1 files changed, 33 insertions, 19 deletions

diff --git a/drivers/net/ethernet/microsoft/mana/mana_en.c b/drivers/net/ethernet/microsoft/mana/mana_en.c
index d2f07e179e86..3d151700f658 100644
--- a/drivers/net/ethernet/microsoft/mana/mana_en.c
+++ b/drivers/net/ethernet/microsoft/mana/mana_en.c
@@ -599,7 +599,11 @@ static void mana_get_rxbuf_cfg(int mtu, u32 *datasize, u32 *alloc_size,
 	else
 		*headroom = XDP_PACKET_HEADROOM;
 
-	*alloc_size = mtu + MANA_RXBUF_PAD + *headroom;
+	*alloc_size = SKB_DATA_ALIGN(mtu + MANA_RXBUF_PAD + *headroom);
+
+	/* Using page pool in this case, so alloc_size is PAGE_SIZE */
+	if (*alloc_size < PAGE_SIZE)
+		*alloc_size = PAGE_SIZE;
 
 	*datasize = mtu + ETH_HLEN;
 }
@@ -1788,7 +1792,6 @@ static void mana_poll_rx_cq(struct mana_cq *cq)
 static int mana_cq_handler(void *context, struct gdma_queue *gdma_queue)
 {
 	struct mana_cq *cq = context;
-	u8 arm_bit;
 	int w;
 
 	WARN_ON_ONCE(cq->gdma_cq != gdma_queue);
@@ -1799,16 +1802,23 @@ static int mana_cq_handler(void *context, struct gdma_queue *gdma_queue)
 		mana_poll_tx_cq(cq);
 
 	w = cq->work_done;
-
-	if (w < cq->budget &&
-	    napi_complete_done(&cq->napi, w)) {
-		arm_bit = SET_ARM_BIT;
-	} else {
-		arm_bit = 0;
+	cq->work_done_since_doorbell += w;
+
+	if (w < cq->budget) {
+		mana_gd_ring_cq(gdma_queue, SET_ARM_BIT);
+		cq->work_done_since_doorbell = 0;
+		napi_complete_done(&cq->napi, w);
+	} else if (cq->work_done_since_doorbell >
+		   cq->gdma_cq->queue_size / COMP_ENTRY_SIZE * 4) {
+		/* MANA hardware requires at least one doorbell ring every 8
+		 * wraparounds of CQ even if there is no need to arm the CQ.
+		 * This driver rings the doorbell as soon as we have exceeded
+		 * 4 wraparounds.
+		 */
+		mana_gd_ring_cq(gdma_queue, 0);
+		cq->work_done_since_doorbell = 0;
 	}
 
-	mana_gd_ring_cq(gdma_queue, arm_bit);
-
 	return w;
 }
 
@@ -1862,10 +1872,12 @@ static void mana_destroy_txq(struct mana_port_context *apc)
 
 	for (i = 0; i < apc->num_queues; i++) {
 		napi = &apc->tx_qp[i].tx_cq.napi;
-		napi_synchronize(napi);
-		napi_disable(napi);
-		netif_napi_del(napi);
-
+		if (apc->tx_qp[i].txq.napi_initialized) {
+			napi_synchronize(napi);
+			napi_disable(napi);
+			netif_napi_del(napi);
+			apc->tx_qp[i].txq.napi_initialized = false;
+		}
 		mana_destroy_wq_obj(apc, GDMA_SQ, apc->tx_qp[i].tx_object);
 
 		mana_deinit_cq(apc, &apc->tx_qp[i].tx_cq);
@@ -1921,6 +1933,7 @@ static int mana_create_txq(struct mana_port_context *apc,
 		txq->ndev = net;
 		txq->net_txq = netdev_get_tx_queue(net, i);
 		txq->vp_offset = apc->tx_vp_offset;
+		txq->napi_initialized = false;
 		skb_queue_head_init(&txq->pending_skbs);
 
 		memset(&spec, 0, sizeof(spec));
@@ -1987,6 +2000,7 @@ static int mana_create_txq(struct mana_port_context *apc,
 
 		netif_napi_add_tx(net, &cq->napi, mana_poll);
 		napi_enable(&cq->napi);
+		txq->napi_initialized = true;
 
 		mana_gd_ring_cq(cq->gdma_cq, SET_ARM_BIT);
 	}
@@ -1998,7 +2012,7 @@ out:
 }
 
 static void mana_destroy_rxq(struct mana_port_context *apc,
-			     struct mana_rxq *rxq, bool validate_state)
+			     struct mana_rxq *rxq, bool napi_initialized)
 
 {
 	struct gdma_context *gc = apc->ac->gdma_dev->gdma_context;
@@ -2013,15 +2027,15 @@ static void mana_destroy_rxq(struct mana_port_context *apc,
 
 	napi = &rxq->rx_cq.napi;
 
-	if (validate_state)
+	if (napi_initialized) {
 		napi_synchronize(napi);
 
-	napi_disable(napi);
+		napi_disable(napi);
 
+		netif_napi_del(napi);
+	}
 	xdp_rxq_info_unreg(&rxq->xdp_rxq);
 
-	netif_napi_del(napi);
-
 	mana_destroy_wq_obj(apc, GDMA_RQ, rxq->rxobj);
 
 	mana_deinit_cq(apc, &rxq->rx_cq);