gianfar: Bundle Rx allocation, cleanup

Use a more common consumer/ producer index design to improve
rx buffer allocation.  Instead of allocating a single new buffer
(skb) on each iteration, bundle the allocation of several rx
buffers at a time.  This also opens the path for further memory
optimizations.

Remove useless check of rxq->rfbptr, since this patch touches
rx pause frame handling code as well.  rxq->rfbptr is always
initialized as part of Rx BD ring init.
Remove redundant (and misleading) 'amount_pull' parameter.
Signed-off-by: Claudiu Manoil <claudiu.manoil@freescale.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

drivers/net/ethernet/freescale/gianfar.c

@@ -116,8 +116,8 @@ static int gfar_start_xmit(struct sk_buff *skb, struct net_device *dev);
 static void gfar_reset_task(struct work_struct *work);
 static void gfar_timeout(struct net_device *dev);
 static int gfar_close(struct net_device *dev);
-static struct sk_buff *gfar_new_skb(struct net_device *dev,
-				    dma_addr_t *bufaddr);
+static void gfar_alloc_rx_buffs(struct gfar_priv_rx_q *rx_queue,
+				int alloc_cnt);
 static int gfar_set_mac_address(struct net_device *dev);
 static int gfar_change_mtu(struct net_device *dev, int new_mtu);
 static irqreturn_t gfar_error(int irq, void *dev_id);
@@ -142,7 +142,7 @@ static void gfar_netpoll(struct net_device *dev);
 int gfar_clean_rx_ring(struct gfar_priv_rx_q *rx_queue, int rx_work_limit);
 static void gfar_clean_tx_ring(struct gfar_priv_tx_q *tx_queue);
 static void gfar_process_frame(struct net_device *dev, struct sk_buff *skb,
-			       int amount_pull, struct napi_struct *napi);
+			       struct napi_struct *napi);
 static void gfar_halt_nodisable(struct gfar_private *priv);
 static void gfar_clear_exact_match(struct net_device *dev);
 static void gfar_set_mac_for_addr(struct net_device *dev, int num,
@@ -169,17 +169,15 @@ static void gfar_init_rxbdp(struct gfar_priv_rx_q *rx_queue, struct rxbd8 *bdp,
 	bdp->lstatus = cpu_to_be32(lstatus);
 }
 
-static int gfar_init_bds(struct net_device *ndev)
+static void gfar_init_bds(struct net_device *ndev)
 {
 	struct gfar_private *priv = netdev_priv(ndev);
 	struct gfar __iomem *regs = priv->gfargrp[0].regs;
 	struct gfar_priv_tx_q *tx_queue = NULL;
 	struct gfar_priv_rx_q *rx_queue = NULL;
 	struct txbd8 *txbdp;
-	struct rxbd8 *rxbdp;
 	u32 __iomem *rfbptr;
 	int i, j;
-	dma_addr_t bufaddr;
 
 	for (i = 0; i < priv->num_tx_queues; i++) {
 		tx_queue = priv->tx_queue[i];
@@ -207,33 +205,18 @@ static int gfar_init_bds(struct net_device *ndev)
 	rfbptr = &regs->rfbptr0;
 	for (i = 0; i < priv->num_rx_queues; i++) {
 		rx_queue = priv->rx_queue[i];
-		rx_queue->cur_rx = rx_queue->rx_bd_base;
-		rx_queue->skb_currx = 0;
-		rxbdp = rx_queue->rx_bd_base;
 
-		for (j = 0; j < rx_queue->rx_ring_size; j++) {
-			struct sk_buff *skb = rx_queue->rx_skbuff[j];
-
-			if (skb) {
-				bufaddr = be32_to_cpu(rxbdp->bufPtr);
-			} else {
-				skb = gfar_new_skb(ndev, &bufaddr);
-				if (!skb) {
-					netdev_err(ndev, "Can't allocate RX buffers\n");
-					return -ENOMEM;
-				}
-				rx_queue->rx_skbuff[j] = skb;
-			}
+		rx_queue->next_to_clean = 0;
+		rx_queue->next_to_use = 0;
 
