] = (u32)(id >> 32);
+	buf->f_fsid    = u64_to_fsid(id);
 	buf->f_namelen = EFS_MAXNAMELEN;	/* max filename length */
 
 	return 0;

fs/erofs/super.c

···
 
 	buf->f_namelen = EROFS_NAME_LEN;
 
-	buf->f_fsid.val[0] = (u32)id;
-	buf->f_fsid.val[1] = (u32)(id >> 32);
+	buf->f_fsid    = u64_to_fsid(id);
 	return 0;
 }
 

fs/exfat/super.c

···
 	buf->f_blocks = sbi->num_clusters - 2; /* clu 0 & 1 */
 	buf->f_bfree = buf->f_blocks - sbi->used_clusters;
 	buf->f_bavail = buf->f_bfree;
-	buf->f_fsid.val[0] = (unsigned int)id;
-	buf->f_fsid.val[1] = (unsigned int)(id >> 32);
+	buf->f_fsid = u64_to_fsid(id);
 	/* Unicode utf16 255 characters */
 	buf->f_namelen = EXFAT_MAX_FILE_LEN * NLS_MAX_CHARSET_SIZE;
 	return 0;

fs/ext2/super.c

···
 	buf->f_namelen = EXT2_NAME_LEN;
 	fsid = le64_to_cpup((void *)es->s_uuid) ^
 	       le64_to_cpup((void *)es->s_uuid + sizeof(u64));
-	buf->f_fsid.val[0] = fsid & 0xFFFFFFFFUL;
-	buf->f_fsid.val[1] = (fsid >> 32) & 0xFFFFFFFFUL;
+	buf->f_fsid = u64_to_fsid(fsid);
 	spin_unlock(&sbi->s_lock);
 	return 0;
 }

fs/ext4/super.c

···
 	buf->f_namelen = EXT4_NAME_LEN;
 	fsid = le64_to_cpup((void *)es->s_uuid) ^
 	       le64_to_cpup((void *)es->s_uuid + sizeof(u64));
-	buf->f_fsid.val[0] = fsid & 0xFFFFFFFFUL;
-	buf->f_fsid.val[1] = (fsid >> 32) & 0xFFFFFFFFUL;
+	buf->f_fsid = u64_to_fsid(fsid);
 
 #ifdef CONFIG_QUOTA
 	if (ext4_test_inode_flag(dentry->d_inode, EXT4_INODE_PROJINHERIT) &&

fs/f2fs/super.c

···
 	}
 
 	buf->f_namelen = F2FS_NAME_LEN;
-	buf->f_fsid.val[0] = (u32)id;
-	buf->f_fsid.val[1] = (u32)(id >> 32);
+	buf->f_fsid    = u64_to_fsid(id);
 
 #ifdef CONFIG_QUOTA
 	if (is_inode_flag_set(dentry->d_inode, FI_PROJ_INHERIT) &&

fs/fat/inode.c

···
 	buf->f_blocks = sbi->max_cluster - FAT_START_ENT;
 	buf->f_bfree = sbi->free_clusters;
 	buf->f_bavail = sbi->free_clusters;
-	buf->f_fsid.val[0] = (u32)id;
-	buf->f_fsid.val[1] = (u32)(id >> 32);
+	buf->f_fsid = u64_to_fsid(id);
 	buf->f_namelen =
 		(sbi->options.isvfat ? FAT_LFN_LEN : 12) * NLS_MAX_CHARSET_SIZE;
 

fs/hfs/super.c

···
 	buf->f_bavail = buf->f_bfree;
 	buf->f_files = HFS_SB(sb)->fs_ablocks;
 	buf->f_ffree = HFS_SB(sb)->free_ablocks;
-	buf->f_fsid.val[0] = (u32)id;
-	buf->f_fsid.val[1] = (u32)(id >> 32);
+	buf->f_fsid = u64_to_fsid(id);
 	buf->f_namelen = HFS_NAMELEN;
 
