how to run fio verify reliably in linux

I am using fio over disks exposed through iscsi. I am giving these params in fio

fio --name=randwrite --ioengine=libaio --iodepth=64 --rw=randrw --rwmixread=50  --bs=4k-2M --direct=1 -filename=data --numjobs=1 --runtime 36000 --verify=md5 --verify_async=4 --verify_backlog=100000 --verify_dump=1 --verify_fatal=1 --time_based --group_reporting

With the above parameters can fio send overlap concurrent writes of size more than page size. If yes, then how does fio verify the checksum because atomicity of io is not guaranteed across page size.