Merge branch 'virtio_rx_merging'

Michael Dalton says:

====================
virtio-net: mergeable rx buffer size auto-tuning

The virtio-net device currently uses aligned MTU-sized mergeable receive
packet buffers. Network throughput for workloads with large average
packet size can be improved by posting larger receive packet buffers.
However, due to SKB truesize effects, posting large (e.g, PAGE_SIZE)
buffers reduces the throughput of workloads that do not benefit from GRO
and have no large inbound packets.

This patchset introduces virtio-net mergeable buffer size auto-tuning,
with buffer sizes ranging from aligned MTU-size to PAGE_SIZE. Packet
buffer size is chosen based on a per-receive queue EWMA of incoming
packet size.

To unify mergeable receive buffer memory allocation and improve
SKB frag coalescing, all mergeable buffer memory allocation is
migrated to per-receive queue page frag allocators.

The per-receive queue mergeable packet buffer size is exported via
sysfs, and the network device sysfs layer has been extended to add
support for device-specific per-receive queue sysfs attribute groups.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>

3 files changed, 36 insertions, 30 deletions

diff --git a/net/core/dev.c b/net/core/dev.c
index f87bedd51eed..288df6232006 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2083,7 +2083,7 @@ int netif_set_real_num_tx_queues(struct net_device *dev, unsigned int txq)
 }
 EXPORT_SYMBOL(netif_set_real_num_tx_queues);
 
-#ifdef CONFIG_RPS
+#ifdef CONFIG_SYSFS
 /**
  *	netif_set_real_num_rx_queues - set actual number of RX queues used
  *	@dev: Network device
@@ -5764,7 +5764,7 @@ void netif_stacked_transfer_operstate(const struct net_device *rootdev,
 }
 EXPORT_SYMBOL(netif_stacked_transfer_operstate);
 
-#ifdef CONFIG_RPS
+#ifdef CONFIG_SYSFS
 static int netif_alloc_rx_queues(struct net_device *dev)
 {
 	unsigned int i, count = dev->num_rx_queues;
@@ -6309,7 +6309,7 @@ struct net_device *alloc_netdev_mqs(int sizeof_priv, const char *name,
 		return NULL;
 	}
 
-#ifdef CONFIG_RPS
+#ifdef CONFIG_SYSFS
 	if (rxqs < 1) {
 		pr_err("alloc_netdev: Unable to allocate device with zero RX queues\n");
 		return NULL;
@@ -6365,7 +6365,7 @@ struct net_device *alloc_netdev_mqs(int sizeof_priv, const char *name,
 	if (netif_alloc_netdev_queues(dev))
 		goto free_all;
 
-#ifdef CONFIG_RPS
+#ifdef CONFIG_SYSFS
 	dev->num_rx_queues = rxqs;
 	dev->real_num_rx_queues = rxqs;
 	if (netif_alloc_rx_queues(dev))
@@ -6385,7 +6385,7 @@ free_all:
 free_pcpu:
 	free_percpu(dev->pcpu_refcnt);
 	netif_free_tx_queues(dev);
-#ifdef CONFIG_RPS
+#ifdef CONFIG_SYSFS
 	kfree(dev->_rx);
 #endif
 
@@ -6410,7 +6410,7 @@ void free_netdev(struct net_device *dev)
 	release_net(dev_net(dev));
 
 	netif_free_tx_queues(dev);
-#ifdef CONFIG_RPS
+#ifdef CONFIG_SYSFS
 	kfree(dev->_rx);
 #endif
 
diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c
index 49843bf7e43e..7eeadeecc5a2 100644
--- a/net/core/net-sysfs.c
+++ b/net/core/net-sysfs.c
@@ -498,17 +498,7 @@ static struct attribute_group wireless_group = {
 #define net_class_groups	NULL
 #endif /* CONFIG_SYSFS */
 
-#ifdef CONFIG_RPS
-/*
- * RX queue sysfs structures and functions.
- */
-struct rx_queue_attribute {
-	struct attribute attr;
-	ssize_t (*show)(struct netdev_rx_queue *queue,
-	    struct rx_queue_attribute *attr, char *buf);
-	ssize_t (*store)(struct netdev_rx_queue *queue,
-	    struct rx_queue_attribute *attr, const char *buf, size_t len);
-};
+#ifdef CONFIG_SYSFS
 #define to_rx_queue_attr(_attr) container_of(_attr,		\
     struct rx_queue_attribute, attr)
 
@@ -543,6 +533,7 @@ static const struct sysfs_ops rx_queue_sysfs_ops = {
 	.store = rx_queue_attr_store,
 };
 
