1 files changed, 110 insertions, 228 deletions

diff --git a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c
index ece8d1804d13..d8c10764f130 100644
--- a/drivers/net/xen-netfront.c
+++ b/drivers/net/xen-netfront.c
@@ -77,7 +77,9 @@ struct netfront_cb {
 
 #define NET_TX_RING_SIZE __CONST_RING_SIZE(xen_netif_tx, PAGE_SIZE)
 #define NET_RX_RING_SIZE __CONST_RING_SIZE(xen_netif_rx, PAGE_SIZE)
-#define TX_MAX_TARGET min_t(int, NET_TX_RING_SIZE, 256)
+
+/* Minimum number of Rx slots (includes slot for GSO metadata). */
+#define NET_RX_SLOTS_MIN (XEN_NETIF_NR_SLOTS_MIN + 1)
 
 /* Queue name is interface name with "-qNNN" appended */
 #define QUEUE_NAME_SIZE (IFNAMSIZ + 6)
@@ -86,10 +88,8 @@ struct netfront_cb {
 #define IRQ_NAME_SIZE (QUEUE_NAME_SIZE + 3)
 
 struct netfront_stats {
-	u64			rx_packets;
-	u64			tx_packets;
-	u64			rx_bytes;
-	u64			tx_bytes;
+	u64			packets;
+	u64			bytes;
 	struct u64_stats_sync	syncp;
 };
 
@@ -137,13 +137,6 @@ struct netfront_queue {
 	struct xen_netif_rx_front_ring rx;
 	int rx_ring_ref;
 
-	/* Receive-ring batched refills. */
-#define RX_MIN_TARGET 8
-#define RX_DFL_MIN_TARGET 64
-#define RX_MAX_TARGET min_t(int, NET_RX_RING_SIZE, 256)
-	unsigned rx_min_target, rx_max_target, rx_target;
-	struct sk_buff_head rx_batch;
-
 	struct timer_list rx_refill_timer;
 
 	struct sk_buff *rx_skbs[NET_RX_RING_SIZE];
@@ -165,7 +158,8 @@ struct netfront_info {
 	struct netfront_queue *queues;
 
 	/* Statistics */
-	struct netfront_stats __percpu *stats;
+	struct netfront_stats __percpu *rx_stats;
+	struct netfront_stats __percpu *tx_stats;
 
 	atomic_t rx_gso_checksum_fixup;
 };
@@ -251,7 +245,7 @@ static void rx_refill_timeout(unsigned long data)
 static int netfront_tx_slot_available(struct netfront_queue *queue)
 {
 	return (queue->tx.req_prod_pvt - queue->tx.rsp_cons) <
-		(TX_MAX_TARGET - MAX_SKB_FRAGS - 2);
+		(NET_TX_RING_SIZE - MAX_SKB_FRAGS - 2);
 }
 
 static void xennet_maybe_wake_tx(struct netfront_queue *queue)
@@ -265,77 +259,55 @@ static void xennet_maybe_wake_tx(struct netfront_queue *queue)
 		netif_tx_wake_queue(netdev_get_tx_queue(dev, queue->id));
 }
 
-static void xennet_alloc_rx_buffers(struct netfront_queue *queue)
+
+static struct sk_buff *xennet_alloc_one_rx_buffer(struct netfront_queue *queue)
 {
-	unsigned short id;
 	struct sk_buff *skb;
 	struct page *page;
-	int i, batch_target, notify;
-	RING_IDX req_prod = queue->rx.req_prod_pvt;
-	grant_ref_t ref;
-	unsigned long pfn;
-	void *vaddr;
-	struct xen_netif_rx_request *req;
 
-	if (unlikely(!netif_carrier_ok(queue->info->netdev)))
-		return;
+	skb = __netdev_alloc_skb(queue->info->netdev,
+				 RX_COPY_THRESHOLD + NET_IP_ALIGN,
+				 GFP_ATOMIC | __GFP_NOWARN);
+	if (unlikely(!skb))
+		return NULL;
 
