[PATCH net-next 04/13] net/mlx5: LAG, allocate port-indexed scratch buffers dynamically

[Tariq Toukan <tariqt@nvidia.com>]

From: Shay Drory <shayd@nvidia.com>

mlx5_lag_print_mapping(), mlx5_infer_tx_affinity_mapping() and
lag_active_port_bits() each used a MLX5_MAX_PORTS-sized stack buffer to
hold per-port working state. Allocate these buffers with
kcalloc(ldev->ports, ...) instead so they track the actual port count
and no longer depend on MLX5_MAX_PORTS.

lag_active_port_bits() can now fail on allocation, so it returns int and
propagates -ENOMEM; a non-negative return is the active-port bitmask as
before. Its two callers, mlx5_cmd_create_lag() and _mlx5_modify_lag(),
check for a negative return and bail out, then narrow the value back
into the u8 they hand to the device.

This removes these buffers' dependency on MLX5_MAX_PORTS.

Signed-off-by: Shay Drory <shayd@nvidia.com>
Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Reviewed-by: Akiva Goldberger <agoldberger@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
---
 .../net/ethernet/mellanox/mlx5/core/lag/lag.c | 100 ++++++++++++++----
 1 file changed, 80 insertions(+), 20 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c b/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c
index 00b0159cb422..93ead5664152 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c
@@ -63,19 +63,25 @@ static int get_port_sel_mode(enum mlx5_lag_mode mode, unsigned long flags)
 	return MLX5_LAG_PORT_SELECT_MODE_QUEUE_AFFINITY;
 }
 
-static u8 lag_active_port_bits(struct mlx5_lag *ldev,
-			       struct lag_tracker *tracker)
+static int lag_active_port_bits(struct mlx5_lag *ldev,
+				struct lag_tracker *tracker)
 {
-	u8 enabled_ports[MLX5_MAX_PORTS] = {};
 	u8 active_port = 0;
+	u8 *enabled_ports;
 	int num_enabled;
 	int idx;
 
+	enabled_ports = kcalloc(ldev->ports, sizeof(*enabled_ports),
+				GFP_KERNEL);
+	if (!enabled_ports)
+		return -ENOMEM;
+
 	mlx5_infer_tx_enabled(tracker, ldev, enabled_ports,
 			      &num_enabled);
 	for (idx = 0; idx < num_enabled; idx++)
 		active_port |= BIT_MASK(enabled_ports[idx]);
 
+	kfree(enabled_ports);
 	return active_port;
 }
 
@@ -105,13 +111,21 @@ static int mlx5_cmd_create_lag(struct mlx5_core_dev *dev, struct mlx5_lag *ldev,
 		MLX5_SET(lagc, lag_ctx, tx_remap_affinity_1, ports[idx0]);
 		MLX5_SET(lagc, lag_ctx, tx_remap_affinity_2, ports[idx1]);
 		break;
-	case MLX5_LAG_PORT_SELECT_MODE_PORT_SELECT_FT:
+	case MLX5_LAG_PORT_SELECT_MODE_PORT_SELECT_FT: {
+		u8 active_port;
+		int ret;
+
 		if (!MLX5_CAP_PORT_SELECTION(dev, port_select_flow_table_bypass))
 			break;
 
-		MLX5_SET(lagc, lag_ctx, active_port,
-			 lag_active_port_bits(ldev, tracker));
+		ret = lag_active_port_bits(ldev, tracker);
+		if (ret < 0)
+			return ret;
+
+		active_port = ret;
+		MLX5_SET(lagc, lag_ctx, active_port, active_port);
 		break;
+	}
 	default:
 		break;
 	}
@@ -240,22 +254,31 @@ static void mlx5_lag_print_mapping(struct mlx5_core_dev *dev,
 				   struct lag_tracker *tracker,
 				   unsigned long flags)
 {
-	char buf[MLX5_MAX_PORTS * 10 + 1] = {};
-	u8 enabled_ports[MLX5_MAX_PORTS] = {};
+	u8 *enabled_ports = NULL;
 	int written = 0;
 	int num_enabled;
+	char *buf;
 	int idx;
 	int err;
 	int i;
 	int j;
 
+	buf = kcalloc(ldev->ports * 10 + 1, sizeof(*buf), GFP_KERNEL);
+	if (!buf)
+		return;
+
 	if (test_bit(MLX5_LAG_MODE_FLAG_HASH_BASED, &flags)) {
+		enabled_ports = kcalloc(ldev->ports, sizeof(*enabled_ports),
+					GFP_KERNEL);
+		if (!enabled_ports)
+			goto free_buf;
+
 		mlx5_infer_tx_enabled(tracker, ldev, enabled_ports,
 				      &num_enabled);
 		for (i = 0; i < num_enabled; i++) {
 			err = scnprintf(buf + written, 4, "%d, ", enabled_ports[i] + 1);
 			if (err != 3)
-				return;
+				goto free_enabled;
 			written += err;
 		}
 		buf[written - 2] = 0;
@@ -267,12 +290,17 @@ static void mlx5_lag_print_mapping(struct mlx5_core_dev *dev,
 				err = scnprintf(buf + written, 10,
 						" port %d:%d", i + 1, ldev->v2p_map[idx]);
 				if (err != 9)
-					return;
+					goto free_enabled;
 				written += err;
 			}
 		}
 		mlx5_core_info(dev, "lag map:%s\n", buf);
 	}
+
+free_enabled:
+	kfree(enabled_ports);
+free_buf:
+	kfree(buf);
 }
 