-			gfar_init_rxbdp(rx_queue, rxbdp, bufaddr);
-			rxbdp++;
-		}
+		/* make sure next_to_clean != next_to_use after this
+		 * by leaving at least 1 unused descriptor
+		 */
+		gfar_alloc_rx_buffs(rx_queue, gfar_rxbd_unused(rx_queue));
 
 		rx_queue->rfbptr = rfbptr;
 		rfbptr += 2;
 	}
-
-	return 0;
 }
 
 static int gfar_alloc_skb_resources(struct net_device *ndev)
@@ -311,8 +294,7 @@ static int gfar_alloc_skb_resources(struct net_device *ndev)
 			rx_queue->rx_skbuff[j] = NULL;
 	}
 
-	if (gfar_init_bds(ndev))
-		goto cleanup;
+	gfar_init_bds(ndev);
 
 	return 0;
 
@@ -1639,10 +1621,7 @@ static int gfar_restore(struct device *dev)
 		return 0;
 	}
 
-	if (gfar_init_bds(ndev)) {
-		free_skb_resources(priv);
-		return -ENOMEM;
-	}
+	gfar_init_bds(ndev);
 
 	gfar_mac_reset(priv);
 
@@ -2704,30 +2683,19 @@ static void gfar_clean_tx_ring(struct gfar_priv_tx_q *tx_queue)
 	netdev_tx_completed_queue(txq, howmany, bytes_sent);
 }
 
-static struct sk_buff *gfar_alloc_skb(struct net_device *dev)
+static struct sk_buff *gfar_new_skb(struct net_device *ndev,
+				    dma_addr_t *bufaddr)
 {
-	struct gfar_private *priv = netdev_priv(dev);
+	struct gfar_private *priv = netdev_priv(ndev);
 	struct sk_buff *skb;
+	dma_addr_t addr;
 
-	skb = netdev_alloc_skb(dev, priv->rx_buffer_size + RXBUF_ALIGNMENT);
+	skb = netdev_alloc_skb(ndev, priv->rx_buffer_size + RXBUF_ALIGNMENT);
 	if (!skb)
 		return NULL;
 
 	gfar_align_skb(skb);
 
-	return skb;
-}
-
-static struct sk_buff *gfar_new_skb(struct net_device *dev, dma_addr_t *bufaddr)
-{
-	struct gfar_private *priv = netdev_priv(dev);
-	struct sk_buff *skb;
-	dma_addr_t addr;
-
-	skb = gfar_alloc_skb(dev);
-	if (!skb)
-		return NULL;
-
 	addr = dma_map_single(priv->dev, skb->data,
 			      priv->rx_buffer_size, DMA_FROM_DEVICE);
 	if (unlikely(dma_mapping_error(priv->dev, addr))) {
@@ -2739,6 +2707,55 @@ static struct sk_buff *gfar_new_skb(struct net_device *dev, dma_addr_t *bufaddr)
 	return skb;
 }
 
+static void gfar_rx_alloc_err(struct gfar_priv_rx_q *rx_queue)
+{
+	struct gfar_private *priv = netdev_priv(rx_queue->dev);
+	struct gfar_extra_stats *estats = &priv->extra_stats;
+
+	netdev_err(rx_queue->dev, "Can't alloc RX buffers\n");
+	atomic64_inc(&estats->rx_alloc_err);
+}
+
+static void gfar_alloc_rx_buffs(struct gfar_priv_rx_q *rx_queue,
+				int alloc_cnt)
+{
+	struct net_device *ndev = rx_queue->dev;
+	struct rxbd8 *bdp, *base;
+	dma_addr_t bufaddr;
+	int i;
+
+	i = rx_queue->next_to_use;
+	base = rx_queue->rx_bd_base;
+	bdp = &rx_queue->rx_bd_base[i];
+
+	while (alloc_cnt--) {
+		struct sk_buff *skb = rx_queue->rx_skbuff[i];
+
+		if (likely(!skb)) {
+			skb = gfar_new_skb(ndev, &bufaddr);
+			if (unlikely(!skb)) {
+				gfar_rx_alloc_err(rx_queue);
+				break;
+			}
+		} else { /* restore from sleep state */
+			bufaddr = be32_to_cpu(bdp->bufPtr);
+		}
+
+		rx_queue->rx_skbuff[i] = skb;
+
+		/* Setup the new RxBD */
+		gfar_init_rxbdp(rx_queue, bdp, bufaddr);
+
+		/* Update to the next pointer */
+		bdp = next_bd(bdp, base, rx_queue->rx_ring_size);
+
+		if (unlikely(++i == rx_queue->rx_ring_size))
+			i = 0;
+	}
+
+	rx_queue->next_to_use = i;
+}
+
 static inline void count_errors(unsigned short status, struct net_device *dev)
 {
 	struct gfar_private *priv = netdev_priv(dev);
@@ -2838,7 +2855,7 @@ static inline void gfar_rx_checksum(struct sk_buff *skb, struct rxfcb *fcb)
 
