Centos7 - 3.10.0-1062 Kernel Crash - during tracing

[tomaszrybicki]

OS: CentOS 7, 3.10.0-1062.el7.x86_64
Crash occurs when workload starts during tracing.

Stack:
BUG: unable to handle kernel NULL pointer dereference at 0000000000000098
IP: [] bio_complete_event+0xac/0x110 [iotrace]

[ 921.400604] Call Trace:
[ 921.400824]
[ 921.401007] [] ? deinit_tracers.isra.3+0x90/0x90 [iotrace]
[ 921.401683] [] iotrace_block_rq_complete+0x4c/0x70 [iotrace]
[ 921.402348] [] blk_update_request+0x116/0x360
[ 921.402885] [] scsi_end_request+0x34/0x1e0
[ 921.403419] [] scsi_io_completion+0x168/0x6a0
[ 921.403955] [] ? irq_exit+0xda/0x110
[ 921.404432] [] scsi_finish_command+0xdc/0x140
[ 921.404966] [] scsi_softirq_done+0x132/0x160
[ 921.405514] [] blk_done_softirq+0x96/0xc0
[ 921.406030] [] __do_softirq+0xf5/0x280
[ 921.406521] [] call_softirq+0x1c/0x30
[ 921.406995] [] do_softirq+0x65/0xa0
[ 921.407479] [] irq_exit+0x105/0x110
[ 921.407938] [] do_IRQ+0x56/0xf0
[ 921.408401] [] common_interrupt+0x162/0x162
[ 921.408914]
[ 921.409110] [] ? ata_sff_pio_task+0x99/0x1d0 [libata]
[ 921.409734] [] process_one_work+0x17f/0x440
[ 921.410271] [] worker_thread+0x126/0x3c0
[ 921.410756] [] ? manage_workers.isra.26+0x2a0/0x2a0
[ 921.411343] [] kthread+0xd1/0xe0
[ 921.411777] [] ? insert_kthread_work+0x40/0x40
[ 921.412338] [] ret_from_fork_nospec_begin+0x21/0x21
[ 921.412917] [] ? insert_kthread_work+0x40/0x40

I narrowed it down to null pointer dereference in bio_complete_event():
bio->bi_bdev
in
dev_id = disk_devt(IOTRACE_BIO_GET_DEV(bio));
(inside the macro)