 	return 0;

fs/hfsplus/super.c

···
 	buf->f_bavail = buf->f_bfree;
 	buf->f_files = 0xFFFFFFFF;
 	buf->f_ffree = 0xFFFFFFFF - sbi->next_cnid;
-	buf->f_fsid.val[0] = (u32)id;
-	buf->f_fsid.val[1] = (u32)(id >> 32);
+	buf->f_fsid = u64_to_fsid(id);
 	buf->f_namelen = HFSPLUS_MAX_STRLEN;
 
 	return 0;

fs/hpfs/super.c

···
 	buf->f_bavail = sbi->sb_n_free;
 	buf->f_files = sbi->sb_dirband_size / 4;
 	buf->f_ffree = hpfs_get_free_dnodes(s);
-	buf->f_fsid.val[0] = (u32)id;
-	buf->f_fsid.val[1] = (u32)(id >> 32);
+	buf->f_fsid = u64_to_fsid(id);
 	buf->f_namelen = 254;
 
 	hpfs_unlock(s);

fs/isofs/inode.c

···
 	buf->f_bavail = 0;
 	buf->f_files = ISOFS_SB(sb)->s_ninodes;
 	buf->f_ffree = 0;
-	buf->f_fsid.val[0] = (u32)id;
-	buf->f_fsid.val[1] = (u32)(id >> 32);
+	buf->f_fsid = u64_to_fsid(id);
 	buf->f_namelen = NAME_MAX;
 	return 0;
 }

fs/minix/inode.c

···
 	buf->f_files = sbi->s_ninodes;
 	buf->f_ffree = minix_count_free_inodes(sb);
 	buf->f_namelen = sbi->s_namelen;
-	buf->f_fsid.val[0] = (u32)id;
-	buf->f_fsid.val[1] = (u32)(id >> 32);
+	buf->f_fsid = u64_to_fsid(id);
 
 	return 0;
 }

fs/namei.c

···
 			return ERR_PTR(error);
 	}
 
-	if (unlikely(nd->flags & LOOKUP_NO_SYMLINKS))
+	if (unlikely(nd->flags & LOOKUP_NO_SYMLINKS) ||
+			unlikely(link->mnt->mnt_flags & MNT_NOSYMFOLLOW))
 		return ERR_PTR(-ELOOP);
 
 	if (!(nd->flags & LOOKUP_RCU)) {

fs/namespace.c

···
 		mnt_flags &= ~(MNT_RELATIME | MNT_NOATIME);
 	if (flags & MS_RDONLY)
 		mnt_flags |= MNT_READONLY;
+	if (flags & MS_NOSYMFOLLOW)
+		mnt_flags |= MNT_NOSYMFOLLOW;
 
 	/* The default atime for remount is preservation */
 	if ((flags & MS_REMOUNT) &&

fs/nilfs2/super.c

···
 	buf->f_files = nmaxinodes;
 	buf->f_ffree = nfreeinodes;
 	buf->f_namelen = NILFS_NAME_LEN;
-	buf->f_fsid.val[0] = (u32)id;
-	buf->f_fsid.val[1] = (u32)(id >> 32);
+	buf->f_fsid = u64_to_fsid(id);
 
 	return 0;
 }

fs/ntfs/super.c

···
 	 * the least significant 32-bits in f_fsid[0] and the most significant
 	 * 32-bits in f_fsid[1].
 	 */
-	sfs->f_fsid.val[0] = vol->serial_no & 0xffffffff;
-	sfs->f_fsid.val[1] = (vol->serial_no >> 32) & 0xffffffff;
+	sfs->f_fsid = u64_to_fsid(vol->serial_no);
 	/* Maximum length of filenames. */
 	sfs->f_namelen	   = NTFS_MAX_NAME_LEN;
 	return 0;