 /* gfar_process_frame() -- handle one incoming packet if skb isn't NULL. */
 static void gfar_process_frame(struct net_device *dev, struct sk_buff *skb,
-			       int amount_pull, struct napi_struct *napi)
+			       struct napi_struct *napi)
 {
 	struct gfar_private *priv = netdev_priv(dev);
 	struct rxfcb *fcb = NULL;
@@ -2849,9 +2866,9 @@ static void gfar_process_frame(struct net_device *dev, struct sk_buff *skb,
 	/* Remove the FCB from the skb
 	 * Remove the padded bytes, if there are any
 	 */
-	if (amount_pull) {
+	if (priv->uses_rxfcb) {
 		skb_record_rx_queue(skb, fcb->rq);
-		skb_pull(skb, amount_pull);
+		skb_pull(skb, GMAC_FCB_LEN);
 	}
 
 	/* Get receive timestamp from the skb */
@@ -2895,27 +2912,30 @@ int gfar_clean_rx_ring(struct gfar_priv_rx_q *rx_queue, int rx_work_limit)
 	struct net_device *dev = rx_queue->dev;
 	struct rxbd8 *bdp, *base;
 	struct sk_buff *skb;
-	int pkt_len;
-	int amount_pull;
-	int howmany = 0;
+	int i, howmany = 0;
+	int cleaned_cnt = gfar_rxbd_unused(rx_queue);
 	struct gfar_private *priv = netdev_priv(dev);
 
 	/* Get the first full descriptor */
-	bdp = rx_queue->cur_rx;
 	base = rx_queue->rx_bd_base;
+	i = rx_queue->next_to_clean;
 
-	amount_pull = priv->uses_rxfcb ? GMAC_FCB_LEN : 0;
+	while (rx_work_limit--) {
 
-	while (!(be16_to_cpu(bdp->status) & RXBD_EMPTY) && rx_work_limit--) {
-		struct sk_buff *newskb;
-		dma_addr_t bufaddr;
+		if (cleaned_cnt >= GFAR_RX_BUFF_ALLOC) {
+			gfar_alloc_rx_buffs(rx_queue, cleaned_cnt);
+			cleaned_cnt = 0;
+		}
 
-		rmb();
+		bdp = &rx_queue->rx_bd_base[i];
+		if (be16_to_cpu(bdp->status) & RXBD_EMPTY)
+			break;
 
-		/* Add another skb for the future */
-		newskb = gfar_new_skb(dev, &bufaddr);
+		/* order rx buffer descriptor reads */
+		rmb();
 
-		skb = rx_queue->rx_skbuff[rx_queue->skb_currx];
+		/* fetch next to clean buffer from the ring */
+		skb = rx_queue->rx_skbuff[i];
 
 		dma_unmap_single(priv->dev, be32_to_cpu(bdp->bufPtr),
 				 priv->rx_buffer_size, DMA_FROM_DEVICE);
@@ -2924,30 +2944,26 @@ int gfar_clean_rx_ring(struct gfar_priv_rx_q *rx_queue, int rx_work_limit)
 			     be16_to_cpu(bdp->length) > priv->rx_buffer_size))
 			bdp->status = cpu_to_be16(RXBD_LARGE);
 
