ivampiresp/Spdk
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
b36face57d
Spdk/test/nvmf/host/fio.sh

82 lines
2.9 KiB
Bash
Raw Normal View History

fio_plugin: add NVMe-oF host support in fio_plugin. Also adds the automation test case for using remote NVMe devices exported by NVMe-oF target. Change-Id: I2b839a4eeec33d5b0c30d654e6013ad8c7949e23 Signed-off-by: Ziye Yang <ziye.yang@intel.com>
2017-03-27 05:33:06 +00:00
#!/usr/bin/env bash
testdir=$(readlink -f $(dirname $0))
rootdir=$(readlink -f $testdir/../../..)
autotest_common.sh: move to test/common This is the first step in reorganizing the spdk test heirarchy. Change-Id: I83467653af2da2a53251649092bed6902d6cb86e Signed-off-by: Seth Howell <seth.howell@intel.com> Reviewed-on: https://review.gerrithub.io/401707 Tested-by: SPDK Automated Test System <sys_sgsw@intel.com> Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2018-02-27 22:14:08 +00:00
source $rootdir/test/common/autotest_common.sh
autotest:Add fio_plugin with nvmf lvol backend(Porting from DTS) To cover fio_plugin as host with nvmf lvol backend. Change-Id: I5c602d95421f1ab6d2dde660c1833c953c2e5c75 Signed-off-by: wenzhong wu <wenzhongx.wu@intel.com> Reviewed-on: https://review.gerrithub.io/390381 Reviewed-by: Changpeng Liu <changpeng.liu@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
2017-12-05 22:07:48 +00:00
source $rootdir/scripts/common.sh
fio_plugin: add NVMe-oF host support in fio_plugin. Also adds the automation test case for using remote NVMe devices exported by NVMe-oF target. Change-Id: I2b839a4eeec33d5b0c30d654e6013ad8c7949e23 Signed-off-by: Ziye Yang <ziye.yang@intel.com>
2017-03-27 05:33:06 +00:00
source $rootdir/test/nvmf/common.sh
rpc_py="python $rootdir/scripts/rpc.py"
set -e
test/nvmf: remove hardcoded NVMF ip address. This is the first in a series of patches to remove network configuration assumptions from the NVMF test suite. Change-Id: I9d8a9daab6910c5e80276db7e6701e1aa8e1d38f Signed-off-by: Seth Howell <seth.howell@intel.com> Reviewed-on: https://review.gerrithub.io/371193 Tested-by: SPDK Automated Test System <sys_sgsw@intel.com> Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2017-07-25 15:39:51 +00:00
RDMA_IP_LIST=$(get_available_rdma_ips)
NVMF_FIRST_TARGET_IP=$(echo "$RDMA_IP_LIST" | head -n 1)
if [ -z $NVMF_FIRST_TARGET_IP ]; then
echo "no NIC for nvmf test"
exit 0
fio_plugin: add NVMe-oF host support in fio_plugin. Also adds the automation test case for using remote NVMe devices exported by NVMe-oF target. Change-Id: I2b839a4eeec33d5b0c30d654e6013ad8c7949e23 Signed-off-by: Ziye Yang <ziye.yang@intel.com>
2017-03-27 05:33:06 +00:00
fi
test/nvmf/host/fio: skip test if FIO is unavailable Change-Id: I7a0e2d433a995d7815634fb60a970b722aa41dcd Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-04-21 17:15:28 +00:00
if [ ! -d /usr/src/fio ]; then
echo "FIO not available"
exit 0
fi
fio_plugin: add NVMe-oF host support in fio_plugin. Also adds the automation test case for using remote NVMe devices exported by NVMe-oF target. Change-Id: I2b839a4eeec33d5b0c30d654e6013ad8c7949e23 Signed-off-by: Ziye Yang <ziye.yang@intel.com>
2017-03-27 05:33:06 +00:00
timing_enter fio
test/nvmf: use stub application The stub application will ensure that each nvmf test does not need to reinitialized DPDK memory and NVMe devices. This drastically cuts down on the amount of time needed to run all of the nvmf tests. Change-Id: I6abad4e1298111884f18026e72e36f5d8b73c4b9 Signed-off-by: cunyinch <cunyin.chang@intel.com> Reviewed-on: https://review.gerrithub.io/362810 Reviewed-by: Jim Harris <james.r.harris@intel.com> Tested-by: SPDK Automated Test System <sys_sgsw@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2017-05-27 00:13:55 +00:00
timing_enter start_nvmf_tgt
fio_plugin: add NVMe-oF host support in fio_plugin. Also adds the automation test case for using remote NVMe devices exported by NVMe-oF target. Change-Id: I2b839a4eeec33d5b0c30d654e6013ad8c7949e23 Signed-off-by: Ziye Yang <ziye.yang@intel.com>
2017-03-27 05:33:06 +00:00
test/nvmf: Do not use config files Use rpc methods instead config file in nvmf tests Change-Id: I75354cae95e1dd5d0e08f5c694b369fb270b92a4 Signed-off-by: Pawel Niedzwiecki <pawelx.niedzwiecki@intel.com> Reviewed-on: https://review.gerrithub.io/419006 Reviewed-by: Pawel Kaminski <pawelx.kaminski@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-07-11 20:28:56 +00:00
$NVMF_APP -m 0xF -w &
fio_plugin: add NVMe-oF host support in fio_plugin. Also adds the automation test case for using remote NVMe devices exported by NVMe-oF target. Change-Id: I2b839a4eeec33d5b0c30d654e6013ad8c7949e23 Signed-off-by: Ziye Yang <ziye.yang@intel.com>
2017-03-27 05:33:06 +00:00
nvmfpid=$!
