net: remove 'fallback' argument from dev->ndo_select_queue()

After the previous patch, all the callers of ndo_select_queue()
provide as a 'fallback' argument netdev_pick_tx.
The only exceptions are nested calls to ndo_select_queue(),
which pass down the 'fallback' available in the current scope
- still netdev_pick_tx.

We can drop such argument and replace fallback() invocation with
netdev_pick_tx(). This avoids an indirect call per xmit packet
in some scenarios (TCP syn, UDP unconnected, XDP generic, pktgen)
with device drivers implementing such ndo. It also clean the code
a bit.

Tested with ixgbe and CONFIG_FCOE=m

With pktgen using queue xmit:
threads		vanilla 	patched
		(kpps)		(kpps)
1		2334		2428
2		4166		4278
4		7895		8100

 v1 -> v2:
 - rebased after helper's name change

Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

drivers/infiniband/hw/hfi1/vnic_main.c

@@ -423,8 +423,7 @@ tx_finish:
 
 static u16 hfi1_vnic_select_queue(struct net_device *netdev,
 				  struct sk_buff *skb,
-				  struct net_device *sb_dev,
-				  select_queue_fallback_t fallback)
+				  struct net_device *sb_dev)
 {
 	struct hfi1_vnic_vport_info *vinfo = opa_vnic_dev_priv(netdev);
 	struct opa_vnic_skb_mdata *mdata;

drivers/infiniband/ulp/opa_vnic/opa_vnic_netdev.c

@@ -95,8 +95,7 @@ static netdev_tx_t opa_netdev_start_xmit(struct sk_buff *skb,
 }
 
 static u16 opa_vnic_select_queue(struct net_device *netdev, struct sk_buff *skb,
-				 struct net_device *sb_dev,
-				 select_queue_fallback_t fallback)
+				 struct net_device *sb_dev)
 {
 	struct opa_vnic_adapter *adapter = opa_vnic_priv(netdev);
 	struct opa_vnic_skb_mdata *mdata;
@@ -106,8 +105,7 @@ static u16 opa_vnic_select_queue(struct net_device *netdev, struct sk_buff *skb,
 	mdata = skb_push(skb, sizeof(*mdata));
 	mdata->entropy = opa_vnic_calc_entropy(skb);
 	mdata->vl = opa_vnic_get_vl(adapter, skb);
-	rc = adapter->rn_ops->ndo_select_queue(netdev, skb,
-					       sb_dev, fallback);
+	rc = adapter->rn_ops->ndo_select_queue(netdev, skb, sb_dev);
 	skb_pull(skb, sizeof(*mdata));
 	return rc;
 }

drivers/net/bonding/bond_main.c

@@ -4114,8 +4114,7 @@ static inline int bond_slave_override(struct bonding *bond,
 
 
 static u16 bond_select_queue(struct net_device *dev, struct sk_buff *skb,
-			     struct net_device *sb_dev,
-			     select_queue_fallback_t fallback)
+			     struct net_device *sb_dev)
 {
 	/* This helper function exists to help dev_pick_tx get the correct
 	 * destination queue.  Using a helper function skips a call to

drivers/net/ethernet/amazon/ena/ena_netdev.c

@@ -2258,8 +2258,7 @@ error_drop_packet:
 }
 
 static u16 ena_select_queue(struct net_device *dev, struct sk_buff *skb,
-			    struct net_device *sb_dev,
-			    select_queue_fallback_t fallback)
+			    struct net_device *sb_dev)
 {
 	u16 qid;
 	/* we suspect that this is good for in--kernel network services that
@@ -2269,7 +2268,7 @@ static u16 ena_select_queue(struct net_device *dev, struct sk_buff *skb,
 	if (skb_rx_queue_recorded(skb))
 		qid = skb_get_rx_queue(skb);
 	else
-		qid = fallback(dev, skb, NULL);
+		qid = netdev_pick_tx(dev, skb, NULL);
 
 	return qid;
 }

drivers/net/ethernet/broadcom/bcmsysport.c

@@ -2274,8 +2274,7 @@ static const struct ethtool_ops bcm_sysport_ethtool_ops = {
 };
 
 static u16 bcm_sysport_select_queue(struct net_device *dev, struct sk_buff *skb,
-				    struct net_device *sb_dev,
-				    select_queue_fallback_t fallback)
+				    struct net_device *sb_dev)
 {
 	struct bcm_sysport_priv *priv = netdev_priv(dev);
 	u16 queue = skb_get_queue_mapping(skb);
@@ -2283,7 +2282,7 @@ static u16 bcm_sysport_select_queue(struct net_device *dev, struct sk_buff *skb,
 	unsigned int q, port;
 
 	if (!netdev_uses_dsa(dev))
-		return fallback(dev, skb, NULL);
+		return netdev_pick_tx(dev, skb, NULL);
 
