ext4: avoid declaring fs inconsistent due to invalid file handles

commit 8a363970d1dc38c4ec4ad575c862f776f468d057 upstream.

If we receive a file handle, either from NFS or open_by_handle_at(2),
and it points at an inode which has not been initialized, and the file
system has metadata checksums enabled, we shouldn't try to get the
inode, discover the checksum is invalid, and then declare the file
system as being inconsistent.

This can be reproduced by creating a test file system via "mke2fs -t
ext4 -O metadata_csum /tmp/foo.img 8M", mounting it, cd'ing into that
directory, and then running the following program.

#define _GNU_SOURCE
#include <fcntl.h>

struct handle {
	struct file_handle fh;
	unsigned char fid[MAX_HANDLE_SZ];
};

int main(int argc, char **argv)
{
	struct handle h = {{8, 1 }, { 12, }};

	open_by_handle_at(AT_FDCWD, &h.fh, O_RDONLY);
	return 0;
}

Google-Bug-Id: 120690101
Signed-off-by: NTheodore Ts'o <tytso@mit.edu>
Cc: stable@kernel.org
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

fs/ext4/ext4.h

@@ -2459,8 +2459,19 @@ int do_journal_get_write_access(handle_t *handle,
 #define FALL_BACK_TO_NONDELALLOC 1
 #define CONVERT_INLINE_DATA	 2
 
-extern struct inode *ext4_iget(struct super_block *, unsigned long);
-extern struct inode *ext4_iget_normal(struct super_block *, unsigned long);
+typedef enum {
+	EXT4_IGET_NORMAL =	0,
+	EXT4_IGET_SPECIAL =	0x0001, /* OK to iget a system inode */
+	EXT4_IGET_HANDLE = 	0x0002	/* Inode # is from a handle */
+} ext4_iget_flags;
+
+extern struct inode *__ext4_iget(struct super_block *sb, unsigned long ino,
+				 ext4_iget_flags flags, const char *function,
+				 unsigned int line);
+
+#define ext4_iget(sb, ino, flags) \
+	__ext4_iget((sb), (ino), (flags), __func__, __LINE__)
+
 extern int  ext4_write_inode(struct inode *, struct writeback_control *);
 extern int  ext4_setattr(struct dentry *, struct iattr *);
 extern int  ext4_getattr(const struct path *, struct kstat *, u32, unsigned int);

fs/ext4/ialloc.c

@@ -1225,7 +1225,7 @@ struct inode *ext4_orphan_get(struct super_block *sb, unsigned long ino)
 	if (!ext4_test_bit(bit, bitmap_bh->b_data))
 		goto bad_orphan;
 
-	inode = ext4_iget(sb, ino);
+	inode = ext4_iget(sb, ino, EXT4_IGET_NORMAL);
 	if (IS_ERR(inode)) {
 		err = PTR_ERR(inode);
 		ext4_error(sb, "couldn't read orphan inode %lu (err %d)",

fs/ext4/inode.c

@@ -4786,7 +4786,9 @@ static inline u64 ext4_inode_peek_iversion(const struct inode *inode)
 		return inode_peek_iversion(inode);
 }
 
-struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
+struct inode *__ext4_iget(struct super_block *sb, unsigned long ino,
+			  ext4_iget_flags flags, const char *function,
+			  unsigned int line)
 {
 	struct ext4_iloc iloc;
 	struct ext4_inode *raw_inode;
@@ -4800,6 +4802,18 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 	gid_t i_gid;
 	projid_t i_projid;
 
+	if (((flags & EXT4_IGET_NORMAL) &&
+	     (ino < EXT4_FIRST_INO(sb) && ino != EXT4_ROOT_INO)) ||
+	    (ino < EXT4_ROOT_INO) ||
+	    (ino > le32_to_cpu(EXT4_SB(sb)->s_es->s_inodes_count))) {
+		if (flags & EXT4_IGET_HANDLE)
+			return ERR_PTR(-ESTALE);
+		__ext4_error(sb, function, line,
+			     "inode #%lu: comm %s: iget: illegal inode #",
+			     ino, current->comm);
+		return ERR_PTR(-EFSCORRUPTED);
+	}
+
 	inode = iget_locked(sb, ino);
 	if (!inode)
 		return ERR_PTR(-ENOMEM);
@@ -4815,18 +4829,26 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 	raw_inode = ext4_raw_inode(&iloc);
 
 	if ((ino == EXT4_ROOT_INO) && (raw_inode->i_links_count == 0)) {
-		EXT4_ERROR_INODE(inode, "root inode unallocated");
+		ext4_error_inode(inode, function, line, 0,
+				 "iget: root inode unallocated");
 		ret = -EFSCORRUPTED;
 		goto bad_inode;
 	}
 