fs/omfs/inode.c

···
 	buf->f_blocks = sbi->s_num_blocks;
 	buf->f_files = sbi->s_num_blocks;
 	buf->f_namelen = OMFS_NAMELEN;
-	buf->f_fsid.val[0] = (u32)id;
-	buf->f_fsid.val[1] = (u32)(id >> 32);
+	buf->f_fsid = u64_to_fsid(id);
 
 	buf->f_bfree = buf->f_bavail = buf->f_ffree =
 		omfs_count_free(s);
···
 		bh = sb_bread(sb, block++);
 		if (!bh)
 			goto nomem_free;
-		*ptr = kmalloc(sb->s_blocksize, GFP_KERNEL);
+		*ptr = kmemdup(bh->b_data, sb->s_blocksize, GFP_KERNEL);
 		if (!*ptr) {
 			brelse(bh);
 			goto nomem_free;
 		}
-		memcpy(*ptr, bh->b_data, sb->s_blocksize);
 		if (count < sb->s_blocksize)
 			memset((void *)*ptr + count, 0xff,
 				sb->s_blocksize - count);

fs/proc_namespace.c

···
 		{ MNT_NOATIME, ",noatime" },
 		{ MNT_NODIRATIME, ",nodiratime" },
 		{ MNT_RELATIME, ",relatime" },
+		{ MNT_NOSYMFOLLOW, ",nosymfollow" },
 		{ 0, NULL }
 	};
 	const struct proc_fs_opts *fs_infop;

fs/qnx4/inode.c

···
 	buf->f_bfree   = qnx4_count_free_blocks(sb);
 	buf->f_bavail  = buf->f_bfree;
 	buf->f_namelen = QNX4_NAME_MAX;
-	buf->f_fsid.val[0] = (u32)id;
-	buf->f_fsid.val[1] = (u32)(id >> 32);
+	buf->f_fsid    = u64_to_fsid(id);
 
 	return 0;
 }

fs/qnx6/inode.c

···
 	buf->f_ffree   = fs32_to_cpu(sbi, sbi->sb->sb_free_inodes);
 	buf->f_bavail  = buf->f_bfree;
 	buf->f_namelen = QNX6_LONG_NAME_MAX;
-	buf->f_fsid.val[0] = (u32)id;
-	buf->f_fsid.val[1] = (u32)(id >> 32);
+	buf->f_fsid    = u64_to_fsid(id);
 
 	return 0;
 }

fs/romfs/super.c

···
 	buf->f_bfree = buf->f_bavail = buf->f_ffree;
 	buf->f_blocks =
 		(romfs_maxsize(dentry->d_sb) + ROMBSIZE - 1) >> ROMBSBITS;
-	buf->f_fsid.val[0] = (u32)id;
-	buf->f_fsid.val[1] = (u32)(id >> 32);
+	buf->f_fsid = u64_to_fsid(id);
 	return 0;
 }
 

fs/squashfs/super.c

···
 	buf->f_files = msblk->inodes;
 	buf->f_ffree = 0;
 	buf->f_namelen = SQUASHFS_NAME_LEN;
-	buf->f_fsid.val[0] = (u32)id;
-	buf->f_fsid.val[1] = (u32)(id >> 32);
+	buf->f_fsid = u64_to_fsid(id);
 
 	return 0;
 }

fs/stat.c

···
  * @path: file to get attributes from
  * @stat: structure to return attributes in
  * @request_mask: STATX_xxx flags indicating what the caller wants
- * @query_flags: Query mode (KSTAT_QUERY_FLAGS)
+ * @query_flags: Query mode (AT_STATX_SYNC_TYPE)
  *
  * Get attributes without calling security_inode_getattr.
  *
···
 
 	memset(stat, 0, sizeof(*stat));
 	stat->result_mask |= STATX_BASIC_STATS;
-	query_flags &= KSTAT_QUERY_FLAGS;
+	query_flags &= AT_STATX_SYNC_TYPE;
 