trap "killprocess $nvmfpid; exit 1" SIGINT SIGTERM EXIT
rpc: add default UNIX domain socket listen address RPC is a default feature required for almost all usages, so enable RPC by default, but with a UNIX domain socket for security reasons. -r can now be used from the command line to specify an alternative RPC listen address from the default /var/tmp/spdk.sock. Remove the Enable parameter from the Rpc config section but still allow specifying an alternative listen address using the Listen parameter as an alternative to the command line option. This keeps backward compatibility for this release for anyone using the configuration file still. Remove the Rpc sections from all configuration files that were using them, except for those that specified alternate TCP ports for multi-process test cases. We can fix these later to use an alternate UNIX domain socket and to use the command line instead. Signed-off-by: Jim Harris <james.r.harris@intel.com> Change-Id: Ife0d03fcab638c67b659f1eb85348ddc2b55c4c4 Reviewed-on: https://review.gerrithub.io/386561 Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com> Tested-by: SPDK Automated Test System <sys_sgsw@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2017-11-09 23:33:29 +00:00
waitforlisten $nvmfpid
test/nvmf: Do not use config files Use rpc methods instead config file in nvmf tests Change-Id: I75354cae95e1dd5d0e08f5c694b369fb270b92a4 Signed-off-by: Pawel Niedzwiecki <pawelx.niedzwiecki@intel.com> Reviewed-on: https://review.gerrithub.io/419006 Reviewed-by: Pawel Kaminski <pawelx.kaminski@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-07-11 20:28:56 +00:00
$rpc_py set_nvmf_target_options -u 8192 -p 4
$rpc_py start_subsystem_init
test/nvmf: use stub application The stub application will ensure that each nvmf test does not need to reinitialized DPDK memory and NVMe devices. This drastically cuts down on the amount of time needed to run all of the nvmf tests. Change-Id: I6abad4e1298111884f18026e72e36f5d8b73c4b9 Signed-off-by: cunyinch <cunyin.chang@intel.com> Reviewed-on: https://review.gerrithub.io/362810 Reviewed-by: Jim Harris <james.r.harris@intel.com> Tested-by: SPDK Automated Test System <sys_sgsw@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2017-05-27 00:13:55 +00:00
timing_exit start_nvmf_tgt
fio_plugin: add NVMe-oF host support in fio_plugin. Also adds the automation test case for using remote NVMe devices exported by NVMe-oF target. Change-Id: I2b839a4eeec33d5b0c30d654e6013ad8c7949e23 Signed-off-by: Ziye Yang <ziye.yang@intel.com>
2017-03-27 05:33:06 +00:00
nvmf: Remove direct mode There is now only virtual mode. Virtual mode has been improved enough to reach feature parity with direct mode and performance benchmarks show no degradation. Simplify the code by always using virtual mode. Change-Id: Id5cdb5d4d8c54e661b245ed7250c2f9d66ca2152 Signed-off-by: Ben Walker <benjamin.walker@intel.com> Reviewed-on: https://review.gerrithub.io/369496 Tested-by: SPDK Automated Test System <sys_sgsw@intel.com> Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-06-27 18:26:19 +00:00
bdevs="$bdevs $($rpc_py construct_malloc_bdev 64 512)"
nvmf: add AllowAnyHost option to subsystems The previous behavior with an empty host NQN whitelist was to allow any host to connect. Change-Id: I5401e52d96642cf20afe0d50c692613e67262edf Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com> Reviewed-on: https://review.gerrithub.io/376432 Tested-by: SPDK Automated Test System <sys_sgsw@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2017-08-30 20:21:12 +00:00
$rpc_py construct_nvmf_subsystem nqn.2016-06.io.spdk:cnode1 "trtype:RDMA traddr:$NVMF_FIRST_TARGET_IP trsvcid:4420" '' -a -s SPDK00000000000001 -n "$bdevs"