-	/*
-	 * Allocate skbuffs greedily, even though we batch updates to the
-	 * receive ring. This creates a less bursty demand on the memory
-	 * allocator, so should reduce the chance of failed allocation requests
-	 * both for ourself and for other kernel subsystems.
-	 */
-	batch_target = queue->rx_target - (req_prod - queue->rx.rsp_cons);
-	for (i = skb_queue_len(&queue->rx_batch); i < batch_target; i++) {
-		skb = __netdev_alloc_skb(queue->info->netdev,
-					 RX_COPY_THRESHOLD + NET_IP_ALIGN,
-					 GFP_ATOMIC | __GFP_NOWARN);
-		if (unlikely(!skb))
-			goto no_skb;
-
-		/* Align ip header to a 16 bytes boundary */
-		skb_reserve(skb, NET_IP_ALIGN);
-
-		page = alloc_page(GFP_ATOMIC | __GFP_NOWARN);
-		if (!page) {
-			kfree_skb(skb);
-no_skb:
-			/* Could not allocate any skbuffs. Try again later. */
-			mod_timer(&queue->rx_refill_timer,
-				  jiffies + (HZ/10));
-
-			/* Any skbuffs queued for refill? Force them out. */
-			if (i != 0)
-				goto refill;
-			break;
-		}
-
-		skb_add_rx_frag(skb, 0, page, 0, 0, PAGE_SIZE);
-		__skb_queue_tail(&queue->rx_batch, skb);
+	page = alloc_page(GFP_ATOMIC | __GFP_NOWARN);
+	if (!page) {
+		kfree_skb(skb);
+		return NULL;
 	}
+	skb_add_rx_frag(skb, 0, page, 0, 0, PAGE_SIZE);
+
+	/* Align ip header to a 16 bytes boundary */
+	skb_reserve(skb, NET_IP_ALIGN);
+	skb->dev = queue->info->netdev;
+
+	return skb;
+}
+
 
-	/* Is the batch large enough to be worthwhile? */
-	if (i < (queue->rx_target/2)) {
-		if (req_prod > queue->rx.sring->req_prod)
-			goto push;
+static void xennet_alloc_rx_buffers(struct netfront_queue *queue)
+{
+	RING_IDX req_prod = queue->rx.req_prod_pvt;
+	int notify;
+
+	if (unlikely(!netif_carrier_ok(queue->info->netdev)))
 		return;
-	}
 
-	/* Adjust our fill target if we risked running out of buffers. */
-	if (((req_prod - queue->rx.sring->rsp_prod) < (queue->rx_target / 4)) &&
-	    ((queue->rx_target *= 2) > queue->rx_max_target))
-		queue->rx_target = queue->rx_max_target;
+	for (req_prod = queue->rx.req_prod_pvt;
+	     req_prod - queue->rx.rsp_cons < NET_RX_RING_SIZE;
+	     req_prod++) {
+		struct sk_buff *skb;
+		unsigned short id;
+		grant_ref_t ref;
+		unsigned long pfn;
+		struct xen_netif_rx_request *req;
 
- refill:
-	for (i = 0; ; i++) {
-		skb = __skb_dequeue(&queue->rx_batch);
-		if (skb == NULL)
+		skb = xennet_alloc_one_rx_buffer(queue);
+		if (!skb)
 			break;
 
-		skb->dev = queue->info->netdev;
-
-		id = xennet_rxidx(req_prod + i);
+		id = xennet_rxidx(req_prod);
 
 		BUG_ON(queue->rx_skbs[id]);
 		queue->rx_skbs[id] = skb;
@@ -345,9 +317,8 @@ no_skb:
 		queue->grant_rx_ref[id] = ref;
 
 		pfn = page_to_pfn(skb_frag_page(&skb_shinfo(skb)->frags[0]));
-		vaddr = page_address(skb_frag_page(&skb_shinfo(skb)->frags[0]));
 