 static int mlx5_lag_netdev_event(struct notifier_block *this,
@@ -668,20 +696,28 @@ static bool __mlx5_lag_is_sd_active(struct mlx5_lag *ldev,
  * If there are ports that are disabled fill the relevant slots
  * with mapping that points to active ports.
  */
-static void mlx5_infer_tx_affinity_mapping(struct lag_tracker *tracker,
-					   struct mlx5_lag *ldev,
-					   u8 buckets,
-					   u8 *ports)
+static int mlx5_infer_tx_affinity_mapping(struct lag_tracker *tracker,
+					  struct mlx5_lag *ldev,
+					  u8 buckets,
+					  u8 *ports)
 {
-	int disabled[MLX5_MAX_PORTS] = {};
-	int enabled[MLX5_MAX_PORTS] = {};
 	int disabled_ports_num = 0;
 	int enabled_ports_num = 0;
+	int *disabled;
+	int *enabled;
+	int err = 0;
 	int idx;
 	u32 rand;
 	int i;
 	int j;
 
+	enabled = kcalloc(ldev->ports, sizeof(*enabled), GFP_KERNEL);
+	disabled = kcalloc(ldev->ports, sizeof(*disabled), GFP_KERNEL);
+	if (!enabled || !disabled) {
+		err = -ENOMEM;
+		goto out;
+	}
+
 	mlx5_ldev_for_each(i, 0, ldev) {
 		if (tracker->netdev_state[i].tx_enabled &&
 		    tracker->netdev_state[i].link_up)
@@ -704,7 +740,7 @@ static void mlx5_infer_tx_affinity_mapping(struct lag_tracker *tracker,
 	/* If all ports are disabled/enabled keep native mapping */
 	if (enabled_ports_num == ldev->ports ||
 	    disabled_ports_num == ldev->ports)
-		return;
+		goto out;
 
 	/* Go over the disabled ports and for each assign a random active port */
 	for (i = 0; i < disabled_ports_num; i++) {
@@ -717,6 +753,11 @@ static void mlx5_infer_tx_affinity_mapping(struct lag_tracker *tracker,
 				mlx5_lag_xa_to_dev_idx(ldev, rand_xa_idx) + 1;
 		}
 	}
+
+out:
+	kfree(enabled);
+	kfree(disabled);
+	return err;
 }
 
 static bool mlx5_lag_has_drop_rule(struct mlx5_lag *ldev)
@@ -816,7 +857,10 @@ static int _mlx5_modify_lag(struct mlx5_lag *ldev,
 		    !MLX5_CAP_PORT_SELECTION(dev0, port_select_flow_table_bypass))
 			return ret;
 
-		active_ports = lag_active_port_bits(ldev, tracker);
+		ret = lag_active_port_bits(ldev, tracker);
+		if (ret < 0)
+			return ret;
+		active_ports = ret;
 
 		return mlx5_cmd_modify_active_port(dev0, active_ports);
 	}
@@ -879,7 +923,14 @@ void mlx5_modify_lag(struct mlx5_lag *ldev,
 	if (!ports)
 		return;
 
-	mlx5_infer_tx_affinity_mapping(tracker, ldev, ldev->buckets, ports);
+	err = mlx5_infer_tx_affinity_mapping(tracker, ldev, ldev->buckets,
+					     ports);
+	if (err) {
+		mlx5_core_err(dev0,
+			      "mlx5_infer_tx_affinity_mapping failed, err = %d\n",
+			      err);
+		goto out;
+	}
 
 	mlx5_ldev_for_each(i, 0, ldev) {
 		for (j = 0; j < ldev->buckets; j++) {
@@ -1045,7 +1096,16 @@ int mlx5_activate_lag(struct mlx5_lag *ldev,
 		return err;
 
 	if (mode != MLX5_LAG_MODE_MPESW) {
-		mlx5_infer_tx_affinity_mapping(tracker, ldev, ldev->buckets, ldev->v2p_map);
+		err = mlx5_infer_tx_affinity_mapping(tracker, ldev,
+						     ldev->buckets,
+						     ldev->v2p_map);
+		if (err) {
+			mlx5_core_err(dev0,
+				      "mlx5_infer_tx_affinity_mapping failed, err = %d\n",
+				      err);
+			return err;
+		}
+
 		if (test_bit(MLX5_LAG_MODE_FLAG_HASH_BASED, &flags)) {
 			err = mlx5_lag_port_sel_create(ldev, tracker->hash_type,
 						       ldev->v2p_map);
-- 
2.44.0