Taihsiang Ho has proposed merging ~taihsiangho/nvidia-dgx-2:peermem-ib-loopback-jammy-mr into nvidia-dgx-2:master.
Requested reviews: nvidia-dgx-2 (nvidia-dgx-2) For more details, see: https://code.launchpad.net/~taihsiangho/nvidia-dgx-2/+git/nvidia-dgx-2/+merge/436833 See the corresponding commit message for more details. It seems there are two loopback IBs connected to hot-koala, and I just pick up one of them for nvidia-peermem-test. I think we can "optimize" the IB network topology later after Taipei lab year end closure. Before that, we may talk about what IB network topology we want, and I will file a lab ticket for it if necessary. -- Your team nvidia-dgx-2 is requested to review the proposed merge of ~taihsiangho/nvidia-dgx-2:peermem-ib-loopback-jammy-mr into nvidia-dgx-2:master.
diff --git a/tests/nvidia-peermem-test/hosts.d/hot-koala b/tests/nvidia-peermem-test/hosts.d/hot-koala new file mode 100644 index 0000000..a76218b --- /dev/null +++ b/tests/nvidia-peermem-test/hosts.d/hot-koala @@ -0,0 +1,7 @@ +SERVER_IFACE=enp132s0 +SERVER_IP=192.168.5.1/24 +SERVER_IB_BDF=0000:84:00.0 + +CLIENT_IFACE=ens1 +CLIENT_IP=192.168.5.2/24 +CLIENT_IB_BDF=0000:05:00.0
-- Mailing list: https://launchpad.net/~sts-sponsors Post to : [email protected] Unsubscribe : https://launchpad.net/~sts-sponsors More help : https://help.launchpad.net/ListHelp
