net: cadence: macb: Implement BQL

Implement byte queue limits to allow queuing disciplines to account for
packets enqueued in the ring buffer but not yet transmitted. There are a
separate set of transmit functions for AT91 that I haven't touched since
I don't have hardware to test on.

Signed-off-by: Sean Anderson <[e-mail]>
Link: https://patch.msgid.link/[e-mail]
Signed-off-by: Jakub Kicinski <[e-mail]>

drivers/net/ethernet/cadence/macb_main.c

···
 						      tx_error_task);
 	bool			halt_timeout = false;
 	struct macb		*bp = queue->bp;
+	u32			queue_index;
+	u32			packets = 0;
+	u32			bytes = 0;
 	struct macb_tx_skb	*tx_skb;
 	struct macb_dma_desc	*desc;
 	struct sk_buff		*skb;
 	unsigned int		tail;
 	unsigned long		flags;
 
+	queue_index = queue - bp->queues;
 	netdev_vdbg(bp->dev, "macb_tx_error_task: q = %u, t = %u, h = %u\n",
-		    (unsigned int)(queue - bp->queues),
-		    queue->tx_tail, queue->tx_head);
+		    queue_index, queue->tx_tail, queue->tx_head);
 
 	/* Prevent the queue NAPI TX poll from running, as it calls
 	 * macb_tx_complete(), which in turn may call netif_wake_subqueue().
···
 					    skb->data);
 				bp->dev->stats.tx_packets++;
 				queue->stats.tx_packets++;
+				packets++;
 				bp->dev->stats.tx_bytes += skb->len;
 				queue->stats.tx_bytes += skb->len;
+				bytes += skb->len;
 			}
 		} else {
 			/* "Buffers exhausted mid-frame" errors may only happen
···
 
 		macb_tx_unmap(bp, tx_skb, 0);
 	}
+
+	netdev_tx_completed_queue(netdev_get_tx_queue(bp->dev, queue_index),
+				  packets, bytes);
 
 	/* Set end of TX queue */
 	desc = macb_tx_desc(queue, 0);
···
 	unsigned int tail;
 	unsigned int head;
 	int packets = 0;
+	u32 bytes = 0;
 
 	spin_lock(&queue->tx_ptr_lock);
 	head = queue->tx_head;
···
 				bp->dev->stats.tx_bytes += skb->len;
 				queue->stats.tx_bytes += skb->len;
 				packets++;
+				bytes += skb->len;
 			}
 
 			/* Now we can safely release resources */
···
 				break;
 		}
 	}
+
+	netdev_tx_completed_queue(netdev_get_tx_queue(bp->dev, queue_index),
+				  packets, bytes);
 
 	queue->tx_tail = tail;
 	if (__netif_subqueue_stopped(bp->dev, queue_index) &&
···
 	/* Make newly initialized descriptor visible to hardware */
 	wmb();
 	skb_tx_timestamp(skb);
+	netdev_tx_sent_queue(netdev_get_tx_queue(bp->dev, queue_index),
+			     skb->len);
 
 	spin_lock_irq(&bp->lock);
 	macb_writel(bp, NCR, macb_readl(bp, NCR) | MACB_BIT(TSTART));
···
 	for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue) {
 		napi_disable(&queue->napi_rx);
 		napi_disable(&queue->napi_tx);
+		netdev_tx_reset_queue(netdev_get_tx_queue(dev, q));
 	}
 
 	phylink_stop(bp->phylink);