net/mlx5: Lag, do bond only if slaves agree on roce state

[ Upstream commit 51ef930 ]

Currently, the driver does not enforce that lag bond slaves must have
matching roce capabilities. Yet, in mlx5_do_bond(), the driver attempts
to enable roce on all vports of the bond slaves, causing the following
syndrome when one slave has no roce fw support:

mlx5_cmd_out_err:809:(pid 25427): MODIFY_NIC_VPORT_CONTEXT(0×755) op_mod(0×0)
failed, status bad parameter(0×3), syndrome (0xc1f678), err(-22)

Thus, create HW lag only if bond's slaves agree on roce state,
either all slaves have roce support resulting in a roce lag bond,
or none do, resulting in a raw eth bond.

Fixes: 7907f23 ("net/mlx5: Implement RoCE LAG feature")
Signed-off-by: Maher Sanalla <[email protected]>
Signed-off-by: Tariq Toukan <[email protected]>
Reviewed-by: Simon Horman <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
Signed-off-by: Sasha Levin <[email protected]>

drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c

@@ -720,6 +720,7 @@ bool mlx5_lag_check_prereq(struct mlx5_lag *ldev)
 	struct mlx5_core_dev *dev;
 	u8 mode;
 #endif
+	bool roce_support;
 	int i;
 
 	for (i = 0; i < ldev->ports; i++)
@@ -746,6 +747,11 @@ bool mlx5_lag_check_prereq(struct mlx5_lag *ldev)
 		if (mlx5_sriov_is_enabled(ldev->pf[i].dev))
 			return false;
 #endif
+	roce_support = mlx5_get_roce_state(ldev->pf[MLX5_LAG_P1].dev);
+	for (i = 1; i < ldev->ports; i++)
+		if (mlx5_get_roce_state(ldev->pf[i].dev) != roce_support)
+			return false;
+
 	return true;
 }
 
@@ -913,8 +919,10 @@ static void mlx5_do_bond(struct mlx5_lag *ldev)
 		} else if (roce_lag) {
 			dev0->priv.flags &= ~MLX5_PRIV_FLAGS_DISABLE_IB_ADEV;
 			mlx5_rescan_drivers_locked(dev0);
-			for (i = 1; i < ldev->ports; i++)
-				mlx5_nic_vport_enable_roce(ldev->pf[i].dev);
+			for (i = 1; i < ldev->ports; i++) {
+				if (mlx5_get_roce_state(ldev->pf[i].dev))
+					mlx5_nic_vport_enable_roce(ldev->pf[i].dev);
+			}
 		} else if (shared_fdb) {
 			int i;