dpll: fix possible deadlock during netlink dump operation

In the Linux kernel, the following vulnerability has been resolved:

dpll: fix possible deadlock during netlink dump operation

Recently, I've been hitting following deadlock warning during dpll pin

dump:

[52804.637962] ======================================================

[52804.638536] WARNING: possible circular locking dependency detected

[52804.639111] 6.8.0-rc2jiri+ #1 Not tainted

[52804.639529] ------------------------------------------------------

[52804.640104] python3/2984 is trying to acquire lock:

[52804.640581] ffff88810e642678 (nlk_cb_mutex-GENERIC){+.+.}-{3:3}, at: netlink_dump+0xb3/0x780

[52804.641417]

but task is already holding lock:

[52804.642010] ffffffff83bde4c8 (dpll_lock){+.+.}-{3:3}, at: dpll_lock_dumpit+0x13/0x20

[52804.642747]

which lock already depends on the new lock.

[52804.643551]

the existing dependency chain (in reverse order) is:

[52804.644259]

-> #1 (dpll_lock){+.+.}-{3:3}:

[52804.644836] lock_acquire+0x174/0x3e0

[52804.645271] __mutex_lock+0x119/0x1150

[52804.645723] dpll_lock_dumpit+0x13/0x20

[52804.646169] genl_start+0x266/0x320

[52804.646578] __netlink_dump_start+0x321/0x450

[52804.647056] genl_family_rcv_msg_dumpit+0x155/0x1e0

[52804.647575] genl_rcv_msg+0x1ed/0x3b0

[52804.648001] netlink_rcv_skb+0xdc/0x210

[52804.648440] genl_rcv+0x24/0x40

[52804.648831] netlink_unicast+0x2f1/0x490

[52804.649290] netlink_sendmsg+0x36d/0x660

[52804.649742] __sock_sendmsg+0x73/0xc0

[52804.650165] __sys_sendto+0x184/0x210

[52804.650597] __x64_sys_sendto+0x72/0x80

[52804.651045] do_syscall_64+0x6f/0x140

[52804.651474] entry_SYSCALL_64_after_hwframe+0x46/0x4e

[52804.652001]

-> #0 (nlk_cb_mutex-GENERIC){+.+.}-{3:3}:

[52804.652650] check_prev_add+0x1ae/0x1280

[52804.653107] __lock_acquire+0x1ed3/0x29a0

[52804.653559] lock_acquire+0x174/0x3e0

[52804.653984] __mutex_lock+0x119/0x1150

[52804.654423] netlink_dump+0xb3/0x780

[52804.654845] __netlink_dump_start+0x389/0x450

[52804.655321] genl_family_rcv_msg_dumpit+0x155/0x1e0

[52804.655842] genl_rcv_msg+0x1ed/0x3b0

[52804.656272] netlink_rcv_skb+0xdc/0x210

[52804.656721] genl_rcv+0x24/0x40

[52804.657119] netlink_unicast+0x2f1/0x490

[52804.657570] netlink_sendmsg+0x36d/0x660

[52804.658022] __sock_sendmsg+0x73/0xc0

[52804.658450] __sys_sendto+0x184/0x210

[52804.658877] __x64_sys_sendto+0x72/0x80

[52804.659322] do_syscall_64+0x6f/0x140

[52804.659752] entry_SYSCALL_64_after_hwframe+0x46/0x4e

[52804.660281]

other info that might help us debug this:

[52804.661077] Possible unsafe locking scenario:

[52804.661671] CPU0 CPU1

[52804.662129] ---- ----

[52804.662577] lock(dpll_lock);

[52804.662924] lock(nlk_cb_mutex-GENERIC);

[52804.663538] lock(dpll_lock);

[52804.664073] lock(nlk_cb_mutex-GENERIC);

[52804.664490]

The issue as follows: __netlink_dump_start() calls control->start(cb)

with nlk->cb_mutex held. In control->start(cb) the dpll_lock is taken.

Then nlk->cb_mutex is released and taken again in netlink_dump(), while

dpll_lock still being held. That leads to ABBA deadlock when another

CPU races with the same operation.

Fix this by moving dpll_lock taking into dumpit() callback which ensures

correct lock taking order.