 	/* DSA tagging layer will have configured the correct queue */
 	q = BRCM_TAG_GET_QUEUE(queue);
@@ -2291,7 +2290,7 @@ static u16 bcm_sysport_select_queue(struct net_device *dev, struct sk_buff *skb,
 	tx_ring = priv->ring_map[q + port * priv->per_port_num_tx_queues];
 
 	if (unlikely(!tx_ring))
-		return fallback(dev, skb, NULL);
+		return netdev_pick_tx(dev, skb, NULL);
 
 	return tx_ring->index;
 }

drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c

@@ -1909,8 +1909,7 @@ void bnx2x_netif_stop(struct bnx2x *bp, int disable_hw)
 }
 
 u16 bnx2x_select_queue(struct net_device *dev, struct sk_buff *skb,
-		       struct net_device *sb_dev,
-		       select_queue_fallback_t fallback)
+		       struct net_device *sb_dev)
 {
 	struct bnx2x *bp = netdev_priv(dev);
 
@@ -1932,7 +1931,7 @@ u16 bnx2x_select_queue(struct net_device *dev, struct sk_buff *skb,
 	}
 
 	/* select a non-FCoE queue */
-	return fallback(dev, skb, NULL) %
+	return netdev_pick_tx(dev, skb, NULL) %
 	       (BNX2X_NUM_ETH_QUEUES(bp) * bp->max_cos);
 }
 

drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h

@@ -498,8 +498,7 @@ int bnx2x_set_vf_spoofchk(struct net_device *dev, int idx, bool val);
 
 /* select_queue callback */
 u16 bnx2x_select_queue(struct net_device *dev, struct sk_buff *skb,
-		       struct net_device *sb_dev,
-		       select_queue_fallback_t fallback);
+		       struct net_device *sb_dev);
 
 static inline void bnx2x_update_rx_prod(struct bnx2x *bp,
 					struct bnx2x_fastpath *fp,

drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c

@@ -979,8 +979,7 @@ freeout:
 }
 
 static u16 cxgb_select_queue(struct net_device *dev, struct sk_buff *skb,
-			     struct net_device *sb_dev,
-			     select_queue_fallback_t fallback)
+			     struct net_device *sb_dev)
 {
 	int txq;
 
@@ -1022,7 +1021,7 @@ static u16 cxgb_select_queue(struct net_device *dev, struct sk_buff *skb,
 		return txq;
 	}
 
-	return fallback(dev, skb, NULL) % dev->real_num_tx_queues;
+	return netdev_pick_tx(dev, skb, NULL) % dev->real_num_tx_queues;
 }
 
 static int closest_timer(const struct sge *s, int time)

drivers/net/ethernet/hisilicon/hns/hns_enet.c

@@ -1964,8 +1964,7 @@ static void hns_nic_get_stats64(struct net_device *ndev,
 
 static u16
 hns_nic_select_queue(struct net_device *ndev, struct sk_buff *skb,
-		     struct net_device *sb_dev,
-		     select_queue_fallback_t fallback)
+		     struct net_device *sb_dev)
 {
 	struct ethhdr *eth_hdr = (struct ethhdr *)skb->data;
 	struct hns_nic_priv *priv = netdev_priv(ndev);
@@ -1975,7 +1974,7 @@ hns_nic_select_queue(struct net_device *ndev, struct sk_buff *skb,
 	    is_multicast_ether_addr(eth_hdr->h_dest))
 		return 0;
 	else
-		return fallback(ndev, skb, NULL);
+		return netdev_pick_tx(ndev, skb, NULL);
 }
 