+	if ((flags & EXT4_IGET_HANDLE) &&
+	    (raw_inode->i_links_count == 0) && (raw_inode->i_mode == 0)) {
+		ret = -ESTALE;
+		goto bad_inode;
+	}
+
 	if (EXT4_INODE_SIZE(inode->i_sb) > EXT4_GOOD_OLD_INODE_SIZE) {
 		ei->i_extra_isize = le16_to_cpu(raw_inode->i_extra_isize);
 		if (EXT4_GOOD_OLD_INODE_SIZE + ei->i_extra_isize >
 			EXT4_INODE_SIZE(inode->i_sb) ||
 		    (ei->i_extra_isize & 3)) {
-			EXT4_ERROR_INODE(inode,
-					 "bad extra_isize %u (inode size %u)",
+			ext4_error_inode(inode, function, line, 0,
+					 "iget: bad extra_isize %u "
+					 "(inode size %u)",
 					 ei->i_extra_isize,
 					 EXT4_INODE_SIZE(inode->i_sb));
 			ret = -EFSCORRUPTED;
@@ -4848,7 +4870,8 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 	}
 
 	if (!ext4_inode_csum_verify(inode, raw_inode, ei)) {
-		EXT4_ERROR_INODE(inode, "checksum invalid");
+		ext4_error_inode(inode, function, line, 0,
+				 "iget: checksum invalid");
 		ret = -EFSBADCRC;
 		goto bad_inode;
 	}
@@ -4905,7 +4928,8 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 			((__u64)le16_to_cpu(raw_inode->i_file_acl_high)) << 32;
 	inode->i_size = ext4_isize(sb, raw_inode);
 	if ((size = i_size_read(inode)) < 0) {
-		EXT4_ERROR_INODE(inode, "bad i_size value: %lld", size);
+		ext4_error_inode(inode, function, line, 0,
+				 "iget: bad i_size value: %lld", size);
 		ret = -EFSCORRUPTED;
 		goto bad_inode;
 	}
@@ -4981,7 +5005,8 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 	ret = 0;
 	if (ei->i_file_acl &&
 	    !ext4_data_block_valid(EXT4_SB(sb), ei->i_file_acl, 1)) {
-		EXT4_ERROR_INODE(inode, "bad extended attribute block %llu",
+		ext4_error_inode(inode, function, line, 0,
+				 "iget: bad extended attribute block %llu",
 				 ei->i_file_acl);
 		ret = -EFSCORRUPTED;
 		goto bad_inode;
@@ -5009,8 +5034,9 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 	} else if (S_ISLNK(inode->i_mode)) {
 		/* VFS does not allow setting these so must be corruption */
 		if (IS_APPEND(inode) || IS_IMMUTABLE(inode)) {
-			EXT4_ERROR_INODE(inode,
-			  "immutable or append flags not allowed on symlinks");
+			ext4_error_inode(inode, function, line, 0,
+					 "iget: immutable or append flags "
+					 "not allowed on symlinks");
 			ret = -EFSCORRUPTED;
 			goto bad_inode;
 		}
@@ -5040,7 +5066,8 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 		make_bad_inode(inode);
 	} else {
 		ret = -EFSCORRUPTED;
-		EXT4_ERROR_INODE(inode, "bogus i_mode (%o)", inode->i_mode);
+		ext4_error_inode(inode, function, line, 0,
+				 "iget: bogus i_mode (%o)", inode->i_mode);
 		goto bad_inode;
 	}
 	brelse(iloc.bh);
@@ -5054,13 +5081,6 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 	return ERR_PTR(ret);
 }
 
-struct inode *ext4_iget_normal(struct super_block *sb, unsigned long ino)
-{
-	if (ino < EXT4_FIRST_INO(sb) && ino != EXT4_ROOT_INO)
-		return ERR_PTR(-EFSCORRUPTED);
-	return ext4_iget(sb, ino);
-}
-
 static int ext4_inode_blocks_set(handle_t *handle,
 				struct ext4_inode *raw_inode,
 				struct ext4_inode_info *ei)

fs/ext4/ioctl.c

@@ -125,7 +125,7 @@ static long swap_inode_boot_loader(struct super_block *sb,
 	    !inode_owner_or_capable(inode) || !capable(CAP_SYS_ADMIN))
 		return -EPERM;
 
-	inode_bl = ext4_iget(sb, EXT4_BOOT_LOADER_INO);
+	inode_bl = ext4_iget(sb, EXT4_BOOT_LOADER_INO, EXT4_IGET_SPECIAL);
 	if (IS_ERR(inode_bl))
 		return PTR_ERR(inode_bl);
 	ei_bl = EXT4_I(inode_bl);

fs/ext4/namei.c

@@ -1571,7 +1571,7 @@ static struct dentry *ext4_lookup(struct inode *dir, struct dentry *dentry, unsi
 					 dentry);
 			return ERR_PTR(-EFSCORRUPTED);
 		}
-		inode = ext4_iget_normal(dir->i_sb, ino);
+		inode = ext4_iget(dir->i_sb, ino, EXT4_IGET_NORMAL);
 		if (inode == ERR_PTR(-ESTALE)) {
 			EXT4_ERROR_INODE(dir,
 					 "deleted inode referenced: %u",
@@ -1613,7 +1613,7 @@ struct dentry *ext4_get_parent(struct dentry *child)
 		return ERR_PTR(-EFSCORRUPTED);
 	}
 
