1 files changed, 125 insertions, 94 deletions

diff --git a/fs/ext4/dir.c b/fs/ext4/dir.c
index e8b365000d73..256fe2c1d4c1 100644
--- a/fs/ext4/dir.c
+++ b/fs/ext4/dir.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0
 /*
  *  linux/fs/ext4/dir.c
  *
@@ -24,12 +25,17 @@
 #include <linux/fs.h>
 #include <linux/buffer_head.h>
 #include <linux/slab.h>
+#include <linux/iversion.h>
+#include <linux/unicode.h>
 #include "ext4.h"
 #include "xattr.h"
 
 static int ext4_dx_readdir(struct file *, struct dir_context *);
 
 /**
+ * is_dx_dir() - check if a directory is using htree indexing
+ * @inode: directory inode
+ *
  * Check if the given dir-inode refers to an htree-indexed directory
  * (or a directory which could potentially get converted to use htree
  * indexing).
@@ -49,6 +55,18 @@ static int is_dx_dir(struct inode *inode)
 	return 0;
 }
 
+static bool is_fake_dir_entry(struct ext4_dir_entry_2 *de)
+{
+	/* Check if . or .. , or skip if namelen is 0 */
+	if ((de->name_len > 0) && (de->name_len <= 2) && (de->name[0] == '.') &&
+	    (de->name[1] == '.' || de->name[1] == '\0'))
+		return true;
+	/* Check if this is a csum entry */
+	if (de->file_type == EXT4_FT_DIR_CSUM)
+		return true;
+	return false;
+}
+
 /*
  * Return 0 if the directory entry is OK, and 1 if there is a problem
  *
@@ -66,35 +84,44 @@ int __ext4_check_dir_entry(const char *function, unsigned int line,
 	const char *error_msg = NULL;
 	const int rlen = ext4_rec_len_from_disk(de->rec_len,
 						dir->i_sb->s_blocksize);
+	const int next_offset = ((char *) de - buf) + rlen;
+	bool fake = is_fake_dir_entry(de);
+	bool has_csum = ext4_has_feature_metadata_csum(dir->i_sb);
 
-	if (unlikely(rlen < EXT4_DIR_REC_LEN(1)))
+	if (unlikely(rlen < ext4_dir_rec_len(1, fake ? NULL : dir)))
 		error_msg = "rec_len is smaller than minimal";
 	else if (unlikely(rlen % 4 != 0))
 		error_msg = "rec_len % 4 != 0";
-	else if (unlikely(rlen < EXT4_DIR_REC_LEN(de->name_len)))
+	else if (unlikely(rlen < ext4_dir_rec_len(de->name_len,
+							fake ? NULL : dir)))
 		error_msg = "rec_len is too small for name_len";
-	else if (unlikely(((char *) de - buf) + rlen > size))
-		error_msg = "directory entry across range";
+	else if (unlikely(next_offset > size))
+		error_msg = "directory entry overrun";
+	else if (unlikely(next_offset > size - ext4_dir_rec_len(1,
+						  has_csum ? NULL : dir) &&
+			  next_offset != size))
+		error_msg = "directory entry too close to block end";
 	else if (unlikely(le32_to_cpu(de->inode) >
 			le32_to_cpu(EXT4_SB(dir->i_sb)->s_es->s_inodes_count)))
 		error_msg = "inode out of bounds";
+	else if (unlikely(next_offset == size && de->name_len == 1 &&
+			  de->name[0] == '.'))
+		error_msg = "'.' directory cannot be the last in data block";
 	else
 		return 0;
 