-		req = RING_GET_REQUEST(&queue->rx, req_prod + i);
+		req = RING_GET_REQUEST(&queue->rx, req_prod);
 		gnttab_grant_foreign_access_ref(ref,
 						queue->info->xbdev->otherend_id,
 						pfn_to_mfn(pfn),
@@ -357,11 +328,16 @@ no_skb:
 		req->gref = ref;
 	}
 
+	queue->rx.req_prod_pvt = req_prod;
+
+	/* Not enough requests? Try again later. */
+	if (req_prod - queue->rx.rsp_cons < NET_RX_SLOTS_MIN) {
+		mod_timer(&queue->rx_refill_timer, jiffies + (HZ/10));
+		return;
+	}
+
 	wmb();		/* barrier so backend seens requests */
 
-	/* Above is a suitable barrier to ensure backend will see requests. */
-	queue->rx.req_prod_pvt = req_prod + i;
- push:
 	RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(&queue->rx, notify);
 	if (notify)
 		notify_remote_via_irq(queue->rx_irq);
@@ -588,7 +564,7 @@ static int xennet_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	unsigned short id;
 	struct netfront_info *np = netdev_priv(dev);
-	struct netfront_stats *stats = this_cpu_ptr(np->stats);
+	struct netfront_stats *tx_stats = this_cpu_ptr(np->tx_stats);
 	struct xen_netif_tx_request *tx;
 	char *data = skb->data;
 	RING_IDX i;
@@ -627,6 +603,9 @@ static int xennet_start_xmit(struct sk_buff *skb, struct net_device *dev)
 				    slots, skb->len);
 		if (skb_linearize(skb))
 			goto drop;
+		data = skb->data;
+		offset = offset_in_page(data);
+		len = skb_headlen(skb);
 	}
 
 	spin_lock_irqsave(&queue->tx_lock, flags);
@@ -692,10 +671,10 @@ static int xennet_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	if (notify)
 		notify_remote_via_irq(queue->tx_irq);
 
-	u64_stats_update_begin(&stats->syncp);
-	stats->tx_bytes += skb->len;
-	stats->tx_packets++;
-	u64_stats_update_end(&stats->syncp);
+	u64_stats_update_begin(&tx_stats->syncp);
+	tx_stats->bytes += skb->len;
+	tx_stats->packets++;
+	u64_stats_update_end(&tx_stats->syncp);
 
 	/* Note: It is not safe to access skb after xennet_tx_buf_gc()! */
 	xennet_tx_buf_gc(queue);
@@ -951,7 +930,7 @@ static int checksum_setup(struct net_device *dev, struct sk_buff *skb)
 static int handle_incoming_queue(struct netfront_queue *queue,
 				 struct sk_buff_head *rxq)
 {
-	struct netfront_stats *stats = this_cpu_ptr(queue->info->stats);
+	struct netfront_stats *rx_stats = this_cpu_ptr(queue->info->rx_stats);
 	int packets_dropped = 0;
 	struct sk_buff *skb;
 
@@ -972,10 +951,10 @@ static int handle_incoming_queue(struct netfront_queue *queue,
 			continue;
 		}
 
-		u64_stats_update_begin(&stats->syncp);
-		stats->rx_packets++;
-		stats->rx_bytes += skb->len;
-		u64_stats_update_end(&stats->syncp);
+		u64_stats_update_begin(&rx_stats->syncp);
+		rx_stats->packets++;
+		rx_stats->bytes += skb->len;
+		u64_stats_update_end(&rx_stats->syncp);
 
 		/* Pass it up. */
 		napi_gro_receive(&queue->napi, skb);
@@ -997,7 +976,6 @@ static int xennet_poll(struct napi_struct *napi, int budget)
 	struct sk_buff_head rxq;
 	struct sk_buff_head errq;
 	struct sk_buff_head tmpq;
-	unsigned long flags;
 	int err;
 
 	spin_lock(&queue->rx_lock);
@@ -1065,27 +1043,16 @@ err:
 
 	work_done -= handle_incoming_queue(queue, &rxq);
 