-	return d_obtain_alias(ext4_iget_normal(child->d_sb, ino));
+	return d_obtain_alias(ext4_iget(child->d_sb, ino, EXT4_IGET_NORMAL));
 }
 
 /*

fs/ext4/resize.c

@@ -1637,7 +1637,7 @@ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input)
 				     "No reserved GDT blocks, can't resize");
 			return -EPERM;
 		}
-		inode = ext4_iget(sb, EXT4_RESIZE_INO);
+		inode = ext4_iget(sb, EXT4_RESIZE_INO, EXT4_IGET_SPECIAL);
 		if (IS_ERR(inode)) {
 			ext4_warning(sb, "Error opening resize inode");
 			return PTR_ERR(inode);
@@ -1965,7 +1965,8 @@ int ext4_resize_fs(struct super_block *sb, ext4_fsblk_t n_blocks_count)
 		}
 
 		if (!resize_inode)
-			resize_inode = ext4_iget(sb, EXT4_RESIZE_INO);
+			resize_inode = ext4_iget(sb, EXT4_RESIZE_INO,
+						 EXT4_IGET_SPECIAL);
 		if (IS_ERR(resize_inode)) {
 			ext4_warning(sb, "Error opening resize inode");
 			return PTR_ERR(resize_inode);

fs/ext4/super.c

@@ -1173,20 +1173,11 @@ static struct inode *ext4_nfs_get_inode(struct super_block *sb,
 {
 	struct inode *inode;
 
-	if (ino < EXT4_FIRST_INO(sb) && ino != EXT4_ROOT_INO)
-		return ERR_PTR(-ESTALE);
-	if (ino > le32_to_cpu(EXT4_SB(sb)->s_es->s_inodes_count))
-		return ERR_PTR(-ESTALE);
-
-	/* iget isn't really right if the inode is currently unallocated!!
-	 *
-	 * ext4_read_inode will return a bad_inode if the inode had been
-	 * deleted, so we should be safe.
-	 *
+	/*
 	 * Currently we don't know the generation for parent directory, so
 	 * a generation of 0 means "accept any"
 	 */
-	inode = ext4_iget_normal(sb, ino);
+	inode = ext4_iget(sb, ino, EXT4_IGET_HANDLE);
 	if (IS_ERR(inode))
 		return ERR_CAST(inode);
 	if (generation && inode->i_generation != generation) {
@@ -4350,7 +4341,7 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 	 * so we can safely mount the rest of the filesystem now.
 	 */
 
-	root = ext4_iget(sb, EXT4_ROOT_INO);
+	root = ext4_iget(sb, EXT4_ROOT_INO, EXT4_IGET_SPECIAL);
 	if (IS_ERR(root)) {
 		ext4_msg(sb, KERN_ERR, "get root inode failed");
 		ret = PTR_ERR(root);
@@ -4620,7 +4611,7 @@ static struct inode *ext4_get_journal_inode(struct super_block *sb,
 	 * happen if we iget() an unused inode, as the subsequent iput()
 	 * will try to delete it.
 	 */
-	journal_inode = ext4_iget(sb, journal_inum);
+	journal_inode = ext4_iget(sb, journal_inum, EXT4_IGET_SPECIAL);
 	if (IS_ERR(journal_inode)) {
 		ext4_msg(sb, KERN_ERR, "no journal found");
 		return NULL;
@@ -5702,7 +5693,7 @@ static int ext4_quota_enable(struct super_block *sb, int type, int format_id,
 	if (!qf_inums[type])
 		return -EPERM;
 
-	qf_inode = ext4_iget(sb, qf_inums[type]);
+	qf_inode = ext4_iget(sb, qf_inums[type], EXT4_IGET_SPECIAL);
 	if (IS_ERR(qf_inode)) {
 		ext4_error(sb, "Bad quota inode # %lu", qf_inums[type]);
 		return PTR_ERR(qf_inode);

fs/ext4/xattr.c

@@ -384,7 +384,7 @@ static int ext4_xattr_inode_iget(struct inode *parent, unsigned long ea_ino,
 	struct inode *inode;
 	int err;
 
-	inode = ext4_iget(parent->i_sb, ea_ino);
+	inode = ext4_iget(parent->i_sb, ea_ino, EXT4_IGET_NORMAL);
 	if (IS_ERR(inode)) {
 		err = PTR_ERR(inode);
 		ext4_error(parent->i_sb,
@@ -1486,7 +1486,8 @@ ext4_xattr_inode_cache_find(struct inode *inode, const void *value,
 	}
 
 	while (ce) {
-		ea_inode = ext4_iget(inode->i_sb, ce->e_value);
+		ea_inode = ext4_iget(inode->i_sb, ce->e_value,
+				     EXT4_IGET_NORMAL);
 		if (!IS_ERR(ea_inode) &&
 		    !is_bad_inode(ea_inode) &&
 		    (EXT4_I(ea_inode)->i_flags & EXT4_EA_INODE_FL) &&