 	if (filp)
 		ext4_error_file(filp, function, line, bh->b_blocknr,
-				"bad entry in directory: %s - offset=%u(%u), "
-				"inode=%u, rec_len=%d, name_len=%d",
-				error_msg, (unsigned) (offset % size),
-				offset, le32_to_cpu(de->inode),
-				rlen, de->name_len);
+				"bad entry in directory: %s - offset=%u, "
+				"inode=%u, rec_len=%d, size=%d fake=%d",
+				error_msg, offset, le32_to_cpu(de->inode),
+				rlen, size, fake);
 	else
 		ext4_error_inode(dir, function, line, bh->b_blocknr,
-				"bad entry in directory: %s - offset=%u(%u), "
-				"inode=%u, rec_len=%d, name_len=%d",
-				error_msg, (unsigned) (offset % size),
-				offset, le32_to_cpu(de->inode),
-				rlen, de->name_len);
+				"bad entry in directory: %s - offset=%u, "
+				"inode=%u, rec_len=%d, size=%d fake=%d",
+				 error_msg, offset, le32_to_cpu(de->inode),
+				 rlen, size, fake);
 
 	return 1;
 }
@@ -108,26 +135,26 @@ static int ext4_readdir(struct file *file, struct dir_context *ctx)
 	struct inode *inode = file_inode(file);
 	struct super_block *sb = inode->i_sb;
 	struct buffer_head *bh = NULL;
-	int dir_has_error = 0;
 	struct fscrypt_str fstr = FSTR_INIT(NULL, 0);
+	struct dir_private_info *info = file->private_data;
 
-	if (ext4_encrypted_inode(inode)) {
-		err = fscrypt_get_encryption_info(inode);
-		if (err && err != -ENOKEY)
-			return err;
-	}
+	err = fscrypt_prepare_readdir(inode);
+	if (err)
+		return err;
 
 	if (is_dx_dir(inode)) {
 		err = ext4_dx_readdir(file, ctx);
-		if (err != ERR_BAD_DX_DIR) {
+		if (err != ERR_BAD_DX_DIR)
 			return err;
+
+		/* Can we just clear INDEX flag to ignore htree information? */
+		if (!ext4_has_feature_metadata_csum(sb)) {
+			/*
+			 * We don't set the inode dirty flag since it's not
+			 * critical that it gets flushed back to the disk.
+			 */
+			ext4_clear_inode_flag(inode, EXT4_INODE_INDEX);
 		}
-		/*
-		 * We don't set the inode dirty flag since it's not
-		 * critical that it get flushed back to the disk.
-		 */
-		ext4_clear_inode_flag(file_inode(file),
-				      EXT4_INODE_INDEX);
 	}
 
 	if (ext4_has_inline_data(inode)) {
@@ -138,14 +165,12 @@ static int ext4_readdir(struct file *file, struct dir_context *ctx)
 			return err;
 	}
 
-	if (ext4_encrypted_inode(inode)) {
-		err = fscrypt_fname_alloc_buffer(inode, EXT4_NAME_LEN, &fstr);
+	if (IS_ENCRYPTED(inode)) {
+		err = fscrypt_fname_alloc_buffer(EXT4_NAME_LEN, &fstr);
 		if (err < 0)
 			return err;
 	}
 
-	offset = ctx->pos & (sb->s_blocksize - 1);
-
 	while (ctx->pos < inode->i_size) {
 		struct ext4_map_blocks map;
 
@@ -154,17 +179,26 @@ static int ext4_readdir(struct file *file, struct dir_context *ctx)
 			goto errout;
 		}
 		cond_resched();
+		offset = ctx->pos & (sb->s_blocksize - 1);
 		map.m_lblk = ctx->pos >> EXT4_BLOCK_SIZE_BITS(sb);
 		map.m_len = 1;
 		err = ext4_map_blocks(NULL, inode, &map, 0);
+		if (err == 0) {
+			/* m_len should never be zero but let's avoid
+			 * an infinite loop if it somehow is */
+			if (map.m_len == 0)
+				map.m_len = 1;
+			ctx->pos += map.m_len * sb->s_blocksize;
+			continue;
+		}
 		if (err > 0) {
-			pgoff_t index = map.m_pblk >>
-					(PAGE_SHIFT - inode->i_blkbits);
+			pgoff_t index = map.m_pblk << inode->i_blkbits >>
+					PAGE_SHIFT;
 			if (!ra_has_index(&file->f_ra, index))
 				page_cache_sync_readahead(
-					sb->s_bdev->bd_inode->i_mapping,
-					&file->f_ra, file,
-					index, 1);
+					sb->s_bdev->bd_mapping,
+					&file->f_ra, file, index,
+					1 << EXT4_SB(sb)->s_min_folio_order);
 			file->f_ra.prev_pos = (loff_t)index << PAGE_SHIFT;
 			bh = ext4_bread(NULL, inode, map.m_lblk, 0);
 			if (IS_ERR(bh)) {
@@ -175,13 +209,6 @@ static int ext4_readdir(struct file *file, struct dir_context *ctx)
 		}
 