fio_plugin: add NVMe-oF host support in fio_plugin. Also adds the automation test case for using remote NVMe devices exported by NVMe-oF target. Change-Id: I2b839a4eeec33d5b0c30d654e6013ad8c7949e23 Signed-off-by: Ziye Yang <ziye.yang@intel.com>
2017-03-27 05:33:06 +00:00
fio_plugin: switch to LD_PRELOAD instead of dynamically loading fio has a race between reap_threads() and free_ioengine(). free_ioengine() will call the ioengine's cleanup routine and then dlclose it if it is dynamically linked (like the spdk fio plugin). free_ioengine() does not set td->io_ops = NULL though until after dlclose() is complete. If reap_threads() tries to dereference td->io_ops after our plugin has been closed but before io_ops was set to NULL, it will segfault. Solution (until an upstream fio fix is available) is to use LD_PRELOAD instead. Signed-off-by: Jim Harris <james.r.harris@intel.com> Change-Id: Ide4eb3cb92a636513289107fc211fdf1f98b616f Reviewed-on: https://review.gerrithub.io/365272 Tested-by: SPDK Automated Test System <sys_sgsw@intel.com> Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Ziye Yang <optimistyzy@gmail.com>
2017-06-13 18:50:32 +00:00
PLUGIN_DIR=$rootdir/examples/nvme/fio_plugin
autotest:Add fio_plugin with nvmf lvol backend(Porting from DTS) To cover fio_plugin as host with nvmf lvol backend. Change-Id: I5c602d95421f1ab6d2dde660c1833c953c2e5c75 Signed-off-by: wenzhong wu <wenzhongx.wu@intel.com> Reviewed-on: https://review.gerrithub.io/390381 Reviewed-by: Changpeng Liu <changpeng.liu@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
2017-12-05 22:07:48 +00:00
# Test fio_plugin as host with malloc backend
LD_PRELOAD=$PLUGIN_DIR/fio_plugin /usr/src/fio/fio $PLUGIN_DIR/example_config.fio --filename="trtype=RDMA adrfam=IPv4 \
traddr=$NVMF_FIRST_TARGET_IP trsvcid=4420 ns=1"
fio_plugin: add NVMe-oF host support in fio_plugin. Also adds the automation test case for using remote NVMe devices exported by NVMe-oF target. Change-Id: I2b839a4eeec33d5b0c30d654e6013ad8c7949e23 Signed-off-by: Ziye Yang <ziye.yang@intel.com>
2017-03-27 05:33:06 +00:00
$rpc_py delete_nvmf_subsystem nqn.2016-06.io.spdk:cnode1
autotest:Add fio_plugin with nvmf lvol backend(Porting from DTS) To cover fio_plugin as host with nvmf lvol backend. Change-Id: I5c602d95421f1ab6d2dde660c1833c953c2e5c75 Signed-off-by: wenzhong wu <wenzhongx.wu@intel.com> Reviewed-on: https://review.gerrithub.io/390381 Reviewed-by: Changpeng Liu <changpeng.liu@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
2017-12-05 22:07:48 +00:00
if [ $RUN_NIGHTLY -eq 1 ]; then
# Test fio_plugin as host with nvme lvol backend
bdfs=$(iter_pci_class_code 01 08 02)
$rpc_py construct_nvme_bdev -b Nvme0 -t PCIe -a $(echo $bdfs | awk '{ print $1 }')
ls_guid=$($rpc_py construct_lvol_store Nvme0n1 lvs_0)
get_lvs_free_mb $ls_guid
lb_guid=$($rpc_py construct_lvol_bdev -u $ls_guid lbd_0 $free_mb)
$rpc_py construct_nvmf_subsystem nqn.2016-06.io.spdk:cnode2 "trtype:RDMA traddr:$NVMF_FIRST_TARGET_IP trsvcid:4420" "" -a -s SPDK00000000000002 -n "$lb_guid"
LD_PRELOAD=$PLUGIN_DIR/fio_plugin /usr/src/fio/fio $PLUGIN_DIR/example_config.fio --filename="trtype=RDMA adrfam=IPv4 \
traddr=$NVMF_FIRST_TARGET_IP trsvcid=4420 ns=1"
$rpc_py delete_nvmf_subsystem nqn.2016-06.io.spdk:cnode2
# Test fio_plugin as host with nvme lvol nested backend
ls_nested_guid=$($rpc_py construct_lvol_store $lb_guid lvs_n_0)
get_lvs_free_mb $ls_nested_guid
lb_nested_guid=$($rpc_py construct_lvol_bdev -u $ls_nested_guid lbd_nest_0 $free_mb)
$rpc_py construct_nvmf_subsystem nqn.2016-06.io.spdk:cnode3 "trtype:RDMA traddr:$NVMF_FIRST_TARGET_IP trsvcid:4420" "" -a -s SPDK00000000000003 -n "$lb_nested_guid"
LD_PRELOAD=$PLUGIN_DIR/fio_plugin /usr/src/fio/fio $PLUGIN_DIR/example_config.fio --filename="trtype=RDMA adrfam=IPv4 \
traddr=$NVMF_FIRST_TARGET_IP trsvcid=4420 ns=1"
