14c0f94c88
verify.state files are not very useful because we fail the tests at any first failure. If there was something wrong with IO verification then autotest.sh would fail. Fio files with metrics output (IOPS, BW, latency, etc) are also not very useful as per-patch tests are not focused on benchmarking. Fixes #2546 Signed-off-by: Kamil Godzwon <kamilx.godzwon@intel.com> Change-Id: I33fbe2fb0749d840f08ffe0d606d206268140991 Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/13019 Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com> Reviewed-by: Karol Latecki <karol.latecki@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Changpeng Liu <changpeng.liu@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
53 lines
607 B
Plaintext
53 lines
607 B
Plaintext
[global]
|
|
thread=1
|
|
group_reporting=1
|
|
direct=1
|
|
norandommap=1
|
|
time_based=1
|
|
do_verify=1
|
|
verify=md5
|
|
verify_backlog=1024
|
|
iodepth=128
|
|
bs=4K
|
|
runtime=10
|
|
size=13%
|
|
verify_state_save=0
|
|
|
|
[job_randwrite]
|
|
rw=randwrite
|
|
name=randwrite
|
|
|
|
[job_randrw]
|
|
offset=13%
|
|
rw=randrw
|
|
name=randrw
|
|
|
|
[job_write]
|
|
offset=26%
|
|
rw=write
|
|
name=write
|
|
|
|
[job_rw]
|
|
offset=39%
|
|
rw=rw
|
|
name=rw
|
|
|
|
[job_unmap_trim_sequential]
|
|
offset=52%
|
|
rw=trim
|
|
trim_verify_zero=1
|
|
name=unmap_trim_sequential
|
|
|
|
[job_unmap_trim_random]
|
|
offset=65%
|
|
rw=randtrim
|
|
trim_verify_zero=1
|
|
name=unmap_trim_random
|
|
|
|
[job_unmap_write]
|
|
stonewall
|
|
offset=52%
|
|
size=26%
|
|
rw=randwrite
|
|
name=unmap_write
|