 		if (!bh) {
-			if (!dir_has_error) {
-				EXT4_ERROR_FILE(file, 0,
-						"directory contains a "
-						"hole at offset %llu",
-					   (unsigned long long) ctx->pos);
-				dir_has_error = 1;
-			}
 			/* corrupt size?  Maybe no more blocks to read */
 			if (ctx->pos > inode->i_blocks << 9)
 				break;
@@ -191,8 +218,7 @@ static int ext4_readdir(struct file *file, struct dir_context *ctx)
 
 		/* Check the checksum */
 		if (!buffer_verified(bh) &&
-		    !ext4_dirent_csum_verify(inode,
-				(struct ext4_dir_entry *)bh->b_data)) {
+		    !ext4_dirblock_csum_verify(inode, bh)) {
 			EXT4_ERROR_FILE(file, 0, "directory fails checksum "
 					"at offset %llu",
 					(unsigned long long)ctx->pos);
@@ -207,7 +233,7 @@ static int ext4_readdir(struct file *file, struct dir_context *ctx)
 		 * readdir(2), then we might be pointing to an invalid
 		 * dirent right now.  Scan from the start of the block
 		 * to make sure. */
-		if (file->f_version != inode->i_version) {
+		if (!inode_eq_iversion(inode, info->cookie)) {
 			for (i = 0; i < sb->s_blocksize && i < offset; ) {
 				de = (struct ext4_dir_entry_2 *)
 					(bh->b_data + i);
@@ -218,7 +244,8 @@ static int ext4_readdir(struct file *file, struct dir_context *ctx)
 				 * failure will be detected in the
 				 * dirent test below. */
 				if (ext4_rec_len_from_disk(de->rec_len,
-					sb->s_blocksize) < EXT4_DIR_REC_LEN(1))
+					sb->s_blocksize) < ext4_dir_rec_len(1,
+									inode))
 					break;
 				i += ext4_rec_len_from_disk(de->rec_len,
 							    sb->s_blocksize);
@@ -226,7 +253,7 @@ static int ext4_readdir(struct file *file, struct dir_context *ctx)
 			offset = i;
 			ctx->pos = (ctx->pos & ~(sb->s_blocksize - 1))
 				| offset;
-			file->f_version = inode->i_version;
+			info->cookie = inode_query_iversion(inode);
 		}
 
 		while (ctx->pos < inode->i_size
@@ -245,7 +272,7 @@ static int ext4_readdir(struct file *file, struct dir_context *ctx)
 			offset += ext4_rec_len_from_disk(de->rec_len,
 					sb->s_blocksize);
 			if (le32_to_cpu(de->inode)) {
-				if (!ext4_encrypted_inode(inode)) {
+				if (!IS_ENCRYPTED(inode)) {
 					if (!dir_emit(ctx, de->name,
 					    de->name_len,
 					    le32_to_cpu(de->inode),
@@ -256,10 +283,20 @@ static int ext4_readdir(struct file *file, struct dir_context *ctx)
 					struct fscrypt_str de_name =
 							FSTR_INIT(de->name,
 								de->name_len);
+					u32 hash;
+					u32 minor_hash;
+
+					if (IS_CASEFOLDED(inode)) {
+						hash = EXT4_DIRENT_HASH(de);
+						minor_hash = EXT4_DIRENT_MINOR_HASH(de);
+					} else {
+						hash = 0;
+						minor_hash = 0;
+					}
 