-	/* If we get a callback with very few responses, reduce fill target. */
-	/* NB. Note exponential increase, linear decrease. */
-	if (((queue->rx.req_prod_pvt - queue->rx.sring->rsp_prod) >
-	     ((3*queue->rx_target) / 4)) &&
-	    (--queue->rx_target < queue->rx_min_target))
-		queue->rx_target = queue->rx_min_target;
-
 	xennet_alloc_rx_buffers(queue);
 
 	if (work_done < budget) {
 		int more_to_do = 0;
 
-		napi_gro_flush(napi, false);
-
-		local_irq_save(flags);
+		napi_complete(napi);
 
 		RING_FINAL_CHECK_FOR_RESPONSES(&queue->rx, more_to_do);
-		if (!more_to_do)
-			__napi_complete(napi);
-
-		local_irq_restore(flags);
+		if (more_to_do)
+			napi_schedule(napi);
 	}
 
 	spin_unlock(&queue->rx_lock);
@@ -1111,18 +1078,22 @@ static struct rtnl_link_stats64 *xennet_get_stats64(struct net_device *dev,
 	int cpu;
 
 	for_each_possible_cpu(cpu) {
-		struct netfront_stats *stats = per_cpu_ptr(np->stats, cpu);
+		struct netfront_stats *rx_stats = per_cpu_ptr(np->rx_stats, cpu);
+		struct netfront_stats *tx_stats = per_cpu_ptr(np->tx_stats, cpu);
 		u64 rx_packets, rx_bytes, tx_packets, tx_bytes;
 		unsigned int start;
 
 		do {
-			start = u64_stats_fetch_begin_irq(&stats->syncp);
+			start = u64_stats_fetch_begin_irq(&tx_stats->syncp);
+			tx_packets = tx_stats->packets;
+			tx_bytes = tx_stats->bytes;
+		} while (u64_stats_fetch_retry_irq(&tx_stats->syncp, start));
 
-			rx_packets = stats->rx_packets;
-			tx_packets = stats->tx_packets;
-			rx_bytes = stats->rx_bytes;
-			tx_bytes = stats->tx_bytes;
-		} while (u64_stats_fetch_retry_irq(&stats->syncp, start));
+		do {
+			start = u64_stats_fetch_begin_irq(&rx_stats->syncp);
+			rx_packets = rx_stats->packets;
+			rx_bytes = rx_stats->bytes;
+		} while (u64_stats_fetch_retry_irq(&rx_stats->syncp, start));
 
 		tot->rx_packets += rx_packets;
 		tot->tx_packets += tx_packets;
@@ -1307,6 +1278,15 @@ static const struct net_device_ops xennet_netdev_ops = {
 #endif
 };
 
+static void xennet_free_netdev(struct net_device *netdev)
+{
+	struct netfront_info *np = netdev_priv(netdev);
+
+	free_percpu(np->rx_stats);
+	free_percpu(np->tx_stats);
+	free_netdev(netdev);
+}
+
 static struct net_device *xennet_create_dev(struct xenbus_device *dev)
 {
 	int err;
@@ -1327,8 +1307,11 @@ static struct net_device *xennet_create_dev(struct xenbus_device *dev)
 	np->queues = NULL;
 
 	err = -ENOMEM;
-	np->stats = netdev_alloc_pcpu_stats(struct netfront_stats);
-	if (np->stats == NULL)
+	np->rx_stats = netdev_alloc_pcpu_stats(struct netfront_stats);
+	if (np->rx_stats == NULL)
+		goto exit;
+	np->tx_stats = netdev_alloc_pcpu_stats(struct netfront_stats);
+	if (np->tx_stats == NULL)
 		goto exit;
 
 	netdev->netdev_ops	= &xennet_netdev_ops;
@@ -1359,7 +1342,7 @@ static struct net_device *xennet_create_dev(struct xenbus_device *dev)
 	return netdev;
 
  exit:
-	free_netdev(netdev);
+	xennet_free_netdev(netdev);
 	return ERR_PTR(err);
 }
 
