| Message ID | 20211026154628.15787-1-marcin.szycik@linux.intel.com |
|---|---|
| State | Accepted |
| Delegated to: | Anthony Nguyen |
| Headers | show |
| Series | [net-next] ice: Clear synchronized addrs when adding VFs in switchdev mode | expand |
>-----Original Message----- >From: Intel-wired-lan <intel-wired-lan-bounces@osuosl.org> On Behalf Of >Marcin Szycik >Sent: Tuesday, October 26, 2021 9:16 PM >To: intel-wired-lan@lists.osuosl.org >Subject: [Intel-wired-lan] [PATCH net-next] ice: Clear synchronized addrs when >adding VFs in switchdev mode > >When spawning VFs in switchdev mode, internal filter list of vsi is cleared, >which includes MAC rules. However MAC entries stay on netdev's multicast >list, which causes error message when bringing link up after spawning VFs >("Failed to delete MAC filters"). __dev_mc_sync() is called and tries to unsync >addresses that were already removed internally when adding VFs. > >This can be reproduced with: >1) Load ice driver >2) Change PF to switchdev mode >3) Bring PF link up >4) Bring PF link down >5) Create a VF on PF >6) Bring PF link up > >Added clearing of netdev's multicast (and also unicast) list when spawning VFs >in switchdev mode, so the state of internal rule list and netdev's MAC list is >consistent. > >Fixes: 1a1c40df2e80 ("ice: set and release switchdev environment") >Signed-off-by: Marcin Szycik <marcin.szycik@linux.intel.com> >--- > drivers/net/ethernet/intel/ice/ice_eswitch.c | 6 ++++++ > 1 file changed, 6 insertions(+) > Tested-by: Sandeep Penigalapati <sandeep.penigalapati@intel.com>
diff --git a/drivers/net/ethernet/intel/ice/ice_eswitch.c b/drivers/net/ethernet/intel/ice/ice_eswitch.c index 6cb50653b18d..d1d7389b0bff 100644 --- a/drivers/net/ethernet/intel/ice/ice_eswitch.c +++ b/drivers/net/ethernet/intel/ice/ice_eswitch.c @@ -19,6 +19,7 @@ static int ice_eswitch_setup_env(struct ice_pf *pf) { struct ice_vsi *uplink_vsi = pf->switchdev.uplink_vsi; + struct net_device *uplink_netdev = uplink_vsi->netdev; struct ice_vsi *ctrl_vsi = pf->switchdev.control_vsi; struct ice_port_info *pi = pf->hw.port_info; bool rule_added = false; @@ -27,6 +28,11 @@ static int ice_eswitch_setup_env(struct ice_pf *pf) ice_remove_vsi_fltr(&pf->hw, uplink_vsi->idx); + netif_addr_lock_bh(uplink_netdev); + __dev_uc_unsync(uplink_netdev, NULL); + __dev_mc_unsync(uplink_netdev, NULL); + netif_addr_unlock_bh(uplink_netdev); + if (ice_vsi_add_vlan(uplink_vsi, 0, ICE_FWD_TO_VSI)) goto err_def_rx;
When spawning VFs in switchdev mode, internal filter list of vsi is cleared, which includes MAC rules. However MAC entries stay on netdev's multicast list, which causes error message when bringing link up after spawning VFs ("Failed to delete MAC filters"). __dev_mc_sync() is called and tries to unsync addresses that were already removed internally when adding VFs. This can be reproduced with: 1) Load ice driver 2) Change PF to switchdev mode 3) Bring PF link up 4) Bring PF link down 5) Create a VF on PF 6) Bring PF link up Added clearing of netdev's multicast (and also unicast) list when spawning VFs in switchdev mode, so the state of internal rule list and netdev's MAC list is consistent. Fixes: 1a1c40df2e80 ("ice: set and release switchdev environment") Signed-off-by: Marcin Szycik <marcin.szycik@linux.intel.com> --- drivers/net/ethernet/intel/ice/ice_eswitch.c | 6 ++++++ 1 file changed, 6 insertions(+)