[Bug] possible deadlock in ptp_clock_unregister

From: Sam Sun
Date: Mon Mar 25 2024 - 13:57:46 EST


Dear developers and maintainers,

We encountered a deadlock in function ptp_clock_unregister(). We
reproduced this bug on the latest upstream kernel(Linux 6.9-rc1). C
repro and kernel config are attached to this email. The bug report is
listed below.
```
============================================
WARNING: possible recursive locking detected
6.7.0-rc7 #1 Not tainted
--------------------------------------------
syz-executor367/8057 is trying to acquire lock:
ffff88801cff4860 (&ptp->n_vclocks_mux){+.+.}-{3:3}, at:
ptp_vclock_in_use drivers/ptp/ptp_private.h:103 [inline]
ffff88801cff4860 (&ptp->n_vclocks_mux){+.+.}-{3:3}, at:
ptp_clock_unregister+0x21/0x240 drivers/ptp/ptp_clock.c:386

but task is already holding lock:
ffff88801ca28860 (&ptp->n_vclocks_mux){+.+.}-{3:3}, at:
n_vclocks_store+0xd8/0x6e0 drivers/ptp/ptp_sysfs.c:216

other info that might help us debug this:
Possible unsafe locking scenario:

CPU0
----
lock(&ptp->n_vclocks_mux);
lock(&ptp->n_vclocks_mux);

*** DEADLOCK ***

May be due to missing lock nesting notation

4 locks held by syz-executor367/8057:
#0: ffff888016a0e418 (sb_writers#5){.+.+}-{0:0}, at:
ksys_write+0x122/0x250 fs/read_write.c:637
#1: ffff888016e5dc88 (&of->mutex){+.+.}-{3:3}, at:
kernfs_fop_write_iter+0x27c/0x500 fs/kernfs/file.c:325
#2: ffff888049f3acf8 (kn->active#63){.+.+}-{0:0}, at:
kernfs_fop_write_iter+0x2a0/0x500 fs/kernfs/file.c:326
#3: ffff88801ca28860 (&ptp->n_vclocks_mux){+.+.}-{3:3}, at:
n_vclocks_store+0xd8/0x6e0 drivers/ptp/ptp_sysfs.c:216

stack backtrace:
CPU: 0 PID: 8057 Comm: syz-executor367 Not tainted 6.7.0-rc7 #1
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
1.13.0-1ubuntu1.1 04/01/2014
Call Trace:
<TASK>
__dump_stack lib/dump_stack.c:88 [inline]
dump_stack_lvl+0xd9/0x150 lib/dump_stack.c:106
check_deadlock kernel/locking/lockdep.c:3062 [inline]
validate_chain kernel/locking/lockdep.c:3856 [inline]
__lock_acquire+0x2152/0x3bb0 kernel/locking/lockdep.c:5137
lock_acquire kernel/locking/lockdep.c:5754 [inline]
lock_acquire+0x1b1/0x530 kernel/locking/lockdep.c:5719
__mutex_lock_common kernel/locking/mutex.c:603 [inline]
__mutex_lock+0x147/0x940 kernel/locking/mutex.c:747
ptp_vclock_in_use drivers/ptp/ptp_private.h:103 [inline]
ptp_clock_unregister+0x21/0x240 drivers/ptp/ptp_clock.c:386
ptp_vclock_unregister+0x119/0x160 drivers/ptp/ptp_vclock.c:228
unregister_vclock+0x108/0x1a0 drivers/ptp/ptp_sysfs.c:178
device_for_each_child_reverse+0x114/0x180 drivers/base/core.c:3984
n_vclocks_store+0x498/0x6e0 drivers/ptp/ptp_sysfs.c:242
dev_attr_store+0x54/0x80 drivers/base/core.c:2366
sysfs_kf_write+0x114/0x170 fs/sysfs/file.c:136
kernfs_fop_write_iter+0x337/0x500 fs/kernfs/file.c:334
call_write_iter include/linux/fs.h:2020 [inline]
new_sync_write fs/read_write.c:491 [inline]
vfs_write+0x96a/0xd80 fs/read_write.c:584
ksys_write+0x122/0x250 fs/read_write.c:637
do_syscall_x64 arch/x86/entry/common.c:52 [inline]
do_syscall_64+0x40/0x110 arch/x86/entry/common.c:83
entry_SYSCALL_64_after_hwframe+0x63/0x6b
RIP: 0033:0x7f86812e5b2d
Code: 28 c3 e8 06 2a 00 00 66 0f 1f 44 00 00 f3 0f 1e fa 48 89 f8 48
89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d
01 f0 ff ff 73 01 c3 48 c7 c1 c0 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007ffda42cf588 EFLAGS: 00000246 ORIG_RAX: 0000000000000001
RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f86812e5b2d
RDX: 0000000000000015 RSI: 0000000020000000 RDI: 0000000000000004
RBP: 00007f868129f4e0 R08: 0000000000000014 R09: 0000000000000000
R10: 00007f868133201e R11: 0000000000000246 R12: 00007f868129f580
R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000
</TASK>
```
According to the report, ptp_clock_unregister() in ptp_clock.c takes
the mutex which was taken before by n_vclocks_store ptp_sysfs.c.

