Description of problem: Allow to set smfs/dmfs modes for mellanox nic in os-net-config . ~~~ devlink dev param set pci/<pci-address> flow_steering_mode value smfs cmode runtime devlink dev eswitch set pci/<pci-address> mode switchdev ~~~ or this: ~~~ devlink dev param set pci/<pci-address> flow_steering_mode value dmfs cmode runtime devlink dev eswitch set pci/<pci-address> mode switchdev ~~~ (above is lifted from BZ 1882480 ) As mentioned in case asked in case 03106576 : "In troubleshooting this feature with RH and Nvidia engineering team, the customer has had to change from SMFS to DMFS steering mode. This change was accomplished manually, but the customer requires a way to override this mode and update it during deployments so the best steering mode may be utilized." In the previous BZ on this matter ( https://bugzilla.redhat.com/show_bug.cgi?id=1882480#c6 ), setting to smfs if switchdev and dmfs otherwise was decided upon . However, this does not address customer's actual need to selectively override SMFS or DMFS defaults for optimum performance . Version-Release number of selected component (if applicable): How reproducible: Steps to Reproduce: 1. 2. 3. Actual results: Expected results: Additional info:
Assigning to networking for dsneddon triage
From Ariel: https://review.opendev.org/c/openstack/os-net-config/+/824971
Verified in RHOS-16.2-RHEL-8-20220301.n.3 Without configuring steering mode in templates (default behaviour) lspci | grep Mella | grep -v Virtual 04:00.0 Ethernet controller: Mellanox Technologies MT27800 Family [ConnectX-5] 04:00.1 Ethernet controller: Mellanox Technologies MT27800 Family [ConnectX-5] devlink dev param show pci/0000:04:00.0 name flow_steering_mode pci/0000:04:00.0: name flow_steering_mode type driver-specific values: cmode runtime value smfs devlink dev param show pci/0000:04:00.1 name flow_steering_mode pci/0000:04:00.1: name flow_steering_mode type driver-specific values: cmode runtime value smfs devlink dev eswitch show pci/0000:04:00.1 pci/0000:04:00.1: mode switchdev inline-mode none encap-mode basic devlink dev eswitch show pci/0000:04:00.0 pci/0000:04:00.0: mode switchdev inline-mode none encap-mode basic Configuring in steering_mode in templates in the following way: - type: ovs_bridge name: br-link0 mtu: 9000 use_dhcp: false members: - type: linux_bond name: mx-bond mtu: 9000 bonding_options: "mode=active-backup" members: - type: sriov_pf name: nic11 numvfs: 10 primary: true promisc: true use_dhcp: false defroute: false link_mode: switchdev steering_mode: dmfs - type: sriov_pf name: nic12 numvfs: 10 promisc: true use_dhcp: false defroute: false link_mode: switchdev steering_mode: dmfs lspci | grep Mella | grep -v Virtual 04:00.0 Ethernet controller: Mellanox Technologies MT27800 Family [ConnectX-5] 04:00.1 Ethernet controller: Mellanox Technologies MT27800 Family [ConnectX-5] devlink dev param show pci/0000:04:00.0 name flow_steering_mode pci/0000:04:00.0: name flow_steering_mode type driver-specific values: cmode runtime value dmfs devlink dev param show pci/0000:04:00.1 name flow_steering_mode pci/0000:04:00.1: name flow_steering_mode type driver-specific values: cmode runtime value dmfs sudo devlink dev eswitch show pci/0000:04:00.1 pci/0000:04:00.1: mode switchdev inline-mode none encap-mode basic sudo devlink dev eswitch show pci/0000:04:00.0 pci/0000:04:00.0: mode switchdev inline-mode none encap-mode basic
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory (Release of components for Red Hat OpenStack Platform 16.2.2), and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2022:1001
The needinfo request[s] on this closed bug have been removed as they have been unresolved for 500 days