severe regression with random I/O from (at least) 4.14?

From: Paolo Valente
Date: Thu Dec 21 2017 - 05:58:01 EST

a few minutes ago I bumped into this apparent severe regression, with 4.15-rc4 and an SSD PLEXTOR PX-256M5S. If, with none as I/O scheduler, I do
fio --name=global --rw=randread --size=512m --name=job1

I get
read : io=524288KB, bw=34402KB/s, iops=8600, runt= 15240msec

This device had to reach at least 40 KIOPS. I retried with mq-deadline too, and then with a 4.14, same outcome in all cases. I asked some colleague to repeat the test with his/her device. Same outcome, so far, also with an SSD 850 EVO 250GB.

I'm sorry if I'm making some colossal mistake.