RDMA/mlx5: Rely on RoCE fw cap instead of devlink when setting profile

When the RDMA auxiliary driver probes, it sets its profile based on
devlink driverinit value. The latter might not be in sync with FW yet
(In case devlink reload is not performed), thus causing a mismatch
between RDMA driver and FW. This results in the following FW syndrome
when the RDMA driver tries to adjust RoCE state, which fails the probe:

"0xC1F678 | modify_nic_vport_context: roce_en set on a vport that
doesn't support roce"

To prevent this, select the PF profile based on FW RoCE capability
instead of relying on devlink driverinit value.
To provide backward compatibility of the RoCE disable feature, on older
FW's where roce_rw is not set (FW RoCE capability is read-only), keep
the current behavior e.g., rely on devlink driverinit value.

Fixes: fbfa97b ("net/mlx5: Disable roce at HCA level")
Reviewed-by: Shay Drory <shayd@nvidia.com>
Reviewed-by: Michael Guralnik <michaelgur@nvidia.com>
Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Maher Sanalla <msanalla@nvidia.com>
Link: https://lore.kernel.org/r/cb34ce9a1df4a24c135cb804db87f7d2418bd6cc.1661763459.git.leonro@nvidia.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>

drivers/infiniband/hw/mlx5/main.c

@@ -4336,7 +4336,7 @@ static int mlx5r_probe(struct auxiliary_device *adev,
 	dev->mdev = mdev;
 	dev->num_ports = num_ports;
 
-	if (ll == IB_LINK_LAYER_ETHERNET && !mlx5_is_roce_init_enabled(mdev))
+	if (ll == IB_LINK_LAYER_ETHERNET && !mlx5_get_roce_state(mdev))
 		profile = &raw_eth_profile;
 	else
 		profile = &pf_profile;

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

@@ -494,6 +494,24 @@ static int max_uc_list_get_devlink_param(struct mlx5_core_dev *dev)
 	return err;
 }
 
+bool mlx5_is_roce_on(struct mlx5_core_dev *dev)
+{
+	struct devlink *devlink = priv_to_devlink(dev);
+	union devlink_param_value val;
+	int err;
+
+	err = devlink_param_driverinit_value_get(devlink,
+						 DEVLINK_PARAM_GENERIC_ID_ENABLE_ROCE,
+						 &val);
+
+	if (!err)
+		return val.vbool;
+
+	mlx5_core_dbg(dev, "Failed to get param. err = %d\n", err);
+	return MLX5_CAP_GEN(dev, roce);
+}
+EXPORT_SYMBOL(mlx5_is_roce_on);
+
 static int handle_hca_cap_2(struct mlx5_core_dev *dev, void *set_ctx)
 {
 	void *set_hca_cap;
@@ -597,7 +615,8 @@ static int handle_hca_cap(struct mlx5_core_dev *dev, void *set_ctx)
 			 MLX5_CAP_GEN_MAX(dev, num_total_dynamic_vf_msix));
 
 	if (MLX5_CAP_GEN(dev, roce_rw_supported))
-		MLX5_SET(cmd_hca_cap, set_hca_cap, roce, mlx5_is_roce_init_enabled(dev));
+		MLX5_SET(cmd_hca_cap, set_hca_cap, roce,
+			 mlx5_is_roce_on(dev));
 
 	max_uc_list = max_uc_list_get_devlink_param(dev);
 	if (max_uc_list > 0)
@@ -623,7 +642,7 @@ static int handle_hca_cap(struct mlx5_core_dev *dev, void *set_ctx)
  */
 static bool is_roce_fw_disabled(struct mlx5_core_dev *dev)
 {
-	return (MLX5_CAP_GEN(dev, roce_rw_supported) && !mlx5_is_roce_init_enabled(dev)) ||
+	return (MLX5_CAP_GEN(dev, roce_rw_supported) && !mlx5_is_roce_on(dev)) ||
 		(!MLX5_CAP_GEN(dev, roce_rw_supported) && !MLX5_CAP_GEN(dev, roce));
 }
 

include/linux/mlx5/driver.h

@@ -1279,16 +1279,17 @@ enum {
 	MLX5_TRIGGERED_CMD_COMP = (u64)1 << 32,
 };
 
-static inline bool mlx5_is_roce_init_enabled(struct mlx5_core_dev *dev)
+bool mlx5_is_roce_on(struct mlx5_core_dev *dev);
+
+static inline bool mlx5_get_roce_state(struct mlx5_core_dev *dev)
 {
-	struct devlink *devlink = priv_to_devlink(dev);
-	union devlink_param_value val;
-	int err;
-
-	err = devlink_param_driverinit_value_get(devlink,
-						 DEVLINK_PARAM_GENERIC_ID_ENABLE_ROCE,
-						 &val);
-	return err ? MLX5_CAP_GEN(dev, roce) : val.vbool;
+	if (MLX5_CAP_GEN(dev, roce_rw_supported))
+		return MLX5_CAP_GEN(dev, roce);
+
+	/* If RoCE cap is read-only in FW, get RoCE state from devlink
+	 * in order to support RoCE enable/disable feature
+	 */
+	return mlx5_is_roce_on(dev);
 }
 
 #endif /* MLX5_DRIVER_H */