-		/* We drop the frame if we failed to allocate a new buffer */
-		if (unlikely(!newskb ||
-			     !(be16_to_cpu(bdp->status) & RXBD_LAST) ||
+		if (unlikely(!(be16_to_cpu(bdp->status) & RXBD_LAST) ||
 			     be16_to_cpu(bdp->status) & RXBD_ERR)) {
 			count_errors(be16_to_cpu(bdp->status), dev);
 
-			if (unlikely(!newskb)) {
-				newskb = skb;
-				bufaddr = be32_to_cpu(bdp->bufPtr);
-			} else if (skb)
+			/* discard faulty buffer */
 			dev_kfree_skb(skb);
+
 		} else {
 			/* Increment the number of packets */
 			rx_queue->stats.rx_packets++;
 			howmany++;
 
 			if (likely(skb)) {
-				pkt_len = be16_to_cpu(bdp->length) -
+				int pkt_len = be16_to_cpu(bdp->length) -
 					  ETH_FCS_LEN;
 				/* Remove the FCS from the packet length */
 				skb_put(skb, pkt_len);
 				rx_queue->stats.rx_bytes += pkt_len;
 				skb_record_rx_queue(skb, rx_queue->qindex);
-				gfar_process_frame(dev, skb, amount_pull,
+				gfar_process_frame(dev, skb,
 						   &rx_queue->grp->napi_rx);
 
 			} else {
@@ -2958,26 +2974,23 @@ int gfar_clean_rx_ring(struct gfar_priv_rx_q *rx_queue, int rx_work_limit)
 
 		}
 
-		rx_queue->rx_skbuff[rx_queue->skb_currx] = newskb;
+		rx_queue->rx_skbuff[i] = NULL;
+		cleaned_cnt++;
+		if (unlikely(++i == rx_queue->rx_ring_size))
+			i = 0;
+	}
 
-		/* Setup the new bdp */
-		gfar_init_rxbdp(rx_queue, bdp, bufaddr);
+	rx_queue->next_to_clean = i;
+
+	if (cleaned_cnt)
+		gfar_alloc_rx_buffs(rx_queue, cleaned_cnt);
 
 	/* Update Last Free RxBD pointer for LFC */
-		if (unlikely(rx_queue->rfbptr && priv->tx_actual_en))
+	if (unlikely(priv->tx_actual_en)) {
+		bdp = gfar_rxbd_lastfree(rx_queue);
 		gfar_write(rx_queue->rfbptr, (u32)bdp);
-
-		/* Update to the next pointer */
-		bdp = next_bd(bdp, base, rx_queue->rx_ring_size);
-
-		/* update to point at the next skb */
-		rx_queue->skb_currx = (rx_queue->skb_currx + 1) &
-				      RX_RING_MOD_MASK(rx_queue->rx_ring_size);
 	}
 
-	/* Update the current rxbd pointer to be the next one */
-	rx_queue->cur_rx = bdp;
-
 	return howmany;
 }
 
@@ -3552,13 +3565,7 @@ static noinline void gfar_update_link_state(struct gfar_private *priv)
 		if ((tempval1 & MACCFG1_TX_FLOW) && !tx_flow_oldval) {
 			for (i = 0; i < priv->num_rx_queues; i++) {
 				rx_queue = priv->rx_queue[i];
-				bdp = rx_queue->cur_rx;
-				/* skip to previous bd */
-				bdp = skip_bd(bdp, rx_queue->rx_ring_size - 1,
-					      rx_queue->rx_bd_base,
-					      rx_queue->rx_ring_size);
-
-				if (rx_queue->rfbptr)
+				bdp = gfar_rxbd_lastfree(rx_queue);
 				gfar_write(rx_queue->rfbptr, (u32)bdp);
 			}
 

drivers/net/ethernet/freescale/gianfar.h

@@ -92,6 +92,8 @@ extern const char gfar_driver_version[];
 #define DEFAULT_TX_RING_SIZE	256
 #define DEFAULT_RX_RING_SIZE	256
 