 	/* allow the fs to override these if it really wants to */
 	/* SB_NOATIME means filesystem supplies dummy atime value */
···
  * @path: The file of interest
  * @stat: Where to return the statistics
  * @request_mask: STATX_xxx flags indicating what the caller wants
- * @query_flags: Query mode (KSTAT_QUERY_FLAGS)
+ * @query_flags: Query mode (AT_STATX_SYNC_TYPE)
  *
  * Ask the filesystem for a file's attributes.  The caller must indicate in
  * request_mask and query_flags to indicate what they want.
···
 EXPORT_SYMBOL(vfs_getattr);
 
 /**
- * vfs_statx_fd - Get the enhanced basic attributes by file descriptor
+ * vfs_fstat - Get the basic attributes by file descriptor
  * @fd: The file descriptor referring to the file of interest
  * @stat: The result structure to fill in.
- * @request_mask: STATX_xxx flags indicating what the caller wants
- * @query_flags: Query mode (KSTAT_QUERY_FLAGS)
  *
  * This function is a wrapper around vfs_getattr().  The main difference is
  * that it uses a file descriptor to determine the file location.
  *
  * 0 will be returned on success, and a -ve error code if unsuccessful.
  */
-int vfs_statx_fd(unsigned int fd, struct kstat *stat,
-		 u32 request_mask, unsigned int query_flags)
+int vfs_fstat(int fd, struct kstat *stat)
 {
 	struct fd f;
-	int error = -EBADF;
-
-	if (query_flags & ~KSTAT_QUERY_FLAGS)
-		return -EINVAL;
+	int error;
 
 	f = fdget_raw(fd);
-	if (f.file) {
-		error = vfs_getattr(&f.file->f_path, stat,
-				    request_mask, query_flags);
-		fdput(f);
-	}
+	if (!f.file)
+		return -EBADF;
+	error = vfs_getattr(&f.file->f_path, stat, STATX_BASIC_STATS, 0);
+	fdput(f);
 	return error;
-}
-EXPORT_SYMBOL(vfs_statx_fd);
-
-static inline unsigned vfs_stat_set_lookup_flags(unsigned *lookup_flags,
-						 int flags)
-{
-	if ((flags & ~(AT_SYMLINK_NOFOLLOW | AT_NO_AUTOMOUNT |
-		       AT_EMPTY_PATH | KSTAT_QUERY_FLAGS)) != 0)
-		return -EINVAL;
-
-	*lookup_flags = LOOKUP_FOLLOW | LOOKUP_AUTOMOUNT;
-	if (flags & AT_SYMLINK_NOFOLLOW)
-		*lookup_flags &= ~LOOKUP_FOLLOW;
-	if (flags & AT_NO_AUTOMOUNT)
-		*lookup_flags &= ~LOOKUP_AUTOMOUNT;
-	if (flags & AT_EMPTY_PATH)
-		*lookup_flags |= LOOKUP_EMPTY;
-
-	return 0;
 }
 
 /**
···
  *
  * 0 will be returned on success, and a -ve error code if unsuccessful.
  */
-int vfs_statx(int dfd, const char __user *filename, int flags,
+static int vfs_statx(int dfd, const char __user *filename, int flags,
 	      struct kstat *stat, u32 request_mask)
 {
 	struct path path;
-	int error = -EINVAL;
-	unsigned lookup_flags;
+	unsigned lookup_flags = 0;
+	int error;
 
-	if (vfs_stat_set_lookup_flags(&lookup_flags, flags))
+	if (flags & ~(AT_SYMLINK_NOFOLLOW | AT_NO_AUTOMOUNT | AT_EMPTY_PATH |
+		      AT_STATX_SYNC_TYPE))
 		return -EINVAL;
+
+	if (!(flags & AT_SYMLINK_NOFOLLOW))
+		lookup_flags |= LOOKUP_FOLLOW;
+	if (!(flags & AT_NO_AUTOMOUNT))
+		lookup_flags |= LOOKUP_AUTOMOUNT;
+	if (flags & AT_EMPTY_PATH)
+		lookup_flags |= LOOKUP_EMPTY;
+
 retry:
 	error = user_path_at(dfd, filename, lookup_flags, &path);
 	if (error)
···
 out:
 	return error;
 }
-EXPORT_SYMBOL(vfs_statx);
 
