The test has two phases: first it runs a bdevperf with rw=verify to check data consistency and then runs bdevperf in the background while sending RPC requests causing the subsystem to be constanty paused/resumed. In-capsule data is set to 0 to make sure all IO requests are using zero-copy. Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com> Change-Id: I1c1d7eb04714c8506307cb95b6cbc5988c8946a3 Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/10797 Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com> |
||
---|---|---|
.. | ||
host | ||
target | ||
common.sh | ||
nvmf.sh | ||
README.md |
NVMe-oF test scripts
The test scripts in this directory hierarchy can be run in isolation by passing the --iso flag when running the test script. This will set up the RDMA NIC for testing and then tear it back down again when the test is completed.