[PATCH v4 7/7] nvme: support rdma transport type
Yi Zhang
yi.zhang at redhat.com
Fri Aug 14 11:45:38 EDT 2020
On 8/14/20 6:49 PM, Yi Zhang wrote:
>
>
> On 8/14/20 2:18 PM, Sagi Grimberg wrote:
>> Signed-off-by: Sagi Grimberg <sagi at grimberg.me>
>> ---
>> tests/nvme/rc | 15 +++++++++++++++
>> 1 file changed, 15 insertions(+)
>>
>> diff --git a/tests/nvme/rc b/tests/nvme/rc
>> index 3e97801bbb30..675acbfa7012 100644
>> --- a/tests/nvme/rc
>> +++ b/tests/nvme/rc
>> @@ -5,6 +5,7 @@
>> # Test specific to NVMe devices
>> . common/rc
>> +. common/multipath-over-rdma
>> def_traddr="127.0.0.1"
>> def_adrfam="ipv4"
>> @@ -25,6 +26,12 @@ _nvme_requires() {
>> _have_modules nvmet nvme-core nvme-tcp nvmet-tcp
>> _have_configfs
>> ;;
>> + rdma)
>> + _have_modules nvmet nvme-core nvme-rdma nvmet-rdma
>> + _have_configfs
>> + _have_program rdma
>> + _have_modules rdma_rxe siw
>
> start_soft_rdma can use siw or rdma_rxe(default one) module, but some
> old distros may not support siw, but suppor rdma_rxe.
> how about:
> _have_modules rdma_rxe || _have_modules siw
>> + ;;
>> *)
>> SKIP_REASON="unsupported nvme_trtype=${nvme_trtype}"
>> return 1
>> @@ -115,6 +122,9 @@ _cleanup_nvmet() {
>> modprobe -r nvmet-${nvme_trtype} 2>/dev/null
>> fi
>> modprobe -r nvmet 2>/dev/null
>> + if [[ "${nvme_trtype}" == "rdma" ]]; then
>> + stop_soft_rdma
>> + fi
>> }
>> _setup_nvmet() {
>> @@ -124,6 +134,11 @@ _setup_nvmet() {
>> modprobe nvmet-${nvme_trtype}
>> fi
>> modprobe nvme-${nvme_trtype}
>> + if [[ "${nvme_trtype}" == "rdma" ]]; then
>> + start_soft_rdma
>> + rdma_intfs=$(rdma_network_interfaces)
>> + def_traddr=$(get_ipv4_addr ${rdma_intfs[0]})
The first rdma_intfs here maybe have DOWN state, which doesn't have an
addr[1], I found similar check code here[2]
[1]
# echo $rdma_intfs
rdma_intfs:eno1
eno2
eno3
eno4
eno49
eno50
# [root at hpe-dl380gen9-01 blktests]# ip a s
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN
group default qlen 1000
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
valid_lft forever preferred_lft forever
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eno1: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc mq state
DOWN group default qlen 1000
link/ether 3c:a8:2a:21:7d:a4 brd ff:ff:ff:ff:ff:ff
3: eno49: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP
group default qlen 1000
link/ether 8c:dc:d4:1e:7a:78 brd ff:ff:ff:ff:ff:ff
inet 10.16.203.27/24 brd 10.16.203.255 scope global dynamic
noprefixroute eno49
valid_lft 84814sec preferred_lft 84814sec
inet6 2620:52:0:10cb:8edc:d4ff:fe1e:7a78/64 scope global dynamic
noprefixroute
valid_lft 2591978sec preferred_lft 604778sec
inet6 fe80::8edc:d4ff:fe1e:7a78/64 scope link noprefixroute
valid_lft forever preferred_lft forever
[2]
# grep -rin rdma_network_ tests/nvmeof-mp/
tests/nvmeof-mp/rc:93: for i in $(rdma_network_interfaces); do
tests/nvmeof-mp/rc:235: ) && for i in $(rdma_network_interfaces); do
>> + fi
>> }
>> _nvme_disconnect_ctrl() {
>
>
> _______________________________________________
> Linux-nvme mailing list
> Linux-nvme at lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-nvme
>
More information about the Linux-nvme
mailing list