+int vfs_fstatat(int dfd, const char __user *filename,
+			      struct kstat *stat, int flags)
+{
+	return vfs_statx(dfd, filename, flags | AT_NO_AUTOMOUNT,
+			 stat, STATX_BASIC_STATS);
+}
 
 #ifdef __ARCH_WANT_OLD_STAT
 

fs/statfs.c

···
 		flags |= ST_NODIRATIME;
 	if (mnt_flags & MNT_RELATIME)
 		flags |= ST_RELATIME;
+	if (mnt_flags & MNT_NOSYMFOLLOW)
+		flags |= ST_NOSYMFOLLOW;
 	return flags;
 }
 

fs/sysv/inode.c

···
 	buf->f_files = sbi->s_ninodes;
 	buf->f_ffree = sysv_count_free_inodes(sb);
 	buf->f_namelen = SYSV_NAMELEN;
-	buf->f_fsid.val[0] = (u32)id;
-	buf->f_fsid.val[1] = (u32)(id >> 32);
+	buf->f_fsid = u64_to_fsid(id);
 	return 0;
 }
 

fs/udf/super.c

···
 			+ buf->f_bfree;
 	buf->f_ffree = buf->f_bfree;
 	buf->f_namelen = UDF_NAME_LEN;
-	buf->f_fsid.val[0] = (u32)id;
-	buf->f_fsid.val[1] = (u32)(id >> 32);
+	buf->f_fsid = u64_to_fsid(id);
 
 	return 0;
 }

fs/ufs/super.c

···
 		? (buf->f_bfree - uspi->s_root_blocks) : 0;
 	buf->f_files = uspi->s_ncg * uspi->s_ipg;
 	buf->f_namelen = UFS_MAXNAMLEN;
-	buf->f_fsid.val[0] = (u32)id;
-	buf->f_fsid.val[1] = (u32)(id >> 32);
+	buf->f_fsid = u64_to_fsid(id);
 
 	mutex_unlock(&UFS_SB(sb)->s_lock);
 

fs/xfs/xfs_super.c

···
 	statp->f_namelen = MAXNAMELEN - 1;
 
 	id = huge_encode_dev(mp->m_ddev_targp->bt_dev);
-	statp->f_fsid.val[0] = (u32)id;
-	statp->f_fsid.val[1] = (u32)(id >> 32);
+	statp->f_fsid = u64_to_fsid(id);
 
 	icount = percpu_counter_sum(&mp->m_icount);
 	ifree = percpu_counter_sum(&mp->m_ifree);

fs/zonefs/super.c

···
 
 	fsid = le64_to_cpup((void *)sbi->s_uuid.b) ^
 		le64_to_cpup((void *)sbi->s_uuid.b + sizeof(u64));
-	buf->f_fsid.val[0] = (u32)fsid;
-	buf->f_fsid.val[1] = (u32)(fsid >> 32);
+	buf->f_fsid = u64_to_fsid(fsid);
 
 	return 0;
 }

include/linux/fcntl.h

···
 /* List of all valid flags for the open/openat flags argument: */
 #define VALID_OPEN_FLAGS \
 	(O_RDONLY | O_WRONLY | O_RDWR | O_CREAT | O_EXCL | O_NOCTTY | O_TRUNC | \
-	 O_APPEND | O_NDELAY | O_NONBLOCK | O_NDELAY | __O_SYNC | O_DSYNC | \
+	 O_APPEND | O_NDELAY | O_NONBLOCK | __O_SYNC | O_DSYNC | \
 	 FASYNC	| O_DIRECT | O_LARGEFILE | O_DIRECTORY | O_NOFOLLOW | \
 	 O_NOATIME | O_CLOEXEC | O_PATH | __O_TMPFILE)
 

include/linux/fs.h

···
 
 extern int iterate_dir(struct file *, struct dir_context *);
 
