| Message ID | 20230607210410.88209-2-saeed@kernel.org (mailing list archive) |
|---|---|
| State | RFC |
| Headers | show |
| Series | [net-next,V2,01/14] RDMA/mlx5: Free second uplink ib port | expand |
Hello: This series was applied to netdev/net-next.git (main) by Saeed Mahameed <saeedm@nvidia.com>: On Wed, 7 Jun 2023 14:03:57 -0700 you wrote: > From: Shay Drory <shayd@nvidia.com> > > The cited patch introduce ib port for the slave device uplink in > case of multiport eswitch. However, this ib port didn't perform > anything when unloaded. > Unload the new ib port properly. > > [...] Here is the summary with links: - [net-next,V2,01/14] RDMA/mlx5: Free second uplink ib port https://git.kernel.org/netdev/net-next/c/962825e534a9 - [net-next,V2,02/14] {net/RDMA}/mlx5: introduce lag_for_each_peer https://git.kernel.org/netdev/net-next/c/222dd185833e - [net-next,V2,03/14] net/mlx5: LAG, check if all eswitches are paired for shared FDB https://git.kernel.org/netdev/net-next/c/4c103aea4bed - [net-next,V2,04/14] net/mlx5: LAG, generalize handling of shared FDB https://git.kernel.org/netdev/net-next/c/86a12124dc02 - [net-next,V2,05/14] net/mlx5: LAG, change mlx5_shared_fdb_supported() to static https://git.kernel.org/netdev/net-next/c/c83e6ab96ef2 - [net-next,V2,06/14] net/mlx5: LAG, block multipath LAG in case ldev have more than 2 ports https://git.kernel.org/netdev/net-next/c/d61bab396115 - [net-next,V2,07/14] net/mlx5: LAG, block multiport eswitch LAG in case ldev have more than 2 ports https://git.kernel.org/netdev/net-next/c/7718c1c8ac32 - [net-next,V2,08/14] net/mlx5: Enable 4 ports VF LAG https://git.kernel.org/netdev/net-next/c/6ec0b55e72a5 - [net-next,V2,09/14] net/mlx5e: Expose catastrophic steering error counters https://git.kernel.org/netdev/net-next/c/a33682e4e78e - [net-next,V2,10/14] net/mlx5e: Remove RX page cache leftovers https://git.kernel.org/netdev/net-next/c/f4692ab13a1f - [net-next,V2,11/14] net/mlx5e: TC, refactor access to hash key https://git.kernel.org/netdev/net-next/c/de1f0a650824 - [net-next,V2,12/14] net/mlx5: Skip inline mode check after mlx5_eswitch_enable_locked() failure https://git.kernel.org/netdev/net-next/c/97bd788efb90 - [net-next,V2,13/14] mlx5/core: E-Switch, Allocate ECPF vport if it's an eswitch manager https://git.kernel.org/netdev/net-next/c/eb8e9fae0a22 - [net-next,V2,14/14] net/mlx5e: simplify condition after napi budget handling change https://git.kernel.org/netdev/net-next/c/803ea346bd3f You are awesome, thank you!
diff --git a/drivers/infiniband/hw/mlx5/ib_rep.c b/drivers/infiniband/hw/mlx5/ib_rep.c index ddcfc116b19a..a4db22fe1883 100644 --- a/drivers/infiniband/hw/mlx5/ib_rep.c +++ b/drivers/infiniband/hw/mlx5/ib_rep.c @@ -126,7 +126,7 @@ mlx5_ib_vport_rep_unload(struct mlx5_eswitch_rep *rep) !mlx5_lag_is_master(mdev)) { struct mlx5_core_dev *peer_mdev; - if (rep->vport == MLX5_VPORT_UPLINK) + if (rep->vport == MLX5_VPORT_UPLINK && !mlx5_lag_is_mpesw(mdev)) return; peer_mdev = mlx5_lag_get_peer_mdev(mdev); vport_index += mlx5_eswitch_get_total_vports(peer_mdev); @@ -146,6 +146,9 @@ mlx5_ib_vport_rep_unload(struct mlx5_eswitch_rep *rep) struct mlx5_core_dev *peer_mdev; struct mlx5_eswitch *esw; + if (mlx5_lag_is_shared_fdb(mdev) && !mlx5_lag_is_master(mdev)) + return; + if (mlx5_lag_is_shared_fdb(mdev)) { peer_mdev = mlx5_lag_get_peer_mdev(mdev); esw = peer_mdev->priv.eswitch;