3.16.58-rc1 review patch. If anyone has any objections, please let me know.
------------------
From: Theodore Ts'o tytso@mit.edu
commit bc890a60247171294acc0bd67d211fa4b88d40ba upstream.
If there is a corupted file system where the claimed depth of the extent tree is -1, this can cause a massive buffer overrun leading to sadness.
This addresses CVE-2018-10877.
https://bugzilla.kernel.org/show_bug.cgi?id=199417
Signed-off-by: Theodore Ts'o tytso@mit.edu [bwh: Backported to 3.16: return -EIO instead of -EFSCORRUPTED] Signed-off-by: Ben Hutchings ben@decadent.org.uk --- fs/ext4/ext4_extents.h | 1 + fs/ext4/extents.c | 6 ++++++ 2 files changed, 7 insertions(+)
--- a/fs/ext4/ext4_extents.h +++ b/fs/ext4/ext4_extents.h @@ -103,6 +103,7 @@ struct ext4_extent_header { };
#define EXT4_EXT_MAGIC cpu_to_le16(0xf30a) +#define EXT4_MAX_EXTENT_DEPTH 5
#define EXT4_EXTENT_TAIL_OFFSET(hdr) \ (sizeof(struct ext4_extent_header) + \ --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -851,6 +851,12 @@ ext4_ext_find_extent(struct inode *inode
eh = ext_inode_hdr(inode); depth = ext_depth(inode); + if (depth < 0 || depth > EXT4_MAX_EXTENT_DEPTH) { + EXT4_ERROR_INODE(inode, "inode has invalid extent depth: %d", + depth); + ret = -EIO; + goto err; + }
/* account possible depth increase */ if (!path) {