-extern int vfs_statx(int, const char __user *, int, struct kstat *, u32);
-extern int vfs_statx_fd(unsigned int, struct kstat *, u32, unsigned int);
+int vfs_fstatat(int dfd, const char __user *filename, struct kstat *stat,
+		int flags);
+int vfs_fstat(int fd, struct kstat *stat);
 
 static inline int vfs_stat(const char __user *filename, struct kstat *stat)
 {
-	return vfs_statx(AT_FDCWD, filename, AT_NO_AUTOMOUNT,
-			 stat, STATX_BASIC_STATS);
+	return vfs_fstatat(AT_FDCWD, filename, stat, 0);
 }
 static inline int vfs_lstat(const char __user *name, struct kstat *stat)
 {
-	return vfs_statx(AT_FDCWD, name, AT_SYMLINK_NOFOLLOW | AT_NO_AUTOMOUNT,
-			 stat, STATX_BASIC_STATS);
+	return vfs_fstatat(AT_FDCWD, name, stat, AT_SYMLINK_NOFOLLOW);
 }
-static inline int vfs_fstatat(int dfd, const char __user *filename,
-			      struct kstat *stat, int flags)
-{
-	return vfs_statx(dfd, filename, flags | AT_NO_AUTOMOUNT,
-			 stat, STATX_BASIC_STATS);
-}
-static inline int vfs_fstat(int fd, struct kstat *stat)
-{
-	return vfs_statx_fd(fd, stat, STATX_BASIC_STATS, 0);
-}
-
 
 extern const char *vfs_get_link(struct dentry *, struct delayed_call *);
 extern int vfs_readlink(struct dentry *, char __user *, int);

include/linux/mount.h

···
 #define MNT_NODIRATIME	0x10
 #define MNT_RELATIME	0x20
 #define MNT_READONLY	0x40	/* does the user want this to be r/o? */
+#define MNT_NOSYMFOLLOW	0x80
 
 #define MNT_SHRINKABLE	0x100
 #define MNT_WRITE_HOLD	0x200
···
 #define MNT_SHARED_MASK	(MNT_UNBINDABLE)
 #define MNT_USER_SETTABLE_MASK  (MNT_NOSUID | MNT_NODEV | MNT_NOEXEC \
 				 | MNT_NOATIME | MNT_NODIRATIME | MNT_RELATIME \
-				 | MNT_READONLY)
+				 | MNT_READONLY | MNT_NOSYMFOLLOW)
 #define MNT_ATIME_MASK (MNT_NOATIME | MNT_NODIRATIME | MNT_RELATIME )
 
 #define MNT_INTERNAL_FLAGS (MNT_SHARED | MNT_WRITE_HOLD | MNT_INTERNAL | \

include/linux/stat.h

···
 #include <linux/time.h>
 #include <linux/uidgid.h>
 
-#define KSTAT_QUERY_FLAGS (AT_STATX_SYNC_TYPE)
-
 struct kstat {
 	u32		result_mask;	/* What fields the user got */
 	umode_t		mode;

include/linux/statfs.h

···
 #define ST_NOATIME	0x0400	/* do not update access times */
 #define ST_NODIRATIME	0x0800	/* do not update directory access times */
 #define ST_RELATIME	0x1000	/* update atime relative to mtime/ctime */
+#define ST_NOSYMFOLLOW	0x2000	/* do not follow symlinks */
 
 struct dentry;
 extern int vfs_get_fsid(struct dentry *dentry, __kernel_fsid_t *fsid);
+
+static inline __kernel_fsid_t u64_to_fsid(u64 v)
+{
+	return (__kernel_fsid_t){.val = {(u32)v, (u32)(v>>32)}};
+}
 
