Spdk/examples/nvmf/nvmf
Alexis Lescouet a71cd5214b event: Add a user option to change the size of spdk_msg_mempool
The spdk_msg_mempool structure has a fixed size, which is not flexible
enough. The size of the memory allocation can now be changed in the
options given to the spdk_app_start function.

Signed-off-by: Alexis Lescouet <alexis.lescouet@nutanix.com>
Change-Id: I1d6524ab8cf23f69f553aedb0f5b0cdc9dde374b
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/11635
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2022-03-09 08:00:28 +00:00
..
.gitignore nvmf_example: add an nvmf example 2019-12-20 10:03:34 +00:00
Makefile event: Shift subsystem initialization code to a separate library 2021-05-24 10:12:50 +00:00
nvmf.c event: Add a user option to change the size of spdk_msg_mempool 2022-03-09 08:00:28 +00:00
README.md Fix Markdown MD026 linter warnings - trailing punctuation in header 2020-02-17 10:07:21 +00:00

NVMe-oF target without SPDK event framework

Overview

This example is used to show how to use the nvmf lib. In this example we want to encourage user to use RPC cmd so we would only support RPC style.

Usage

This example's usage is very similar with nvmf_tgt, difference is that you must use the RPC cmd to setup the nvmf target.

First, start this example app. You can use the -m to specify how many cores you want to use. The other parameters you can use -h to show. ./nvmf -m 0xf -r /var/tmp/spdk.sock

Then, you need to use the RPC cmd to config the nvmf target. You can use the -h to get how many RPC cmd you can use. As this example is about nvmf so I think you can focus on the nvmf cmds and the bdev cmds. ./scripts/rpc.py -h

Next, You should use the RPC cmd to setup nvmf target. ./scripts/rpc.py nvmf_create_transport -t RDMA -g nvmf_example ./scripts/rpc.py nvmf_create_subsystem -t nvmf_example -s SPDK00000000000001 -a -m 32 nqn.2016-06.io.spdk:cnode1 ./scripts/rpc.py bdev_malloc_create -b Malloc1 128 512 ./scripts/rpc.py nvmf_subsystem_add_ns -t nvmf_example nqn.2016-06.io.spdk:cnode1 Malloc1 ./scripts/rpc.py nvmf_subsystem_add_listener -t rdma -f Ipv4 -a 192.168.0.10 -s 4420 -p nvmf_example nqn.2016-06.io.spdk:cnode1

Last, start the initiator to connect the nvmf example target and test the IOs $ROOT_SPDK/example/nvme/perf/perf -q 64 -o 4095 -w randrw -M 30 -l -t 60
-r "trtype:RDMA adrfam:IPv4 traddr:192.168.0.10 trsvcid:4420 subnqn:nqn.2016-06.io.spdk:cnode1"