 static const struct net_device_ops hns_nic_netdev_ops = {

drivers/net/ethernet/intel/ixgbe/ixgbe_main.c

@@ -8483,8 +8483,7 @@ static void ixgbe_atr(struct ixgbe_ring *ring,
 
 #ifdef IXGBE_FCOE
 static u16 ixgbe_select_queue(struct net_device *dev, struct sk_buff *skb,
-			      struct net_device *sb_dev,
-			      select_queue_fallback_t fallback)
+			      struct net_device *sb_dev)
 {
 	struct ixgbe_adapter *adapter;
 	struct ixgbe_ring_feature *f;
@@ -8514,7 +8513,7 @@ static u16 ixgbe_select_queue(struct net_device *dev, struct sk_buff *skb,
 			break;
 		/* fall through */
 	default:
-		return fallback(dev, skb, sb_dev);
+		return netdev_pick_tx(dev, skb, sb_dev);
 	}
 
 	f = &adapter->ring_feature[RING_F_FCOE];

drivers/net/ethernet/mellanox/mlx4/en_tx.c

@@ -685,16 +685,15 @@ static void build_inline_wqe(struct mlx4_en_tx_desc *tx_desc,
 }
 
 u16 mlx4_en_select_queue(struct net_device *dev, struct sk_buff *skb,
-			 struct net_device *sb_dev,
-			 select_queue_fallback_t fallback)
+			 struct net_device *sb_dev)
 {
 	struct mlx4_en_priv *priv = netdev_priv(dev);
 	u16 rings_p_up = priv->num_tx_rings_p_up;
 
 	if (netdev_get_num_tc(dev))
-		return fallback(dev, skb, NULL);
+		return netdev_pick_tx(dev, skb, NULL);
 
-	return fallback(dev, skb, NULL) % rings_p_up;
+	return netdev_pick_tx(dev, skb, NULL) % rings_p_up;
 }
 
 static void mlx4_bf_copy(void __iomem *dst, const void *src,

drivers/net/ethernet/mellanox/mlx4/mlx4_en.h

@@ -698,8 +698,7 @@ void mlx4_en_arm_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq);
 
 void mlx4_en_tx_irq(struct mlx4_cq *mcq);
 u16 mlx4_en_select_queue(struct net_device *dev, struct sk_buff *skb,
-			 struct net_device *sb_dev,
-			 select_queue_fallback_t fallback);
+			 struct net_device *sb_dev);
 netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev);
 netdev_tx_t mlx4_en_xmit_frame(struct mlx4_en_rx_ring *rx_ring,
 			       struct mlx4_en_rx_alloc *frame,

drivers/net/ethernet/mellanox/mlx5/core/en.h

@@ -769,8 +769,7 @@ struct mlx5e_profile {
 void mlx5e_build_ptys2ethtool_map(void);
 
 u16 mlx5e_select_queue(struct net_device *dev, struct sk_buff *skb,
-		       struct net_device *sb_dev,
-		       select_queue_fallback_t fallback);
+		       struct net_device *sb_dev);
 netdev_tx_t mlx5e_xmit(struct sk_buff *skb, struct net_device *dev);
 netdev_tx_t mlx5e_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb,
 			  struct mlx5e_tx_wqe *wqe, u16 pi);

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

@@ -110,11 +110,10 @@ static inline int mlx5e_get_dscp_up(struct mlx5e_priv *priv, struct sk_buff *skb
 #endif
 
 u16 mlx5e_select_queue(struct net_device *dev, struct sk_buff *skb,
-		       struct net_device *sb_dev,
-		       select_queue_fallback_t fallback)
+		       struct net_device *sb_dev)
 {
+	int channel_ix = netdev_pick_tx(dev, skb, NULL);
 	struct mlx5e_priv *priv = netdev_priv(dev);
-	int channel_ix = fallback(dev, skb, NULL);
 	u16 num_channels;
 	int up = 0;
 

drivers/net/ethernet/qlogic/qede/qede.h

@@ -498,8 +498,7 @@ struct qede_reload_args {
 /* Datapath functions definition */
 netdev_tx_t qede_start_xmit(struct sk_buff *skb, struct net_device *ndev);
 u16 qede_select_queue(struct net_device *dev, struct sk_buff *skb,
-		      struct net_device *sb_dev,
-		      select_queue_fallback_t fallback);
+		      struct net_device *sb_dev);
 netdev_features_t qede_features_check(struct sk_buff *skb,
 				      struct net_device *dev,
 				      netdev_features_t features);