 					/* Directory is encrypted */
 					err = fscrypt_fname_disk_to_usr(inode,
-						0, 0, &de_name, &fstr);
+						hash, minor_hash, &de_name, &fstr);
 					de_name = fstr;
 					fstr.len = save_len;
 					if (err)
@@ -278,14 +315,11 @@ static int ext4_readdir(struct file *file, struct dir_context *ctx)
 			goto done;
 		brelse(bh);
 		bh = NULL;
-		offset = 0;
 	}
 done:
 	err = 0;
 errout:
-#ifdef CONFIG_EXT4_FS_ENCRYPTION
 	fscrypt_fname_free_buffer(&fstr);
-#endif
 	brelse(bh);
 	return err;
 }
@@ -362,14 +396,17 @@ static inline loff_t ext4_get_htree_eof(struct file *filp)
 static loff_t ext4_dir_llseek(struct file *file, loff_t offset, int whence)
 {
 	struct inode *inode = file->f_mapping->host;
+	struct dir_private_info *info = file->private_data;
 	int dx_dir = is_dx_dir(inode);
-	loff_t htree_max = ext4_get_htree_eof(file);
+	loff_t ret, htree_max = ext4_get_htree_eof(file);
 
 	if (likely(dx_dir))
-		return generic_file_llseek_size(file, offset, whence,
+		ret = generic_file_llseek_size(file, offset, whence,
 						    htree_max, htree_max);
 	else
-		return ext4_llseek(file, offset, whence);
+		ret = ext4_llseek(file, offset, whence);
+	info->cookie = inode_peek_iversion(inode) - 1;
+	return ret;
 }
 
 /*
@@ -384,11 +421,11 @@ struct fname {
 	__u32		inode;
 	__u8		name_len;
 	__u8		file_type;
-	char		name[0];
+	char		name[] __counted_by(name_len);
 };
 
 /*
- * This functoin implements a non-recursive way of freeing all of the
+ * This function implements a non-recursive way of freeing all of the
  * nodes in the red-black tree.
  */
 static void free_rb_tree_fname(struct rb_root *root)
@@ -405,18 +442,15 @@ static void free_rb_tree_fname(struct rb_root *root)
 	*root = RB_ROOT;
 }
 
-
-static struct dir_private_info *ext4_htree_create_dir_info(struct file *filp,
-							   loff_t pos)
+static void ext4_htree_init_dir_info(struct file *filp, loff_t pos)
 {
-	struct dir_private_info *p;
-
-	p = kzalloc(sizeof(struct dir_private_info), GFP_KERNEL);
-	if (!p)
-		return NULL;
-	p->curr_hash = pos2maj_hash(filp, pos);
-	p->curr_minor_hash = pos2min_hash(filp, pos);
-	return p;
+	struct dir_private_info *p = filp->private_data;
+
+	if (is_dx_dir(file_inode(filp)) && !p->initialized) {
+		p->curr_hash = pos2maj_hash(filp, pos);
+		p->curr_minor_hash = pos2min_hash(filp, pos);
+		p->initialized = true;
+	}
 }
 
 void ext4_htree_free_dir_info(struct dir_private_info *p)
@@ -440,14 +474,13 @@ int ext4_htree_store_dirent(struct file *dir_file, __u32 hash,
 	struct rb_node **p, *parent = NULL;
 	struct fname *fname, *new_fn;
 	struct dir_private_info *info;
-	int len;
 
 	info = dir_file->private_data;
 	p = &info->root.rb_node;
 
 	/* Create and allocate the fname structure */
-	len = sizeof(struct fname) + ent_name->len + 1;
-	new_fn = kzalloc(len, GFP_KERNEL);
+	new_fn = kzalloc(struct_size(new_fn, name, ent_name->len + 1),
+			 GFP_KERNEL);
 	if (!new_fn)
 		return -ENOMEM;
 	new_fn->hash = hash;
@@ -456,7 +489,6 @@ int ext4_htree_store_dirent(struct file *dir_file, __u32 hash,
 	new_fn->name_len = ent_name->len;
 	new_fn->file_type = dirent->file_type;
 	memcpy(new_fn->name, ent_name->name, ent_name->len);
-	new_fn->name[ent_name->len] = 0;
 
 	while (*p) {
 		parent = *p;
@@ -492,7 +524,7 @@ int ext4_htree_store_dirent(struct file *dir_file, __u32 hash,
 
