[v5.15] BUG: sleeping function called from invalid context in sb_start_write

[syzbot]

Hello,

syzbot found the following issue on:

HEAD commit: 6139f2a02fe0 Linux 5.15.148
git tree: linux-5.15.y
console output: https://syzkaller.appspot.com/x/log.txt?x=16fe7e94180000
kernel config: https://syzkaller.appspot.com/x/.config?x=c170eb20d8be8542
dashboard link: https://syzkaller.appspot.com/bug?extid=3afbc278969bca713e9a
compiler: Debian clang version 15.0.6, GNU ld (GNU Binutils for Debian) 2.40
syz repro: https://syzkaller.appspot.com/x/repro.syz?x=16979b34180000

Downloadable assets:
disk image: https://storage.googleapis.com/syzbot-assets/2ac68f24aed3/disk-6139f2a0.raw.xz
vmlinux: https://storage.googleapis.com/syzbot-assets/e8982a3768c5/vmlinux-6139f2a0.xz
kernel image: https://storage.googleapis.com/syzbot-assets/eb272dd019ce/bzImage-6139f2a0.xz

IMPORTANT: if you fix the issue, please add the following tag to the commit:
Reported-by: syzbot+3afbc2...@syzkaller.appspotmail.com

BUG: sleeping function called from invalid context at include/linux/percpu-rwsem.h:49
in_atomic(): 1, irqs_disabled(): 0, non_block: 0, pid: 3536, name: udevd
no locks held by udevd/3536.
Preemption disabled at:
[<ffffffff8174f806>] is_module_text_address+0x16/0x140 kernel/module.c:4758
CPU: 1 PID: 3536 Comm: udevd Not tainted 5.15.148-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/25/2024
Call Trace:
<TASK>
__dump_stack lib/dump_stack.c:88 [inline]
dump_stack_lvl+0x1e3/0x2cb lib/dump_stack.c:106
___might_sleep+0x547/0x6a0 kernel/sched/core.c:9626
percpu_down_read include/linux/percpu-rwsem.h:49 [inline]
__sb_start_write include/linux/fs.h:1785 [inline]
sb_start_write+0x28/0x1c0 include/linux/fs.h:1855
mnt_want_write+0x3b/0x80 fs/namespace.c:377
vfs_utimes+0x1a4/0x760 fs/utimes.c:35
do_utimes_path fs/utimes.c:98 [inline]
do_utimes+0x1a4/0x270 fs/utimes.c:144
__do_sys_utimensat fs/utimes.c:163 [inline]
__se_sys_utimensat fs/utimes.c:147 [inline]
__x64_sys_utimensat+0x14b/0x250 fs/utimes.c:147
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x3d/0xb0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x61/0xcb
RIP: 0033:0x7f9975310fcc
Code: d8 64 89 02 48 83 c8 ff 89 ef 48 89 44 24 08 e8 99 00 fa ff 48 8b 44 24 08 48 83 c4 30 5d c3 c3 41 89 ca b8 18 01 00 00 0f 05 <48> 3d 00 f0 ff ff 76 10 48 8b 15 2d 6e 0d 00 f7 d8 64 89 02 48 83
RSP: 002b:00007fffdfa0c798 EFLAGS: 00000202 ORIG_RAX: 0000000000000118
RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f9975310fcc
RDX: 0000000000000000 RSI: 000055ba9fef3d10 RDI: 00000000ffffff9c
RBP: 0000000000000000 R08: 0000000000000000 R09: 000055ba9fef1b10
R10: 0000000000000000 R11: 0000000000000202 R12: 0000000000006180
R13: 0000000000000030 R14: 000055ba9fef3d10 R15: 000055ba9ff01cf0
</TASK>
BUG: scheduling while atomic: udevd/3536/0x00000002
no locks held by udevd/3536.
Modules linked in:
Preemption disabled at:
[<ffffffff8174f806>] is_module_text_address+0x16/0x140 kernel/module.c:4758


---
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.

If the report is already addressed, let syzbot know by replying with:
#syz fix: exact-commit-title

If you want syzbot to run the reproducer, reply with:
#syz test: git://repo/address.git branch-or-commit-hash
If you attach or paste a git patch, syzbot will apply it before testing.

If you want to overwrite report's subsystems, reply with:
#syz set subsystems: new-subsystem
(See the list of subsystem names on the web dashboard)

If the report is a duplicate of another one, reply with:
#syz dup: exact-subject-of-another-report

If you want to undo deduplication, reply with:
#syz undup