============================================
WARNING: possible recursive locking detected
syzkaller #0 Not tainted
--------------------------------------------
kworker/u8:5/10261 is trying to acquire lock:
ffff88810734bd58 (&qdisc_xmit_lock_key#4){+.-.}-{3:3}, at: sch_direct_xmit+0x15f/0x4c0

but task is already holding lock:
ffff888113385d58 (&qdisc_xmit_lock_key#4){+.-.}-{3:3}, at: sch_direct_xmit+0x15f/0x4c0

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

       CPU0
       ----
  lock(&qdisc_xmit_lock_key#4);
  lock(&qdisc_xmit_lock_key#4);

 *** DEADLOCK ***

 May be due to missing lock nesting notation

13 locks held by kworker/u8:5/10261:
 #0: ffff88802330e948 ((wq_completion)bond44){+.+.}-{0:0}, at: process_scheduled_works+0x9d4/0x17a0
 #1: ffffc9000292fb80 ((work_completion)(&(&bond->alb_work)->work)){+.+.}-{0:0}, at: process_scheduled_works+0xa0f/0x17a0
 #2: ffffffff8e341aa0 (rcu_read_lock){....}-{1:3}, at: bond_alb_monitor+0xf8/0x17e0
 #3: ffffffff8e341b00 (rcu_read_lock_bh){....}-{1:3}, at: __dev_queue_xmit+0x2aa/0x3230
 #4: ffff88816a86b228 (dev->qdisc_tx_busylock ?: &qdisc_tx_busylock#3){+...}-{3:3}, at: __dev_queue_xmit+0xf1a/0x3230
 #5: ffff888113385d58 (&qdisc_xmit_lock_key#4){+.-.}-{3:3}, at: sch_direct_xmit+0x15f/0x4c0
 #6: ffffffff8e341aa0 (rcu_read_lock){....}-{1:3}, at: ip_output+0x5b/0x450
 #7: ffffffff8e341aa0 (rcu_read_lock){....}-{1:3}, at: __icmp_send+0x216/0x1470
 #8: ffff888168699fe0 (k-slock-AF_INET){+.-.}-{3:3}, at: __icmp_send+0x9a1/0x1470
 #9: ffffffff8e341aa0 (rcu_read_lock){....}-{1:3}, at: ip_output+0x5b/0x450
 #10: ffffffff8e341aa0 (rcu_read_lock){....}-{1:3}, at: ip_finish_output2+0x44d/0x1180
 #11: ffffffff8e341b00 (rcu_read_lock_bh){....}-{1:3}, at: __dev_queue_xmit+0x2aa/0x3230
 #12: ffff888174cea228 (dev->qdisc_tx_busylock ?: &qdisc_tx_busylock#3){+...}-{3:3}, at: __dev_queue_xmit+0xf1a/0x3230

stack backtrace:
CPU: 0 UID: 0 PID: 10261 Comm: kworker/u8:5 Not tainted syzkaller #0 PREEMPT(full) 
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.2-debian-1.16.2-1 04/01/2014
Workqueue: bond44 bond_alb_monitor
Call Trace:
 <TASK>
 dump_stack_lvl+0x1a9/0x280
 print_deadlock_bug+0x279/0x290
 __lock_acquire+0x253f/0x2cf0
 lock_acquire+0x116/0x340
 _raw_spin_lock+0x2e/0x40
 sch_direct_xmit+0x15f/0x4c0
 __dev_queue_xmit+0x13b1/0x3230
 ip_finish_output2+0xd5d/0x1180
 ip_output+0x29f/0x450
 ip_push_pending_frames+0xbe/0x150
 __icmp_send+0x10c5/0x1470
 ip_fragment+0x17c/0x220
 ip_output+0x29f/0x450
 iptunnel_xmit+0x59f/0xa00
 ip_tunnel_xmit+0x1cea/0x2430
 __gre_xmit+0x19e/0x240
 erspan_xmit+0xa87/0x1440
 dev_hard_start_xmit+0x2cd/0x7f0
 sch_direct_xmit+0x251/0x4c0
 __dev_queue_xmit+0x13b1/0x3230
 alb_send_lp_vid+0x348/0x550
 alb_send_learning_packets+0x12c/0x300
 bond_alb_monitor+0x3d5/0x17e0
 process_scheduled_works+0xaec/0x17a0
 worker_thread+0x89f/0xd90
 kthread+0x726/0x8b0
 ret_from_fork+0x5b0/0xb20
 ret_from_fork_asm+0x1a/0x30
 </TASK>
netdevsim netdevsim1 eth0: unset [0, 0] type 1 family 0 port 8472 - 0
netdevsim netdevsim1 eth0: unset [1, 0] type 2 family 0 port 20000 - 0
netdevsim netdevsim1 eth1: unset [0, 0] type 1 family 0 port 8472 - 0
netdevsim netdevsim1 eth1: unset [1, 0] type 2 family 0 port 20000 - 0
netdevsim netdevsim1 eth2: unset [0, 0] type 1 family 0 port 8472 - 0
netdevsim netdevsim1 eth2: unset [1, 0] type 2 family 0 port 20000 - 0
netdevsim netdevsim1 eth3: unset [0, 0] type 1 family 0 port 8472 - 0
netdevsim netdevsim1 eth3: unset [1, 0] type 2 family 0 port 20000 - 0
