CVE-2023-53763

In the Linux kernel, the following vulnerability has been resolved:

Revert "f2fs: fix to do sanity check on extent cache correctly"

syzbot reports a f2fs bug as below:

UBSAN: array-index-out-of-bounds in fs/f2fs/f2fs.h:3275:19 index 1409 is out of range for type '__le32[923]' (aka 'unsigned int[923]') Call Trace: __dump_stack lib/dump_stack.c:88 [inline] dump_stack_lvl+0x1e7/0x2d0 lib/dump_stack.c:106 ubsan_epilogue lib/ubsan.c:217 [inline] __ubsan_handle_out_of_bounds+0x11c/0x150 lib/ubsan.c:348 inline_data_addr fs/f2fs/f2fs.h:3275 [inline] __recover_inline_status fs/f2fs/inode.c:113 [inline] do_read_inode fs/f2fs/inode.c:480 [inline] f2fs_iget+0x4730/0x48b0 fs/f2fs/inode.c:604 f2fs_fill_super+0x640e/0x80c0 fs/f2fs/super.c:4601 mount_bdev+0x276/0x3b0 fs/super.c:1391 legacy_get_tree+0xef/0x190 fs/fs_context.c:611 vfs_get_tree+0x8c/0x270 fs/super.c:1519 do_new_mount+0x28f/0xae0 fs/namespace.c:3335 do_mount fs/namespace.c:3675 [inline] __do_sys_mount fs/namespace.c:3884 [inline] __se_sys_mount+0x2d9/0x3c0 fs/namespace.c:3861 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd

The issue was bisected to:

commit d48a7b3a72f121655d95b5157c32c7d555e44c05 Author: Chao Yu [email protected] ↗ Date: Mon Jan 9 03:49:20 2023 +0000

f2fs: fix to do sanity check on extent cache correctly

The root cause is we applied both v1 and v2 of the patch, v2 is the right fix, so it needs to revert v1 in order to fix reported issue.

v1: commit d48a7b3a72f1 ("f2fs: fix to do sanity check on extent cache correctly") https://lore.kernel.org/lkml/[email protected]/

v2: commit 269d11948100 ("f2fs: fix to do sanity check on extent cache correctly") https://lore.kernel.org/lkml/[email protected]/

A low impact flaw was found in the Linux kernel's f2fs filesystem implementation. When mounting a specially crafted f2fs filesystem image, an out-of-bounds array access can occur in the f2fs inline-data handling (reported by UBSAN in inline_data_addr() in fs/f2fs/f2fs.h while reading an inode), which Red Hat assessed as having low availability impact only. The root cause was that upstream had applied both v1 and v2 of the same f2fs extent-cache sanity-check fix; the upstream fix reverted the incorrect v1 patch. All Red Hat Enterprise Linux kernel and kernel-rt streams, as well as Red Hat In-Vehicle OS, are not affected by this issue because the faulty v1 commit was never backported into Red Hat kernels (vulnerable code not present). While the upstream CVE record rates this issue as Important (CVSS 7.8) assuming high confidentiality, integrity, and availability impact, Red Hat rates it Low (CVSS 3.3): triggering the issue requires local access with privileges sufficient to mount an attacker-crafted f2fs image, and the assessed consequence is limited to availability.

The CVSS score(s) listed for this vulnerability do not reflect the associated product's status, and are included for informational purposes to better understand the severity of this vulnerability.