Skip to contentSkip to conversation

xfs: don't walk off the end of a directory data block

PATCH Superseded lei lu 2 participants 2 messages Jun 13, 2024 XFS filesystem

v4 of 5 Superseded Latest is v5 →

lore
Share & tools
lore

lei lu

author
#0 lore raw

PATCH v4 1/1 xfs: don't walk off the end of a directory data block

This adds sanity checks for xfs_dir2_data_unused and xfs_dir2_data_entry to make sure don't stray beyond valid memory region. Before patching, the loop simply checks that the start offset of the dup and dep is within the range. So in a crafted image, if last entry is xfs_dir2_data_unused, we can change dup->length to dup->length-1 and leave 1 byte of space. In the next traversal, this space will be considered as dup or dep. We may encounter an out of bound read when accessing the fixed members.

In the patch, we make sure that the remaining bytes large enough to hold an unused entry before accessing xfs_dir2_data_unused and xfs_dir2_data_unused is XFS_DIR2_DATA_ALIGN byte aligned. We also make sure that the remaining bytes large enough to hold a dirent with a single-byte name before accessing xfs_dir2_data_entry.

---
 fs/xfs/libxfs/xfs_dir2_data.c | 30 +++++++++++++++++++++++++-----
 fs/xfs/libxfs/xfs_dir2_priv.h |  7 +++++++
 2 files changed, 32 insertions(+), 5 deletions(-)
fs/xfs/libxfs/xfs_dir2_data.c +25 -5
index dbcf58979a59..feb4499f70e2 100644
@@ -177,6 +177,14 @@ __xfs_dir3_data_check(
 	while (offset < end) {
 		struct xfs_dir2_data_unused	*dup = bp->b_addr + offset;
 		struct xfs_dir2_data_entry	*dep = bp->b_addr + offset;
+		unsigned int	reclen;
+
+		/*
+		 * Are the remaining bytes large enough to hold an
+		 * unused entry?
+		 */
+		if (offset > end - xfs_dir2_data_unusedsize(1))
+			return __this_address;
 
 		/*
 		 * If it's unused, look for the space in the bestfree table.
@@ -186,9 +194,12 @@ __xfs_dir3_data_check(
 		if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) {
 			xfs_failaddr_t	fa;
 
+			reclen = xfs_dir2_data_unusedsize(be16_to_cpu(dup->length));
 			if (lastfree != 0)
 				return __this_address;
-			if (offset + be16_to_cpu(dup->length) > end)
+			if (be16_to_cpu(dup->length) != reclen)
+				return __this_address;
+			if (offset + reclen > end)
 				return __this_address;
 			if (be16_to_cpu(*xfs_dir2_data_unused_tag_p(dup)) !=
 			    offset)
@@ -206,10 +217,18 @@ __xfs_dir3_data_check(
 				    be16_to_cpu(bf[2].length))
 					return __this_address;
 			}
-			offset += be16_to_cpu(dup->length);
+			offset += reclen;
 			lastfree = 1;
 			continue;
 		}
+
+		/*
+		 * This is not an unused entry. Are the remaining bytes
+		 * large enough for a dirent with a single-byte name?
+		 */
+		if (offset > end - xfs_dir2_data_entsize(mp, 1))
+			return __this_address;
+
 		/*
 		 * It's a real entry.  Validate the fields.
 		 * If this is a block directory then make sure it's
@@ -218,9 +237,10 @@ __xfs_dir3_data_check(
 		 */
 		if (dep->namelen == 0)
 			return __this_address;
-		if (!xfs_verify_dir_ino(mp, be64_to_cpu(dep->inumber)))
+		reclen = xfs_dir2_data_entsize(mp, dep->namelen);
+		if (offset + reclen > end)
 			return __this_address;
-		if (offset + xfs_dir2_data_entsize(mp, dep->namelen) > end)
+		if (!xfs_verify_dir_ino(mp, be64_to_cpu(dep->inumber)))
 			return __this_address;
 		if (be16_to_cpu(*xfs_dir2_data_entry_tag_p(mp, dep)) != offset)
 			return __this_address;
@@ -244,7 +264,7 @@ __xfs_dir3_data_check(
 			if (i >= be32_to_cpu(btp->count))
 				return __this_address;
 		}
-		offset += xfs_dir2_data_entsize(mp, dep->namelen);
+		offset += reclen;
 	}
 	/*
 	 * Need to have seen all the entries and all the bestfree slots.
fs/xfs/libxfs/xfs_dir2_priv.h +7 -0
index 1db2e60ba827..263f77bc4cf8 100644
@@ -188,6 +188,13 @@ void xfs_dir2_sf_put_ftype(struct xfs_mount *mp,
 extern int xfs_readdir(struct xfs_trans *tp, struct xfs_inode *dp,
 		       struct dir_context *ctx, size_t bufsize);
 
+static inline unsigned int
+xfs_dir2_data_unusedsize(
+	unsigned int	len)
+{
+	return round_up(len, XFS_DIR2_DATA_ALIGN);
+}
+
 static inline unsigned int
 xfs_dir2_data_entsize(
 	struct xfs_mount	*mp,
Signature · 1 line
2.34.1
Reviewed-by on v4 1/1
#1 lore raw
Quoted 40 lines · lei lu

This adds sanity checks for xfs_dir2_data_unused and xfs_dir2_data_entry to make sure don't stray beyond valid memory region. Before patching, the loop simply checks that the start offset of the dup and dep is within the range. So in a crafted image, if last entry is xfs_dir2_data_unused, we can change dup->length to dup->length-1 and leave 1 byte of space. In the next traversal, this space will be considered as dup or dep. We may encounter an out of bound read when accessing the fixed members.