+#ifdef CONFIG_RPS
 static ssize_t show_rps_map(struct netdev_rx_queue *queue,
 			    struct rx_queue_attribute *attribute, char *buf)
 {
@@ -718,16 +709,20 @@ static struct rx_queue_attribute rps_cpus_attribute =
 static struct rx_queue_attribute rps_dev_flow_table_cnt_attribute =
 	__ATTR(rps_flow_cnt, S_IRUGO | S_IWUSR,
 	    show_rps_dev_flow_table_cnt, store_rps_dev_flow_table_cnt);
+#endif /* CONFIG_RPS */
 
 static struct attribute *rx_queue_default_attrs[] = {
+#ifdef CONFIG_RPS
 	&rps_cpus_attribute.attr,
 	&rps_dev_flow_table_cnt_attribute.attr,
+#endif
 	NULL
 };
 
 static void rx_queue_release(struct kobject *kobj)
 {
 	struct netdev_rx_queue *queue = to_rx_queue(kobj);
+#ifdef CONFIG_RPS
 	struct rps_map *map;
 	struct rps_dev_flow_table *flow_table;
 
@@ -743,6 +738,7 @@ static void rx_queue_release(struct kobject *kobj)
 		RCU_INIT_POINTER(queue->rps_flow_table, NULL);
 		call_rcu(&flow_table->rcu, rps_dev_flow_table_release);
 	}
+#endif
 
 	memset(kobj, 0, sizeof(*kobj));
 	dev_put(queue->dev);
@@ -763,25 +759,36 @@ static int rx_queue_add_kobject(struct net_device *net, int index)
 	kobj->kset = net->queues_kset;
 	error = kobject_init_and_add(kobj, &rx_queue_ktype, NULL,
 	    "rx-%u", index);
-	if (error) {
-		kobject_put(kobj);
-		return error;
+	if (error)
+		goto exit;
+
+	if (net->sysfs_rx_queue_group) {
+		error = sysfs_create_group(kobj, net->sysfs_rx_queue_group);
+		if (error)
+			goto exit;
 	}
 
 	kobject_uevent(kobj, KOBJ_ADD);
 	dev_hold(queue->dev);
 
 	return error;
+exit:
+	kobject_put(kobj);
+	return error;
 }
-#endif /* CONFIG_RPS */
+#endif /* CONFIG_SYFS */
 
 int
 net_rx_queue_update_kobjects(struct net_device *net, int old_num, int new_num)
 {
-#ifdef CONFIG_RPS
+#ifdef CONFIG_SYSFS
 	int i;
 	int error = 0;
 
+#ifndef CONFIG_RPS
+	if (!net->sysfs_rx_queue_group)
+		return 0;
+#endif
 	for (i = old_num; i < new_num; i++) {
 		error = rx_queue_add_kobject(net, i);
 		if (error) {
@@ -790,8 +797,12 @@ net_rx_queue_update_kobjects(struct net_device *net, int old_num, int new_num)
 		}
 	}
 
-	while (--i >= new_num)
+	while (--i >= new_num) {
+		if (net->sysfs_rx_queue_group)
+			sysfs_remove_group(&net->_rx[i].kobj,
+					   net->sysfs_rx_queue_group);
 		kobject_put(&net->_rx[i].kobj);
+	}
 
 	return error;
 #else
@@ -1155,9 +1166,6 @@ static int register_queue_kobjects(struct net_device *net)
 	    NULL, &net->dev.kobj);
 	if (!net->queues_kset)
 		return -ENOMEM;
-#endif
-
-#ifdef CONFIG_RPS
 	real_rx = net->real_num_rx_queues;
 #endif
 	real_tx = net->real_num_tx_queues;
@@ -1184,7 +1192,7 @@ static void remove_queue_kobjects(struct net_device *net)
 {
 	int real_rx = 0, real_tx = 0;
 
-#ifdef CONFIG_RPS
+#ifdef CONFIG_SYSFS
 	real_rx = net->real_num_rx_queues;
 #endif
 	real_tx = net->real_num_tx_queues;
diff --git a/net/core/sock.c b/net/core/sock.c
index 85ad6f0d3898..b3f7ee3008a0 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -1836,9 +1836,7 @@ bool skb_page_frag_refill(unsigned int sz, struct page_frag *pfrag, gfp_t prio)
 		put_page(pfrag->page);
 	}
 
-	/* We restrict high order allocations to users that can afford to wait */
-	order = (prio & __GFP_WAIT) ? SKB_FRAG_PAGE_ORDER : 0;
-
+	order = SKB_FRAG_PAGE_ORDER;
 	do {
 		gfp_t gfp = prio;