 /*
  * This is a helper function for ext4_dx_readdir.  It calls filldir
- * for all entres on the fname linked list.  (Normally there is only
+ * for all entries on the fname linked list.  (Normally there is only
  * one entry on the linked list, unless there are 62 bit hash collisions.)
  */
 static int call_filldir(struct file *file, struct dir_context *ctx,
@@ -527,14 +559,9 @@ static int ext4_dx_readdir(struct file *file, struct dir_context *ctx)
 	struct dir_private_info *info = file->private_data;
 	struct inode *inode = file_inode(file);
 	struct fname *fname;
-	int	ret;
+	int ret = 0;
 
-	if (!info) {
-		info = ext4_htree_create_dir_info(file, ctx->pos);
-		if (!info)
-			return -ENOMEM;
-		file->private_data = info;
-	}
+	ext4_htree_init_dir_info(file, ctx->pos);
 
 	if (ctx->pos == ext4_get_htree_eof(file))
 		return 0;	/* EOF */
@@ -567,15 +594,15 @@ static int ext4_dx_readdir(struct file *file, struct dir_context *ctx)
 		 * cached entries.
 		 */
 		if ((!info->curr_node) ||
-		    (file->f_version != inode->i_version)) {
+		    !inode_eq_iversion(inode, info->cookie)) {
 			info->curr_node = NULL;
 			free_rb_tree_fname(&info->root);
-			file->f_version = inode->i_version;
+			info->cookie = inode_query_iversion(inode);
 			ret = ext4_htree_fill_tree(file, info->curr_hash,
 						   info->curr_minor_hash,
 						   &info->next_hash);
 			if (ret < 0)
-				return ret;
+				goto finished;
 			if (ret == 0) {
 				ctx->pos = ext4_get_htree_eof(file);
 				break;
@@ -606,14 +633,7 @@ static int ext4_dx_readdir(struct file *file, struct dir_context *ctx)
 	}
 finished:
 	info->last_pos = ctx->pos;
-	return 0;
-}
-
-static int ext4_dir_open(struct inode * inode, struct file * filp)
-{
-	if (ext4_encrypted_inode(inode))
-		return fscrypt_get_encryption_info(inode) ? -EACCES : 0;
-	return 0;
+	return ret < 0 ? ret : 0;
 }
 
 static int ext4_release_dir(struct inode *inode, struct file *filp)
@@ -632,7 +652,7 @@ int ext4_check_all_de(struct inode *dir, struct buffer_head *bh, void *buf,
 	unsigned int offset = 0;
 	char *top;
 
-	de = (struct ext4_dir_entry_2 *)buf;
+	de = buf;
 	top = buf + buf_size;
 	while ((char *) de < top) {
 		if (ext4_check_dir_entry(dir, NULL, de, bh,
@@ -648,7 +668,19 @@ int ext4_check_all_de(struct inode *dir, struct buffer_head *bh, void *buf,
 	return 0;
 }
 
+static int ext4_dir_open(struct inode *inode, struct file *file)
+{
+	struct dir_private_info *info;
+
+	info = kzalloc(sizeof(*info), GFP_KERNEL);
+	if (!info)
+		return -ENOMEM;
+	file->private_data = info;
+	return 0;
+}
+
 const struct file_operations ext4_dir_operations = {
+	.open		= ext4_dir_open,
 	.llseek		= ext4_dir_llseek,
 	.read		= generic_read_dir,
 	.iterate_shared	= ext4_readdir,
@@ -657,6 +689,5 @@ const struct file_operations ext4_dir_operations = {
 	.compat_ioctl	= ext4_compat_ioctl,
 #endif
 	.fsync		= ext4_sync_file,
-	.open		= ext4_dir_open,
 	.release	= ext4_release_dir,
 };