@@ -1401,7 +1384,7 @@ static int netfront_probe(struct xenbus_device *dev,
 	return 0;
 
  fail:
-	free_netdev(netdev);
+	xennet_free_netdev(netdev);
 	dev_set_drvdata(&dev->dev, NULL);
 	return err;
 }
@@ -1638,11 +1621,6 @@ static int xennet_init_queue(struct netfront_queue *queue)
 	spin_lock_init(&queue->tx_lock);
 	spin_lock_init(&queue->rx_lock);
 
-	skb_queue_head_init(&queue->rx_batch);
-	queue->rx_target     = RX_DFL_MIN_TARGET;
-	queue->rx_min_target = RX_DFL_MIN_TARGET;
-	queue->rx_max_target = RX_MAX_TARGET;
-
 	init_timer(&queue->rx_refill_timer);
 	queue->rx_refill_timer.data = (unsigned long)queue;
 	queue->rx_refill_timer.function = rx_refill_timeout;
@@ -1665,7 +1643,7 @@ static int xennet_init_queue(struct netfront_queue *queue)
 	}
 
 	/* A grant for every tx ring slot */
-	if (gnttab_alloc_grant_references(TX_MAX_TARGET,
+	if (gnttab_alloc_grant_references(NET_TX_RING_SIZE,
 					  &queue->gref_tx_head) < 0) {
 		pr_alert("can't alloc tx grant refs\n");
 		err = -ENOMEM;
@@ -1673,7 +1651,7 @@ static int xennet_init_queue(struct netfront_queue *queue)
 	}
 
 	/* A grant for every rx ring slot */
-	if (gnttab_alloc_grant_references(RX_MAX_TARGET,
+	if (gnttab_alloc_grant_references(NET_RX_RING_SIZE,
 					  &queue->gref_rx_head) < 0) {
 		pr_alert("can't alloc rx grant refs\n");
 		err = -ENOMEM;
@@ -2141,30 +2119,18 @@ static const struct ethtool_ops xennet_ethtool_ops =
 };
 
 #ifdef CONFIG_SYSFS
-static ssize_t show_rxbuf_min(struct device *dev,
-			      struct device_attribute *attr, char *buf)
+static ssize_t show_rxbuf(struct device *dev,
+			  struct device_attribute *attr, char *buf)
 {
-	struct net_device *netdev = to_net_dev(dev);
-	struct netfront_info *info = netdev_priv(netdev);
-	unsigned int num_queues = netdev->real_num_tx_queues;
-
-	if (num_queues)
-		return sprintf(buf, "%u\n", info->queues[0].rx_min_target);
-	else
-		return sprintf(buf, "%u\n", RX_MIN_TARGET);
+	return sprintf(buf, "%lu\n", NET_RX_RING_SIZE);
 }
 
-static ssize_t store_rxbuf_min(struct device *dev,
-			       struct device_attribute *attr,
-			       const char *buf, size_t len)
+static ssize_t store_rxbuf(struct device *dev,
+			   struct device_attribute *attr,
+			   const char *buf, size_t len)
 {
-	struct net_device *netdev = to_net_dev(dev);
-	struct netfront_info *np = netdev_priv(netdev);
-	unsigned int num_queues = netdev->real_num_tx_queues;
 	char *endp;
 	unsigned long target;
-	unsigned int i;
-	struct netfront_queue *queue;
 
 	if (!capable(CAP_NET_ADMIN))
 		return -EPERM;
@@ -2173,97 +2139,15 @@ static ssize_t store_rxbuf_min(struct device *dev,
 	if (endp == buf)
 		return -EBADMSG;
 
-	if (target < RX_MIN_TARGET)
-		target = RX_MIN_TARGET;
-	if (target > RX_MAX_TARGET)
-		target = RX_MAX_TARGET;
+	/* rxbuf_min and rxbuf_max are no longer configurable. */
 