+#define GFAR_RX_BUFF_ALLOC	16
+
 #define GFAR_RX_MAX_RING_SIZE   256
 #define GFAR_TX_MAX_RING_SIZE   256
 
@@ -640,6 +642,7 @@ struct rmon_mib
 };
 
 struct gfar_extra_stats {
+	atomic64_t rx_alloc_err;
 	atomic64_t rx_large;
 	atomic64_t rx_short;
 	atomic64_t rx_nonoctet;
@@ -1015,9 +1018,9 @@ struct rx_q_stats {
 /**
  *	struct gfar_priv_rx_q - per rx queue structure
  *	@rx_skbuff: skb pointers
- *	@skb_currx: currently use skb pointer
  *	@rx_bd_base: First rx buffer descriptor
- *	@cur_rx: Next free rx ring entry
+ *	@next_to_use: index of the next buffer to be alloc'd
+ *	@next_to_clean: index of the next buffer to be cleaned
  *	@qindex: index of this queue
  *	@dev: back pointer to the dev structure
  *	@rx_ring_size: Rx ring size
@@ -1027,19 +1030,18 @@ struct rx_q_stats {
 
 struct gfar_priv_rx_q {
 	struct	sk_buff **rx_skbuff __aligned(SMP_CACHE_BYTES);
-	dma_addr_t rx_bd_dma_base;
 	struct	rxbd8 *rx_bd_base;
-	struct	rxbd8 *cur_rx;
 	struct	net_device *dev;
 	struct	gfar_priv_grp *grp;
-	struct rx_q_stats stats;
-	u16	skb_currx;
+	u16 rx_ring_size;
 	u16 qindex;
-	unsigned int	rx_ring_size;
-	/* RX Coalescing values */
+	u16 next_to_clean;
+	u16 next_to_use;
+	struct rx_q_stats stats;
+	u32 __iomem *rfbptr;
 	unsigned char rxcoalescing;
 	unsigned long rxic;
-	u32 __iomem *rfbptr;
+	dma_addr_t rx_bd_dma_base;
 };
 
 enum gfar_irqinfo_id {
@@ -1295,6 +1297,23 @@ static inline void gfar_clear_txbd_status(struct txbd8 *bdp)
 	bdp->lstatus = cpu_to_be32(lstatus);
 }
 
+static inline int gfar_rxbd_unused(struct gfar_priv_rx_q *rxq)
+{
+	if (rxq->next_to_clean > rxq->next_to_use)
+		return rxq->next_to_clean - rxq->next_to_use - 1;
+
+	return rxq->rx_ring_size + rxq->next_to_clean - rxq->next_to_use - 1;
+}
+
+static inline struct rxbd8 *gfar_rxbd_lastfree(struct gfar_priv_rx_q *rxq)
+{
+	int i;
+
+	i = rxq->next_to_use ? rxq->next_to_use - 1 : rxq->rx_ring_size - 1;
+
+	return &rxq->rx_bd_base[i];
+}
+
 irqreturn_t gfar_receive(int irq, void *dev_id);
 int startup_gfar(struct net_device *dev);
 void stop_gfar(struct net_device *dev);

drivers/net/ethernet/freescale/gianfar_ethtool.c

@@ -61,6 +61,8 @@ static void gfar_gdrvinfo(struct net_device *dev,
 			  struct ethtool_drvinfo *drvinfo);
 
 static const char stat_gstrings[][ETH_GSTRING_LEN] = {
+	/* extra stats */
+	"rx-allocation-errors",
 	"rx-large-frame-errors",
 	"rx-short-frame-errors",
 	"rx-non-octet-errors",
@@ -74,6 +76,7 @@ static const char stat_gstrings[][ETH_GSTRING_LEN] = {
 	"tx-underrun-errors",
 	"rx-skb-missing-errors",
 	"tx-timeout-errors",
+	/* rmon stats */
 	"tx-rx-64-frames",
 	"tx-rx-65-127-frames",
 	"tx-rx-128-255-frames",