 #endif

include/uapi/linux/mount.h

···
 #define MS_REMOUNT	32	/* Alter flags of a mounted FS */
 #define MS_MANDLOCK	64	/* Allow mandatory locks on an FS */
 #define MS_DIRSYNC	128	/* Directory modifications are synchronous */
+#define MS_NOSYMFOLLOW	256	/* Do not follow symlinks */
 #define MS_NOATIME	1024	/* Do not update access times. */
 #define MS_NODIRATIME	2048	/* Do not update directory access times */
 #define MS_BIND		4096

tools/testing/selftests/mount/.gitignore

···
 # SPDX-License-Identifier: GPL-2.0-only
 unprivileged-remount-test
+nosymfollow-test

tools/testing/selftests/mount/Makefile

···
 CFLAGS = -Wall \
          -O2
 
-TEST_PROGS := run_tests.sh
-TEST_GEN_FILES := unprivileged-remount-test
+TEST_PROGS := run_unprivileged_remount.sh run_nosymfollow.sh
+TEST_GEN_FILES := unprivileged-remount-test nosymfollow-test
 
 include ../lib.mk

tools/testing/selftests/mount/nosymfollow-test.c

···
+// SPDX-License-Identifier: GPL-2.0
+#define _GNU_SOURCE
+#include <errno.h>
+#include <fcntl.h>
+#include <limits.h>
+#include <sched.h>
+#include <stdarg.h>
+#include <stdbool.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/mount.h>
+#include <sys/stat.h>
+#include <sys/types.h>
+#include <sys/vfs.h>
+#include <unistd.h>
+
+#ifndef MS_NOSYMFOLLOW
+# define MS_NOSYMFOLLOW 256     /* Do not follow symlinks */
+#endif
+
+#ifndef ST_NOSYMFOLLOW
+# define ST_NOSYMFOLLOW 0x2000  /* Do not follow symlinks */
+#endif
+
+#define DATA "/tmp/data"
+#define LINK "/tmp/symlink"
+#define TMP  "/tmp"
+
+static void die(char *fmt, ...)
+{
+	va_list ap;
+
+	va_start(ap, fmt);
+	vfprintf(stderr, fmt, ap);
+	va_end(ap);
+	exit(EXIT_FAILURE);
+}
+
+static void vmaybe_write_file(bool enoent_ok, char *filename, char *fmt,
+		va_list ap)
+{
+	ssize_t written;
+	char buf[4096];
+	int buf_len;
+	int fd;
+
+	buf_len = vsnprintf(buf, sizeof(buf), fmt, ap);
+	if (buf_len < 0)
+		die("vsnprintf failed: %s\n", strerror(errno));
+
+	if (buf_len >= sizeof(buf))
+		die("vsnprintf output truncated\n");
+
+	fd = open(filename, O_WRONLY);
+	if (fd < 0) {
+		if ((errno == ENOENT) && enoent_ok)
+			return;
+		die("open of %s failed: %s\n", filename, strerror(errno));
+	}
+
+	written = write(fd, buf, buf_len);
+	if (written != buf_len) {
+		if (written >= 0) {
+			die("short write to %s\n", filename);
+		} else {
+			die("write to %s failed: %s\n",
+				filename, strerror(errno));
+		}
+	}
+
+	if (close(fd) != 0)
+		die("close of %s failed: %s\n", filename, strerror(errno));
+}
+
+static void maybe_write_file(char *filename, char *fmt, ...)
+{
+	va_list ap;
+
+	va_start(ap, fmt);
+	vmaybe_write_file(true, filename, fmt, ap);
+	va_end(ap);
+}
+
+static void write_file(char *filename, char *fmt, ...)
+{
+	va_list ap;
+
+	va_start(ap, fmt);
+	vmaybe_write_file(false, filename, fmt, ap);
+	va_end(ap);
+}
+
+static void create_and_enter_ns(void)
+{