drivers/net/ethernet/qlogic/qede/qede_fp.c

@@ -1696,8 +1696,7 @@ netdev_tx_t qede_start_xmit(struct sk_buff *skb, struct net_device *ndev)
 }
 
 u16 qede_select_queue(struct net_device *dev, struct sk_buff *skb,
-		      struct net_device *sb_dev,
-		      select_queue_fallback_t fallback)
+		      struct net_device *sb_dev)
 {
 	struct qede_dev *edev = netdev_priv(dev);
 	int total_txq;
@@ -1705,7 +1704,7 @@ u16 qede_select_queue(struct net_device *dev, struct sk_buff *skb,
 	total_txq = QEDE_TSS_COUNT(edev) * edev->dev_info.num_tc;
 
 	return QEDE_TSS_COUNT(edev) ?
-		fallback(dev, skb, NULL) % total_txq :  0;
+		netdev_pick_tx(dev, skb, NULL) % total_txq :  0;
 }
 
 /* 8B udp header + 8B base tunnel header + 32B option length */

drivers/net/ethernet/renesas/ravb_main.c

@@ -1615,8 +1615,7 @@ drop:
 }
 
 static u16 ravb_select_queue(struct net_device *ndev, struct sk_buff *skb,
-			     struct net_device *sb_dev,
-			     select_queue_fallback_t fallback)
+			     struct net_device *sb_dev)
 {
 	/* If skb needs TX timestamp, it is handled in network control queue */
 	return (skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) ? RAVB_NC :

drivers/net/ethernet/sun/ldmvsw.c

@@ -101,8 +101,7 @@ static struct vnet_port *vsw_tx_port_find(struct sk_buff *skb,
 }
 
 static u16 vsw_select_queue(struct net_device *dev, struct sk_buff *skb,
-			    struct net_device *sb_dev,
-			    select_queue_fallback_t fallback)
+			    struct net_device *sb_dev)
 {
 	struct vnet_port *port = netdev_priv(dev);
 

drivers/net/ethernet/sun/sunvnet.c

@@ -234,8 +234,7 @@ static struct vnet_port *vnet_tx_port_find(struct sk_buff *skb,
 }
 
 static u16 vnet_select_queue(struct net_device *dev, struct sk_buff *skb,
-			     struct net_device *sb_dev,
-			     select_queue_fallback_t fallback)
+			     struct net_device *sb_dev)
 {
 	struct vnet *vp = netdev_priv(dev);
 	struct vnet_port *port = __tx_port_find(vp, skb);

drivers/net/hyperv/netvsc_drv.c

@@ -308,7 +308,7 @@ static inline int netvsc_get_tx_queue(struct net_device *ndev,
  * If a valid queue has already been assigned, then use that.
  * Otherwise compute tx queue based on hash and the send table.
  *
- * This is basically similar to default (__netdev_pick_tx) with the added step
+ * This is basically similar to default (netdev_pick_tx) with the added step
  * of using the host send_table when no other queue has been assigned.
  *
  * TODO support XPS - but get_xps_queue not exported
@@ -331,8 +331,7 @@ static u16 netvsc_pick_tx(struct net_device *ndev, struct sk_buff *skb)
 }
 
 static u16 netvsc_select_queue(struct net_device *ndev, struct sk_buff *skb,
-			       struct net_device *sb_dev,
-			       select_queue_fallback_t fallback)
+			       struct net_device *sb_dev)
 {
 	struct net_device_context *ndc = netdev_priv(ndev);
 	struct net_device *vf_netdev;
@@ -344,10 +343,9 @@ static u16 netvsc_select_queue(struct net_device *ndev, struct sk_buff *skb,
 		const struct net_device_ops *vf_ops = vf_netdev->netdev_ops;
 
 		if (vf_ops->ndo_select_queue)
-			txq = vf_ops->ndo_select_queue(vf_netdev, skb,
-						       sb_dev, fallback);
+			txq = vf_ops->ndo_select_queue(vf_netdev, skb, sb_dev);
 		else
-			txq = fallback(vf_netdev, skb, NULL);
+			txq = netdev_pick_tx(vf_netdev, skb, NULL);
 
 		/* Record the queue selected by VF so that it can be
 		 * used for common case where VF has more queues than