diff options
| author | Shay Drory <shayd@nvidia.com> | 2023-02-05 15:44:40 +0300 |
|---|---|---|
| committer | Saeed Mahameed <saeedm@nvidia.com> | 2023-06-02 22:10:47 +0300 |
| commit | 18e31d42267556fd98590d91dda161f2a39a1def (patch) | |
| tree | 27c0b74820e6b512d77284ff661b38110aede285 /drivers/net | |
| parent | 9be6c21fdcf8a7ec48262bb76f78c17ac2761ac6 (diff) | |
| download | linux-18e31d42267556fd98590d91dda161f2a39a1def.tar.xz | |
net/mlx5: E-switch, enlarge peer miss group table
There is an implicit assumption that peer miss group table
require to handle only a single peer.
Also, there is an assumption that total_vports of the master
is greater or equal to the total_vports of each peer.
Change the code to support peer miss group for more than one peer.
Signed-off-by: Shay Drory <shayd@nvidia.com>
Reviewed-by: Mark Bloch <mbloch@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Diffstat (limited to 'drivers/net')
| -rw-r--r-- | drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c | 7 |
1 files changed, 4 insertions, 3 deletions
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index a767f3d52c76..ca69ed487413 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -1573,6 +1573,7 @@ esw_create_peer_esw_miss_group(struct mlx5_eswitch *esw, u32 *flow_group_in, int *ix) { + int max_peer_ports = (esw->total_vports - 1) * (MLX5_MAX_PORTS - 1); int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); struct mlx5_flow_group *g; void *match_criteria; @@ -1599,8 +1600,8 @@ esw_create_peer_esw_miss_group(struct mlx5_eswitch *esw, MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, *ix); MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, - *ix + esw->total_vports - 1); - *ix += esw->total_vports; + *ix + max_peer_ports); + *ix += max_peer_ports + 1; g = mlx5_create_flow_group(fdb, flow_group_in); if (IS_ERR(g)) { @@ -1702,7 +1703,7 @@ static int esw_create_offloads_fdb_tables(struct mlx5_eswitch *esw) * total vports of the peer (currently is also uses esw->total_vports). */ table_size = MLX5_MAX_PORTS * (esw->total_vports * MAX_SQ_NVPORTS + MAX_PF_SQ) + - esw->total_vports * 2 + MLX5_ESW_MISS_FLOWS; + esw->total_vports * MLX5_MAX_PORTS + MLX5_ESW_MISS_FLOWS; /* create the slow path fdb with encap set, so further table instances * can be created at run time while VFs are probed if the FW allows that. |