UBSAN: shift-out-of-bounds in diReadSpecial

[syzbot]

Hello,

syzbot found the following issue on:

HEAD commit: 614cb589 Merge tag 'acpi-5.11-rc1-2' of git://git.kernel.o..
git tree: upstream
console output: https://syzkaller.appspot.com/x/log.txt?x=10e4ffbd500000
kernel config: https://syzkaller.appspot.com/x/.config?x=bf519e1e96191576
dashboard link: https://syzkaller.appspot.com/bug?extid=d0d8d1bfd19a85de7703
compiler: gcc (GCC) 10.1.0-syz 20200507

Unfortunately, I don't have any reproducer for this issue yet.

IMPORTANT: if you fix the issue, please add the following tag to the commit:
Reported-by: syzbot+d0d8d1...@syzkaller.appspotmail.com

================================================================================
UBSAN: shift-out-of-bounds in fs/jfs/jfs_imap.c:445:7
shift exponent -19 is negative
CPU: 0 PID: 11254 Comm: syz-executor.2 Not tainted 5.10.0-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
Call Trace:
__dump_stack lib/dump_stack.c:79 [inline]
dump_stack+0x107/0x163 lib/dump_stack.c:120
ubsan_epilogue+0xb/0x5a lib/ubsan.c:148
__ubsan_handle_shift_out_of_bounds.cold+0xb1/0x181 lib/ubsan.c:395
diReadSpecial.cold+0x1b/0x3e fs/jfs/jfs_imap.c:445
jfs_mount+0x83/0x3d0 fs/jfs/jfs_mount.c:86
jfs_fill_super+0x5b1/0xbc0 fs/jfs/super.c:562
mount_bdev+0x34d/0x410 fs/super.c:1366
legacy_get_tree+0x105/0x220 fs/fs_context.c:592
vfs_get_tree+0x89/0x2f0 fs/super.c:1496
do_new_mount fs/namespace.c:2875 [inline]
path_mount+0x13ad/0x20c0 fs/namespace.c:3205
do_mount fs/namespace.c:3218 [inline]
__do_sys_mount fs/namespace.c:3426 [inline]
__se_sys_mount fs/namespace.c:3403 [inline]
__x64_sys_mount+0x27f/0x300 fs/namespace.c:3403
do_syscall_64+0x2d/0x70 arch/x86/entry/common.c:46
entry_SYSCALL_64_after_hwframe+0x44/0xa9
RIP: 0033:0x460b9a
Code: b8 a6 00 00 00 0f 05 48 3d 01 f0 ff ff 0f 83 ad 89 fb ff c3 66 2e 0f 1f 84 00 00 00 00 00 66 90 49 89 ca b8 a5 00 00 00 0f 05 <48> 3d 01 f0 ff ff 0f 83 8a 89 fb ff c3 66 0f 1f 84 00 00 00 00 00
RSP: 002b:00007f2ed35f4a78 EFLAGS: 00000202 ORIG_RAX: 00000000000000a5
RAX: ffffffffffffffda RBX: 00007f2ed35f4b10 RCX: 0000000000460b9a
RDX: 0000000020000000 RSI: 0000000020000100 RDI: 00007f2ed35f4ad0
RBP: 00007f2ed35f4ad0 R08: 00007f2ed35f4b10 R09: 0000000020000000
R10: 0000000000000000 R11: 0000000000000202 R12: 0000000020000000
R13: 0000000020000100 R14: 0000000020000200 R15: 000000002006d200
================================================================================


---
This report is generated by a bot. It may contain errors.
See https://goo.gl/tpsmEJ for more information about syzbot.
syzbot engineers can be reached at syzk...@googlegroups.com.

syzbot will keep track of this issue. See:
https://goo.gl/tpsmEJ#status for how to communicate with syzbot.

[Dave Kleikamp]

I suspect Randy Dunlap's fix for
https://syzkaller.appspot.com/bug?extid=36315852ece4132ec193
will take care of this one as well:
https://github.com/kleikamp/linux-shaggy/commit/3bef198f1b17d1bb89260bad947ef084c0a2d1a6

sbi->l2nbperpage is derived from sbi->l2bsize, which is sanity checked
by Randy's patch.

#syz dup: UBSAN: shift-out-of-bounds in chkSuper

Shaggy