2 files changed, 21 insertions, 11 deletions

diff --git a/drivers/net/ethernet/huawei/hinic/hinic_rx.c b/drivers/net/ethernet/huawei/hinic/hinic_rx.c
index 1d4f712b15a8..e2e5cdc7119c 100644
--- a/drivers/net/ethernet/huawei/hinic/hinic_rx.c
+++ b/drivers/net/ethernet/huawei/hinic/hinic_rx.c
@@ -26,6 +26,7 @@
 #include <linux/skbuff.h>
 #include <linux/dma-mapping.h>
 #include <linux/prefetch.h>
+#include <linux/cpumask.h>
 #include <asm/barrier.h>
 
 #include "hinic_common.h"
@@ -171,11 +172,10 @@ static int rx_alloc_pkts(struct hinic_rxq *rxq)
 	struct hinic_sge sge;
 	dma_addr_t dma_addr;
 	struct sk_buff *skb;
-	int i, alloc_more;
 	u16 prod_idx;
+	int i;
 
 	free_wqebbs = hinic_get_rq_free_wqebbs(rxq->rq);
-	alloc_more = 0;
 
 	/* Limit the allocation chunks */
 	if (free_wqebbs > nic_dev->rx_weight)
@@ -185,7 +185,6 @@ static int rx_alloc_pkts(struct hinic_rxq *rxq)
 		skb = rx_alloc_skb(rxq, &dma_addr);
 		if (!skb) {
 			netdev_err(rxq->netdev, "Failed to alloc Rx skb\n");
-			alloc_more = 1;
 			goto skb_out;
 		}
 
@@ -195,7 +194,6 @@ static int rx_alloc_pkts(struct hinic_rxq *rxq)
 					  &prod_idx);
 		if (!rq_wqe) {
 			rx_free_skb(rxq, skb, dma_addr);
-			alloc_more = 1;
 			goto skb_out;
 		}
 
@@ -211,9 +209,7 @@ skb_out:
 		hinic_rq_update(rxq->rq, prod_idx);
 	}
 
-	if (alloc_more)
-		tasklet_schedule(&rxq->rx_task);
-
+	tasklet_schedule(&rxq->rx_task);
 	return i;
 }
 
@@ -357,7 +353,7 @@ static int rxq_recv(struct hinic_rxq *rxq, int budget)
 	}
 
 	if (pkts)
-		tasklet_schedule(&rxq->rx_task); /* hinic_rx_alloc_pkts */
+		tasklet_schedule(&rxq->rx_task); /* rx_alloc_pkts */
 
 	u64_stats_update_begin(&rxq->rxq_stats.syncp);
 	rxq->rxq_stats.pkts += pkts;
@@ -417,6 +413,8 @@ static int rx_request_irq(struct hinic_rxq *rxq)
 	struct hinic_dev *nic_dev = netdev_priv(rxq->netdev);
 	struct hinic_hwdev *hwdev = nic_dev->hwdev;
 	struct hinic_rq *rq = rxq->rq;
+	struct hinic_qp *qp;
+	struct cpumask mask;
 	int err;
 
 	rx_add_napi(rxq);
@@ -432,7 +430,9 @@ static int rx_request_irq(struct hinic_rxq *rxq)
 		return err;
 	}
 
-	return 0;
+	qp = container_of(rq, struct hinic_qp, rq);
+	cpumask_set_cpu(qp->q_id % num_online_cpus(), &mask);
+	return irq_set_affinity_hint(rq->irq, &mask);
 }
 
 static void rx_free_irq(struct hinic_rxq *rxq)
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_tx.c b/drivers/net/ethernet/huawei/hinic/hinic_tx.c
index abe3e38cd342..9128858479c4 100644
--- a/drivers/net/ethernet/huawei/hinic/hinic_tx.c
+++ b/drivers/net/ethernet/huawei/hinic/hinic_tx.c
@@ -212,10 +212,19 @@ netdev_tx_t hinic_xmit_frame(struct sk_buff *skb, struct net_device *netdev)
 
 	sq_wqe = hinic_sq_get_wqe(txq->sq, wqe_size, &prod_idx);
 	if (!sq_wqe) {
-		tx_unmap_skb(nic_dev, skb, txq->sges);
-
 		netif_stop_subqueue(netdev, qp->q_id);
 
+		/* Check for the case free_tx_poll is called in another cpu
+		 * and we stopped the subqueue after free_tx_poll check.
+		 */
+		sq_wqe = hinic_sq_get_wqe(txq->sq, wqe_size, &prod_idx);
+		if (sq_wqe) {
+			netif_wake_subqueue(nic_dev->netdev, qp->q_id);
+			goto process_sq_wqe;
+		}
+
+		tx_unmap_skb(nic_dev, skb, txq->sges);
+
 		u64_stats_update_begin(&txq->txq_stats.syncp);
 		txq->txq_stats.tx_busy++;
 		u64_stats_update_end(&txq->txq_stats.syncp);
@@ -223,6 +232,7 @@ netdev_tx_t hinic_xmit_frame(struct sk_buff *skb, struct net_device *netdev)
 		goto flush_skbs;
 	}
 
+process_sq_wqe:
 	hinic_sq_prepare_wqe(txq->sq, prod_idx, sq_wqe, txq->sges, nr_sges);
 
 	hinic_sq_write_wqe(txq->sq, prod_idx, sq_wqe, skb, wqe_size);