]> git.baikalelectronics.ru Git - kernel.git/commit
f2fs: fix to do sanity check on i_extra_isize in is_alive()
authorChao Yu <chao@kernel.org>
Mon, 14 Nov 2022 16:08:47 +0000 (00:08 +0800)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Thu, 9 Feb 2023 10:28:27 +0000 (11:28 +0100)
commitb5de82d1bc895464b444d91ef337189de2445ad1
tree6b53b8b9c408cf786265b587b7370f60fd96cbb4
parent114858a06baf9759aa9eaff2b6ecbf06a4f00911
f2fs: fix to do sanity check on i_extra_isize in is_alive()

commit d3b7b4afd6b2c344eabf9cc26b8bfa903c164c7c upstream.

syzbot found a f2fs bug:

BUG: KASAN: slab-out-of-bounds in data_blkaddr fs/f2fs/f2fs.h:2891 [inline]
BUG: KASAN: slab-out-of-bounds in is_alive fs/f2fs/gc.c:1117 [inline]
BUG: KASAN: slab-out-of-bounds in gc_data_segment fs/f2fs/gc.c:1520 [inline]
BUG: KASAN: slab-out-of-bounds in do_garbage_collect+0x386a/0x3df0 fs/f2fs/gc.c:1734
Read of size 4 at addr ffff888076557568 by task kworker/u4:3/52

CPU: 1 PID: 52 Comm: kworker/u4:3 Not tainted 6.1.0-rc4-syzkaller-00362-gfef7fd48922d #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 10/26/2022
Workqueue: writeback wb_workfn (flush-7:0)
Call Trace:
<TASK>
__dump_stack lib/dump_stack.c:88 [inline]
dump_stack_lvl+0xcd/0x134 lib/dump_stack.c:106
print_address_description mm/kasan/report.c:284 [inline]
print_report+0x15e/0x45d mm/kasan/report.c:395
kasan_report+0xbb/0x1f0 mm/kasan/report.c:495
data_blkaddr fs/f2fs/f2fs.h:2891 [inline]
is_alive fs/f2fs/gc.c:1117 [inline]
gc_data_segment fs/f2fs/gc.c:1520 [inline]
do_garbage_collect+0x386a/0x3df0 fs/f2fs/gc.c:1734
f2fs_gc+0x88c/0x20a0 fs/f2fs/gc.c:1831
f2fs_balance_fs+0x544/0x6b0 fs/f2fs/segment.c:410
f2fs_write_inode+0x57e/0xe20 fs/f2fs/inode.c:753
write_inode fs/fs-writeback.c:1440 [inline]
__writeback_single_inode+0xcfc/0x1440 fs/fs-writeback.c:1652
writeback_sb_inodes+0x54d/0xf90 fs/fs-writeback.c:1870
wb_writeback+0x2c5/0xd70 fs/fs-writeback.c:2044
wb_do_writeback fs/fs-writeback.c:2187 [inline]
wb_workfn+0x2dc/0x12f0 fs/fs-writeback.c:2227
process_one_work+0x9bf/0x1710 kernel/workqueue.c:2289
worker_thread+0x665/0x1080 kernel/workqueue.c:2436
kthread+0x2e4/0x3a0 kernel/kthread.c:376
ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:306

The root cause is that we forgot to do sanity check on .i_extra_isize
in below path, result in accessing invalid address later, fix it.
- gc_data_segment
 - is_alive
  - data_blkaddr
   - offset_in_addr

Reported-by: syzbot+f8f3dfa4abc489e768a1@syzkaller.appspotmail.com
Link: https://lore.kernel.org/linux-f2fs-devel/0000000000003cb3c405ed5c17f9@google.com/T/#u
Signed-off-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
fs/f2fs/gc.c