$rpc_py delete_nvmf_subsystem nqn.2016-06.io.spdk:cnode3
sync
# Delete lvol_bdev and destroy lvol_store.
lvol: add destroy_lvol_bdev RPC method This is the first step toward fixing the behavior of the lvol bdev module so that it handles spdk_bdev_unregister() correctly. Currently, when an lvol bdev is unregistered, this causes the associated lvol to be deleted; this isn't the desired behavior, since spdk_bdev_unregister() is just meant to drop the currently-exposed bdev, as in a hot-unplug event, not destroy the backing storage. The current implementation of the new destroy_lvol_bdev method still calls into the exsting (broken) spdk_bdev_unregister() path, but this at least lets us fix the existing delete_bdev RPC callers that really intended to destroy the lvol. The next step is to make vbdev_lvol_destroy() call spdk_lvol_destroy() instead of unregistering the bdev, which will require cleanup in the generic lvol layer to remove the close_only flag. Change-Id: Ic4cb51305d31f6d7366a82ba7bf7b19ebbcd3e9c Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com> Reviewed-on: https://review.gerrithub.io/408084 Tested-by: SPDK Automated Test System <sys_sgsw@intel.com> Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2018-04-11 22:12:13 +00:00
$rpc_py destroy_lvol_bdev "$lb_nested_guid"
autotest:Add fio_plugin with nvmf lvol backend(Porting from DTS) To cover fio_plugin as host with nvmf lvol backend. Change-Id: I5c602d95421f1ab6d2dde660c1833c953c2e5c75 Signed-off-by: wenzhong wu <wenzhongx.wu@intel.com> Reviewed-on: https://review.gerrithub.io/390381 Reviewed-by: Changpeng Liu <changpeng.liu@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
2017-12-05 22:07:48 +00:00
$rpc_py destroy_lvol_store -l lvs_n_0
lvol: add destroy_lvol_bdev RPC method This is the first step toward fixing the behavior of the lvol bdev module so that it handles spdk_bdev_unregister() correctly. Currently, when an lvol bdev is unregistered, this causes the associated lvol to be deleted; this isn't the desired behavior, since spdk_bdev_unregister() is just meant to drop the currently-exposed bdev, as in a hot-unplug event, not destroy the backing storage. The current implementation of the new destroy_lvol_bdev method still calls into the exsting (broken) spdk_bdev_unregister() path, but this at least lets us fix the existing delete_bdev RPC callers that really intended to destroy the lvol. The next step is to make vbdev_lvol_destroy() call spdk_lvol_destroy() instead of unregistering the bdev, which will require cleanup in the generic lvol layer to remove the close_only flag. Change-Id: Ic4cb51305d31f6d7366a82ba7bf7b19ebbcd3e9c Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com> Reviewed-on: https://review.gerrithub.io/408084 Tested-by: SPDK Automated Test System <sys_sgsw@intel.com> Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2018-04-11 22:12:13 +00:00
$rpc_py destroy_lvol_bdev "$lb_guid"
autotest:Add fio_plugin with nvmf lvol backend(Porting from DTS) To cover fio_plugin as host with nvmf lvol backend. Change-Id: I5c602d95421f1ab6d2dde660c1833c953c2e5c75 Signed-off-by: wenzhong wu <wenzhongx.wu@intel.com> Reviewed-on: https://review.gerrithub.io/390381 Reviewed-by: Changpeng Liu <changpeng.liu@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
2017-12-05 22:07:48 +00:00
$rpc_py destroy_lvol_store -l lvs_0
$rpc_py delete_bdev "Nvme0n1"
fi
fio_plugin: add NVMe-oF host support in fio_plugin. Also adds the automation test case for using remote NVMe devices exported by NVMe-oF target. Change-Id: I2b839a4eeec33d5b0c30d654e6013ad8c7949e23 Signed-off-by: Ziye Yang <ziye.yang@intel.com>
2017-03-27 05:33:06 +00:00
trap - SIGINT SIGTERM EXIT
killprocess $nvmfpid
timing_exit fio
Reference in New Issue Copy Permalink