In the patch, we make sure that the remaining bytes large enough to hold an unused entry before accessing xfs_dir2_data_unused and xfs_dir2_data_unused is XFS_DIR2_DATA_ALIGN byte aligned. We also make sure that the remaining bytes large enough to hold a dirent with a single-byte name before accessing xfs_dir2_data_entry.

---
 fs/xfs/libxfs/xfs_dir2_data.c | 30 +++++++++++++++++++++++++-----
 fs/xfs/libxfs/xfs_dir2_priv.h |  7 +++++++
 2 files changed, 32 insertions(+), 5 deletions(-)
fs/xfs/libxfs/xfs_dir2_data.c +8 -0
index dbcf58979a59..feb4499f70e2 100644
@@ -177,6 +177,14 @@ __xfs_dir3_data_check(
 	while (offset < end) {
 		struct xfs_dir2_data_unused	*dup = bp->b_addr + offset;
 		struct xfs_dir2_data_entry	*dep = bp->b_addr + offset;
+		unsigned int	reclen;
+
+		/*
+		 * Are the remaining bytes large enough to hold an
+		 * unused entry?
+		 */
+		if (offset > end - xfs_dir2_data_unusedsize(1))
+			return __this_address;
 
 		/*
 		 * If it's unused, look for the space in the bestfree table.

fs/xfs/libxfs/xfs_dir2_data.c __xfs_dir3_data_check( · line 197

Quoted 4 lines
fs/xfs/libxfs/xfs_dir2_data.c +1 -0
@@ -186,9 +194,12 @@ __xfs_dir3_data_check(
 		if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) {
 			xfs_failaddr_t	fa;
 
+			reclen = xfs_dir2_data_unusedsize(be16_to_cpu(dup->length));

Overly long line here.

			reclen = xfs_dir2_data_unusedsize(
					be16_to_cpu(dup->length));

With that fixed up, I think this is good to go. Thanks for working on this bug report!

--D

Quoted 73 lines
fs/xfs/libxfs/xfs_dir2_data.c +16 -5
 			if (lastfree != 0)
 				return __this_address;
-			if (offset + be16_to_cpu(dup->length) > end)
+			if (be16_to_cpu(dup->length) != reclen)
+				return __this_address;
+			if (offset + reclen > end)
 				return __this_address;
 			if (be16_to_cpu(*xfs_dir2_data_unused_tag_p(dup)) !=
 			    offset)
@@ -206,10 +217,18 @@ __xfs_dir3_data_check(
 				    be16_to_cpu(bf[2].length))
 					return __this_address;
 			}
-			offset += be16_to_cpu(dup->length);
+			offset += reclen;
 			lastfree = 1;
 			continue;
 		}
+
+		/*
+		 * This is not an unused entry. Are the remaining bytes
+		 * large enough for a dirent with a single-byte name?
+		 */
+		if (offset > end - xfs_dir2_data_entsize(mp, 1))
+			return __this_address;
+
 		/*
 		 * It's a real entry.  Validate the fields.
 		 * If this is a block directory then make sure it's
@@ -218,9 +237,10 @@ __xfs_dir3_data_check(
 		 */
 		if (dep->namelen == 0)
 			return __this_address;
-		if (!xfs_verify_dir_ino(mp, be64_to_cpu(dep->inumber)))
+		reclen = xfs_dir2_data_entsize(mp, dep->namelen);
+		if (offset + reclen > end)
 			return __this_address;
-		if (offset + xfs_dir2_data_entsize(mp, dep->namelen) > end)
+		if (!xfs_verify_dir_ino(mp, be64_to_cpu(dep->inumber)))
 			return __this_address;
 		if (be16_to_cpu(*xfs_dir2_data_entry_tag_p(mp, dep)) != offset)
 			return __this_address;
@@ -244,7 +264,7 @@ __xfs_dir3_data_check(
 			if (i >= be32_to_cpu(btp->count))
 				return __this_address;
 		}
-		offset += xfs_dir2_data_entsize(mp, dep->namelen);
+		offset += reclen;
 	}
 	/*
 	 * Need to have seen all the entries and all the bestfree slots.
fs/xfs/libxfs/xfs_dir2_priv.h +7 -0
index 1db2e60ba827..263f77bc4cf8 100644
@@ -188,6 +188,13 @@ void xfs_dir2_sf_put_ftype(struct xfs_mount *mp,
 extern int xfs_readdir(struct xfs_trans *tp, struct xfs_inode *dp,
 		       struct dir_context *ctx, size_t bufsize);
 
+static inline unsigned int
+xfs_dir2_data_unusedsize(
+	unsigned int	len)
+{
+	return round_up(len, XFS_DIR2_DATA_ALIGN);
+}
+
 static inline unsigned int
 xfs_dir2_data_entsize(
 	struct xfs_mount	*mp,
Signature · 1 line
2.34.1