Hi Bart On 05/11/2018 11:26 PM, Bart Van Assche wrote: > The bug is in the above trace_printk() call: blk_rq_deadline() must only be used > for the legacy block layer and not for blk-mq code. If you have a look at the value > of the das.deadline field then one can see that the value of that field is correct: blk_rq_deadline is used by __blk_add_timer, and __blk_add_timer is also used by blk_mq_add_timer. This is why I used blk_rq_deadline to print out the value in my test. on 64bit system, das.deadline will be a correct value, what if the jiffies has crossed the 32bit boundary ? > 0xffff550c - 0xffff37c0 = 7500 = 30 * 250. Does that mean that HZ = 250 on the system > on which you ran this test? > Yes, the HZ is 250 Thanks Jianchao