-	for (i = 0; i < num_queues; ++i) {
-		queue = &np->queues[i];
-		spin_lock_bh(&queue->rx_lock);
-		if (target > queue->rx_max_target)
-			queue->rx_max_target = target;
-		queue->rx_min_target = target;
-		if (target > queue->rx_target)
-			queue->rx_target = target;
-
-		xennet_alloc_rx_buffers(queue);
-
-		spin_unlock_bh(&queue->rx_lock);
-	}
 	return len;
 }
 
-static ssize_t show_rxbuf_max(struct device *dev,
-			      struct device_attribute *attr, char *buf)
-{
-	struct net_device *netdev = to_net_dev(dev);
-	struct netfront_info *info = netdev_priv(netdev);
-	unsigned int num_queues = netdev->real_num_tx_queues;
-
-	if (num_queues)
-		return sprintf(buf, "%u\n", info->queues[0].rx_max_target);
-	else
-		return sprintf(buf, "%u\n", RX_MAX_TARGET);
-}
-
-static ssize_t store_rxbuf_max(struct device *dev,
-			       struct device_attribute *attr,
-			       const char *buf, size_t len)
-{
-	struct net_device *netdev = to_net_dev(dev);
-	struct netfront_info *np = netdev_priv(netdev);
-	unsigned int num_queues = netdev->real_num_tx_queues;
-	char *endp;
-	unsigned long target;
-	unsigned int i = 0;
-	struct netfront_queue *queue = NULL;
-
-	if (!capable(CAP_NET_ADMIN))
-		return -EPERM;
-
-	target = simple_strtoul(buf, &endp, 0);
-	if (endp == buf)
-		return -EBADMSG;
-
-	if (target < RX_MIN_TARGET)
-		target = RX_MIN_TARGET;
-	if (target > RX_MAX_TARGET)
-		target = RX_MAX_TARGET;
-
-	for (i = 0; i < num_queues; ++i) {
-		queue = &np->queues[i];
-		spin_lock_bh(&queue->rx_lock);
-		if (target < queue->rx_min_target)
-			queue->rx_min_target = target;
-		queue->rx_max_target = target;
-		if (target < queue->rx_target)
-			queue->rx_target = target;
-
-		xennet_alloc_rx_buffers(queue);
-
-		spin_unlock_bh(&queue->rx_lock);
-	}
-	return len;
-}
-
-static ssize_t show_rxbuf_cur(struct device *dev,
-			      struct device_attribute *attr, char *buf)
-{
-	struct net_device *netdev = to_net_dev(dev);
-	struct netfront_info *info = netdev_priv(netdev);
-	unsigned int num_queues = netdev->real_num_tx_queues;
-
-	if (num_queues)
-		return sprintf(buf, "%u\n", info->queues[0].rx_target);
-	else
-		return sprintf(buf, "0\n");
-}
-
 static struct device_attribute xennet_attrs[] = {
-	__ATTR(rxbuf_min, S_IRUGO|S_IWUSR, show_rxbuf_min, store_rxbuf_min),
-	__ATTR(rxbuf_max, S_IRUGO|S_IWUSR, show_rxbuf_max, store_rxbuf_max),
-	__ATTR(rxbuf_cur, S_IRUGO, show_rxbuf_cur, NULL),
+	__ATTR(rxbuf_min, S_IRUGO|S_IWUSR, show_rxbuf, store_rxbuf),
+	__ATTR(rxbuf_max, S_IRUGO|S_IWUSR, show_rxbuf, store_rxbuf),
+	__ATTR(rxbuf_cur, S_IRUGO, show_rxbuf, NULL),
 };
 
 static int xennet_sysfs_addif(struct net_device *netdev)
@@ -2320,9 +2204,7 @@ static int xennet_remove(struct xenbus_device *dev)
 		info->queues = NULL;
 	}
 
-	free_percpu(info->stats);
-
-	free_netdev(info->netdev);
+	xennet_free_netdev(info->netdev);
 
 	return 0;
 }