general protection fault in shmem_free_inode

[syzbot]

Hello,

syzbot found the following crash on:

HEAD commit: dcae9fa1 ANDROID: squashfs: resolve merge conflict with 4...
git tree: android-4.9
console output: https://syzkaller.appspot.com/x/log.txt?x=161260ae400000
kernel config: https://syzkaller.appspot.com/x/.config?x=e4e70395f75b2239
dashboard link: https://syzkaller.appspot.com/bug?extid=9170b1e11929e38ee693
compiler: gcc (GCC) 8.0.1 20180413 (experimental)
syz repro: https://syzkaller.appspot.com/x/repro.syz?x=122d8421400000

IMPORTANT: if you fix the bug, please add the following tag to the commit:
Reported-by: syzbot+9170b1...@syzkaller.appspotmail.com

VFS: Busy inodes after unmount of tmpfs. Self-destruct in 5 seconds. Have
a nice day...
kasan: CONFIG_KASAN_INLINE enabled
kasan: GPF could be caused by NULL-ptr deref or user memory access
general protection fault: 0000 [#1] PREEMPT SMP KASAN
Dumping ftrace buffer:
(ftrace buffer empty)
Modules linked in:
CPU: 1 PID: 26958 Comm: syz-executor6 Not tainted 4.9.125+ #37
task: ffff8801d05b8000 task.stack: ffff8801c9788000
RIP: 0010:[<ffffffff81442144>] [<ffffffff81442144>]
shmem_free_inode.isra.1+0x24/0x90 mm/shmem.c:214
RSP: 0018:ffff8801c978fc50 EFLAGS: 00010206
RAX: dffffc0000000000 RBX: 0000000000000000 RCX: 1ffff10039183e09
RDX: 000000000000000c RSI: ffffffff8144212f RDI: 0000000000000060
RBP: ffff8801c978fc60 R08: ffff8801d05b88f8 R09: 0000000000000001
R10: 0000000000000000 R11: 0000000000000001 R12: ffffffff81448150
R13: ffff8801c8c1f090 R14: ffff8801c8c1f048 R15: ffff8801c8c1f048
FS: 00000000012d8940(0000) GS:ffff8801db700000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 000000000070cf70 CR3: 00000001d24c5000 CR4: 00000000001606b0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Stack:
ffff8801c7eb2200 ffffffff81448150 ffff8801c978fca8 ffffffff814482fd
ffff8801c978fc88 ffff8801c8c1f0b8 ffff8801c8c1f090 ffffffff81448150
ffff8801c8c1f168 ffff8801c8c1f0b8 ffffffff8289db20 ffff8801c978fce0
Call Trace:
[<ffffffff814482fd>] shmem_evict_inode+0x1ad/0x5c0 mm/shmem.c:1067
[<ffffffff81541e3e>] evict+0x22e/0x4f0 fs/inode.c:553
[<ffffffff81542f31>] iput_final fs/inode.c:1516 [inline]
[<ffffffff81542f31>] iput+0x371/0x900 fs/inode.c:1543
[<ffffffff815bfd31>] fsnotify_detach_mark+0x251/0x2f0 fs/notify/mark.c:170
[<ffffffff815c10ac>] fsnotify_detach_group_marks+0x5c/0xd0
fs/notify/mark.c:506
[<ffffffff815beae2>] fsnotify_destroy_group+0x62/0x120 fs/notify/group.c:70
[<ffffffff815c2827>] inotify_release+0x37/0x50
fs/notify/inotify/inotify_user.c:282
[<ffffffff814f06f3>] __fput+0x263/0x700 fs/file_table.c:208
[<ffffffff814f0c15>] ____fput+0x15/0x20 fs/file_table.c:244
[<ffffffff81138c4c>] task_work_run+0x10c/0x180 kernel/task_work.c:116
[<ffffffff81003e49>] tracehook_notify_resume include/linux/tracehook.h:191
[inline]
[<ffffffff81003e49>] exit_to_usermode_loop+0x129/0x150
arch/x86/entry/common.c:161
[<ffffffff8100570d>] prepare_exit_to_usermode arch/x86/entry/common.c:191
[inline]
[<ffffffff8100570d>] syscall_return_slowpath arch/x86/entry/common.c:260
[inline]
[<ffffffff8100570d>] do_syscall_64+0x35d/0x480 arch/x86/entry/common.c:287
[<ffffffff8278c193>] entry_SYSCALL_64_after_swapgs+0x5d/0xdb
Code: 1f 84 00 00 00 00 00 55 48 89 e5 41 54 53 48 89 fb e8 21 73 ec ff 48
8d 7b 60 48 b8 00 00 00 00 00 fc ff df 48 89 fa 48 c1 ea 03 <80> 3c 02 00
75 54 48 83 7b 60 00 75 0a e8 fa 72 ec ff 5b 41 5c
RIP [<ffffffff81442144>] shmem_free_inode.isra.1+0x24/0x90 mm/shmem.c:214
RSP <ffff8801c978fc50>
---[ end trace 5c366bd5137db2f6 ]---


---
This bug 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 bug report. See:
https://goo.gl/tpsmEJ#status for how to communicate with syzbot.
syzbot can test patches for this bug, for details see:
https://goo.gl/tpsmEJ#testing-patches