If you have any questions, please contact us.
Reported by: Yue Sun <samsun1006219@xxxxxxxxx>
Reported by: xingwei lee <xrivendell7@xxxxxxxxx>

Best Regards,
Yue

Attachment: config
Description: Binary data

// autogenerated by syzkaller (https://github.com/google/syzkaller)

#define _GNU_SOURCE

#include <endian.h>
#include <stdint.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <sys/syscall.h>
#include <sys/types.h>
#include <unistd.h>

uint64_t r[2] = {0xffffffffffffffff, 0xffffffffffffffff};

int main(void)
{
syscall(__NR_mmap, /*addr=*/0x1ffff000ul, /*len=*/0x1000ul, /*prot=*/0ul,
/*flags=MAP_FIXED|MAP_ANONYMOUS|MAP_PRIVATE*/ 0x32ul, /*fd=*/-1,
/*offset=*/0ul);
syscall(__NR_mmap, /*addr=*/0x20000000ul, /*len=*/0x1000000ul,
/*prot=PROT_WRITE|PROT_READ|PROT_EXEC*/ 7ul,
/*flags=MAP_FIXED|MAP_ANONYMOUS|MAP_PRIVATE*/ 0x32ul, /*fd=*/-1,
/*offset=*/0ul);
syscall(__NR_mmap, /*addr=*/0x21000000ul, /*len=*/0x1000ul, /*prot=*/0ul,
/*flags=MAP_FIXED|MAP_ANONYMOUS|MAP_PRIVATE*/ 0x32ul, /*fd=*/-1,
/*offset=*/0ul);
intptr_t res = 0;
memcpy((void*)0x20000000, "/sys/devices/virtual/ptp/ptp0/n_vclocks\000", 40);
res = syscall(__NR_openat, /*fd=*/0xffffffffffffff9cul, /*file=*/0x20000000ul,
/*flags=*/2ul, /*mode=*/0ul);
if (res != -1)
r[0] = res;
sprintf((char*)0x20000080, "%020llu", (long long)3);
*(uint8_t*)0x20000094 = 0;
syscall(__NR_write, /*fd=*/r[0], /*buf=*/0x20000080ul, /*len=*/0x15ul);
memcpy((void*)0x20000000, "/sys/devices/virtual/ptp/ptp0/n_vclocks\000", 40);
res = syscall(__NR_openat, /*fd=*/0xffffffffffffff9cul, /*file=*/0x20000000ul,
/*flags=*/2ul, /*mode=*/0ul);
if (res != -1)
r[1] = res;
sprintf((char*)0x20000080, "%020llu", (long long)0);
*(uint8_t*)0x20000094 = 0;
syscall(__NR_write, /*fd=*/r[1], /*buf=*/0x20000080ul, /*len=*/0x15ul);
return 0;
}