ivampiresp/Spdk
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

test/nvmf: lower number of io queues in connect_disconnect nightly test

Browse Source 
By default nvme-cli will try to connect to subsystem using
number of io queues equal to number of available cores.
In case of more powerful CPUs and HT enabled this results in
more time needed to allocate resources.

Using "-i" option for nvme connect allows us to control the
number of IO queues and avoid timeout issues when testing.

Fixes issue #927

Change-Id: Ibb9b89b2d8be84fba29360db0f781cc99ae5c5b1
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/466389
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
This commit is contained in:
Karol Latecki 2019-08-27 09:59:24 +02:00 committed by Jim Harris
parent d40944a256
commit 02f1526086
1 changed files with 2 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
test/nvmf/target/connect_disconnect.sh
View File

@ -26,13 +26,14 @@ $rpc_py nvmf_subsystem_add_listener nqn.2016-06.io.spdk:cnode1 -t $TEST_TRANSPOR
if [ $RUN_NIGHTLY -eq 1 ]; then
num_iterations=200
IO_QUEUES="-i 8"
else
num_iterations=10
fi
set +x
for i in $(seq 1 $num_iterations); do
nvme connect -t $TEST_TRANSPORT -n "nqn.2016-06.io.spdk:cnode1" -a "$NVMF_FIRST_TARGET_IP" -s "$NVMF_PORT"
nvme connect -t $TEST_TRANSPORT -n "nqn.2016-06.io.spdk:cnode1" -a "$NVMF_FIRST_TARGET_IP" -s "$NVMF_PORT" "$IO_QUEUES"
waitforblk "nvme0n1"
nvme disconnect -n "nqn.2016-06.io.spdk:cnode1"
waitforblk_disconnect "nvme0n1"