+	uid_t uid = getuid();
+	gid_t gid = getgid();
+
+	if (unshare(CLONE_NEWUSER) != 0)
+		die("unshare(CLONE_NEWUSER) failed: %s\n", strerror(errno));
+
+	maybe_write_file("/proc/self/setgroups", "deny");
+	write_file("/proc/self/uid_map", "0 %d 1", uid);
+	write_file("/proc/self/gid_map", "0 %d 1", gid);
+
+	if (setgid(0) != 0)
+		die("setgid(0) failed %s\n", strerror(errno));
+	if (setuid(0) != 0)
+		die("setuid(0) failed %s\n", strerror(errno));
+
+	if (unshare(CLONE_NEWNS) != 0)
+		die("unshare(CLONE_NEWNS) failed: %s\n", strerror(errno));
+}
+
+static void setup_symlink(void)
+{
+	int data, err;
+
+	data = creat(DATA, O_RDWR);
+	if (data < 0)
+		die("creat failed: %s\n", strerror(errno));
+
+	err = symlink(DATA, LINK);
+	if (err < 0)
+		die("symlink failed: %s\n", strerror(errno));
+
+	if (close(data) != 0)
+		die("close of %s failed: %s\n", DATA, strerror(errno));
+}
+
+static void test_link_traversal(bool nosymfollow)
+{
+	int link;
+
+	link = open(LINK, 0, O_RDWR);
+	if (nosymfollow) {
+		if ((link != -1 || errno != ELOOP)) {
+			die("link traversal unexpected result: %d, %s\n",
+					link, strerror(errno));
+		}
+	} else {
+		if (link < 0)
+			die("link traversal failed: %s\n", strerror(errno));
+
+		if (close(link) != 0)
+			die("close of link failed: %s\n", strerror(errno));
+	}
+}
+
+static void test_readlink(void)
+{
+	char buf[4096];
+	ssize_t ret;
+
+	bzero(buf, sizeof(buf));
+
+	ret = readlink(LINK, buf, sizeof(buf));
+	if (ret < 0)
+		die("readlink failed: %s\n", strerror(errno));
+	if (strcmp(buf, DATA) != 0)
+		die("readlink strcmp failed: '%s' '%s'\n", buf, DATA);
+}
+
+static void test_realpath(void)
+{
+	char *path = realpath(LINK, NULL);
+
+	if (!path)
+		die("realpath failed: %s\n", strerror(errno));
+	if (strcmp(path, DATA) != 0)
+		die("realpath strcmp failed\n");
+
+	free(path);
+}
+
+static void test_statfs(bool nosymfollow)
+{
+	struct statfs buf;
+	int ret;
+
+	ret = statfs(TMP, &buf);
+	if (ret)
+		die("statfs failed: %s\n", strerror(errno));
+
+	if (nosymfollow) {
+		if ((buf.f_flags & ST_NOSYMFOLLOW) == 0)
+			die("ST_NOSYMFOLLOW not set on %s\n", TMP);
+	} else {
+		if ((buf.f_flags & ST_NOSYMFOLLOW) != 0)
+			die("ST_NOSYMFOLLOW set on %s\n", TMP);
+	}
+}
+
+static void run_tests(bool nosymfollow)
+{
+	test_link_traversal(nosymfollow);
+	test_readlink();
+	test_realpath();
+	test_statfs(nosymfollow);
+}
+
+int main(int argc, char **argv)
+{
+	create_and_enter_ns();
+
+	if (mount("testing", TMP, "ramfs", 0, NULL) != 0)
+		die("mount failed: %s\n", strerror(errno));
+
+	setup_symlink();
+	run_tests(false);
+
+	if (mount("testing", TMP, "ramfs", MS_REMOUNT|MS_NOSYMFOLLOW, NULL) != 0)
+		die("remount failed: %s\n", strerror(errno));
+
+	run_tests(true);
+
+	return EXIT_SUCCESS;
+}

tools/testing/selftests/mount/run_nosymfollow.sh

···
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+./nosymfollow-test